Rose rosette disease is a major disease of rose (Rosa spp.; Rosaceae) in the landscape (Jeppson et al. 1975; Crowe 1983; Amrine 1996; Babu et al. 2014; Pemberton et al. 2018) and poses a threat to production in nurseries. Rose rosette disease is caused by the rose rosette virus (Laney et al. 2011), and an eriophyid mite, Phyllocoptes fructiphilus Keifer (Acari: Eriophyidae), transmits rose rosette virus (Allington et al. 1968; Amrine et al. 1988). In the US, roses are valued > $203 million, where Florida, California, Texas, North Carolina, Arizona, and Georgia are the top rose-producing states (USDA-NASS 2017). Rose rosette disease and P. fructiphilus have been established in western states such as Kansas, California, and Arizona for many yr (Wagnon & Nichols 1966, 1970; Jeppson et al. 1975; Crowe 1983), but not until recently was rose rosette disease reported from New England, mid-Atlantic and southeastern states in the US (Tipping & Sindermann 2000; Babu et al. 2014; Grant 2019; RRD EDD 2020), excluding Florida (Hoy 2013).

The rose rosette virus (Emaravirus) infects multiflora and ornamental roses. Rose rosette disease causes foliar mosaic and mottling, flower and leaf malformation, excessive thorniness, increased lateral shoot formation, young shoot reddening, stem thickening, and plant death (Jeppson et al. 1975). To date, once plants exhibit disease symptoms, there is no cure.

Phyllocoptes fructiphilus is approximately 140 to 175 µm long and is tapered at both ends (Otero-Colina et al. 2018). Eggs are oviposited on leaf bracts and sepals of the rose flower, where trichomes are abundant (Otero-Colina et al. 2018). Within 4 d, the eggs hatch, and larvae reach adulthood approximately 1 wk after molting through 2 larval and nymphal stages. Male P. fructiphilus release spermatophores on the plant surface, and females gather them from the plant surface through their genital opening and store them (Otero-Colina et al. 2018). The fertilized eggs develop into females, whereas nonfertilized eggs develop into males. Sexually mature adult females oviposit an egg per d and live up to 12 d. The P. fructiphilus population builds up from midsummer into Sep (Amrine 1996). Although eriophyid mites have been found on symptomatic roses and healthy plants, the numbers usually are very high on symptomatic roses (Amrine 1996; Otero-Colina et al. 2018). Phyllocoptes fructiphilus overwinters inside the ovary of rose fruit (Amrine 1996).

Previous research has shown that biological functions such as oviposition, reproduction, survival, etc., of arthropods are impacted by changes in relative humidity, e.g., western flower thrips, Frankliniella occidentalis (Pergande) (Thysanoptera: Thripidae) on rose (Fatnassi et al. 2015), the wood-boring bamboo beetle, Dinoderus minutus Fabricius (Coleoptera: Bostrichidae) (Norhisham et al. 2013), storage mites, Tyrophagus putrescentiae (Schrank) (Acari: Acaridae), Tyrophagus neiswanderi Johnston and Bruce (Acari: Acaridae), and Acarus farris (Oudemans) (Acari: Acaridae) (Sánchez-Ramos et al. 2007), and the spider mite Tetranychus telarius L. (Acari: Tetranychidae) (Boubreaux 1958). Relative humidity varies drastically across various regions impacted by rose rosette disease in the US (Barreca 2012). The biology of P. fructiphilus under various relative humidity regimes is understood poorly. In 1940, P. fructiphilus was first described on Rosa californica Cham. & Schldl. (Rosaceae) in California, USA, and it still is not reported from major rose nurseries in the southern San Joaquin Valley in California, USA. Although the exact reason is unclear, perhaps low relative humidity in the southern San Joaquin Valley does not favor population growth of P. fructiphilus compared to that in eastern regions of the US. Thus, it is critical to understand the performance of P. fructiphilus under varying relative humidity conditions. This information will improve our understanding of the ecology of P. fructiphilus to develop an effective, tailored, and region-specific integrated pest management program. The major objective of the current study was to determine the effect of relative humidity on both the abundance of P. fructiphilus, as well as on the incidence and severity of rose rosette disease symptoms.

Materials and Methods

The experiment was conducted in environmentally controlled chambers at the University of Georgia, Griffin campus, Georgia, USA, in 2018. Potted ‘Pink Double Knock-Out’ rose plants (3.7 L) were obtained from a wholesale nursery in Dearing, Georgia, USA. There was no incidence of rose rosette disease in the nursery or on the roses planted in surrounding landscapes. The random sepal samples were collected from rose plants and were devoid of P. fructiphilus. The experiment was conducted in 3 environmentally controlled growth chambers and was programmed at 20, 60 and 95% relative humidity. The temperature of all 3 chambers was set at 28 °C with a 16:8 h (L:D) photoperiod. The 4 potted rose plants were introduced into each chamber and were acclimated to the conditions in chambers for 7 d before P. fructiphilus was introduced. The relative humidity regimes 20, 60, and 95% were the treatments, and the 4 rose plants were the replications. To prevent desiccation, all the rose plants were monitored daily to ensure sufficient moisture in the potting soil, and were irrigated as needed for the duration of the experiment. Before introducing P. fructiphilus into the chamber, rose terminal samples were sampled randomly from the potted rose plants to confirm the absence of any mites on them. There were no rose rosette disease symptoms on rose plants before introducing P. fructiphilus. Phyllocoptes fructiphilus-infested rose terminals (about 7 cm long) were collected from rose rosette disease symptomatic rose shrubs in the landscape within the Griffin campus. The rose rosette virus on the rose shrubs was confirmed using a modified real time reverse transcription PCR (RT-qPCR) using primers developed by Babu et al. (2016) at the plant disease diagnostic laboratory at the University of Florida North Florida Research and Education Center in Quincy, Florida, USA. On 15 Oct 2018, each plant in the chamber was infested with P. fructiphilus by attaching field-collected rose terminals on 2 terminals of each rose plant in the chamber using paper clips. Each rose plant received about 20 P. fructiphilus individuals. A rose terminal consisted of an opened flower bud and 3 leaves.

After 2 wk of infestation, 2 sepals were collected from each potted rose plant to determine the number of P. fructiphilus individuals. The sampling continued at 2-wk intervals for up to 12 wk. The number of P. fructiphilus individuals was quantified directly per sepal under 40× magnification using a dissecting microscope (Leica Microsystems, Wetzlar, Germany). Phyllocoptes fructiphilus has a subtriangular shield that tapers out to a point approaching the anterior end. This shield, on the mite's ventral side, allows distinction among species (Otero-Colina et al. 2018).

Approximately 10 wk after P. fructiphilus infestation, rose plants started to show rose rosette disease symptoms. The number of rose rosette disease symptomatic and total terminals was counted, and then the proportion of symptomatic terminals per plant was determined. The severity of rose rosette disease symptoms was assessed using the Horsfall-Barratt scale starting 10 wk post-infestation (Horsfall & Barratt 1945). To confirm rose rosette virus, the rose rosette disease symptomatic and non-symptomatic leaves were collected randomly from rose plants in the chambers, and leaf tissues were tested using the modified RT-qPCR technique mentioned above at 12 wk post-infestation.

The P. fructiphilus data were subjected to ANOVA by wk using the PROC GLM general linear model procedure in SAS (SAS 2012). The data were square-root transformed to establish homogeneity of variance using the PROC Univariate procedure in SAS (2012) before analysis, where relative humidity was the treatment factor. To determine the overall effect of relative humidity treatments, the number of P. fructiphilus collected over 12 wk was pooled, square-root transformed, and subjected to ANOVA by wk using the general linear model procedure in SAS (2012). The data on the incidence of rose rosette disease symptoms were log-transformed (ln[x + 1]), whereas the severity of rose rosette disease symptoms assessed using the Horsfall-Barratt scale was arcsine square-root transformed before analysis. The transformed data were subjected to ANOVA by wk using the general linear model procedure in SAS (2012). The means were separated using Tukey's HSD test for treatment comparisons. All statistical comparisons were considered significant at α = 0.05.

Results

The number of P. fructiphilus was significantly greater in sepals with 60% relative humidity than in those with 20% relative humidity at 2 wk after infestation (F = 4.1; df = 2, 14; P = 0.040; Fig. 1A). At 4 wk after infestation, there was no significant difference in the number of P. fructiphilus among relative humidity levels (F = 2.5; df = 2, 14; P = 0.116). A significantly greater number of P. fructiphilus occurred on sepals under 60% relative humidity than under 20% or 95% relative humidity at 6 wk after infestation (F = 6.1; df = 2, 14; P = 0.013; Fig. 1A). The number of P. fructiphilus was not significantly different in sepals among relative humidity levels at 8 wk after infestation (F = 3.1; df = 2, 14; P = 0.082). Overall, the number of P. fructiphilus was significantly greater on sepals under 60% relative humidity than under 20 or 95% relative humidity (F = 5.7; df = 2, 14; P = 0.015; Fig. 1B).

At 10 wk after infestation, there was no difference in the incidence of rose rosette disease (F = 4.2; df = 2, 6; P = 0.073; Fig. 2A). The Horsfall-Barratt scale showed that the severity of rose rosette disease on plants was significantly greater under 95% relative humidity than under 20% relative humidity (F = 8.9; df = 2, 6; P = 0.016; Fig. 2B). At 12 wk after infestation, the incidence of rose rosette disease symptoms on plants was significantly greater under 95% relative humidity than under 20 and 60% relative humidity (F = 8.7; df = 2, 6; P = 0.017; Fig. 2A). The Horsfall-Barratt scale showed that the severity of rose rosette disease on plants was significantly greater under 95% relative humidity than under 20% relative humidity (F = 10.3; df = 2, 6; P = 0.012; Fig. 2B). At 10 and 12 wk after infestation, the severity of rose rosette disease on plants was not significantly different between 60 and 95% relative humidity. At 14 wk after infestation, the incidence of rose rosette disease was not significantly different on 14 wk after infestation (F = 3.8; df = 2, 6; P = 0.087). Rose rosette virus was confirmed on symptomatic plant samples under various relative humidity regimes. The severity of rose rosette disease on plants was significantly greater under 95% relative humidity than under 20 and 60% relative humidity (F = 13.6; df = 2, 6; P = 0.005; Fig. 2B).

Fig. 1.

Mean (± SE) number of Phyllocoptes fructiphilus under various relative humidity regimes (A) by wk and (B) for the duration of the experiment. The same letters within a wk after infestation or bars are not significantly different (ANOVA followed by Tukey's HSD test; α = 0.05). Where no differences were observed, no letters are included.

Discussion

The results show that the number of P. fructiphilus individuals was greater under a moderate relative humidity (60%) than under a high relative humidity (95%) or low relative humidity (20%). Previous studies on a spider mite, T. telarius, showed that the total oviposition of female T. telarius was reduced by more than half when the relative humidity increased to 95% from 35% (Boubreaux 1958). Variation in relative humidity also can affect biological control activity. The predaceous mite Phytoseiulus persimilis Athias-Henriot (Acari: Phytoseiidae) provided greater pest control on the two-spotted spider mite Tetranychus urticae (Koch) (Acari: Tetranychidae) under 60 to 85% relative humidity than under lower relative humidity (Stenseth 1979). This suggests that the performance of a specific biological control agent should be studied carefully to match when the abundance of P. fructiphilus is high across various stages. The success of biological control could be increased if the agent(s) could reduce the P. fructiphilus to very low density, because few individuals could transmit the rose rosette virus through feeding.

In the current study, the incidence and severity of rose rosette disease symptoms were noticeably higher under the high (95%) relative humidity regimen than under the low (20%) relative humidity regimen. This result could impact rose breeding programs. Rose genotypes and cultivars are field-tested for various horticultural attributes and resistance to diseases, including rose rosette disease (Byrne 2015). The data suggest that rose genotypes tested for horticultural attributes and disease resistance to other than rose rosette disease research may not necessarily express rose rosette disease symptoms if infected with rose rosette virus, and viable studies could be conducted. In contrast, rose rosette disease genotype or cultivar screening should be conducted under high relative humidity conditions to ensure reliable results. The rose plants maintained in the greenhouse at 21 °C and about 40% relative humidity with no exposure to rose rosette virus and P. fructiphilus did not elicit any rose rosette disease symptoms. In Florida, P. fructiphilus populations developing on roses without rose rosette virus infection were asymptomatic (Fife et al. 2020).

Fig. 2.

Mean (± SE) (A) proportion of rose rosette disease symptomatic terminals and (B) value of the Horsfall-Barratt scale on the severity of rose rosette disease. The same letters within a wk after infestation are not significantly different (ANOVA followed by Tukey's HSD test; α = 0.05). Where no differences were observed, no letters are included.

Relative humidity can vary drastically from western states to eastern states in the US (Barreca 2012). The data suggest that region-specific integrated pest management strategies for P. fructiphilus and rose rosette disease should be developed. For example, based on the data, rose rosette virus infected rose plants in a region with high relative humidity ranges have a high probability of rose rosette disease symptom expression. Thus, these rose plants may need comprehensive management to reduce rose rosette virus transmission. In contrast, those plants shipped from the regions with low relative humidity should be subjected to rigorous diagnostic tests, especially before being shipped to regions with high relative humidity. More studies are warranted to determine the effects of relative humidity under low- and high-temperature regimes to improve the management of vector and reduce rose rosette virus transmission.