INTRODUCTION

Black poplar (Populus nigra L.) is found in large areas of Europe and Asia and is typically a pioneer species found in riverine areas [1]. In Iran it is of both social and economic interest. It is planted as windbreaks for the protection of agricultural land, as ornamentals and for wood production [2].

In 2017, a bark canker with copious brownish exudations was observed on mature black popular trees in Hamadan province, Iran. The exudates were observed on the trunk, crotch or pruning scars in spring, summer and early autumn. When the bark was removed, reddish lesions within the woody tissue was observed (Fig. 1).

Fig. 1.
figure 1

Symptoms of bark canker on Poplar mature trees caused by Brenneria sp. in Iran. (a–d) The symptoms of bark canker on poplar trees in the field. (e) The symptoms caused by Brenneria strain YMAPO10 in the greenhouse on excised stems 4 days after inoculation. (f) Control water-cultured excised stems inoculated with SDW. (g) The symptoms caused by Brenneria strain YMAPO10 in greenhouse on excised stems after 3 days after inoculation. (h) Control excised stems inoculated with SDW. (i) The symptoms caused by Brenneria strain YMAPO10 in greenhouse on shoots of poplar seedling after 4 days after inoculation. (j) The symptoms caused by Brenneria strain YMAPO10 in the greenhouse and on the trunk of poplar mature trees in the field 18 days after inoculation. (k) Control mature trees inoculated with SDW only in the field. (l) Control mature trees inoculated with B. goodwinii strain ICMP21594 in the field.

Full size image

Diseases, especially bark cankers, reduce the quantity and quality of poplar wood. Plant pathogenic bacteria are one of the most important causes of bark cankers on poplar trees. Bark canker symptoms have been observed on Populus monilifera, P. nigra, and P. monilifera in the Netherlands, Belgium, France, England, and Italy [3]. Similar symptoms were reported in Spain on Populus × euramericana and the causal agent described as Brenneria sp. [4]. Recently, the causal agent of this disease on this host was reported to be Lonsdalea populi in Spain [5], Hungary [6] and China [7]. Bacterial bark canker caused by Xanthomonas populi has previously been reported on poplar trees in Europe and the USA [8]. To date, none of these bacteria have been reported from Iran.

The genus Brenneria was first described by Hauben et al. [9]. Almost all species of the genus are plant pathogens and cause blight, canker, wilt, and necrosis of deciduous trees, including oak, willow, and walnut [9–12]. There are species in this genus which can cause bark cankers with stem bleeding. These include Brenneria nigrifluens and B. rubrifaciens causing bark canker and deep bark canker of walnut, respectively [11, 13], B. alni causing bark canker of alder [14], Brenneria salicis causing watermark disease (WMD) on willow [15–17], and Brenneria sp. causing bark canker on poplar trees [4]. Brenneria goodwinii and B. roseae have been isolated from symptomatic oak and hornbeam tissues with stem bleeding and have found to be associated with acute oak decline (AOD) [18–20]. B. goodwinii was, however, recently reported to be one of the key causal agents of tissue necrosis characterizing AOD in Britain [12].

Considering that nothing is known about the causative bacteria of bark canker on poplar trees in Iran, the aim of this study was to determine the causal agent of the bark canker disease of Populus nigra in Iran.

MATERIALS AND METHODS

Symptoms, Sampling and Bacterial Isolation

In 2017 and 2018, samples were collected from 16 symptomatic P. nigra trees in Hamadan province of Iran (Table 1). In 2017, a bleeding canker with abundant brownish exudates was observed on poplar trees in Iran. Exudation occurred particularly on the tree trunk from pruning cuts, bark cracks, and branch crotches. Bleed points occurred at 0.5 to 2.5 m above ground level. Active bleeding canker was usually seen from spring to early autumn. After removing the outer bark behind the flow of exudates on the trunk, a necrosis tissue was observed on the inner bark and sapwood (Figs. 1a–1d). Other symptoms on affected trees were chlorosis, defoliation, progressive loss of vigor, but the death of the trees was not observed. Each sample consisted of a single panel of outer bark, inner bark and sapwood tissue. Also, two non-symptomatic P. nigra trees were sampled. The collected samples were placed in paper bags and transferred to Plant Pathology Laboratory of Bu Ali Sina University in an ice-box.

Table 1.   Isolation details from symptomatic poplar trees (Populus nigra) sampled in this study
Full size table

For each sample, the outer bark was removed due to the presence a large number of saprophytic microorganisms and the samples were then washed under tap water. For isolation of bacteria from diseased trees, the pieces of tissue (5 × 30 mm) were cut from the interface between diseased and healthy phloem and sapwood tissue using a sterile scalpel. For isolation of bacteria from healthy trees, the pieces of tissue (5 × 30 mm) were also cut from apparently healthy tissue.

Both diseased and healthy pieces were surface-disinfected individually by dipping in 70% ethanol for 2 min, and rinsed in three changes (3 min each) of sterile distilled water (SDW). Each piece was placed in a sterile petri-dish and further fragmented manually in 6 ml SDW using a sterile scalpel. The resulting suspension was transferred to a test tube and shaken for about 30 min at 25°C. A loop-full of the resulting suspension was streaked on nutrient agar (nutrient agar; Merck, Darmstadt, Germany) and incubated in the dark at 28°C for 7 days. Single dominant colonies were selected and sub-cultured on NA again for purification [20]. The isolates were kept at 4°C for the short term and in 30% (v/v) glycerol at –80°C for long term storage [21].

Phenotypic Characteristics of Isolates Obtained from Poplar Trees

Phenotypic features of the bacterial strains were characterized based on standard bacteriological methods. Sensitivity to 3% KOH [22], oxidative/fermentative test [23] and the oxidase reaction [24] was determined for each isolate. Formation of levan, catalase production, hydrolysis of gelatin, aesculin and starch, production arginine dihydrolase, H2S production from cysteine, nitrate reduction, urease production, and production of fluorescent pigment on King’s B medium were performed according to the methods described by Schaad et al. [21]. In addition, assimilation of carbon sources was tested using the basal medium of Ayers et al. [25] supplemented with 0.1–0.3% of the carbon compounds.

Genotypic Characteristics of Bacterial Isolates from Symptomatic Trees DNA Preparation

Three bacterial isolates obtained from symptomatic trees (YMAP5, YMAP10, and YMAP25) were used for genotypic tests. Bacteria were cultured on NA medium and kept at 28°C for 24 h. Bacterial suspensions (108 CFU/ml) were prepared in SDW and lysed by the addition of 1:10 volume of 10% KOH. They were heated at 100°C for 5 min and then placed at 4°C until cooled. Lysates were centrifuged at 10 000 × g for 10 min, and the supernatants were used directly for PCR or stored at –20°C until used [26]. The quality of DNA was determined by electrophoresis on agarose gel [27].

BOX-PCR

BOX-PCR fingerprints were produced using BOX-A1R primer (5'-CTA CGGCAAGGCGACGCTGACG-3') [28]. The total volume of the reaction was 25 μL which contained 2.5 μL 10X PCR buffer (100 mM Tris-HCl, 500 mM KCl, pH 8.4), 2.5 mM MgCl2, 0.4 mM dNTPs, 0.2 μM BOXA1R primer, 1.25 U Taq DNA polymerase (CinnaGen, Iran) and 2 μL DNA. Amplification was performed in a Techne TC-512 thermal cycler with initial denaturation for 7 min at 94°C, followed by 35 amplification cycles consisting denaturation (94°C, 1 min), annealing (59°C, 1.5 min) and extension (72°C, 2 min), with a final extension at 72°C for 10 min [29]. The PCR products were separated by electrophoresis in a horizontal 1.5% agarose gel in 1X TBE buffer (100 mM Tris, 500 mM boric acid and 1 mM EDTA), at 80 V for 120 min and stained with ethidium bromide. A 1 Kb GeneRuler™ ladder (GeneRuler, Fermentas, Vilnius, Lithuania) was used as a base pair size marker. The banding patterns of isolates were visualized and photographed using a gel documentation system DigiDoc H101 (Iran).

16S rRNA and Multilocus Sequence Analysis (MLSA)

The 16S rRNA gene of isolates was amplified using primers fD1 and rP2 [30]. PCR mix and amplification were performed according to Moradi-Amirabad and Khodakaramian [26] using XP thermal cycler TC-XP-G (Bioer, Germany). DNA gyrase B sub-unit (gyrB), initiation translation factor 2 (infB), and RNA polymerase β subunit (rpoB) genes was amplified using the primers and conditions as described by Brady et al. [31]. Partial amplification of the glyceraldehyde-3-phosphate dehydrogenase A (gapA) gene was conducted as described by Ma et al. [32].

PCR products were electrophoresed in 1.2% agarose gel, stained with ethidium bromide and observed with gel documentation system DigiDoc H101 (Iran).The amplified fragments were sequenced by the dideoxy termination method by BIONEER Co. (South Korea) on both strands employing the primers used for amplification. The sequence data were trimmed with BioEdit version 7.2.5 program [33] and compared with those from GenBank with the BLASTN program [34]. Clustal W was used for creating the sequence alignments [35]. Neighbor-joining tree was constructed employing MEGA X [36]. Bootstrap analysis was performed to test the stability on randomly chosen sets of positions with 1000 replicates [37]. The sequence data were deposited in GenBank and accession numbers are presented in Table 2. The representative isolates were deposited a culture collection in the Plant Pathology Laboratory of Bu Ali Sina University.

Table 2.   The sequences accession numbers of poplar strains of Brenneria obtained from this study
Full size table

Pathogenicity Tests

For pathogenicity tests, the representative isolates (YMAPO1, YMAPO5, YMAPO7, YMAPO10, YMAPO14, and YMAPO25) were grown on NA for 48 h at 28°C (Table 1). Cells were suspended in SDW and their density adjusted to 107 CFU/mL. Four pathogenicity trials were undertaken. In the first trial, pathogenicity tests were performed on excised stems (2 cm in diameter and 30 cm in length) of P. nigra trees in a greenhouse (28°C incubator). The upper end of stems were wrapped with plastic film to prevent drying and lower cut surface of them were placed in saucers containing water. The stems were surface sterilized with 70% ethanol and cross-shaped wounds (5 mm long) were made in the bark with a sterile knife. 100 μL bacterial suspension was introduced into the wound with a sterile dropper. The inoculated segments were covered with polyethylene bags to maintain the humidity and were kept in darkness at 28°C [7]. In the second trial, the wound was made using a 10 mm cork-borer on excised stems of P. nigra trees. Half a loop of 24 h NA culture of each representative isolate was inserted into the wound and the stems placed in a greenhouse (28°C). The outer bark plug was pulled back into the inoculation point and wrapped with parafilm and sealed with duct tape. Inoculated stems were incubated at 25°C in the growth chamber with a 12 h photoperiod [12]. The third pathogenicity trial was conducted on shoots of poplar seedlings by injecting 50 μL bacterial suspension (107 CFU/mL) as described by Schaad et al. [21].

The forth pathogenicity trial was conducted in the field on mature trees of P. nigra at Bu-Ali Sina University, Hamadan where the disease first observed. Two cm long wounds were made in the trunk bark using a sterile knife. Two hundred ul of the bacterial suspension (107 CFU/mL) was introduced into the wound. The wounds were wrapped with parafilm and unwrapped after three days [7]. Also, the first, second, and third pathogenicity trials were conducted to assess the potential pathogenicity of isolates obtained from healthy trees. Inoculated stems were assessed for symptom daily. In all trials, control excised stems and plants were inoculated with SDW and Brenneria goodwinii ICMP21594. Each trial was conducted for each isolate in three repetitions. At the end of the trials, re-isolations were performed at the point of inoculation and/or at the interface between healthy and diseased tissue.

RESULTS

Sampling and Bacterial Isolation

From 16 symptomatic trees, a total of 32 bacterial strains were isolated from difference region of Hamadan province in Iran (Table 1). Only, Brenneria-like colony appeared on the culture medium, after 20 days. All of colonies were 1–2 mm in diameter on NA after 4 days at 20°C, pale cream, circular, convex, smooth with entire margins and shining (Fig. 2a). Colonies of all isolates had a metalic green sheen on EMB agar (Fig. 2b).

Fig. 2.
figure 2

Colonies of the Brenneria sp. from Iran (a) NA medium. (b) Eosin methylene-blue culture (EMB) medium showed metallic green colony colour.

Full size image

Phenotypic Characteristics of Bacterial Isolates from Symptomatic Trees

Strains were Gram-negative, and facultatively anaerobic. They were positive for catalase reaction, and hydrolysis of aesculin, and H2S production, but negative for tobacco hypersensitivity reaction, oxidase, urease production, and lysine decarboxylase, fluorescent production on King B, starch hydrolysis, and indole production. All isolates produced acid from D-glucose, fructose, L-rhamnose, cellobiose, sucrose, maltose and D-mannitol, but they didn’t produce acid from L-arabinose, lactose, D-raffinose, D-xylose, D-galactose, trehalose, adonitol, D-sorbitol, and dulcitol. Other phenotypic characteristics are summarized in Table 3.

Table 3.   Phenotypic characteristics distinguishing strains of Brenneria sp. isolated from symptomatic poplar trees in Iran and in comparison with described taxa of the genera Brenneria and Lonsdalea
Full size table

Phenotypic Characteristics of Bacterial Isolates from Healthy Trees

Also, two healthy poplar trees were sampled and 17 bacterial isolates were obtained. They were mostly Gram-positive bacteria (eight strains), Pseudomonas spp. (five strains), and members of the Enterobacteriales (three strains). None of the isolates obtained from healthy tree were identified as Brenneria sp. (Table 4).

Table 4.   Isolates obtained from non-symptomatic Populus nigra trees sampled in this study
Full size table

GENOTYPIC CHARACTERISTICS OF BACTERIAL ISOLATES FROM SYMPTOMATIC TREES

BOX-PCR

The genomic fingerprinting pattern of symptomatic poplar isolates were generated using the BOXA1R primers. The amplified bands in BOX-PCR ranged from 350 to 3000 bp (Fig. 3). The genomic fingerprint pattern was re-generated upon repeat of the procedure. Poplar isolates were grouped into one group based on binding pattern in BOX-PCR. All of the poplar isolates obtained in this study showed same banding pattern based on BOX-PCR fingerprint and were placed in one group (Fig. 3).

Fig. 3.
figure 3

BOX-PCR patterns of Brenneria causal agent of bark canker on poplar trees in Iran.

Full size image

16S rRNA Sequences Analysis

The three isolates (YMAPO5, YMAPO10, and YMAPO25) from symptomatic poplar trees demonstrate 100% 16S rRNA gene sequence pairwise similarity to each other and Brenneria strain NCPPB 4297, and 99.48 and 99.93% to NCPPB4296 and NCPPB4299, and 99.19 % to the type strains of B. salicis LMG 2698. The poplar isolates and NCPPB4296, NCPPB 4297 and NCPPB 4299 formed a single cluster with 91% bootstrap support on a separate branch in the 16S rRNA phylogenetic tree, and demonstrated a close phylogenetic relationship to B. salicis strain LMG 2698 (the type species of Brenneria) with 100% bootstrap in neighbour-joining (Fig. 4).

Fig. 4.
figure 4

The optimal neighbour-joining phylogenetic tree based on 16S rRNA sequences of poplar isolates (YMAPO5, YMAPO10, and YMAPO25) and the reference strains of Brenneria, Lonsdalea Dickeya, and Pectobacterium species obtained from GenBank (sum of branch length = 0.35196674). The GenBank accession numbers are given in parentheses behind the species names and strain numbers. Numbers at the nodes are the bootstrap values obtained for 1000 replicates. The sequence alignment consisted of 1263 characters. Bar, 0.005 substitution per nucleotide position. The tree is rooted to Erwinia amylovora LMG 30165T.

Full size image

MLSA

The nucleotide sequences of housekeeping genes were obtained for the representative isolates. PCR amplification and sequencing of the housekeeping gene of the strains YMAPO5, YMAPO10, and YMAPO25 yielded partial sequences. Pairwise alignment revealed 100% sequence similarity between strains for all three housekeeping genes (gyrB, infB, and rpoB). They were 97.99, 99.35, and 97.96% identical with those of reference strain B. salicis LMG 2698 for gyrB, infB, and rpoB respectively. To compare poplar isolates with Brenneria strain NCPPB 4296, NCPPB 4297, and NCPPB 4299, no sequence of gyrB, infB and rpoB genes was available in GenBank. InMLSA based on the nucleotide sequences of the three housekeeping genes (gyrB, infB, and rpoB), the poplar isolates formed a separate branch from B. salicis strain LMG 2698 with 100% bootstrap in neighbour-joining (Fig. 5).

Fig. 5.
figure 5

The optimal neighbour-joining phylogenetic tree based on based on concatenated gyrB, infB and rpoB sequences of poplar isolates (YMAPO5, YMAPO10, and YMAPO25) and the reference isolates of Brenneria, Lonsdalea Dickeya, and Pectobacterium species obtained from GenBank (sum of branch length = 1.61974421). The GenBank accession numbers are given in parentheses behind the species names and strain numbers. Numbers at the nodes are the bootstrap values obtained for 1000 replicates. The sequence alignment consisted of 1868 characters. Bar, 0.02 substitution per nucleotide position. The tree is rooted to Erwinia amylovora LMG 2024.

Full size image

gapA Sequences Analysis

The representative isolates (YMAPO5, YMAPO10, and YMAPO25) demonstrated 100% similarity to each other in pairwise comparisons of the gapA gene sequences and 100 and 96.32% to NCPPB 4299, and B. salicis LMG 2698, respectively. In a phylogenetic tree based on the nucleotide sequences of gapA gene, poplar isolates clustered with those strains of Brenneria strain NCPPB 4296, NCPPB 4297, and NCPPB 4299 Separated from other species with 100% bootstrap in neighbour-joining (Fig. 6).

Fig. 6.
figure 6

The optimal neighbour-joining phylogenetic tree based on based on gapA sequences of poplar isolates (YMAPO5, YMAPO10, and YMAPO25) and the reference isolates of Brenneria, Lonsdalea Dickeya, and Pectobacterium species obtained from GenBank (sum of branch length = 0.75672445). The GenBank accession numbers are given in parentheses behind the species names and strain numbers. Numbers at the nodes are the bootstrap values obtained for 1000 replicates. The sequence alignment consisted of 397 characters. Bar, 0.01 substitution per nucleotide position. The tree is rooted to Erwinia amylovora LMG 2024.

Full size image

Pathogenicity Tests

Pathogenicity test 1: In the five days test period, all isolates obtained from symptomatic trees caused symptoms of bleeding canker on the_inoculated 30‑cm-long stems. The outer and inner bark were macerated at the site of inoculation 2 days after inoculation, and a transparent yellowish liquid was exudated 5 days after inoculation (Fig. 1e).

Pathogenicity test 2: Fig. 1g shows the symptoms caused by isolates inoculated into excised stems. In the 4 days test period, the bark was macerated and after removing the bark, sapwood was necrotic at the site of inoculation; also. A transparent liquid flowed from the inoculation site. Pathogenicity test 3: inoculated isolates of symptomatic trees caused necrosis on shoots of poplar seedlings in greenhouse after 5 days inoculation (Fig. 1i).

In the Pathogenicity tests 1, 2, and 3, the inoculated stems with the bacterial isolates obtained from healthy trees showed no symptoms after 14 days. Therefore, they were not used in the pathogenicity test 4.

Pathogenicity test 4: typical field symptoms were reproduced including bleeding cankers with brownish exudates by symptomatic poplar isolates inoculated into mature P. nigra trees one month after inoculation (Fig. 1j). Re-isolations from the inoculated plants were done, and the isolated strains were confirmed to be Brenneria based on morphological characteristics and biochemical tests. No bacteria were re-isolated from the control plants. Inoculation with B. goodwinii ICMP21594 (Fig. and the control (SDW) did not cause any symptoms.

DISCUSSION

During 2017 and 2018, sampling was done from poplar trees in different poplar plantations of Hamadan province, Iran. The disease was observed only on mature trees. The external symptoms were bleeding cankers with abundant dark brownish and slimy exudations which stained the trunk bark. The symptoms were similar to those caused by Brenneria spp. on the trunk of walnut, alder, poplar, and oaks trees [4, 11, 12, 14, 20]. When the outer bark was removedfrom the canker, the inner bark and sapwood showed a reddish color progressing to necrotic lesions. In Hungary [6] and China [7], the bark is vertically cracked in symptomatic trees, and a sticky, brown-colored fluid exudates from the canker. In Spain, two types of symptoms were observed on poplar trees: [1] the bark canker is characterized by cankers with copious brownish exudations staining the trunk bark and reddish inner lesions progressing to necrosis that was caused by Brenneria sp. strain NCPPB 4296, NCPPB 4297, and NCPPB 4299 [4]; (2) the bark of symptomatic trees was vertically cracked and copious white frothy fluid and creamy slime was observed caused by Lonsdalea populi [5]. The symptoms observed in Iran were similar to bark canker of poplar trees caused by Brenneria sp. in Spain previously described by Biosca et al. [4]. None of the bacterial strains isolated from healthy trees were Brenneria spp. which suggests that they do not occur as endophytes. In addition, no fungi were isolated from diseased tissue. Thus the Brenneria sp. identified in this study from diseased trees is the causal agents of bleeding cankers on poplar trees in Iran.

Uniform colonies were always isolated from all the symptomatic trees (Fig. 2a). The repeated isolation of Brenneria sp. from necrotic tissues in diseased trees confirmed the etiology of this symptoms in poplar plantations in Iran. All of the poplar isolates obtained in this study were placed in one group based on phenotypic characteristics (Table 3) and BOX-PCR (Fig. 3). Several molecular methods are used to rapidly identify and classify pathogenic bacteria including BOX-PCR. There are three families of repetitive sequences for distinguish bacterial strains below the level of species including repetitive exteragenic palindromic (REP) sequence, enterobacterial repetitive intergenic consensus (ERIC) sequence and BOX element [28]. Previous studies have shown a high correlation between BOX-PCR fingerprinting and DNA-DNA homology data [38]. The pathogenicity tests on excised stems and trunk of poplar trees reproduced the bark cankers observed in naturally infected poplar. The inoculations on shoots of poplar trees caused necrosis on them which was consistent with the Biosca et al. [4].

The poplar isolates from Iran were identified in this study as a previously undescribed Brenneria sp. Based on phenotypic and molecular characteristics poplar isolates were clustered with Brenneria sp. strains NCPPB 4296, 4297, and 4299 and are most closely related to Brenneria salicis based on 16S rRNA gene analysis (Fig. 4). They formed a separate cluster close to B. salicis based on MLSA analysis that this confirmed their undescribed status (Fig. 5). Nevertheless, the poplar isolates were differentiated phenotypically from their closest phylogenetic neighbours. MLSA is a useful tool for the investigation of bacterial taxonomic relationships [31, 32, 39] and an appropriate technique for species delineation and for distinguishing relationships at the intraspecific level. Because housekeeping genes are less conserved than rRNA genes, the high degree of sequence divergence of housekeeping genes is superior for identification purposes. MLSA was recently used with great success to resolve the phylogenetic position Brenneria species based on partial sequencing of gyrB, rpoB, infB and atpD genes [39]. The phylogenetic trees revealed that the strains isolated from bark canker on P. nigra in Iran (YMAPO5, YMAPO10, ad YMAPO25) and Populus = euramericana in Spain (NCPPB 4296, 4297, and 4299) clustered together and close to the type strain of B. salicis LMG2698. The Poplar isolates were different in some phenotypic tests with those references strains of the Brenneria species including Brenneria salicis. Unfortunately, No information has been published on phenotype characteristics of NCPPB 4296, 4297, and 4299. The results of phylogenetic analysis suggest that the bacterial strains obtained from poplar canker in Iran is similar to Brenneria sp. NCPPB 4296, 4297, and 4299 obtained from poplar canker in Spain [7]. Despite the similarities and differences, it cannot be stated that Brenneria sp. strains from Iran are a new species of Brenneria or a subspecies of Brenneria salicis. Whole genome sequencing (WGS) would need to be undertaken in order to formally describe this species.

Populus nigra is one of the most widely grown poplar cultivars in Iran. The bacterial canker disease observed to occur in Hamadan province of Iran. The occurrence of the bacterial canker disease on P. nigra emerged as a barrier for expanding cultivation of the poplar trees. In this study a previously undescribed Brenneria sp. was identified as causal agent bark canker of P. nigra in Iran and to our knowledge, this is the first report of the occurrence of this bacterial species in Iran, and on this host species globally.