Commercial ginseng fields from which root samples were obtained were located in Kamloops, BC and were sampled during the months of June to September. Roots showing longitudinal stripes (see Figure 1) were selected primarily from one-year-old plantings (in the year following seeding the previous fall season) where the symptoms predominated on seedling roots. Most of the roots were collected in the later part of the growing season (August-September) in two consecutive years. Harvested roots (approx. 100 in total) were placed inside plastic bags and transported to the laboratory in a cooler and stored at 4˚C and used for the experiments below. Field sites where striped roots were observed were marked and re-visited the following year to observe progression of symptom development.

Segments of root tissues approx. 2 mm² with visible stripe symptoms (Figure 1) were cut from affected roots and placed in a fixative for 72 hr (formalin: acetic acid: alcohol, 1:1:1) and then dehydrated in an ethanol series (70% to 100%, 2 hr in each). The tissue pieces were pre-infiltrated in a 100% ethanol: 2-hydroxyethylmetha- crylate mixture (1:1, v/v) (Technovit 7100, Marivac Ltd., Halifax, NS). Samples were then infiltrated in Technovit 7100 (100 ml plus 1 g hardner) for 24 hr at 4˚C and then embedded in Technovit. Cross-sections of the tissue were made using an LKB pyramitome (Diversified Equipment Corp., Lorton, VA) to a thickness of 4 - 8 μm. The sections were stained with Toluidine Blue 0 (0.05%) for 30 sec and rinsed in water. The sections were placed on microscope slides, dried at 50˚C for 10 min, and sealed under a coverslip using Permount (Fisher Scientific Canada, Ottawa, ON). The sections were examined at magnifications of 100 - 400 X under a Zeiss light microscope. Control tissues taken from roots without any striping were included for comparison. A minimum of 5 sections of each root sample was examined.

Samples of small root segments (1 mm²) containing striped tissue were prepared for scanning electron microscopy (SEM) and energy-dispersive spectroscopy analysis (EDX) to observe morphological changes on the root

surface and identify metal compounds in epidermal cells, respectively. Healthy root tissues were included as a control. For SEM, the tissues samples were fixed in 0.1 M cacodylate buffer containing 2% glutaraldehyde, 4% formaldehyde and 4% sucrose using standard microwave processing [11] . Samples were post-fixed in 2% (w/v) tannic acid, followed by fixation in 1% osmium tetroxide in the above buffer for 1 h at room temperature, and 2% aqueous uranyl acetate for 1 h at room temperature in the dark. The samples were then dehydrated in a graded ethanol series and critical point-dried using CO₂. Samples were mounted on stubs using colloidal silver, sputter-coated with gold-palladium in a Nanotech SEMPrep II Sputter Coater and examined in a scanning electron microscope (Hitachi S4700). For EDX, the tissue samples were immediately frozen in liquid nitrogen and freeze-dried for 48 hr, after which the samples were adhered to a metallic stub and carbon-coated to make the sample conductive. The samples were analyzed in a Hitachi S-3000 N, EDX QuartzX one (Tokyo, Japan).

The roots selected for this assay were first gently washed in deionized water several times and blotted dry. The striped tissues were gently scraped from the root surface using a sharp scalpel and bulked to obtain a composite sample. Healthy root tissues were included as a control. Air-dried (0.2 g) tissue was digested with 40% (v/v) hydrofluoric acid mixed with nitric acid and made up to 100 ml with 1% nitric acid. Aliquots of this solution were mixed with an internal standard and analyzed by Inductively Coupled Plasma Atomic Emission Spectrometry (ICP-AES) as described in [12] . The results were compared to a calibration curve constructed using multi-element standards and expressed in ppm. Two replicate samples were included for analysis and the procedure was repeated once using a different set of roots.

Total phenolic compounds present in striped and healthy root tissues was determined using the Folin-Ciocaleau method [13] . Using a sharp scalpel, the striped areas were gently scraped off the surface of multiple roots and bulked to provide 2 g (fresh weight) which was then frozen in liquid nitrogen and ground using a mortar and pestle. The tissues were mixed with 80% methanol containing 8% formic acid, stirred at 37˚C in a water bath for 30 min, and centrifuged at 8000 × g for 10 min. One ml of the supernatant was added to 5 ml of diluted (1:10, v/v) F-C reagent and mixed for 30 sec. After 5 min, 4 ml of 7.5% Na₂CO₃ solution was added and the mixture was incubated for 2 hr in the dark. Absorbance was measured at 740 nm in a Thermo Spectronic Helios UV visible spectrophotometer (Cambridge, UK) by using the reagent mix as a reference. The mean absorbance was compared to a calibration graph prepared using a range of concentrations (10, 25, 50 and 100 μg∙ml⁻¹) of gallic acid and the phenolic concentration was expressed as μg∙g⁻¹ gallic acid equivalents (GAE). Two replicate samples were included for analysis and the procedure was repeated once using a different set of roots.

To determine specific phenolic compounds, the above supernatant extract was evaporated to dryness using a rotary evaporator. The residue was dissolved in 5 ml solvent (methanol, formic acid and water) and centrifuged at 12,000 × g for 10 min and filtered through a 0.2 μm hydrophilic Durapore membrane filter (Millipore Corp., Bedford, MA). Fifty microliters of the extract, which constituted the mobile phase for HPLC, was injected through through an Atlantis^TM dC18 column in an LC-module 1 Waters HPLC machine (Mildford, MS) at a flow rate 1.2 ml/min⁻¹ and the absorbance was recorded at 280 nm over a 24 min retention time.

Striped tissues were gently scraped from the surface of multiple roots and bulked to provide 2 g (fresh weight), frozen immediately in liquid N₂, and homogenized with 20 ml of extraction buffer on ice. The buffer was prepared with 80 ml Tris-HCl (0.1 M, pH 7.8), containing 15 ml of 10 mM β-mercaptoethanol, 5 ml 0.5% ascorbate and three drops of glycerol. The homogenate was centrifuged at 37,000 × g for 10 min at 4˚C. The supernatant was kept on ice until the assay for enzymes was done. Three replicate samples were included in each assay.

Phenylalanine ammonia lyase (PAL). The assay mixture contained 900 μl of 10 mM L-phenylalanine in Tris?HCl (0.1 M, pH 8.8), and 100 μl extract. The mixture was placed in a water bath at 37˚C for 60 min. Enzyme activity was determined using a spectrophotometer (Thermo Spectronic Helios UV visible) by the increase in absorbance at 290 nm over a 60 min interval. The change in absorbance was found to be proportional to the amount of enzyme, which produces an increase in absorbance at 290 nm of 1.0 per hour with a 1 cm light path. Specific enzyme activity was expressed as units per milligram protein of the enzyme extract. Protein concentration was measured according to Bradford [14] using bovine serum albumin (BSA) as a standard.

Peroxidase (POD). The tissue extracts were diluted 10-fold with the homogenization buffer. Pyrogallol solution (0.05 M) was prepared in 5 mM Na2(PO₄)₃ buffer (pH 6.0). One hundred microliters of the diluted extracts was mixed with 800 μl pyrogallol. The reaction was started by adding 100 μl of 0.6 M H₂O₂ to the mixture, and the rate of increase in absorbance at 470 nm was measured at 20 sec intervals for a minute. Specific enzyme activity was expressed as units per milligram protein of the enzyme extract. One unit is defined as the amount that will form 1.0 mg purpurogallin (PPG) from pyrogallol in 20 sec at pH 6.0 at 20˚C by increasing the absorbance by 1.0.

Ginseng roots with stripe symptoms were distinctly different from healthy roots that showed no external discoloration (Figure 1(a)). Striped roots had varying degrees of reddish-orange longitudinal stripes that varied in length from 2 cm to up to 8 cm and extended down from the upper crown to more than halfway down the length of the root (Figure 1(b)). The stripes consisted of ridges of tissues comprised of reddish-brown material (Figure 1(c)). Light microscopic examination of the tissues showed that the cell layers stained a bluish-green with Toluidine blue 0 (Figure 2(a)), indicating the presence of phenolic compounds. The underlying cortical cells remained unstained. The ridges appeared to have formed as a result of epidermal fissures that extended down into the cortical cells, causing them to push outwards (Figure 2(b)). The fissures could be seen forming over the entire root surface, leading to multiple ridges forming on the root (Figure 3(a)). At subsequent later stages of development, the epidermal cell layers that were stained with Toluidine blue and several cortical cell layers were sloughed off by the formation of an abscission zone (Figure 3(b)). These cell layers were not seen in control roots (Figure 3(c)).

Images of the surface of striped root tissues examined under a SEM confirmed that the epidermis had ruptured and cracks (fissures) were formed (Figure 4(a)), creating ridges of tissue comprised of the epidermis on either side and exposing the underlying cortical cells (Figure 4(b)). These ridges of darkly-pigmented tissues could also be seen as longitudinal lines (stripes) that made their way down the root (Figure 4(c)). Examination of the epidermal tissues using SEM-EDX revealed that there was a significant increase in the levels of silicon, aluminum and iron as seen in the spectrum analysis when compared with healthy epidermal tissues, in which only potassium was significantly higher (Figure 5). When the compositional levels of all elements were expressed as

concentration (wt [%]), the levels of aluminum, calcium, iron and silicon were found to be higher and potassium lower in striped tissues (Table 1). As well, magnesium and phosphorus were slightly higher in striped tissues.

Analysis of striped root tissues and healthy tissues by Inductively Coupled Plasma Atomic Emission Spectrometry confirmed the results obtained by SEM-EDX. Levels of aluminum and iron were markedly higher in striped tissues, and magnesium and manganese levels were slightly higher (Table 2). In healthy tissues, levels of copper and zinc were increased.

Total phenolic levels in extracts from striped root tissues were almost 80% higher compared to healthy tissues (69 μg/g⁻¹ gallic acid equivalents (GAE) compared to 38 μg/g⁻¹, respectively). When examined by HPLC analysis,

Data are the average from two replicate samples and are expressed as concentration [wt. (%)]. The SEM-EDX profiles are shown in Figure 5. Carbon and oxygen were estimated to be 55% and 35%, respectively.

Data are the average of two replicate samples. The analysis was conducted twice using tissues bulked from different roots. Representative data from one analysis are shown.

several phenolic compounds showed higher peaks at specific retention times between 10 and 25 min in extracts from striped root tissues compared to healthy roots (Figure 6). By using specific phenolic compounds as standards, some of these higher peaks on the chromatogram were identified as chlorogenic acid, quercetin, protocatechuic acid, O-coumaric acid, catechin and gallic acid (data not shown). The levels of the enzymes phenyalalanine ammonium lyase (PAL) and peroxidase (POD) were also found to be higher in extracts from striped tissues compared to healthy tissues (24.5 and 16.5 units of PAL and 7.2 and 4.6 units of POD, respectively. The data were averaged from 3 replicate samples).