Molecular Tools Developed for Nicotiana attenuata

A comprehensive toolbox for the N. attenuata ecological expression system has been developed in our Department. The most important compartment of this toolbox is a set of transgenic N. attenuata lines, transformed with our stable transformation system, post-transcriptionally silencing or ectopically overexpressing more than 200 different genes (table below) that are important for N. attenuata’s ecological performance. A large proportion of the targeted genes originate from “ask the plant” experiments in which the transcriptomes of plants involved in an ecological interaction (herbivory, pollination, competition, mycorrhization, root microbial colonization etc.) were queried by DDRT-PCR, SuperSage, various cDNA and oligo microarrays and RNAseq to identify differentially regulated genes. Many of the differentially regulated genes were first silenced with a virus induced gene silencing system based on the Tobacco Rattle Virus optimized for rapid gene silencing in N. attenuata (VIGS below), before being stably transformed.

The collection of transformed lines represents a unique tool that is being used to study the function of ecologically relevant genes in field experiments (Fig. 1a), tents, mesocosms, glasshouse (Fig. 1b) and even the wind tunnels of our institute (Ecological toolbox). All of these lines are available in T2 or T3 generations as fully characterized homozygous lines harboring single insertions of the silencing/overexpression constructs and hence can be crossed to make all binary combinations of transformants in hemizygous lines. In most cases, the silencing phenotype is comparable between homozygous single construct- and hemizygous binary construct-harboring lines.

Fig. 1: Experiments with transgenic plants from our Molecular toolbox. Each color of a flag or label indicates a different transgenic line. (a) Field experiments in the native habitat of N. attenuata in Utah (USA) Picture: A. Weinhold. (b) N. attenuata plants (same age) in the glasshouse growing in a paired competition design with two different genotypes competing in a single pot.

To complement the above-mentioned reverse-genetics approach using RNAi for the identification of ecologically important genes, a forward genetics approach has been adopted using an AI-RIL (Advanced Intercross-Recombinant Inbred Lines) population created from a cross of the two best studied N. attenuata ecotypes (“Utah” and “Arizona”) and a MAGIC (Multiparent Advanced Generation InterCross) population created from 26 different ecotypes that expressed unique defense, metabolomics, and morphological phenotypes (MAGIC below). Both of these RILs are in the final stages of breeding and phenotyping and we are now planning field experiments both at our field station in Utah as well as at field sites in Arizona.

The assembly of the almost completed sequences of the N. attenuata and N. obtusifolia genomes and transcriptomes are available (Genomic databases below). This sequence information was used to design complete transcriptomic microarrays based on AGILENT´s 4x44k and 8x60k platforms, which allow for the identification and transcriptional analyses of ecologically relevant genes. For the same purpose SAGE libraries and qPCR tools have been developed. Moreover, genomic and transcriptomic databases from different accessions of N. attenuata and different Nicotiana species and RNAseq data from different plant tissues and plants attacked by different insect species (Ecological tool box) are available. A microsatellite based method for the identification of different N. attenuata populations, ecotypes and individuals has also been developed. To study signaling between shoots and roots, a micrografting method can be applied for N. attenuata.


We established a MAGIC (Multiparent Advanced Generation InterCross) population from 26 different N. attenuata ecotypes. N. attenuata accessions were collected over 20 years from more than 100 natural populations (Fig. 2).

Fig. 2: Collection sites of seeds from different natural N. attenuata populations in the past 20 years. Individuals from these collections were phenotypically characterized to choose the ecotypes with the most extreme phenotypes.

For the selection of the parental plants used in the MAGIC cross, morphological and biochemical traits of 424 individuals from 230 accessions were characterized. To guarantee genetic diversity, the 26 individuals with the most extreme morphological, chemical, and defensive phenotypes, representing 26 different ecotypes including “Utah” and “Arizona” were selected as parents.

For the MAGIC cross first all 325 possible combinations of the 26 parents were performed in a half diallel cross without parents or reciprocal crosses, ♀ and ♂ were chosen randomly (Fig. 3a). The offspring with each 2 of the 26 chosen parents was used in four subsequent rounds of crossings, designed so that each of the 325 combinations served as ♀ and most as ♂, so that no reciprocal crosses occurred and that each offspring had 4, 8, 16 and all 26 different parents after 1, 2, 3, 4 rounds of crossing, respectively (Fig. 3b). To obtain the final MAGIC Recombinant Inbred Lines (RILs), the 325 obtained lines will be inbred for 6 consecutive generations (Fig. 3c).

Fig. 3: Crossing strategy for the MAGIC population.
(a) Step 1, half diallel cross of all 26 parental lines
(b) Step 2, four rounds of designed crosses, each offspring has 26 different parental lines
(c) Step 3, six rounds of inbreeding, Recombinant Inbred Lines with 26 different parental lines

Virus induced Gene Silencing (VIGS)

A VIGS system based on a tobacco rattle virus containing vector has been adapted for N. attenuata (Saedler & Baldwin, 2004) which allows for the rapid silencing of endogenous genes (Fig. 4) in plants in a matter of weeks. Another application of this VIGS system is Plant mediated RNAi (PmRNAi) to transiently silence insect genes of the plant’s specialist herbivores, including Manduca sexta (Kumar et al., 2012). In this example, a 300 bp fragment of the target gene is cloned in the VIGS vector in antisense orientation. After Agro-infiltration of N. attenuata with this vector, dsRNA of this fragment is formed as a result of TRV replication. Native M. sexta larvae feeding on plants stabily transformed with this construct in the field were silenced in the expression the targeted midgut gene.

At present 250 different constructs for VIGS are available in the department. Since the current VIGS procedure only works at relatively low temperatures of around 20°C we are now working on the development of a temperature resistant VIGS system that can be used under field conditions at temperatures of more than 30°C in the native habitat of N. attenuata.

Fig. 4: Virus Induced Gene Silencing (VIGS) of the phytoene desaturase (pds) gene in N. attenuata. Knocking down the pds gene disturbs the antenna pigment formation of the plant’s photosystem and leads to bleaching. Whereas VIGS in (a) was induced by the commonly used method of inoculation with Agrobacterium tumefaciens, VIGS in (b) was induced by inoculation with Agrobacterium free sap harvested from N. attenuata plants with VIGS silenced pds gene. Pictures: Jamilur Rahman.

Genomic databases

In an ongoing project with the Max-Planck-Institute for Molecular Genetics, the genomes of N. attenuata (2.4 Gb) and N. obtusifolia (1.5 Gb) are being sequenced. At present a draft N. attenuata version with an N50 contigs length of 27.6 kb comprising 2.2 Gb has been established and is available on our department´s internal BLAST server. A variety of tools for gene prediction, annotation and analysis of the genomic data has been established, among them the prediction tool AUGUSTUS, the annotation tool WebApollo and our GALAXY browser allowing access to tools for sequence analyses. Our current efforts are focused at improving the assembly. For this, we are using Pacific Biosciences (PacBio) sequencing and BioNano Genomics DNA optical mapping technologies. PacBio method results in long reads up to 60 kb and allows assembling of DNA regions, particularly the repeat regions which yields suboptimal results using Illumina short reads sequencing. This assembly would be further improved by 10x Genomics linked read technology, which links short Illumina sequence reads using DNA barcoding, and results in large contiguous assemblies. Additionally, BioNano optical maps generated from high molecular weight DNA provides a complete graphic map of the genome, allowing chromosome length scaffolding of assembled contigs to the near exact positions in the N. attenuata chromosomes. We anticipate that a combination of all these methods with our current sequence data will lead to a near complete N. attenuata genome, which will soon be made publicly available. In addition, the genome of the related species Nicotiana obtusifolia has been sequenced with lower coverage. The availability of both sequences allows for studies of the evolution of ecologically relevant traits in the genus Nicotiana. Link to the NaDH database.

High throughput pyrosequencing of microbial communities

Roots and leaves of field-grown N. attenuata plants can be colonized by diverse microbial communities living as endophytes (within the plant) or as epiphytes (on the plant surface). We isolated DNA from field grown plants to amplify bacterial 16S rDNA using universal bacterial primers. For the comparison of microbial communities among different genotypes we used high throughput 454 pyrosequencing offered by an external company (Santhanam, et al., 2014). We also investigated whether the composition of fungal communities is influenced by the calcium calmodulin-dependent protein kinase gene (CCaMK) using a transgenic irCCaMK knock down line and primers for the fungal internal transcribed spacer (ITS) region.

Bisulfite genomic sequencing

We used bisulfite genomic sequencing for promoter methylation analyses in N. attenuata plants (Weinhold et al., 2013). Primers were designed to be universally suitable for the analysis of the NOS promoter of most transgenic plant lines in our department. Transgenic N. attenuata plants can show a very dynamic increase in promoter methylation during vegetative development which is inherited to the following generation and leads to epigenetic gene silencing.

Transcriptomic databases

The sequences of the nearly whole transcriptome of N. attenuata are available. mRNA from different tissues of N. attenuata collected at different time points after elicitations, was isolated, labeled and sequenced using Illumina and 454 technologies. From these data, the nearly complete transcriptome of N. attenuata was assembled de novo. To determine transcript abundance for each gene in specific tissues, N. attenuata tissues from different organs or growth stages (leaves, roots, stems, flower parts, seeds) were tagged individually prior to Illumina sequencing. Additionally, de novo transcriptome assembly is performed on all the 26 MAGIC parents and a comparitive transcriptomic analysis between these ecotypes will give insight in intraspecific genome variations and adaptation. We are now in the process of sequencing the transcriptome of one set of MAGIC population post elicitation in the field, which would then enable us to perform eQTL and identify varition in gene expression associated with herbivory. The nearly complete transcriptome sequence of the related species N. obtusifolia is also available and was established in cooperation with Vertis Biotechnologie AG. mRNA was isolated, normalized and sequenced using the 454 technology. All next generations sequencing transcriptome assemblies can be currently analyzed on our group internal BLAST server.

In an ongoing project transcriptomic data from Nicotiana x obtusiata, the allotetraploid cross of N. attenuata and N. obtusifolia, and N. quadrivalvis and N. clevelandii were generated to investigate allopolyploid species formation. These are being used to study how the genetic architecture of complex ecological adaptions are altered during allopolyploidization to better understand why this common form of speciation is so strongly associated with adaptive radiations [(Pearse et al., (2006); Anssour et al., (2008) Anssour & Baldwin (2010); Bezzi et al., (2010)].

To enable studies of plant-insect interactions the transcriptomes of the most common herbivores (Manduca sexta, M. quinquemaculata, Corimelaena extensa, Empoasca spec., Tupiocoris spec., Oecanthus fultoni and Trichobaris mucorea) feeding on N. attenuata and of the predator Geocoris spec., have been sequenced using Illumina technology. We anticipate using these sequences in numerous future PmRNAi experiments in the field and laboratory to unravel the mysteries of how these insects manage to feed on N. attenuata or the herbivores of N. attenuata. Link to the NaDH database.

Tools to study smRNA mediated interactions

In close collaboration with Shree Pandey’s group at IISER Kolkata we are investigating the world of regulatory small-RNAs. RNA-directed RNA polymerases (RdRs), DICER-Like proteins (DCLs) and the Argonautes (AGOs) from the three most important components of the small-RNA pathway. We have generated knock-down-lines for all the three functional RdRs (irRdR1, 2 and 3; Pandey & Baldwin, 2007; Pandey & Baldwin, 2008; Pandey et al., 2008), the four DCLs (Bozorov et al., 2011) and eight AGOs. Using NGS approach, we have developed an expression atlas of N. attenuata smRNAs (Pandey et al., 2008). The Pandey group is in parallel developing databases and computational resources (such as Srivastava et al., 2014) for N. attenuata smRNA genomics that is proving helpful to understand their function and evolution (Singh et al., 2015).

Transgenic Plant Collection and Stable Transformation System

The standard transformation procedures developed for cultivated tobacco did not work for N. attenuata and hence a completely new Agrobacterium based transformation procedure (Krügel et al., 2002) with new transformation vectors optimized for the transformation of the elongated hypocotyl and the regeneration of plants from the resulting callus was developed. We constructed around 150 binary plant transformation vectors each containing an inverted repeat sequence of a gene to be silenced constitutively and around 30 vectors for the constitutive ectopic overexpression of full length genes. Our latest cloning vectors for these types of constructs are the pRESC8/pRESC9 plasmids (Fig. 5) with a hygromycin plant selectable marker and with the nptII gene as bacterial selectable marker, which allows flanking sequence at the insertion site to be recovered. Other vectors with nourseothricin as a selectable marker are also available, which allow plants to be transformed twice.

Fig. 5: pRESC8 binary plant transformation vector backbone. This vector series is used for the inverted repeat silencing of N. attenuata genes (‘goi’). In the pRESC9 series the ‘goi’ cassette is replaced by the coding sequence of a gene, thus allowing its ectopic overexpression.

Recently we were able to implement the highly sensitive LhGR/DEX – pOp6-promoter inducible system (Fig. 6a/6b) into our N. attenuata transformation system (Schäfer et al., 2013).

Fig. 6: The LhGR/DEX – pOp6-promoter inducible system. (a) Vector pSOL9LHGRC for the constitutive synthesis of the LHGR protein. (b) Backbone of the pPOP6 vector series for DEX induced gene (goi) overexpression. By replacing the goi sequence with an inverted repeat arrangement (as in Fig. 5) the pPOP6 backbone can be used for induced gene silencing.

This inducible promoter is very valuable for experiments under native field conditions, because in contrast to constitutive promoters it allows gene expression mediating ecological interactions to be manipulated in real time during an interaction and in a dose-dependent and spatially explicit manner. Moreover, these DEX-inducible constructs allow genetic traits that play essential roles in development to be manipulated without collateral damage to developmental programs (Fig. 7). This allows us to ask completely new research questions. More than 230 different transgenic plant lines have been generated, among them a set of about 30 different vectors for the pOp6-promoter driven induced silencing or overexpression of genes. We still continue to produce new lines for newly identified genes with high ecological relevance. An overview of the most important transgenic N. attenuata lines available in our group can be found in the table at the end of the chapter.

Fig. 7: Dexamethasone (DEX) induced gene silencing of the phytoene desaturase (pds) gene in N. attenuata under field conditions approximately 2 weeks after induction. (a) Bleached plant after treatment with 100µM DEX. (b) Treatment of different branches with different local DEX concentrations results in different degrees of bleaching. Pictures: Martin Schäfer.

Publication: Gase, K., Weinhold, A., Bozorov, T. A., Schuck, S., Baldwin, I. T. (2011). Efficient screening of transgenic plant lines for ecological research. Molecular Ecology Resources, 11(5), 890-902. doi:10.1111/j.1755-0998.2011.03017.x.


The availability of nearly complete transcriptomic and genomic sequence data for N. attenuata enabled the development of 8x60k microarrays for the Agilent Technologies microarray platform covering the complete N. attenuata transcriptome. A 4x44k microarray for the AGILENT platform [GEO Accession GPL13527] based on a 454 sequencing approach [BioProject PRJNA223344] is also available. For this microarray the 43,533 N. attenuata contigs with the best coverage were used to design 60mer oligonucleotides. Statistical tools for the evaluation of the microarray data have been developed. A database with the results from hundreds of hybridizations from our ecological experiments is available.

SAGE (Serial Analysis of Gene Expression)

We have generated in collaboration with GenXPro GmbH two SAGE libraries with more than 350,000 total tags from N. attenuata wounded and FAC (Fatty Acid Conjugates) elicited leaves (Gilardoni et al., 2010). These tags represent ca. 23,000 unique transcripts and approx. 500 of these tags are statistically > 2-fold up-regulated after FAC treatment compared to wounding whereas 250 are > 2-fold down-regulated. Some of these tags have been selected for gene function characterization in N. attenuata using VIGS and RNAi gene-specific silencing techniques. Consistent with the high sensitivity of the SAGE technique to detect very low abundant transcripts (potentially encoding for regulatory factors), a large number of differentially expressed tags in FAC treated samples correspond to low copy number tags, and preliminary sequence alignment analysis has classified them as encoding for potential regulatory components (e.g., transcription factors, protein kinases, protein phosphatases, calcium binding proteins).


Quantitative PCR can be performed on two MX3005P cyclers. For transcriptional studies, primers and probes for many ecologically important genes are available.


Genotyping Primer pairs amplifying N. tabacum microsatellite markers (Bindler et al., A microsatellite marker based linkage map of tobacco. Theor Appl Genet. 2007; 114:341–349) were screened with N. attenuata gDNA from different populations for amplification of polymorphic loci. By using fluorescent-labeled primers, size determination of the identified PCR fragments (Fig. 8)

Fig. 8: GeneScan Plot on an ABI 3130xl sequencer (Fig. 9) confirmed that 6 primer pairs are suitable for genotyping.

Fig. 9: Applied Biosystems Genetic Analyzer 3130xl

These will allow us to genotype individual N. attenuata plants that we work with in natural habitats. The initial plant material grinding for the HTP gDNA isolation procedure is performed in a SPEX SamplePrep 2000 Geno/Grinder® (Fig. 10).

Fig. 10: SPEX SamplePrep 2000 Geno/Grinder®

Micrografting A micrografting method for the connection of shoots and roots from different transgenic N. attenuata plants has been developed and established (Fragoso et al., 2011). This simple procedure (Fig. 11) allows us to study signal transduction between the above- and below-ground organs of the plant and is also extensively used for experiments in the field (Ecological platform).

Fig. 11: All transformed plants are now fully characterized by a customized NanoString-nCounter module that rigorously establishes insertion copy number and T-DNA integration characteristic.



Fig. 11: Micrografting

Molecular Biology Standard Lab equipment

Our lab is well equipped with standard molecular biology instrumentation. Here is an overview of the equipment:

Modification and Preparation of DNA and RNA

  • 6 PCR-Machines
  • 8 Incubators
  • 1 Concentrator
  • 6 Micro-Centrifuges
  • 3 Table-Centrifuges (refrigerated)
  • 1 Gel Documentation System

Cultivation of Bacteria

  • 3 Table Shaking Incubators
  • 1 Horizontal Shaking Incubator

Analysis of the DNA and RNA

  • 1 Agilent 2100 Bioanalyzer (Fig. 9)
  • 1 Tecan Infinite 200
  • 1 Tecan Microplate Reader
  • 2 Photometers
  • 1 ABI Genetic Analyzer 3100 (Fig. 12)
  • 1 Isotope Imaging System (Phosphoimager)
  • 2 qPCR Machines Mx3005P

Fig 12: ABI Genetic Analyzer 3100

A selection of available transgenic N. attenuata lines



Experimental trait


Circadian clock




adagio-protein 2

clock-mediated traits

P6 ir


late elongated hypocotyl

clock-mediated traits

ir/ov, P6 ir/ov


timing of CAB expression

clock-mediated traits

ir, P6 ir



clock-mediated traits

ir, P6 ir


pseudo-response regulator

clock-mediated traits

ir/ov, P6 ir/ov

Cytokinin pathway




cytokinin oxidase


ov, P6 ov


cytokinin receptor




suppressor of G-two allele of SKP1




isopentenyl transferase from A. tumefaciens


S1 ov, P6 ov


isopentenyl transferase 2-4 from A. thaliana


S1 ov


histidine kinase



Ethylene pathway




1-aminocyclopropane 1-carboxylic acid oxidase

ethylene biosynthesis



ir silencing LOX2 and ov ETR1-1

ethylene signaling



ethylene receptor mutant

ethylene reception


Floral traits




chalcone synthase

benzyl acetone synthesis






sucrose transporter

nectar production







cell wall invertase inhibitor


ir, P6 ir


pectin methyl esterase

methanol production



rapid alkalization factor

root development



cinnamyl alcohol dehydrogenase

lignin biosynthesis


Indirect/direct defense





lignin/polyamine biosynthesis

ir, P6 ir


hydroxycinnamoyl CoA quinate transferase

chlorogenic acid biosynthesis



farnesyl pyrophosphate synthase

terpene biosynthesis



geranylgeranyl pyrophosphate synthase

DTG biosynthesis




DTG glycosylation



terpene synthase

Sesquiterpene biosynthesis



hydroperoxide lyase

GLV biosynthesis



putrescine n-methyl transferase

nicotine biosynthesis



trypsin protease inhibitor

TPI/nectarine secretion


Kinase signaling




salicylic acid-activated MAP kinase

herbivore related signaling



calcium/calmodulin-dependent protein kinase

Mycorrhiza interaction



calcium dependent protein kinase

JA biosynthesis



lectin-domain receptor-like kinase

herbivore related signaling


MEK1, 2

mitogen activated protein kinase kinase

herbivore related signaling



mitogen-activated protein kinase

herbivore related signaling



MAP kinase kinase

herbivore related signaling



wound inducible protein kinase

herbivore related signaling



sucrose non fermenting protein

beta-subunit of SNRK1


Light signaling





light reception


CRY1a, 2

cryptochrome 1a, 2

light reception

ir, P6 ir

PHYA, B1/2

phytochrome A, B1, B2

Far red response



constitutively photomorphogenic

ir, P6 ir


UVB receptor

UV response


Oxylipin signaling





2-HOT biosynthesis



coronatine insensitive 1

Jasmonate reception



coronatine insensitive 2

Jasmonate reception



jasmonic acid carboxyl methyltransferase

Jasmonate metabolism


JAR4, 6

jasmonic acid resistance protein

JA pathway



jasmonate ZIM domain A, C-E

Jasmonate metabolism



jasmonate ZIM domain H

Jasmonate metabolism

ir, P6 ir


jasmonoyl-L-isoleucine hydrolase

JA pathway



methy jasmonate esterase

JA pathway



ir silencing MJE and ovAtJMT

JA pathway



transforms 12-oxo-PDA to OPC-8:3

JA pathway


LOX1-3, 6


fatty acid peroxidation



acyl-CoA oxidase

JA biosynthesis



allene oxide cyclase

JA biosynthesis



allene oxide synthase

JA biosynthesis



threonine deaminase

isoleucine for JA conjugation


Pathogen related



DEF1, 2


antimicrobial peptide



non responding pathogen resistance protein

SA receptor



enhanced desease susceptibility

pathogen defesnse



isochorismate synthase

SA biosynthesis



salycilate hydroxylase

SA metabolism






phytoene desaturase

ir, P6 ir


NAPDH oxidase



RuBPCase activase

and oxylipin signaling








Small RNAs







dicer like



RNA dependent RNA polymerase


Transcription factors




MYB transcripion factors

phenylamide defense



WRKY transcription factors

herbivore related signaling



WRKY transcription factor

herbivore related signaling








herbivore/pathogen resistance



alternative oxidase

stress response



up frameshift protein





pathogen resistance




ir, P6 ir



terpene synthesis



herbivore elicitor regulated

ABA sensitivity



regulator for POP6 promoter

POP6 promoter induction



DR5 element

IAA signaling



ortholog of sugar beet Hs1pro-1




luminal binding protein

ER folding stress


as: antisense silencing (35S promoter)

ir: inverted repeat silencing (35S promoter)

ov: overexpression (35S promoter)

P6 ir: inverted repeat silencing (pOp6 promoter, DEX inducible)

P6 ov: overexpression (pOp6 promoter, DEX inducible)

S1 ov: overexpression (SAG1 promoter, senescence associated)



Selected literature using stably transformed lines (for a full list see: departmental publications):

  • Circadian clock:
    Yon, F., Seo, P. J., Ryu, J. Y., Park, C.-M., Baldwin, I. T., Kim, S.-G. (2012). Identification and characterization of circadian clock genes in a native tobacco, Nicotiana attenuata. BMC Plant Biology, 12: 172. doi:10.1186/1471-2229-12-172.

  • Ethylene pathway
    von Dahl, C. C., Winz, R., Halitschke, R., Kühnemann, F., Gase, K., Baldwin, I. T. (2007). Tuning the herbivore-induced ethylene burst: the role of transcript accumulation and ethylene perception in Nicotiana attenuata. The Plant Journal, 51(2), 293-307. doi:10.1111/j.1365-313X.2007.03142.x.

  • Floral traits
    Kessler, D., Gase, K., Baldwin, I. T. (2008). Field experiments with transformed plants reveal the sense of floral scents. Science, 321(5893), 1200-1202. doi:10.1126/science.1160072.

  • Kinase signaling
    Schäfer, M., Meza Canales, I. D., Navarro-Quezada, A., Brütting, C., Radomira, V., Baldwin, I. T., Meldau, S. (2014). Cytokinin levels and signaling respond to wounding and the perception of herbivore elicitors in Nicotiana attenuata. Journal of Integrative Plant Biology. doi:10.1111/jipb.12227.

  • Oxylipin signaling
    Oh, Y., Baldwin, I. T., Galis, I. (2013). A jasmonate ZIM-domain protein NaJAZd regulates floral jasmonic acid levels and counteracts flower abscission in Nicotiana attenuata plants. PLoS One, 8(2): e57868. doi:10.1371/journal.pone.0057868.

  • Photosynthesis
    Mitra, S., Baldwin, I. T. (2014). RuBPCase activase mediates growth-defense tradeoffs: silencing RCA redirects jasmonic acid (JA) flux from JA-Ile to MeJA to attenuate induced defense responses in Nicotiana attenuata. New Phytologist, 201, 1385-1395. doi:10.1111/nph.12591.

  • Small RNAs
    Bozorov, T. A., Pandey, S. P., Dinh, T. S., Kim, S.-G., Heinrich, M., Gase, K., Baldwin, I. T. (2012). Dicer-like proteins and their role in plant-herbivore interactions in Nicotiana attenuata. Journal of Integrative Plant Biology, 54(3), 189-206. doi:10.1111/j.1744-7909.2012.01104.x.