Optimization of cultural practices to reduce the development of Pseudomonas syringae pv. actinidiae, causal agent of the bacterial canker of kiwifruit

2.1Biological material and phytopathometric evaluations

2.2Influence of fertilization on disease development

The influence of nitrogen form, rating and mode of application (root or foliar) were investigated. In addition, the role of micronutrients deficiency or excess on the disease development was studied.

Psa inoculation on plants was done by spraying a 10⁸ CFU mL^–1 bacterial suspension in 10 mM MgSO₄, one week after treating with the first dose of nitrogen, two weeks after transplanting on micronutrient-deficient medium, or two weeks after the first excess micronutrient supply, according to the experiments. Symptoms developments were monitored weekly.

One month after inoculation, the endophytic population of the pathogen inside each plant was measured. Plant weight, number of new leaves and chlorophyll fluorescence (measured with the PSI equipment, Photon Systems Instruments, Czech Republic) were monitored. On micropropagated plants, carbon exchange was also measured by means of the EGM-4 equipment (PP System international, Amesbury, USA). Each experiment was replicated at least twice over two years. Statistical significance of differences among treatments was assessed by ANOVA followed by Fisher’s Least Significant Difference test with P < 0.05.

2.3Influence of irrigation on disease development

2.4Role of pruning in the disease cycle and curative pruning

The experiments aimed to investigate the role of pruning cuts as possible entry points for Psa, the influence of the different commercial pruning strategies on the disease occurrence, and the migration rate of Psa inside the plant organs.

2.5Effect of the training system on disease incidence

A commercial field of A. deliciosa cv. Hayward located in Castel Bolognese (Italy, 44° 20’ 16” N; 11° 48’ 58” E) was split in two plots, where the vines were trained according with the two typical training systems in Emilia-Romagna (‘Pergola’ and ‘Geneva Double Curtain’, GDC). Symptom development was monitored monthly for 1 year. During the experiment, microclimatic data inside the canopy were recorded at hourly intervals. Two data logger sensors (EasyLog-UBS-1) were placed in each plot, at the level of the leader and on the outer side in mid-length of the canes to collect temperature and humidity data for six months (June-November). The data obtained from the sensors were expressed as an average of 10-day cumulates, and the significance of their differences was assessed with the Student’s t test (P < 0.05). Finally, the penetration of water sprays inside the canopy was evaluated as previously described.

2.6Influence of the use of bioregulators on disease development

The experiments were performed in both controlled and field conditions. The influence of synthetic auxins and cytokinins on plant susceptibility to the disease was evaluated. 3,5,6-trichloro-2-pyridyloxyacetic acid (TPA, marketed as Maxim^®, L. Gobbi S.r.l., Genova, Italy) was used as the synthetic auxin and 1-(2-chloropyridin-4-yl)-3-phenylurea (CPPU or Forchlorfenuron, marketed as Sitofex^®, AlzChem AG, Trostberg, Germany) was used as the synthetic cytokinin.

2.7Effect of Psa infection on fruit production, quality and storability

In 2014, 100 fruits were collected from symptomatic and symptomless plants in commercial orchards of A. deliciosa cv. Hayward (44° 15’ 13”N; 11° 50’ 52” E), A. chinensis cv. Hort16A (44° 14’ 10” N; 11° 49’ 35” E),and A. chinensis cv. Jintao (44° 20’ 1” N; 11° 49’ 45” E) located in Faenza. The fruits were analyzed at harvest, after two months in cold storage and at the end of cold storage (4 months post harvest). Differences among samples from different plants were analyzed by Student’s t test with P < 0.05.

In addition, A. deliciosa cv. Hayward fruits were collected from symptomless and symptomatic plants. From the latter ones, fruit samples were also taken specifically from canes where symptoms were present. Qualitative tests were performed at harvest, after two months in cold storage, and after one week in shelf life. ANOVA and the Fisher’s Least Significant Difference test were applied to assess the significance of differences forP < 0.05.

The following parameters were measured: fruit weight, DA Index (I_AD), flesh firmness, ethylene production, flesh colour, soluble solid contents, dry matter and titrable acidity. I_AD was measured with “kiwi-meter” (TR, Forlí, Italy) on the peel (“skin DA Index”), and after peeling out 1 (for A. deliciosa) or 2 mm (for A. chinensis) (“flesh DA index”) [10]. The flesh firmness was measured by using a penetrometer (Güss Fruit Texture Analyzer, Cape Town, South Africa) with 8 mm tip on two orthogonal faces of the fruit, after peeling. Ethylene was measured by gas chromatography (Dani 86.10 HT, Milano, Italy) on two-fruits samples, closed in 1.7 L vessels for two days. Flesh colour was determined by colorimeter (CR-400, Konica Minolta, Tokyo, Japan) after removing 1 mm (for A. deliciosa) or 2 mm (for A. chinensis) of peel on two portions of the fruit placed orthogonally. Soluble solid content was measured by a digital refractometer Atago (Optolab, Modena, Italy). Dry matter was determined by drying a 2-mm thick fruit slice at 60°C for 48 hours. Acidity was measured by titration with 0.25 N NaOH, using a Crison Compact titrator (Crison Instruments, Barcelona, Spain).

2.8Statistical analyses

The STATISTICA version 5 software (StatSoft Inc, Tulsa, USA) was used for calculation of ANOVA, Fisher’s Least Significant Difference test and Student’s t test. The software SPPA 2.0 was used for geostatic analysis of the data.

3.1Effect of nitrogen form, dosages and mode of application in controlled conditions

Due to the high variability of response inside each of the fertilization treatment, the differences among samples treated with different nitrogen compounds or ratings were not statistically significant. However, the plant receiving NPK fertilization had the lowest disease incidence and severity, whereas the ones fertilized with ammonium nitrate showed incidence and severity vaules up to five and three times higher, respectively. All the recorded biometric parameters (production of new leaves, plant weight and quantum yield) were not influenced by the form of nitrogen applied to the plants. The highest nitrogen doses were associated to the mildest symptomatology, but also with the highest Psa endophytic populations. 100% nitrogen ratings decreased the dry matter content in leaves and it also caused a significant leaf drop. No differences in the symptom development or in the bacterial endophytic population were noted in relation to root or foliar application.

3.2Effect of micronutrients deficiency or excess

The depletion of any single microelement, among those tested, did not cause statistically significant differences compared to control plants with regard to symptoms, although the deficiency of specific micronutrients, such as Ca and Mn, slightly increased symptoms severity (Fig. 1A). The depletion of Ca, Fe and B generally caused an increase in Psa endophytic population (Fig. 1B). The excess of micronutrients did not statistically influence symptoms development (Fig. 2A), nor Psa endophytic population (data not shown). However, the excess of B and Ca slightly increased symptom severity. QY was significantly reduced by the excess of B and Mn (Fig. 2B).

3.3Influence of irrigation on disease development

Water deprivation increased disease symptom severity (Fig. 3), but not incidence. The source of the water supply used for irrigation influenced the disease incidence: for the whole duration of the monitoring, orchards irrigated with internal water sources presented a higher disease incidence than those receiving water from running water natural rivers or artificial canals (Fig. 4).

3.4Role of pruning in the disease cycle

The wound caused by pruning represented a possible entry point for Psa for longer than 1 month both in A. deliciosa and A. chinensis. Pruning performed when the plant is completely dormant resulted in the lowest disease incidence and severity, while late pruning, performed at the beginning of the vegetative season (bleeding sap), induced the highest disease incidence and severity (Table 1). Among the three different pruning types (standard, short, and zero leaves), tested on the Pergola training system, the short pruning granted a more effective water spray penetration at full leaf development (’medium’ drop size in June and September, compared to ‘fine’ or ‘very fine’ in the other pruning systems; Table 2). No significant changes in temperature and leaf wetness were recorded in the three cases (data not shown).

In three infected commercial orchards monitored over three years, a significant correlation was found between the position of diseased plants along each row in the orchard. In addition, the closest association between diseased plants was found below 5 meters, corresponding to the distance between rows.

3.5Bacterial migration and host colonization

Seven, 14, 21 and 120 days after stab-inoculation, self-rooted or grafted A. chinensis (cv. Jintao) plants were processed to verify the maximal length of bacterial migration. During 21 days, the maximum basipetal and acropetal migrations were, respectively, 3 and 2 cm. At the maximal distance of migration, the endophytic Psa population size was generally comparable to the one at the infection point. In 4 months, the pathogen had completely invaded the plant tissues in the apex direction, whereas its basipetal migration was approx. 138 cm (Table 3). Furthermore, Psa completely colonized the roots as well, overcoming the grafting point. Psa also migrated through the branching points (nodes) in the different directions (data not shown). The presence of the pathogen was confirmed in infected A. chinensis (cv. Jintao) plants after uprooting, regardless of whether the plant was grafted or self-rooted. In fact, the roots of infected plants hosted average populations of 4.27×10⁸ CFU g^–1 in grafted plants and 1.15×10⁸ CFU g^–1 in self-rooted plants, with bacterial populations in radicles even higher than in the aerial parts (Table 4).

3.6Effect of the training system on disease incidence

The influence of two training systems (pergola and GDC) on the microclimatic conditions inside the canopies and the penetration of water sprays was assessed. Water drop dispersion, after the full development of leaves, was most effective in the short pruning system (Table 2). Microclimatic data collected within the canopies showed no significant differences between two different training systems. However, GDC-trained canopies were constantly slightly warmer and less humid (Fig. 5).

3.7Influence of the use of bioregulators on disease development

The effects of forchlorfenuron (CPPU) and TPA on Psa are summarized in Fig. 6. When supplied 30 days before inoculation, both CPPU and TPA induced lower disease index scores than water-treated plants. When the bioregulators were applied 15 days before inoculation, disease incidence and severity were slightly reduced by the higher CPPU rate, but increased by TPA. Treatments performed 7 days before inoculation resulted ineffective (for CPPU) or detrimental (for TPA). Although subjected to high variability, bacterial populations were in agreement with symptomatology. No differences were noted between the application of pure active ingredients and commercial formulates. In field trials, both CPPU and TPA incremented the disease incidence (although differences were not statistically significant) when applied to increase fruit production (Table 5).

3.8Effect of Psa infection on fruit production, quality and storability

The analysis of fruits from symptomatic and asymptomatic plants (assessed by visual evaluation of symptoms) showed that the quality parameters are influenced by the disease and the changes in fruit quality can be observed both at harvest and after cold storage. Fruits harvested from symptomatic plants had a lower weight, particularly in A. chinensis cultivars (Table 6). Flesh colour at harvest is also affected by the disease, being enhanced in A. chinensis cultivars, and clearer in A. deliciosa cv. Hayward. Differences are even more evident between the fruits taken from asymptomatic plants and those taken from symptomatic branches of diseased plants (Table 7). Ethylene production by fruits harvested from diseased plants was also increased (Fig. 7).

4.1Influence of fertilization on disease development

Nutritional imbalances may cause stress in the plant making it more susceptible to pathogens attack. Our findings showed that nitrogen fertilization promotes bacterial growth and disease development, like for other biotrophic pathogens exploiting host metabolism for their multiplication [11]. Furthermore, in field conditions, high nitrogen fertilization results in denser canopies, which may provide favourable microclimatic conditions for Psa epiphytic growth. Denser canopy may also have an indirect positive effect for the disease development: in fact, they reduce the penetration of the copper sprays generally used for Psa control. Finally, nitrogen stimulates shoot growth, which, in turn, leads to the need of more pruning interventions, thus increasing the risk of infection through pruning wounds. The effect of nitrogen on Psa did not depend on its application mode (root or spray), nor its form (ammonium and/or nitrate). No differences in lignification (which could modify the tissue susceptibility to pathogens) were noticed among the treatments. In conclusion, our results suggest to reduce the use of nitrogen fertilization. In general, micronutrient deficiency promotes bacterial growth, especially in the case of iron. Iron is a key nutrient for Pseudomonas spp. [12]. However, we found that also the deficiency promotes the growth of the bacterium in planta, possibly due to the induction of early senescence, facilitating Psa growth by the inactivation of plant defences and the demolition of nutrient storages [13].

4.2Influence of irrigation on disease development

Leaf wetness and relative humidity are crucial to allow Psa epiphytic survival and growth [4]. In addition, we observed that water stress induced a more severe symptomatology. This may be explained by the worsening of drought stress due to the colonization of vascular tissues by Psa, reducing the capacity of vessels. Indeed, on drought stressed plants in greenhouse conditions, we observed that Psa symptoms often result in large desiccated areas harbouring a consistent bacterial population, which differ from the most common angular leaf spots. The spread of Pseudomonas syringae pv. syringae has been demonstrated to depend on the water cycle [14]. Our research showed a positive correlation between disease incidence and the use of irrigation water pumped from artificial basins collecting the rainwater from the infected orchards, thus suggesting a possible role of irrigation water in the local spread of the disease. However, further experiments are needed to fully elucidate the role of irrigation water in vectoring Psa. For example, in the area where the experiments have been performed, the only irrigation method used is drop irrigation to the root system and there is no evidence of Psa infecting the plants via the root system. A tentative explanation could be that irrigation by water containing Psa may increase the pathogen population in the litter where it can survive, even in harsh conditions, for months [15]. The increased environmental Psa population in the orchard may facilitate the spread of the pathogen to susceptible host parts such as leaves and flowers.

4.3Role of pruning in the disease cycle and curative pruning

Pruning cuts are possible entry points for Psa and, in the period when pruning is usually done (late winter) they need more than a month to heal. Therefore, the minimization of pruning interventions is suggested. In addition, post-pruning treatments able to seal the cuts, or the application of protective and/or disinfectant compounds to decrease Psa population may further reduce the risk of infection. On the other hand, pruning performed during vegetative dormancy are associated with the lowest disease risk.

The comparison of two common training systems (GDC and Pergola) showed similarities in the microclimatic conditions inside the canopy, although the GDC training system allowed slightly higher temperatures and lower humidity compared to the pergola, possibly due to the increased light penetration and air flow into the canopy. Psa shows the highest aggressiveness between 12–18°C, while its virulence is inhibited by temperatures above 25°C [16]. Thus, even though no difference in Psa incidence was observed in field conditions, the GDC training system seems to provide less favourable conditions to Psa infection. However, a reduction of the disease risk may also be achieved with the more widespread Pergola training system, by implementing specific pruning strategies. For example, the short pruning allowed the best accessibility of canopies to water sprays, while in the zero leaves system, slightly higher temperature values were observed in canes area, probably due to the greater penetration of light inside the canopy. These data suggest that the short and zero leaves pruning forms allow better aeration and irradiation of the foliage.

To set up an effective curative pruning, Psa rate and direction of migration inside the plant tissues are crucial information. After an initial slow migration, in 4 months, the bacterium was able to colonize the whole plant both acropetally and basipetally. Therefore a prompt pruning of the infected limbs may prevent the systemic migration of the pathogen. Nonetheless, when symptoms (i.e. cankers) are evident on canes, the infection has probably spread systemically. The branching nodes and the grafting point do not prevent bacterial migration. Therefore, coppicing and re-grafting are not suitable to reconstitute a productive plant.

Finally, our research also aimed to evaluate whether the current pruning practices may contribute to disease spread. In order to contain disease spread in open field, a prompt sanitary pruning and eradication of whole infected plants are crucial practices. The mapping of affected orchards demonstrated that, during the early stages of infection, the diseased plants mostly aggregated along the row rather than among adjacent rows. Thus, mechanical pruning performed along the row or, in case of manual pruning, the ineffective disinfection of tools may be regarded as a major issue for the bacterial spread.

4.4Influence of bioregulators on disease development

In controlled conditions, the synthetic cytokinin (Forchlorfenuron), applied up to one month prior inoculation, may induce a moderate resistance against Psa. On the other hand, the auxinic compound TPA enhanced disease development. Evaluations of treatments with synthetic auxins and cytokinins agree with previous observations, where auxin and cytokinins showed, respectively, antagonistic and synergistic effects on SA signalling [17]. In addition, the application of TPA in young plants caused phytotoxic effects, including deformation of shoots and hyperthrophic growth of the shoot tips, which also resulted in diffuse micro and meso-lesions of the plant surface. These lesions may facilitate Psa infection and spread. In field conditions, the use of bioregulators slightly increased Psa disease incidence. However, the disease incidence in all the 4 orchards considered for the experiments was very low. Thus, further experiments are needed to fully elucidate the influence of bioregulators in field conditions.

4.5Effect of Psa infection on fruit production, quality and storability

In order to verify whether fruit quality, besides crop yield, was affected by the bacterial canker, fruits were separately harvested from healthy and symptomatic plants and, in the latter case, at least for A. deliciosa, specifically from symptomatic canes. The quality parameters at harvest and during storage of these fruits were monitored. In general, fruits from symptomatic plants were smaller and with lower sugar content. Although not a typical climacteric fruit, kiwifruit produces ethylene in its post-harvest life [18]. In this work, a higher ethylene production was recorded by fruits from the symptomatic plants, and in particular by those from the most infected parts, compared to fruits from asymptomatic plants. The higher ethylene production may induce a faster ripening of fruits from infected plants during post-harvest with consequent storage disorders.

4.6Conclusions

This work provided evidence that different agricultural practices may significantly contribute to the control of bacterial canker. In particular, balanced fertilization, sufficient irrigation from non-contaminated water sources, and the limitation of pruning may contribute to reduce Psa disease incidence and severity. In addition, less dense training systems, such as GDC, are suggested when planting new kiwifruit orchards.