Rupam Kumar Bhunia1, Anirban Chakraborty1, Ranjeet Kaur1*, T. Gayatri1*, Jagannath

Seed-specific increased expression of 2S albumin promoter of sesame qualifies it as a useful genetic tool for fatty acid metabolic engineering and related transgenic intervention in sesame and other oil seed crops

Rupam Kumar Bhunia1, Anirban Chakraborty1, Ranjeet Kaur1*, T. Gayatri1*, Jagannath Bhattacharyya1, Asitava Basu1, Mrinal K. Maiti1,2 and Soumitra Kumar Sen1†

1Advanced Laboratory for Plant Genetic Engineering, Indian Institute of Technology, Kharagpur-721302, India

2Department of Biotechnology, Indian Institute of Technology, Kharagpur-721302, India

*Authors contributed equally

†Corresponding Author:

Ph. No. 91-3222277028;

Fax No. 91-3222277890

Email:

Abstract:(223words)

The sesame 2S albumin (2Salb) promoter was evaluated for its capacity to express the reporter gusA gene encoding β-glucuronidase in transgenic tobacco seeds relative to the soybean fad3C gene promoter element. Results revealed increased expression of gusA gene in tobacco seed tissue when driven by sesame 2S albumin promoter. Prediction based deletion analysis of both the promoter elements confirmed the necessary cis-acting regulatory elements as well as the minimal promoter element for optimal expression in each case. The results also revealed that cis-regulatory elements might have been responsible for high level expression as well as spatio-temporal regulation of the sesame 2S albumin promoter. Transgenic over-expression of a fatty acid desaturase (fad3C) gene of soybean driven by 2S albumin promoter resulted in seed-specific enhanced level of α-Linolenic acid (ALA) in sesame. The present study, for the first time helped to identify that the sesame 2S albumin promoter is a promising endogenous genetic element in genetic engineering approaches requiring spatio-temporal regulation of gene(s) of interest in sesame and can also be useful as a heterologous genetic element in other important oil seed crop plants in general for which seed oil is the harvested product. The study also established the feasibility of fatty acid metabolic engineering strategy undertaken to improve quality of edible seed oil in sesame using the 2Salbumin promoter as regulatory element.

Keywords: Sesame;2S albumin; fatty acid desaturase (fad3);gusA, α-Linolenic acid (ALA);

Seed-specific expression

Introduction:(1465 words)

The identification of a suitable promoter element is the most important attribute for any transgenic approach that demands a moderate to high level of target gene expression in a spatio-temporal manner to achieve agronomic or commercial benefit. Many plant promoters are already characterized and in use for both basic studies of gene expression, and also for value-added transgenic plant generation. In this aspect, both constitutive as well as tissue-specific promoter elements have been used in past (Potenza et al., 2004). In plants, viral CaMV35S promoter remains the most commonly used genetic element for regulation of transgene expression in a constitutive manner. However, in some cases, engineering of new traits was made possible by the expression of gene(s) during seed development (Tan et al., 2011). For example, it is now possible to successfully modify the fatty acid content of plant oils via transgenic metabolic engineering to generate something approaching ‘designer oil’ (Napier and Graham, 2010). This gains more importance in the context that the imbalanced ω-6:ω-3 fatty acid ratio in diet has been implicated for the prevalence of cancer, cardiovascular and inflammatory/autoimmune diseases (Kankaanpaaet al., 1999 and Poudyal et al., 2011). Increased level of ω-3 Polyunsaturated Fatty Acid (PUFA) in diet has suppressive effects towards these diseases (Simopoulos, 2002). Presently the primary sources of ω-3 fatty acids consist of some deep sea fish and some specific oilseed plants (e.g. flax, soybean, rape, walnut and perilla). The decline in the world fish supply accompanied by the rise of heavy metal pollution in the sea have further limited the common dietary sources of ω-3 fatty acids. Under the given circumstances,any transgenic metabolic engineering approach leading to increased α-linolenic acid (ALA, C18:3, ω-3) content in seed oil of oilseed plants is highly desirable. Sesame (Sesamum indicum L.)qualifies to be a major target plant in this regard.Sesame has superior oil quality and quantity in comparison to other vegetable oil available in market.The oil content ranges from 35% to 55% in different cultivars, the average being generally about 50% of the seed weight (Ashri, 1989). However, values up to 63.2% have been reported in some varieties (Baydar et al., 1999). The oil is comparatively stable due to the presence of a variety of lignans or antioxidants (Kamal-Eldin and Appelqvist, 1994), which are known to play several health beneficial roles also. Sesame seed oil also contains high level of both oleic acid (OA, C18:1, ω-9) and linoleic acid (LA, C18:2, ω-6). Despite these nutritional and agronomic benefits, it contains substantially low amount (<1%)of ALA(Mondal et al., 2010).In this context, sesame and other oil seed plantswith improved content of ALA could ameliorate much of the global dietary ω-3 fatty acid deficiency. Genetic engineering of this kind necessitates identification and characterization of newer seed-specific promoter elements which can drive the expression of target gene(s) of interest in such plants. Also, there is a need for identification and expression of suitable heterologous or endogenous candidate gene(s),implicated in fatty acid biosynthetic pathway, regulated by such seed-specific promoter elements to validate both the promoter efficiency as well as the genetic engineering strategy.

Seed maturation in higher plants is associated with the deposition of storage reserves, such as oil (triacylglycerol), carbohydrates, and proteins (Baud and Lepiniec, 2010). Seed proteins are generally synthesized in the cotyledon and the embryo of dicotyledonous plants or the endosperm of monocotyledonous plants, accumulating in organelles called protein bodies, which are surrounded by a single membrane of tonoplast or ER origin (Pernollet, 1978). The proteins are then utilized as carbon and nitrogen sources by the emerging seedling during germination (Argoset al., 1982). Genes encoding seed storage proteins (SSP) have been isolated from the major crop plants and evaluation of their biological roles, their ancestral origins, and their modes of synthesis and deposition have been investigated (Shewry et al., 1995). Most of the studied seed-specific promoters are isolated from such SSP encoding genes, such as rice glutelin and globulin(Hwang et al., 2002 and Qu et al., 2008), soybean lectin and and β-conglycinin (Cho et al., 1995 and Qinggele et al., 2007) and Brassica napin(Kridl et al., 1991). However, a number of genes encoding non-SSP proteins are also abundantly expressed in seeds during late embryogenesis. These include oleosins (Huang, 1996), late embryogenesis abundant (LEA) proteins (Fujiwara et al., 2002), oleate desaturases (Kim et al., 2006), sucrose synthases (Rasmussen and Donaldson, 2006) which were isolated and characterized from different plant species. 2S albumin is the major soluble storage protein of sesame seeds. It accumulates in discrete vesicles of protein bodies and constitutes a major fraction of the proteins found in the mature seed (Orruno and Morgan, 2011). Thus, the 5’ regulatory region of 2S albumin gene is expected to drive moderate to high level seed-specific expression of the target gene placed under its controlin oilseed crops in general andparticularly in sesame. On the other hand, it is well known that in higher plants the conversion of LA to ALA occurs both in the plastids and in the endoplasmic reticulum (ER) by omega-3/Δ15 desaturases. Among them ALA synthesis in seeds is mainly catalyzed by microsomal ω-3 FAD3 encoded by fad3 gene (Ohlrogge and Browse, 1995).It was found that approximately 1,000 bp upstream region from translational start site of microsomal fad3 gene was capable of directing seed-specific expression in flax seeds (Vrinten et al., 2005). Among the available vegetable oil in the nature, soybean oil carries optimum quantity of ALA and balanced proportion of ω-3: ω-6 ratio (Giacomelli et al., 2006). Three soybean microsomal ω-3 fatty acid desaturase (fad3) genes, designated as Gmfad3A, Gmfad3B, and Gmfad3C, are reported that contribute to seed ALA levels (Bilyeu et al., 2003). In developing seeds, the Gmfad3Ahas highest expression level (60% of house-keeping gene expression) with Gmfad3C having moderate expression. Fatty acid metabolic engineering in sesame and related oil seed plants being our major focus, very high level of seed-specific expression of target gene may lead to skewed ω-6: ω-3 ratio, which is not desirable in normal diet. Thus, Gmfad3Cgene promoter element of soybean could serve as another potential candidate for employment in genetic engineering strategies in oilseed crop plants particularly in sesame. This is more relevant in the context that in plant genetic engineering approaches, both native and heterologous promoter elements are successfully used.

The mechanism of fatty acid biosynthesis and oil formation in plant cells have been extensively studied (Ohlrogge et al., 1991and Ohlrogge and Browse, 1995). Omega-3 fatty acid desaturase enzymes introduce the third double bond into LA precursors to produce ALA precursors.Arabidopsis contains two plastidial desaturases, FAD7 and FAD8 (Browse et al., 1986 and McConn et al., 1994), which desaturate both hexadecadienoic acid (C16:2, ω-6) and LA, and a single microsomal FAD3 (Browse et al., 1993), which acts only on LA. The vital role of FAD3 is reflected in Arabidopsis, where the level of ALA decreased from ~20% of total seed fatty acids in wild-type plants to 1% to 2% in fad3 mutant lines (James and Dooner, 1990 and Lemieux et al., 1990), whereas leaf ALA contentwasonly slightly reduced (Lemieux et al., 1990 and Browse et al., 1993). Puttick et al. (2009) reported that the ALA content of seed increased from 19% to nearly 40% of total fatty acids in Arabidopsis by seed-specific over-expression of endogenous fad3 gene.Ectopic expression of heterologous fad3 in soybean seed also greatly increased both the ALA content and the ω-3:ω-6 ratio (Damude et al., 2006 and Eckert et al., 2006).So, over-expression of either endogenous or heterologous fad3gene might help to accumulate the ALA in sesame seeds leading to enhanced nutritional property.

In the present study, we isolated the sesame 2S albumin (2Salb) promoter and compared its capacity with fad3C gene promoter of soybean to drive transgene expression in seed tissue of model plant tobacco. Our aim was to make a case for the merits of one endogenous promoter of sesame versus one heterologous promoter element. Additionally, a comparative analysis was documented between the minimal promoter elements identified in each case in moderate to high level seed-specific expression. Based on the outcome of such analyses, microsomalfad3C gene from soybean was introduced into sesame under the control of predominantly seed-specific promoter, 2S albumin (isolated during the present study), to evaluate the efficiency of the isolated 5’ regulatory element in conferring moderate to high level seed-specific expression and also to test the feasibility of genetic engineering strategy to increase ALA content within health-permissible range in transgenic sesame seeds or any other oil seed crop in general. The significance of this study has direct bearing on our goal to develop genetically modified oilseed plants, with particular focus on sesame to improve the quality of seed oil for human consumption.

Materials and Methods: (1650 words)

Bacterial materials:

The super virulent EHA105 (Hood et al., 1993)Agrobacterium tumefaciens strain and binary vector pCAMBIA1300 were used in the tobacco and sesame transformation experiments. Escherichia coli strain DH10B, and cloning bacterial vector pUC18 and TA cloning vector (Invitrogen) were applied in gene cloning.

Plant materials:

Sesame (Sesamum indicum variety “Var-9”), Soybean (Glycine max) and tobacco (Nicotianatabacum) constituted the experimental plant materials in the present study.

Isolation of 5’ flanking DNA sequence of 2S albumin gene from sesame and fad3C gene from soybean:

Genome walking was carried out to isolate 5’ upstream sequence of 2S albumin gene of sesame using GenomeWalkerTM universal kit (Clontech) according to the manufacturer’s instructions. 2.5 μg aliquot of sesame genomic DNA was digested with EcoRV at 370C for overnight. Digested genomic DNA was ligated to GenomeWalker™ Adaptors. The primary PCR used the outer adaptor primer 1 (AP1) provided in the GenomeWalker universal kit and an inner, gene-specific primer 1 (Si2Salb GSP1). The primary PCR mixture was used as a template for a nested PCR with the nested adaptor primer 2 (AP2) and a nested gene-specific primer 2 (Si2Salb GSP2). The detailed thermal cycling using the two-step cycle parameters was as follows-

7 cycles: 94°C/ 25 sec, 72°C /3 min

32 cycles: 94°C /25 sec, 67°C/ 3 min

67°C for an additional 7 min after the final cycle

The amplified secondary PCR product was cloned into TA cloning vector (Invitrogen) and sequenced.

In another case, genomic DNA from experimental soybean plant was used as a template for PCR. Primer sets used for the isolation of the Gmfad3C putative promoter were pfad3 FP and pfad3 RP. The primers were designed on the basis of 5’ flanking sequence of fad3C gene available in PhytozomeV9 database( The amplified product was cloned in pUC18 vector and sequenced. The 5’ deletion fragments of the full-length promoters were generated by polymerase chain reaction (PCR) using specific primers (for sesame 2S albuminpromoter: delp2S albumin FP and delp2S albumin RP and for soybean fad3C promoter: delpfad3 FP and pfad3 RP). The bioinformatic analysis of both the promoter elements was carried out using Jellyfish 3.3.1 and PlantCARE (Lescot et al., 2002)softwares. The isolated full length promoter elements were submitted to NCBI under the following accession numbers: Sesamum indicum2S albumin promoter(KC414203) and Glycine max fad3C promoter (KC414201).

Isolation of total cellular RNA and synthesis of 1st strand cDNA:

Glycine max seed tissues were used as the source of total RNA isolation using RNeasy Mini Kit (Qiagen) according to manufacturer’s instruction.2 μg of total RNA was used for cDNA synthesis with Gmfad3 reverse primerusing Transcriptor 1st strand cDNA synthesis kit version 6.0 (Roche Molecular biochemicals) in a total volume of 20 μl using manufacturer’s protocol. cDNA was used for RT-PCR mediated isolation of the fad3C gene Coding DNA Sequence (CDS) from soybean.

Amplification of soybean fad3Cgene by RT-PCR

Isolation of 1140 bp soybean fad3C gene (Accession no: AY204712) CDS was carried out byRT-PCRusing 2 µl of the 1st strand cDNA with primer sets Gmfad3 Forward primer and Gmfad3 reverse primerby the following thermal profile: initial denaturation at 98°C for 2 min, followed by 30 cycles of 98°C/15s, 62°C/1min 30 secs, 72°C/1 min (1 min/Kb) and a final extension at 72°C for 10 min in a Applied Biosystems Veriti Gradient thermocycler.

Construction of chimeric gusA gene expression cassettes controlled by two different full length promoters and their deletions and chimeric fad3Cgene over-expression cassette controlled by sesame 2S albumin promoter in the binary vector pCAMBIA 1300:

The HindIII–BamHI fragment containing the full length and deletedpromoter elements in each case was fused in a translational frame at the BamHI site of the gusA gene in the pCAMBIA 1300 binary vector to generate the pCAM1300::2Salb/gusA, pCAM1300::del2Salb/gusA, pCAM1300::fad3C/gusA and pCAM1300::delfad3C/gusA binary vectors, respectively. For over-expression cassette, HindIII–BamHI fragment containing the 2S albuminpromoter element (isolated and characterized during the present study, Accession number: KC414203; denoted by sip2Salbumin) was fused in a translational frame with the BamHI-SacIfragment of the fad3CgeneCDS in the pCAMBIA 1300 binary vector to generate the pCAM1300::2Salb/fad3C binary vector.The expression vectors were transferred to Agrobacterium tumefaciens strain EHA105 by the freeze–thaw method (Burow et al., 1990) and subsequently transformed in tobacco and sesame plants, as the case may be.

Plant transformation

Agrobacterium tumefaciens mediated transformation using tobacco leaf discs and selection for primary transformant lines were essentially the same as described previously with slight modifications (Horsch et al., 1985). Transformed plants were selected on MS medium containing Hygromycin (35 mg/L), transferred to soil and grown to maturity under glasshouse conditions. The seeds from T0plants were then germinated in hygromycin (35 mg/L) containing media and the surviving plantlets were transferred to soil to obtain the T1 generation plants.

On the other hand, despite late progress (Yadav et al., 2010 andChowdhury et al., 2014), there was no well established protocol for sesame transformation.We have generated and optimized a protocol in our lab (unpublished data). As a brief outline, two day old germinated seedlings from the plumule tip were used as explants for Agrobacterium- mediated transformation. About 8-12 shoots were obtained from each explant and they were screened against hygromycin (35 mg/L). The surviving shoots were subjected to root development in rooting medium. The plantlets with well developed roots were finally acclimatized in soil under sterile condition for 1 week and then finally established in glasshouse. The T1 plants were generated as described above.

Southern hybridization

Plant DNA was isolated following the protocol of Doyle and Doyle (1987). Southern hybridization was performed according to Sambrook et al. (1989) at 650C in CHURCH buffer using 1.8 Kb gusA gene probe for tobacco transformant lines and 1140 bpfad3C gene probe for sesame transformant lines. In each case, 15 µg of genomic DNA was digested with BamHI and run on 0.8% gel overnight. The radiolabeling of the probe DNA was carried out with P32 dCTP (3500 ci/mmol) by random priming using rediprime II DNA labeling system (GE Healthcare, USA) following manufacturer’s instructions.

Gus activity

Histochemical and fluorometric analyses of GUS activity were conducted following the method of Jefferson et al. (1987). Fluorometric estimation was carried out with the help of spectrofluorometer (PerkinElmer LS45) according to manufacturer’s protocol.

Real Time PCR analysis of transgenic tobacco and sesame lines:

Real time PCR was carried out in an Eppendorf Realplex2 Master Cycler using SYBR green based relative quantification method using 5 prime kit (Eppendorf) according to Wang et al. (2006) in 20μl reaction volume. Total RNA was isolated from respective tissues according to hot-phenol extraction method of Verwoerd et al. (1989). 1st strand cDNA was synthesized by using gusA RP for tobacco transformants and Gmfad3 qRT RP for sesame transformants using transcriptor 1st strand cDNA synthesis kit (Roche). PCR was carried out using primer sets,gusA FP and gusA RP for gusA gene and Gmfad3qRT FP and Gmfad3 qRTRP for fad3C gene over-expression. Thermal cycling conditions were 2min at 94°C followed by 40 cycles of 94°C for 30s, 50°C for 15s, and 68°C for 30s. For each reading, duplicate CT values were averaged. Melting curve analysis in each case confirmed the amplification of specific product.

Data analysis of real time PCR using the 2-ΔΔCT method:

The 2-ΔΔCT method for relative quantification (Livak and Schmittgen, 2001) was adapted in the present study to estimate the relative expression levels of gusA gene in different tobacco tissue parts and of fad3C gene in seeds of transgenic sesame plants. β-actin(for tobacco Accession no:AB158612; for sesame Accession no: JQ658353)gene was used as the internal control for normalization in each case.The average CT was calculated for both reference and target genes and the ΔCt (Ct,target gene – Ct, β-actin) was determined.