Aster leafhopper survival and reproduction, and Aster yellows transmission under static and fluctuating temperatures, using ddPCR for phytoplasma quantification

Aster yellows (AY) is an important disease of Brassica crops and is caused by Candidatus Phytoplasma asteris and transmitted by the insect vector, Aster leafhopper (Macrosteles quadrilineatus). Phytoplasma-infected Aster leafhoppers were incubated at various constant and fluctuating temperatures ranging from 0 to 35 °C with the reproductive host plant barley (Hordium vulgare). At 0 °C, leafhopper adults survived for 18 days, but failed to reproduce, whereas at 35 °C insects died within 18 days, but successfully reproduced before dying. Temperature fluctuation increased thermal tolerance in leafhoppers at 25 °C and increased fecundity of leafhoppers at 5 and 20 °C. Leafhopper adults successfully infected and produced AY-symptoms in canola plants after incubating for 18 days at 0–20 °C on barley, indicating that AY-phytoplasma maintains its virulence in this temperature range. The presence and number of AY-phytoplasma in insects and plants were confirmed by droplet digital PCR (ddPCR) quantification. The number of phytoplasma in leafhoppers increased over time, but did not differ among temperatures. The temperatures associated with a typical crop growing season on the Canadian Prairies will not limit the spread of AY disease by their predominant insect vector. Also, ddPCR quantification is a useful tool for early detection and accurate quantification of phytoplasma in plants and insects.

'Ca. Phytoplasma asteris' by Macrosteles quadripunctulatus Kirschbaum was investigated at only four constant temperatures 9 , while another study used two constant temperatures to report that the multiplication kinetics of Flavescence doree doubled at 25 °C compared to 20 °C 10 . Nevertheless, experiments that incorporate incremental changes in temperature are more ecologically relevant than those using constant temperatures and ensure that the studied species have time to express their physiological coping mechanisms 11 .
Studies at constant extreme temperatures versus fluctuating temperature showed differences in the thermal limits of insects 12,13 . A variable environment may result in different natural patterns than are evident under the average environmental conditions predicted by Jensen's inequality theory 14 . The sooty copper butterfly, Lycaena tityrus (Lepidoptera: Lycaenidae), develops faster at fluctuating temperatures than it does at constant temperatures 15 . Also, the development time of the diamondback moth, Plutella xylostella (Lepidoptera: Plutellidae), followed the same pattern of shorter development time at temperatures that fluctuate around the mean (0-14 °C) than at a constant mean of 7 °C 13 . Experiments with fluctuating as opposed to constant temperatures may be a more accurate way to assess the effect of temperature on the biology of insects or pathogens 16 . Another potential limitation of other studies is the investigation of a single organism instead of the complex of organisms that are involved in the epidemiology of a disease. Parasites and their hosts often exist in a balanced equilibrium of population densities and their resulting disease epidemiology may also be at equilibrium 17,18 . Provided that synchrony between the development of a parasite and its host is affected by temperature, changing temperatures may be more beneficial to one species than another. Hence, the synchrony between phytoplasmas and their occurrence in an insect vector may be influenced by climatic factors 19 . Other studies have investigated the effect of temperature on phytoplasma and its host plant 11 , and phytoplasma and its insect vector 19 ; however no study has investigated thermal effects on a whole disease system comprising phytoplasma, leafhopper and the host plant.
Various techniques, including ELISA, PCR and RT-qPCR, are used to quantify and monitor the distribution and movement of phytoplasmas in plants [20][21][22] . Droplet digital PCR (ddPCR) technique is becoming more widely used to quantify microorganisms by determining the number of specific genetic markers [23][24][25][26][27] . Relative to other PCR techniques, ddPCR provides advantages in dynamic range, including detection of very low concentration of target molecules in a sample, with a reduced per sample cost 28,29 . Unlike analog quantitative PCR (qPCR) where a standard curve is used to estimate target concentration, ddPCR is an endpoint and absolute measurement approach through which target copy number can be determined using the Poisson distribution 29 . Incidentally, ddPCR is more sensitive than qPCR for the detection of rare target molecules and more accurate at low target copy numbers 23 . The use of ddPCR as a tool to quantify phytoplasma was validated by Mehle et al. 30 , who quantified Flavescence dorée phytoplasma in grapevine by estimating the numbers of phytoplasma per µL of DNA. Perez-Lopez et al. 31 also used ddPCR to quantify phytoplasmas in berries and periwinkle plants. However, no internal control assays were used in these studies. No studies exist on the quantification of phytoplasmas in a vector insect using ddPCR.
The objectives of this study were to investigate: (1) the survival and multiplication of the AY-phytoplasma; (2) the survival, reproduction and disease transmission capacity of its insect vector M. quadrilineatus and (3) the survival of its plant food and reproductive host, barley (Hordeum vulgare L) (Poaceae) through a range of constant and variable temperatures which simulated a natural climate using a unique thermal gradient cell array. The number of AY-phytoplasma present in the plants and insects were quantified using ddPCR allowing absolute quantification 23 .

Results
To assess the effects of temperature, AY-phytoplasma infected and healthy leafhoppers on barley plants were incubated at different constant (0, 5, 10, 15, 20, 25, 30 and 35 °C) and fluctuating temperatures (0-10 °C with a mean of 5 °C, 15-25 °C with a mean of 20 °C, 20-30 °C with a mean of 25 °C, and 25-35 °C with a mean of 30 °C) in individual cells of a thermal gradient system. The survival and reproduction of leafhopper, and the numbers of phytoplasma in insect and plant are presented in this section.
Effect of temperature on leafhopper survival. There was a significant effect (P < 0.001) of temperature on the survival of leafhoppers by the 7 th day, but AY-infection had no effect on leafhopper survival (P = 0.10), with no significant interaction between AY-infection and time (P = 0.12). Similarly, on the 14 th day, there was no significant interaction between AY-infection and time (P = 0.10) and no effect of AY on the survival of leafhoppers (P = 0.50), but there was a significant effect (P < 0.001) of temperature on leafhopper survival. This trend continued until the 18 th day, with no significant interaction between AY-infection and time (P = 0.10), and no effect of AY-infection on survival (P = 0.50), but a significant effect (P < 0.001) of temperature on leafhopper survival (Table 1). On all three sampling days, the highest percentages of surviving leafhoppers were found in assays occurring at constant temperatures of 5, 10, 15, 20 °C and at fluctuating temperatures between 5 and 20 °C. After 18 days, no leafhoppers survived at a constant temperature of 35 °C or a fluctuating temperature with a mean of 30 °C (Table 1). Overall, temperature fluctuation did not have a significant effect (7 th day P = 0.88; 14 th day P = 0.92; 18 th day P = 0.96) on the survival of leafhoppers, with the exception of 25 °C on the 14 th and 18 th days, where significantly more leafhoppers survived at the fluctuating temperature regime than the constant regime ( Table 1).
Effect of temperature on leafhopper reproduction. Some leafhopper adults survived at a constant temperature of 0 °C, but did not reproduce as no eggs or nymphs were found on the barley plants. At

Detection of AY-phytoplasma in barley plants. Barley plants infested with either AY-infected or
-uninfected leafhoppers survived at all temperatures except 35 °C; however, phytoplasma were not detected in any of the barley plants.

Detection and quantification of AY-phytoplasma in Aster leafhopper.
There was no interaction (P = 0.25) between time, temperature fluctuation and temperature on the mean number of phytoplasma present in a single leafhopper. There was a significant increase in the number of AY-phytoplasma in leafhoppers after two weeks of incubation compared to one week of incubation (P = 0.005). However, there was no difference (P = 0.96) in phytoplasma number among the temperatures (Fig. 2). After 18 days of incubation, AY-infected surviving leafhopper adults recovered from bioassay chambers at constant temperatures of 0 to 20 °C, as well as at a fluctuating temperature with a mean of 5 °C, were transferred to healthy B. napus plants. After six weeks, all plants showed symptoms typical of AY disease (Fig. 3). The presence of AY-phytoplasma in those plants was confirmed by ddPCR. There was no significant differences (P = 0.47) in the number of phytoplasma in plants infested with the leafhoppers incubated at various temperatures (Fig. 4).

Discussion
In this study, we used a range of constant and fluctuating temperatures that the causal agent of Aster yellows disease, AY-phytoplasma, the vector, Aster leafhopper, M. quadrilineatus, and its host plants would encounter during a typical growing season on the Canadian prairies 32 . This study represents the first use of ddPCR to quantify the number of AY-phytoplasma in plants (B. napus and H. vulgare) as well as in the insect vector M. quadrilineatus. In addition to phytoplasma detection, the ddPCR assay allows for absolute quantification of the number of AY-phytoplasma in an individual leafhopper specimen and in plants. This quantitative information is crucial for determining the titre of AY-phytoplasma per leafhopper, which signifies the capacity for transmitting phytoplasma to plants. With the B. napus ddPCR primers/probe proven to function here, the absolute quantification of AY-phytoplasma from samples of canola plants is also possible. This information will help to determine to what degree AY-symptom development in canola is related to the AY-phytoplasma present in affected plant tissues. This will allow for predictions of AY-symptoms and impact on yield. The results presented here allowed the evaluation of the replication of AY-phytoplasma in both the Aster leafhopper and two host plants, over time and over a wide range of constant and fluctuating temperatures. Temperature had a significant effect on the survival of leafhopper adults and on their reproductive capacity. Very few leafhoppers survived at temperatures ≥30 °C, while 20% of the leafhoppers survived at 0 °C over 18 days. This indicates that the upper thermal limit for Aster leafhopper survival is approximately 30 °C. A preliminary experiment also indicated that Aster leafhoppers can survive freezing temperatures; however, in this study the lower thermal limit for Aster leafhopper was not evaluated. Viable eggs, subsequently hatched into nymphs, were observed on barley plants grown under a constant temperature of 35 °C, but not at 0 °C. A greater number of eggs were observed on plants grown under constant temperatures of 5, 10, 15, 20 or 25 °C, as compared to ≥30 °C. Thus, the most suitable temperature range for Aster leafhopper survival and reproduction is 5 to 20 °C. Interestingly, the Aster leafhopper is capable of laying viable eggs even at 5 °C, which contrasts with the potato leafhopper, Empoasca fabae (Harris), another leafhopper that migrates into Western Canada, which does not oviposit below 9 °C 33 .
A trend toward better fitness was observed at fluctuating temperatures compared to constant temperatures. For example, the incidence of egg hatches increased at fluctuating temperatures with a mean of 5 and 20 °C, as compared to constant temperatures of 5 and 20 °C. At a constant temperature of 30 °C, all plants had eggs and nymphs, but at temperatures fluctuating around 30 °C the proportion of plants with eggs and nymphs decreased, suggesting that the upper thermal tolerance for Aster leafhopper reproduction is between 31 and 35 °C. The current study also showed that leafhopper eggs start hatching within 18 days at a temperature of 20 °C and above, which is consistent with the findings of Falzoi et al. 34 , who reported that the optimum temperature for egg hatching of the American grapevine leafhopper, Scaphoideus titanus (Hemiptera: Cicadellidae) was 22 °C.
Temperature significantly affected leafhopper survival, but not AY-phytoplasma replication in the leafhopper. On all three sampling days, the highest percentages of surviving leafhoppers were found in assays occurring at constant and fluctuating temperatures between 5 and 20 °C. There was a trend of increasing AY-phytoplasma numbers in leafhopper at constant 5 °C and fluctuating temperatures around 5 and 20 °C. In this study, leafhopper survival was not affected by AY-phytoplasma infection. In contrast, Beanland 35 reported that female leafhoppers lived longer (28 days) on plants infected with AY-phytoplasma than on uninfected plants (19 days). An interaction between temperature and AY-phytoplasma infection of leafhoppers could influence survival. In other studies, survival between infected and uninfected leafhoppers did not differ at constant temperatures of 15 and 20 °C, but there was a significant increase in survival of infected leafhoppers compared to uninfected leafhoppers at 25 and 30 °C 36 . However, the opposite phenomenon occurred when M. quadripunctulatus was infected with Chrysanthemum yellows phytoplasma 37 . Tully et al. 38 reported that most known mycoplasmas failed to survive at temperatures above 34 °C, but the current study found phytoplasma in the bodies of dead leafhoppers incubated at 35 °C. Unfortunately, the death of the leafhoppers at this temperature prevented transferring the leafhoppers to B. napus plants to determine if the phytoplasma was still active. Considering all life cycle parameters measured in this study the temperature range of 5-20 °C was the most suitable for Aster leafhoppers, and whenever leafhoppers survived they successfully transmitted AY-phytoplasma to healthy plants. Within this range, more eggs were laid by infected leafhoppers and more eggs hatched at the warmer temperatures. In S. titanus, an increase in winter temperature induces asynchrony between egg-hatching and budburst of grapevines (Vitis vinifera) showing that warmer temperatures, up to a critical point, accelerate egg hatch 19 .  Temperature-dependent transmission efficiency is a common feature of many vector-borne diseases 39 . In the current study, leafhoppers were able to transmit AY-phytoplasma to healthy B. napus plants and elicit AY-disease symptoms at all temperatures that permitted leafhopper survival. Thus, the insect host and phytoplasma shared the same level of fitness and temperature tolerance under the environmental conditions tested. Similarly, Galetto et al. 40 established that environmental conditions have an effect on phytoplasma multiplication and that this effect is dependent on both the plant and insect host. In the current study, the potential disease-transferring capacity of leafhoppers could not be determined at the three temperature extremes (0, 30, 35 °C) as too few leafhoppers survived to permit transfer to B. napus. Further studies are recommended that begin with a larger number of leafhoppers or a reduction of the length of incubation to ensure that sufficient leafhoppers survive to be transferred to plants. In a preliminary experiment, adult M. quadrilineatus leafhoppers could survive for one week at temperatures as low as -4 °C; therefore, further studies at lower extreme temperatures are also warranted to elucidate the lower threshold for AY-phytoplasma transmission by this leafhopper.
A combination of suitable temperature and wind trajectory brings leafhoppers and AY to the Canadian Prairies 41 . The crop must also be at a susceptible growth stage for infected, migrant leafhoppers to cause an economic loss through AY-phytoplasma infection 42 . Temperatures may not be favourable for leafhopper establishment in years with a long winter and a delayed spring on the prairies. If leafhoppers arrive on early southern winds during unfavourable conditions, they would be unable to survive or reproduce and, hence, would not transmit AY-phytoplasma. This situation is a reasonable explanation for the occurrence of only occasional outbreaks of AY in susceptible crops on the Canadian Prairies 43 . The information in this study enhances our understanding of Aster yellows disease epidemiology. A long winter with a delayed spring may reduce the potential for AY outbreaks for that crop growing season. Similarly, the sharp decline in leafhopper survival at high temperatures may result in a reduction in the spread of AY-phytoplasma later in the growing season under extreme heat conditions. At higher temperatures, even though leafhoppers reproduce before dying, their progeny may not survive the latent period of AY-phytoplasma amplification needed for leafhoppers to be effective vectors. The common temperature range of a typical growing season on the Canadian prairies, though, will not limit AY-phytoplasma transmission via the vector insect, Aster leafhopper.

Materials and Methods
Phytoplasma, vector insect and host plants. The  Barley, periwinkle and canola plants (Brassica napus L. cv. AC Excel) were grown in a growth chamber (temp: 20 ± 1 °C, 16 h L: 8 h D photo regime) in 15-cm-diameter plastic pots containing soil-less mix (modified after Stringham 44 ). Slow release granular fertilizer (26: 13: 0) was added (2 g/L) and the soil was watered daily until seeds germinated and as needed afterwards. For B. napus plants, 3-5 seeds were initially planted and the healthiest seedling was allowed to grow in each pot. Experimental design. Each replicate consisted of 10 barley plants in a 35 ml plastic cup with 30 ml of soilless growing medium 44 . Five days after germination, each cup was placed into a 370 ml plastic transparent plastic cup and 20, 5 day-old AY-infected-adult Aster leafhoppers, collected from the colony cage using a battery operated aspirator were added in each plastic cup (Fig. 5). The plants and insects were incubated at eight different constant temperatures (0, 5,10,15,20,25,30  and several clones were sequenced. The B. napus actin-2 gene (GenBank Accession FJ529167.1) was used as a reference to enumerate the number of AY-phytoplasma vs. plant genome equivalents. The ddPCR primers and probes for the phytoplasma 16S rRNA and B. napus actin-2 genes were designed using PrimerQuest software (Integrated DNA Technologies, Inc.) in accordance with the MIQE guidelines 46 (Table 2). Phytoplasma in leafhopper was quantified using the primer and probe for 16S rRNA and calculated as the phytoplasma copy number per leafhopper.  To quantify the number of phytoplasma in plants, a duplex ddPCR reaction mixture composed of 12.5 µl 2X ddPCR Super Mix for Probes, 16S and actin-2 primers (final concentration of each primer was 900 nM) and probe (final concentration of each primer was 250 nM), 1 µl of digested DNA (10 ng) made to a final volume of 25 µl with ddH 2 O was used. A 20 µl aliquot was used to generate droplets in an 8-well cartridge using a QX100 droplet generator (Bio-Rad, Pleasanton, CA). Droplets were transferred to a 96-well ddPCR plate, sealed with heat seal (Bio-Rad) and amplified in a conventional PCR thermal cycler. The thermal cycling conditions were as follows: 10 min denaturation at 95 °C, 50 cycles of a two-step thermal profile of 30 sec denaturation at 94 °C and 1 min annealing/extension at 58 °C. After amplification, the products were heated to 98 °C for 10 min to harden the droplets and then cooled to 12 °C. Droplets were quantified in a QX100 droplet reader (Bio-Rad, Pleasanton, CA). Data acquisition and analysis were performed using QuantaSoft software (Bio-Rad, Pleasanton, CA). Positive droplets containing amplification products were discriminated from negative droplets by setting the fluorescence amplitude threshold to the lowest point of the positive droplet cluster.
To quantify the copy number of phytoplasma in insects, a single-plex ddPCR was performed with a reaction mixture composed of 12.5 µl 2X ddPCR Super Mix for Probes, 16S primers (final concentration of the primer was 900 nM) and probe (final concentration of each primer was 250 nM), 1 µl of digested DNA (10 ng) made to a final volume of 25 µl with ddH 2 O. The thermal cycling conditions were the same as the plant samples.

Statistical analysis.
To analyze the effects of temperature and AY-infection on the survival of leafhopper adults, a two-way ANOVA was performed in R 47 . Means were separated with a Tukey's HSD test (α = 0.05). The number of phytoplasma in leafhoppers and plants was analyzed with PROC Logistic analysis using the Generalized Linear Model in SAS (SAS Version 9.1).