Integration of deep transcriptome and proteome analyses reveals the components of alkaloid

RESEARCH ARTICLE Open Access

Integration of deep transcriptome and proteomeanalyses reveals the components of alkaloidmetabolism in opium poppy cell culturesIsabel Desgagneacute-Penix1 Morgan F Khan23 David C Schriemer23 Dustin Cram23 Jacek Nowak23 Peter J Facchini1

Abstract

Background Papaver somniferum (opium poppy) is the source for several pharmaceutical benzylisoquinolinealkaloids including morphine the codeine and sanguinarine In response to treatment with a fungal elicitor thebiosynthesis and accumulation of sanguinarine is induced along with other plant defense responses in opiumpoppy cell cultures The transcriptional induction of alkaloid metabolism in cultured cells provides an opportunityto identify components of this process via the integration of deep transcriptome and proteome databasesgenerated using next-generation technologies

Results A cDNA library was prepared for opium poppy cell cultures treated with a fungal elicitor for 10 h Using454 GS-FLX Titanium pyrosequencing 427369 expressed sequence tags (ESTs) with an average length of 462 bpwere generated Assembly of these sequences yielded 93723 unigenes of which 23753 were assigned GeneOntology annotations Transcripts encoding all known sanguinarine biosynthetic enzymes were identified in theEST database 5 of which were represented among the 50 most abundant transcripts Liquid chromatography-tandem mass spectrometry (LC-MSMS) of total protein extracts from cell cultures treated with a fungal elicitor for50 h facilitated the identification of 1004 proteins Proteins were fractionated by one-dimensional SDS-PAGE anddigested with trypsin prior to LC-MSMS analysis Query of an opium poppy-specific EST database substantiallyenhanced peptide identification Eight out of 10 known sanguinarine biosynthetic enzymes and many relevantprimary metabolic enzymes were represented in the peptide database

Conclusions The integration of deep transcriptome and proteome analyses provides an effective platform tocatalogue the components of secondary metabolism and to identify genes encoding uncharacterized enzymesThe establishment of corresponding transcript and protein databases generated by next-generation technologies ina system with a well-defined metabolite profile facilitates an improved linkage between genes enzymes andpathway components The proteome database represents the most relevant alkaloid-producing enzymescompared with the much deeper and more complete transcriptome library The transcript database containedfull-length mRNAs encoding most alkaloid biosynthetic enzymes which is a key requirement for the functionalcharacterization of novel gene candidates

BackgroundOpium poppy (Papaver somniferum) remains our mostimportant source for several pharmaceutical benzylisoqui-noline alkaloids (BIAs) including the narcotic analgesicmorphine the anti-tussive drug codeine the vasodilatorpapaverine and the antimicrobial agent sanguinarine In

opium poppy plants most BIAs (eg morphine codeineand papaverine) occur in the cytoplasm (ie latex) of spe-cialized cells known as laticifers that are associated withthe phloem in all organs Although latex and thus mostBIAs are most abundant on shoot organs of opium poppythe antimicrobial alkaloid sanguinarine accumulates con-stitutively in roots possibly in association with cell typesother than laticifers Although opium poppy cell culturesdo not produce BIAs constitutively the biosynthesis ofsanguinarine is induced in response to treatment of the

Correspondence pfacchinucalgaryca1Department of Biological Sciences University of Calgary Calgary AlbertaT2N 1N4 CanadaFull list of author information is available at the end of the article

Desgagneacute-Penix et al BMC Plant Biology 2010 10252httpwwwbiomedcentralcom1471-222910252

copy 2010 Desgagneacute-Penix et al licensee BioMed Central Ltd This is an Open Access article distributed under the terms of the CreativeCommons Attribution License (httpcreativecommonsorglicensesby20) which permits unrestricted use distribution andreproduction in any medium provided the original work is properly cited

cells with a fungal elicitor As such opium poppy cell cul-tures provide an effective model system to investigate theinducible regulation of BIA metabolism and other plantdefense pathways The induction of sanguinarine biosynth-esis and supporting metabolism in elicitor-treated opiumpoppy cell cultures has been characterized using a varietyof technologies including EST and species-specific micro-array analyses to analyze the transcriptome [1] LC-MSMS to survey the proteome [2] and Fourier-transformion-cyclotron resonance-mass spectrometry (FT-ICR-MS)to profile the metabolome [13] Although these studiesprovided valuable insights into the response of opiumpoppy cell cultures to fungal elicitor treatment the tech-nologies used to generate the various databases werelimited in terms of the depth of penetration into the tran-scriptome proteome and metabolome of the culturedcells For example LC-MSMS peptide analysis of 340spots isolated by two-dimensional SDS-PAGE led to theidentification of 219 proteins using a combination of pub-lic and species-specific sequence databases More exten-sive genomics resources for opium poppy would improvethe downstream identification and discovery of enzymesinvolved in alkaloid biosynthesis New sequencing technol-ogies such as 454 pyrosequencing and advances in LC-MSMS-based proteomics and bioinformatics will expandthe application of genomics methodologies to a vast arrayof non-model plants that produce interesting and valuablemetabolitesThe biosynthesis of BIAs in opium poppy starts with

the condensation of two tyrosine derivatives dopamineand 4-hydroxyphenylacetaldehyde (4-HPAA) by norco-claurine synthase (NCS) to yield (S)-norcoclaurine(Figure 1) [4-6] The formation of dopamine involvesthe decarboxylation of tyrosine andor dihydrophenylala-nine (DOPA) by tyrosineDOPA decarboxylase (TYDC)[7] (S)-Norcoclaurine is then methylated by the norco-claurine 6-O-methyltransferase (6OMT) and coclaurineN-methyltransferase (CNMT) to yield (S)-methylco-claurine [8-11] The P450-dependent monooxygenase(S)-N-methylcoclaurine-3rsquo-hydroxylase (NMCH orCYP80B3) catalyzes the 3rsquo-hydroxylation of (S)-N-methylcoclaurine prior to the formation of (S)-reticulineby the 3rsquo-hydroxy-N-methylcoclaurine 4rsquo-O-methyltrans-ferase (4rsquoOMT) [811-14] (S)-Reticuline is the centralintermediate in the biosynthesis of most BIA structuraltypes including morphinans (eg morphine) benzophe-nanthridines (eg sanguinarine) and substituted benzyli-soquinolines (eg laudanine and papaverine) (AdditionalFile 1)The berberine bridge enzyme (BBE) converts (S)-

reticuline to (S)-scoulerine as the first committed stepin sanguinarine biosynthesis (Figure 1) [1315-17] TwoP450-dependent enzymes cheilanthifoline synthase(CheSyn) and the stylopine synthase (StySyn) catalyze

the formation of two methylenedioxy bridges and yield(S)-stylopine [1819] Subsequently tetrahydroprotober-berine cis-N-methyltransferase (TNMT) converts(S)-stylopine to (S)-cis-N-methylstylopine [20] Twoadditional P450-dependent enzymes N-methylstylopine14-hydroxylase (MSH) [21] and protopine 6-hydroxylase(P6H) are responsible for the conversion of (S)-cis-N-methylstylopine to 6-hydroxyprotopine which sponta-neously rearranges to yield dihydrosanguinarine [2122]Finally dihydrosanguinarine is oxidized to sanguinarineby the oxygen-dependent oxidoreductase dihydro-benzophenanthridine oxidase (DBOX) [2324] Thebiosynthesis of morphine involves the epimerization of(S)-reticuline to (R)-reticuline followed by a series ofC-C phenol coupling two reductions O-acetylation andtwo O-demethylations [25-29] Reticuline 7-O-methyl-transferase (7OMT) converts (S)-reticuline to (S)-lauda-nine [9] whereas norreticuline 7-O-methyltransferase(N7OMT) yields norlaudanine from norreticuline [30](Additional File 1) Cognate cDNAs have been reportedfor all of the aforementioned enzymes with the excep-tion of MSH P6H and DBOXThe standard approach to establish genomics resources

for non-model plant species involves the random genera-tion of expressed sequence tags (ESTs) from a cDNAphagemid library using dideoxy chain-termination (San-ger) sequencing technology Next-generation technolo-gies such as 454 pyrosequencing have the potential todramatically increase the availability of sequence data[3132] The redundancy and depth of coverage of 454pyrosequencing also provides and unbiased representa-tion of transcript abundance which is useful for relativegene expression analysis especially in non-model plantsthat lack complete genome sequence information [3334]However despite the fundamental importance of tran-scriptome analysis in genomics-based research the fre-quent incongruity between steady-state protein levels andthe abundance of cognate gene transcripts [35] is crucialto the interpretation of relative gene expression profilesin the context of systems biology or gene discovery appli-cations Complementary analysis of the most abundantproteins combined with a comprehensive transcriptomedatabase provides an important validation tool for therelative importance of gene transcripts within a givencell tissue or organ Transcript and protein sequencedatabases have recently been reported for opium poppycell cultures using Sanger sequencing of randomlyselected cDNAs and first-generation LC-MSMS analysisof proteins isolated by two-dimensional SDS-PAGE [12]In terms of the components of sanguinarine metabolismtranscripts corresponding to all known biosyntheticgenes were present in the EST database although somewere represented by relatively few sequence reads Com-bined with a low-throughput two-dimensional sampling

Page 2 of 17

Figure 1 Biosynthetic pathway from tyrosine to sanguinarine Biosynthesis of sanguinarine from tyrosine Enzymes for which cognate cDNAshave been isolated are shown in black Abbreviations TYDC tyrosinedopa decarboxylase NCS norcoclaurine synthase 6OMT (S)-norcoclaurine6-O-methyltransferase CNMT (S)-coclaurine N-methyltransferase NMCH (S)-N-methylcoclaurine 3rsquo-hydroxylase 4rsquoOMT (S)-3rsquo-hdroxy-N-methylcoclaurine 4rsquo-O-methyltransferase BBE berberine bridge enzyme CheSyn cheilanthifoline synthase StySyn stylopine synthase TNMTtetrahydroprotoberberine N-methyltransferase MSH methylstylopine hydroxylase P6 H protopine 6-hydroxylase DBOXdihydrobenzophenanthridine oxidase StySyn and CheSyn cDNAs were functionally characterized in plant species other than opium poppy

Page 3 of 17

approach the limited availability of sequence data moreseverely hampered the identification of known sanguinar-ine biosynthetic enzymes using LC-MSMS analysiswhich yielded only 6OMT [2]We report the integration of next-generation 454 pyr-

osequencing and advanced LC-MSMS analysis to estab-lish a deep survey of the transcriptome and proteome ofopium poppy cell cultures in response to treatment witha fungal elicitor The effectiveness of 454 pyrosequen-cing is compared to standard Sanger sequencing withrespect to depth of penetration into the transcriptomeof elicitor-treated opium poppy cells Use of theenhanced EST database for the identification of corre-sponding peptide sequences obtained by one-dimen-sional SDS-PAGE and LC-MSMS peptide analysisfacilitated the identification of more than 1000 peptidesand polypeptides Most of the known enzymes involvedin sanguinarine biosynthesis and many components ofprimary metabolic pathways that support alkaloid pro-duction are present in the protein database Several can-didate proteins and transcripts that potentially representnovel biosynthetic enzymes involved in the biosynthesisof sanguinarine and other BIAs are also represented

ResultsInduction of sanguinarine accumulation inopium poppy cell culturesThe content of reticuline protopine and sanguinarinewere determined at several time points after elicitortreatment of opium poppy cell cultures to facilitate acorrelation of the occurrence of specific transcript andproteins with the abundance of intermediate and end-product alkaloids (Additional File 2) Reticuline wasdetected at low levels over the entire time course butthe levels of this central pathway intermediate decreasedbeginning 50 h after elicitor treatment Protopine andsanguinarine were not detected at early time points afterelicitor treatment but both accumulated later in thetime course Sanguinarine levels began to increaserapidly between 10 and 50 h after the addition of elici-tor and reach a concentration of more than 5 μggfresh-weight of cells by the end of the 100-h timecourse At 100 h post-elicitation the level of sanguinar-ine was 40-fold greater than that of reticuline or proto-pine An inverse correlation between the levels ofreticuline and protopinesanguinarine was apparent overthe duration of the time course

Properties of the transcript database generatedby 454 pyrosequencingA total of 427 369 high-quality expressed sequence tags(ESTs) with an average read length of 462 bp were gen-erated by 454 pyrosequencing of half a plate using GSFLX Titanium system (Table 1) The assembly of

overlapping sequences yielded a total of 93723 uni-genes of which 37329 (398) were composed of twoor more contiguous ESTs (ie contigs) whereas 56394(602) consisted of only a single unique sequence (iesingletons) (Table 1) BLASTx analysis showed thatthese unigenes could be classified into two groups Thefirst group contained 73496 (784) unigenes that dis-played similarity to known genes (BLASTx expectationvalue of e lt 10-5) The second group consisted of 20227(216) unigenes that showed no similarity with anygene in the public UniProt database Unigenes in thelatter group could represent previously uncharacterizedor unknown genes sequences specific to opium poppyor gene fragments that are too short to annotateThe discovery rate of new unigenes reach saturation

after approximately 250000 pyrosequencing reads sug-gesting that near-complete representation of the elicitor-treated opium poppy cell culture transcriptome wasachieved (Additional File 3A) The majority of unigeneswere between 200 and 600 bp in length and although thepercentage of unigenes longer than 600 bp was consider-ably lower 1716 contigs showed greater than 90 cover-age of predicted open reading frames among knowngenes identified by BLASTx analysis (Additional File 3Band 3C) Errors in sequencing and the assembly of con-tigs could have resulted in the apparently low representa-tion of full-length transcripts The possibility that greateroverall sequence coverage was present in the database issupported by the frequent occurrence of several indepen-dent unigenes encoding the same gene product Forexample opium poppy TNMT was represented by 19unigenes (Figure 2 and Additional File 4) The mostabundant of these unigenes (ie Contig1) was assembledfrom 534 independent 454 pyrosequencing reads andalthough it covered the entire open reading frame encod-ing TNMT the predicted protein showed only 96amino acid identity compared with the publishedsequence (Figure 2) [20] Two other independent uni-genes (ie Contig2 and Contig3) displayed 100 aminoacid sequence identity with respect to the published

Table 1 Summary of the expressed sequence tagdatabases for elicitor-treated opium poppy cell culturesobtained using 454 GS-FLX Titanium pyrosequencing

Feature Number

Total number of EST sequences clustered 427 369

Average length of EST sequences (bp) 462

Number of contigs 37 329

Number of singletons 56 394

Total number of unigenes 93 723

Number of unigenes blasted with no hits 20 27

After removal of sub-standard sequence

Sum of contigs and singletons

BLASTx search of the UniProt Plants v148 database (e-value cutoff of 10-5)

Page 4 of 17

Figure 2 Clustering of 454 pyrosequencing data annotated as TNMT Various examples representing assembly of ESTs for TNMT annotatedunigenes found in the 454 database The upper bar corresponds to the translated TNMT protein (Accession number Q108P1_PAPSO) The lowerbar represents the unigene found in the 454 database and labelled with the contig number The white region reflects the TNMT open readingframe See Additional File 4 for a summary of unigenes shown in this figure

Page 5 of 17

sequence but did not show complete open reading framecoverage The multiple contigs likely represent differentTNMT isoforms of independent genes expressed in elici-tor-treated opium poppy cell cultures Alternatively thelarge number of unigenes might reflect sequencing andor assembly errorsAn anticipated feature of the EST database is the sub-stantially greater coverage of the 3rsquo-ends of several con-tigs resulting from (1) the proportionately largernumber of partial cDNAs in the library and (2) the rela-tive inefficiency of shearing short cDNAs prior to 454pyrosequencing (Figure 2) In most cases and as demon-strated for TNMT (Figure 2) relatively few unigeneswere assembled from the majority of sequence readscorresponding to the same gene product Taking thisinto account the gene density of opium poppy can beestimated by only dividing the number of contigs (ie37329) by the size of the opium poppy genome esti-mated at 3724 Mbp [36] which yields a gene density ofapproximately 10 genes per Mbp

Deep transcriptome analysis of elicitor-treated opiumpoppy cell culturesThe large number of reads generated by 454 pyrose-quencing allows a robust comparison of the relativeexpression of different genes Moreover the saturationof newly discovered unigenes after the analysis ofapproximately 250000 ESTs (Additional File 2A) sug-gests near-complete coverage of the elicitor-treatedopium poppy cell culture transcriptome However accu-rate identification of each EST is essential to facilitatethe quantification of all reads corresponding to selectedgenes The unigenes were mapped to UniProt version148 (minus genomic sequences from Vitis viniferawhich have not been annotated) using BLASTx analysiswith an expectation highly expressed value of e lt 10-5

included in the high-scoring segment pair The 50 mostabundant unigenes represented in the transcriptomeaccounted for approximately 9 of the transcriptome(Table 2) The single most abundant transcript wassampled 3165 times and annotated as a senescence-associated protein a putative cytochrome P450 monoox-ygenase from pea The remaining top-50 most highlyrepresented transcripts were sampled between 1579 and409 times and encode proteins involved in metabolismdefense signaling transport and cellular structure(Table 2) Transcripts encoding several enzymesinvolved in the biosynthesis of S-adenosylmethionine(SAM) such as SAM synthetase S-adenosylhomocytei-nase and methionine synthase were highly representedin the database Abundant transcripts encoding defense-response proteins encoded chitinase b-lactamase poly-phenol oxidase xyloglucanase inhibitor peroxidase andpathogenesis-related (PR) proteins Transcripts encoding

components of the proteosome and polyubiquitin alongwith several housekeeping proteins including an elonga-tion factor ubiquitin and ribosomal proteins were alsoabundant Most importantly six of the top-50 mostabundant transcripts encoded known enzymes involvedin sanguinarine biosynthesis NCS CNMT BBE StySynand two putative TNMT isoforms (Table 2)Local BLASTx analysis was performed to identify

ESTs encoding all known BIA biosynthetic enzymes(Figure 3) In the pathway from tyrosine to (S)-reticuline(Figure 1) sequence reads corresponding to knownenzymes were relatively abundant with CNMT showingthe highest transcript level and the P450-dependentenzyme NMCH displaying the lowest transcript abun-dance (Figure 3) Several unigenes showed less than 90amino acid sequence identity compared with 6OMT and4rsquoOMT suggesting that the corresponding transcriptsencode O-methyltransferases that potentially accept BIAsubstrates other than norcoclaurine and 3rsquo-hydroxy-N-methylcoclaurine respectively In the branch pathwayfrom (S)-reticuline to sanguinarine sequence readsencoding known enzymes were similarly abundant withthe P450-dependent enzymes CheSyn and StySyn show-ing the lowest transcript levels It should be noted thatthe transcript levels shown in Figure 3 reflect the totalof all putative isoforms (ie unigenes with gt 90 aminoacid identity compared with functionally verified genes)thus individual NCS CNMT BBE StySyn and TNMTunigenes were assembled from the largest number ofsequence reads (Table 2 Figure 2) although the totalnumber of ESTs corresponding to all putative isoformswas higher for certain other enzymes (Figure 3) Forexample no individual unigenes encoding TYDC werefound among the 50 most abundant transcripts (Table 2)although the total number of reads for all unigenesencoding TYDC was substantial suggesting the occur-rence of several different isoforms (Figure 3) Unigenesencoding CheSyn and StySyn were identified on the basisof their similarity to characterized cDNAs fromEschscholzia californica[1819] and their sequences havebeen deposited in GenBank accession numbersGU325749 and GU325750 respectivelyTranscripts encoding three out of six known enzymes

involved in the conversion of (S)-reticuline to morphinewere not detected in the database with ESTs corre-sponding to the other three enzymes represented atsubstantially lower levels than enzymes involved in san-guinarine biosynthesis (Figure 3 Additional File 1)Similarly transcripts encoding two other known BIAbiosynthetic enzymes from opium poppy were repre-sented at low levels (eg 7OMT N7OMT) or were notfound in the database (Figure 3 Additional File 1) Thefailure of elicitor-treated opium poppy cell cultures toexpress genes encoding SalSyn SalR and CODM

Page 6 of 17

Table 2 The fifty most abundant unigenes in the opium poppy 454 G S-F L X Titanium pyrosequencing database

Rank Numberof reads

Annotation Proteinscore

Plant species Accession number

1 3165 Senescence-associated protein 859 Pisum sativum Q9AVH2_PEA

2 1579 S-Adenosylmethionine synthetase 609 Nicotiana suaveolens Q069K3_9SOLA

3 1579 S-Adenosylmethionine synthetase 1846 Solanum tuberosum METK2_SOLTU

4 1323 Multiprotein bridging factor 600 Solanum tuberosum Q9LL86_SOLTU

5 1220 Heat shock protein 2973 Cucurbita maxima Q8GSN4_CUCMA

6 1176 Chitinase class IV 903 Nepenthes alata A9ZMK1_NEPAL

7 1154 Berberine bridge enzyme 2698 Papaver somniferum RETO_PAPSO

8 1120 60 S ribosomal protein L6 827 Mesembryanthemum crystallinum RL6_MESCR

9 1106 Elongation factor 1a 2227 Lilium longiflorum Q9SPA1_LILLO

10 1009 Beta lactamase 1388 Zea mays Q285M4_MAIZE

11 978 Heat shock protein 90 2846 Nicotiana tabacum Q14TB1_TOBAC

12 950 40 S ribosomal protein S9 862 Solanum demissum Q60CZ2_SOLDE

13 863 Methionine synthase 3422 Carica papaya A6YGE7_CARPA

14 855 Coclaurine N-methyltransferase 1351 Papaver somniferum Q7XB08_PAPSO

15 847 Polyphenol oxidase 1498 Malus domestica PPO_MALDO

16 844 Fructose-bisphosphate aldolase 1644 Solanum tuberosum Q2PYX3_SOLTU

17 841 Nodulin protein 457 Oryza sativa subsp japonica Q5VRN2_ORYSJ

18 807 Proteasome component protein 69 Medicago truncatula A2Q5C5_MEDTR

19 757 Nectarin IVxyloglucanase inhibitor 1484 Nicotiana langsdorffii times N sanderae Q3KU27_NICLS

20 735 Cellulose synthase 1669 Zea mays B6SW15_MAIZE

21 729 Luminal-binding protein 5 2728 Nicotiana tabacum BIP5_TOBAC

22 717 Elongation factor 1a 2231 Prunus persica B6V864_PRUPE

23 686 Uncharacterized protein 864 Arabidopsis thaliana Q9LZN8_ARATH

24 680 Peroxidase 1154 Medicago truncatula A4UN76_MEDTR

25 652 Adenosylhomocysteinase 2300 Medicago sativa SAHH_MEDSA

26 647 Pathogenesis-related protein 383 Solanum lycopersicum Q53U35_SOLLC

27 630 Norcoclaurine synthase 1 1114 Papaver somniferum Q4QTJ2_PAPSO

28 621 Sterol dehydrogenase 1133 Arabidopsis thaliana O22856_ARATH

29 572 ADP ribosylation factor 936 Daucus carota Q38JU3_DAUCA

31 534 Tetrahydroprotoberberine N-methyltransferase 1780 P apaver somniferum Q108P1_PAPSO

32 528 Polyubiquitin 740 Euphorbia esula Q9M5X0_EUPES

33 507 ABC transporter 1930 Oryza sativa subsp japonica Q84ZB2_ORYSJ

34 506 Polyphenol oxidase 1866 Annona cherimola A0A168_ANNCH

35 479 S-Adenosylmethionine synthetase 1928 Vitis vinifera METK2_VITVI

36 468 Lipid transfer protein 239 Oryza sativa subsp japonica Q6L4H1_ORYSJ

37 465 Glycoprotein 515 Daucus carota Q05929_DAUCA

38 463 b-D-glucosidase 2574 Gossypium hirsutum Q7XAS3_GOSHI

39 456 Cysteine proteinase 1690 Elaeis guineensis var tenera A6N8F8_ELAGV

40 449 Ripening-regulated protein 857 Oryza sativa subsp japonica Q6ZJI2_ORYSJ

41 448 Stylopine synthase 1996 Eschscholzia californica Q50LH3_ESCCA

43 435 Calreticulin 1738 Berberis stolonifera CALR_BERST

44 434 FAD-dependent oxidoreductase 1206 Arabidopsis thaliana O64743_ARATH

45 434 Xyloglucanase inhibitor 1485 Solanum tuberosum Q7XJE7_SOLTU

46 434 Uncharacterized protein 1245 Arabidopsis thaliana Q8VZ33_ARATH

47 422 Tetrahydroprotoberberine N- methyltransferase 1413 Papaver somniferum Q108P1_PAPSO

48 419 ATPase AAA-type 1486 Arabidopsis thaliana Q9FKM3_ARATH

49 415 Spindle disassembly protein 2459 Nicotiana tabacum Q1G0Z1_TOBAC

Refers to a measure of similarity between a previously characterized protein with the listed annotation and an amino acid sequence translated from the contigA high score indicates substantial amino acid identity between the two proteins

Page 7 of 17

explains the absence of morphine in dedifferentiatedcells It is notable that all unigenes encoding enzymesinvolved in sanguinarine biosynthesis displayed full-length open reading frames which further supports thedepth of transcriptome coverage that is possible using454 pyrosequencing

Deep proteome analysis of elicitor-treatedopium poppy cell culturesA total protein extract from elicitor-treated opiumpoppy cells was fractionated by one-dimensional SDS-PAGE and the gel was cut into 12 equal-sized fragments(Figure 4) Proteins in each gel slice were digested withtrypsin and subjected to LC-MSMS Peptide fragmentspectra were used to search both the public NCBI non-redundant green plant protein database and the opiumpoppy database created by 454 pyrosequencing Using astringent cut-off 288 peptides and polypeptides were

identified using the public protein database of which177 were represented by two or more peptides In con-trast 1004 peptides and polypeptides were identifiedusing the opium poppy-specific 454 pyrosequencingdatabase of which 571 were represented by two ormore peptides (Additional File 5) The species-specificityand depth of coverage offered by the 454 pyrosequen-cing database added substantial identification power tothe analysisAnnotated unigenes in the 454 pyrosequencing data-

base and proteins identified by LC-MSMS peptide ana-lysis were classified into functional categories based ontheir putative roles in cellular processes (Figure 5) Puta-tive GO annotations could be assigned to 72 of thepeptides and polypeptides with corresponding ESTswhereas the remaining 28 belong to unknown uncate-gorized and no hit categories (Figure 5B) The mostabundant category (eg metabolism) represented 23 of

Figure 3 Number of 454 pyrosequence reads representing gene transcripts corresponding to known benzylisoquinoline alkaloidbiosynthetic enzymes The cDNA library used for 454 pyrosequencing was prepared from opium poppy cell cultures treated with a fungalelicitor for 10 h Sequence counts include unigenes encoding predicted proteins with gt 90 amino acid sequence identity to known opiumpoppy enzymes except for CheSyn and StySyn which were compared with known enzymes from Eschscholzia californica Black bars representunigenes encoding enzymes involved in the conversion of precursor tyrosine to the central intermediate (S)-reticuline Red bars refer tounigenes encoding enzymes involved in the formation of sanguinarine blue bars represent unigenes encoding enzymes involved in thebiosynthesis of morphine and green bars correspond to other enzymes with a role in benzylisoquinoline alkaloid metabolism Abbreviations areas indicated in Figure 1 and Additional File 1

Page 8 of 17

all identified proteins and included enzymes involved inprimary metabolism such as SAM synthetase methio-nine synthase and enzymes involved in central metabolicpathways such as glycolysis and the tricarboxylic acidcycle (Additional File 5) A number of peptides andpolypeptides (14) were associated with protein synth-esis and modification suggesting a substantial role forthese processes in elicitor-treated opium poppy cell

cultures (Figure 5B) Chaperones and heat shock pro-teins (5) defense proteins (3) and proteins involvedin protein turnover (7) were also well representedAlmost all enzymes involved in BIA biosynthesis thatwere represented in the 454 pyrosequencing databasewere identified by LC-MSMS peptide analysis (Addi-tional File 5) including NCS 6OMT CNMT NMCH4rsquoOMT BBE StySyn and TNMT The morphine bio-synthetic enzyme COR1 was also identified Interest-ingly TYDC and CheSyn were not identified despitesimilar transcript levels compared with other solubleand P450-dependent enzymes respectively (Figure 3)

Integration of transcriptome and proteome databasesA broad survey of cellular metabolism involved in theconversion of sucrose to sanguinarine resulted in theidentification of transcripts or proteins corresponding toa substantial number of metabolic enzymes (Figure 6)With the exception of 3-dehydroquinate dehydratase allenzymes required for the formation of tyrosine andSAM were represented in the 454 pyrosequencing data-base whereas 20 out of 36 enzymes were found in theLC-MSMS-generated protein database (Additional File5) Many of these enzymes were also among the top-50most abundant unigenes (Table 2) Notably four inde-pendent unigenes in the top 50 encoded enzymesinvolved in the metabolism of SAM the methyl donorfor the various O-and N-methyltransferases in BIAbiosynthesisThe remaining sanguinarine biosynthetic enzymes for

which cognate cDNAs have not been isolated catalyzethree of the four steps involved in the formation of dopa-mine and 4-HPAA and the final three conversions from(S)-cis-N-methylstylopine to sanguinarine (Figure 6)Some of these enzymes likely belong to known proteinfamilies including the cytochromes P450 MSH and P6H[2122] and the oxidoreductase DBOX [2324] Candidateproteins with substantial identity to oxidoreductase andother enzyme categories potentially involved in BIAmetabolism were found in the LC-MSMS-generatedpeptide and polypeptide database (Additional File 6)

DiscussionIntegration of 454 pyrosequencing and LC-MSMS pep-tide analysis were used to survey the transcriptome andproteome respectively of elicitor-treated opium poppycell cultures The depth of each database provides newinsights into the regulation of BIA metabolism andplant defense responses establishes valuable resourcesfor the discovery of new alkaloid biosynthetic genesand allows an assessment of next-generation-omicstechnologies as tools to study natural product biosynth-esis in plants that currently lack genome sequenceresources [37]

Figure 4 Fractionation of the gel containing proteins separatedby SDS-PAGE prior to LC-MSMS Coomassie stained gel of a totalprotein extract (10 μg) from opium poppy cell cultures treated with afungal elicitor for 50 h Each of the 12 gel slices was treated withtrypsin and independently analyzed by LC-MSMS peptide analysis

Page 9 of 17

Treatment of cell cultures with the elicitor for 10 and50 h maximized the accumulation of BIA biosyntheticgene transcripts and cognate enzymes respectivelyUsing microarray and northern blot analyses the maxi-mum induction of BIA biosynthetic genes was pre-viously shown to occur 10 h after the elicitor treatmentof opium poppy cell cultures [1] Corresponding westernblot analysis showed that BIA biosynthetic enzyme levelswere highest 50 h after elicitor treatment [238] The eli-citor-induced accumulation profiles of protopine andsanguinarine (Additional File 2) were in agreement withthe temporal induction of BIA products and pathwayintermediates determined using FT-ICR-MS [1]The EST database generated by 454 pyrosequencing

was compared with that established by random sequen-cing of clones from an elicitor-treated opium poppy cellculture cDNA library using dideoxy chain-termination(Sanger) technology [1] Although the average readlength produced by 454 pyrosequencing was less thanthat of Sanger-based sequencing (462 bp and 653 bprespectively) substantially more ESTs were generated(427369 and 10224 respectively) which led to theacquisition of a large number of additional unigenes(93723 and 7225 respectively Table 1) The mostabundant transcripts in the 454 pyrosequencingdatabase encoded enzymes and proteins involved inmetabolism defense signaling transport and cellularstructure (Table 2) Transcripts encoding biosyntheticenzymes involved in the regeneration of (S)-adenosyl-methionine (ie SAM synthetase S-adenosylhomocytei-nase and methionine synthase) were among the mostabundant in the database which is in agreement with

their widespread occurrence in an EST database gener-ated by Sanger sequencing [1]An EST database was established using 454 pyrose-

quencing to investigate the flavonoid pathway in theChinese medicinal plant Epimedium sagittatum[39] Atotal of 217380 reads with an average length of 225 bpwere assembled into 76459 unigenes consisting of17231 contigs and 59228 singletons Similarly 454 pyr-osequencing of cDNA obtained from the glandular tri-chomes of Artemisia annua which produces theantimalarial sesquiterpene artemisinin yielded 406044reads with and average length of 210 bp asembling into42678 contigs and 147699 singletons [40] Real time-PCR confirmed the expression of all known terpenoidbiosynthetic genes and revealed several novel gene tran-scripts in the 454 pyrosequencing database includingputative sesquiterpene synthase homologues Our opiumpoppy cell culture library is comparable or superior tothese reported databases in terms of sequence coverage(Table 1) The large number singletons in the opiumpoppy cell culture database was also reported in theother systems and might result from assembly errorsdue or from the occurrence of low abundance tran-scripts As shown for TNMT (Figure 2 Additional file4) several unigenes were found for most BIA biosyn-thetic enzymes (ie using the criterion of gt 90 aminoacid identity) which could reflect the occurrence ofmultiple gene family members Separate unigenesencoding proteins with 100 amino acid identity couldalso result from the improper assembly of contigs (Fig-ure 2 Additional file 4) Clearly the reported 93723unigenes is a substantial overestimate of the actual

Figure 5 Functional categories of (A) trans cripts represented in the 454 pyrosequence database and (B) peptides identified byLC-MSMS (A) GO annotations were assigned for 23753 contigs and singletons out of a total of 93723 unigenes in the opium poppy 454pyrosequencing database (B) GO annotations were assigned for a total of 1004 putative opium poppy proteins identified by LC-MSMS peptideanalysis

Page 10 of 17

Figure 6 Metabolic networks from sucrose to sanguinarine and morphine Gene transcripts corresponding to enzymes shown in black orred were identified in the 454 pyrosequencing database whereas those written in grey were not Enzymes written in red were found amongproteins identified by LC-MSMS peptide analysis Cognate cDNAs have not been isolated for enzymes shown in blue

Page 11 of 17

number of transcripts expressed in those cells The lackof assembly potentially caused by sequencing errorscould account for the high number of orphaned ESTsalthough some could represent low-expression levelgenesOur 454 pyrosequencing reads showed a substantial

bias for the 3rsquo-end of gene transcripts most likely due tothe priming of the first-strand cDNA synthesis usingoligo-dT (Figure 2) Moreover the nebulization of rela-tively short partial cDNAs process was also inefficient(Figure 2) Comparison of 454 pyrosequencing usingCalifornia poppy (Eschscholzia californica) cDNAlibraries prepared using oligo-dT or random-primersconfirmed that the 3rsquo-end bias resulted from primingfirst-strand cDNA synthesis with oligo-dT [41] Assem-bly of both the oligo-dT and random-primed ESTsgenerated from two full plate of GS-FLX 454 pyrose-quencing resulted in 120585 unigenes with an averagelength of 157 bp which assembled into 30603 contigsand 89892 singletons [41] Many of the partial ESTs arealso truncated at common points at their 5rsquo-end (Figure2) Genomic 454 pyrosequencing of Escherichia colishowed that 80 of single nucleotide polymorphismswere falsely linked to reads having the same startingpoint [42] The high number of replicate sequences sug-gested that the phenomenon was not random and wasassociated with the emulsion PCR step and not withnebulization Similar artefacts were also reported in ametagenomics study on several species which foundsystematic errors in genomes sequenced by 454 pyrose-quencing technologies (ie GS20 or GS-FLX) [43] Mul-tiple reads from a single template were suggested tooccur when amplified DNA attaches to empty beadsduring emulsion PCR Although duplicate sequences areoccasionally removed prior to assembly clearly some arestill present in our database (Figure 2)Since the number of 454 pyrosequencing reads in each

contig is directly proportional to the abundance of spe-cific cDNAs in the library quantification of the dataprovides an accurate measure of the relative expressionlevel of selected transcripts Comparisons of 454 pyrose-quencing and hybridization-based gene expression ana-lyses (ie DNA microarray or northern blot) have shownreproducible correlations [323544] The induction of allknown genes encoding BIA biosynthetic enzymesinvolved in the formation of sanguinarine has beenshown previously [1] However 454 pyrosequencingprovides the ability to quantify the relative abundance ofdifferent gene transcripts (Figure 3) In the conversionof tyrosine to sanguinarine (Figure 1)454 pyrosequen-cing reads encoding CNMT and TNMT were mostabundant and were represented among the top-50 mosthighly expressed unigenes (Table 2) In contrast readscorresponding to the P450-dependent enzymes NMCH

CheSyn and StySyn were the least abundant (Figure 3)Overall elicitor-treated opium poppy cell cultures showa strong commitment to sanguinarine biosynthesis asdemonstrated by the cumulative abundance of 454 pyro-sequencing reads corresponding to genes involved insanguinarine biosynthesis in agreement with previouslypublished reports comparing control and elicitor-treatedopium poppy cell cultures [1] The total number ofreads (ie 8505) with greater than 90 amino acid iden-tity to all known BIA biosynthetic enzymes representapproximately 20 of the transcriptome Of these625 encoded enzymes leading to (S)-reticuline 350corresponded to enzymes involved in the conversion of(S)-reticuline to sanguinarine 24 represented enzymesin the morphinan alkaloid branch pathway and 01were relevant to the formation of other BIAs such aslaudanine (Figure 1 and Additional File 1) The absenceof any 454 pyrosequncing reads corresponding to three(ie SalSyn SalR and CODM) out of the six enzymesleading from (S)-salutaridine to morphine explains thelack of codeine or morphine in elicitor-treated opiumpoppy cell cultures (Figure 3) Interestingly genesencoding other known biosynthetic enzymes in the mor-phinan alkaloid branch pathway (ie SalAT T6ODMand COR) were expressed albeit at low levels comparedwith those involved in the formation of (S)-reticuline orsanguinarine (Figure 3) DNA microarray and northernblot analyses have previously shown that SalAT mRNAlevels were induced in opium poppy cell cultures inresponse to elicitor treatment whereas COR transcriptsare constitutive [1] Moreover the recently identifiedT6ODM cDNA was represented in a Sanger-basedopium poppy cell culture EST database whereas theCODM was only found in an opium poppy stem ESTdatabase [29] The differential expression of genesencoding morphinan biosynthetic enzymes in elicitor-treated opium poppy cells could reflect a requirementfor specific cellular or developmental conditions notpresent in dedifferentiated cell cultures [3845]Two-dimensional (2D) SDS-PAGE is the most com-

monly used method to separate proteins for plant pro-teomics applications However penetration into theproteome is limited owing mostly to low abundanceproteins that are difficult isolate from the 2D gel Ourprevious application of LC-MSMS to analyze the pro-teome of elicitor-treated opium poppy cell culturesresulted in the identification of 219 proteins based onpeptide fragment fingerprint searches using a combina-tion of public and opium poppy EST (ie Sangersequenced) databases [2] A total of 340 spots were iso-lated by 2D SDS-PAGE and 6OMT was the only BIAbiosynthetic enzyme represented among the identifiedproteins [2] Recently the mass spectral analysis of pro-teins partially separated by one-dimensional SDS-PAGE

Page 12 of 17

was used to assemble an impressive proteome map forArabidopsis thaliana for different organs developmentalstages and undifferentiated cultured cells [46] A totalof 86456 peptide matches yielded 13029 identified pro-teins Proteomics in many plants is typically forced torely on cross-species identification owing to the lack ofgenome or transcriptome sequence information whichleads to relatively low numbers of unambiguously identi-fied proteins and the potential for false-positive identifi-cation For example proteome analysis of banana whichis distantly related to most plant species with substantialDNA sequence data was used to compare various pro-tein extraction methods and one-dimensional versus 2DSDS-PAGE techniques within the context of cross-spe-cies matching of peptide mass spectra One-dimensionalSDS-PAGE on proteins extracted in chloroformmetha-nol (54) followed by LC-MSMS facilitated the ide-ntification of the most proteins including severalhydrophobic proteins that were underrepresented when2D SDS-PAGE was used [47]Our one-dimensional SDS-PAGE LC-MSMS proteo-

mics approach coupled with the unambiguous identifica-tion of peptide spectra using an extensive opium poppyEST database generated by 454 pyrosequencing led tothe identification of five-fold more peptides and poly-peptides (1004 in total) than our previous effort [2]Most identified proteins are involved in metabolismdefense signalling transport and cellular structure(Figure 5 Additional file 5) The majority of identifiedtranscripts (Figure 5A) and proteins (Figure 5B) areinvolved in metabolism and include enzymes of primarymetabolic pathways such as glycolysis and the tricar-boxylic acid cycle and intermediary metabolic enzymessuch as SAM synthetase and methionine synthase(Additional File 5 Figure 6) Almost all enzymesinvolved in BIA biosynthesis were identified by LC-MSMS peptide analysis (Additional File 5 Figure 6) Sincethe likelihood of identifying a certain protein is propor-tional to the abundance of specific peptides the absenceof TYDC in the identified protein list (Additional File 5)could be due to the large number of isoforms encodedby the approximately 15-member TYDC gene family [7]Alternatively the absence of TYDC in the identifiedprotein list (Additional File 5) could result from issuesrelated to solubility extraction efficiency digestion orthe nature of peptide sequencesWith one exception (ie 3-dehydroquinate dehydra-

tase) transcripts encoding all enzymes required for theformation of tyrosine and SAM were represented in theEST database (Figure 6) and several were also amongthe 50 most abundant unigenes (Table 2) Similarlymany of these enzymes were also found in the proteindatabase (Additional File 5 Figure 6) Altogether theseresults further demonstrate the metabolic commitment

of elicitor-treated opium poppy cell cultures to the over-all biosynthesis of sanguinarineBased on the abundance of characterized mRNAs and

enzymes involved in BIA metabolism the remainingcomponents of sanguinarine biosynthesis for which cog-nate cDNAs have not been isolated should be repre-sented in the transcript and protein databases at similarlevels The penultimate and third-to-last enzymes insanguinarine biosynthesis which catalyze the conversionof (S)-cis-N-methylstylopine to dehydrosanguinarine(Figure 1) are cytochromes P450 [2122] Although thetranscript database contains numerous sequences thatannotated as cytochromes P450 a more focused numberof candidate cytochromes P450 were represented in theprotein database (Additional File 6) Cytochromes P450represented in the 454 pyrosequencing database at levelssimilar to transcripts encoding NMCH CheSyn and Sty-Syn coupled with the corresponding representation ofthe cognate enzymes in the LC-MSMS protein databaseprovides a basis for the selection of genes putativelyinvolved in BIA biosynthesis

ConclusionsThe integration of state-of-the-art 454 GS-FLX Tita-nium pyrosequencing and LC-MSMS-based proteinprofiling technologies is an effective strategy to establishdeep transcriptome and proteome databases for theinvestigation of natural product metabolism in non-model plant systems A near-complete transcriptomefrom relatively homogenous elicitor-treated opiumpoppy cell cultures could be achieved by the assemblyof fewer than one-half million 454 pyrosequencing readsusing the GS-FLX Titanium technology All known san-guinarine biosynthetic gene transcripts were representedin the database along with a multitude of transcriptsencoding primary and intermediary metabolic enzymesproviding precursors and co-substrates in support ofalkaloid production The absence of several transcriptsencoding morphinan alkaloid biosynthetic enzymesreveals the transcriptional basis for the lack of morphineproduction in dedifferentiated opium poppy cell cul-tures The establishment of tandem mass spectra derivedfrom predicted peptides represented in the deep tran-script database facilitated the empirical identification ofa large number of corresponding proteins fractionatedby one-dimensional SDS-PAGE The depth of proteomecoverage was dramatically greater than that achievedusing 2D SDS-PAGE to isolate individual proteins baseon the representation of most sanguinarine biosyntheticenzymes and a substantial number of metabolic enzymesrelevant to BIA metabolism The integration of a near-complete and species-specific transcript database wascrucial to the successful identification of multiple pep-tides Profiling of the more abundant proteins in

Page 13 of 17

elicitor-treated opium poppy cell cultures also revealed anumber of uncharacterized enzymes that potentially cat-alyze steps in sanguinarine biosynthesis

MethodsCell culture and elicitor treatmentCell suspension cultures of opium poppy (Papaversomniferum) cv Marianne cell line 2009 [48] weregrown at 23degC on a gyrotary shaker at 125 rpm inGamborg 1B5C medium [49] containing B5 salts andvitamins 20 gL sucrose 1 gL casein hydrolysate and1 mgL 24-dichlorophenoxyacetic acid The cell cul-tures were sub-cultured biweekly using a 13 dilutionof inoculum to fresh medium Elicitor treatment wasperformed by adding 1 ml of Botrytis cinerea homoge-nate to 50 ml of cultured cells in rapid growth phase(2 days after sub-culture) and grown for an additional10 h or 50 h for the isolation of RNA or proteinrespectively Cells were collected by vacuum filtrationand stored at -80degC The fungal elicitor was preparedby inoculating 50 ml of 1B5C medium lacking 24-dichlorophenoxyacetic acid with 1 cm3 of B cinereamycelium The fungal culture was grown at 120 rpmon a gyratory shaker at 22degC in the dark for 1 weekFungal mycelia and medium were homogenized in ablender (Waring Instruments Torrington CT) auto-claved at 121degC for 20 min and stored at -20degC

R NA extraction cDNA library construction454 pyrosequencing and data processingCell cultures treated with the fungal elicitor for 10 hwere ground to a fine powder under liquid nitrogen andtotal RNA was isolated according to [50] Poly(A)+ RNAwas extracted by two rounds of Dynabeads oligo(dT)-based purification according to the manufacturerrsquosinstruction (Invitrogen Carlsbad CA) A double-stranded cDNA library was prepared using a protocoloptimized for 454 pyrosequencing developed by theJoint Genome Institute httpwwwjgidoegov[51] ThecDNA library was randomly sheared and sequencedusing Titanium FLX series reagents on a GenomeSequencer FLX instrument (454 Life Sciences BranfordCT) A total of 427369 high-quality expressed sequencetags (ESTs) with an average read length of 462 bp wereobtained after processing the raw data to eliminate low-quality sequences and poly(A) tails Pre-processing ofthe data included the removal of reads with BLASTnexpectation values of e lt 10-20 with respect to a plantrepeat sequence database [52] the trimming of poly(AT)tails the removal of low-complexity sequences using themdust filtering program httpwwwtigrorgtdbtgisoft-ware and the elimination of reads shorter than 40 bpClustering was done using the TGI Clustering Toolssoftware httpcompbiodfciharvardedutgisoftware

Unigenes were annotated by BLASTx analysis against theUniprot Plants 92 public database

Protein extraction and SDS-PAGEOpium poppy cells (1 g) treated with the fungal elicitorfor 50 h were ground to a fine powder under liquidnitrogen and extracted in 05 M Tris-HCl pH 7550 mM EDTA 1 (wv) SDS and 2 (wv) 2-mercap-toethanol The extract was centrifuged at 15000 g andthe supernatant extracted with an equal volume of phe-nol Subsequently the emulsion was centrifuged at15000 g to separate the phases The aqueous phase wasdiscarded and an equal volume of extraction buffer wasvigorously mixed with the phenol phase The emulsionwas centrifuged and the phenol phase was recoveredFive volumes of methanol containing 01 M ammoniumacetate and 0068 (vv) 2-mercaptoethanol were addedand the mixture incubated overnight at -20 degC Precipi-tated proteins were collected by centrifugation at 15000g and washed twice with the methanol solution Thepellet was dried and dissolved in rehydration buffer (7M urea 2 M thiourea 56 mM dithiothreitol and 25(vv) 3-[(3-cholamidopropyl)dimethylammonio]-1-pro-pane-sulfonic acid (CHAPS)) Protein concentration wasdetermined using the RC DC protein assay (BioRadHercules CA) Ten micrograms of total proteins wereseparated by SDS-PAGE To evaluate the quality of theelectrophoretic separation the gel was stained withCoomassie Brilliant Blue The lane on the gel containingprotein was cut into 12 equal segments (Figure 2) Pro-teins in each gel segment were digested with trypsinprior to LC-MSMS

Mass spectrometry and spectrum data analysisTryptic protein digests were analyzed using an Agilent1100 LC-Ion-trap-XCT-Ultrasystem (Agilent Technolo-gies Santa Clara CA) fitted with an integrated fluidiccartridge for peptide capture separation and nanospray-ing (HPLC-Chip technology) as described previously [2]Injected protein samples were trapped and desalted on apre-column channel (40-nl volume Zorbax 300 SC-C18)for 5 min with 02 (vv) formic acid delivered by anauxiliary pump at 4 μlmin The peptides were thenreverse-eluted from the trapping column and separatedon the analytical channel (43-mm channel length Zor-bax 300 SC-C18) at 03 μlmin Peptides were elutedusing a 5-70 (vv) acetonitrile gradient in 02 (vv)formic acid over 10 min MSMS spectra were collectedby data-dependent acquisition with parent ion scans of8100 Ths over mz 400-2000 and MSMS scans at thesame rate over mz 100-2200 Peak-list data wereextracted from these files by the DataAnalysis softwarefor the 6300 series ion trap v34 (build 175) Mascotv21 (Matrix Science Boston MA) was used to search

Page 14 of 17

the MSMS data using the following parameters 16 Daprecursor ion mass tolerance 08 Da fragment ion masstolerance 1 potential missed cleavage carbamidomethylmodification of cysteine and variable oxidation ofmethionine Peptide sequence data was used to searchthe Viridiplantae (green plants) database (containing468052 sequences) in NCBI httpwwwncbinlmnihgov The peptide sequence data was then used to querythe elicitor-treated opium poppy cell culture 454 ESTdatabase (containing 427369 sequences) in all potentialopen reading frames using Mascot v21 Results wereindexed with the aid of a prior-clustering and annota-tion exercise In all cases human and Botrytis cinereaproteins were included in the searches to avoid contami-nant-based erroneous assignment of the dataProtein hits were scored based on the quality and abun-

dance of the underlying peptide MSMS data and theirscores A cut-off score (p lt 0012) of 56 with a false dis-covery rate of 1 was used for all peptides identifiedthrough matches in the public databases and a cut-offscore (p lt 005) of 46 with a false discovery rate of 113was used for all peptides identified through matches in the454 EST database The resulting MSMS spectra weremanually assessed for consistency with the proposedsequences and distance from the next highest scoring pep-tide(s) The protein names associated with each hit weredetermined by selecting the highest scoring entry and themost common name representing the dataset

Metabolite extraction and HPLC analysisFrozen cell cultures (1 g) were ground to a fine powderunder liquid nitrogen and extracted for 2 h in 100 (vv)methanol at room temperature The extracts were centri-fuged for 10 min to pellet debris and the supernatantswere reduced to dryness under reduced pressure Pelletswere resuspended in 100 μl 100 (vv) methanol Tenmicroliters of each extract was diluted in 100 μl of 98(vv) H2O 2 (vv) acetonitrile 004 (vv) H3PO4 andanalyzed using a System Gold HPLC and photodiodearray detector (Beckman-Coulter Mississauga Canada)All separations were performed at a flow rate of 15mlmin on a LiChrospher RP-Select B 5μ column 150 times46 mm (Alltech Illinois USA) Separation was achievedusing a gradient of solvent A [98 (vv) H2O 2 (vv)acetonitrile 004 (vv) H3PO4] and solvent B [98 (vv)acetonitrile 2 (vv) H2O 004 (vv) H3PO4] Chroma-tography was initiated and maintained for 5 min using90 solvent A Subsequently the gradient was ramped to35 solvent B over 40 min and then to 100 solvent Bover 5 min Peaks corresponding to reticuline protopineand sanguinarine were monitored at 210 nm and identi-fied on the basis of their retention times and UV spectracompared with authentic standards Dextromethorphan

was used as an internal standard for the quantificationof data

Accession numbersThe sequences described in this paper have beensubmitted to GenBank under the accession numbersGU325749 and GU325750

Additional material

Additional file 1 Biosynthetic pathways leading to morphine (A)laudanine (B) and norlaudanine (C) Enzymes for which cognatecDNAs have been isolated are shown in black Abbreviations DRS 12-dehydroreticuline synthase DRS 12-dehydroreticuline reductase SalSynsalutaridine synthase SalR salutaridine reductase SalAT salutaridinol 7-O-acetyltransferase THS thebaine synthase T6ODM thebaine 6-O-demethylase COR1 codeinone reductase 1 CODM codeine O-demethylase 7OMT (RS)-reticuline 7-O-methyltransferase N7OMT (RS)-norreticuline 7-O-methyltransferase

Additional file 2 Alkaloid content of opium poppy cells afterelicitor treatment Reticuline (blue) protopine (yellow) and sanguinarine(red) levels in opium poppy cell cultures at various times after elicitortreatment

Additional file 3 Summary of characteristics for the 454pyrosequencing database (A) Number of new unigenes discovered per10000 sequences (B) Frequency distribution of unigene length aftersequence assembly (C) Frequency distribution of the percentage of full-length open reading frame coverage among unigenes with gt 50 orhigher amino acid identity

Additional file 4 Unigenes that annotate as TNMT in the 454pyrosequencing database

Additional file 5 List of peptides and polypeptides identified byLC-MSMS analysis Color coding known enzymes involved insanguinarine biosynthesis (yellow) enzymes involved in the primarymetabolism relevant to sanguinarine biosynthesis (green) candidateenzymes potentially involved in benzylisoquinoline alkaloid metabolism

Additional file 6 Candidate proteins identified by LC-MSMS andpotentially involved in benzylisoquinoline alkaloid metabolism inopium poppy cell cultures

AcknowledgementsThis work was funded by a Natural Sciences and Engineering ResearchCouncil of Canada Strategic Project Grant to PJF DCS holds the CanadaResearch Chair in Chemical Biology PJF holds the Canada Research Chair inPlant Metabolic Processes Biotechnology

Author details1Department of Biological Sciences University of Calgary Calgary AlbertaT2N 1N4 Canada 2Department of Biochemistry and Molecular BiologyUniversity of Calgary Calgary Alberta T2N 4N1 Canada 3National ResearchCouncil-Plant Biotechnology Institute Saskatoon Saskatchewan S7N 0W9Canada

Authorsrsquo contributionsIDP carried out all experimental work with the exception of thebioinformatics and proteomics DC and JN performed the bioinformatics onthe 454 pyrosequencing data MFK and DCS conducted the LC-MSMSpeptide analysis IDP and PJF designed the experiments IDP wrote themanuscript and PJF was its primary editor All authors read and approvedthe final manuscript

Received 2 June 2010 Accepted 18 November 2010Published 18 November 2010

Page 15 of 17

References1 Zulak KG Cornish A Daskalchuk TE Deyholos MK Goodenowe DB

Gordon PP Klassen D Pelcher LE Sensen CW Facchini PJ Gene transcriptand metabolite profiling of elicitor-induced opium poppy cell culturesreveals the coordinate regulation of primary and secondary metabolismPlanta 2007 2251085-1106

2 Zulak KG Khan MF Alcantara J Schriemer DC Facchini PJ Plant defenseresponses in opium poppy cell cultures revealed by liquidchromatography tandem mass spectrometry proteomics Molecular ampCellular Proteomics 2009 886-98

3 Zulak KG Weljie AM Vogel HJ Facchini PJ Quantitative 1H NMRmetabolomics reveals extensive metabolomic reprogramming ofprimary and secondary metabolism in elicitor-treated opium poppy cellcultures BMC Plant Biology 2008 85-21

4 Samanani N Facchini PJ Purification and characterization ofnorcoclaurine synthase The first committed enzyme inbenzylisoquinoline alkaloid biosynthesis in plants J Biol Chem 200227733878-33883

5 Samanani N Liscombe DK Facchini PJ Molecular cloning andcharacterization of norcoclaurine synthase an enzyme catalyzing thefirst committed step in benzylisoquinoline alkaloid biosynthesis Plant J2004 40302-313

6 Liscombe DK MacLeod BP Loukanina N Nandi OI Facchini PJ Evidencefor the monophyletic evolution of benzylisoquinoline alkaloidbiosynthesis in angiosperms Phytochemistry 2005 662501-2520

7 Facchini PJ DeLuca V Differential and tissue-specific expression of agene family for tyrosinedopa decarboxylase in opium poppy J BiolChem 1994 26926684-26690

8 Facchini PJ Park S-U Developmental and inducible accumulation of genetranscripts involved in alkaloid biosynthesis in opium poppyPhytochemistry 2003 64177-186

9 9 Ounaroon A Decker G Schmidt J Lottspeich F Kutchan TM (RS)-Reticuline 7-O-methyltransferase and (RS)-norcoclaurine 6-O-methyltransferase of Papaver somniferum-cDNA cloning andcharacterization of methyl transfer enzymes of alkaloid biosynthesis inopium poppy Plant J 2003 36808-819

10 Morishige T Choi K-B Sato F In vivo bioconversion oftetrahydroisoquinoline by recombinant coclaurine N-methyltransferaseBiosci Biotechnol Biochem 2004 68939-941

11 Ziegler J Diacuteaz-Chaacutevez ML Kramell R Ammer C Kutchan TM Comparativemacroarray analysis of morphine containing Papaver somniferum andeight morphine free Papaver species identifies an O-methyltransferaseinvolved in benzylisoquinoline biosynthesis Planta 2005 222458-471

12 Pauli HH Kutchan TM Molecular cloning and functional heterologousexpression of two alleles encoding (S)-N-methylcoclaurine 3rsquo-hydroxylase(CYP80b1) a new methyl jasmonate-inducible cytochrome P-450-dependent monooxygenase of benzylisoquinoline alkaloid biosynthesisPlant J 1998 13793-801

13 Huang FC Kutchan TM Distribution of morphinan and benzo[c]phenanthridine alkaloid gene transcript accumulation in Papaversomniferum Phytochemistry 2000 53555-564

14 Morishige T Tsujita T Yamada Y Sato F Molecular characterization of theS-adenosyl-L-methionine 3rsquo-hydroxy-N-methylcoclaurine 4rsquo-O-methyltransferase involved in isoquinoline alkaloid biosynthesis in Coptisjaponica J Biol Chem 2000 27523398-23405

15 Dittrich H Kutchan TM Molecular cloning expression and induction ofberberine bridge enzyme an enzyme essential to the formation ofbenzophenanthridine alkaloids in the response of plants to pathogenicattack Proc Natl Acad Sci USA 1991 889969-9973

16 Facchini PJ Penzes C Johnson AG Bull D Molecular characterization ofberberine bridge enzyme genes from opium poppy Plant Physiol 19961121669-1677

17 Samanani N Park SU Facchini PJ Cell type-specific localization oftranscripts encoding nine consecutive enzymes involved inprotoberberine alkaloid biosynthesis Plant Cell 2005 17915-926

18 Ikezawa N Iwasa K Sato F Molecular cloning and characterization ofmethylenedioxy bridge-forming enzymes involved in stylopinebiosynthesis in Eschscholzia californica FEBS J 2007 2741019-1035

19 Ikezawa N Iwasa K Sato F CYP719A subfamily of cytochrome P450oxygenases and isoquinoline alkaloid biosynthesis in E schscholziacalifornica Plant Cell Rep 2009 28123-133

20 Liscombe DK Facchini PJ Molecular cloning and characterization oftetrahydroprotoberberine cis-N-methyltransferase an enzyme involvedin alkaloid biosynthesis in opium poppy J Biol Chem 200728214741-14751

21 Rueffer M Zenk MH Enzymatic formation of protopines by a microsomalcytochrome P-450 system of Corydalis vaginans Tetrahedron Lett 1987285307-5310

22 Tanahashi T Zenk MH Elicitor induction and characterization ofmicrosomal protopine-6-hydroxylase the central enzyme inbenzophenanthridine alkaloid biosynthesis Phytochemistry 1990291113-1122

23 Ignatov A Clark WG Cline SD Psenak M Krueger J Coscia CJ Elicitation ofdihydrobenzophenanthridine oxidase in Sanguinaria canadensis cellcultures Phytochemistry 1996 431141-1144

24 Schumacher H-M Zenk MH Partial purification and characterization ofdihydrobenzophenanthridine oxidase from Eschscholtzia californica cellsuspension cultures Plant Cell Rep 1988 743-46

25 Gesell A Rolf M Ziegler J Diacuteaz Chaacutevez ML Huang F-C Kutchan TMCYP719B1 is salutaridine synthase the C-C phenol-coupling enzyme ofmorphine biosynthesis in opium poppy J Biol Chem 200928424432-24442

26 Ziegler J Voigtlander S Schmidt J Kramell R Miersch O Ammer C Gesell AKutchan TM Comparative transcript and alkaloid profiling in Papaverspecies identifies a short chain dehydrogenasereductase involved inmorphine biosynthesis Plant J 2006 48177-192

27 Grothe T Lenz R Kutchan TM Molecular characterization of thesalutaridinol 7-O-acetyltransferase involved in morphine biosynthesis inopium poppy Papaver somniferum J Biol Chem 2001 27630717-30723

28 Unterlinner B Lenz R Kutchan TM Molecular cloning and functionalexpression of codeinone reductase the penultimate enzyme inmorphine biosynthesis in the opium poppy Papaver somniferum Plant J1999 18465-475

29 Hagel JM Facchini PJ (2010) Novel O-demethylases of morphinebiosynthesis in opium poppy Nature Chem Biol 2010 6273-275

30 Pienkny S Brandt W Schmidt J Ziegler J Functional characterization of anovel benzylisoquinoline-O-methyltransferase suggests its involvementin papaverine biosynthesis in opium poppy (Papaver somniferum L)Plant J 2009 6056-67

31 Weber AP Weber KL Carr K Wilkerson C Ohlrogge JB Sampling theArabidopsis transcriptome with massively parallel pyrosequencing PlantPhysiol 2007 14432-42

32 Droege M Hill B The Genome Sequencer FLX System-longer reads moreapplications straight forward bioinformatics and more complete datasets J Biotechnol 2008 1363-10

33 Andreas PM Weber KL Weber KC Wilkerson C Ohlrogge JB Sampling theArabidopsis transcriptome with massively parallel pyrosequencing PlantPhysiol 2007 14432-42

34 Torres TT Metta M Ottenwalder B Schlotterer C Gene expression profilingby massively parallel sequencing Genome Res 2008 18172-177

35 Gygi SP Rochon Y Franza BR Aebersold R Correlation between proteinand mRNA abundance in yeast Mol Cell Biol 1999 191720-1730

36 Bennett MD Smith JB Nuclear DNA amounts in angiosperms Phil TransRoyal Soc London B 1976 274227-274

37 Emrich SJ Barbazuk WB Li L Schnable PS Gene discovery and annotationusing LCM-454 transcriptome sequencing Genome Res 2007 1769-73

38 Alcantara J Bird DA Franceschi VR Facchini PJ Sanguinarine biosynthesisis associated with the endoplasmic reticulum in cultured opium poppycells after elicitor treatment Plant Physiol 2005 138173-183

39 Zeng S Xiao G Guo J Fei Z Xu Y Roe BA Wang Y Development of a ESTdataset and characterization of EST-SSR s in a traditional Chinesemedicinal plant Epimedium sagittatum (Sieb et Zucc) Maxim BMCGenomics 2010 1194-42

40 Wang W Wang Y Zhang Q Qi Yan Guo D Global characterization ofArtemisia annua glandular trichome using 454 pyrosequencing BMCGenomics 2009 10465-475

41 Wall PK Leebens-Mack J Chanderbali AS Barakat A Wolcott E Liang HLandherr L Tomsho LP Hu Y Carlson JE Ma H Schuster SC Soltis DESoltis PS Altman N dePamphilis CW Comparison of next generationsequencing technologies for transcriptome characterization BMCGenomics 2009 10347-365

Page 16 of 17

42 Klopp C Pyrosequencing read bioas evidences and correction proposalfor genome sequencing 2009 [httpwwweadgeneinfoEventsNextGenSeqWorkshop2009tabid375Defaultaspx]

43 Gomez-Alvarez V Teal TK Schmidt TM Systematic artifacts inmetagenomes from complex microbial communities ISME J 200931314-1317

44 Hornshoslashj H Bendixen E Conley LN Andersen PK Hedegaard J Panitz FBendixen C Transcriptomic and proteomic profiling of two porcinetissues using high-throughput technologies BMC Genomics 2009 1030

45 Bird DA Franceschi VR Facchini PJ A tale of three cell types alkaloidbiosynthesis is localized to sieve elements in opium poppy Plant Cell2003 152626-2635

46 Baerenfaller K Grossmann J Grobei MA Hull R Hirsch-Hoffman MYalovsky S Zimmermann P Grossniklaus U Gruissem W Baginsky SGenome-scale proteomics reveals Arabidopsis thaliana gene models andproteome dynamics Science 2008 320938-941

47 Vertommen A Panis B Swennen R Carpentier SC Evaluation ofchloroform methanol extraction to facilitate the study of membraneproteins of non-model plants Planta 2010 2311113-1125

48 Eilert U Kurz WGW Constabel F Stimulation of sanguinarineaccumulation in Papaver somniferum cell cultures by fungal elicitors JPlant Physiol 1985 11965-76

49 Gamborg OL Miller RA Ojima K Nutrient requirements of suspensioncultures of soybean root cells Exp Cell Res 1968 50151-158

50 Meisel L Fonseca B Gonzalez S Baeza-Yates R Cambiazo V Campos RGonzalez M Orellana A Retamales J Silva H A rapid and efficient methodfor purifying high quality total RNA from peaches (Prunus persica) forfunctional genomics analyses Biol Res 2005 3883-88

51 Zhao ZJ Ng D cDNA library creation protocol 2007 [httpmyjgidoegovgeneralprotocolsSOP_DRAFT_cDNA_library_creation_454]

52 Ouyang S Buell CR (2004) The TIGR Plant Repeat Databases A collectiveresource for identification of repetitive sequences in plants NAR 2004 32 Database D360-363

doi1011861471-2229-10-252Cite this article as Desgagneacute-Penix et al Integration of deeptranscriptome and proteome analyses reveals the components ofalkaloid metabolism in opium poppy cell cultures BMC Plant Biology2010 10252

Submit your next manuscript to BioMed Centraland take full advantage of

bull Convenient online submission

bull Thorough peer review

bull No space constraints or color figure charges

bull Immediate publication on acceptance

bull Inclusion in PubMed CAS Scopus and Google Scholar

bull Research which is freely available for redistribution

Submit your manuscript at wwwbiomedcentralcomsubmit

Page 17 of 17

Abstract
- Background
- Results
- Conclusions
- - Background
  - Results
  - - Induction of sanguinarine accumulation in opium poppy cell cultures
    - Properties of the transcript database generated by 454 pyrosequencing
    - Deep transcriptome analysis of elicitor-treated opium poppy cell cultures
    - Deep proteome analysis of elicitor-treated opium poppy cell cultures
    - Integration of transcriptome and proteome databases
    - - Discussion
      - Conclusions
      - Methods
      - Cell culture and elicitor treatment
        
        R NA extraction cDNA library construction 454 pyrosequencing and data processing
        
        Protein extraction and SDS-PAGE
        
        Mass spectrometry and spectrum data analysis
        
        Metabolite extraction and HPLC analysis
        
        Accession numbers
        
        Acknowledgements
        
        Author details
        
        Authors contributions
        
        References

Page 2: Integration of deep transcriptome and proteome analyses reveals the components of alkaloid

cells with a fungal elicitor As such opium poppy cell cul-tures provide an effective model system to investigate theinducible regulation of BIA metabolism and other plantdefense pathways The induction of sanguinarine biosynth-esis and supporting metabolism in elicitor-treated opiumpoppy cell cultures has been characterized using a varietyof technologies including EST and species-specific micro-array analyses to analyze the transcriptome [1] LC-MSMS to survey the proteome [2] and Fourier-transformion-cyclotron resonance-mass spectrometry (FT-ICR-MS)to profile the metabolome [13] Although these studiesprovided valuable insights into the response of opiumpoppy cell cultures to fungal elicitor treatment the tech-nologies used to generate the various databases werelimited in terms of the depth of penetration into the tran-scriptome proteome and metabolome of the culturedcells For example LC-MSMS peptide analysis of 340spots isolated by two-dimensional SDS-PAGE led to theidentification of 219 proteins using a combination of pub-lic and species-specific sequence databases More exten-sive genomics resources for opium poppy would improvethe downstream identification and discovery of enzymesinvolved in alkaloid biosynthesis New sequencing technol-ogies such as 454 pyrosequencing and advances in LC-MSMS-based proteomics and bioinformatics will expandthe application of genomics methodologies to a vast arrayof non-model plants that produce interesting and valuablemetabolitesThe biosynthesis of BIAs in opium poppy starts with

the condensation of two tyrosine derivatives dopamineand 4-hydroxyphenylacetaldehyde (4-HPAA) by norco-claurine synthase (NCS) to yield (S)-norcoclaurine(Figure 1) [4-6] The formation of dopamine involvesthe decarboxylation of tyrosine andor dihydrophenylala-nine (DOPA) by tyrosineDOPA decarboxylase (TYDC)[7] (S)-Norcoclaurine is then methylated by the norco-claurine 6-O-methyltransferase (6OMT) and coclaurineN-methyltransferase (CNMT) to yield (S)-methylco-claurine [8-11] The P450-dependent monooxygenase(S)-N-methylcoclaurine-3rsquo-hydroxylase (NMCH orCYP80B3) catalyzes the 3rsquo-hydroxylation of (S)-N-methylcoclaurine prior to the formation of (S)-reticulineby the 3rsquo-hydroxy-N-methylcoclaurine 4rsquo-O-methyltrans-ferase (4rsquoOMT) [811-14] (S)-Reticuline is the centralintermediate in the biosynthesis of most BIA structuraltypes including morphinans (eg morphine) benzophe-nanthridines (eg sanguinarine) and substituted benzyli-soquinolines (eg laudanine and papaverine) (AdditionalFile 1)The berberine bridge enzyme (BBE) converts (S)-

reticuline to (S)-scoulerine as the first committed stepin sanguinarine biosynthesis (Figure 1) [1315-17] TwoP450-dependent enzymes cheilanthifoline synthase(CheSyn) and the stylopine synthase (StySyn) catalyze

the formation of two methylenedioxy bridges and yield(S)-stylopine [1819] Subsequently tetrahydroprotober-berine cis-N-methyltransferase (TNMT) converts(S)-stylopine to (S)-cis-N-methylstylopine [20] Twoadditional P450-dependent enzymes N-methylstylopine14-hydroxylase (MSH) [21] and protopine 6-hydroxylase(P6H) are responsible for the conversion of (S)-cis-N-methylstylopine to 6-hydroxyprotopine which sponta-neously rearranges to yield dihydrosanguinarine [2122]Finally dihydrosanguinarine is oxidized to sanguinarineby the oxygen-dependent oxidoreductase dihydro-benzophenanthridine oxidase (DBOX) [2324] Thebiosynthesis of morphine involves the epimerization of(S)-reticuline to (R)-reticuline followed by a series ofC-C phenol coupling two reductions O-acetylation andtwo O-demethylations [25-29] Reticuline 7-O-methyl-transferase (7OMT) converts (S)-reticuline to (S)-lauda-nine [9] whereas norreticuline 7-O-methyltransferase(N7OMT) yields norlaudanine from norreticuline [30](Additional File 1) Cognate cDNAs have been reportedfor all of the aforementioned enzymes with the excep-tion of MSH P6H and DBOXThe standard approach to establish genomics resources

for non-model plant species involves the random genera-tion of expressed sequence tags (ESTs) from a cDNAphagemid library using dideoxy chain-termination (San-ger) sequencing technology Next-generation technolo-gies such as 454 pyrosequencing have the potential todramatically increase the availability of sequence data[3132] The redundancy and depth of coverage of 454pyrosequencing also provides and unbiased representa-tion of transcript abundance which is useful for relativegene expression analysis especially in non-model plantsthat lack complete genome sequence information [3334]However despite the fundamental importance of tran-scriptome analysis in genomics-based research the fre-quent incongruity between steady-state protein levels andthe abundance of cognate gene transcripts [35] is crucialto the interpretation of relative gene expression profilesin the context of systems biology or gene discovery appli-cations Complementary analysis of the most abundantproteins combined with a comprehensive transcriptomedatabase provides an important validation tool for therelative importance of gene transcripts within a givencell tissue or organ Transcript and protein sequencedatabases have recently been reported for opium poppycell cultures using Sanger sequencing of randomlyselected cDNAs and first-generation LC-MSMS analysisof proteins isolated by two-dimensional SDS-PAGE [12]In terms of the components of sanguinarine metabolismtranscripts corresponding to all known biosyntheticgenes were present in the EST database although somewere represented by relatively few sequence reads Com-bined with a low-throughput two-dimensional sampling

Page 2 of 17

Page 3 of 17

Feature Number

Page 4 of 17

Page 5 of 17

Page 6 of 17

Rank Numberof reads

Page 7 of 17

Page 8 of 17

Page 9 of 17

Page 10 of 17

Page 11 of 17

Page 12 of 17

Page 13 of 17

Page 14 of 17

Additional material

Page 15 of 17

Page 16 of 17

Page 17 of 17

Abstract
- Background
- Results
- Conclusions
- - Background
  - Results
    - - Discussion
      - Conclusions
      - Methods
        
        
        
        
        
        Accession numbers
        
        Acknowledgements
        
        Author details
        
        
        References

Page 3: Integration of deep transcriptome and proteome analyses reveals the components of alkaloid

Page 3 of 17

Feature Number

Page 4 of 17

Page 5 of 17

Page 6 of 17

Rank Numberof reads

Page 7 of 17

Page 8 of 17

Page 9 of 17

Page 10 of 17

Page 11 of 17

Page 12 of 17

Page 13 of 17

Page 14 of 17

Additional material

Page 15 of 17

Page 16 of 17

Page 17 of 17

Abstract
- Background
- Results
- Conclusions
- - Background
  - Results
    - - Discussion
      - Conclusions
      - Methods
        
        
        
        
        
        Accession numbers
        
        Acknowledgements
        
        Author details
        
        
        References

Page 4: Integration of deep transcriptome and proteome analyses reveals the components of alkaloid

Feature Number

Page 4 of 17

Page 5 of 17

Page 6 of 17

Rank Numberof reads

Page 7 of 17

Page 8 of 17

Page 9 of 17

Page 10 of 17

Page 11 of 17

Page 12 of 17

Page 13 of 17

Page 14 of 17

Additional material

Page 15 of 17

Page 16 of 17

Page 17 of 17

Abstract
- Background
- Results
- Conclusions
- - Background
  - Results
    - - Discussion
      - Conclusions
      - Methods
        
        
        
        
        
        Accession numbers
        
        Acknowledgements
        
        Author details
        
        
        References

Page 5: Integration of deep transcriptome and proteome analyses reveals the components of alkaloid

Page 5 of 17

Page 6 of 17

Rank Numberof reads

Page 7 of 17

Page 8 of 17

Page 9 of 17

Page 10 of 17

Page 11 of 17

Page 12 of 17

Page 13 of 17

Page 14 of 17

Additional material

Page 15 of 17

Page 16 of 17

Page 17 of 17

Abstract
- Background
- Results
- Conclusions
- - Background
  - Results
    - - Discussion
      - Conclusions
      - Methods
        
        
        
        
        
        Accession numbers
        
        Acknowledgements
        
        Author details
        
        
        References

Page 6: Integration of deep transcriptome and proteome analyses reveals the components of alkaloid

Page 6 of 17

Rank Numberof reads

Page 7 of 17

Page 8 of 17

Page 9 of 17

Page 10 of 17

Page 11 of 17

Page 12 of 17

Page 13 of 17

Page 14 of 17

Additional material

Page 15 of 17

Page 16 of 17

Page 17 of 17

Abstract
- Background
- Results
- Conclusions
- - Background
  - Results
    - - Discussion
      - Conclusions
      - Methods
        
        
        
        
        
        Accession numbers
        
        Acknowledgements
        
        Author details
        
        
        References

Page 7: Integration of deep transcriptome and proteome analyses reveals the components of alkaloid

Rank Numberof reads

Page 7 of 17

Page 8 of 17

Page 9 of 17

Page 10 of 17

Page 11 of 17

Page 12 of 17

Page 13 of 17

Page 14 of 17

Additional material

Page 15 of 17

Page 16 of 17

Page 17 of 17

Abstract
- Background
- Results
- Conclusions
- - Background
  - Results
    - - Discussion
      - Conclusions
      - Methods
        
        
        
        
        
        Accession numbers
        
        Acknowledgements
        
        Author details
        
        
        References

Page 8: Integration of deep transcriptome and proteome analyses reveals the components of alkaloid

Page 8 of 17

Page 9 of 17

Page 10 of 17

Page 11 of 17

Page 12 of 17

Page 13 of 17

Page 14 of 17

Additional material

Page 15 of 17

Page 16 of 17

Page 17 of 17

Abstract
- Background
- Results
- Conclusions
- - Background
  - Results
    - - Discussion
      - Conclusions
      - Methods
        
        
        
        
        
        Accession numbers
        
        Acknowledgements
        
        Author details
        
        
        References

Page 9: Integration of deep transcriptome and proteome analyses reveals the components of alkaloid

Page 9 of 17

Page 10 of 17

Page 11 of 17

Page 12 of 17

Page 13 of 17

Page 14 of 17

Additional material

Page 15 of 17

Page 16 of 17

Page 17 of 17

Abstract
- Background
- Results
- Conclusions
- - Background
  - Results
    - - Discussion
      - Conclusions
      - Methods
        
        
        
        
        
        Accession numbers
        
        Acknowledgements
        
        Author details
        
        
        References

Page 10: Integration of deep transcriptome and proteome analyses reveals the components of alkaloid

Page 10 of 17

Page 11 of 17

Page 12 of 17

Page 13 of 17

Page 14 of 17

Additional material

Page 15 of 17

Page 16 of 17

Page 17 of 17

Abstract
- Background
- Results
- Conclusions
- - Background
  - Results
    - - Discussion
      - Conclusions
      - Methods
        
        
        
        
        
        Accession numbers
        
        Acknowledgements
        
        Author details
        
        
        References

Page 11: Integration of deep transcriptome and proteome analyses reveals the components of alkaloid

Page 11 of 17

Page 12 of 17

Page 13 of 17

Page 14 of 17

Additional material

Page 15 of 17

Page 16 of 17

Page 17 of 17

Abstract
- Background
- Results
- Conclusions
- - Background
  - Results
    - - Discussion
      - Conclusions
      - Methods
        
        
        
        
        
        Accession numbers
        
        Acknowledgements
        
        Author details
        
        
        References

Page 12: Integration of deep transcriptome and proteome analyses reveals the components of alkaloid

Page 12 of 17

Page 13 of 17

Page 14 of 17

Additional material

Page 15 of 17

Page 16 of 17

Page 17 of 17

Abstract
- Background
- Results
- Conclusions
- - Background
  - Results
    - - Discussion
      - Conclusions
      - Methods
        
        
        
        
        
        Accession numbers
        
        Acknowledgements
        
        Author details
        
        
        References

Page 13: Integration of deep transcriptome and proteome analyses reveals the components of alkaloid

Page 13 of 17

Page 14 of 17

Additional material

Page 15 of 17

Page 16 of 17

Page 17 of 17

Abstract
- Background
- Results
- Conclusions
- - Background
  - Results
    - - Discussion
      - Conclusions
      - Methods
        
        
        
        
        
        Accession numbers
        
        Acknowledgements
        
        Author details
        
        
        References

Page 14: Integration of deep transcriptome and proteome analyses reveals the components of alkaloid

Page 14 of 17

Additional material

Page 15 of 17

Page 16 of 17

Page 17 of 17

Abstract
- Background
- Results
- Conclusions
- - Background
  - Results
    - - Discussion
      - Conclusions
      - Methods
        
        
        
        
        
        Accession numbers
        
        Acknowledgements
        
        Author details
        
        
        References

Page 15: Integration of deep transcriptome and proteome analyses reveals the components of alkaloid

Additional material

Page 15 of 17

Page 16 of 17

Page 17 of 17

Abstract
- Background
- Results
- Conclusions
- - Background
  - Results
    - - Discussion
      - Conclusions
      - Methods
        
        
        
        
        
        Accession numbers
        
        Acknowledgements
        
        Author details
        
        
        References

Page 16: Integration of deep transcriptome and proteome analyses reveals the components of alkaloid

Page 16 of 17

Page 17 of 17

Abstract
- Background
- Results
- Conclusions
- - Background
  - Results
    - - Discussion
      - Conclusions
      - Methods
        
        
        
        
        
        Accession numbers
        
        Acknowledgements
        
        Author details
        
        
        References

Page 17: Integration of deep transcriptome and proteome analyses reveals the components of alkaloid

Page 17 of 17

Abstract
- Background
- Results
- Conclusions
- - Background
  - Results
    - - Discussion
      - Conclusions
      - Methods
        
        
        
        
        
        Accession numbers
        
        Acknowledgements
        
        Author details
        
        
        References

Date post:	12-Sep-2021
Category:	Documents
Upload:	others
View:	1 times
Download:	0 times

Integration of deep transcriptome and proteome analyses reveals the components of alkaloid

Documents