Drought and heat stress can result in aflatoxin contamination of peanuts especially when this occurs during the last three to six wk of the growing season. Identifying drought-tolerant genotypes may aid in development of peanuts that are less susceptible to aflatoxin contamination. Research was conducted to phenotype seven peanut genotypes based on their response to drought stress. Six peanut genotypes that have exhibited lower aflatoxin and/or drought tolerance in previous researches (Tifguard, Tifrunner, Florida-07, PI 158839, NC 3033, C76-16) were compared to an aflatoxin-susceptible genotype, A72. The phenotyping methods included visual ratings, chlorophyll fluorescence (PIABS, φEO, and Fv/Fm), SPAD chlorophyll meter reading (SCMR), normalized difference vegetation index (NDVI), canopy temperature (CT), canopy temperature depression (CTD), and pod yield. Based on these traits, Tifguard and Tifrunner exhibited greater drought tolerance mechanisms than the other genotypes and may be good candidates to be incorporated in future drought tolerance studies. After the aflatoxin content of the different genotypes was measured, aflatoxin contamination showed high correlations with visual ratings (0.85), CTD (0.81), NDVI (0.79), and CT (0.73), and moderate correlations with Fv/Fm (0.62) and SCMR (0.57) (P ≥ 0.05). These easily measurable, rapid and cost-effective phenotyping methods may be used as alternative to more tedious and costly methods of identifying genotypes that are less susceptible to aflatoxin contamination. Using a combination of these methods is beneficial but not always practical. The combined use of visual ratings, CTD and NDVI is advised for initial evaluation of drought tolerance in peanut genotypes.
Peanut (
Aflatoxin contamination in several crops has been repeatedly reported to have adverse effects in livestock and human health. This includes reduced immune system function against infections and diseases, lesser productivity such as reduced milk yield in cattle and decreased egg production in poultry, hepatocellular carcinoma (liver cancer), and death. This consequence has led to significant economic problems for the international peanut trade and high losses to international and domestic producers (
Early breeding efforts for the selection of drought-tolerant genotypes were based on pod yield alone. High-yielding cultivars that continued to produce well under drought conditions were selected as a priority to enable stable production (
Rainout shelters were set up at the National Environmentally Sound Production Agriculture Laboratory (NESPAL) (31°28′32.2″N 83°31′46.2″W) and at the Gibbs Farm (31°26′04.7″N 83°35′18.3″W), Tift County, Georgia during the summers of 2012 and 2013. Both locations have Tifton sandy loam soil (fine loamy, kaolinitic, thermic Plintic Kandiudults) composed of 87% sand, 7% silt, 6% clay and <1% organic matter. Seven peanut genotypes, namely: Tifguard (
The NESPAL trial was conducted in 24 two-row field microplots (1.7 m long by 1.4 m wide with 0.9 m spacing in between rows) equipped with an automated rain-out shelter that was manually closed when rain was expected and moved to their original position when rain was unlikely. Each microplot was installed with four soil moisture (model 200SS, Watermark, Riverside, CA) and three temperature (model 200TS, Watermark) sensors at the pod zone (upper soil layer, 28 cm), four soil moisture sensors at the root zone (lower soil layer, 0.5 m), and PVC pipe planting frames (10.2 cm by 27.9 cm by 1.4 m) on each row. A CR3000 Micrologger (Campbell Scientific, Logan, Utah) was used to monitor and collect hourly data from the soil moisture and temperature sensors. The root and pod zones were separated by an elastic rubber sheet (1 mm thick) stretching from the border walls of the microplot to the outer sides of the planting frames. Two-week old peanuts grown in the greenhouse were transplanted in the middle of the planting frames at a final plant density of 20 plants/m row. The peanut roots penetrated into the lower soil layer without entering the pod zone as prevented by the rubber sheet, while the pegs grew and spread over the upper soil layer. The rubber sheet also prevented the movement of water between zones. The experiment was arranged in split-plot design where three water treatments served as the main plot factor and six peanut genotypes (Tifguard, Tifrunner, C76-16, Florida-07, PI 158839, and A72) as the subplot factor. For the first 100 d after planting (DAP), irrigation at field capacity was supplied through drip irrigation at the root zone and manual spray irrigation at the pod zone. Starting at 100 DAP, treatments included: pod-zone stress (PZS) = drought and heat stress at the pod zone, irrigated at the root zone but not at the pod zone; whole plant stress (WPS) = heat stress imposed and no irrigation provided at the pod and root zones; and well-watered (WW) = no heat stress imposed, irrigated at the pod and root zones. There were three replicates per genotype for PZS and WPS and two replicates per genotype for the irrigated plots. Increased soil temperatures were achieved by passing hot water (30 C) through PVC pipes (1.5 cm dia. by 1.3 cm) installed on both sides of each row (12.7 cm from the middle of the row). Irrigation was provided each time water tension reached 200 kPa in order to bring it back to 180 kPa. In 2013, the authors were interested to assess NC 3033 because of its reported drought tolerance characteristics. Since it was observed from the 2012 trial that PI 158839 exhibited high visual drought stress ratings under WPS, low yield and considerably high aflatoxin contamination, PI 158839 was replaced with NC 3033 at the NESPAL shelter for the 2013 trial.
The Gibbs Farm trial in 2012 was arranged in a randomized complete block design with eight replications. Tifguard, Tifrunner, C76-16, Florida-07, PI 158839, and A72 were planted in two-row rectangular plots (1.5 m long by 1.2 m wide with 0.9 m spacing) at a final plant density of 20 plants/m row. For the first 100 d of planting, irrigation (1.9 to 3.8 cm) was provided as needed. The amount of water provided was calculated based on the amount of rainfall received (measured by a rain gauge set-up within the trial) and the water use required by peanuts at certain DAP for Tifton soil series (
The fungal inocula were prepared using the organic matrix method (
Plant drought stress was rated on a scale of 1 to 5 based on the criteria described in
General criteria used for the visual rating of drought stress on a scale ranging from 1‐5, where: 1 = healthy plants; no symptoms of drought stress; leaves are raised, turgid, green/bright green in color (A); 1.5 = top leaves start to fold/curl/wave; leaf color changes to lighter green/yellowish green (some genotypes may not show color change) (B); 2 = upper branches bend downwards (C); 3 = whole plant bends downwards; leaves start to dry and turn brown (D); 4 = upper canopy dries up; leaves become brittle and thin (E); and, 5 = plants are severely wilted and/or (nearly) dead (F).
Chlorophyll
Chlorophyll content was measured using a SPAD meter sensor (SPAD-502, Minolta, Tokyo, Japan). Readings were taken from the second fully-expanded penultimate leaf of the main stem (
Spectral chlorophyll reflectance was measured using a handheld active sensor reflectance meter (Crop Circle model ACS-210, Holland Scientific, Lincoln, NE). Readings were taken over the middle of each plot from a height of 60 to 90 cm above the canopy at nadir position (0° angle) (
Temperature was measured using an infrared thermometer (Extech IR400, Extech Instruments, Nashua, NH). Four readings were taken from the same side of each plot at an angle of approximately 45° from the horizontal plane, ensuring that different regions of the plot were sampled and the laser was striking the plant leaves (
This was calculated using the equation:
Ambient temperature at the NESPAL shelter was determined using air temperature sensors (model 083E, Met One Instruments, Grants Pass, OR) placed at the corners of the shelter. Thermal imaging taken through the use of a FLIR Thermal Imager/Camera (FLIR bXX series, FLIR Systems, Wilsonville, OR) was also used to measure CT and compute CTD in 2013. Data were downloaded and analyzed using the FLIR QuickReport software. At the Gibbs Farm shelter, AT was instantly measured after four readings from each plot (
At the NESPAL shelter, peanut pods were harvested from each plot at 140 DAP for Tifguard, C76-16, PI 158839, NC3033 and A72 (medium maturing genotypes) and 10 d later for Tifrunner and Florida-07 (medium to late maturing genotypes). At Gibbs Farm shelter, peanut pods were harvested concurrently at 140 DAP. Pods were manually dug and hand-picked at the NESPAL shelter. Harvesting at the Gibbs Farm shelters was done mechanically using a tractor and picker. Pod yield was determined by weighing the harvested pods after cleaning from rocks, soil and other materials then drying to 7% moisture. Yield from the Gibbs Farm shelter was used for quantification of aflatoxin contamination. The NESPAL shelter, however, provided very low yield. Thus, pods were not sufficient for aflatoxin quantification and were saved for a separate study.
Harvested peanut pods from each entire plot at the Gibbs Farm shelter were shelled, ground and mixed thoroughly before representative samples (100 g) were obtained. Aflatoxin content was measured through the standard Vicam fluorometry method. Briefly, the 100 g sample of shelled and ground peanuts were mixed with 10 g NaCl and 200 ml of methanol/water (80:20 v/v), homogenized using a Waring blender (Vicam, Milford, MA) at high speed for 1 min, and filtered through qualitative fluted filter paper (24 cm dia., 25 µm pore size) (Fisher Scientific, Pittsburgh, PA). Five ml filtrate was transferred into 16 mm by 125 mm glass test tube (Fisher Scientific), diluted with 20 ml HPLC water (Fisher Scientific) then re-filtered through glass microfiber filter (11 cm dia., 1.6 µm pore size) (Fisher Scientific). A 10 ml filtrate aliquot was purified with immunoaffinity columns (Vicam) containing aflatoxin-specific (B1, B2, G1 and G2) monoclonal antibodies and washed with 10 ml HPLC water. Column wash with 10 ml HPLC water was repeated then aflatoxin was eluted with 1 ml HPLC grade methanol (Fisher Scientific). The eluted fraction was collected in 12 mm by 75 mm borosilicate glass tubes (Vicam), added with 1 ml fresh Aflatest Developer (Vicam) then swirled in a Vortex mixer (Fisher Scientific) at low speed. Finally, the glass tube was inserted into the fluorometer (series 4EX Fluorometer, Vicam) for aflatoxin quantification. The fluorometer was calibrated using the instructions and aflatoxin calibration standards provided by the manufacturer at the beginning of each day that samples were measured. The HPLC water and reagent blank (1 ml methanol + 1 ml Developer) were also checked on the calibrated fluorometer to make sure that the readings were at 0 ppb.
Individual plots were considered as experimental units where measurements were taken once a week for NDVI and twice a week for visual ratings, chlorophyll fluorescence, SCMR, CT and CTD. Pod yield and aflatoxin contamination were measured from each plot at the end of each season. All the data collected from the NESPAL shelter and the 2013 trial at Gibbs Farm were analyzed using mixed effects ANOVA in SAS 9.2 (SAS Institute, Cary, NC). Water treatment and genotype were considered as fixed effects while replication was considered as a random effect. The data collected from the Gibbs Farm shelter in 2012 were analyzed using one-way PROC ANOVA in SAS where replication was considered as a random effect. Means separation for all analysis was conducted using Fisher LSD test at P ≤ 0.05. Correlation analysis was performed using PROC CORR in SAS to determine relationships among the different phenotyping methods used and with levels of aflatoxin contamination.
Data analysis showed significant genotype by treatment (G × T) effects at the NESPAL shelter in 2012 and 2013. Thus, data are presented separately for each genotype and each treatment (
Meana morning (8:00 hr) and afternoon (13:00 hr) visual drought stress ratings of the peanut genotypes exposed to pod zone stress (PZS), whole plant stress (WPS), and well-watered (WW) treatments and to WPS and WW conditions at the NESPAL and the Gibbs Farm rainout shelters, respectively.
Morning ratings at the Gibbs Farm shelter for Tifguard, Tifrunner, C76-16 and Florida-07 were significantly lower than A72 in 2012 (
The stressed plants exhibited permanent foliage wilting, loss of turgor, leaf color change, leaf shedding, and receding of canopy between rows which were similar to the descriptions of
Some plant species adapt to conditions of drought, high temperature, and high light (typical of field scenarios) by dissipating excess excitation energy thermally and down regulating photosystem II (PSII) activity in order to protect the photosynthetic apparatus (
Results from the NESPAL shelter showed significant G × T effect in both years for PIABS, while only in 2013 for φEO and Fv/Fm. The plants exposed to PZS and WPS generally showed reductions in PIABS, φEO and Fv/Fm as compared to WW conditions (
Meana performance index (PIABS), quantum yield of electron transport (φEO) and maximum quantum yield of photosystem II (Fv/Fm) of the peanut genotypes exposed to pod zone stress (PZS), whole plant stress (WPS), and well-watered (WW) treatments and to WPS and WW conditions at the NESPAL and the Gibbs Farm rainout shelters, respectively.
The observed few differences in the results of PIABS, Fv/Fm, and φEO regarding genotypic responses to drought stress at the NESPAL shelter was also reported by
Correlation between aflatoxin contamination, yield and drought stress evaluation methods of the peanut genotypes planted at the Gibbs Farm shelter.
Maintenance of chlorophyll content under water-limited conditions has been suggested as a mechanism for drought tolerance in peanut. SCMR measures the chlorophyll content per unit area of a leaf through the light absorbance and/or transmittance characteristic of a leaf. SCMR is positively correlated with chlorophyll content and chlorophyll density, thus it can be used to screen for genotypic variation in photosynthetic capacity (
No significant G × T effect was observed from the NESPAL shelter in 2012, but G × T interaction was significant in 2013. A general decrease in SCMR was observed when plants were exposed to PZS and WPS as compared to WW conditions, except for Tifguard which maintained similar SCMR under WPS and WW treatments (
Meana soil plant analysis development (SPAD) chlorophyll meter reading (SCMR) and normalized difference vegetation index (NDVI) of the peanut genotypes exposed to pod zone stress (PZS), whole plant stress (WPS), and well-watered (WW) treatments and to WPS and WW conditions at the NESPAL and the Gibbs Farm rainout shelters, respectively.
This tool uses the visible and near-infrared bands of the electromagnetic spectrum to analyze remote sensing measurements and assess live green vegetation. Healthy vegetation, which correlates to higher photosynthetic capacity and indicated by a higher NDVI, is detected as it absorbs most of the visible light that hits it and reflects a large portion of the near-infrared light. On the other hand, unhealthy or sparse vegetation reflects more visible light and less near-infrared light (
No significant G × T effect was observed at the NESPAL shelter in 2012 where Tifrunner exhibited the highest NDVI (
Canopy temperature examines drought tolerance based on the negative correlation between leaf temperature and transpirational cooling (
Evaluation of CT using an infrared thermometer at the NESPAL shelter in 2012 showed significant G × T effect, but no significant differences among genotypes under PZS and WW treatments (
Meana canopy temperature (CT) and canopy temperature depression (CTD) of the peanut genotypes under pod zone stress (PZS), whole plant stress (WPS), and well-watered (WW) treatments and WPS and WW conditions at the NESPAL and the Gibbs Farm rainout shelters, respectively.
At the Gibbs Farm shelter, the lowest CT in 2012 was also observed from Tifrunner but was not significantly different from most of the other genotypes including A72 (
In relation to CT, CTD measures the deviation of plant temperature from AT. It is used to indicate overall plant water status resulting from the effects of several biochemical and morphophysiological features acting at the stomata, leaf, and canopy levels (
Evaluation of CTD at the NESPAL shelter in 2012 showed significant G × T effect but no significant differences among genotypes under PZS (
The harvested average pod yields ranged from 2049 to 4202 kg/ha (
Mean pod yield (kg/ha)a and aflatoxin content (ppb) ab of the peanut genotypes harvested from the Gibbs Farm rainout shelter under whole plant stress (WPS) and well-watered (WW) conditions.
A wide range of aflatoxin values were obtained from the plots (
All the evaluation methods used, except PIABS and φEO, produced significant correlations with aflatoxin contamination (
Significant correlations were also observed among the evaluation methods. This suggests the interrelatedness of these traits in the plant's mechanism for tolerating drought stress. The significant correlations between visual ratings and Fv/Fm, SCMR, CT, CTD, and NDVI indicate that the effect of drought on other plant physiological traits will likely affect the visual appearance of the plant. The positive correlations between Fv/Fm and all the other evaluation methods indicate that the photosynthetic efficiency of PSII is affected by the chlorophyll content (SCMR), CT, difference between CT and AT (CTD), and NDVI. Similar to the report of
All the evaluation methods used in this study (visual ratings, chlorophyll fluorescence, SCMR, CT, CTD, NDVI, and pod yield) showed significant variation among genotypes in both rainout shelter locations suggesting differences in their sensitivity to detect differences in genotypic response to drought tolerance. Significant G × T effects were frequently obtained from the NESPAL shelter data indicating that the genotypes may behave differently depending on water conditions. This large G × T interaction has been reported to be very common in aflatoxin research and is acknowledged as the main reason for the inconsistent performance of peanut genotypes in response to aflatoxin contamination (
Each evaluation method has its own advantages and disadvantages in the evaluation of drought stress. Visual rating offers an advantage over the other methods considering that no equipment is needed during plant evaluation. However, certain genotypes behaved differently in response to drought stress. Hence, the results can depend on the rater's subjective assessment of the crop status. The equipment used to measure chlorophyll fluorescence, SCMR, CT, CTD, and NDVI are light-weight, easy to use, rapid in giving measurements, and inexpensive. Yet, certain challenges were also faced using these methods. The evaluation of chlorophyll fluorescence required ratings before dawn, and thus, can be very challenging when measuring a large amount of genotypes or plant populations. Results from CT at the NESPAL shelter showed that both infrared thermometer and thermal imaging can be used effectively. The choice of device will be dependent on the amount of area to be measured and the availability of equipment. Using an infrared thermometer is simpler in that one can obtain a measure of CT immediately, whereas infrared images must be analyzed using specialized software at a later date. In contrast, infrared thermometer measurements must be taken quickly as a change in atmospheric factors over time can cause a change in the CT of the plants. Thermal imaging offers the advantage of taking an image and recording the CT measurements of several plant canopies in one shot, thereby eliminating variability in canopy temperature due to slight differences in time of measurement from one plot to the next. However, additional equipment and creative methods are needed to take images at an angle (usually above the plots) that can encompass the plants to be measured. The use of NDVI was very useful but can reflect various plant growth factors instead of exclusively reflecting the effect of one parameter, i.e. water availability (
This project was partially supported by the Agriculture and Food Research Initiative competitive grant 2012-85117-19435 of the USDA National Institute of Food and Agriculture.