Eur J Plant Pathol (2018) 152:711–722 https://doi.org/10.1007/s10658-018-1513-9 Risks for infection of strawberry plants with an aerosolized inoculum of Xanthomonas fragariae J. M. van der Wolf & A. Evenhuis & P. Kastelein & M. C. Krijger & V. Z. Funke & W. van den Berg & A. F. Moene Accepted: 23 May 2018 /Published online: 31 May 2018 The Author(s) 2018 Abstract Xanthomonas fragariae is the causative agent Elsanta and cv Sonata. Results indicate that there is a of angular leaf spot of strawberry, a quarantine organism considerable risk on infections of strawberry plants ex- in plant propagation material in the European Union. posed to aerosolized inoculum. Field experiments were conducted to assess the risks for infection of strawberry plants through dispersal of an . . Keywords Angular leaf spot Air sampling Particle aerosolized inoculum. In practice, pathogen aerosols . . . counters Infection thresholds TaqMan assay Fragaria canbeformedduringmowing ofaninfectedcropor xananasa by water splashing on symptomatic plants during over- head irrigation or rain. In our experiments, aerosols were generated by spraying suspensions of X. fragariae with 8 −1 a density of 10 cfu ml or water under pressure verti- Introduction cally up into the air. In strawberry plants (cv Elsanta) placed at 1.3, 5 and 10 m distance downwind from the Xanthomonas fragariae (Kennedy and King 1962a; spray boom, infections were found, as evidenced with a Kennedy and King 1962b), is the causative agent of combination of dilution–plating and molecular tech- angular leaf spot of strawberry. At present, in Europe niques, but more frequently in plants wetted prior to the pathogen is listed as a quarantine organism on straw- inoculation than in plants kept dry. A logarithmic de- berry plants intended for planting (Anonymous 2006). crease in infection incidence was found with the dis- Infections can result in high economic losses as plants tance to the inoculum source. Symptomatic plants were should be removed and destroyed upon disease out- found up to 5 m distance from the inoculum source. No breaks (Desmet et al. 2006). In the Netherlands, one of infected plants were found in plants placed 4 m upwind the biggest producers of strawberry planting material in or treated with water. In glasshouse studies, it was Europe (Lieten 2014), legislation is in place about the shown that under conditions favorable for disease de- radius of the buffer zone to be cleared around disease foci. velopment, spray-inoculation of strawberry plants with Management of the disease is mainly based on ex- estimated densities of X. fragariae as low as 2000 cfu clusion of the pathogen via cultivation practices and per plant were able to cause symptoms both in cv hygienic measures. No chemical or biological control agents are currently available to control the disease, but on a small scale, thermotherapy is applied to reduce : : : J. M. van der Wolf (*) A. Evenhuis P. Kastelein infection pressure (Turechek and Peres 2009;Van : : : M. C. Krijger V. Z. Funke W. van den Berg A. F. Moene Kruistum et al. 2015). All commercial cultivars are Wageningen University and Research, PO Box 16, susceptible to X. fragariae although at a various level 6700 Wageningen, AA, Netherlands e-mail: Jan.vanderWolf@wur.nl (Desmet et al. 2009; Turechek and Peres 2009). 712 Eur J Plant Pathol (2018) 152:711–722 Strawberry cultivation starts with in vitro material and temperature (Kennedy and King 1962b; and the first generations of strawberry plants for plant- Hildebrand et al. 2005). Long periods of rain, irrigation ing, representing the highest classes, are grown in aphid- or dew favour the disease (Maas 1998). The assumption free glasshouses, where the risk for infections with is that leaf wetness is necessary for a successful infec- X. fragariae is considered low (Van der Gaag et al. tion. Moist conditions also favour exudation of the 2013). In Europe, the last two generations of plant pathogen from lesions (Maas 1998). material are often grown in the field which increases The aim of this study was to assess the risks for the risk for infection. infections and disease development after spread of aero- A field-grown crop that is initially free of the patho- solized inoculum onto strawberry plants at various dis- gen can become infected via different pathways which tances from the source. In field experiments bacteria may include contact with contaminated machineries, were released and the maximum distance estimated at materials, animals, shoes and clothes (Maas 2004), which dissemination resulted in an infection of straw- drenching of planting material in preventive fungicide berry plants. Strawberry plants were either wetted or baths (Melis et al. 2012), use of contaminated irrigation kept dry prior to inoculation, to vary in leaf wetness water, carry-over contamination from infected crops conditions. In addition, glasshouse experiments were grown nearby via splashing water, or via aerosols conducted to determine the minimum inoculum pres- (EPPO 1997; Van der Wolf et al. 2017). sure to establish an infection. Contaminated aerosols can be generated during splashing of water on symptomatic plants, which can exude large quantities of X. fragariae (up to 10 cfu) Materials and methods during spraying of plants with an excess of water (un- published data). Aerosols can also be generated during Xanthomonas fragariae and culturing mowing of a strawberry crop at the end of the growing season (Van der Wolf et al. 2017). Mowing of strawber- A natural Rifampicin resistant strain (designated ry crops is a cultivation practice to lower transpiration IPO3488) of X. fragariae isolate PD 3145, obtained in rate (Rätsep et al. 2015), to renovate plants after the first 1997 from strawberry in Spain, was used in the field harvest (Rätsep et al. 2015) or to remove the excess of experiments. Strain IPO3488 was kindly supplied by leaves prior to low temperature storage of so-called frigo Dr. H. Koenraadt of the Netherlands Inspection Service plants. for Horticulture (Naktuinbouw). In preceding pathoge- Dispersal of pathogens via aerosols has been shown nicity tests, strain 3488 proved as virulent as the parental to play a role in the epidemiology of various plant wild type strain (data not shown). pathogenic bacteria including Pseudomonas syringae Strain 3488 was stored on beads (Protect bacterial (Morris et al. 2007), Pectobacterium and Dickeya spe- preservers, TS/70; Technical Service Consultants Ltd., cies (Perombelon et al. 1979; Franc and DeMott 1998) Lancashire, UK) at −80 °C. Three to 4 weeks before and bacterial pathogens of tomato (McInnes et al. 1988). starting the experiment the strain was revived on Tryptic It was hypothesized that aerosols can be responsible for Soy Agar (Difco, USA) at 25 °C and maintained at dissemination over a distance of at least 100 m from the 17 °C by monthly transfers on YDC medium (Duchefa −1 inoculum source (Perombelon et al. 1979). It was found Biochemie, NL) with 50 mg l Rifampicin (Duchefa that plant pathogenic bacteria, i.e. Pectobacterium spe- Biochemie). Inoculum was prepared by growing the cies, can act as cloud condensation nuclei (Franc and strain on glycine amended R2A Agar (R2AG: −1 −1 DeMott 1998). If aerosolized bacteria are transported 18.12 g l R2A Agar, Difco USA, and 25 mg l gly- −1 into cloud systems, they can move over much larger cine; Sigma-Aldrich, USA) with 50 mg l Rifampicin distances before they will be deposited in precipitation. (Duchefa Biochemie, NL) for 3 days at 25 °C and Several factors are described that are involved in washing the cells from the agar with a quarter-strength disease development which include the cultivar (Pérez- Ringer solution (Oxoid, UK). Jiménez et al. 2012; Rivera-Zabala et al. 2017), the To check for the presence of X. fragariae in air or susceptibility of the plant (Kennedy and King 1962b), leaves, 50 μl concentrated air sample or leaf extract was the virulence of the pathogen (Rivera-Zabala et al. 2017) plated undiluted and ten-fold diluted in quarter-strength −1 and environmental conditions, in particular humidity Ringer solution on R2AGRC; R2AG with 50 mg l Eur J Plant Pathol (2018) 152:711–722 713 −1 Rifampicin and 200 mg l Cycloheximide (Duchefa and severity in plants for each cultivar separately. For Biochemie). Plates were incubated for 8–10 days at the analysis of the effects on incidences angular- 25 °C before inspection for the presence of transformed values were used. Duncan’s new multiple Xanthomonas-like colonies (circular, convex, glistening range test was used for evaluating the significance of and translucent to pale-yellow). differences between averages within cultivars. Glasshouse experiments Field experiment Two experiments were conducted to determine the low- On 25 May and 13 June 2016, during dry spells on two est inoculum density needed to cause angular leaf spot rainy days, an experiment was conducted to assess the in strawberry leaves. The first experiment was conduct- risk for infection of strawberry plants after aerosol dis- ed in March–April 2013 and the second in October– persal of X.fragariae. November 2014. Plants of cultivars Elsanta and Sonata were grown in a glasshouse at 17 °C and 65–70% RH. Strawberry plants and cultivation In 2014, daylight was prolonged to 14 h. Approximately 20 days after planting of the cold (−1.5 °C) stored On two time points in April 2016 cold (−1.5 °C) stored certified waiting bed plants, 1-L pots containing Lentse certified waiting bed plants of cv Elsanta were planted in potting soil no.3 (Horticoop, NL), two – three fully 11x11x12 cm TEKU pots (Pöppelmann, DE) filled with expanded leaves were present. During plant growth, Lentse potting soil. The interval between both planting stolons and inflorescences were dissected from plants. dates was 2 weeks. In 2014, plants were treated against mildew with The first 4 weeks after potting the plants were placed in Bupirimate (Nimrod; Adama, IL) according the manu- the open air on weed control fabric. Thereafter the plants facturer’s instructions. Approximately 35 days after were grown under a rain shelter with roofing of polyeth- planting, when the first flower branch was dissected, ylene greenhouse film. Initially the plants were kept on plants were inoculated with different inoculum densities benches, but after being used in an aerosol experiment of X. fragariae. A stock suspension in 0.3% (v/v) Silwet they were placed on the floor of insect-proof nylon cages 719 (Momentive, USA) was set to an absorbance value in the same rain shelter. Till their use in aerosol experi- 8 −1 of A600nm = 0.1 (approximately 10 cfu ml )bydi- ments the plants were watered and fertilized in line with luting bacterial inoculum prepared in the lab to the prevailing horticultural standards. After the aerosol exper- desired inoculum density. Inoculation was done by at- iment the plants in the insect-proof cages were watered omizing either undiluted, 100 x, 10,000 x, or 1,000,000 only via irrigation mats to avoid splash dispersal of x diluted stock suspension onto the abaxial side of the X. fragariae by overhead irrigation. leaves using a high pressure plant sprayer (Gardena, DE). Twenty to 25 ml of bacterial suspension was Experimental site and weather measurement applied per plant. Mock inoculated plants were sprayed with 0.3% Silwet 719. After inoculation, strawberry The experiment were carried out on a well-cut lawn at plants were placed in a plastic tent for maintaining high Nergena experimental farm near Wageningen, the Neth- moisture conditions. One day after inoculation (dpi), the erlands in an area of the Netherlands were no straw- tent was removed and plants were distributed in the berries are grown on a commercial scale. A moveable glasshouse in five blocks. At 14 and 28 dpi every leaflet weather station (Decagon Devices Inc., USA; EM50 was examined for the presence of symptoms of angular datalogger) was placed on the experimental site to mea- leaf spot. Furthermore, at 28 dpi for each leaflet the sure wind speed, wind direction at 50, 111 and 220 cm percentage of necrotic leaf area was estimated and the above the soil level (Davis cup anemometer). Further- more, air temperature and relative humidity were re- severity indexes calculated, i.e. the average percentage necrotic leaf surface times the number of affected leaf- corded on site at 150 cm above soil level (Decagon lets per plant. EHT sensor). Data of other atmospheric variables (sun- Analysis of variance (ANOVA; Genstat 18.1, VSN shine duration and global radiation) were obtained from International, UK) was used to analyse the effect of the Veenkampen weather station at 2.9 km beeline dis- X. fragariae inoculum density on disease incidence tance from the experimental site. 714 Eur J Plant Pathol (2018) 152:711–722 Release of inoculum At each release of inoculum, a new batch of straw- berry target plants was used. The youngest fully expand- One day in the experiment consisted of successively a ed leaves, which are most sensitive to infection (Hazel dummy run, a run with spraying tap water and two runs and Civerolo 1980;Hildebrand et al. 2005), had been with spraying X. fragariae suspensions into the air. tagged a few days before the experiments were carried About 15 to 20 min before the release of aerosols, air out, to support disease assessments and sampling of was sampled for 10 min to investigate if X. fragariae exposed leaves later on. was naturally present in the air. Thereafter, on each day Plants exposed to aerosols were removed from the one 2 L portion of tap water and two 2 L portions of experimental site before starting the next spray session. X. fragariae suspension set to an absorbance value of They were temporarily placed at a site out of reach of A600nm = 0.1 (approximately 10 colony-forming units newly generated aerosols. During transfer dripping of −1 ml ), were sprayed vertically up into the air with a water from leaves and contact between plants was spray boom kept in inverted position. Inoculum was avoided. Directly after removal of exposed plants, the sprayed with a pressure of 3 bar through 6 Teejet XR same high pressure plant sprayer was used as in the 110–02 VP extended rate flat spray nozzles (Teejet glasshouse experiment to atomize X. fragariae suspen- Technologies, USA) with in-between distances of sion on both sides of the labelled leaves of three control 60 cm. The spray boom was situated at a height of strawberry plants until runoff to assess the susceptibility 23 cm above ground level and the water droplets re- for infection during actual field conditions. At the end of leased by the spray nozzles reached a height of approx- the day all the plants were transported back to the rain imately 150 cm. With each 2 L portion of water or shelter and the different groups (distance from the spray X. fragariae suspension pulses of aerosols were gener- boom and leaf treatment) of plants were placed in sep- arate insect-proof cages. Furthermore, a sample of the ated during approximately 30 s in the first experiment. In the second experiment, due to variation in wind X. fragariae suspension used that day was atomized on direction, pulses of aerosols of irregular duration were the abaxial side of the tagged leaf of three plants kept in given over a period of 30 to 120 s. Thus, in two days a total of four runs were carried out with X. fragariae suspensions being sprayed into the air. Upwind Downwind (80) Strawberry plants were placed at various distances from the inoculum source in order to establish whether (40) aerolised X. fragariae was infectious. Strawberry plants were arranged along three curved lines at various dis- (10) (6) tances downwind and one curved line upwind of a spray boom serving as aerosol source. During spraying tap water with the spray boom (source) six strawberry plants (target plants) were located at 4 m windward and ten strawberry plants at 1.3 m leeward of the spray boom. When X. fragariae suspension was sprayed, additional- ly 40 strawberry plants were placed along curves at 5 m 1.3 m distance leeward from the spray boom and 80 plants at 10 m distance. The target plants were placed along a 4 m semi-arc such that each plant was placed at a similar 5m distance from the source. The lay-out of the experiment is described in Fig. 1. The number of plants increased with distance from the source to cover the expansion of 10 m the aerosol plume with increasing distance. To mimic Fig. 1 Design of the field experiment. Strawberry plants were recently-fallen rain, the foliage of half of the plants had placed at one distance upwind and at various distances downwind been wetted with tap water shortly before setting them from the spray boom used to release inoculum (suspension of out in the field. Wet and dry plants were arranged Xanthomonas fragariae). Between brackets: number of plants alternating without coming in contact with each other. per distance Spray boom Eur J Plant Pathol (2018) 152:711–722 715 a humid chamber to check for the infectivity of the at 1.3, 5, 10, 25 and 50 m distance leeward from the X. fragariae suspensions. source, at a height of 45 cm above ground level was recorded with Dylos DC1700 air quality monitors Air sampling and quantification of air particles (Dylos Corporation, USA). Furthermore, an air quality monitor at 4 m windward of the aerosol source was used Information on the spread of X. fragariae in aerosols to record the natural background level of particles in the was quantified by using two methods. Coriolis Micro air air. The particle counters assess small (> 0.5 μm) parti- samplers (Bertin Technologies, FR) were used to collect cles and large particles (>2.5 μm). Recordings are the aerosolized X. fragariae Dylos DC1700 air quality average of 10 s measurements. Each minute 6 readings monitors (Dylos Corporation, USA) were used to quan- of the number of particles in the air are stored in the data tify particles in the air. Equipment was placed at various base. During the experiment continuous readings were distances from the inoculum source. made. A background number of particles was recorded During the spray runs, two Coriolis samplers were during the experiment when no suspension was released situated at 1.3 and 5 m distance leeward from the inoc- in the air. When the suspension was released the number ulum source. The capacity of the Coriolis sampler was of particles recorded peaked. To quantify peaks in the set at assessing 300 L air per minute. During a period of particle counts the number of particles at the time of 3 min starting at the beginning of aerosol generation the suspensions release visible as peak values were added microflora present in total of 0.9 m air at a height of (two readings). To compensate for the background par- 47 cm above ground level was collected in sterile cones ticle density naturally present in the air the number of filled with 15 ml RT (quarter-strength Ringer solution particle counts before the onset of the peak and directly with 0.01% Tween20; ThermoFischer Scientific, USA). after the release event were subtracted from the peak values. Thus an estimate of the particles in the air due to The density of the bacterial cells in the air sam- ples collected by the Coriolis Micro air sampler was the release of a X. fragariae suspension was calculated. concentrated 10-fold by centrifugation. Samples were transferred to sterile 50 ml Nunc conical centrifuge Sampling of strawberry target plants and processing tubes (ThermoFisher Scientific), spun at RCF 8867 x g for 10 min at 10 °C in a Fiberlite F-15-6x100y Three weeks after the release of aerolised inoculum and rotor of a SL40R benchtop centrifuge (ThermoFisher deposition on the strawberry target plants, the tagged Scientific), after which the supernatant was drained leaves and the leaves unfolded just before and just after and the pellet suspended in 1.5 ml quarter-strength the tagged leaf, were inspected for the occurrence of Ringer solution. To check for the presence of symptoms of angular leaf spot. The number of infected X. fragariae in the air sample, undiluted and 100× plants with and without symptoms was assessed. After diluted suspension was plated on R2AGRC. After the disease assessment the tagged leaves were sampled incubation as described above in ‘Xanthomonas to test for X. fragariae infections. From each target plant fragariae and culturing’ the plates were inspected that had been located 4 m windward or 1.3 m leeward of for presence of Xanthomonas-like colonies. The the spray boom the complete leaf was cut off and proc- identity of a random selection of Xanthomonas-like essed. From target plants that had been located 5 or 10 m colonies was checked by TaqMan assay. Based on leeward of the spray boom only one leaflet of the tagged the colony counts the density of X. fragariae cfu in trifoliate leaves was cut off. These leaflets were proc- air was estimated. In the second experiment at 1.3 m essed in batches of 4 leaflets for plants at 5 m and 8 the number of bacterial cfu was higher than 33,000, leaflets for plants at 10 m. the upper threshold. For the statistical analysis this Each one-leaf or composite leaflet sample was trans- upper threshold was used. ferred to a universal extraction bag (Bioreba, CH) and crushed using a hammer. Directly after crushing, a vol- Particle sampling in the air and processing ume of Wilbrink’s solution (Koike 1965)equivalent to 5 mL plus 1.3 times the sample weight was mixed During the spray runs the extent of decrease in the through the macerated tissue. Wilbrink’s solution −1 −1 density of water droplets in the air, due to the expansion consisted of 10 g L sucrose (Sigma-Aldrich), 5 g L −1 of the aerosol plumes and evaporation of water droplets, proteose peptone (Oxoid), 0.5 g L K HPO (Sigma- 2 4 716 Eur J Plant Pathol (2018) 152:711–722 −1 Aldrich), 0.25 g L MgSO .7H O (Sigma-Aldrich), composite samples tested, and n the number of straw- 4 2 −1 0.25 g L NaNO (Sigma-Aldrich). To check for the berry leaflets combined into a composite sample (De presence of X. fragariae in theleafextract,anundiluted Boer 2002). and 100× diluted suspension was plated on R2AGRC. Analysis of variance (ANOVA; Genstat 18.1, VSN After incubation as described above in ‘Xanthomonas International) was used to analyse the effect of fragariae and culturing’ the plates were inspected for X. fragariae inoculum density on symptomatic infec- presence of Xanthomonas-like colonies. The identity of tions of strawberry plants cultivars Elsanta and Sonata a random selection of Xanthomonas-like colonies was under greenhouse conditions. Fisher protected pairwise checked by a TaqMan assay. T-tests were used for evaluating the significance of differences between pairs of averages within cultivars. ANOVA (with angular-transformed incidences; TaqMan assay Genstat 18.1) was used to analyse the effects of distance from the inoculum source and leaf wetness on the oc- A colony-TaqMan assay was used to confirm the identity currence of (symptomatic) infections with X. fragariae, of Xanthomonas-like colonies growing on R2AGRC Fisher protected pairwise T-tests were used for evaluat- plates seeded with an air sample or leaf extract. Bacterial ing the significance of differences between pairs of cells from Xanthomonas-like colonies were sampled averages. Water controls were omitted from the analysis. with an inoculation needle, suspended in 1 mL sterile ANOVA (with log-transformed numbers) was used water in 1.2 mL collection tubes (QIAGEN). In addition, to analyse the effect of distance from the X. fragariae a so-called bio-TaqMan assay was used to verify the source on the number X. fragariae cfu in air assessed presence or absence of X. fragariae colonies on plates with Coriolis air samplers at 1.3 and 5 m. The untreated on which no Xanthomonas-like colonies were detected control as a measure of the background X. fragariae during visual inspection. In the bio-TaqMan assay, plates population was included in the analysis. were flooded with 3 mL sterile water and the bacterial colonies dislodged from the agar with the aid of an L- shaped spreader. Depending on the number of colonies Results 1 mL undiluted suspension, or diluted to a slightly clouded suspension, was transferred to a 1.2 mL collec- Glasshouse experiments tion tube. Next bacterial suspensions were centrifuged for 15 min at 5800 RCF in a 4–15 C centrifuge (Sigma) and In 2013, strawberry plants of varieties Elsanta and So- 980 μl supernatant was removed from each tube before nata were spray-inoculated with suspensions of storage of the pellets at −20 °C until further processing, 2 4 6 8 −1 X. fragariae of either 0, 10 ,10 ,10 or 10 cfu ml . DNA extraction from the pellets and the TaqMan At 15 dpi, even the lowest inoculum density resulted in assays were conducted as described by Kastelein et al. symptomatic plants, although at a low level (Fig. 2a). A (2014). Suspensions of which amplification plots positive relation was found between the level of the showed CT-values >35 were considered negative. inoculum density and the percentage of affected plants. At the lowest inoculum density, Sonata had a higher Data processing and statistics disease incidence, i.e. percentage affected leaves −1 plant , than Elsanta, but at the highest inoculum den- The data of the disease assessments were used to calcu- sity, Elsanta (P =0.05) was more affected. late the disease incidence (expressed as percentage) of At the two highest inoculum densities, the disease symptomatic plants at the three distances leeward from severity, i.e. amount of leaf necrosis, was higher for the source of infection. The results of plating leaf ex- Sonata than for Elsanta (P = 0.05) (Fig. 2b). After tracts were used to estimate the infection incidence (I)of mock-inoculation, no symptomatic plants were found. strawberry plants using the formula In 2014, the experiment was repeated. In this exper- no 1=n iment, only for Elsanta symptomatic plants were found I ¼ 1–½ ðÞ N –p =N 100 at the lowest inoculum density (Fig. 2-c). In this exper- where p is the number of composite samples that tested iment, overall the disease severity was higher for Elsanta positive for X. fragariae, N the total number of than for Sonata (Fig. 2-d). Eur J Plant Pathol (2018) 152:711–722 717 cc A C dc cb cc 50 50 40 40 Elsanta Elsanta bb 30 30 Sonata Sonata ab 20 20 ba ab a 10 10 aa aa 0 0 0 10E2 10E4 10E6 10E8 0 10E2 10E4 10E6 10E8 Inoculum dose (cfu/ml) Inoculum dose (cfu/ml) 1800 1800 b c B D 1600 1600 1400 1400 ab 1000 1000 bb 800 Elsanta Elsanta 600 600 Sonata Sonata 200 200 aa aa aa aa aa aa a a 0 10E2 10E4 10E6 10E8 0 10E2 10E4 10E6 10E8 Inoculum dose (cfu/ml) Inoculum dose (cfu/ml) −1 Fig. 2 Disease incidences, percentages affected leaflets plant , results obtained in 2013 and C and D in 2014. A and C show determined at 15 dpi and severity indexes, i.e. average percentage disease incidences and B and D severity indexes. Statistical anal- −1 necrotic leaf surface times number of affected leaflets plant ,at40 ysis was done per cultivar, per year, separately for incidences and dpi observed in two-years glasshouse experiments using two severity indexes (P = 0.05). Error bars show standard deviations. cultivars of strawberry plants (cv. Elsanta and cv. Sonata) after Average disease incidence and severity index values for the same spray inoculation with a hundred-fold serial dilution series of a cultivar with the same letters are not significantly different 10 cfu/ml suspension of Xanthomonas fragariae. Aand Bare (Duncan’s multiple range test, P = 0.05) Field experiment 220° during the experiment but was stable at the time scale of an individual dispersion experiment (single re- Weather conditions during the field experiment lease). On 13 June the wind direction was more constant between experiments (250° to 265°) but was much more The weather conditions during experimental work in the variable on the time scale of a single release. This expe- field are summarized in Table 1. On the first day of the rience in the field is not clearly reflected in the standard experiment, 25 May 2016, the sky was overcast and no deviation of wind direction as shown in the table, prob- sunshine was observed. On the second day, 13 ably due to the limitations of the wind vane to follow the June 2016, the sky was mostly overcast, but with variations in wind direction. Because of these variable patches of a thinner layer of clouds, allowing for an wind directions on the second day, spraying of water and insolation that was overall higher than on May 25. On X. fragariae-suspension into the air was interrupted sev- both days the relative humidity during the experiment eral times to achieve a more regular dispersion of aero- was around 80% but the days differed in terms of the air sols in the main wind direction. temperature during the experiment: around 13 °C on May 25, and 17–19.5 °C on June 13. Splash and aerosol dispersal Another marked difference between both days was the wind speed at 2.20 m height. During the first day, a low The use of a spray boom resulted in the dispersal of both wind speed was measured ranging from 1.2 to 2.0 m/s, small sized (aerosols) and larger droplets. The larger whereas on the second day the wind speed, especially droplets were dispersed over a distance of at least during the two X. fragariae sprays, was higher at around 1.3 m, as droplets were observed on strawberry plants 4 m/s. The wind direction at 25 May ranged from 180° to at that distance. However, after the experiment, no water Avg % necrosis x aﬀected leaﬂets Avg % aﬀected leaﬂets/plant Avg % aﬀected leaﬂets/plant Avg % necrosis x aﬀected leaﬂets 718 Eur J Plant Pathol (2018) 152:711–722 Table 1 Characteristics of weather conditions during experimen- direction (Udir, relative to North) and standard deviation of wind tal work in the field: temperature (T), relative humidity (RH), wind direction (StdUdir) speed at 2.20 m (U), standard deviation of wind speed (stdU), wind Date Time Spray T RH U stdU Udir stdUdir °C % m/s m/s ° ° 25–5-2016 12:35 Water 13.2 80 1.2 0.43 207 6.4 13:10 X. fragariae 13.2 81 2.0 0.29 181 17.5 13:40 X. fragariae 13.1 82 1.7 0.29 219 10.7 13–6-2016 15:20 Water 18.7 80 2.3 0.20 250 15.6 16:10 X. fragariae 19.1 76 4.0 0.69 263 9.7 17:20 X. fragariae 17.3 77 3.7 0.52 251 12.4 Values given are 15-min averages centred around the time of aerosol release, based on data sampled at 1 sample per minute local time halfway 10–15 min experimental work droplets were found on dry strawberry plants placed at a 10 m were (almost) absent and consequently the particle distance of 5 and 10 m. Obviously, on pre-wetted straw- counts in the peaks could not be assessed reliably. The berry plants water droplets were present at all distances. reason for this absence of a clear peak is two-fold. First, On both days, no X. fragariae was detected by dilu- on the second day of the experiment the release was not tion plating in any of the air samples collected with continuous and spread over a longer time so that the Coriolis samplers prior to release of the inoculum, or plume of particles was not as clearly defined in space when water was sprayed. When X. fragariae suspensions and time as for the first experiment with a continuous were released into the air the pathogen was detected in air release. Secondly, the high variability of the wind speed samples collected at both 3 and 5 m from the source. The during the second day (standard deviation of 0.5 m/s as density of X. fragariae in air at 1.3 m was 3.0 × opposed to 0.25 m/s on May 25) may have caused the 4 −1 10 cfu L and was significantly (F . = 0.05) higher plume to meander and the particle concentrations to vary prob 4 −1 than at 5mwhichwas 1.3×10 cfu L (Table 2). at such short time scales that the 10-s mean concentra- On the first day of the experiment, a high linear tions did not show a clear peak upon arrival of the plume relation (r = 0.9958, y = −78,539× + 851,311) was at the particle counter. found between the particle counts of the Dylos air quality None of the strawberry plants exposed to aerosols monitors and the distance from the source in meters (data during spraying water into the air became infected by not shown). On the second day, clear peaks at 1.3, 5 and X. fragariae, nor did plants upwind of X. fragariae con- taining aerosols. However, plants leeward of the source were found infected after release of inoculum, even up to −1 Table 2 Densities of Xanthomonas fragariae (cfu L ) in air a distance of 10 m from the inoculum source (Table 3). samples collected 4 m upwind (−4 m) and at 1.3 and 5 m down- Generally, the X. fragariae infection incidence of wind from the inoculum, a bacterial suspension released with a strawberry plants was comparable for both days of the spray boom experiment and both runs within that day. This allowed Treatment Day 1 Day 2 Days 1 + 2 us to regard the data of the four runs as data derived from b a a one experiment with four repetitions. The infection inci- −4m 0 a 0a 0a 4 4 4 dence of strawberry plants was significantly (F . prob 1.3 m 2.7 10 c3.310 c 3.0 10 c 3 4 4 <0.05) higher at 1.3 m distance from the source in 5 m 9.7 10 b1.710 b 1.3 10 b comparison to infection incidences at 5 and 10 m dis- The experiment was conducted on two days. The back trans- tance. Furthermore, infection incidences were signifi- formed mean after angular transformation per day (N = 2) are cantly (F . <0.05) higher in plants of which the leaves prob shown and the mean of both days b had been wetted shortly before the exposure to aerosols Means without common characters within the same column of the pathogen than in plants of which the leaves indicate significant differences between treatments (Fisher protected pairwise T-tests, P =0.05) remained dry, if the distance from the source was not Eur J Plant Pathol (2018) 152:711–722 719 Table 3 Infection incidence of infections of strawberry target fragariae released with a spray boom. The strawberry leaves were plants at 4 m upwind (−4 m) and at 1.3, 5 and 10 m downwind either dry or wet at the time of aerosol dispersion from the inoculum source, a suspension of Xanthomonas Treatment Distance Xanthomonas fragariae Day 1 Day 2 Days 1 + 2 dry −40 a 0a 0 a dry 1.3 61.0 d 71.0 c 65.8 d dry 5 8.4 bc 11.9 b 10.1 bc dry 10 8.4 bc 6.2 ab 7.2 b wet −4 0 a0 a0 a wet 1.3 100.0 e 100.0 d 100.0 e wet 5 16.0 c 21.6 b 18.7 c wet 10 0.7 ab 9.0 b 3.7 b The experiment was conducted on two days. The back transformed means after angular transformation per day (N = 2) are shown and the means of both days Means without common characters within the same column indicate significant differences between treatments (Fisher protected pairwise T-tests) taken into account. The infection incidence of wetted source no symptomatic plants were observed, regardless plants was also significantly higher than dry plants at of the leaf wetness condition of the plant. Symptom 1.3 m. At 5 and 10 m, the infection incidence of wetted expression was more pronounced in the plants of the plants was comparable to dry plants (Table 3). Most first day compared to those of the second day, whereas remarkable is that strawberry plants which had not been wetted became infected by X. fragariae at each distance tested. At 1.3 m the dry strawberry leaves were partly Table 4 Incidence of angular leaf spot of strawberry target plants wetted by the suspension. However, at 5 and 10 m dry at 4 m upwind (−4) and 1.3, 5 and 10 m downwind from the inoculum source, a suspension of Xanthomonas fragariae released plants were not visibly wetted by the sprayed suspension. with a spray boom Although leeward from the spray boom many plants got infected, relatively few plants developed symptoms Treatment Distance Symptomatic disease incidence of angular leaf spot. In addition, disease severities were Day 1 Day 2 Days 1 + 2 very small (data not shown). The disease incidence varied more than the infection incidence between both dry −4.0 0 a 0a 0 a days of the experiment (Table 4). On the first day, at dry 1.3 0 a 0 a 0 a 1.3 m the disease incidence was 40% on wetted straw- dry 5 0 a 5.4 a 1.4 ab berry plants and 0% on dry plants, whereas in the plants dry 10 0 a 0 a 0 a of the second day no symptomatic plants were found on wet −4.0 0 a 0 a 0 a both wetted and dry plants. At 5 m from the inoculum wet 1.3 39.0 b 0 a 11.0 b source, the disease incidence on wetted plants was 5.4 wet 5 1.4 a 5.4 a 3.0 ab and 0% for day 1 and 2, respectively. On dry plants the wet 10 0 a 0 a 0 a incidences were 0 and 5.4%. Combining the data of both days of the experiment, The strawberry leaves were either dry or wet at the time of aerosol dispersion symptom expression was restricted to on average 20% The experiment was conducted on two days. The back trans- of the plants at 1.3 m and 1.4% at 5 m from the source on formed means after angular transformation per day (N = 2) are the pre-wetted strawberry plants (Table 4). On dry shown and the means of both days plants, symptom expression was found at 5 m distance Means without common characters within the same column from the source but not at 1.3 m; the disease incidence at indicate significant differences between treatments (Fisher 5 m was on average 2.7%. At 10 m from the inoculum protected pairwise T-tests, P = 0.05) 720 Eur J Plant Pathol (2018) 152:711–722 infection incidences bon both days of the experiment Infections of plants at a short distance of 1.3 m from were largely comparable. the infection source may have been caused by aerosols Seven control plants kept in a humid chamber and but also by splash dispersal released by the spray boom, spray-inoculated with the bacterial suspensions used for as larger droplets were observed on the leaves after the inoculating plants in the field were all infected three experiment. This may explain the high infection inci- weeks after inoculation; two plants were symptomatic. dence at this distance. At 5 and 10 m distance splash This indicated that the inoculum used was able to cause dispersal is unlikely. No water droplets were observed angular leaf spot. Similarly, in total 12 control plants were on the dry target plants supporting the lack of splash inoculated with the bacterial suspension using the spray dispersal. boom directly after removal of exposed plants. They were Experiments were conducted at temperatures of 13 °C all infected three weeks after inoculation and ten plants during the first and ranging between 17 and 20 °C during showed symptoms. This indicated that the conditions in the second part of the experiment. According to the the field were suitable to cause angular leaf spot. literature, the highest number of lesions on leaves are found at moderate temperatures between 16 and 25 °C (Kennedy and King 1962b; Kennedy-Fisher 1997; Discussion Hildebrand et al. 2005). Despite the more optimal tem- perature conditions in the second experiment infection X. fragariae released in the form of aerosolized cells can incidences were not higher. Possibly the variable wind infect strawberry plants, minimally up to a distance of during release of the inoculum has resulted in a lower 10 m of the inoculum source. This assessment on risk infection pressure. At a low temperature of 5 °C and high for infection was supported by the detection of temperatures above 30 °C no lesions are formed, but the bacteria will not disappear (Hildebrand et al. 2005; culturable cells of X. fragariae in sampled air and by particle counts during the release of inoculum which Roberts et al. 1996). During the experiments, a relative exponentially decreased with the distance from the high humidity of between 75 and 85% was found. A high source as found on the first day of the experiment. The humidity is also important for infection, disease develop- decrease is a consequence of a Gaussian dispersal of ment and production of bacterial ooze (Kennedy and particles in the open air from a point source as described King 1962b;Hildebrandetal. 2005). by Spijkerboer et al. (2002). The infections must have been established within a In the field experiments, the conditions were condu- short time after deposition of aerosols on leaves. There cive for infection and symptom development as plants are no indications for an epiphytic phase of X. fragariae placed at a distance of 1.3 m from the infection source, (Hildebrand et al. 2005; Kastelein et al. 2014), as has of which the leaves had been wetted just before the start been found for a number of other phytopathogenic bacteria, including Pseudomonas syringae, of the experiment, were found infected. On a relatively high percentage (20%) of these plants symptoms devel- Xanthomonas axonopodis pv. phaseoli and Erwinia oped. Control plants inoculated in the field with a amylovora (Hirano and Upper 1983). Here, epiphytes houseplant mist sprayer developed symptoms of angular are defined as organisms that can grow or at least reside leaf spot as well, also indicating that the circumstances on the host. In glasshouse experiments in which spray- in the field were conducive for infection and that the inoculations of leaves with high densities of X. fragariae inoculum used was viable. were conducted, a strong decline of culturable cells in Control plants placed up-wind from the inoculum wash water of leaves of at least a 100.000 times was source remained free from X. fragariae. This indicates found in the first week after inoculation (Kastelein et al. that no natural inoculum source was present and no 2014). Only upon the development of symptoms, pop- inoculum was disseminated upwind or at least not suf- ulation densities in the wash water increased. It is ex- ficiently to establish an infection. It further indicates that pected that in the field, where X. fragariae is subjected infections of the target plants were from the released to UV radiation and desiccation, population densities inoculum not from an unknown source in the surround- will drop even faster than under glasshouse conditions ing. This is supported by the fact that before each release (Beattie and Lindow 1995). Consequently, if air-borne no X. fragariae was detected in the air samples collected inoculum is deposited on leaves, within a short time free for 10 min with the Coriolis sampler. water is required to establish an infection. If free water is Eur J Plant Pathol (2018) 152:711–722 721 present, the bacterium colonizes internal leaf tissues demonstrated in the glasshouse experiments in rapidly, as treatment of leaves with a biocide within which the percentage of successful infections was one hour after inoculation did not result in an effective strongly correlated with the inoculum density ap- control (data unpublished). As a consequence, a higher plied. Furthermore, even at low bacterial counts of infection incidence for plants was found that were wet- 2000 cfu per plant, infection could occur. ted prior to inoculation than for plants kept dry. Never- This study shows that under field conditions, wind- theless, infections also occurred on plants kept dry dur- blown water droplets loaded with X. fragariae, can infect ing inoculation. It may be that the aerosols provided efficiently strawberry plants at a distance of at least 10 m sufficient water for the pathogen to migrate through from the inoculum source. Under favourable conditions, stomata which are identified as main port of entrance a low inoculum pressure is sufficient to cause an infec- for the pathogen (Hildebrand et al. 2005). Alternatively, tion. Management of X. fragariae should therefore in- the relatively high air humidity due to water on irriga- clude roguing of symptomatic plants to reduce the inoc- tion mats and tempered sunshine may have allowed the ulum pressure, avoidance of the release of contaminated survival of the pathogen on leaves for a prolonged time. aerosols through cultivation practices such as mowing Dew may have provided the water required for the and maintaining sufficient distances between strawberry infection. It was found that even at a relative humidity cultivation plots. The infection risk is less when leaves of only 50% a water film can be formed on leaves are air dry compared to wetted plants. Therefore, cultiva- (Burkhardt and Eiden 1994). Possibly small particles tion practices should be taken preferably in a dry crop on leaves act as condensation centres during dew for- rather than in a crop with a wet canopy. mation at a low humidity (Eiden et al. 1994). Acknowledgements This research received funding from the Symptom expression depends on bacterial density in Dutch growers of strawberry planting material, the Dutch Ministry plant tissues and growth conditions of the plant. For all of Economic Affairs (PPS project KV) and the European Union plants used on each day of the experiment the growth Seventh Framework (FP7/ 2007–2013) under the grand agreement conditions during the experiment and incubation period n° 613678 (DROPSA). were the same. Therefore, we assume that differences in Compliance with ethical standards symptom expression within the plants used on the same day were not caused by the conditions, but must be a Conflict of interest The corresponding author (Jan van der function of bacterial deposition density. Obviously the Wolf) has received research funding from the Dutch growers of infection incidence itself is a prerequisite for symptom strawberry planting material. All other authors herewith declare expression. Bio-TaqMan analysis revealed that infected that they have no conflict of interest. plants were present at each distance although the infection Human and animals studies This study does not contain stud- incidence became less with distance. Air samples showed ies with human participants or animals performed by any of the that X. fragariae was present at 1.3 and 5 m although at a authors. significantly lower density at a lager distance which coin- cided with a lower infection incidence. In this study, at Open Access This article is distributed under the terms of the 10 m from the inoculum source no air was sampled for the Creative Commons Attribution 4.0 International License (http:// presence of X. fragariae, however, earlier aerosol experi- creativecommons.org/licenses/by/4.0/), which permits unrestrict- ments showed the presence of X. fragariae at that distance ed use, distribution, and reproduction in any medium, provided (data unpublished). For the first day of the experiment a you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if good relation between the number of small particles in the changes were made. air and the infection incidence was found. In the second experiment however, no clear peaks were observed in particle sampling, possible due to the higher variation in References thewindspeedandwinddirectionduringreleaseof inoc- ulum, leading to decrease in infection pressure. Inoculum Anonymous. (2006). Xanthomonas fragariae. EPPO Bulletin, 36, pressure as expressed by the number of viable cells of 135–144. X. fragariae deposited on the leaves might be correlated Beattie, G. A., & Lindow, S. E. (1995). The secret life of foliar to the number of successful infections and subse- bacterial pathogens on leaves. Annual Review of quently to symptom expression. This was clearly Phytopathology, 33,145–172. 722 Eur J Plant Pathol (2018) 152:711–722 Burkhardt, J., & Eiden, R. (1994). Thin water films on coniferous Mcinnes, T. B., Gitaitis, R. D., Mccarter, S. M., Jaworski, C. A., & needles. Atmospheric Environment, 28,2001–2011. Phatak, S. C. (1988). Airborne dispersal of bacteria in tomato De Boer, S. (2002). Relative incidence of Erwinia carotovora and pepper transplant fields. Plant Disease, 72,575–579. subsp. atroseptica in stolon end and peridermal tissue of Melis, P., Baets, W., Verjans, W., Deckers, T., Stragier, P., De Vos, potato tubers in Canada. Plant Disease, 86,960–964. P., Vandroemme, J., & Maes, M. (2012). Xanthomonas Desmet, E., Van Vaerenbergh, J., & Denruyter, L. (2006). fragariae in de aardbeiteelt – deel 2: leefwijze en Bacteriebladvlekkenziekte bij aardbei. (Xanthomonas verspreiding van. Xanthomonas in de plant en het veld. fragariae). Proeftuinnieuws, 14/15,22–23. Proeftuinnieuws, 7, 44–46. Desmet, E. M., Maes, M., Van Vaerenbergh, J., Verbraeken, L., & Morris, C. E., Kinkel, L. L., Xiao, K., Prior, P., & Sands, D. C. Baets, W. (2009). Sensitivity screening of commonly grown (2007). Surprising niche for the plant pathogen Pseudomonas strawberry cultivars towards angular leaf spot caused by syringae. Infection, Genetics and Evolution, 7,84–92. Xanthomonas fragariae. Acta Horticulturae, (842), 275–278. Pérez-Jiménez, R. M., De Cal, A., Melgarejo, P., et al. (2012). Eiden, R., Burkhardt, J., & Burkhardt, O. (1994). Atmospheric aero- Resistance of several strawberry cultivars against three dif- sol particles and their role in the formation of dew on the surface ferent pathogens. Spanish Journal of Agricultural Research, of plant leaves. Journal of Aerosol Science, 25, 367–376. 10,502–512. EPPO (1997). EPPO datasheet Xanthomonas fragariae.In: Perombelon, M., Fox, R., & Lowe, R. (1979). Dispersion of Quarantine pests for Europe, Second. Edition (Smith, I.M, Erwinia carotovora in aerosols produced by the pulverization McNamara, D.G., Scott, P.R., & Holderness, M. Ed), pp. of potato haulm prior to harvest. JournalofPhytopathology, 1124–1128. European and Mediterranean Plant Protection 94,249–260. Organization (EPPO); CABI Publishing, OX. Fragariae ord Rätsep, R., Moor, U., Vool, E., & Karp, K. (2015). Effect of post- (GB) harvest flame-defoliation on strawberry (Fragaria x Franc, G. D., & Demott, P. J. (1998). Cloud activation character- ananassa Duch.) growth and fruit biochemical composition. istics of airborne Erwinia carotovora cells. Journal of Zemdirbyste Agriculture, 102,403–410. Applied Meteorology, 37,1293–1300. Rivera-Zabala, N., Ochoa-Martinez, D. L., Rojas-Martinez, Hazel, W., & Civerolo, E. (1980). Procedures for growth and R. I., Rodriguez-Martinez, D., Aranda-Ocampo, S., & inoculation of Xanthomonas fragariae, causal organism of Zapien-Macias,J.M.(2017). Xanthomonas fragariae angularleafspot of strawberry. Plant Disease, 64,178–181. genetic variability and its severity on strawberry Hildebrand, P. D., Braun, P. G., Renderos, W. E., Jamieson, A. R., genotyes (Fragaria ananassa Duch). Agrociencia, 51, Mcrae, K. B., & Binns, M. R. (2005). A quantitative method 329–341. for inoculating strawberry leaves with Xanthomonas Roberts, P. D., Jones, J. B., Chandler, C. K., Stall, R. E., & Berger, fragariae, factors affecting infection, and cultivar reactions. R. D. (1996). Survival of Xanthomonas fragariae on straw- Canadian Journal of Plant Pathology, 27,16–24. berry in summer nurseries in Florida detected by specific Hirano, S. S., & Upper, C. D. (1983). Ecology and epidemiology primers and nested polymerase chain reaction. Plant of foliar bacterial plant pathogens. Annual Review of Disease, 80,1283–1288. Phytopathology, 21,243–269. Spijkerboer, H., Beniers, J., Jaspers, D., et al. (2002). Ability of the Kastelein, P., Krijger, M., Czajkowski, R., et al. (2014). Gaussian plume model to predict and describe spore dispersal Development of Xanthomonas fragariae populations and over a potato crop. Ecological Modelling, 155,1–18. disease progression in strawberry plants after spray- Turechek, W. W., & Peres, N. A. (2009). Heat treatment effects on inoculation of leaves. Plant Pathology, 63,255–263. strawberry plant survival and angular leaf spot, caused by Kennedy, B. W., & King, T. H. (1962a). Angular leaf spot of Xanthomonas fragariae, in nursery production. Plant strawberry caused by Xanthomonas fragariae sp. nov. Disease, 93,299–308. Phytopathology, 52,873–875. Kennedy, B. W., & King, T. H. (1962b). Studies on epidemiology Van Der Gaag, D.J., Bergsma-Vlami, M., Van Vaerenbergh, of bacterial angular leafspot on strawberry. Plant Disease J., Vandroemme, J., & Maes, M. (2013). Pest risk Reporter, 40,360–363. analysis for Xanthomonas fragariae. Netherlands food Kennedy-Fisher, S.D. (1997). The effect of copper sulphate and and consumer product safety authority, Utrecht, the host variety on angular leaf spot. (Xanthomonas fragariae) of Netherlands - Insitute for agricultural and fisheries strawberry. MSc thesis, Dalhousie University Halifax, Nova research, Merelbeke, Belgium, 50 pages. Scotia, Canada. Resource document. http://www. Van der Wolf, J., Kastelein P., Evenhuis B., & Moene, A. collectionscanada.gc.ca/obj/s4/f2/dsk3/ftp04/mq24862.pdf (2017). Dissemination of Xanthomonas fragariae in a th Koike, H. (1965). Aluminium cap method for testing sugar cane strawberry field crop. 12 European Foundation for th varieties against leaf scald disease. Phytopathology, 55,317– Plant Pathology and 10 French Society for Plant Pathology, 29 May 2017 to 2 June 2017, Dunkerque, Lieten, P. (2014). The strawberry nursery industry in the France, Abstract. Netherlands: An update. Acta Horticulturae, (1049), 99–106. Van Kruistum, G., Evenhuis, A., Hoek, J., Kastelein, P., Van der Maas, J.L. (1998). Compendium of strawberry diseasese second Wolf, J. M., & Verschoor, J. A. (2015). CATT: A new and ed. APS Press, 98 p. non-chemical pest and nematode control method in strawber- Maas, J. L. (2004). Strawberry disease management. In S. A. M. H. ry planting stock. Acta Horticulturae, (1105), 189–196. Naavi (Ed.), Diseases of fruits and vegetables, Volume II (pp. 441–483). the Netherlands: Kluwer Academic Publishers.
European Journal of Plant Pathology – Springer Journals
Published: May 31, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera