JEQ Journal of Natural Resources and Life Sciences Education
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 5 April 2007
Published in J Environ Qual 36:709-717 (2007)
DOI: 10.2134/jeq2006.0231
© 2007 American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by von Kiparski, G. R.
Right arrow Articles by Gillespie, A. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by von Kiparski, G. R.
Right arrow Articles by Gillespie, A. R.
Agricola
Right arrow Articles by von Kiparski, G. R.
Right arrow Articles by Gillespie, A. R.
Related Collections
Right arrow Fate
Right arrow Agroforestry

TECHNICAL REPORTS

Plant and Environment Interactions

Occurrence and Fate of the Phytotoxin Juglone in Alley Soils under Black Walnut Trees

Guntram R. von Kiparskia, Linda S. Leeb,* and Andrew R. Gillespiea

a Dep. of Forestry and Natural Resources, Purdue Univ., West Lafayette, IN 47907-2051. G.R. von Kiparski, current address, Center for Accelerator Mass Spectrometry, Lawrence Livermore National Lab., Livermore, CA 94550-2051
b Dep. of Agronomy, Purdue Univ., West Lafayette, IN 47907-2054

* Corresponding author (lslee{at}purdue.edu)

Received for publication June 16, 2006.

   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 SUMMARY AND SIGNIFICANCE
 REFERENCES
 
Juglone (5-hydroxy-1,4-napthoquinone) is a chemical released by walnut trees, which can be toxic at various levels to several plant species. A balance among competing source and sink mechanisms and rates will ultimately determine whether juglone is capable of attaining sufficient levels to be allelopathic to intercrops in a walnut tree agroforestry system. In this study, juglone's release, accumulation, and decline in soil are explored using data from soil beneath a black walnut tree (Juglans nigra L) alley cropping system, greenhouse pot studies, and laboratory sorption/degradation studies. Juglone pore water concentrations estimated from extracts of surficial soil from beneath the alley cropping system exceeded the lowest solution culture toxicity levels reported for some plants of 10–7 M, but did not exceed the inhibition threshold reported for typical intercrops such as maize and soybeans 10–5 M. Further assessment of the likely persistence of juglone in soils indicated that juglone is both microbially and abiotically degraded, and that it will be particularly short-lived in soils supporting microbial activity. However, walnut seedlings planted in sand-filled pots clearly showed that juglone is released in measurable quantities to the soil's rhizosphere. Therefore, juglone accumulation in low fertility soils is plausible, and may still be worthy of consideration in management of alley agroforestry systems.

Abbreviations: ACN, acetonitrile • Cs,ext, extracted sorbed concentration • Cw, equilibrium solution concentration • HPLC, high performance liquid chromatography • Kd, linear sorption coefficient • Koc, organic carbon normalized sorption coefficient • MeOH, methanol • OC, organic carbon content • pKa, acid dissociation constant • UV, ultraviolet


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 SUMMARY AND SIGNIFICANCE
 REFERENCES
 
THE detrimental effect of walnut trees on certain plants has been a concern since the days of the Roman Period (Bostock et al., 1718) and is still acknowledged in current horticultural extension literature today (Dana and Lerner, 1994; Tentinger, 2003). Juglone (5-hydroxy-1,4-naphthoquinone) (Table 1) is believed to be released by walnut trees (Juglans spp) into the soil and accumulate to toxic levels sufficient to limit the growth of competing plants near the trees. Evidence of juglone toxicity comes from: (1) field observation of walnut toxicity toward certain plants causing symptoms collectively referred to as walnut wilt (Dana and Lerner, 1994; Tentinger, 2003); (2) hydroponics-based solution culture toxicity testing, which demonstrated that juglone is potentially toxic to certain plant species (Funk et al., 1979; Rietveld, 1983; Segura-Aguilar et al., 1992; Jose and Gillespie, 1998b; Kocacaliskan and Terzi, 2001; Hejl and Koster, 2004); and (3) field evidence of juglone accumulation in soils (Ponder and Tadros, 1985; De Scisciolo et al., 1990; Jose and Gillespie, 1998a). Whether juglone toxicity occurs at the target plant or not depends on its mode of toxic action and whether sufficient juglone concentrations in soil are available to invoke a toxic response.


View this table:
[in this window]
[in a new window]

  		Table1. Molecular properties of juglone (5-hydroxy-1,4-naphthoquinone).

 
For the agronomic crops maize (Zea mays L.) and soybeans (Glycine max L. Merr.), exposure-based data show that juglone can inhibit shoot- and root-relative growth rates (Jose and Gillespie, 1998b; Hejl and Koster, 2004); leaf photosynthesis (Hejl et al., 1993; Jose and Gillespie, 1998b), transpiration (Jose and Gillespie, 1998b), stomatal conductance (Jose and Gillespie, 1998b), and leaf and root respiration (Perry, 1967; Koeppe, 1972; Hejl et al., 1993; Jose and Gillespie, 1998b). Juglone toxicity has also been observed against fungal species (Hedin et al., 1980; Tripathi et al., 1980) and N-fixing soil bacteria (Neave and Dawson, 1989). Juglone's association with these phenomena appears to be related to enzyme inhibition, irreversible complexation with cellular proteins, and disruption of electron transport chains. Juglone's mode of toxic action could involve mixed mechanisms including electrophilic reactivity of its quinoidal and phenolic functionalities (Table 1) toward cellular nucleophiles (e.g., alcohol, thiol, and amino groups) and an ability to undergo oxidation–reduction cycles between oxidized quinones, partially reduced semiquinones, and fully reduced phenols (O'Brien, 1991; Rath et al., 1996). The redox-active nature of juglone's functional groups provides properties known to disrupt physiologically important electron transfer chains in plants (Oettmeier et al., 1988).

Concentration levels and persistence within soil and soil pore water of juglone or its precursors are initially determined by source and input locations (e.g., root exudation belowground or litterfall and leaf leaching aboveground). Juglone from black walnut trees to soil is thought to exist primarily as {alpha}-hydrojuglone glucoside, which is relatively benign, but can be rapidly oxidized to juglone (Daglish, 1950; Willis, 2000). Release of juglone from the parent glucoside in plants is considered to be a two-step process with ß-glucosidase, a common soil enzyme (Alef and Nannipieri, 1995; Busto and Perez-Mateos, 2000), catalyzing the hydrolysis to hydrojuglone (Duroux et al., 1998) followed by the fast chemical oxidation to the quinonoidal compound juglone (Duroux et al., 1998; Müller and Leistner, 1976). Potential juglone abundance estimated in walnut leaves, hulls, and roots ranges from less than 0.1% dry wt. basis to as much as 5% dry wt. basis depending on when samples were taken in the growing season and the extraction techniques employed (Willis, 2000 and citations within). With regards to the latter, some investigators targeted only juglone, whereas others first attempted to oxidize any precursors present to juglone before or during extraction. Although both above and below ground sources of juglone are recognized, rhizodeposition has been hypothesized to be the major source of juglone entry to soil from walnut trees (Jose and Gillespie, 1998a). Jose and Gillespie (1998a) found significantly lower soil juglone concentrations in the cropping alleys associated with a tree root barrier treatment compared to the "no barrier" control. The root barrier treatment also resulted in significantly higher juglone levels within the tree row, where tree roots were concentrated relative to the "no barrier" control. Only juglone concentrations were targeted in this study; however, given that the precursors {alpha}-hydrojuglone glucoside and hydrojuglone are expected to rapidly convert to juglone in the aerobic rhizosphere soil environment, their assessment which supports the significance of root exudation over aboveground sources is likely valid.

When juglone from black walnut trees enters the soil, interactions with soil particles combined with microbial activity can reduce its concentration in soil pore water, which is where juglone would be in intimate contact with plant roots and soil organisms, thus directly available to affect their growth and behavior. Juglone in soil released from black walnut trees first encounters the target plant at the plant root surface. In solution culture, Hejl and Koster (2004) reported that juglone interference with maize plants occurred at the roots of target plants by disrupting plasma membrane K+–ATPase activity and associated plant water and nutrient uptake. Some plants appear to have a protective capacity against oxidative stress from juglone by emitting enzymes that metabolize the compound to less toxic hydroquinones (Segura-Aguilar et al., 1992; Matvienko et al., 2001). Juglone that has been released by source plants into the soil, if not irreversibly sorbed or degraded, will accumulate on soil particles in dynamic equilibrium with the solution phase and serve as an additional pool of juglone for the target plant/organism. If juglone in soils can contact with plant roots over a period of time, then both soil pore water concentrations and potentially reversibly-sorbed soil juglone may serve in toxicity. If the contact time in the target plant is short-lived, then time for desorption from soil surfaces to the aqueous phase is critical to whether a certain amount of the compound can reach the target plant.

Understanding soil-phytotoxin interactions has been of increasing interest in determining plant-available pools of toxins (Cheng, 1995; Blum et al., 1999; Dalton, 1999; Reigosa et al., 1999; Jose, 2002), but research into juglone's interaction with soil is limited and contradictory (Fisher, 1978; De Scisciolo et al., 1990). Fisher (1978) found that juglone was persistent in the water-extractable phase in a sandy loam soil even after a 90-d laboratory incubation period. In contrast, De Scisciolo et al. (1990) reported rapid (1.5 h) and apparently abiotic juglone loss in both sterile (autoclaved) and unsterile silt loam soil. Microbial degradation of juglone in nutrient cultures with selected bacteria is well known (Rettenmaier et al., 1983; Schmidt, 1988; Müller and Lingens, 1988); however, the relative contribution of abiotic and microbial processes in soils is not known. Coupled interactions between soil minerals, organic matter, and microorganisms are expected to govern the fate of juglone entering soils.

A balance among competing source and sink mechanisms and rates will ultimately determine whether juglone is capable of attaining sufficient levels to be allelopathic to intercrops in a walnut tree agroforestry system (Willis, 2000 and citations within). In this study, estimates were made of juglone levels in the soil pore water beneath a black walnut alley cropping system and compared to published solution culture toxicity threshold levels. In addition, juglone accumulation in sand-filled pots containing black walnut seedlings was measured in a greenhouse study to determine if seedling root exudation can serve as a measurable source of juglone to soil. Finally, sorption and degradation under sterile and unsterile conditions were measured for juglone in two soils with similar parent materials but differing organic carbon content and acidity to assess persistence of juglone in soils. In light of the data from this study, the environmental significance of juglone is evaluated.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 SUMMARY AND SIGNIFICANCE
 REFERENCES
 
Chemicals
Juglone (Table 1) was purchased from Sigma-Aldrich Chemical Co. HPLC-grade (99.9% purity) acetonitrile was purchased from Mallinckrodt and HPLC-grade methanol, anhydrous calcium chloride (CaCl2), and phosphoric acid (85% v/v) were purchased from Fisher Scientific.

Soils
Native soil juglone concentrations were measured in Ryker soils collected from the A horizon beneath a black walnut alley cropping system established in 1985 (SEPAC; Butlerville, IN; 39°03' N, 85°30' W). Further details of the site are given by Gillespie et al. (2000) and Jose (1997). The black walnut trees were interplanted with maize (Zea mays L.) or soybeans (Glycine max (L.) Merr.) in parallel (north-south) rows in the alleys and managed using annual chemical weed control and NPK fertilization. During 2002 through 2004, the alley cropping system was fallowed. The silt loam soil (fine silty, mixed, active, mesic Fragaquic Paleudult) was formed under forest vegetation and now has a visible plow layer from intermittent agricultural cultivation. The soil had been previously characterized as a Parke soil (Jose, 1997), but was later reclassified to account for the incipient fragipan within the lower B horizon and is described for this research as a Ryker series soil (D. Marshall, personal communication, 2003). The soil mineral composition consists of kaolinite and mica from loess with smaller quantities of quartz (USDA NRCS reference data from soil characterization on site; USDA Soil Survey Staff, 2004). The strongly leached and weathered soil has a low organic matter content, low native fertility, acidic reddish subsoil, and no carbonates in the upper 1 m of soil. Abundant iron and manganese films are present on the surface of soil peds in the subsoil. Accumulation of tree litter is sparse with <1% of the surface area covered by partially decomposed walnut leaf raches and <1 cm diam. twigs.

For determination of native juglone concentrations, soils were sampled in relation to spatial distance from the black walnut tree row. In mid-August 2004, two positions were sampled in the alley cropping plots in which juglone was known to accumulate differently (Jose and Gillespie, 1998a): the alley edge (1.2 m from the tree row) and the mid alley (4.2 m from the tree row). Six A horizon (0 to 20 cm depth) soil samples were collected two meters apart along each of the two distances from the tree row, transported on ice to the laboratory, and stored at 4°C for 1 wk before extraction and analysis. Likewise, soils were also collected from the A horizon of an adjacent agricultural field as a control that did not support walnut trees and was expected to lack juglone. Organic carbon (OC) analysis was performed on a composite (subsamples of each of the six samples were mixed) for each of the two distances (edge and mid-alley).

For the sorption and degradation studies, two soils (Table 2) in the Ryker soil series, an epipedon (A horizon; 0 to 20 cm depth) and associated subsoil (B horizon; 4 to 65 cm depth), were collected (3 kg each) from an area adjacent to but at least 50 m and upwind (avoiding tree litterfall inputs) from a black walnut tree (Juglans nigra L.) agroforestry plantation (Gillespie et al., 2000). The entire area was under corn–fallow rotation for approximately 50 yr before the start of the walnut plantation in 1985.


View this table:
[in this window]
[in a new window]

  		Table 2. Selected properties of the two bulk Ryker silt loam soils (Fragaquic Paleudult) from the Southeastern Purdue Agricultural Center (SEPAC; Butlerville, IN, USA) used for the juglone (5-hydroxy-1,4-naphthoquinone) sorption and degradation experiments.

 
Juglone Determination in Field Soils
Soils collected for juglone analysis were sieved through a 4-mm mesh screen immediately before extraction to homogenize soils before subsampling and any roots and other visible undecomposed organic materials were removed. Soils were extracted for 3 h with 1:1 (v/v) acetonitrile/methanol at a soil/solvent ratio (m/v) of 1:5, which was assumed to include both reversibly-sorbed (Cs,ext) and aqueous phase (Cw) juglone. Additional extractions or shaking for longer periods (up to 24 h) improved compound recovery. Controls (juglone addition to reaction vessels with no soil but with solvent or water) exhibited <1% juglone degradation after 3 d of incubation. Tubes were centrifuged at 1500 g for 25 min, and supernatant aliquots were transferred to 2-mL amber vials for juglone analysis. Differences, if any, in the estimated total extractable juglone content of soils from the two agroforestry alley positions (1.2 and 4.2 m) were assessed using the Student's t test (SAS Institute, 2001). The sample means were considered significant if p < 0.05.

Juglone Accumulation in Potted Soils Growing Black Walnut Seedlings
Juglone release to soil was measured for 1-yr-old black walnut seedlings grown in pots for approximately 4 mo in the greenhouse. Washed sand composed of >99.9% silica was used for the potting media with no fertilizer added. The black walnut seedlings (1/0 stock) were planted into 5-L-capacity plastic pots (with drainage hole) in early April 2004 as part of another research project (F. Salifu, personal communication, 2004). All seedlings received water at the same rate and it was sufficient to meet evaporative demand. Water was added at the soil surface to avoid the possibility of leaf-leaching additions of juglone. The pot study was hypothesized to represent a worse case scenario with regards to accumulation of juglone concentrations given the low sorptivity, reactivity, and fertility (no fertilizer addition) of the sand used.

A subset of the potted seedlings was harvested to measure soil juglone content in mid-August 2004. Soil samples (~200 g) from each of three (n = 3) individually-potted seedlings were collected in duplicate from two locations within each pot. The first location represented the bulk soil and the second location sampled only rhizosphere soil. The rhizosphere soils were collected after first carefully excavating the black walnut tree seedling from the pot and then gently shaking off soil surrounding and clinging to the tree roots into a sterilized container. Samples were transferred back to the laboratory and extracted within 25 min. Bulk soil was collected from the soil materials remaining after rhizosphere soil sampling. Before sample extraction, soil samples were inspected to remove roots and other differentiable organic materials. Juglone concentrations from soil extracts were compared between bulk and rhizosphere soil using the Student's t test (SAS Institute, 2001). Significant differences in the means between the bulk and rhizosphere soils were reported for p < 0.05.

Juglone Sorption and Degradation in Soils
Portions (500 g each) of the A and B horizon soils were sieved through 2-mm mesh, air-dried, and stored in air-tight plastic containers for approximately 2 mo before conducting the batch sorption experiments. A subset of each of the soils was sterilized by irradiating air-dried soils for 38 h using 60Cobalt-{gamma}-irradiation (Purdue University radiation environmental management facility) achieving a 3.7 Mrads dose. All glassware was autoclaved, juglone solutions were filter-sterilized (0.2 µm), and care was taken to avoid contamination of items coming into contact with the sterile soil and the reaction vials. Sorption tests with sterile and unsterile soils were conducted randomly distributed over time to remove a potential source of bias.

Sorption isotherms at 22°C were constructed in duplicate for each soil using a batch equilibrium technique. Samples (5 g air-dried soil) were equilibrated in Teflon-lined screw-capped 40-mL glass vials with 20-mL aqueous solution containing 50, 100, 150, or 250 µg of juglone in 0.0025 M CaCl2 to create a four-point isotherm for each replicate isotherm. Aqueous juglone solutions were prepared from a juglone stock solution of 1:1 (v/v) ACN/MeOH matrix such that the organic solvent never exceeded 0.2% of volume in the aqueous solutions. Samples were kept in complete darkness between sampling times, and extracted under low lighting conditions to prevent photodegradation. Samples were shaken for 24 h on an orbital-action shaker followed by centrifugation (1500 g) for 25 min. Aqueous supernatants (Cw, mg L–1) were removed for analysis and soil was then extracted with 1:1 (v/v) ACN/MeOH as previously described for the field soils to determine juglone concentrations (Cs,ext, mg kg–1), which was considered juglone that could be available for desorption and transport (i.e., reversibly sorbed). Solvent extract concentrations were corrected for residual aqueous-phase juglone by assuming juglone concentrations in the bulk aqueous supernatant were equal to the concentrations in the residual water and knowing the weight of the residual water.

Sorption isotherms (Cs,ext vs. Cw) were fit with a model to estimate the linear reversible distribution coefficients (Kd, L kg–1): Kd = Cs,ext Cw–1, where Cs,ext (mg kg–1) is extractable juglone sorbed on solids and Cw (mg L–1) is the juglone concentration in 0.0025 M CaCl2. Juglone among mass fractions (solvent-extractable, aqueous phase, unrecoverable, total recovered) is reported as a percentage of applied juglone. Differences in soil juglone partitioning and degradation between soils and treatments were evaluated based on differences in Kd values and total juglone recoveries.

Juglone Analysis
Juglone was analyzed using a PerkinElmer high performance liquid chromatograph (HPLC) with a 40-µL injection loop, 5-µm ODS C-18 Microsorb-MV reverse-phase column (4.6 x 250 mm) (Varian Inc.), a mobile phase of 55:45 v/v acetonitrile/0.5% H3PO4 at 0.8 mL min–1, and a UV-Vis detector (PerkinElmer Model LC-95). A visible wavelength of 427 nm peak was chosen for optimal detection of juglone to minimize interference from co-extractants expected at the shorter wavelengths also suitable for juglone. Juglone identification in samples was confirmed by comparison to the UV spectrum of a commercial standard and by co-injection. Quantification of juglone was made based on comparisons to an external calibration curve of juglone prepared in 0.0025 M CaCl2 or the extracting solvent as appropriate. The limit of quantitation (LOQ) varied from 10 to 20 µg juglone L–1.


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
SUMMARY AND SIGNIFICANCE
 REFERENCES
 
Juglone Concentrations in Black Walnut Agroforestry Soils
Total extractable juglone in the Ryker silt loam soil (A horizon) collected in August 2004 beneath the 19-yr-old black walnut alley cropping system varied from 0 to 0.55 mg kg–1 at the alley edge (1.2 m from the tree row) to 0 to 5.65 mg kg–1 at the mid alley (4.2 m from the tree row) (Table 3). Juglone was not detected in any soil samples taken from the adjacent agronomic field where no black walnut trees were present. A simple t test revealed no significant differences in the spatial patterns of extractable juglone within the black walnut alleys (p = 0.15); however, juglone concentrations in three of six samples taken at the mid alley location were nearly an order of magnitude higher than any of the samples take at the alley edge (e.g., 5.65 vs. 0.50 mg kg–1) (Table 3). Eight years earlier in late July, Jose and Gillespie (1998a) had done a more detailed study in which they obtained composite soils samples (n = 10) for two transects in each of three blocks within the same site sampled in the current study. They observed decreasing juglone concentrations with increasing distance from the trees with 1.4 ± 0.2 and 0.5 ± 0.2 µg juglone g–1 dry soil in samples taken 0.9 and 4.3 m from the tree row, respectively, in contrast to the apparent trend in the current study. During the period between the two studies, tree root biomass at the 0- to 30-cm soil depth increased by a factor of 5 from 187 to 940 kg ha–1 at 4.2 m from the tree row and only by about 35% from 996 to 1350 kg ha–1 at the 1.2-m distance from the tree row (von Kiparski and Gillespie, 2006). Further support that the increased rooting between sampling periods contributed to the elevated concentrations in the mid alley is the significant reduction in juglone levels observed by Jose and Gillespie (1998a) in soils sampled from companion plots where root barriers were installed. In the current study, half the mid alley samples had low or no detectable concentrations (Table 3), which may reflect point sources from litter decomposition; however, litter accumulation was sparse to none at the time of sampling. Also researchers have shown low concentrations of juglone in senescent leaves (Borazjani et al., 1983; Cline and Neely, 1984).


View this table:
[in this window]
[in a new window]

  		Table 3. Total extractable juglone (5-hydroxy-1,4-naphthoquinone) from the Ryker soil A horizon (0–20 cm depth interval) beneath the black walnut (Juglans nigra L.) alley cropping system at SEPAC (Butlerville, IN, USA) in August 2004.

 
Overall, the juglone concentrations observed in late summer of the current study include the range reported by Jose and Gillespie (1998a) of 0 to 2.90 mg kg–1 in earlier years for all plots and sampling times (spring, summer, and early winter) on this same site. Values also include the maximum value (3.95 mg kg–1) for a silt loam soil beneath a 14-yr-old black walnut tree plantation (Ponder and Tadros, 1985) and the range of 0 to 3.25 mg kg–1 in silt loam to sandy soils supporting 73-yr-old black walnut trees (De Scisciolo et al., 1990), respectively. In the latter study conducted in New York, seasonal variations were observed with juglone soil concentrations being highest in soils sampled in spring and early winter (April and November) and lowest in mid-summer samples (July). On the contrary, Jose and Gillespie (1998a) found no statistically significant differences between samples taken in the same months for the site in Indiana. De Scisciolo et al. (1990) also found no statistically significant correlation between soil moisture, pH, organic matter content, and soil texture. Therefore, predicting when seasonal trends are expected remains obscure.

Juglone Accumulation in Sand-Filled Pots Supporting Walnut Seedlings
In the sand-filled pot studies supporting black walnut seedlings for 4 mo, juglone was observed in only the sand immediately surrounding and clinging to the tree roots (4.44 ± 0.36, 4.85 ± 3.30, and 4.18 ± 1.83 mg kg –1 for the three replicate pots); no juglone was detected in any of the bulk sand outside the rhizophere. During the 4-mo growing period, plants did not indicate foliar nutrient deficiency symptoms; however, they were visually stunted relative to full nutrient plants of the same age in the greenhouse. This greenhouse study is the first known documentation of rhizosphere juglone levels and supports that rhizodeposition is involved in juglone release and accumulation in soil.

Juglone Sorption by Soils
The relative distribution of juglone among the reversibly sorbed fraction (solvent-extractable) and the aqueous fractions as a function of applied juglone mass in the batch sorption experiments is summarized in Table 4. Reversible sorption isotherms for juglone constructed by plotting solvent-extractable juglone concentrations (Cs,ext) vs. aqueous phase concentrations (Cw) after a 24-h equilibration are shown in Fig. 1 along with the linear sorption coefficients and the regression correlation coefficients. No juglone was detected in aqueous or solvent extracts of the unsterile A horizon soils. The OC-normalized sorption coefficients (Koc; Koc = Kd foc–1 where foc is the fraction of OC) for the A and B horizon soils are 102.56 and 103.44 L kg–1 OC, respectively. The higher Koc in the B horizon relative to the A horizon is typical of lower OC soils in which contributions to sorption by domains other than organic matter become increasingly significant, e.g., sorption to mineral particles in subsurface soils. The Ryker subsoil has high clay content and low OC content (Table 2) with considerable amounts of oxides and amorphous hydroxides of iron, aluminum and manganese evident in the soil fabric (from pedological analysis on site). These minerals were not quantified for the Ryker soil, but pedogenically-similar soils (Cincinnati series) located nearby contained abundant citrate-dithionate-extractable iron (~7 g kg–1 soil) and aluminum- (~1.5 g kg–1 soil) oxides (Duncan and Franzmeier, 1999). These soil constituents have been implicated as dominant sorption sites for organic acids in soil (Huang et al., 1977; van Hees et al., 2003). Therefore, strong attractions between juglone and the hydroxylated mineral surfaces are likely given juglone's quinoidal and phenolic functional groups. Although juglone is ionizable, pH-dependent speciation is likely to have no impact on sorption; juglone will exist as a neutral species in both soil horizons (pKa > pH +2).


View this table:
[in this window]
[in a new window]

  		Table 4. The distribution of the solute juglone (5-hydroxy-1,4-naphthoquinone) in a Ryker silt loam soil (sterile and unsterile) showing means and standard errors among different sorbed and extractable fractions after a 24-h equilibration in 0.0025 M CaCl2 at T = 22°C.

 

Figure 1
View larger version (16K):
[in this window]
[in a new window]

  		Fig. 1. Reversible sorption isotherms (T = 22°C) for juglone on unsterile and 60Co {gamma} irradiated Ryker A and B horizon soils after a 24-h equilibration in 0.0025 M CaCl2. Reversibly sorbed juglone (Cs,ext) was determined by direct extraction with 1:1 (v/v) ACN/MeOH. No juglone was detected in aqueous and solvent extracts of the unsterile A horizon soil. Kd (L kg–1) is the linear sorption coefficient and R2 is the regression correlation coefficient.

 
Microbial versus Abiotic Degradation during Sorption
In the sorption study, no juglone was recovered in either the solvent-extractable or aqueous fractions of the unsterile Ryker A horizon soil whereas 49 to 66% was recovered from the corresponding sterilized A horizon soil (Table 4). For the B horizon soils, 62 to 70% of the juglone applied was recovered in both sterile and unsterilized soils. Complete disappearance of juglone in the unsterile Ryker A horizon soils relative to the sterile soils exemplifies that microorganisms acclimated with the necessary enzyme systems for degrading juglone were present. Juglone is similar in nature to many soil organic matter components (Maximov and Glebko, 1971). Microbial activity in the surficial Ryker soil samples may be enhanced by the presence of more readily-available organic carbon sources such as plant root products (exudates and extracts) consisting of easily degradable growth substrates such as amino acids, organic acids, or sugars (Rentz et al., 2004). The apparent lack of microbial contributions in the B horizon soil could be related to inherently low soil fertility (e.g., low organic matter and nitrogen content), thus a concomitant effect on microbial activity relative to the A horizon results. The organic carbon content and nitrogen contents of the B horizon soil were low in general and much less than in the A horizon soil (Table 2). Additionally, the low pH in the subsoil (pH = 4.2) relative to the A horizon soil (pH = 6.1) may have played a role in the differences in juglone persistence between the two unsterile soil horizons. De Scisciolo et al. (1990) also noted a higher recovery (thus greater persistence) of juglone from a slightly acidic (pH 5.6) low OC (<0.25%) soil compared to a more neutral (pH = 6.6) high OC (~2.7%) soil (70 vs. 20% recovery after a 2-h equilibration period time using chloroform as the extracting solvent).

For both sterile horizons, substantial amounts (30 to 40%) of juglone were not recovered after the 24-h equilibration. For an autoclaved silt loam soil, De Scisciolo et al. (1990) reported <40% recovery of juglone after 4 h and nearly 100% loss after a 43-h incubation. Juglone is a redox-active compound (Mukherjee, 1987). The quinone functional groups attached to aromatic rings in juglone (Table 1) are characteristically reactive because of their strong affinity for electrons and ability to undergo reversible one-electron reduction to semiquinones (Öllinger and Brunmark, 1991). Juglone's phenolic moiety also takes part in these electron transfer reactions that can result in cross-coupling or binding of molecules to humic constituents through oxidative coupling reactions involving generation of free radicals and catalyzed by oxidoreductases, such as laccases (Chefetz et al., 1998), peroxidases (Dec and Bollag, 2000), or abiotically by metal oxides and clay minerals (Lehmann et al., 1987; Stone, 1987; Huang, 1990). Irreversible binding and transformation processes will reduce juglone availability in soil pore water and reduce toxicity risks to other biota. No additional chromatographic peaks were noted in this study; however, a more thorough investigation is needed to better assess the presence of transformation metabolites as well as their availability and toxicity.

Estimated Pore Water Concentrations of Juglone
To obtain estimates of juglone pore water concentrations (Cpw, mg L–1) for the A horizon under equilibrium conditions, an equation was derived to estimate Cpw from total juglone concentrations in field moist soils (Cs*, mg kg–1), which includes both juglone in pore water and reversibly sorbed (Cs,ext):

Formula 1[1]
Using Eq. [1], the Koc values determined for the Ryker A horizon soil (Koc = 102.56), the average foc value from the bulk soils (1.44% OC for alley edge soils and 1.34% OC for mid alley soils), and a pore water volume to soil mass ratio (Vpw/ms) estimated from the average volumetric water content ({theta}w, 0.28 cm3 cm–3) and bulk density ({rho}b, 1.45 g cm–3) of the field samples, i.e., Vpw/ms = {theta}w/{rho}b, provides a quantitative relationship between Cpw and Cs* for the Ryker A horizon soils (Fig. 2).


Figure 2
View larger version (22K):
[in this window]
[in a new window]

  		Fig. 2. Estimated juglone pore water concentrations (Cpw) in equilibrium with total extractable juglone (Cs*) from Ryker A horizon soils beneath a black walnut agroforestry system at two locations within the cropping alley (1.2 and 4.2 m from the tree row). The solid line is the prediction from Eq. [1] in which the reversible sorption coefficient was estimated using for average % organic carbon values of the two alley positions. Symbols are predicted Cpw values for specifically measured soil concentrations from the field for reference.

 
Cpw values predicted using the Cs* values measured from Ryker A horizon soils beneath the black walnut trees (Table 3) range from 0 to 0.1 mg L–1 (0 to 5.8 x 10–7 M) in the alley edge and from 0 to 1.12 mg L–1 (0 to 6.4 x 10–6 M) in the mid alley row. All these Cpw values exceed the lowest hydroponics threshold concentration reported for juglone of 10–7 M for pine (Pinus strobus L.); however, even the highest value Cpw did not exceed the inhibition threshold of 10–5 M reported from solution plant cultures of typical intercrops such as maize and soybeans (Jose and Gillespie, 1998b) as well as various shrubs and herbs (Rietveld, 1983). The Cpw value estimated from Eq. [1] and used in deriving Fig. 2 is the concentration of juglone that would be present in the pore water of a given juglone-contaminated soil at equilibrium, which is often not the case in the field. Juglone released to soil pore water may be taken up by nearby plant roots before interacting with the soil in which case roots may be exposed to higher concentrations. Alternatively, juglone pore water concentrations may remain lower than estimated equilibrium concentrations if microbial degradation occurs at rates faster than juglone desorption from soil. Overall it is likely that Cpw value estimated from soil concentrations assuming equilibrium likely over estimate the actual exposure concentrations.

Several things, however, should be considered when extrapolating juglone toxicity levels determined from hydroponics studies, which is all that is currently available for assessing juglone toxicity to a variety of plants. In most if not all of the hydroponics studies assessing juglone toxicity, plants are grown for a few days under lighting targeted to simulate natural photoperiods in stagnant solutions prepared at a given juglone concentration. The actual juglone concentrations are usually not quantified initially or over time. Given that juglone is photosensitive and easily degraded biologically, the actual concentration of juglone being experienced by the plants likely decreases substantially over even a few days such that the prepared concentration reported as a toxicity threshold level may be drastically over estimated. In addition, stressed plants may be inhibited by lower juglone concentrations than found to be toxic in otherwise healthy plants. Also juglone may be effective at lower concentrations in the presence of other phytotoxins such as phenolic acids (Blum et al., 1989; Blum, 1998) due to an added toxic activity effect, or as a baseline response from cumulative partitioning to biological membranes in the receptor plant. In the field, decomposition of plant tissues results in release of water-soluble phenolic and polyphenolic compounds (Kuiters and Sarink, 1986; Oliva et al., 2003). For walnut trees, these water-soluble compounds include a variety of phenolic compounds (Gupta et al., 1972) as well as terpenoids (Nahrstedt et al., 1981) and other naphthoquinones (Hirakawa et al., 1986; Binder et al., 1989).


   SUMMARY AND SIGNIFICANCE
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 SUMMARY AND SIGNIFICANCE
REFERENCES
 
Evidence of juglone accumulation under field conditions as exemplified in the current study by measurable and sometimes substantial juglone concentrations in alley cropping system soils clearly shows that release rates of juglone from black walnut trees can be greater than the abiotic and microbial juglone transformation rates. Beneath 19-yr-old walnut plantations, soil juglone concentrations yielded estimated pore water concentrations that approached, but did not exceed, the inhibition solution threshold of crops typically considered for intercropping. However, as noted, there are several factors such as plant stress that may result in inhibition at much lower concentrations. In addition substantially higher levels of juglone can be retained and reversibly sorbed by soil (up to 13 mg kg–1 exemplified in this study), although concentrations will be attenuated by both microbial degradation and abiotic transformation reactions. From greenhouse studies where walnut tree seedlings were planted in sand-filled pots, it is also clear that substantial concentrations of juglone are released into the rhizosphere soils with minimal horizontal transport to the bulk soil. This may also occur in the field; however, rhizosphere soils also tend to have higher microbial activity as well, thus juglone may not accumulate substantially (unlike in the unfertile sand used in the greenhouse pot studies) (Helal and Sauerbeck, 1986). Juglone persistence is particularly short-lived in soils supporting microbial activity. Therefore, the likelihood of juglone-induced inhibition will increase if the rooting systems between alley crops and the trees overlap. It is also plausible that rhizosphere processes of alley crops may invoke remobilization of reversibly sorbed juglone; however, this process has not been assessed. Juglone may also accumulate in subsurface soils due to reduced microbial degradation, thus be available for uptake by deep-rooting plants or upward transport via evapotranspiration to the intercrop root zone. Therefore, although there are many natural processes that will minimize juglone persistence in the field, this research has shown that juglone concentrations approaching levels of concern can occur, particularly in acidic soils that are low in organic carbon and fertility, and should continue to be given some consideration in walnut alley agroforestry system management.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 SUMMARY AND SIGNIFICANCE
 REFERENCES
 





This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by von Kiparski, G. R.
Right arrow Articles by Gillespie, A. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by von Kiparski, G. R.
Right arrow Articles by Gillespie, A. R.
Agricola
Right arrow Articles by von Kiparski, G. R.
Right arrow Articles by Gillespie, A. R.
Related Collections
Right arrow Fate
Right arrow Agroforestry

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Crop Science
Journal of Natural Resources
and Life Sciences Education		 Vadose Zone Journal
Soil Science Society of America Journal Journal of Plant Registrations The Plant Genome