A peptide from insects protects transgenic tobacco from a parasitic weed

Springer 2005

Transgenic Research (2005) 14:227–236 DOI 10.1007/s11248-004-7546-1

A peptide from insects protects transgenic tobacco from a parasitic weed Noureddine Hamamouch1, James H. Westwood1,*, Idit Banner2, Carole L. Cramer1, Shimon Gepstein2 & Radi Aly3 1

Virginia Tech, Department of Plant Pathology, Physiology, and Weed Science, Blacksburg VA 24061, USA The Faculty of Biology, Technion-Israel Institute of Technology, Haifa 32000, Israel 3 Agricultural Research Organization (ARO), Department of Weed Research, Newe Ya’ar Research Center, Ramat Yishay 30095, Israel 2

Received 13 May 2004; revised 28 October 2004; accepted 8 November 2004

Key words: Egyptian broomrape, HMG2, Orobanche aegyptiaca, resistance, sarcotoxin IA, tobacco

Abstract Parasitic plants present some of the most intractable weed problems for agriculture in much of the world. Species of root parasites such as Orobanche can cause enormous yield losses, yet few control measures are effective and affordable. An ideal solution to this problem is the development of parasite-resistant crops, but this goal has been elusive for most susceptible crops. Here we report a mechanism for resistance to the parasitic angiosperm Orobanche based on expression of sarcotoxin IA in transgenic tobacco. Sarcotoxin IA is a 40-residue peptide with antibiotic activity, originally isolated from the fly, Sarcophaga peregrina. The sarcotoxin IA gene was fused to an Orobanche-inducible promoter, HMG2, which is induced locally in the host root at the point of contact with the parasite, and used to transform tobacco. The resulting transgenic plants accumulated more biomass than non-transformed plants in the presence of parasites. Furthermore, plants expressing sarcotoxin IA showed enhanced resistance to O. aegyptiaca as evidenced by abnormal parasite development and higher parasite mortality after attachment as compared to non-transformed plants. The transgenic plants were similar in appearance to non-transformed plants suggesting that sarcotoxin IA is not detrimental to the host.

Introduction Parasitic weeds of the genus Orobanche (broomrapes) are obligate holoparasites that attack the roots of many economically-important crops throughout the semiarid regions of the world, especially the Mediterranean and Middle East where they are endemic. The parasites form vascular connections with the host and act as strong sinks for the uptake of water, nutrients and photosynthates, often causing severe losses in crop yield and quality (Parker and Riches, 1993).

* Author for correspondence E-mail: [email protected]

Orobanche control is difficult because the parasites are closely associated with the host root and are concealed underground for most of their life cycle. These parasites are not controlled effectively by traditional cultural or herbicidal weed control strategies (Foy et al., 1989). The most effective current control method is soil fumigation with methyl bromide (Jacobsohn, 1994), but this is expensive and hazardous to the environment. Another promising Orobanche control strategy involves herbicide-resistant crops, which allow herbicides to translocate harmlessly through the host plant to accumulate in, and kill, the parasite (Joel et al., 1995; Surov et al., 1998). However, this approach depends on commercial availability of herbicide-resistant varieties of host

228 crops in affected regions and has only recently been adapted to protect maize from Striga spp. parasitism in Africa (Kanampiu et al., 2002). The best long-term strategy for limiting damage by Orobanche is the development of parasite-resistant crops that would preclude the need for herbicides or other control measures (Cubero, 1991; Ejeta et al., 1991). However, generation of resistant crops through conventional means has been challenging and slow. Genetic engineering may offer new insights and strategies to facilitate progress toward this goal. Sarcotoxin IA is an anti-microbial peptide from the flesh fly (Sarcophaga peregrina). It is a cecropin-type peptide that interacts with the bacterial cell membrane, causing a loss of electrochemical potential (Nakajima et al., 1987; Iwai et al., 1993). Sarcotoxin IA has been shown to have potent cytotoxic effects against several bacteria, including plant pathogenic bacteria (Aly et al., 1999), and confer resistance to both bacterial (Erwinia carotovora subsp. carotovora and Pseudomonas syringae pv. tabaci) and fungal (Rhizoctonia solani and Pythium aphanidermatum) pathogens when expressed in tobacco plants (Ohshima et al., 1999; Mitsuhara et al., 2000). Aly and coworkers observed that transgenic tobacco plants expressing sarcotoxin IA under the control of the constitutive, root-specific TobRB7 promoter (Yamamoto et al., 1991) had reduced parasitism by Orobanche (Aly R., Plakhin D., Achdari G. and Banner, I., unpublished). However, resistance in these plants was incomplete, perhaps due to the low level of expression driven by the TobRB7 promoter. Independently, Westwood et al. (1998) used GUS (b-glucuronidase) fusions with the promoter from HMG2, a defense-specific isogene of 3-hydroxy-3-methylglutaryl CoA reductase from tomato, to show that O. aegyptiaca parasitism induced expression of this promoter in transgenic tobacco. The HMG2 promoter exhibits a number of desirable features with respect to engineering parasite resistance in that it is induced immediately following penetration of the host root and expression occurs specifically in the area immediately surrounding the point of attachment, continuing for at least the first four weeks of parasite development. Activation in response to O. aegyptiaca parasitism may be related to its role in providing isoprenoid intermediates for phyto-

alexin synthesis, as HMG2 expression is induced in response to wounding and pathogen invasion (Chappell, 1995). Expression of HMG2-driven transgenes in the absence of wound/pathogen/parasite stimuli should be limited because the HMG2 promoter is not active in healthy tissues, with the specific exceptions of cotyledons, trichomes of young leaves, sites of lateral root initiation, and developing anthers (Cramer et al., 1993). In order to demonstrate the effect of sarcotoxin IA on Orobanche, we fused the sarcotoxin IA gene to the HMG2 promoter and generated transgenic tobacco plants. The ability of Orobanche to parasitize and grow on these plants was assayed under soil- and hydroponic-based growth conditions.

Materials and methods Vector construction and plant transformation A 189 bp XbaI – SstI fragment of the sarcotoxin IA (SARCO) gene (Matsumoto et al., 1986, Genbank accession No. M25612) containing the signal peptide (SSP) was excised from a pET-3 vector (courtesy of E. Galun, Weizmann Institute, Israel) and ligated to the 30 end of the 450 bp HMG2 promoter (courtesy of CropTech Corp., Blacksburg, VA) in a pBC plasmid (Figure 1(a)). The identity, orientation, and junctions of the resulting HMG2:SARCO construct were confirmed by DNA sequencing. The gene construct was subcloned into the pBIBhyg vector (Becker, 1990) using HindIII and SstI restriction sites and the construct was introduced into Agrobacterium tumefaciens strain LBA4404. Tobacco (Nicotiana tabacum cv. Xanthi) was transformed as described by MedinaBolivar and Cramer (2004). Nine independent lines were regenerated for the HMG2:SARCO construct. Plants were grown to seed and the progeny (T1 plants) were used in subsequent analyses. DNA extraction and PCR analysis Seeds of transformed plants were surfacesterilized and germinated on MS medium containing 50 mg l)1 hygromycin and 500 mg l)1 carbenicillin. Surviving plants were transplanted to Metro-Mix 360 potting medium (Scotts-Sierra

229 Horticultural, Marysville, OH) and the presence of the transgene was confirmed by PCR. Genomic DNA was extracted as described by Edwards et al. (1991) and PCR was performed using the primers, 50 -AAGTCCAGCGCGGCAACCGC-30 and 50 -CTAGAGCTCTCAACCTCTGGCTGTAGCAGC-30 , which anneal within the HMG2 promoter and at the 30 end of the sarcotoxin IA gene, respectively, yielding a 586 bp fragment. PCR conditions consisted of 94 C for 4 min, 35 cycles of 94 C for 2 min, 62 C for 1 min, and 72 C for 1 min, and a 72 C final extension for 7 min. Reaction products were fractionated on a 1.8% (w/v) agarose gel and visualized with ethidium bromide. DNA blot hybridization

Figure 1. Integration and expression of the sarcotoxin IA gene in transgenic tobacco lines. (a) Map of the construct and binary vector used in transformations. (b) PCR amplification of the transgene using primers specific to HMG2 and SARCO showing a product around the expected 586 bp size in L03 and L07 but not the non-transformed (NT) line. (c) DNA blot hybridization showing transgene incorporation and copy number in the tobacco genome. The probe was made to the hygromycin resistance gene. (d) Gene-specific RT-PCR detection of sarcotoxin IA mRNA from wounded leaves; ()) PCR of total RNA without reverse-transcriptase; (+) PCR of cDNA made from total RNA with reverse-transcriptase. (e) Protein immunoblot of total protein from Orobanche parasitized roots of transformed lines and a non-transformed line using sarcotoxin IA-specific antibody. Lane labeled Sarco contains 45 ng of sarcotoxin IA peptide.

DNA blot analysis was performed on total genomic DNA from leaves of the HMG2:SARCOtransformed plants to confirm integration into the genome and determine transgene copy number. Genomic DNA (30 lg) was digested overnight at 37 C with HindIII, which cuts once in the T-DNA region (Figure 1(a)). The digested DNA was fractionated on a 0.8% agarose gel and transferred to a Hybond–N+ charged nylon membrane (Amersham Bioscience, Piscataway, NJ) according to Sambrook and Russell (2001). The hygromycin phosphotransferase (hpt) gene (1025 bp) used as a probe was amplified from the pBIBhyg vector by PCR, and 75 ng of the gel-purified PCR product was labeled using the Prime-It Random Primer Labeling Kit (Stratagene, La Jolla, CA) with 32P-labeled dCTP (PerkinElmer Life Sciences, Boston, MA). Hybridization was conducted overnight at 65 C in Denhardt’s reagent (Denhardt, 1966), followed by triplicate washes at 65 C in 0.1X SSC and 0.1% SDS (Sambrook & Russell, 2001). Signal was detected using Kodak X-Omat AR-5 Scientific Imaging Film (Eastman Kodak, Rochester, NY). RNA extraction and RT-PCR of the sarcotoxin IA gene The HMG2 promoter was induced by wounding fully-expanded leaves of transgenic plants by passage through a pasta-maker. The resulting leaf strips (2 mm wide) were incubated for 6 h at room temperature. Total RNA was extracted

230 from the wounded leaves with RNeasy Plant Mini Kit (QIAGEN, Valencia, CA) according to the manufacturer’s directions. Total RNA (1 lg) was reverse-transcribed with an oligo dT primer using SuperscriptTM RNase H) Reverse Transcriptase according to the manufacturer’s instructions (Invitrogen Life Technologies, Carlsbad, CA). The resulting cDNA was used as a template in a PCR to amplify the sarcotoxin IA gene using the sarcotoxin-specific primers, 50 -GCAGGTACCATATGAATTTCCAGAAC-30 and 50 -CTGAGCTATACCCAAACCTTGTATG30 . PCR conditions consisted of 94 C for 4 min, 35 cycles of 94 C for 1 min, 48 C for 30 s, and 72 C for 1 min, and a 72 C final extension for 7 min. The expected 167 bp product was visualized on a 0.8% gel stained with ethidium bromide. Immunoblot analysis Transgenic and non-transgenic tobacco plants were cultivated in a hydroponic growth system comprised of glass fiber sheets held in polyethylene bags (Westwood, 2000), and were inoculated with O. aegyptiaca seeds. Thirty-five days after inoculation, total soluble proteins were extracted as described by Ohshima et al. (1999) from 100 mg of transgenic (lines L03 and L07) and non-transgenic tobacco roots adjacent to the parasite penetration site (1.5 cm from each side of the parasite tubercles including the attached parasite tubercles). Proteins were separated by tricine–SDS–PAGE as described by Scha¨gger and von Jagow (1987). After semi-dry electroblotting (Kyhse-Anderson, 1984) onto a nylon membrane (Hybond N+, Amersham Bioscience, Piscataway, NJ) the immobilized sarcotoxin IA peptide was probed using anti-sarcotoxin IA rabbit polyclonal antibodies (1:1000 dilution) (Ohshima et al., 1999). Sarcotoxin IA peptide was detected with alkaline phosphatase-conjugated anti-rabbit IgG (1:10,000) (Sigma-Aldrich Israel, Rehovot) as the secondary antibody. Purified sarcotoxin IA peptide (45 ng) (Mitsuhara et al., 2000) was used as a positive control. Assays for resistance to Orobanche Transgenic (T1) tobacco seeds were surfacesterilized and grown on MS selection medium

(50 mg l)1 hygromycin and 500 mg l)1 carbenicillin). Antibiotic-resistant seedlings were confirmed for the presence of the transgene by PCR as described above, and then analyzed for resistance to Orobanche. Non-transformed plants were treated similarly except that they were grown on media lacking antibiotics. Plants were evaluated for resistance to Orobanche using two experimental systems, inoculated soil and the polyethylene bag system. Orobanche seeds used were collected in Israel and the same seed lot was used for all experiments. Tests in soil For experiments in the US, transgenic and non-transgenic tobacco plants were initially transplanted from selection medium to small pots containing Metro-Mix 360 potting medium without Orobanche seeds. Two to three weeks later, plants were transplanted into 11 cm · 11 cm plastic pots containing potting medium inoculated with O. aegyptiaca or O. ramosa seeds (400 mg l)1) in a layer occupying the middle half of each pot, with the bottom and top quarters being filled with non-inoculated soil to minimize the chance of escape of the seeds (work with Orobanche in the US is regulated by USDA quarantine procedures). The pots were watered as needed and fertilized one time with 5 g Osmocote slow-release fertilizer (Scotts-Sierra Horticultural Products). Plants were held in a greenhouse under ambient sunlight at a temperature of 26 ± 4 C. Shortly after the emergence of Orobanche shoots (approximately 7 weeks later) tobacco plants were harvested by removing shoots and washing medium from the roots to determine parasite numbers and fresh weights. Tobacco shoot dry weights were determined after drying for 24 h at 80 C. The experimental design was a randomized complete block with three replications of each line for O. aegyptiaca and five replications of each line for O. ramosa. The transgenic lines were compared to non-transformed lines using Student’s t-test. Because Orobanche is endemic in Israel, challenge experiments do not require quarantine conditions and can be conducted on a larger scale. Therefore, for experiments conducted at Newe Ya’ar, 10 plants of each transgenic and nontransformed line were transplanted into 10-l pots filled with soil (light-medium clay with 63% sand,

231 12% silt and 22% clay) that was artificially inoculated with O. aegyptiaca seeds (40 mg kg)1 soil). Plants were arranged in a random block design, grown in a greenhouse under natural lighting with an average of 14 h daylight and a temperature of 27 ± 3 C, and watered and fertilized as needed. Sixty-five days after planting, the tobacco plants were removed from pots, the roots washed carefully to remove soil, and the numbers and dry weights of parasites on each host plant determined. Tests in polyethylene bags Transgenic tobacco were transplanted into the polyethylene bag system and kept moist with halfstrength nutrient solution (Hoagland & Arnon, 1950). Plant growth conditions were 25 C, with 14 h light at 100 lE s)1 m)2). Surface-sterilized O. aegyptiaca seeds were applied on and around the host roots with care taken to achieve even inoculation of all plant roots. After allowing 7 days for parasite seed preconditioning, 10 ml of 2 mg l)1 GR-24, a germination stimulant (Mangnus et al., 1992), was added to each bag in order to synchronize germination of Orobanche seeds. After 2–3 weeks, parasitism was evaluated by counting the number of live and dead tubercles on each plant using a binocular microscope. Dead tubercles are characterized by dark color and necrotic tissues. Each line was replicated 10 times and data were analyzed using JMP software (ver 4.0.3, SAS Institute Inc., Cary, NC). Averages were compared by Student’s t-test. Data discussed is from one experiment conducted in Israel, although identical experiments conducted in Israel and the U.S. supported these results.

Results and Discussion Generation of sarcotoxin IA-expressing tobacco plants Nine independent lines of transgenic tobacco containing HMG2:SARCO were developed through Agrobacterium-mediated transformation. The resulting plants appeared normal and were fertile. Because anti-sarcotoxin IA antibodies were not available when the first transformants were regenerated, transgenic lines were initially screened based on presence of the transgene and then on biomass accumulation while growing in

O. aegyptiaca-inoculated soil. Lines L03 and L07 were thus selected for further characterization. The presence of the HMG2:SARCO construct in these lines was confirmed by PCR using specific primers spanning the HMG2 promoter and sarcotoxin IA coding region (Figure 1(b)). To confirm stable integration of the transgene into the tobacco genome and estimate transgene copy number, DNA hybridization analysis was conducted using a probe specific to the hpt gene. This probe was used because at 1 kb in length it provided a stronger signal than a smaller probe made to the sarcotoxin IA gene. The hpt probe detected one transgene copy in line L03 and at least three copies in L07 (Figure 1(c)). Expression of the transgene was verified by reverse-transcriptase (RT)-PCR of total RNA from transgenic leaves that had been wounded for 6 h to induce expression of the HMG2 promoter. Reactions performed using sarcotoxin IAspecific primers yielded a product of the expected size (186 bp) with the transformed lines, but not with the non-transformed control (Figure 1(d)). Confirmation that the PCR product resulted from cDNA template rather than genomic DNA contamination was provided by PCR on the same RNA preparations used for RT-PCR, but which had not been reverse-transcribed (-lanes in Figure 1(d)). Accumulation of the sarcotoxin IA peptide in lines L03 and L07 was verified by immuno-blot analysis of total proteins obtained from roots at sites of parasitism (Figure 1(e)). The band observed at 8.2 kDa is consistent with that reported from transgenic sarcotoxin IAexpressing tobacco by Mitsuhara et al. (2000) and likely represents a dimer of the 4.1 kDa sarcotoxin IA mature peptide. These results confirm the generation of HMG2:SARCO transgenic lines that produce sarcotoxin IA mRNA and protein. Impact of HMG2: sarcotoxin IA expression on host resistance to Orobanche In order to evaluate the effect of sarcotoxin IA on host resistance to parasitism by Orobanche, experiments were conducted in soil and in a polyethylene bag system. Although experiments in soil more closely approximate field conditions, the polyethylene bag system is valuable because it allows visualization of Orobanche tubercles during development.

232 Transformed tobacco lines L03 and L07 had significantly higher biomass accumulation as compared to non-transgenic tobacco lines when grown in soil inoculated with O. aegyptiaca (Figure 2(a)). This was confirmed in a similar experiment conducted in Israel, in which transgenic tobacco lines showed greater growth than nontransformed controls when challenged with O. aegyptiaca (Figure 2(c)). When L03 and L07 plants were grown in the absence of O. aegyptiaca, their growth was greater than that of the same lines growing in inoculated soil (Figure 2(c)), indicating that the transgene conferred an intermediate level of Orobanche-resistance between absolute resistance and complete susceptibility. However, the presence of the transgene

itself did not carry a productivity penalty for the tobacco, as we observed no differences in growth and appearance between transformed and nontransformed lines (Figure 2(c)). To quantify this observation, several transgenic lines were compared to non-transformed plants in non-inoculated soil, and no significant differences were detected in biomass accumulation. For example after three months of growth, L03 and non-transformed plants accumulated 18.4 g (SE 3.5) and 15.0 g (SE 1.1) dry weight, respectively (n ¼ 8). The effect of the transgene on parasite growth generally reflected an increased host resistance, although results were not entirely consistent. The numbers of O. aegyptiaca tubercles that established on transgenic hosts did not differ signifi-

Figure 2. Response of HMG2:SARCO transformed tobacco plants to O. aegyptiaca in a soil-based assay. Experiments were conducted in the US (a, b) and Israel (c, d). (a) Dry weight of host shoots from non-transformed (NT) and transformed lines L03 and L07. (b) Number and fresh weight of Orobanche plants attached to tobacco in (a). Bars are means of three replicates and vertical lines indicate SE. (c) Photo showing phenotypes of NT, L03, andL07 plants in non-inoculated soil ()Oro) and soil inoculated with O. aegyptiaca (+Oro). Arrows show O. aegyptiaca flowers emerged on NT tobacco. (d) Numbers and dry weights of O. aegyptiaca parasitizing NT, L03, and L07 lines from the experiment in (c). Bars represent means of 10 replicates and vertical lines indicate SE. For both experiments * and ** indicate means different from NT as determined by Student’s t-test with a ¼ 0.05 and 0.01, respectively.

233 cantly from those on non-transformed plants (Figure 2(b) and (d)). However, in the experiment conducted in Israel, parasite biomass was significantly lower on lines L03 and L07 compared to non-transformed controls (Figure 2(d)). This sharp decline in biomass of parasites suggests that sarcotoxin IA affects parasite growth after attachment has occurred, rather than inhibiting the attachment process itself. This effect was not observed in the experiment conducted in the US (Figure 2(b)), but this experiment may have been influenced by the high inoculation rate of O. aegyptiaca seeds (400 mg l)1 soil). This would represent an extreme O. aegyptiaca infestation, and may have approached the point of overwhelming the host’s ability to respond. In addition, US experiments must be harvested earlier than those in Israel to comply with quarantine restrictions on avoiding parasite reproduction, so late-emerging effects on parasite mortality could have been missed. To better understand the effect of sarcotoxin IA on the parasite, we studied parasitism of lines L03 and L07 in the polyethylene bag system. Transformed plants had higher proportions of necrotic and dead tubercles as compared to the non-transformed plants. Specifically, the mean proportion of necrotic tubercles on nontransformed plants was 0.6% (SE 0.3), whereas L03 and L07 had 41.4% (SE 4.9) and 67.1% (SE 9.4), respectively. This was a significant increase (t-test, a ¼ 0.01) in mortality rate and agrees with the soil-based experiment described above in which parasites attached to the roots of both transgenic and non-transformed lines, but a large proportion of the parasites on transgenic lines subsequently died. These experiments support a post-attachment mechanism of sarcotoxin IA action, which is consistent with expectations given that the HMG2 promoter controlling sarcotoxin IA is highly expressed only after the parasite has penetrated the host root (Westwood et al., 1998). One of the potential concerns with engineering Orobanche resistance is that the resistance mechanism may not be effective against all races of the parasite. This has been documented for sunflower resistance to O. cumana, in which new resistant varieties have been repeatedly overcome by races of the parasite (Alonso, 1998; Tang et al., 2003). Although no such races of O. ae-

gyptiaca have been reported, we tested a closely related species, O. ramosa, to evaluate the robustness of the sarcotoxin IA-mediated resistance. O. ramosa shares much in common with O. aegyptiaca in that it is also an important parasite of Solanaceous crops and is among the most destructive of Orobanche species (Parker and Riches, 1993). Transgenic sarcotoxin IA-expressing tobacco lines exhibited resistance to O. ramosa that was similar to that observed in experiments with O. aegyptiaca. Line L03, in particular, showed greater biomass accumulation in the presence of the parasite (Figure 3(a)). L07 and L03 plants also showed a 50 and 90% reduction, respectively, in fresh weight or O. ramosa shoots compared to those growing on non-transformed plants and L03 had significantly fewer tubercles (Figure 3(b)). Equally informative were the phenotypes of tubercles growing on the transgenic lines. The majority of tubercles parasitizing L03 and L07 plants were small, dark, necrotic, or malformed (Figure 3(b)), in contrast to tubercles attached to non-transformed plants, which were light in color, apparently healthy, and in several cases, produced floral shoots that emerged from the soil by the end of the experiment. Some parasites growing on transgenic plants were also able to develop to the point of producing floral shoots (e.g., inset photo of O. ramosa on L07 in Figure 3(b)), but these were few in number and may represent escapes from the effect of sarcotoxin IA. The interaction between host and parasite is complex, and is influenced by the host size, the number and size of the parasites, and other factors that affect relative sink strengths in host and parasite (Graves, 1995), but even a few surviving parasites appear to impact growth of the host as observed with L07 (Figure 3(a)). The response of transgenic lines to O. ramosa and O. aegyptiaca was similar, with an overall stronger effect in L03 than L07, suggesting that this mechanism may have broad application across Orobanche species. This research indicates that the HMG2: SARCO construct confers an intermediate level of resistance to Orobanche. Although the lines described here fall short of the resistance levels that would be required for reducing Orobanche infestations in the field, sarcotoxin IA represents a new lead in the search for parasite resistance and understanding the mechanism of action of

234

Figure 3. Response of transgenic HMG2:SARCO plants to O. ramosa in soil. (a) Dry weight of host shoots from NT and transgenic lines L03 and L07. (b) Number and fresh weight of O. ramosa plants attached to tobacco in A. Inset photos show phenotypes of parasites from one representative plant of the indicated lines. Bars are means of five plants with vertical lines indicating SE. * and ** indicate means different from NT as determined by Student’s t-test with a ¼ 0.05 and 0.01, respectively.

this peptide against Orobanche will facilitate optimization of this approach. Sarcotoxin IA preferentially disrupts bacterial membranes more than eukaryotic membranes (Nakajima et al., 1987; Iwai et al., 1993), but Orobanche is an angiosperm so there is no obvious reason why sarcotoxin IA should be more toxic to the parasite than to the host. Two formal possibilities could explain this selectivity. First, sarcotoxin IA could act within the host root to induce host defenses or otherwise disrupt host cells such that they cease to effectively transmit nutrients to the parasite. Alternatively, sarcotoxin IA may be translocated into the parasite to disrupt Orobanche plasma membranes or even move into Orobanche cells to disrupt organelle membranes. To address this latter possibility we have studied apoplastic movement of macromolecules from host to parasite and documented the transfer of proteins up to 27 kDa and fluorescent

dextrans up to 70 kDa, both of which greatly exceed the 4 kDa size of sarcotoxin IA (Hamamouch et al., unpublished data). The sarcotoxin IA used contains the signal peptide to direct the protein to the extracellular space where it would likely move into the parasite with the bulk flow of water. Although we have no data on the movement of sarcotoxin IA itself, Orobanche plants are strong sinks (Aber et al., 1983; Nandula et al., 1999) so the potential exists for accumulation of macromolecules such as sarcotoxin IA in Orobanche tissues. Regardless of the site of sarcotoxin IA localization, its inhibitory effect is likely concentrationdependent. Ohashi and colleagues (Okamoto et al., 1998; Ohshima et al., 1999; Mitsuhara et al., 2000) have identified promoter strength and sarcotoxin IA stability as key features in obtaining sufficient accumulation of the peptide to confer bacterial and fungal resistance on

235 transgenic tobacco. The HMG2 promoter used in the current study appears to be strongly induced by Orobanche parasitism, although its expression is restricted to a small region of host root around the point of parasite entry (Westwood et al., 1998). We have characterized additional promoters for induction by Orobanche that could be used to supplement HMG2-driven sarcotoxin IA expression (Griffitts et al., 2004). Identification of such parasite-inducible promoters is important because our work has already demonstrated that neither the Cauliflower Mosaic Virus (CaMV35S) nor the pathogenesis-related (PR-1a) gene promoters used in the generation of sarcotoxin IAmediated pathogen-resistant tobacco (Ohshima et al., 1999; Mitsuhara et al., 2000) are highly expressed in Orobanche-parasitized tobacco roots (Westwood et al., 1998; Griffitts et al., 2004). The issue of peptide stability is equally important. Sarcotoxin IA appears to be highly susceptible to proteases present in the intercellular fluid of tobacco plants (Mitsuhara, personal communication), so it may be degraded before accumulating to toxic levels. Instability of sarcotoxin IA has been overcome by fusing it to another protein such as GUS (Okamoto et al., 1998). Thus, several possibilities exist to increase sarcotoxin IA-mediated Orobanche resistance above the current levels. This research describes a strategy for genetically engineering enhanced resistance to a parasitic plant. Considering the importance of parasitic weeds to world agriculture and the difficulty in obtaining resistance by conventional methods, this represents a significant advance. It is noteworthy that researchers have screened over one thousand lines of tobacco (including related species and mutated populations) for resistance to O. aegyptiaca, O. ramosa, and O. cernua over the past 30 years with almost no success (Alonso, 1998). Even the few promising lines identified were parasitized to some extent, for example the most resistant variety claimed by one team still had 13–61% of plants parasitized (Raju, 1996). In this context of partial resistance, the effect of the sarcotoxin IA-expressing plants on Orobanche is remarkable because it was conferred by the addition of just a single gene, and suggests that new mechanisms of parasite resistance are possible. We are still at the beginning of understanding how sarcotoxin IA acts on the parasite, and

further research will be required to understand and fully optimize this mechanism of resistance.

Acknowledgements We thank E. Galun for the sarcotoxin IA gene and I. Mitsuhara for the synthetic sarcotoxin IA and anti-sarcotoxin IA antibody. Thanks to Jeannine Flagg and Achdari Guy for technical assistance. This research was supported by Research Grant No. IS-3048-98 from BARD, The United States – Israel Binational Agricultural Research and Development Fund. Additional support was provided by USDA NRI award 2001-35320-10900 to J. H. W. and C. L. C., and USDA Hatch Project Nos. 135657 to J. H. W. and 129016 to C. L. C.

References Aber M, Fer A and Salle´ G (1983) Etude du transfert des substances organiques de l’hoˆte (Vicia faba) vers le parasite (Orobanche crenata Forsk.). Z Pflanzenphysiol Bd 112: 297– 308. Alonso LC (1998) Resistance to Orobanche and resistance breeding: a review. In: Wegmann K, Musselman LJ and Joel DM (eds), Proceedings of the Fourth International Workshop on Orobanche (pp. 233–257) Institute for Wheat and Sunflower ‘‘Dobroudja’’, Albena, Bulgaria. Aly R, Granot D, Mahler-Slasky Y, Halpern N and Galun E (1999) Saccharomyces cerevisiae cells, harboring the gene encoding Sarcotoxin IA secrete a peptide that is toxic to plant pathogenic bacteria. Protein Expression and Purification 16: 120–124. Becker D (1990) Binary vectors which allow the exchange of plant selectable markers and reporter genes. Nucl Acids Res 18: 203. Chappell J (1995) Biochemistry and molecular biology of the isoprenoid biosynthetic pathway in plants. Annu Rev Plant Physiol Plant Mol Biol 46: 521–548. Cramer CL, Weissenborn DL, Cottingham CK, Denbow CJ, Eisenback JD, Radin DN and Yu X (1993) Regulation of defense-related gene expression during plant-pathogen interactions. J Nematol 25: 507–518. Cubero JI (1991) Breeding for resistance to Orobanche species: a review. In: Wegmann K and Musselman LJ (eds), Progress in Orobanche Research (pp. 257–277) EberhardKarls-Univrersitat, Tubingen, FRG. Denhardt DT (1966) A membrane filter technique for determination of compementary DNA. Biochem Biophys Res Com 23: 641–646. Edwards K, Johnstone C and Thompson C (1991) A simple and rapid method for the preparation of plant genomic DNA for PCR analysis. Nucleic Acids Res 19: 1349. Ejeta G, Butler LG, Hess DE and Vogler RK (1991) Genetic and breeding strategies for Striga resistance in sorghum. In:

236 Ransom JK, Musselman LJ, Worsham AD and C. P (eds), Proceedings of the 5th International Symposium of Parasitic Weeds (pp. 539–544) CIMMYT, Nairobi, Kenya. Foy CL, Jain R and Jacobsohn R (1989) Recent approaches for chemical control of broomrape (Orobanche spp.). Rev Weed Sci 4: 123–152. Graves JD (1995) Host-plant responses to parasitism. In: Press MC and Graves JD (eds), Parasitic Plants (pp. 206– 225) Chapman & Hall, London. Griffitts AA, Cramer CL and Westwood JH (2004) Host Gene Expression in Response to Egyptian Broomrape (Orobanche aegyptiaca). Weed Sci 52: 697–703. Hoagland DR and Arnon DI (1950) The water-culture method for growing plants without soil. Calif Agric Exp Sta Circ 347: 1–32. Iwai H, Nakajima Y, Natori S, Arata Y and Shimada I (1993) Solution conformation of an antibacterial peptide, sarcotoxin IA, as determined by H-NMR. Eur J Biochem 217: 639–644. Jacobsohn R (1994) The broomrape problem in Israel and an integrated approach to its control. In: Pieterse AH, Verkleij JAC and ter Borg SJ (eds), Biology and Management of Orobanche, Proc 3rd Int Workshop on Orobanche and Related Striga Res (pp. 652–658) Royal Tropical Inst., Amsterdam, The Netherlands. Joel DM, Kleifeld Y, Losner-Goshen D and Gressel J (1995) Transgenic crops against parasites. Nature 374: 220–221. Kanampiu FK, Ransom JK, Friesen D and Gressel J (2002) Imazapyr and pyrithiobac movement in soil and from maize seed coats to control Striga in legume intercropping. Crop Prot 21: 611–619. Kyhse-Anderson J (1984) Electroblotting of multiple gels: A simple apparatus without buffer tank for rapid transfer of protein from polacrylamide to nitrocellulose. J Biochem Biophys Meth 10: 203–209. Mangnus EM, Stommen PLA and Zwanenburg B (1992) A standardized bioassay for evaluation of potential germination stimulants for seeds of parasitic weeds. J Plant Growth Regul 11: 91–98. Matsumoto N, Okada M, Takahashi H, Ming QX, Nakajima Y, Nakanishi Y, Komano H and Natori S (1986) Molecular cloning of a cDNA and assignment of the C-terminal of sarcotoxin IA, a potent antibacterial protein of Sarcophaga peregina. Biochem J 239: 717–722. Medina-Bolivar F and Cramer CL (2004) Production of recombinant proteins in hairy roots cultured in plastic sleeve bioreactors. In: Balbas P and Lorence A (eds), Recombinant Gene Expression: Reviews and Protocols, second edition (pp. 351–363) Humana Press, Totowa, NJ. Mitsuhara I, Matsufuru H, Ohshima M, Kaku H, Nakajima Y, Murai N, Natori S and Ohashi Y (2000) Induced expression of sarcotoxin IA enhanced host resistance against both bacterial and fungal pathogens in transgenic tobacco. Mol Plant Microbe Interact 13: 860–868.

Nakajima Y, Qu XM and Natori S (1987) Interaction between liposomes and sarcotoxin IA, a potent antibacterial protein of Sarcophaga peregrina (flesh fly). J Biol Chem 262: 1665–1669. Nandula VK, Foy CL and Orcutt DM (1999) Glyphosate for Orobanche aegyptiaca control in Vicia sativa and Brassica napus. Weed Sci 47: 486–491. Ohshima M, Misuhara I, Okamoto M, Sawano S, Nishiyama K, Kaku H, Natori S and Ohashi Y (1999) Enhanced resistance to bacterial diseases of transgenic tobacco plants overexpressing sarcotoxin IA, a bactericidal peptide of insect. J Biochem 125: 431–435. Okamoto M, Mitsuhara I, Ohshima M, Natori S and Ohashi Y (1998) Enhanced expression of an antibiotic peptide sarcotoxin IA by GUS fusion in transgenic tobacco plants. Plant Cell Physiol 39: 57–63. Parker C and Riches CR (1993) Parasitic Weeds of the World: Biology and Control. pp. 111–164. CAB International, Wallingford. Raju CA (1996) Variability in tobacco germplasm towards Orobanche infection. In: Moreno MT, Cubero JI, Berner D, Joel DM, Musselman LJ and Parker C (eds), Advances in Parasitic Plant Research (pp. 609–613) Junta de Andalucia, Cordoba, Spain. Sambrook J and Russell DW (2001) Molecular Cloning, 3rd ed. pp. 6.51–6.55. Cold Springs Harbor Laboratory Press, Cold Spring Harbor, N.Y. Scha¨gger H and von Jagow G (1987) Tricine-sodium dodecyl sulfate-polyacrylamide gel electrophoresis for the separation of proteins in the range from 1 to 100 kDa. Anal Biochem 166: 368–379. Surov T, Aviv D, Aly R, Joel DM, Goldman-Guez T and Gressel J (1998) Generation of transgenic asulam-resistant potatoes to facilitate eradication of parasitic broomrapes (Orobanche spp.), with the sul gene as the selectable marker. Theor Appl Genet 96: 132–137. Tang S, Heesacker A, Kishore VK, Fernandez A, Sadik ES, Cole G and Knapp SJ (2003) Genetic mapping of the Or5 gene for resistance to Orobanche race E in sunflower. Crop Sci 43: 1021–1028. Westwood JH (2000) Characterization of the OrobancheArabidopsis system for studying parasite-host interactions. Weed Sci 48: 742–748. Westwood JH, Yu X, Foy CL and Cramer CL (1998) Expression of a defense-related 3-hydroxy-3-methylglutaryl CoA reductase gene in response to parasitization by Orobanche spp. Mol Plant-Microbe Interact 11: 530–536. Yamamoto YT, Taylor CG, Acedo GN, Cheng C-L and Conkling MA (1991) Characterization of cis-acting sequences regulating root-specific gene expression in tobacco. Plant Cell 3: 371–382.