Plant Cell Rep (2009) 28:1399–1409 DOI 10.1007/s00299-009-0739-6
ORIGINAL PAPER
Phenotypic and molecular evaluation of cotton hairy roots as a model system for studying nematode resistance Martin J. Wubben Æ Franklin E. Callahan Æ Barbara A. Triplett Æ Johnie N. Jenkins
Received: 24 March 2009 / Revised: 17 June 2009 / Accepted: 21 June 2009 / Published online: 4 July 2009 Ó Springer-Verlag 2009
Abstract Agrobacterium rhizogenes-induced cotton (Gossypium hirsutum L.) hairy roots were evaluated as a model system for studying molecular cotton–nematode interactions. Hairy root cultures were developed from the root-knot nematode (RKN) (Meloidogyne incognita [Kofoid and White] Chitwood, race 3)-resistant breeding line M315 and from the reniform nematode (RN) (Rotylenchulus reniformis Linford & Oliveira)-resistant accession GB713 (G. barbadense L.) and compared to a nematode-susceptible culture derived from the obsolete cultivar DPL90. M315, GB713, and DPL90 hairy roots differed significantly in their appearance and growth potential; however, these differences were not correlated with transcript levels of the A. rhizogenes T-DNA genes rolB and aux2 which help regulate hairy root initiation and proliferation. DPL90 hairy roots were found to support both RKN and RN reproduction in tissue culture, whereas M315 and GB713 hairy roots were resistant to RKN and RN, respectively. M315 hairy roots showed constitutive up-regulation of the defense gene MIC3 (Meloidogyne Induced Cotton3) compared to M315 whole-plant roots and
Mention of trade names or commercial products in this article is solely for the purpose of providing specific information and does not imply recommendation or endorsement by the United States Department of Agriculture. Communicated by J. Register. M. J. Wubben (&) F. E. Callahan J. N. Jenkins Crop Science Research Laboratory, USDA/ARS, Mississippi State, MS 39762, USA e-mail:
[email protected] B. A. Triplett Southern Regional Research Center, USDA/ARS, New Orleans, LA 70124, USA
DPL90 hairy roots. Our data show the potential use of cotton hairy roots in maintaining monoxenic RKN and RN cultures and suggest hairy roots may be useful in evaluating the effect of manipulated host gene expression on nematode resistance in cotton. Keywords Agrobacterium rhizogenes Cotton Hairy roots Root-knot nematode Reniform nematode Plant defense Resistance
Introduction The root-knot nematode (RKN) (Meloidogyne incognita [Kofoid and White] Chitwood, race 3) and the reniform nematode (RN) (Rotylenchulus reniformis Linford & Oliveira) are obligate, biotrophic root parasites of Upland cotton (Gossypium hirsutum L.). During 2005, RKN and RN infection accounted for a combined loss of 1,087,535 cotton bales having an estimated US market value of $232 million (Blasingame 2006). Currently, US cotton producers employ genetic resistance, nematicide application, and rotation with non-host crops in an effort to limit nematoderelated yield losses (Starr et al. 2007). Undoubtedly, the most economical and environmentally friendly means of nematode control would be through the use of resistant cotton cultivars (Starr et al. 2007). The commercial cultivar ‘Acala NemX’ provides moderate resistance to RKN field populations in California (Oakley 1995); however, no equivalent cultivars are available that are adapted to other geographic areas in the US (Creech et al. 1998). Cotton germplasm lines resistant to RKN have been available for breeding purposes for a number of years (Creech et al. 1998; Shepherd et al. 1996); however, the multigenic nature of the resistance and the difficulty in accurately scoring
123
1400
the resistant phenotype in segregating populations has hampered the transfer of this resistance into a commercial elite cultivar by traditional breeding methods. Similar to RKN, there are currently no cultivars available that significantly suppress RN field populations (Robinson 2007); however, it is known that accessions of the diploid cotton G. longicalyx (Hutch & Lee) are immune to RN (Yik and Birchfield 1984) and substantial progress has been made in introgressing this immunity into tetraploid cotton (Robinson et al. 2007). Resistance has also been identified in accessions of G. barbadense (L.) (Robinson et al. 2004). As an alternative to resistance, a small number of cultivars tolerant to RN have been identified which show some promise in minimizing yield loss under high RN infection conditions (Stetina et al. 2009). As obligate biotrophs, successful RKN and RN parasitism is contingent upon the establishment of a specialized feeding site within the host root from which the nematode can ingest nutrients necessary for survival. In the case of RKN, this process involves the formation of hypertrophied, multinucleate ‘giant cells’ that are initiated through the action of esophageal gland secretions produced by the infective second-stage juvenile (J2) nematode (Davis et al. 2004; Williamson and Hussey 1996). As opposed to giant cells, RN forms a syncytium within the host root (Robinson 2007). The syncytium begins as a single initial feeding cell, usually an endodermal cell, that eventually fuses with neighboring cells by partial cell wall dissolution (Robinson 2007). As in the case with giant cell formation, syncytium initiation and development are mediated by esophageal gland secretions that are injected directly into or in proximity to the host root cells through the stylet (Davis et al. 2004; Williamson and Hussey 1996). A considerable amount of effort has been expended toward identifying RKN and cyst nematode (Heterodera spp.) genes that encode esophageal gland proteins or nongland proteins that are essential for metabolism, development, or reproduction (Alkharouf et al. 2007; Gao et al. 2003; Huang et al. 2003). Consequently, there has been significant progress in the development of transgenic host plant resistance to RKN and cyst nematodes through the in planta overexpression of dsRNA constructs that silence a target nematode gene through RNA interference (Fairbairn et al. 2007; Huang et al. 2006a; Yadav et al. 2006). In addition, genome-wide expression profiling experiments have identified numerous plant genes that may be involved in feeding site formation or that potentially mediate host resistance to nematode infection (Ithal et al. 2007; Jammes et al. 2005; Schaff et al. 2007). Despite the progress that has been made in understanding plant parasitism by sedentary, endoparasitic nematodes, there remains little information regarding compatible or incompatible cotton–nematode interactions
123
Plant Cell Rep (2009) 28:1399–1409
at the molecular level. In fact, only one cotton gene, MIC3 (Meloidogyne Induced Cotton3), has been identified as being induced within immature galls formed on resistant plants following RKN infection (Callahan et al. 1997, 2004; Wubben et al. 2008; Zhang et al. 2002). Molecular studies of cotton–nematode interactions, particularly studies designed to elucidate host gene function, are hampered by difficulties in whole-plant cotton transformation and by the root-parasitic nature of the pathogen. As a potential solution to these obstacles, Agrobacterium rhizogenesinduced hairy roots have been used extensively in studying plant–nematode interactions including nematode inoculum propagation (Savka et al. 1990), plant promoter analyses (Alpizar et al. 2006; Hansen et al. 1996; Mazarei et al. 2004; Preiszner et al. 2001; Wang et al. 2007), and gene functional analyses (Cai et al. 2003; Doyle and Lambert 2003; Gal et al. 2006; Huang et al. 2006b; Samuelian et al. 2004; Urwin et al. 1995). Infection of plant tissues by the gram-negative bacterium A. rhizogenes results in the proliferation of adventitious roots that originate at the site of infection and grow in an agravitropic manner. These transformed ‘‘hairy roots’’ harbor at least one copy of the T-DNA that originates from the root-inducing plasmid (reviewed by Nilsson and Olsson 1997). Recently, a protocol has been established for generating hairy roots from the allotetraploid cotton species G. hirsutum and G. barbadense (Triplett et al. 2008). Consequently, there is an opportunity to investigate whether cotton hairy roots can serve as a model system for studying cotton–nematode interactions, as exemplified by a number of other plant–nematode pathosystems. In addition to hairy roots, plant root cultures derived from excised root tips have been used to test for RKN resistance in tomato (Orion and Pilowsky 1984) and for maintaining monoxenic nematode cultures in onion and dandelion (Mitkowski and Abawi 2002). However, we determined that excised cotton root tips were not amenable to these protocols and failed to grow at all in the absence of A. rhizogenes transformation (unpublished results). In this report, we describe the phenotypic and molecular characterization of cotton hairy root cultures developed from a RKN-resistant breeding line (M315) and from a RN-resistant G. barbadense accession (GB713) in comparison to a culture derived from a susceptible control genotype. Our data show that (1) cotton hairy roots can be used for maintaining monoxenic RKN or RN cultures, (2) RKN resistance and RN resistance were retained in the M315 and GB713 hairy root cultures, respectively, and (3) hairy root cultures derived from independent transformation events can show significant variation in morphology, growth, and steady-state defense gene expression. We conclude that, while cotton hairy roots do not exactly mimic whole-plant root tissue, they nevertheless show
Plant Cell Rep (2009) 28:1399–1409
potential as a system for studying the effects of host transgene overexpression or silencing on the cotton–nematode interaction.
Materials and methods Plant material Delinted seeds of G. hirsutum (M315 RNR) and G. barbandense (GB713) were surface sterilized by washing in 10% bleach for 20 min followed by three consecutive washes with sterile Milli-Q water (Millipore, Bedford, MA, USA). Seeds were germinated in glass Magenta GA-7 vessels (Sigma-Aldrich, St. Louis, MO) on tissue culture medium containing Murashige–Skoog (MS) salts (Murashige and Skoog 1962) plus vitamins [Research Products International (RPI) Corp., Mt. Prospect, IL], 3% sucrose, and 0.8% Daishin agar (RPI Corp., Mt. Prospect, IL) at pH 5.8. Germination occurred under a 16 h light/8 h dark regimen at 27°C in a growth chamber. Fully expanded cotyledons generally developed 7–10 days after germination. Inoculation of cotton cotyledons and co-culture with A. rhizogenes Agrobacterium rhizogenes strain 15834 was ordered from the American Type Culture Collection (Manassas, VA) under APHIS permit #74117 and received as a freeze-dried pellet. The A. rhizogenes pellet was re-constituted with 400 lL Luria-Bertani (LB) broth, streaked-out on LB agar plates, and incubated at 28°C. A single colony was used to inoculate a 10 mL LB broth culture that was incubated at 28°C with shaking (200 rpm) for 4 days. The 10 mL culture was re-streaked on LB agar plates and four colonies were used for colony-polymerase chain reaction (PCR) using rolB-specific primers (forward 50 –30 : GCTCTTGCA GTGCTAGATTT, reverse 50 –30 : GAAGGTGCAAGCTA CCTCTC, product size = 423 bp) to confirm the identity of the culture as A. rhizogenes. 25% glycerol stocks of the 10 mL LB broth culture were made and stored at -70°C. Expanded cotton cotyledons were inoculated with A. rhizogenes using the method described by Triplett et al. (2008) with some modification. An A. rhizogenes glycerol stock was used to inoculate 10 mL of Yeast-ExtractPeptone (YEP) liquid medium (10 g/L bacto peptone, 10 g/L bacto yeast extract, 5 g/L sodium chloride, pH 6.1). The culture was incubated at 28°C with shaking (200 rpm) for 3 days. The entire 10 mL culture was then combined with an additional 240 mL of YEP liquid medium and incubated as before until the OD600 reached 0.25–0.3. The 250 mL culture was then divided among five sterile 50 mL tubes (Greiner Bio-One, Monroe, NC, USA) and the cells
1401
were pelleted by centrifugation (10 min @ 3,2209g). The supernatants were discarded and each bacterial pellet was resuspended in 9 mL MS salts ? vitamins liquid medium (minus sucrose). The five 9 mL cell suspensions were combined and the OD600 determined. The A. rhizogenes cell suspension was diluted to OD600 * 0.3 and 25 mL poured into a 100 9 25 mm Petri dish (Fisher Scientific, Pittsburgh, PA, USA). Cotton cotyledons were excised from seedlings with a sterile scalpel and submerged in the A. rhizogenes suspension. The Petri dish was placed inside a bell jar and a vacuum applied for 30 s. The A. rhizogenesinfiltrated cotyledons were then transferred to new Petri dishes (two cotyledons/dish) containing multiple pieces of Whatman filter paper (Whatman Inc., Florham Park, NJ, USA) that had been pre-soaked in MS salts ? vitamins liquid medium (minus sucrose). Plates were sealed with MicroporeTM surgical tape (3M, St. Paul, MN). Co-cultivation was performed in a growth chamber (16 h light/8 h dark, 27°C) for 3 days. Cotyledons were then washed for 20 min in MS salts ? vitamins liquid medium (minus sucrose) containing 500 lg/mL ampicillin. After washing, cotyledons were transferred to MS salts ? vitamins/3% sucrose agar plates (two cotyledons/plate) containing 500 lg/mL ampicillin, sealed with surgical tape, and placed back in the growth chamber. Hairy root culture establishment and growth comparisons Hairy roots began to emerge from cotyledons 7–9 days after A. rhizogenes infiltration. To remove residual A. rhizogenes contamination, hairy roots (*2 cm long) were excised from cotyledons with a sterile scalpel and transferred to new MS salts ? vitamins/3% sucrose agar plates containing 500 lg/mL ampicillin and placed in a 25°C incubator under continuous darkness. After 2 weeks, roots were transferred to new ampicillin plates and transferred again 2 weeks later. After 2 weeks of growth on the third round of ampicillin plates, roots were transferred to fresh MS salts ? vitamins/3% sucrose agar plates lacking ampicillin. Hairy root cultures were then maintained by transferring new growth to fresh plates every 4–6 weeks. Comparisons of hairy root culture growth were accomplished as follows: for each hairy root genotype examined, five white hairy roots, each *1.5 cm long, were excised from stock culture plates and transferred (arranged in parallel) to a fresh MS salts ? vitamins/3% sucrose agar plate. A total of five plates were measured for each hairy root genotype in each of two experiments. Plates were sealed with surgical tape and stored at 25°C for 40 days in an incubator under continuous darkness. All hairy root tissues were then removed from each plate, and the fresh
123
1402
weight (g) was determined. The mean ± SE fresh root weight of all five plates for each hairy root genotype was calculated and differences (P B 0.05) between the means were determined by t test. Nematode inoculations A RKN field population, originally acquired from Dr. Terry Kirkpatrick (Univ. Arkansas), was propagated on greenhouse-grown kenaf (Hibiscus cannabinus), cotton (G. hirsutum), or tomato (Lycopersicon esculentum) plants. RKN eggs were collected from plant roots by washing the roots in 1% sodium hypochlorite for 3 min and collecting the eggs on a 25 lm sieve. RKN eggs were purified from contaminating soil and plant tissue by sucrose floatation and were hatched to collect infective second-stage juvenile (J2) nematodes as described by Barker (1985). A RN field population, originally acquired from Dr. Forest Robinson (USDA-ARS), was propagated on cotton in a growth chamber (16 h light/8 h dark, 27°C). Vermiform RN was collected from infested sand using a Baermann funnel (Barker 1985). All nematodes used for hairy root inoculation were surface sterilized according to Baum et al. (2000). For nematode inoculation, 1.5 cm white hairy roots were excised from stock cultures and transferred to fresh MS salts ? vitamins/3% sucrose plates (4–5 roots per plate arranged in parallel, 4–5 plates total per experiment). Two days later, 20 lL of surface-sterilized nematodes, suspended in sterile Milli-Q water, was applied directly onto the tip of each root. Approximately, 100 J2/root tip was used for the RKN experiments and *300 vermiform nematodes/root tip was used for the RN experiments. Because only RN females infect host roots and male:female ratios are generally 1:1 (Robinson 2007), a greater number of RN was used for hairy root inoculation compared to the RKN experiments. After inoculation, plates were left uncovered in a laminar flow hood until the water had completely soaked into the medium. Plates were then sealed with MicroporeTM surgical tape. RKN-inoculated hairy roots were maintained in darkness at 25°C, while RN-inoculated hairy roots were incubated at 28°C. Acid fuchsin staining of RKN-infected hairy roots was accomplished according to Hussey (1990). Acid fuchsin staining of RN-infected hairy roots was performed using the same method with the tissue clearing step omitted. Photographs were taken with a Nikon Coolpix 4300 digital camera mounted onto a Nikon SMZ1000 stereomicroscope (Nikon Inc., Melville, NY, USA).
Plant Cell Rep (2009) 28:1399–1409
tube and flash frozen in liquid nitrogen. Tissue was ground using a plastic pestle and total RNA was extracted using the RNeasy Plant Mini-Kit (Qiagen, Valencia, CA, USA). Total RNA extraction from uninoculated whole-plant roots was accomplished as previously described (Wubben et al. 2008). Contaminating genomic DNA was removed using the DNA-free Kit (Ambion, Austin, TX, USA). First-strand cDNA was synthesized from 1 lg of DNase-treated total RNA using the SuperScript III First-Strand cDNA Synthesis System for RT-PCR (Invitrogen, Carlsbad, CA, USA). qRT-PCR was performed on an Applied Biosystems SDS7000 as previously described (Wubben et al. 2008). Relative transcript abundances of rolB (gb|X03433), aux2 (gb|DQ782955), and MIC3 (gb|AY072783.1) were normalized using the geometric mean relative starting quantities of G. hirsutum a-tubulin (gb|EV485208.1), ubiquitin conjugating protein (gb|ES800665.1), and RNA helicase (gb|ES804313.1) according to the method described by Vandesompele et al. (2002). For each experiment, mean ± SE rolB, aux2, and MIC3 relative transcript abundances were calculated from three independent biological samples and differences (P B 0.05) between hairy root lines were determined by t test. Forward (F) and reverse (R) primer sequences used for qRT-PCR were as follows (50 -30 ): RNA helicase (F-AGCCTGATCGATA TGTGGAGGGAT, R-TCAGGAAGGTTTGGCCATCTT GGA), ubiquitin conjugating protein (F-CGGAAAGA GGTGAAGATGTCAAC, R-GGATCTTGCTGCAACCT CTTAAA), a-tubulin (F-GATCTCGCTGCCCTGGAA, R-ACCAGACTCAGCGCCAACTT), rolB (F-AGTGACC AACGTTTACGGGAGAGA, R-GGTGCCGCAAGCTAC AACATCATA), aux2 (F-ATACTTGCAACATCCGGA CTGCGA, R-AGCTTTCCGACTGCCATCTACGAA), MIC3 (F-TTAGGGTTAAGTGGATTATTCTTTGG, R-T GGTTTTCGGTCGGAATGATCTTAGT). Total protein extraction and western blotting Total protein extraction from hairy roots and native plant roots, immunoblotting, and MIC3 protein detection were performed as described previously (Callahan et al. 2004). c-Tubulin was detected with anti-c-tubulin (Sigma, St. Louis, MO, USA) at 1:2,000 dilution and served as a loading control for the immunoblots.
Results
RNA extraction and quantitative reverse transcriptionPCR (qRT-PCR)
Morphological and molecular characterization of cotton hairy roots derived from nematode-resistant germplasm
Independent biological samples were defined as 3–5 white hairy roots that were collected into a 1.5 mL centrifuge
Agrobacterium rhizogenes strain 15834 was vacuum infiltrated into cotyledons of the RKN-resistant G. hirsutum
123
Plant Cell Rep (2009) 28:1399–1409
1403
Fig. 1 Morphology of hairy root cultures derived from the nematode-susceptible cultivar DPL90 (G. hirsutum), M315 and M315-X (RKN-resistant) (G. hirsutum), and GB713 (RN-resistant) (G. barbadense)
germplasm line M315 RNR (Shepherd et al. 1996) and into cotyledons of the RN-resistant accession G. barbadense 713 (GB713) (Robinson et al. 2004). Of 40 independent M315 hairy root lines initially selected, only two (‘‘M315’’ and ‘‘M315-X’’) continued to show growth after three rounds of selection on 500 lg/mL ampicillin plates to remove residual A. rhizogenes. Similarly, only 1 of 20 independent GB713 hairy root lines continued to show growth following A. rhizogenes decontamination. The obsolete G. hirsutum cultivar Deltapine90 (DPL90) is highly susceptible to both RKN and RN. A DPL90 hairy root culture had previously been developed by Triplett et al. (2008), and this culture was used in our experiments as a susceptible control. The absence of A. rhizogenes from the established M315, M315-X, and GB713 hairy root cultures was verified by PCR using primers specific to the A. rhizogenes virG locus and hairy root genomic DNA as template (data not shown).
The DPL90, M315, M315-X, and GB713 hairy root cultures exhibited the plagiotropic and agravitropic growth that is characteristically seen in A. rhizogenes-induced hairy roots; however, qualitative morphological differences between the cultures were observed (Fig. 1). DPL90 and GB713 hairy roots showed greater elongation than those formed by M315 and M315-X (Fig. 1). In addition, M315 and M315-X hairy roots were thinner and less robust than those formed by DPL90 and GB713 (Fig. 1). The relative growth rates of the cultures were different based on the fresh weight of the root tissue obtained after 40 days of growth (Fig. 2). DPL90 hairy roots were the most prolific followed by GB713 and M315 (Fig. 2). Hairy root initiation and proliferation are influenced by the expression of the A. rhizogenes rolB and aux2 genes which are integrated into the host plant genome as part of the TL-DNA and TR-DNA, respectively (Nilsson and Olsson 1997). To determine whether differences in growth
123
1404
Plant Cell Rep (2009) 28:1399–1409 1.8
4.5
a
4.0 3.5 3.0 2.5
c 2.0 1.5
b
1.0 0.5 0.0 DPL90
M315
GB713
Fig. 2 Growth comparisons between the DPL90, M315, and GB713 hairy root cultures. Presented are the mean ± SE hairy root fresh weight (g)/petri dish as calculated from five petri dishes for each culture starting with equivalent amounts of root material. Total hairy root fresh weight was measured after 40 days of growth. Means showing different letter designations are different at P B 0.05 as determined by two-sample t test
between the hairy root cultures could be explained by differential expression of rolB and/or aux2, the relative abundance of rolB and aux2 transcripts was determined for each culture by qRT-PCR. Transcript abundances of rolB and aux2 did not differ within each culture, with the exception of GB713 which lacked the aux2 gene (Fig. 3). DPL90 showed the highest levels of rolB and aux2 transcripts (Fig. 3) which seemed to correlate with its increased growth compared to the other cultures. The M315 and M315-X cultures showed similar levels of rolB and aux2 transcripts (Fig. 3). rolB and aux2 transcript levels were approx. 1.5-fold greater in DPL90 compared to M315 (Fig. 3); however, DPL90 growth was approx. 5-fold greater compared to M315. GB713, which showed a growth capacity that was intermediate to that of DPL90 and M315, exhibited the lowest amount of rolB transcript (approx. 4-fold less than DPL90 and approx. 2.5-fold less than M315) (Fig. 3).
Relative mean transcript starting quantity
Hairy root fresh weight (g)/petri plate
5.0
a
Ar_rolB Ar_aux2
1.6
A 1.4
b
b 1.2
B
B
1.0 0.8 0.6
c 0.4 0.2 0.0 DPL90
M315
M315-X
GB713
Fig. 3 Assessment of A. rhizogenes rolB (Ar_rolB) and aux2 (Ar_aux2) transcript levels in DPL90, M315, M315-X, and GB713 hairy roots by quantitative reverse transcription-PCR. Presented are the normalized relative mean ± SE transcript starting quantities for Ar_rolB and Ar_aux2 as calculated from three independent biological samples for each hairy root line examined. No difference (P [ 0.05; two-sample t test) between Ar_rolB and Ar_aux2 mean transcript starting quantities within any one culture was detected. Differences (P \ 0.05; two-sample t test) in either Ar_rolB or Ar_aux2 mean transcript starting quantities between cultures are denoted by different lowercase or uppercase letter designations, respectively
(Fig. 4b). Acid fuchsin staining showed sedentary thirdstage RKN juveniles within DPL90 galls at 13 DAI (Fig. 4c). In contrast to DPL90, M315 hairy roots predominantly showed slightly swollen sedentary J2 nematodes at 13 DAI (Fig. 4d). The mean number of sedentary RKN and egg masses per hairy root was determined for DPL90 and M315 at 40 DAI. These data showed that while an average of 2.2 ± 0.27 egg masses was observed to form on each DPL90 hairy root, no egg masses were found on the M315 hairy roots, confirming the resistance of M315 hairy roots to RKN (Fig. 4e).
RKN inoculation of cotton hairy roots RN inoculation of cotton hairy roots DPL90 (RKN-susceptible) and M315 (RKN-resistant) hairy roots were inoculated with surface-sterilized RKN second-stage juveniles (J2) in Petri dishes. At 7 days after inoculation (DAI), DPL90 and M315 hairy roots showed extensive J2 infection as revealed by acid fuchsin staining (data not shown). By 13 DAI, however, a phenotypic difference between DPL90 and M315 hairy roots was apparent (Fig. 4a–d). A galling response was readily observed on DPL90 hairy roots at 13 DAI (Fig. 4a), whereas M315 showed minimal galling or no galling at all
123
Hairy roots of DPL90 and M315 (RN-susceptible) and GB713 (RN-resistant) were excised from stock cultures and transferred to Petri dishes containing fresh growth medium. Two to 3 days after transfer, each root tip was inoculated with *300 vermiform RN in the same manner as described for RKN. By 11 DAI, numerous sedentary RN could be observed feeding on DPL90 hairy roots (Fig. 5a, b) with egg mass formation occurring by 32 DAI (Fig. 5c). The mean number of sedentary, reniform-shaped nematodes per hairy
Plant Cell Rep (2009) 28:1399–1409
a
1405
a
b
c
b c
e
d
12
DPL90 M315
Mean number/root tip
10
8
5.0 6
4
2.2 2
* 0.0 0 Sedentary nematodes
Egg masses
Fig. 4 RKN infection of cotton hairy roots. a Gall formation on a RKN-infected DPL90 hairy root at 13 days after inoculation (DAI) (89). b RKN-infected M315 hairy root showing no visible gall formation at 13 DAI (89). c Acid fuchsin-stained third-stage juvenile RKN routinely observed in galls formed on DPL90 hairy roots by 13 DAI (609). d Acid fuchsin-stained M315 hairy root showing delayed RKN development as evident by the presence of second-stage juvenile nematodes at 13 DAI (609) as denoted by black arrows. e Mean ± SE number of sedentary RKN and egg masses/hairy root tip for DPL90 (n = 19) and M315 (n = 13) as determined at 40 DAI for one representative experiment. Sample mean values are listed above their respective bars. *P \ 0.05 as determined by two-sample t test
root was determined for DPL90, M315, and GB713. As shown in Fig. 5d, DPL90 and M315 hairy roots supported similar numbers of sedentary RN, whereas GB713 hairy roots supported fewer RN than either DPL90 or M315 (Fig. 5d).
d Mean number sedentary females/root tip
8.8
20 18 16 14 12 10
*
8 6 4 2 0
DPL90
M315
GB713
Fig. 5 RN infection of cotton hairy roots. a Acid fuchsin-stained (259) and b not stained (809) DPL90 hairy roots showing sedentary female RN at 11 days after inoculation (DAI). Sedentary female RN is designated by arrows in (a). c Mature RN female with egg mass on a DPL90 hairy root at 32 DAI (809). d Mean ± SE number of sedentary RN/root tip for DPL90 (n = 16), M315 (n = 13), and GB713 (n = 15) hairy roots for one representative experiment. The GB713 sample mean is different (*P \ 0.05) from the DPL90 and M315 sample means as determined by two-sample t test
MIC3 transcript and protein abundance in cotton hairy root cultures The morphological and growth differences between the hairy root cultures raised the question as to whether hairy
123
Discussion The enhancement of Upland cotton cultivars by the incorporation of RKN and RN resistance remains one of the major challenges facing the cotton research community. In addition to molecular breeding efforts, the development of nematode-resistant cotton would be facilitated by a clearer understanding of the roles of specific cotton genes and signal transduction pathways in mediating compatible and incompatible interactions. In order to facilitate molecular biology-based experiments that could be conducted under sterile conditions, we investigated whether A. rhizogenes-induced cotton hairy roots could substitute for whole plants in studying the cotton–nematode pathosystem. To meet this objective, we conducted morphological, nematode infection, and molecular assays
123
Relative mean MIC3 transcript starting quantity
root cultures accurately model whole-plant roots at the molecular level. To answer this question, we measured MIC3 transcript abundance and MIC3 protein levels in the DPL90, M315, and M315-X hairy root lines and compared these values to those obtained for whole-plant roots from RKN-resistant and RKN-susceptible plants of varying age. MIC3 transcript and protein accumulation is tightly correlated with the onset of cotton resistance to RKN; however, MIC3 protein is routinely undetectable in uninoculated roots of either resistant or susceptible plants (Callahan et al. 1997, 2004; Wubben et al. 2008; Zhang et al. 2002). Basal MIC3 transcript levels differed between the hairy root cultures (Fig. 6a). For example, M315-X showed *6-fold more MIC3 transcript compared to DPL90. In contrast, basal MIC3 transcript levels did not differ between RKN-resistant (M315) and RKN-susceptible (M8) whole-plant roots within any of the observed time-points corresponding to increasing seedling age (Fig. 6b). We determined that MIC3 protein levels also varied between the hairy root lines and between experiments (Fig. 7). While undetectable in DPL90 hairy roots (lane 1), MIC3 protein was detected within the M315 and M315-X hairy roots in Experiment 1 and only M315-X in Experiment 2 (lanes 2 vs. 3, Fig. 7). Surprisingly, the level of MIC3 protein in M315-X in Experiment 2 (lane 3) was roughly equivalent to that detected in young galls collected from RKN-infected M315 plants (lane 6). The observed MIC3 protein accumulation in the M315-X hairy root line agrees with the finding that MIC3 transcript levels were greatest in M315-X compared to the other hairy root lines (Fig. 6a). MIC3 protein was not detectable in control roots of whole plants of either DPL90 or M315 (lanes 4 and 5, Fig. 6) in both experiments as expected from previously published work (Callahan et al. 1997, 2004; Zhang et al. 2002).
Plant Cell Rep (2009) 28:1399–1409
Relative mean MIC3 transcript starting quantity
1406 4.5
a b
4.0 3.5 3.0 2.5
a/b
2.0 1.5 1.0
a
0.5 0.0 DPL90
0.5
M315
M315-X
b
M8 M315
0.4
0.3
0.2
0.1
0.0 14 DAP
21 DAP
28 DAP
35 DAP
Fig. 6 Basal MIC3 transcript levels in cotton hairy roots and wholeplant roots as determined by quantitative reverse transcription-PCR. a Normalized relative mean ± SE MIC3 transcript starting quantities present in uninoculated DPL90, M315, and M315-X hairy roots as calculated from three independent biological replicates. Sample means sharing the same letter designation are not different (P [ 0.05; two-sample t test). b Normalized relative mean ± SE MIC3 transcript starting quantities present in uninoculated wholeplant roots of M8 (RKN-susceptible) and M315 at 14, 21, 28, and 35 days after planting (DAP). No differences (P [ 0.05; two-sample t test) in MIC3 expression between M8 and M315 were detected at any time-point
on hairy roots derived from a RKN-resistant breeding line (M315) and from a RN-resistant G. barbadense accession (GB713) in comparison to the nematode-susceptible control hairy root culture DPL90. Vacuum infiltration of M315 and GB713 cotyledons with a wild-type A. rhizogenes strain produced numerous, independent primary hairy roots that were selected for culturing; however, in both instances, only 5% of the primary cultures continued to show growth after three rounds of selection on 500 lg/mL ampicillin tissue culture
Plant Cell Rep (2009) 28:1399–1409
1
1407
2
Experiment 1 3 4
5
6
1
2
Experiment 2 3 4
5
6
γ-tubulin
MIC3 Fig. 7 Assessment of basal MIC3 protein levels in cotton hairy roots and whole-plant roots normalized to c-tubulin. Immunoblot analysis of MIC3 (*14 kDa) and c-tubulin (*40 kDa) protein in two independent experiments. Lane 1 uninoculated DPL90 hairy root,
lane 2 uninoculated M315 hairy root, lane 3 uninoculated M315-X hairy root, lane 4 uninoculated DPL90 whole-plant root, lane 5 uninoculated M315 whole-plant root, and lane 6 M315 gall tissue from RKN-inoculated plant roots
medium. This value is lower than that previously reported for cotton where 50–60% of the primary hairy root cultures survived antibiotic selection (Triplett et al. 2008). Because the same A. rhizogenes strain was used by Triplett et al. (2008) and in this study, it is unlikely that genetic differences in virulence or T-DNA transfer efficiency are the cause. However, genotypic differences in the progenitor cotton lines may partially explain the observed disparity. Nematode resistance may predispose M315 and GB713 hairy roots to an increased sensitivity to ampicillin compared to the nematode-susceptible genotypes used by Triplett et al. (2008). The efficiency of hairy root induction can vary considerably between genotypes of a given species and between plant species in general. For example, Mazarei et al. (1998) observed a range in hairy root induction efficiency of 5–90% across nine soybean genotypes. A similar phenomenon was observed by Savka et al. (1990) across ten soybean genotypes. In contrast, a high rate of hairy root induction was found using carrot disks; however, only 19% of independent hairy root cultures continued to show growth after 3 weeks of sub-culturing and only 4% after 2 years (Guivarc’h et al. 1999). We observed differences in morphology and growth between the DPL90, M315, and GB713 hairy root cultures. Hairy root cultures derived from independent transformation events can show morphological variation or differences in growth rate even when the same progenitor genotype is used. Triplett et al. (2008) reported significant variation in growth between 26 independently derived G. barbadense hairy root cultures. A similar phenomenon has been observed in other plant species including carrot (Guivarc’h et al. 1999), soybean (Cho et al. 2000), sugar beet (Paul et al. 1990), and tomato (Plovie et al. 2003). We investigated whether transformation-dependent differences in morphology and growth capacity between the DPL90, M315, and GB713 hairy root cultures were mediated by differences in the expression of genes that reside within the A. rhizogenes T-DNA. In planta overexpression of rolB, located on the TL-DNA, causes greatly increased adventitious root formation, while aux2, which resides on the TR-DNA, is believed to promote hairy root formation through its action as a 3-indoleacetamide hydrolase
(Altamura 2004; Nilsson and Olsson 1997). In support of our hypothesis, we found that rolB and aux2 expression was highest in the DPL90 hairy root culture which also showed the greatest amount of growth. However, GB713 hairy roots, which showed the second greatest amount of growth, exhibited the lowest level of rolB expression and did not have the aux2 gene. These observations do not support the hypothesis that differences in growth between the DPL90, M315, and GB713 cultures are due to correlative differences in T-DNA gene expression. Similarly, Alpizar et al. (2008) was unable to correlate the presence or absence of rol and aux genes in coffee hairy root cultures with phenotypic differences in morphology or growth rate between independent cultures. Cho et al. (2000) attributed variation in growth rate, branching, and degree of geotropism between soybean hairy root cultures to the genotypic differences in the cultivars used. It is possible, however, that the morphological and growth differences we have observed in cotton hairy roots are not due entirely to the progenitor genotype since the site of T-DNA integration into the host genome and T-DNA copy number can also strongly affect host gene expression (Nilsson and Olsson 1997). Inoculation of M315 hairy roots with infective RKN juveniles resulted in a complete lack of egg mass development and suggested that the RKN resistance of M315 had been retained in hairy roots. However, upon further analysis by qRT-PCR and immunoblotting, we determined that the M315 and M315-X hairy root cultures constitutively expressed elevated levels of MIC3 transcript and protein compared to the DPL90 control hairy roots. In previous characterizations of MIC3, and verified in the present study, MIC3 transcript and protein quantities between resistant and susceptible cotton plant roots were not found to differ prior to RKN inoculation (Callahan et al. 1997; Wubben et al. 2008; Zhang et al. 2002). MIC3 transcript and protein accumulation has been shown to occur at the onset of gall development in resistant plant roots (Callahan et al. 1997; Wubben et al. 2008). It cannot be determined at this time the nature of the mechanism governing MIC3 up-regulation in M315 hairy roots. One possibility is that the act of transformation induces MIC3 as
123
1408
a response to A. rhizogenes infection. Alternatively, the requisite elevation in auxin metabolism during hairy root formation and sustained growth may be an inducer of MIC3 expression. Regardless of the mechanism, it is interesting to note that neither DPL90 nor GB713, which are susceptible to RKN, showed this MIC3 up-regulation. This observation further corroborates the association between increased MIC3 expression and RKN resistance as previously reported (Callahan et al. 1997, 2004; Wubben et al. 2008; Zhang et al. 2002). We also determined that resistance to RN was retained in hairy roots derived from GB713. Unlike the M315 hairy root cultures, GB713 hairy roots more closely resembled the DPL90 control in morphology and growth rate. There is currently no information regarding the identity of any gene that is directly involved in RN resistance or whose expression is correlated with the onset of resistance. Consequently, it is difficult to determine whether the resistance mechanism within GB713 hairy roots is the same as that at work in whole plants. Further characterization of the GB713 resistance phenotypes in whole plants and hairy roots, including histological analyses of syncytia, will help resolve this issue. In summary, the data presented here demonstrate the potential use of cotton hairy roots as an experimental system for studying nematode resistance in cotton. We have shown that both RKN and RN will infect and reproduce on a susceptible culture, thereby, providing a way to maintain sterile monoxenic nematode cultures for research purposes. In addition, we have shown that RKN-resistant and RN-resistant hairy roots can be distinguished from a susceptible control. These results suggest that experiments involving the manipulation of host gene expression, e.g., overexpression, can be evaluated statistically and effects detected. The observed differences in cotton hairy root culture morphology, growth, and host gene expression, however, indicate that thorough preliminary analyses of control and experimental cultures should be conducted to identify potentially confounding phenotypes of the cultures themselves that are unrelated to the trait of interest. Acknowledgments The authors would like to thank Dr. Dirk Charlson (University of Arkansas) and Dr. Jeff Wilkinson (Mississippi State University) for their critical review of the manuscript.
References Alkharouf NW, Klink VP, Matthews BF (2007) Identification of Heterodera glycines (soybean cyst nematode [SCN]) cDNA sequences with high identity to those of Caenorhabditis elegans having lethal mutant or RNAi phenotypes. Exp Parasitol 115:247–258 Alpizar E, Dechamp E, Espeout S, Royer M, Lecouls AC, Nicole M, Bertrand B, Lashermes P, Etienne H (2006) Efficient production
123
Plant Cell Rep (2009) 28:1399–1409 of Agrobacterium rhizogenes-transformed roots and composite plants for studying gene expression in coffee roots. Plant Cell Rep 25:959–967 Alpizar E, Dechamp E, Lapeyre-Montes F, Guilhaumon C, Bertrand B, Jourdan C, Lashermes P, Etienne H (2008) Agrobacterium rhizogenes-transformed roots of coffee (Coffea arabica): conditions for long-term proliferation, and morphological and molecular characterization. Ann Bot 101:929–940 Altamura MM (2004) Agrobacterium rhizogenes rolB and rolD genes: regulation and involvement in plant development. Plant Cell Tiss Org Cult 77:89–101 Barker KR (1985) Nematode extraction and bioassays. In: Barker KR et al (eds) An advanced treatise on Meloidogyne, vol II, methodology. North Carolina State University Graphics, Raleigh, NC, pp 19–38 Baum TJ, Wubben MJE, Hardy KA, Su H, Rodermel SR (2000) A screen for Arabidopsis thaliana mutants with altered susceptibility to Heterodera schachtii. J Nematol 32:166–173 Blasingame D (2006) 2005 cotton disease loss estimate committee report. 2006 Beltwide cotton conferences, San Antonio, TX, January 3–6 Cai D, Thurau T, Tian Y, Lange T, Yeh K-W, Jung C (2003) Sporamine-mediated resistance to beet cyst nematodes (Heterodera schachtii Schm.) is dependent on trypsin inhibitory activity in sugar beet (Beta vulgaris L.) hairy roots. Plant Mol Biol 51:839–849 Callahan FE, Jenkins JN, Creech RG, Lawrence GW (1997) Changes in cotton root proteins correlated with resistance to root-knot nematode development. J Cot Sci 1:38–47 Callahan FE, Zhang X-D, Ma D-P, Jenkins JN, Hayes RW, Tucker ML (2004) Comparison of MIC-3 protein accumulation in response to root-knot nematode infection in cotton lines displaying a range of resistance levels. J Cot Sci 8:186–190 Cho H-J, Farrand SK, Noel GR, Widholm JM (2000) High-efficiency induction of soybean hairy roots and propagation of the soybean cyst nematode. Planta 210:195–204 Creech RG, Jenkins JN, Lawrence GW, McCarty JC Jr (1998) Nematode resistance in cotton. In: Bajaj YPS (ed) Biotechnology in agriculture and forestry, vol 42, cotton. Springer-Verlag, Berlin, pp 250–259 Davis EL, Hussey RS, Baum TJ (2004) Getting to the roots of parasitism by nematodes. Trends Parasitol 20:134–141 Doyle EA, Lambert KN (2003) Meloidogyne javanica chorismate mutase 1 alters plant cell development. Mol Plant–Microbe Interact 16:123–131 Fairbairn DJ, Cavallaro AS, Bernard M, Mahalinga-Iyer J, Graham MW, Botella JR (2007) Host-delivered RNAi: an effective strategy to silence genes in plant parasitic nematodes. Planta 226:1525–1533 Gal TZ, Aussenberg ER, Burdman S, Kapulnik Y, Koltai H (2006) Expression of a plant expansin is involved in the establishment of root knot nematode parasitism in tomato. Planta 224:155–162 Gao B, Allen R, Maier T, Davis EL, Baum TJ, Hussey RS (2003) The parasitome of the phytonematode Heterodera glycines. Mol Plant–Microbe Interact 16:720–726 Guivarc’h A, Boccara M, Prouteau M, Chriqui D (1999) Instability of phenotype and gene expression in long-term culture of carrot hairy root clones. Plant Cell Rep 19:43–50 Hansen E, Harper G, McPherson MJ, Atkinson HJ (1996) Differential expression patterns of the wound-inducible transgene wun1-uidA in potato roots following infection with either cyst or root knot nematodes. Physiol Mol Plant Pathol 48:161–170 Huang G, Gao B, Maier T, Allen R, Davis EL, Baum TJ, Hussey RS (2003) A profile of putative parasitism genes expressed in the esophageal gland cells of the root-knot nematode Meloidogyne incognita. Mol Plant–Microbe Interact 16:376–381
Plant Cell Rep (2009) 28:1399–1409 Huang G, Allen R, Davis EL, Baum TJ, Hussey RS (2006a) Engineering broad root-knot resistance in transgenic plants by RNAi silencing of a conserved and essential root-knot nematode parasitism gene. Proc Natl Acad Sci USA 103:14302–14306 Huang G, Dong R, Allen R, Davis EL, Baum TJ, Hussey RS (2006b) A root-knot nematode secretory peptide functions as a ligand for a plant transcription factor. Mol Plant–Microbe Interact 19:463– 470 Hussey RS (1990) Staining nematodes in plant tissue. In: Zuckerman BM et al (eds) Plant nematology laboratory manual. The Univ of Mass Agri Exp Stn, Amherst, MA, pp 190–193 Ithal N, Recknor J, Nettleton D, Maier T, Baum TJ, Mitchum MG (2007) Developmental transcript profiling of cyst nematode feeding cells in soybean roots. Mol Plant–Microbe Interact 20:510–525 Jammes F, Lecomte P, de Almeida-Engler J, Bitton F, MartinMagniette M-L, Renou JP, Abad P, Favery B (2005) Genomewide expression profiling of the host response to -knot nematode infection in Arabidopsis. Plant J 44:447–458 Mazarei M, Ying Z, Houtz RL (1998) Functional analysis of the Rubisco large subunit N-methyltransferase promoter from tobacco and its regulation by light in soybean hairy roots. Plant Cell Rep 17:907–912 Mazarei M, Lennon KA, Puthoff DP, Rodermel SR, Baum TJ (2004) Homologous soybean and Arabidopsis genes share responsiveness to cyst nematode infection. Mol Plant Pathol 5:409–423 Mitkowski NA, Abawi GS (2002) Monoxenic maintenance and reproduction of root-knot nematode (Meloidogyne hapla) on multiple-species in vitro root culture systems. Plant Cell Rep 21:14–23 Murashige T, Skoog F (1962) A revised medium for rapid growth and bio-assays with tobacco tissue cultures. Physiol Plant 15:473– 497 Nilsson O, Olsson O (1997) Getting to the root: the role of the Agrobacterium rhizogenes rol genes in the formation of hairy roots. Physiol Plant 100:463–473 Oakley SR (1995) CPCSD Acala C-225: a new nematode-resistant Acala variety for California’s San Joaquin Valley. In: Proceedings of Beltwide Cotton Production Research Conference, National Cotton Council of America, Memphis, TN, p 39 Orion D, Pilowsky M (1984) Excised tomato root culture as a tool for testing root-knot nematode resistance. Phytoparasitica 12:71–73 Paul H, van Deelen JEM, Henken B, de Bock ThSM, Lange W, Krens FA (1990) Expression in vitro of resistance to Heterodera schachtii in hairy roots of an alien monotelosomic addition plant of Beta vulgaris, transformed by Agrobacterium rhizogenes. Euphytica 48:153–157 Plovie E, De Buck S, Goeleven E, Tanghe M, Vercauteren I, Gheysen G (2003) Hairy roots to test for transgenic nematode resistance: think twice. Nematology 5:831–841 Preiszner J, VanToai TT, Huynh L, Bolla RI, Yen HH (2001) Structure and activity of a soybean Adh promoter in transgenic hairy roots. Plant Cell Rep 20:763–769 Robinson AF (2007) Reniform in US cotton: where, when, why, and some remedies. Annu Rev Phytopathol 45:263–288 Robinson AF, Bridges AC, Percival AE (2004) New sources of resistance to the reniform (Rotylenchulus reniformis Linford and Oliveira) and root-knot (Meloidogyne incognita (Kofoid & White) Chitwood) nematode in upland (Gossypium hirsutum L.) and sea island (G. barbadense L.) cotton. J Cot Sci 8:191–197
1409 Robinson AF, Bell AA, Dighe ND, Menz MA, Nichols RL, Stelly DM (2007) Introgression of resistance to nematode Rotylenchulus reniformis into upland cotton (Gossypium hirsutum) from Gossypium longicalyx. Crop Sci 47:1865–1877 Samuelian S, Kleine M, Ruyter-Spira CP, Klein-Lankhorst RM, Jung C (2004) Cloning and functional analyses of a gene from sugar beet up-regulated upon cyst nematode infection. Plant Mol Biol 54:147–156 Savka MA, Ravillion B, Noel GR, Farrand SK (1990) Induction of hairy roots on cultivated soybean genotypes and their use to propagate the soybean cyst nematode. Phytopathology 80:503– 508 Schaff JE, Nielsen DM, Smith CP, Scholl EH, DMcK Bird (2007) Comprehensive transcriptome profiling in tomato reveals a role for glycosyltransferase in Mi-mediated nematode resistance. Plant Physiol 144:1079–1092 Shepherd RL, McCarty JC Jr, Jenkins JN, Parrott WL (1996) Registration of nine cotton germplasm lines resistant to root-knot nematode. Crop Sci 36:820 Starr JL, Koenning SR, Kirkpatrick TL, Robinson AF, Roberts PA, Nichols RL (2007) The future of nematode management in cotton. J Nematol 39:283–294 Stetina SR, Sciumbato GL, Young LD, Blessitt JA (2009) Cotton cultivars evaluated for tolerance to reniform nematode. Online. Plant Health Progress. Available via http://www. plantmanagementnetwork.org/php Triplett BA, Moss SC, Bland JM, Dowd MK (2008) Induction of hairy root cultures from Gossypium hirsutum and Gossypium barbadense to produce gossypol and related compounds. In Vitro Cell Develop Biol Plant 44:508–517 Urwin PE, Atkinson HJ, Waller DA, McPherson MJ (1995) Engineered oryzacystatin-I expressed in transgenic hairy roots confers resistance to Globodera pallida. Plant J 8:121–131 Vandesompele J, De Preter K, Pattyn F, Poppe B, Van Roy N, De Paepe A, Speleman F (2002) Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome Biol 3:34.1–34.11 Wang X, Replogle A, Davis EL, Mitchum MG (2007) The tobacco Cel7 gene promoter is auxin-responsive and locally induced in nematode feeding sites of heterologous plants. Mol Plant Pathol 8:423–436 Williamson VM, Hussey RS (1996) Nematode pathogenesis and resistance in plants. Plant Cell 8:1735–1745 Wubben MJ, Callahan FE, Hayes RW, Jenkins JN (2008) Molecular characterization and temporal expression analyses indicate that the MIC (Meloidogyne Induced Cotton) gene family represents a novel group of root-specific defense-related genes in upland cotton (Gossypium hirsutum L.). Planta 228:111–123 Yadav BC, Veluthambi K, Subramaniam K (2006) Host-generated double stranded RNA induces RNAi in plant-parasitic nematodes and protects the host from infection. Mol Biochem Parasitol 148:219–222 Yik CP, Birchfield W (1984) Resistant germplasm in Gossypium species and related plants to Rotylenchulus reniformis. J Nematol 6:97–108 Zhang X-D, Callahan FE, Jenkins JN, Ma D-P, Karaca M, Saha S, Creech RG (2002) A novel root-specific gene, MIC-3, with increased expression in nematode-resistant cotton (Gossypium hirsutum L.) after root-knot nematode infection. Biochim Biophys Acta 1576:214–218
123
