
We modified the reported DNA extraction methods using a commercial DNA extraction kit: the cetyl trimethylammonium bromide (CTAB) method13 with the addition of skim milk to prevent the absorption of DNA and a bead beating method14. In this report, this method is named the modified CTAB-bead method. The proposed real-time PCR assay may be suitable for the quantification of P. infestans population densities, at least in Japanese upland soils, because P. infestans DNA from various kinds of upland soils was well quantified, and there were no false positives in the negative control plots. Thus, we conclude that the P. infestans population density can be represented by the quantity of DNA determined using real-time PCR. One udifluvent and udult soil quantified slightly small amounts of DNA, and there were small differences among soil types at the same population densities. However, this should not be of great consequence because the differences compared with the other upland soils are within tenfold; thus, these small differences are likely due to the soil characteristics. A previous study reported that no single method of cell lysis or purification is appropriate for all soils15. Thus, the proposed real-time PCR assay is available to quantify the pathogen densities in soils such that most soil samples containing 4–400 zoosporangia/g soil plots except decomposed granite soil and sea sand were quantified as approximately 1–100 pg/g soil. Although this method can be used to quantify P. infestans DNA levels in soil, not all soil samples containing the same number of zoosporangia yielded similar results, as the amount of DNA absorbed was dependent on the soil type. Thus, a calibration curve may be required when a new soil type is tested in which a zoosporangia suspension or P. infestans DNA is added to nondiseased soil. Regarding decomposed granite soil and sea sand, which are not upland soils and not suitable for potato cultivation, the reason for the small DNA quantities may be that a large amount of DNA is absorbed onto silica under Na+– or Ca2+-rich conditions16. If the soil type is sandy or clayey, the DNA quantities may be smaller than those in other soil types. For further development of this method, the addition of an internal control, such as GFP-induced plasmid DNA17, to correct the raw data might be effective. Additionally, changing the glass beads used in this method to zirconia or iron beads may also be effective due to the powerful homogenization and lower amount of DNA absorption achieved with the latter two bead types. However, these improvements may be unnecessary because the proposed assay has a small detection limit such that samples containing only 4 zoosporangia/g soil were detected and quantified. Ristaino et al.18 reported that real-time LAMP and droplet digital PCR can be used to quantify P. infestans DNA from plant tissue. Compared with these tools, the proposed real-time PCR assay has some advantages, such as a wide dynamic range. For this reason, this assay may be widely applied to upland soils.
This is the first report of the quantification of P. infestans population densities in naturally infested soil samples, and changes in the population densities were analysed using real-time PCR. These results also showed that this quantitative method provides reproducible results, because changes in P. infestans DNA were correlated with symptom development throughout the growth periods. DNA quantities during the epidemics (5 and 18 August 2017 and 2018, respectively) were converted into P. infestans population densities in zoosporangia equivalents based on the results obtained for udant B (experimental field, HARC), as shown in Fig. 1; thus, there were approximately 104–105 and 103–104 zoosporangia in the ridgetop soils. These results indicate that a large amount of P. infestans existed in the field ridgetop soils where the plants were blighted. Quantified DNA may be from zoosporangia, mycelia, or small residue or free DNA but not from oospores. Because the A1 mating type has been dominant in Japan since 200519, sexual reproduction rarely occurs, at least in potato fields. We have not verified the availability of soils containing oospores. In future studies, inoculated soil containing oospores should be tested. However, soils containing P. infestans oospores might be quantified using the proposed assay because previous studies reported that soils containing three potato pathogens and oospores of Pythium spp. have been quantified using CTAB and bead beating methods20,21. If the proposed assay can quantify P. infestans DNA from oospores in soil, we might apply this assay to soil diagnosis before planting.
A previous study reported that the inoculum potentials of soil decreased as foliage lesions became less abundant2. Our study corresponds to this previous study because the quantities of P. infestans DNA in soil were consistent with foliage symptom development (in 2017 and 2018) and the number of lesions per plant (in 2018). Hence, the proposed real-time PCR method can be an alternative to bioassays and used as a method to quantify the P. infestans population density. Bioassays require special knowledge and techniques of plant pathology because researchers have to judge whether inoculated tubers were rotted due to P. infestans. On the other hand, real-time PCR assays are easy and require only minimal knowledge and techniques of molecular biology. In 2018, symptom development stopped from late July to early August due to a heat wave. The DNA quantities were reflected in foliage symptoms, with small quantities of DNA estimated during this period. These results imply that this method is highly sensitive for estimating even weekly population changes. The quantities of DNA were decreased to one-tenth of their former numbers in a week after the desiccation of the foliage. As indicated by the decrease, most P. infestans zoospores or zoosporangia cannot survive in/on the soil and quickly die and are degraded by microorganisms and DNase6,22,23. However, if new A2 strains migrated into Japan and oospores were found in field soil, DNA would be detected for a long time even in the noncultivation period. Surprisingly, an infinitesimal quantity of DNA was detected one month after the foliage had disappeared in 2017. This DNA may have been from another plant out of experimental fields or DNA absorbed to some kind of soil material and may persist against DNase24.
Figure 5 shows a positive correlation between the quantity of P. infestans DNA and the inoculum potential. Thus, the proposed real-time PCR method is suitable for indirect estimation of P. infestans inoculum potential. In this analysis, two data sets containing zero values were eliminated as outliers because a zero value in this experiment signifies “below the detection limit”; we cannot determine the exact value. Thus, data sets containing zero values cannot be included in the analysis to evaluate the applicability of the proposed real-time PCR assay instead of the bioassay for estimating inoculum potential. Previous estimation methods, such as bioassays, require an incubation period of approximately 2–3 weeks, expert knowledge of P. infestans and incubation space. On the other hand, real-time PCR requires several hours to estimate the population densities, minimal knowledge of molecular biology and no incubation space. For these reasons, we can more easily estimate P. infestans inoculum potential with real-time PCR than with bioassays. In the experiments using commercial potato fields, a larger amount of DNA was quantified from ridge bottom soils than from any other location. This result agrees with a previous report that most rainwater was deposited at the bottoms of ridges, and the rainwater contained fewer than 500 zoosporangia when blight was present on the crop25. According to this result, soils sampled from the bottom of a ridge are suitable for whole field estimation of P. infestans population densities.
In this study, the quantities of DNA and inoculum potentials were larger in field A than in fields B and C. This result suggests that the proposed real-time PCR assay may be suitable for comparison among potato fields. In field A, late blight occurred because farmers could not conduct chemical control due to heavy rain. Field B was a non-controlled commercial field, and incomplete chemical spraying gave rise to a non-controlled spot in field C. Fields B and C were not perfectly managed for preventing late blight; however, some control, such as cultural or chemical control, was performed in some part. On the other hand, field A did not receive control measures at all. This might be why field A had larger DNA quantities and inoculum potentials than fields B and C.
For the results from non-controlled fields, P. infestans did not percolate through the soil but instead remained at the surface because most soil samples from ridgetops contained larger amounts of DNA than those from the tuber periphery. A previous study reported that more than half of the tubers in the top 5.1 cm of soil were blighted, and the population of blighted tubers decreased with increasing depth26. We can show the same conclusions using real-time PCR. However, in commercial fields, all soils sampled from the tuber periphery contained larger amounts of DNA and had a lower inoculum potential than those from the ridge surfaces. Rainy and cold conditions (approximately at 13 °C) continued from 14 to 18 August 2018, several days before sampling27. The weather might have sustained indirect germination, and many zoospores were released and percolated through the soil. However, zoospores are motile for only a short time28 and cannot survive for a long time. Thus, much of the quantified DNA was from dead P. infestans or free DNA, and less inoculum potential was found near tubers. Soil samples from ridgetops showed larger inoculum potential than those from the tuber periphery. This may be because ridgetop samples contain a large amount of fresh P. infestans from foliage lesions.
In this study, we successfully developed a real-time PCR assay to estimate the P. infestans densities in upland soils, and the proposed assay is available not only for the estimation of population density but also inoculum potential. In the future, this research can provide to a new decision support system for predicting and preventing potato storage rot. The P. infestans soil population density is the most important factor influencing potato storage rot. The possibilities or severities of potato storage rot may be predicted by estimating the P. infestans population density in soil before harvesting. Previous storage planning suggested that potato storage rot might occur if many foliage lesions occur during the growing season. In this study, most P. infestans DNA from foliage lesions degraded within one week. Thus, the possibility and severity of storage rot may be low if the quantity of P. infestans DNA immediately before harvesting is small, even if many foliage lesions occur during the growing season. Additionally, the previous quantitative method (bioassay) requires an incubation period of approximately one week or more2,7. On the other hand, the real-time PCR assay does not require an incubation period, and it takes only several hours to quantify the P. infestans population density in the sample soil. Potato storage rot may be reduced because the storage plan can be selected accurately and rapidly by using real-time PCR compared with previous methods. For example, tubers harvested from fields harbouring high levels of P. infestans DNA can be shipped as soon as possible to prevent potato storage rot. However, many other factors may be involved in the spread of this disease, such as surface injury5. A decision support system would allow potato storage companies to evaluate and address factors associated with potato storage rot and establish appropriate countermeasures to prevent economic losses.