Introduction

Total global citrus production exceeds 100 million tons/year, and oranges comprise about two thirds of the production (Anonymous 2016). In Vietnam, over one million tons of oranges (Citrus sinensis) and mandarins (C. reticulata) were traded in 2018 (Anonymous 2018). Oranges have been grown for a long time in the Mekong Delta (Thanh et al. 2018), and in recent decades, production of oranges and mandarins has expanded into northern Vietnam (Quang 2013; Tien 2013).

Root diseases and citrus gummosis cause substantial loss in production, and Phytophthora spp. are the most widespread causal agents (Graham and Feichtenberger 2015; Timmer et al. 2003). A number of species have been implicated, e.g., P. nicotianae and P. palmivora in Florida in the United States (Graham et al. 2003; Widmer et al. 1998), P. nicotianae and P. cryptogea in Tunisia (Boughalleb-M’hamdi et al. 2018), P. nicotianae in Thailand (Watanarojanaporn et al. 2011), and P. palmivora in Japan (Tashiro et al. 2012), Egypt (Ahmed et al. 2014), and Vietnam (Hung et al. 2015b). In northern Vietnam, Phytophthora palmivora is the main pathogen causing root rot disease in Citrus reticulata and C. sinensis in Quang Ninh Province (Chi et al. 2019).

Chemicals, live organisms (McGrath 2009), and changes in horticultural practices are among the wide range of methods used to mitigate diseases caused by Phytophthora spp. including. Metalaxyl-M, mancozeb, ethaboxam, mandipropamid, fluopicolide, benzethonium chloride, and potassium phosphonate have been used commercially to reduce or prevent disease (Graham 2011; Hao et al. 2019; Lawrence et al. 2017; Ramallo et al. 2019; Rolando et al. 2017; Vawdrey et al. 2004). Trichoderma spp. have been used to reduce Phytophthora diseases in fruit (de Oliveira et al. 2018; Sriwati et al. 2019), and three species of Chaetomium were used to reduce P. palmivora and P. nicotianae root rot in citrus (Hung et al. 2015b, a). Application of arbuscular mycorrhizal fungi (Acaulospora tuberculata and Glomus etunicatum) also reduced P. nicotianae root rot in citrus (Watanarojanaporn et al. 2011). Preventative measures to control Phytophthora root rot in citrus include enacting phytosanitary requirements for producing pathogen-free nursery trees in enclosed structures and propagation from indexed and certified pathogen-free sources (Graham and Feichtenberger 2015), reducing flooding in orchards (Chaudhary et al. 2016), and using resistant rootstocks (Lima et al. 2018; Mourão Filho et al. 2008; Widmer et al. 1998; Yan et al. 2017).

Thanh et al. (2004) noted the need for more research on Phytophthora species in citrus in Vietnam. The present study focused on developing practical solutions to control root rot disease caused by P. palmivora in mandarin (C. reticulata) in northern Vietnam and preventing tree decline and death.

Materials and methods

Evaluation of agents for inhibition of P. palmivora in vitro

Five fungicides and five biological agents were evaluated for their efficacy to inhibit the growth of P. palmivora (isolate QN729) compared to a water control (see Table 1 for the chemical and biological formulations). QN729 was isolated from diseased roots of mandarin in 2018 in An Sinh Ward, Dong Trieu District, Quang Ninh Province and was identified as P. palmivora based on the ITS sequence (MT113313) (Chi et al. 2019). Pathogenicity was previously tested by inoculating Citrus sinensis and C. reticulata seedlings, and QN729 was shown to be an aggressive isolate (Chi et al. 2019). The different treatments against P. palmivora were evaluated in vitro using the method of Singh and Tripathi (1999) with some modifications. Mycelial plugs (5 mm diameter) of QN729 were placed at three places on potato dextrose agar (PDA, Merck KGaA, Darmstadt, Germany) at equal distances from each other on the perimeter of each Petri dish. One 5-mm-diameter well was created in the center of each Petri dish, and 50 μl of the test solution was placed into each well (Table 1). Each treatment was applied to six replicate plates, and the experiment was done five times, for a total of 30 plates for each treatment and the control. Plates were placed into an incubator at 22 ℃ for 7 days, then the diameter of any inhibition zone of P. palmivora growth was measured across each well and classified as D = 0 cm (no inhibition); 0 cm < D ≤ 1.0 cm (weak); 1.0 cm < D ≤ 2.0 cm (average); 2.0 cm < D ≤ 3.0 cm (strong); D > 3.0 cm (very strong).

Table 1 Formulations tested for control of Phytophthora palmivora
Full size table

Evaluation of agents for reducing P. palmivora disease in mandarin seedlings

Metalaxyl-M, potassium phosphonate, mancozeb, and Trichoderma viride had the greatest inhibitory effect against P. palmivora in vitro and were tested further against P. palmivora in 6-month-old C. reticulata seedlings. Seeds of a single family of C. reticulata were sown in sterilized substrate (8 parts soil, 2 parts compost) in containers, and seedlings were grown in a glasshouse for 6 months to a height of 50–60 cm. Complete liquid fertilizer was applied every 15 days to promote healthy growth. To prepare the inoculum, a solid medium comprising 250 g of rice bran, 250 g boiled unhusked rice, 17.5 ml of vegetable juice broth, and 3 g CaCO3 was placed in 500 ml Erlenmeyer flasks and autoclaved at 121 ℃ for 30 min. Agar plugs with mycelia of isolate QN729 were added to the cooled medium, and the flasks were incubated at 22 ℃. After 5 weeks, 2.0 g of the culture was placed on the surface of the potting medium (2/seedling) in each pot, the pots were briefly flooded with water, then the surface was covered with 1–2 cm of autoclaved soil. Each of the five treatments (control was sterile water) was applied with a hand sprayer to 15 6-month-old seedlings, 50–60 cm in height, and the experiment was done four times. At 40 days after inoculation with QN729, and again 15 days later, the treatments were sprayed on the foliage, stem and base of each seedling (200 ml/seedling) and also applied to the soil around the base (200 ml/seedling). Dosages are given in Table 1.

Symptoms on foliage were evaluated 40 days after inoculation, and foliage and root damage were assessed after 90 days. Disease severity was then classified from 0 to 4 using the methods of Chi et al. (2019) where 0 = healthy foliage/healthy roots; 1 = yellowing and sparse foliage/ < 25% of roots diseased; 2 = top of the branches wilting/25–50% roots diseased; 3 = plant completely wilted/51–75% roots diseased; 4 = plant dead/ > 75% roots diseased. These ratings were then used to calculate a disease index (described later in the disease analysis section). Representative samples were collected, the pathogen was reisolated, and hyphal and sporangial traits were examined with a light microscope to confirm P. palmivora was present.

Evaluation of agents for reducing P. palmivora disease in a mandarin orchard

The four treatments tested for control of P. palmivora on seedlings in the nursery were then evaluated against P. palmivora in a commercial mandarin orchard. Mandarin orchards in An Sinh Ward, Dong Trieu District, Quang Ninh Province were surveyed for disease incidence and disease index, and a 2-ha property with 625 trees/ha and about 40% disease incidence was selected (21151′03"N, 106516′05"E, 68 m a.s.l.) for the experiment. The orchard had previously hosted one rotation of citrus, and P. palmivora had been isolated from scions (15 trees) and rootstocks (45 trees) in this orchard (Chi et al. 2019) where mandarin was grafted onto pomelo (C. maxima) seedlings. The orchardist had chosen to use pomelo because it is alleged by producers to be “stronger” than mandarin, but it is known to be susceptible to P. palmivora (Hung et al. 2015b). The location has a mean annual temperature of 23.8 ℃, 1,600 h sunshine/year, and average rainfall of 1,850 mm/year. The terrain is flat, and the soil is a red-yellow ferralite > 50 cm in depth. The trees were trickle-irrigated daily using groundwater, the weeds were cut by hand, and the soil was cultivated monthly. The fertilizer schedule for the orchard was February, 30 kg of organic fertilizer + 500 g NPK/tree; May, 500 g NPK/tree; and August, 400 g NPK/tree. The presence of drainage ditches prevents flooding in the orchard during the wet season. The experiment was set up with three blocks, each contained five plots (500 m2), and each plot had 30 trees. The treatments were randomly assigned to the plots. The plots were separated by buffer strips of at least 10 m. Trees in the plots were planted in 2017, 2.5–3.0 m tall and first produced fruit in 2019.

The fungicides and biopesticide were sprayed (electric sprayer, 2 l/tree) on foliage, branches, and trunk (including the rootstock), and they were also applied to the soil within 50 cm of the trunk (2 l/tree) (dosages: Table 1). The treatments were applied in the late afternoon, without rain. The four treatments were applied twice, 15 days apart, in April and repeated in July 2019. This regime was chosen based on government recommendations on the use of fungicides for preventing gummosis disease on citrus.

The trees were assessed twice: when the trial was established (April 2019) and after 6 months (October 2019). Foliage symptoms were used to assess trees in April and October. The level of disease was classified at five levels (0 to 4), where 0 = healthy plant; 1 = yellowing and sparse foliage; 2 = foliage wilting; 3 = plant completely wilted; 4 = plant dead.

Disease analysis

Using the results of symptom classification, the disease incidence (P%) for each treatment was determined as P% = (n/N) × 100, where n = number of diseased plants; N = total number of plants assessed.

The disease index (DI) value was calculated as DI = (Ʃni × vi)/N, where vi = class of disease severity; ni = number of plants recorded in this class; N = total number of plants assessed. The DI was then classified according to the range in values: DI = 0 (no infection); 0 < DI ≤ 1 (low); 1 < DI ≤ 2 (medium); 2 < DI ≤ 3 (high); 3 < DI ≤ 4 (very high).

The level of disease inhibition (E) of each agent was calculated as:

$$E = \left( {1 - \frac{{C_{a} T_{b} }}{{C_{b} T_{a} }}} \right) \times 100,$$

where E = inhibitory effect (%); Ca = disease incidence/disease index of the control treatment at the start of the trial; Ta = disease incidence/disease index of the agent treatment at the start of the trial; Cb = disease incidence/disease index of the control treatment at the end of the trial; and Tb = disease incidence/disease index of the agent treatment at the end of the trial.

The recovery of trees (R%) of each agent was calculated as R% = [(Nb − Na)/Nb] × 100, where Nb = number of diseased plants at the time of application of treatments; Na = number of diseased plants after application of treatments.

Statistical analyses

Data were analysed using GenStat Release 12.1 software package (VSN International Ltd., Hemel Hempstead, UK). Analysis of variance (ANOVA) was used to test for significant effect of treatments or site factors, followed by Duncan’s multiple range test for comparisons of means. Differences were considered to be significant at P < 0.05 or very significant at P < 0.01.

Results

Inhibition of P. palmivora in vitro

There was a significant effect (Fpr < 0.001) of chemical and biological treatments on P. palmivora growth in vitro (Table 2). The chemicals metalaxyl-M (Fig. 1a) and potassium phosphonate (Fig. 1b) very strongly inhibited growth, followed by mancozeb and Trichoderma viride (Fig. 1c) which strongly inhibited growth compared to the control (Fig. 1d). Chlorothalonil and cytosinpeptidemycin were weak inhibitors.

Table 2 Comparison of chemical and biological treatments for their inhibition of growth of Phytophthora palmivora in vitro
Full size table
Fig. 1
figure 1

Inhibition of Phytophthora palmivora on PDA by a metalaxyl-M; b potassium phosphonate; c Trichoderma viride; d control

Full size image

Reducing P. palmivora disease in C. reticulata seedlings

After 40 days of inoculation 90.0–93.3% of the C. reticulata seedlings were infected with a mild and homogeneous disease index (DI = 1.00–1.05). Seedlings with uniform symptoms of leaf yellowing (Fig. 2) were selected, ensuring homogeneity before application of inhibitory treatments.

Fig. 2
figure 2

Effect of Phytophthora palmivora infection on Citrus reticulata seedlings a, d above- and belowground at 40 days after inoculation; b, c, e, f 50 days after treatment: b, e diseased seedling after metalaxyl-M above- and belowground; c, f control above- and belowground

Full size image

After 50 days, the disease incidence (P%) and disease index (DI) differed significantly (Fpr < 0.001) between treatments (Table 3). Approximately 50% of the control seedlings were wilted compared to 6% of seedlings in the metalaxyl-M treatment. Plants treated with metalaxyl-M and potassium phosphonate had the lowest P% after 90 days with 26.7% and 28.3% of seedlings, respectively, followed by mancozeb (40.0%). The mean disease index of these fungicides was also small, ranging from 0.57 to 0.93. In comparison, the biofungicide T. viride was only mildly effective with P% of 60.0%. The untreated control had a disease incidence of 96.7% and disease index of 2.82.

Table 3 Disease incidence (P%) and disease index (DI) for foliage on 6-month-old Citrus reticulata seedlings after various chemical or biological treatment
Full size table

The pathogen was reisolated from roots, and all isolates (15 per treatment) on agar produced lumpy-branching hyphae with hyphal swellings. Sporangia were papillate and caducous with short pedicels up to 6 µm long. Sporangial shape varied from ellipsoid, ovoid, pyriform, obpyriform to near spherical. The characteristics were the same as described previously by Chi et al. (2019).

Reducing P. palmivora disease in a mandarin orchard

Disease classification of the canopy before treatments were first applied indicated that the trees were infected mildly and homogeneously (DI = 0.50–0.51; disease incidence P% = 40.0–42.2%, the trees were yellowing and had sparse foliage, ensuring homogeneity for the experiment before application of treatments (Fig. 3a). The P% and DI for plants after 6 months differed significantly (Fpr < 0.001) among treatments (Table 4).

Fig. 3
figure 3

Appearance of Citrus reticulata trees in commercial orchard after infection with Phytophthora palmivora. a When the trial was established, approximately 40% of the trees were yellow. b Six months after potassium phosphonate treatment, the trees were green, and 50% had recovered. c Six months after Trichoderma viride treatment, the trees had some new shoots, but only 5.4% had recovered. d Untreated control trees had yellow leaves, dead branches or loss of canopy, and symptoms were more severe than 6 months before

Full size image
Table 4 Phytophthora palmivora disease incidence (P%) and disease index (DI) for foliage on Citrus reticulata trees in a commercial orchard at the time of application various chemical or biological treatments and 6 months after application
Full size table

After 6 months, trees treated with potassium phosphonate or metalaxyl-M yielded the lowest P%, 21.1% and 22.2% of trees, respectively. The mean DI of these treatments decreased to 0.30 to 0.37. T. viride was only weakly effective with P% of 38.9% and DI of 0.77, the latter higher than before the treatment was applied (Fig. 3c). The P% and DI in the control treatment increased significantly to 66.7% and 1.20, respectively (Fig. 3d). Disease symptoms were the most reduced by potassium phosphonate (Fig. 3b).

Discussion

This is the first study to compare the effectiveness of fungicides and biological agents against P. palmivora, the cause of Phytophthora disease in C. reticulata in Vietnam. The results showed that metalaxyl-M, potassium phosphonate, mancozeb, and T. viride strongly inhibited the growth of P. palmivora in vitro and markedly reduced disease in inoculated seedlings. Furthermore, potassium phosphonate and metalaxyl-M applied by spraying the canopy and soil led to the recovery about 50% of grafted mandarin trees in a commercial orchard. These findings provide an interim option for producers to better manage Phytophthora disease in Vietnam.

The fungicides identified in the present study with the greatest inhibitory effect against P. palmivora have been tested in numerous studies for their efficacy against a wide range of diseases in crops and trees. Potassium phosphonate has been widely used for decades to control root rot caused by P. cinnamomi in many different forest, plantation, and horticultural tree species (González et al. 2017). Potassium phosphonate has been used to control lemon brown rot caused by P. citrophthora both in vitro and in orchards. For example, it reduced by 40–60% the incidence of rot in fruit (Ramallo et al. 2019). Metalaxyl-M + mancozeb, fosetyl-Al, and phosphorous acid have been used to control Phytophthora crown and root rot caused by P. palmivora in apricot and cherry (Türkölmez and Derviş 2017). Metalaxyl-M and mancozeb inhibited mycelial growth of Phytophthora species in vitro (Rolando et al. 2017) and in pot trials (Reglinski et al. 2009), and they have been used for management of Botryosphaeria causing stem cankers in field-grown Protea magnifica (Denman et al. 2004). In our study, metalaxyl-M, potassium phosphonate, and mancozeb strongly inhibited the growth of P. palmivora in vitro. After screening of fungicides in vitro, it is important that field trials are established as efficacy of fungicides can be changed in field conditions for control of pathogens.

The same three fungicides were also effective in reducing symptoms of P. palmivora disease in nursery seedlings and in a commercial mandarin orchard. The disease incidence on 6-month-old inoculated seedlings was less than 40% following treatment, whereas the control seedlings were severely diseased (P% = 96.7%). In the commercial orchard, the trees showed symptoms of disease in the years before treatment (2017: P% = 8.0–10.0%; 2018: P% = 15.0–20.0%), and at the time of treatment, the trees were just coming into bearing age (Chi et al. 2019). The disease incidence of mandarin trees was reduced to less than 28% after treatment with potassium phosphonate, metalaxyl-M, and mancozeb, and the untreated trees in the control treatment remained severely diseased (66.7% disease incidence).

We showed that the biological agent T. viride had some potential to inhibit the growth of P. palmivora in vitro and reduce disease on seedlings and in mandarin trees. However, it was not as effective as potassium phosphonate, metalaxyl-M, or mancozeb. On our results so far, the putative biological control agents that were tested are not recommended for commercial use in mandarin orchards in northern Vietnam. So far, most commercial biological control agents apparently are ineffective in controlling Phytophthora in vivo.

In this study, we only tested a single concentration of each treatment, and therefore, we were unable to determine the most effective concentration of each treatment. Further research is required not only to determine the optimal concentrations for control of P. palmivora disease in C. reticulata in the field, but also to identify the number and timing of applications to optimize disease control.

As new fungicides are approved for use in Vietnam, they can be evaluated in local conditions. Recently, registration for the fungicides ethaboxam and mandipropamid has been requested, and two other fungicides, fluopicolide and oxathiapiprolin, have received full registration for the control of Phytophthora root rot in citrus in the United States (Hao et al. 2019).

Another factor to be considered is the mode of application of fungicides to citrus in Vietnam. P. cinnamomi root rot in avocado was controlled by spraying or drenching with phosphorous acid and a 20% solution of partially neutralised phosphorus acid injected into the trunks of 7-year-old diseased trees resulted in the rapid recovery of trees (Pegg et al. 1985). Trunk injections are now routinely being applied to many tree species. Holderness (1992) showed that trunk injection of potassium phosphonate gave control of P. palmivora pod rot comparable to that obtained with conventional metalaxyl sprays. Furthermore, potassium phosphonate foliar sprays gave poor control of both canker and pod rot and were less effective than either trunk injection or metalaxyl sprays.

More attention should be paid to hygiene, selection of rootstocks, orchard design, and management. Pomelo seedlings are routinely used as rootstock in northern Vietnam, but they are susceptible to P. palmivora and P. nicotianae (Hung et al. 2015b, a). The local mandarin in China (C. reticulata) called Guanggan is strongly resistant to P. nicotianae (Yan et al. 2017) and could be evaluated as a rootstock in Vietnam. Phytophthora disease is known to be more prevalent when citrus orchards are flooded (Chaudhary et al. 2016). The orchard used in our study was well drained, but many orchards in the region have water ponds between trees after monsoonal rain. Future work to develop sustainable integrated management strategies for local conditions will greatly benefit growers and the environment.