Some aspects of biology and host range of Biscogniauxia mediterranea, one of the causal agent of oak charcoal disease

INTRODUCTION

Zagros forests (ZFs) extend from northwest to southeast of Iran and their total area is about 4 million hectares (Sabeti 1994). These forests play an important role in protecting soil, water and the environment. Human activities such as using forest wood for fuel, conversion of forests to agricultural lands and other purposes and also unconditional grazing of livestock have noticeably destroyed some parts of ZFs. Moreover, deleterious effects of pests, plant diseases and parasites and also severe drought have increased By, increasing the concentration of the micron-sized dust and increasing their establishment time in different areas of ZFs, different adverse effects on the plant's physiology and forest trees have observed. It is believed that the weakening of trees and shrubs in recent years due to successive droughts has led to increase their susceptibility to pests and diseases in these areas.

Most of the trees in the Zagros mountains are mainly from oak (Quercus L.), including Q. brantii Lindl., Q. infectoria Oliv. and Q. libani Oliv. (Jazirehi & Ebrahimi Rostaghi 2003). Charcoal disease caused by Biscogniauxia mediterranea (De Not.) Kuntze has been reported as the most common and important disease of oak trees in Zagros forests (Mirabulfathy 2013). In subsequent studies, Mirabolfathy et al. (2013) showed that in some Zagros areas, the signs macroscopically similar to that of charcoal disease on some oak trees are related to the species Obolarina persica Mirabolfathy, Ju, Hsieh & Rogers. Following this research, Safaee et al. (2017) designed a multiplex PCR using two pairs of specific primers to detect and identify the two species in oak tree tissue before the disease sign appears.

Biscogniauxia mediterranea has been in the center of attention in most of the oak forests over the world (Jurc & Ogris 2006, Vasilyeva et al. 2007, Whalley 1996, Vanini & Valentini 1994). Most important aspects of the pathogenicity of B. mediterranea are the relationship between the role of drought stress and the disease outbreak (Vannini et al. 2009, Vanini & Valentini, 1994, Vannini & Scarascia Mugnozza, 1991, Ragazzi et al. 2001, Linaldeddu et al. 2011, Luchi et al. 2005), the molecular identification of the cause and tracing of the disease before the sign appears (Mazzaglia et al. 2001, Luchi et al. 2005, Safaee et al. 2017).

Vanini et al. (1999) Study the genetic structure of B. mediterranea using Random Amplified Polymorphic DNA (RAPD) showed that there is a great genetic diversity within the population derived from a single stroma and among various isolates collected from a limited area, therefore, they proposed a heterothallic mating type system for the fungus.

Biscogniauxia mediterranea has a long endophytic stage in which the infected trees have no symptoms. The fungus acts in an aggressive manner under drought stress and high abnormal temperature. The considerable effect of water content on the activity of B. mediterranea has been proved. It is believed that the development of the fungus in the endophytic phase depends on the reduction of the host water potential (Vannini & Scarascia Mugnozza 1991).

Vannini et al. (1996) investigated the factors influencing the release and germination of B. mediterraneua (H. mediterraneum) ascospores on Turkish oak (Q. cerris) under natural and laboratory conditions and concluded that ascospore release and germination depend on the presence of free water or high relative humidity ,however they did not test ascospore pathogenicity.

In Spain, Martin et al. (2005) investigated the correlation between wrong operations for the harvesting of cork oak trees (Q. suber) and perforation caused by beetles and contamination to B. mediteranea in oak forests. They realized that there is a strong correlation (r = 0.86) between wrong harvesting operations and holes made by beetles (Cerambyx spp.), and also between the damage caused by beetles and contamination to B. mediteranea (r = 0.91).

Toxins extraction and identification from B. mediterranea implemented by Evidente et al. (2005) and the results showed that in addition to the two previously known toxins, phenylacetic acid and 5-methylmellein, this fungus produces another toxin called biscopyran, which is non-specific and makes epinasty in Q. suber.

Lack of research and information about the biology of B. mediterranea caused some difficulties in the biology of this species, however the role of ascospore and conidium in pathogenicity, the survival duration (viability) of ascospores, development of sexual and asexual stages and host range of the species were investigated.

MATERIALS AND METHODS

Fungal Isolates

Samples of stromata containing perithecia of charcoal disease agents were collected from different regions of ZFs in Kurdistan, Kermanshah, Ilam and Lorestan provinces during 2014-2016. Pure cultures were obtained from single ascospore. To this, the contents of few adjacent perithecia were added to 1 ml of sterilized water by a sterile needle and after shaking, 10 μL of suspension was spread over the WA medium. After 24 to 48 hours, one germinated ascospore was transferred to a PDA plate. All isolates were kept on slant PDA tubes at 4°C.

Identification of oak charcoal disease agent

To identify the species, the characteristics of the teleomorph (stroma, perithecium, ascus and ascospore) were investigated and compared with descriptions of the genus and species in valid mycological references (Ju & Rogers 1996, Ju et al. 1998, Ju & Rogers 2001). To confirm the morphological identification of the species based on molecular characteristics, a pair of specific primers MED1 (forward: 5́-GCTATAAGGGATCACCGCC TC-3′) and MED2 (reverse: 5́-TGCAGAAGCGAGA GTGTGG-3′), designed by Mazaglya et al. (2001) were used. DNA from colonies of single spore isolates was extracted by using Chelex 100 (BIORAD, USA) based on the method of Hirata & Takamatsu (1996) and its concentration was determined using a spectrophotometer (Epoch 2, BioTek, USA). Polymerase chain reaction was performed in 200-μl microtubes containing 25 μl mixtures of reaction. The reaction mixture consists of a PCR buffer (1X), 0.2 mM of dNTPs, 2 mM of MgCl₂, 0.6 μM of each of the MED1 and MED2 primers, 0.8 units of hot start Taq DNA polymerase (Thermo Scientific, Lithuania) and 2 ng of DNA extracted from each isolate. Moreover, a negative control (reaction mixture without DNA) and a positive control (reaction mixture containing DNA extracted from an identified B. mediterranea) were used. The reaction was performed in a thermocycler device (BIO-RAD, USA). The thermal program including 36 cycles was performed as follows. An initial denaturing at 95°C for 4 min, two cycles of  denaturing at 95°C for 40 s, annealing at 62°C for 30 s, extension at 72°C for 30 s, two cycles of 95°C for 40 s,  61°C for 30 s,  72°C for 30 s, 32 cycles of 95°C for 40 s, 60°C for 30 s, 72°C for 30 s and a final extension  at 72°C for five min. Amplified DNA fragments and a 100-bp DNA marker (Sinaclon, Iran) were passed through Agarose gel 1.5% (Sinoclon, Iran) containing DNA safe stain  (Sinoclon, Iran) in an electrophoresis device. After electrophoresis, the DNA bands in the agarose gel were visualized and photographed using the UV-gel documentation system (UVITEC, France).

Determination of the host range Based on the field observation:During investigating charcoal disease on oak trees (Quercus spp.), its possible presence on other forest species including Crataegus spp., Amygdalus spp., Acer monspessulanum, Cerasus microcarpa,  Pistacia spp., Rosa spp., Ulmus sp., and Pyrus spp. was also considered.

Based on inoculation on excised branches: According to Safaee et al. (2015), from healthy trees of Quercus spp., Crataegus spp., Amygdalus spp., Cerasus microcarpa, Pistacia spp., Rosa spp., Ulmus sp., and Pyrus spp. in the areas without charcoal disease, some branches with a diameter of about 2 to 4 cm and a length of about 1 to 1.5 meters were cut and transferred to the laboratory. These branches were segmented into pieces of 30 cm in length and were first washed with tap water and then completely immersed in 0.5% sodium hypochlorite for 10 min and then soaked in sterilized water for 15 min. Inoculation was performed with and without making a wound in the bark. The wound was made in the medial of the excised branch by removing one square centimeter of bark. A 5-mm diameter mycelium plug of the fungus was placed into each wound and the inoculated area wrapped with Parafilm. In none-wounded treatments, after placement of the inoculum on the bark surface, it was covered in the same above manner. In control treatments, autoclaved colony plugs were used. The experiment was conducted as a factorial based on a completely randomized design. The treatments consisted of, 1) with wounding 2) without wounding and 3) control treatments in 5 replications (each replication included two excised branches). The inoculated excised branches were placed in a sterilized plastic container and held in an incubator at 25 ± 1°C. The length of necrosis tissue under the bark was measured after a month and used for comparison. To re-isolate the pathogen, fragments from the margin of healthy and diseased tissue were removed aseptically and cultured on PDA containing streptomycin sulfate.

Based on seedling inoculation: Regarding the importance of oak species, the pathogenicity test of B. mediterranea on seedlings of three species of oak (Q. brantii, Q. infectoria and Q. libani) which are most frequent in Zagros forests, was conducted on 20-month-old seedlings in 3.5-liter plastic pots containing forest soil. An experiment was conducted as factorial on the basis of a completely randomized design at five replications, each replication including two pots and each pot including one seedling. Inoculation of seedlings was performed by placing a plug of the fungal colony into the wounded stem (in bark-wounded treatments) and on intact stem surface (in non-wounded treatments), then the inoculated area wrapped with Parafilm. The autoclaved inoculum was used in controls. The pots were held in the greenhouse for two months after inoculation, and then necrosis extension in inner bark was evaluated. Pathogenicity proof was carried out by re-isolation of the fungus.

Pathogenicity and survival

Preparation of ascospores and conidium suspension: Of the sixteen stromatic samples containing live ascospores, six samples were randomly selected. Ascospores of three pieces of each sample were collected according to Vannini et al., (1996). The outer layer of the pieces (about 2 × 2 centimeters in size) was removed from the perithecial stroma so that the perithecium opening appeared. Then these pieces were set on a sterilized wet tissue paper into Petri dishes and placed in an incubator at 8°C. After 48 hours, the ascospores released on the inner surface of the Petri dishes were collected by adding sterile distilled water and pipetting, resulting in a concentration of about 10⁶ ml^-1.

Of the sixteen single spore isolates of B. mediterranea, six isolates were randomly selected and cultured on oatmeal agar (OMA) for production of conidia. The cultures were placed under a 12-hour light/dark condition with a mixture of fluorescent light and near ultraviolet at 25± 3°C for 10 days. Conidia were washed from the colony surface with sterilized distilled water, filtered through glass wool to remove mycelial fragments. The concentrations of conidia and small passed mycelial fragments (SPMFs) were determined by microscopic count with a hemocytometer which resulted in concentrations of 10⁵ and 50 ml^-1 conidium and SPMFs respectively.

The rate of ascospore and conidium germination: The amount of 5 μl of ascospore suspension of six above samples were spread on PDA, MEA and WA media with three replications and incubated at 25°C for 48 hours. Fifty ascospores were randomly observed microscopically (Olympus-BH2, Japon) at 400X at each replication on plates. When germ tube emerged from ascospores, they were regarded as germinated ascospores.

According to this method, the rate of conidium germination of six above conidium suspensions was calculated.

Inoculation of seedlings with ascospores and conidia: Eighty seedlings of 20 months old seedlings of Q.brantii, approximately equal in height, were inoculated with each ascospore and conidia just as the previous method for seedling inoculation except that inoculum was a suspension of ascospores or conidia at concentrations of 10⁶ and 10⁵ ml^-1 respectively. Pathogenicity assessment was performed in the same way as before.

Survival of ascospores: In the protected zone of Bisotun (Kermanshah province), three Q.brantii trees containing stroma of B. mediterranea were marked. From each tree, a branch containing stroma was cut so that half of the stroma remained on the tree and the other half was transferred to the laboratory. Ascospore suspensions were prepared from each of them, and their rate of germination was calculated and considered as initial germination. After that, we split the stroma-bearing branches and put a part of them in the paper bags and kept in the room condition and hung the other part on a tree in the yard of Kermanshah Agricultural and Natural Resources Research and Education Center (KANRREC), which provided almost conditions similar to the presence of stroma on the trees under natural conditions. After one year, the percentage of germination (secondary germination) of ascospores on marked oak trees, the hung parts in the yard and the parts in paper bags were estimated, and their mean differences were compared by Duncan's test. The percentage reduction of germination after one year was calculated with the following formula:

Ascospore germination = [(Percent of primary germination - Percent of secondary germination)/Percent of primary germination] × 100

Development of sexual and asexual stages

The time of the onset of the symptoms after penetrating B. mediterranea in the oak tissue, the rate of expansion in the tissue and the development of a sexual and asexual form of the fungus were studied under the following conditions.

Under field conditions: In one of the forested regions of Kermanshah province, in which charcoal disease was prevalent, we selected 12 Q. brantii trees with trunk diameter around 17-21 cm at breast height, in an area of about 2000 m². These trees did not have stroma of charcoal disease and also its causal agent was not isolated from their tissue. In six of them, a piece of trunk bark (approximately 5 × 5 cm) was removed with a chisel (sterilized with alcohol and flame between each use). Then some of the inoculum (colony of B. mediterranea on a PDA of about 9 cm² in size) was placed on the wood surface and the removed bark was returned to its place so that the inoculum was set between the tissue of the inner bark and the wood. The inoculated spot was first covered with wet autoclaved cheesecloth and then with parafilm. Six other trees were inoculated with autoclaved fungus colony as a control. These trees were surveyed in autumn, winter, spring, and summer every 15, 30, 10 and 20 days, respectively during 24 months. In these periods, we investigated the possibility of forming of the stromata containing perithecia and/or the emergence of the colonies bearing conidia in the site of inoculation and surrounding tissues. By removing the tissue at 5 cm intervals from the margin of the inoculated wound, the extension of necrosis in the tissue was assayed and some samples for re-isolation of the fungus were provided. At the end of the experiment, the outer bark was removed around the inoculation site until the necrosis continued to measure the longitudinal extension (along the trunk). Also, we sampled some inner bark and wood tissues in the opposite side of the inoculation site to detect possible invasion of the fungus across the trunk.

On the excised branches: Three excised branches of each Q. brantii and Q. infectoria were inoculated by B. mediterranea as mentioned above and kept in the incubator at 25°C for one month. Subsequently, all of them were placed below a wired cloth box (Fig. 1) in the garden of the KANRREC beside which several pieces bearing B. mediterranea perithecia were placed. At the above-mentioned intervals, their status was examined. After 24 months they were individually placed in a paper bag and kept at room temperature.

In greenhouse: Ten 2-year-old seedlings of Q. brantii were inoculated with B. mediterranea by placing the mycelial plug into the wound on the stem as earlier mentioned and five seedlings were used as a control. The inoculated plants were held under normal conditions in the greenhouse. All the treatments were monitored almost every week for 34 months to record fungal and disease development. At the end of the experiment, re-isolation of the fungus from the margin of the healthy and infected tissues was performed.

Fig. 1. The excised branches of Q. brantii and Q. infectoria inoculated by B. mediterranea and placed on sand surrounded by a wire cloth.

RESULTS

Identification of fungal isolates

Thirty-one stromatic samples were collected from oak trees in Kurdistan, Kermanshah, Ilam and Lorestan provinces whose teleomorph characteristics were consistent with the description of B. mediterranae. Of these, the stromata of the seven specimens were rotting and their ascospores were not able to germinate. Of the remaining specimens, 24 single spore isolates were obtained. Specific primers (MED1 and MED2) were able to produce fragments of 377 bp in a polymerase chain reaction mixture containing DNA of identified isolates, and their morphological identification was confirmed.

Determination of the host range

Signs of charcoal disease (black stroma containing perithecia of B. mediterranea) were observed only on Q. brantii and Q. infectoria. While in some sampled regions, the charcoal disease sign on oak trees was prevalent, it was not observed on other forest tree species including Crataegus spp., Amygdalus spp., Acer monspessulanum, Cerasus microcarpa, Pistacia spp., Rosa spp., Ulmus sp. and Pyrus spp.

In the inner bark tissue of the excised branches of Amygdalus sp., Acer monspessulanum, Cerasus microcarpa, Pistacia sp. and Rosa sp., B. mediterranea created extensive brown necrosis with a length of about 8-15 centimeters, but the disease did not occur in Crataegus sp. and Ulmus sp. from the margins between healthy and necrotic tissue, the fungus was reisolated and according to this method, its pathogenicity was proved mentioned above species.

Biscogniauxia mediterranea created brown necrosis in the inner bark tissue of Q. brantii, Q. infectoria and Q. libani seedlings only in bark-wounded treatments. Extension of necrosis was significantly different in the treatments contained disease symptoms compared to control. Although Q. infectoria was more sensitive to the disease compared to Q. libani, it was not statistically significant (α = 0.05).

Germination of conidia and ascospores

The average of ascospores germination of three isolates of B. mediterranea, collected from different locations was varied from 37.2 to 95.8%, (Table 1). The growth of the mycelium produced by the ascospores continued and resulted in the production of colonies. Only 1.1 percent of conidia (0-1.1 percent) germinated (Table 2), however, the growth of the germ tube stopped quickly and its length did not exceed that of the conidium. The small segments of the mycelium which were approximately the same size as conidia and passed through the filter, germinated and formed colonies.

Table 1. The average of ascospore germination percentage of three isolates of Biscogniauxia. mediterranea on three different culture media.

Table 2. The average of conidium germination percentage of three isolates of Biscogniauxia. mediterranea on three different culture media.

Pathogenicity ability of ascospore andconidium

Ascospore inoculation of Q. brantiis seedlings caused fungal infection and invasion of the inner bark tissue in bark-wounded treatments. The pathogenicity of B. mediterranea ascospores was approved by re-isolating the fungus from the margin between diseased and healthy tissue.

It should be noted that ascospores did not produce disease in wound-free treatment. The incidence of necrosis by inoculation of ascospores was significantly higher than that of the control and the necrosis caused by the conidium (α = 0.01).

Of the 10 seedlings that were inoculated with conidium suspension in the wound on stems, only one seedling showed disease symptoms, however, none of the wound-free treatments developed the disease.

Survival of ascospores

Reduction of germination percentage of B. mediterranea ascospores after one year under the condition of placing stroma containing ascospores in the natural state (on the oak tree in forest conditions), similar to natural condition (hanging in the yard of KANRREC) and in the paper bag in the room was 17.3%, 13.5% and 7.4%, respectively, and the difference between initial and secondary germination of two first cases was significant at α=0.05 (Table 3).

Table 3. Mean comparison of initial germination percentage of ascospores with their germination percentage after one year under forest conditions, similar to the forest and in-room condition by Duncan multiple range tests (α = 0.05).

* Difference among means with the same letter is not statically significant at α = 0.05

Development of sexual and asexual stages

Biscogniauxia mediterranea caused brown to dark brown necrosis in the inner bark of Q. brantii trees and gradually extended along the trunk length. Necrosis penetration depth often expanded to the center of the trunk and its width was about twice as the width of the inoculation site (about 5 cm). The fungus was re-isolated five months after inoculation, from 7 to 10 centimeters and two years later, from 40 to 60 centimeters distance from the inoculation site along the length of the trunk. On the opposite side of the inoculated area, the symptoms of the disease were not detected in both inner bark and wood. In all examined trees, no signs and symptoms were seen on the surface of the outer bark under which inner bark had become necrotic, and had a normal appearance. There was a highly significant difference between treatments with a control.

From the inoculation site on the excised branches, the symptom of the disease began as brown to dark brown necrosis in the inner bark and wood and it extended along the length, width (approximately half of the cross-section periphery) and depth (up to center) of excised branches. Some white colonies were usually formed at two ends of the excised branches but at this stage, the bark surface had no signs and symptoms of the disease. Seven months after inoculation, white colonies containing conidia of B. mediterreanea emerged around the wounded site of inoculation. Under the outer bark, a thin layer (about 200 µm) of black stroma without teleomorph emerged, which gradually extended and an olivaceous brown mass of conidiophores and conidia formed on it, in this stage some parts of outer bark was peeling from lower tissue (Fig 2 a). The growth of the germ tube of these conidia (Fig 2 b) on PDA and WA media stopped quickly like those from a pure culture. One year after inoculation, the excised branches were investigated on which no teleomorph appeared. Then they were kept in the paper bag in room conditions. Two months later, we examined them again and found that a two-part stroma containing prethicia of B. mediterranea had been formed on one of the excised branches of Q. infectoria. (Fig 3 a-b).

From the inoculation site, the color of the inner bark and surface of wood tissue began to turn brownish to dark brown and extended on both sides of the inoculation site along the seedling stem. The progression of the necrosis within the stem breadth was about six millimeters (nearly twice the width of the wound site). Conidiophores, conidia and stroma of B. mediterranea have not been formed on seedlings since inoculation time (for 34 months).