TIMING OF FUNGICIDE APPLICATION FOR CONTROL OF PERONOSPORA TABACINA, CAUSAL AGENT OF TOBACCO BLUE MOLD, AFFECTS EFFICACY AND FUNGICIDE RESIDUES IN CONNECTICUT CIGAR WRAPPER TOBACCO
Tobacco blue mold, caused by Peronospora tabacina, can quickly devastate cigar wrapper tobacco crops and must be managed with protectant fungicides. We conducted experiments with field-grown broadleaf wrapper and shade-grown Connecticut wrapper tobacco types to evaluate efficacy against blue mold and resulting fungicide residues in cured leaves. Fungicides were applied as weekly season-long programs versus front-loaded schedules where the same total amount of fungicide was applied but with higher rates early in the season. Efficacy against disease was evaluated, and leaves were harvested and cured. Fungicide residues were determined with the QuEChERS method. Blue mold was severe in the first year. Front-loading fungicides reduced disease and fungicide residues in broadleaf and shade tobacco. Blue mold did not occur in broadleaf in the second year, but front-loading fungicides and reducing rates again reduced fungicide residues in cured leaves. Blue mold occurred in only the second half of shade tobacco harvests, and disease severity was greater for front-loaded treatments, which coincided with reduced fungicide residues from those harvested leaves. Models were developed to describe the decline of dimethomorph, azoxystrobin, and mandipropamid to predict the time required since the last application to achieve target ppm residues in cured leaves. Target residues were achieved for broadleaf tobacco but not for shade-grown tobacco in these experiments.
INTRODUCTION
Shade-grown and broadleaf tobacco (Nicotiana tabacum, L.) cigar wrapper types grown in the Connecticut River Valley of Connecticut and Massachusetts are prized as high-quality natural cigar wrapper leaves. However, these leaves must be disease and blemish free to be marketable. Foliar diseases such as blue mold and target spot can result in severe disease and quickly ruin crops under conducive environmental conditions. Blue mold, caused by the oomycete pathogen Peronospora tabacina Adam (Peronospora hyoscyami de Bary), can cause complete loss of cigar wrapper shade tobacco types within 2 weeks (9). Target spot (caused by Thanatephorus cucumeris) was observed at nearly 100% incidence and resulted in complete loss of a broadleaf cigar wrapper crop for certain growers (10). Growers therefore rely on fungicide programs to protect their crops throughout the season, but at the same time attempt to have the lowest fungicide residue possible in harvested cured leaves.
The timing of fungicide application can be critical for both disease management and fungicide residues present at harvest and after curing. Blue mold is most severe and damaging when it occurs early in the season and has longer periods of time for epidemic development than when it occurs late, assuming that conditions are suitable for disease (7, 9, 13). Fungicide applications made after disease was initiated were less effective for sunflower and plum rusts (4, 5) and for downy mildews such as blue mold (6). Multiple fungicide applications are required as new unprotected tissues are produced by ongoing plant growth and fungicide concentrations in and on plant tissues decrease with time. Two types of cigar wrapper tobaccos are grown in Connecticut. Broadleaf wrapper is a topped and stalk-cut type that is sun-grown and harvested once at the end of the season. Shade-grown Connecticut wrapper tobacco is a longer-season crop grown under shade cover that is harvested by removing individual leaves as they ripen from the bottom of the plant to the top. Three leaves per week are usually picked (primed) weekly over 6 weeks. Both harvest approaches have their own challenges in relation to fungicide efficacy against disease and fungicide residues in harvested cured leaves. We conducted experiments in shade-grown and field-grown broadleaf cigar wrapper tobaccos to evaluate efficacy against blue mold and resulting fungicide residues in cured leaves.
MATERIALS AND METHODS
Field-Grown Broadleaf Wrapper Tobacco.
Single-row plots were transplanted with 10 plants per row of Connecticut broadleaf tobacco cultivar “C9” at the Connecticut Agricultural Experiment Station Valley Laboratory Research Farm in Windsor, CT. Plant spacing was 1 m between rows with plants 0.6 m apart within rows. There were 6 replicate plots for each treatment, and plots were bordered by a single row of broadleaf tobacco that was not treated with fungicides. Fertilizer and pesticide applications were performed as per commercial practice. Each year, all plots received nitrogen incorporated pre-plant (224 kg/ha) and nitrogen side-dressed (56 kg/ha) at 24 d after transplanting (5.9-2.8-6.1 N-P-K cottonseed meal base). A standard season-long fungicide protection program was compared to a front-loaded program where the same total amount of fungicide was applied to treatments over the season, but higher rates were used early, and late sprays were omitted or reduced. Fungicide sprays were applied to the appropriate plots using hand-pump backpack sprayers (Solo, Newport News, VA) at 175 to 200 kPa. Dimethomorph (Forum, 43.5% a.i., BASF Corp., Research Triangle Park, NC) and azoxystrobin (Quadris SC, 22.9% a.i., Syngenta Crop Protection, Inc., Greensboro, NC) were evaluated for blue mold management in tobacco field plots. Forum was applied at rates ranging from 147 to 440 ml/ha. Quadris, which is also efficacious against target spot, was applied at a rate of 585 ml/ha at all applications. Transplants were set on July 2, 2012, and plants were topped on August 9 and stalk cut for harvest on August 29. The number of blue mold lesions on wrapper leaves per plot were counted at harvest. Plants were air cured over 8 weeks in a tobacco shed, and cured leaves were analyzed for fungicide residues. The timing of fungicide applications and rates used in 2012 are presented in Table 1. In 2013 the number of replicates was increased to 12, and an additional front-load treatment consisting of half-rate combinations of Forum and Revus (active ingredient mandipropamid 23.3%; Syngenta Crop Protection, Inc., Greensboro, NC) was included. The timing of applications and rates used in 2013 are presented in Table 2. Transplants were set on June 27, 2013, and plants were topped on August 6 and stalk cut for harvest on August 21. Plants were air cured over 8 weeks in a tobacco shed, and cured leaves were analyzed for fungicide residues.
Shade-Grown Wrapper Tobacco.
Experiments were conducted in a shade tent at the Connecticut Agricultural Experiment Station Valley Laboratory Research Farm in Windsor, CT. Shade-grown tobacco in 5 × 10 m plots consisting of 4 planted rows was fertilized annually with cottonseed meal nitrogen incorporated pre-plant (224 kg/ha) and nitrogen side-dressed (56 kg/ha) at 24 days after transplanting (5.9-2.8-6.1 N-P-K cottonseed meal base). Plots were planted with blue-mold-susceptible shade tobacco cultivar “8212” transplants in early June in 4 rows (30 plants per row 30 cm apart in rows with 39 cm between rows). Pendimethalin (Prowl 3.3 E at 2.9 L/ha) was applied as a lay-by-directed spray using an 8004E nozzle at 175 kPa in late June to control weeds. Plants were hand suckered and tied with a string to an overhead wire running the length of the row at the end of June, and the string was wrapped around the plant multiple times as the plant grew for stalk support in early and mid-July.
Fungicide sprays were applied to the 2 inside plot rows using hand-pump backpack sprayers (Solo, Newport News, VA) at 175 to 200 kPa in 225 L/ha (first spray date), 450 L/ha (second spray), and 900 L/ha (sprays 3 through 6) to achieve coverage of spray rows. Fungicides applied included Forum, Quadris, and Revus. The 2 outside rows were unsprayed borders in each plot. Treatments were arranged in a randomized block design with 5 replicate plots of each treatment. Fungicides were applied at approximately 7-day intervals dependent on weather conditions starting 3 weeks after transplanting. The timing of fungicide applications and rates used in 2012 are presented in Table 3. Transplants were set on June 12, 2012, and plants were sprayed with fungicide treatments on June 29, July 6, 13, 20, and 27, and August 3, 9, 16, 23, and 31. Three leaves per plant were harvested (primed) August 9, 16, 23, and 31 prior to fungicide application and on September 6 and 13.
The experiment was repeated in 2013 with fewer fungicide applications at higher rates for front-load treatments, and the timing of fungicide applications and rates used are shown in Table 4. Transplants were set on June 17, 2013, and plants were sprayed with fungicide treatments on June 25, July 3, 9, 16, 24, and 31, and August 6, 12, and 20. Three leaves per plant were harvested (primed) August 6, 12, 20, and 27 and September 3 and 11. Leaves were picked prior to fungicide application. Leaves were sewn, hung from a lath, and air cured in a shed.
For all shade-grown tobacco experiments, ripe leaves were picked, 3 leaves from each plant at each harvest date, and blue mold or target spot lesions, if present, were counted on picked leaves for all plants in the spray rows. The number of lesions per plot on picked leaves and number of healthy leaves (no visible lesions) per plot were recorded for each picking. Border row plants were not picked. Leaves were sewn, hung from a lath, and air cured in a shed. Cured leaves were sampled by cutting a 5-cm strip from middle of the leaf from the leaf margin to but not including the midvein. Samples were dried and crushed, and fungicide residues determined. Data were tested for normality and analyzed by analysis of variance, and means were separated by Fisher’s LSD Multiple Comparison Test (NCSS 12 Statistical Software, Kaysville, UT). In addition, fungicide residue data from the 2013 shade experiment front-load treatments was used to regress decline in residues as a function of weeks since the last application.
Analytical Methods.
Extraction.
Samples were extracted using a modified version of the QuEChERs procedure (1). Dried and crushed tobacco (0.5 g) was combined with 15 ml reverse osmosis water, 15 ml acetonitrile, 6 g of MgSO4, and 1.5 g sodium acetate. After shaking for 30 min the mixture was centrifuged, and an aliquot of the upper acetonitrile layer was filtered and taken for analysis. After injection it was noted that some samples were out of the instrument calibration range. Those samples were further diluted with acetonitrile to bring them into the calibration range (usually 100 μl diluted to 1 ml).
Analytical detection.
Samples were analyzed using an Agilent 1200 Rapid Resolution Liquid Chromatograph interface to a Thermo-LTQ linear ion trap mass spectrometer. The LC was operated with a Zorbax SB-C18 2.0 × 150 mm column that was eluted with a multistep gradient using water and methanol (both with 0.1% formic acid). The gradient went from 95% water to 100% methanol over the course of 16 min. The LTQ was operated in the positive electrospray mode at 3.5 kV energy. Two distinct scan functions were used to monitor each compound. The first provided MS/MS spectra for each compound, and as these spectra yielded a single ion for 2 of the compounds, an additional MS/MS/MS scan was used to further confirm the compounds. All collision energies were set at 35 (arbitrary units). Monitored ions are shown in Table 5.
A tobacco blank was extracted and then used to prepare a set of matrix matched standards ranging from 0.1 to 1.0 ppm for calibration and quantitation. Each compound was quantified in both the MS/MS and MS/MS/MS modes and then the 2 modes were averaged for the value reported and used for all analyses.
RESULTS
Field-Grown Broadleaf Wrapper Tobacco.
Blue mold was first observed in plots about July 25, 2012, and had 4 weeks to infect plants before they were stalk cut on August 29. Untreated control plots averaged over 500 lesions per plot, significantly more disease than for standard or front-load fungicide programs (Table 6). A standard spray schedule reduced the number of lesions by about 80%. Front-loading fungicides earlier in the season further reduced disease, reducing lesions by nearly 90% compared to the control, significantly better than the standard fungicide schedule. Also, as less fungicide was applied late in the season, we observed significantly lower dimethomorph residues in cured leaves when compared to the standard spray schedule. The residues in the untreated controls were not included in the statistical analyses as the zero data artificially reduced the variance.
In 2013, blue mold was not present on the farm until mid-August, a week prior to cutting broadleaf plants. No disease was observed in the experiment. However, front-loading dimethomorph fungicide again significantly reduced fungicide residues in the cured leaves compared to a standard spray schedule (Table 7). Replacing dimethomorph fungicide (Forum) with a half rate each of dimethomorph and mandipropamid (Revus) fungicides further reduced dimethomorph residues and resulted in low residues of mandipropamid.
The decline in dimethomorph residues was examined by regression of residues in cured leaves from 2012 and 2013 against the number of weeks since the last application of full rate dimethomorph fungicide until harvest. The equation describing dimethomorph decline over time (weeks since last application) for full-rate 585 ml/ha application to broadleaf tobacco was
Shade-Grown Wrapper Tobacco.
Blue mold was severe in the shade tent in 2012. The disease was present for more than 2 weeks before the first harvest, and the unsprayed border plants were a persistently strong inoculum source. As a result, disease was present in all primes, and the best fungicide treatments yielded only about 35% of healthy harvested leaves over the 6-week harvest period (Table 8). Front-loading fungicides significantly increased marketable wrapper leaf yields over standard weekly applications. The use of half rates of Forum and Revus was significantly better than using a full rate of Forum alone. Front-loading fungicides increased residues in cured leaves in the early harvests and reduced residues in late harvests compared to the standard fungicide program, which had higher residues in the later harvested leaves (Table 9).
Blue mold disease severity was reduced in 2013. The pathogen was introduced later in the season, about the time of the second prime, 1 week after the last fungicide application in a standard program and 5 weeks after the last application in the front-load treatments. As a result, the disease occurred only on leaves harvested during the last 3 primes (Table 10). In 2013, disease was more severe in front-loaded plots (about 15% of leaves diseased in primes 5 and 6) than in standard plots (about 5% of leaves diseased in primes 5 and 6), which had more recent fungicide application. The disease severity in front-load plots for primes 5 and 6 was about half of disease in the unsprayed border plants, which averaged 280 lesions in the same number of plants (data not shown). This is reflected in the fungicide residues recovered from cured leaves in 2013 (Table 11). In front-load treatments, it took about 5 weeks from the last dimethomorph application to get below 10 ppm residue in cured leaves.
The decline in azoxystrobin, dimethomorph, and mandipropamid residues was examined by regression of residues in cured leaves for each prime against the number of weeks since the last application of each fungicide (Figure 1). As there appeared to be a break point about 4 weeks after the last fungicide application, each fungicide was modeled by 2 equations: for 2 to 4 weeks and for 4 to 7 weeks after the last application.
Citation: Tobacco Science 58, 1; 10.1111/TOBSCI-D-21-00001
The equations modeling Azoxystrobin decline over time for full-rate 585 m/ha application are
The equations modeling Dimethomorph decline over time for full-rate 585 m/ha application are
The equations modeling Dimethomorph decline over time for half-rate 293 ml/ha application are
The equations modeling Mandipropamid decline over time for the half-rate 293 ml/ha application are
DISCUSSION
Management of a potentially devastating tobacco disease such as blue mold that can occur anytime throughout the season while keeping fungicide residues low is an extremely difficult challenge. This is especially true when leaves are harvested rather than washable fruit or tubers and even more so when the leaves are cured and dried, effectively increasing fungicide concentrations by approximately tenfold.
These experiments have demonstrated that timing of fungicide application is critical both for disease management and for management of fungicide residue in cured leaves. Blue mold is most severe and damaging when it occurs early and has a longer time for epidemic development than when it occurs late in the season. Fungicide application after a disease epidemic has started is generally less effective (5, 6), and that is especially true for tobacco blue mold. Multiple fungicide applications are also periodically required as new unprotected tissue is produced by ongoing growth and fungicide concentration and efficacy decreases over time.
Dimethomorph, the active ingredient in Forum fungicide, is a locally systemic a.i. that is xylem but not phloem translocated and can be moved into leaves by absorption into leaf petioles and perhaps stems (3). Concentration in leaf tissues increases over time, suggesting that diffusion through plant and leaf tissue occurs, consistent with reports that soil drench application effectively protected tomatoes from downy mildew (3), and consistent with our regression results indicating a slower decline in dimethomorph concentration for the first 4 weeks after application, perhaps due to redistribution from stalks and petioles into leaves, than for times after 4 weeks.
The Coresta Guidance Residue Level (GRL) for dimethomorph is 2 ppm in cured leaves (2), which would be approximately 0.2 ppm in fresh leaves. This concentration in fresh leaf tissue would be borderline effective at controlling the blue mold pathogen. We have demonstrated reduced sensitivity to dimethomorph by P. tabacina and shown that concentrations of greater than 1.2 ppm in green leaves are required (8). This concentration would result in greater than 12 ppm dimethomorph in a cured leaf. Label rates of dimethomorph are efficacious, but leaf residues resulting from label use rates have been shown to exceed GRLs. This is not unique to cigar wrapper tobacco. Leidy et al. (11) demonstrated that dimethomorph residues in flue-cured tobacco ranged from 2 to 60 ppm 2 to 3 weeks after application and 4 to 10 ppm 6 weeks after the last application at label rates. Six to 8 weeks were required for dimethomorph residues to decline below 2 ppm. Residues in burley tobacco also ranged from 5 to 43 ppm 26 days after application. Our results reported here are consistent with these findings. In shade-grown tobacco, it took 5.7 weeks for dimethomorph residues from full label rate applications to decline below 2 ppm and 5 weeks from half-rate applications to achieve the desired GRL. Dimethomorph concentrations in broadleaf tobacco declined more quickly, reaching the 2 ppm GRL 4.3 weeks after the last full label rate application. This may be due to differences in environmental conditions between broadleaf tobacco grown in full sun compared to shade-grown conditions of 30% shade and higher plant densities (14). The Forum label allows applications of up to 585 ml/ha and a total of 2.19 L/ha per season with zero days to harvest interval, after the spray has dried for crops such as tobacco and pepper. The dissipation curve in pepper predicts a half-life of 2 to 4 days and concentrations decreasing to less than 10% by 14 days after application (12). They concluded that pepper residues would be less than 0.5 ppm, within the range considered safe for consumption, within 7 days of the last application at 1.5 times the recommended rate. This differs dramatically from tobacco in that the initial concentration in pepper 2 hr after application was only 1.8 to 2.8 ppm in fresh tissue and demonstrates that uptake, dissipation, and efficacy differ with different plants. One also needs to note that had the tissue been dried as with tobacco, concentrations would be increased by about 10 times.
Our results demonstrate that it is unlikely to achieve excellent disease control in shade-grown tobacco over an extended 6-week harvest period while reducing fungicide concentrations in cured leaf tissue to levels at or lower than required to be effective. However, these experiments also established the efficacy of front-loading fungicides in broadleaf tobacco to control disease during the period of highest risk while significantly reducing fungicide residues in harvested cured leaves. Azoxystrobin can be effective against both blue mold and target spot and in these experiments effectively controlled target spot in broadleaf with a single application as the first spray after transplanting. Azoxystrobin may cause flecking in broadleaf tobacco and cannot be applied to what will become wrapper leaves without risk of significant injury. Two sprays were applied without damage to shade tobacco, effectively managing target spot throughout the season. Residues were predicted to be below the GRL of 16 ppm (2) for shade tobacco 4 weeks after the last application. Additional research will be needed to investigate methods needed to manage disease with minimal fungicide application in shade-grown tobacco.
Fungicide residue changes in cured shade tobacco leaves over time since the last application. Azoxystrobin and dimethomorph high rate applied at 585 ml/ha; mandipropamid and dimethomorph low rate applied at 293 ml/ha (Table 4). As there appeared to be a break point about 4 weeks after the last fungicide application, each fungicide was modeled by 2 equations; for 2 to 4 weeks and for 4 to 7 weeks after the last application.
Contributor Notes