Introduction

Brazil has 9.55 million hectares of planted forests for industrial purposes, and approximately 78% of this area is occupied by Eucalyptus plantations (Ibá 2021). The country leads the global ranking of forest productivity with an average production of 35.7 m³/ha/year, a value equivalent to twice that of countries in the northern hemisphere. The main destinations for Brazilian planted forest woods are the pulp and paper industry, steel and charcoal, panels, and laminate flooring (Ibá 2021).

To optimize the production of clonal seedlings, new methodologies of vegetative propagation were developed and popularized in Eucalyptus, such as mini-cutting and micro-cutting techniques (Xavier and Silva 2009). These techniques improved the commercialization of clonal genotypes that were difficult to root (Ferreira et al. 2004), allied to other advantages of vegetative propagation in clonal mini-gardens such as lower costs of implantation, transportation, and maintenance, as well as greater ease of harvest, irrigation control, and rooting speed (Mafia et al. 2005).

Notwithstanding the recent advances in forest biotechnology, Eucalyptus cloning carried out in mini-gardens still faces some challenges, such as the losses caused by diseases favoured by the high temperature and humidity of the forest nurseries. Recently, fungal structures resembling those in the Cryphonectriaceae were found associated with several Eucalyptus mini-stumps in clonal mini-gardens in Brazil, for the first time. The symptoms observed were wilt that evolved into partial or complete drying of the canopy and subsequent death of the mini-stumps. The internal symptom was characterized by an upward lesion along the stem, and no fungal structures were observed in fresh samples.

Chrysoporthe Gryzenh. & M.J. Wingf. is a genus of fungi belonging to the family Cryphonectriaceae, order Diaporthales, phylum Ascomycota. The genus encompasses various and important pathogens known for causing typical symptoms of canker in species of the Myrtaceae and Melastomataceae families (Barreto et al. 2006; Gryzenhout et al. 2006; Myburg et al. 2003). Initially, the causal agent of Eucalyptus canker disease was described as Diaporthe cubensis (Bruner 1917), which was later transferred to the genus Cryphonectria, as Cryphonectria cubensis (Bruner) Hodges, due to similar cultural and morphological characteristics with Cryphonectria species (Hodges et al. 1980). Subsequently, based on morphological characters and phylogenetic analyses, isolates of Cr. cubensis were shown to be distinct from Cryphonectria (Gryzenhout et al. 2004; Myburg et al. 2003). As a result, the genus Chrysoporthe was created to accommodate the single species Ch. cubensis (Bruner) Gryzenhout & M.J. Wingf., (Gryzenhout et al. 2004).

In additional to Ch. cubensis, nine other Chrysoporthe species are currently known: Ch. austroafricana, Ch. colombiana, Ch. deuterocubensis, Ch. doradensis, Ch. hodgesiana, Ch. inopina, Ch. puriensis, Ch. syzygiicola, and Ch. zambiensis (Gryzenhout et al. 2004, 2005, 2006; Chungu et al. 2010; van der Merwe et al. 2010; Oliveira et al. 2021; Suzuki et al. 2023). All these species were recorded in association with Eucalyptus, except Ch. colombiana, Ch. hodgesiana, Ch. inopina, Ch. puriensis, and Ch. syzygiicola that were found in the Melastomataceae (Gryzenhout et al. 2004, 2006; Oliveira et al. 2021; Suzuki et al. 2023). It was, however, demonstrated by inoculation that Ch. colombiana, Ch. puriensis, and Ch. syzygiicola are able to cause symptoms on Eucalyptus (Chungu et al. 2010; Oliveira et al. 2021; Suzuki et al. 2023). On the other hand, only Ch. cubensis, Ch. puriensis, and Ch. doradensis have been reported in Brazil (Hodges et al. 1976; Soares et al. 2018; Oliveira et al. 2021). Furthermore, there is strong evidence that Ch. cubensis and Ch. puriensis are native to Brazil, which is supported by the fact that these species occur commonly on native Tibouchina spp. and have a high level of genetic diversity (Oliveira et al. 2022).

Chrysoporthe cubensis is able to infect Eucalyptus at different ages. In young plants, the fungus may cause occasional death due to girdling of stem at the base of the trunk. In older plants, the symptoms vary from lesions with diverse degrees of bark cracking to typical cankers at different heights of the trunk (Ferreira 1989). A typical canker is a deep lesion bordered by calluses, as a consequence of cambium death in an attempt to prevent girdling of the stem (Alfenas et al. 2009). Canker caused by Ch. cubensis significantly impacted the development of Eucalyptus plantations in the tropics and southern hemisphere in the 1970s (Ferreira 1989), which motivated the first studies on the disease and boosted genetic breeding of Eucalyptus to obtain resistant species to the pathogen. Despite the extensive research on breeding, Eucalyptus canker caused by Ch. cubensis is still one of the most harmful diseases in commercial plantations in Brazil (Ferreira and Milani 2004; Gryzenhout et al. 2009).

To date, there are no studies that demonstrate the occurrence of Chrysoporthe spp. associated with Eucalyptus mini-stumps worldwide. Based on this, efforts to disclose the identity of the wilt disease and to evaluate how different clones react to the causal agent are crucial to the definition of control strategies in clonal mini-gardens. Therefore, the aims of the present study were (i) to elucidate the causal agent of the wilt disease on Eucalyptus mini-stumps through morphological/molecular analyses and pathogenicity tests on healthy eucalypt seedlings; (ii) to evaluate the susceptibility of seedlings of different commercial clones of Eucalyptus to the pathogen; and (iii) to evaluate the influence of temperature on the development of the disease.

Materials and methods

Fungal isolates and preservation of the cultures

Symptomatic mini-stumps of five hybrid clones of E. urophylla × E. grandis (CNB 005, CNB 007, CNB 010, CNB 029, and CNB 030) were collected from the clonal mini-garden at a forest company in the state of Minas Gerais, located in the Southeast region of Brazil. Symptoms included wilt, canopy drying, xylem lesion, and death (Fig. 1.A-H). Fungal structures were rarely found on mini-stumps surfaces, only after 72 h of incubation of segments from the bark (Fig. 1.I). All samples were first examined at 20× magnification under a stereomicroscope Motic® SMZ-140 for observation of fungal structures and longitudinal cuts were made with a sterilized scalpel, from the root system to the tip of the stem, in order to find internal symptoms.

Fig. 1
figure 1

Chrysoporthe cubensis on Eucalyptus mini stumps. (A-B) Losses due to wilt caused by Ch. cubensis in a clonal-mini garden; (C) asymptomatic mini-stump; (D) early stage of drying; (E) late stage of drying; (F-G) lesion in the xylem; (H) mini-canker in inoculated plant; (I) Conidiomata exuding conidia on the surface of mini-stump bark

Full size image

For fungal isolation, surface disinfection of symptomatic stem fragments was done by stepwise washing in 70% ethanol for 30 s and 2% sodium hypochlorite solution for two minutes. Fragments were transferred to Potato Dextrose Agar (PDA) and incubated at 28 °C with a 12-hour photoperiod until mycelial growth. Additional isolates obtained out of symptomatic mini-stumps from clonal mini-gardens in the states of São Paulo and Mato Grosso do Sul were provided from the Collection of Laboratório de Patologia Florestal. All pure cultures obtained from hyphal tips on PDA were preserved and deposited in the Octavio de Almeida Drumond Collection (COAD) at Universidade Federal de Viçosa (UFV), Brazil under collection numbers COAD 3400–3401, COAD 3421–3423 and COAD 3478–3486 (Table 1).

Table 1 GenBank accession numbers of DNA sequences of Chrysoporthe and outgroup species used in the phylogenetic analyses
Full size table

DNA extraction and amplification

The fungal isolates were grown on PDA at 28 °C under a 12-hour photoperiod for 7–10 days to obtain sufficient fungal biomass for DNA extraction. Mycelia was scraped off with a sterilized wooden toothpick and transferred to a 2 mL microcentrifuge tube. The extraction was performed by mechanical disruption using stainless steel beads in an L-Beader-3 (Loccus Biotecnologia, SP, Brazil). Total DNA was extracted using the Wizard® Genomic DNA Purification Kit (Promega Corporation, WI, USA) according to the manufacturer’s instructions.

The primer pairs ITS1 and ITS4 (White et al. 1990), Bt1a and Bt1b (Glass and Donaldson 1995), and ACT-512 and ACT-783R (Carbone and Kohn 1999) were used to amplify the nuclear rDNA internal transcribed spacers (ITS1-5.8 S-ITS2 = ITS), β-tubulin (TUB), and actin (ACT) genes, respectively. All amplification reactions were prepared in a final volume of 12.5 µl. PCR was performed with 6 µL of Dream Taq TM PCR Master Mix 2× (MBI Fermentas, Vilnius, Lithuania);5 µmol/l of each forward and reverse primer; 0.5 µL of dimethyl sulfoxide (DMSO, Sigma–Aldrich, St. Louis, MO); 1 mg/µL Bovine Serum Albumin (BSA, Sigma–Aldrich,);30 ng of genomic DNA and 2.5 µL of nuclease-free water. Amplification was performed with an initial denaturing step at 94 °C for 5 min, followed by 38 cycles of denaturation at 94 °C for 30 s, primer annealing at 54 °C for 30 s (for both TUB and ITS), or 60 °C for 30 s (ACT), and extension at 72 °C for 45 s, with an additional final extension at 72 °C for 7 min. The presence of PCR products was confirmed by 0.8% agarose gel electrophoresis. The amplified fragments were purified and sequenced by Macrogen Inc., South Korea.

Data editing and phylogenetic analyses

The sequences were edited and manually checked with the BioEdit software program version 7.2 (Hall 2014). Nucleotide arrangements with ambiguous positions were clarified by analyzing the sequences obtained with forward and reverse primers. Based on a BLAST search to check for similarity, 29 additional reference sequences were obtained from the NCBI GenBank database (Table 1). One dataset was built for each locus using the sequences obtained in this study, as well as the representative for the other known species of Chrysoporthe and Amphilogia gyrosa (outgroup), and then aligned individually using the MAFFT algorithm (Katoh and Standley 2013) implemented in Aliview software version 1.28 (Larsson 2014). The alignment concatenation (ITS + TUB + ACT) was performed in Mesquite version 3.61 (Maddison and Maddison 2018). Substitution models were determined separately for each gene partition using jModelTest 2.1.10 (Darriba et al. 2012), in which they were selected according to the Akaike Information Criterion (AIC).

Bayesian Inference (BI) trees were generated employing the Markov Chain Monte Carlo (MCMC) algorithm. The phylogenetic analyses were performed through CIPRES Science Gateway (Miller et al. 2010) using MrBayes 3.2.6 on the XSEDE tool (Ronquist and Huelsenbeck 2003). Four MCMC chains were run simultaneously from random trees for ten million generations and sampled every 1,000 generations. The first 2,500 trees were discarded as the burn-in phase of each analysis. The posterior probabilities were determined from the remaining trees. Additionally, Maximum Likelihood (ML) analyses were generated for each separate gene and multi-locus alignment using RAxML-HPC ver. 8.2.12 (Stamatakis 2014). The chain robustness was assessed through the bootstrap re-sampling strategy with 1,000 bootstrap test replicates. The trees obtained from single genes and multi-locus alignments were compared along with their performance in species recognition. The resulting trees were visualized in FigTree (Rambaut 2009) and exported to Inkscape v. 0.91 (www.inkscape.org) for editing of the layout.

Pathogenicity test

The pathogenicity of the five isolates from the state of Minas Gerais (COAD 3400–3423) was assessed on 6-month-old seedlings of six commercial hybrid clones of E. urophylla × E. grandis (CNB 005, CNB 007, CNB 010, CNB 029, CNB 030, and CNB 032). At the base of the seedling stem, a 1 cm wound was made with a scalpel to remove the bark and expose the xylem. A mycelium plug of 5 mm diameter from a 7-day-old culture was placed facing the wound to allow the contact between mycelium and the xylem. To avoid desiccation, the wound was covered with plastic film for 30 days, according to the method described by Ferreira and Milani (2004). Control treatment consisted of a sterilized PDA plug.

For each combination of Eucalyptus clone and isolate (including control), 10 plants were inoculated, resulting in a total of 360 plants divided into 36 treatments of clone × isolate combinations. Inoculated plants were kept under greenhouse conditions for 70 days. All seedlings were evaluated by measuring the lesion length in the xylem. Koch’s postulates were performed by re-isolating the fungi from the inoculated areas, and the cultures were compared to the original ones. The experiment was repeated once.

Effect of temperature on pathogenicity of Ch. cubensis on Eucalyptus clones

The effect of temperature was tested on 6-month-old seedlings of two commercial hybrid clones of E. urophylla × E. grandis (CNB 010 and CNB 032), which were inoculated with five isolates: COAD 3400–3401 and COAD 3421–3423. Inoculations were performed similar to those of the pathogenicity tests. For each combination of Eucalyptus clone and isolate (including control), 10 plants were inoculated, resulting in a total of 120 plants divided into 12 treatments of clone × isolate. Subsequent to the inoculation, half of the plants of each treatment were transferred to a growth chamber at 19 °C and the other half to 28 °C to compare the effect of temperature on disease development. At the end of 70 days, all seedlings were evaluated by measuring the lesion length in the xylem and the treatments were compared to each other. Koch’s postulates were fulfilled by re-isolating the fungi from the inoculated areas, and the cultures were compared to the original ones. The experiment was not repeated.

Experimental design and data analysis

Tests for pathogenicity and the influence of temperature on disease development were carried out in a completely randomized design, with 10 biological replicates. The effects of the main factors for the pathogenicity test data (clones) and temperature tests (clones and temperature, and the interaction between them) were evaluated in a linear mixed modeling framework. Isolates and replicates were treated as random effects in our model. The model was fitted with the lmer function of the package lmer4 (Bates et al. 2015) of the software R version 4.2.2 (R Core Team 2022). Data were subjected to a type III analysis of variance with the Anova function on the car package (Fox and Weisberg 2019). The emmeans package was used to obtain the estimated marginal means and respective 95% confidence intervals (Lenth 2021). The function cld from multcomp package was used for multiple comparisons of treatment means via the Tukey test at 5% significance (Hothorn et al. 2008).

Results

Fungal isolates and preservation of the cultures

Fourteen isolates resembling species of Chrysoporthe were obtained from symptomatic mini-stumps of Eucalyptus in three states of Brazil, of which five were obtained in this study, one from each Eucalyptus clone, and nine from the Collection of Laboratório de Patologia Florestal (Table 1). Symptoms included wilt, canopy drying, xylem lesion, and death (Fig. 1.A-H). Fungal structures were rarely found on mini-stumps surfaces, only after 72 h of incubation of segments from the bark (Fig. 1.I).

Phylogenetic analyses

Phylogenetic analyses were performed on single-locus and multi-loci datasets of ACT, TUB and ITS. Single-locus trees (Suppl. material 1) showed low topological divergence, congruence for species delimitation, and high support value for the Ch. cubensis clade. The single-locus tree based on TUB sequences was the best in delimitating all the currently accepted species of Chrysoporthe. The ITS region alone did not accurately delineate Ch. cubensis, Ch. austroafricana, Ch. syzygiicola, Ch. deuterocubensis, Ch. doradensis, Ch. zambiensis. ACT failed only in separating Ch. deuterocubensis from Ch. inopina. The concatenated dataset consisted of 27 sequences of known species of Chrysoporthe, 14 isolates included in this study and two sequences of Amphilogia gyrosa as the outgroup. The final alignment was 1370 in length (457, 273, 640 for TUB, ACT, and ITS, respectively) of which 1210 sites were conserved, 126 variable, 63 parsimony informative, and 62 singletons. The best nucleotide substitution models for the Bayesian Inference were HKY for ACT and HKY + I for TUB and ITS, according to the Akaike Information Criterion (AIC). According to the multi-locus trees obtained through BI and ML methods, all the isolates collected in this study grouped with the ex-type strain of Ch. cubensis (Fig. 2).

Fig. 2
figure 2

Multilocus phylogenetic tree based on Bayesian Inference (BI) using the alignment of combined sequences (TUB + ACT + ITS) of Chrysoporthe species. Bayesian posterior probabilities and bootstrap values for BI and Maximum Likelihood higher than 0.5/80 are shown respectively at the nodes (BI/ML). Thickened branches denote Bayesian posterior probabilities higher than 0.99 and Maximum Likelihood bootstrap support higher than 90%. The tree was rooted with Amphilogia gyrosa CMW 10,469 and YMJ91123101. Ex-type strains are emphasized with an asterisk (*). The isolates obtained in this study are in bold

Full size image

Pathogenicity test

All isolates of Ch. cubensis were pathogenic to Eucalyptus regardless of the clone, although there was a significant variation in lesion length for each isolate (P < 0.05) (Fig. 3A). There was also a significant difference in lesion length among the Eucalyptus clones, regardless of the isolate (P < 0.05) (Fig. 3B). Comparing all Eucalyptus clones, CNB 007 was the most susceptible, whereas CNB 005 and CNB 030 were the least susceptible. CNB 010, CNB 029, and CNB 032 were moderately susceptible (Fig. 3B).

Fig. 3
figure 3

 A, Distribution of the length of lesion caused by Chrysoporthe cubensis isolates on seedlings of Eucalyptus urophylla × Eucalyptus grandis clones under greenhouse conditions. B, Estimates of the marginal means and respective 95% confidence intervals by a multilevel model fitted to lesion length data

Full size image

In all plants, a lesion developed in the xylem and the fungus was recovered from the inoculated tissue (Fig. 1.F-G). Most inoculated plants did not present any external symptoms or fungal structures, however, the formation of mini-cankers was observed in few plants. The fungal colonies recovered from the lesions exhibited white and fluffy growth when younger and orange when older, which is the same pattern of the colonies used for inoculation, fulfilling Koch’s postulates. No symptoms were observed in the control plants.

Effect of temperature on pathogenicity of ch. Cubensis on Eucalyptus clones

The development of the disease occurred on both clones regardless of the temperature and isolates. There was no significant interaction between Eucalyptus clones and temperature (P = 0.831). However, significant differences were observed for the effects of the factors clone (P = 0.02) and temperature (P = 0.005) individually. Regardless of the temperature, the clone CNB 010 was more susceptible to the pathogen (Fig. 4). Overall, a temperature of 28 °C favored the progress of the disease more so than did 19 °C (Fig. 4).

Fig. 4
figure 4

Distribution of the length of lesions caused by Chrysoporthe cubensis isolates on seedlings of two clones (CNB 010 and CNB 032) of Eucalyptus urophylla × Eucalyptus grandis incubated under 19 and 28°C

Full size image

Discussion

This study reports Ch. cubensis as the causal agent of wilt on Eucalyptus mini-stumps. It is the first record of Chrysoporthe species causing wilt disease on Eucalyptus mini-stump worldwide.

A previous report of Ch. cubensis on Eucalyptus mini-stumps exists, but is associated with death by stem girdling. It was also based exclusively on morphological identification and Koch’s postulates were not fulfilled (Alfenas et al. 2009).

The most common symptomatology presented by trees and shrubs infected by Chrysoporthe spp. are stem canker, girdling of the stem, cracking of the bark, dying branches and presence of fruiting bodies typical (Chungu et al. 2010; Ferreira et al. 1989; Oliveira et al. 2021). Therefore, Ch. cubensis related to wilt on Eucalyptus mini-stumps is a new symptom amid others considered a hallmark for that pathogen. Pathogen structures (conidiomata) have rarely been found associated with the bark.

Phylogenetic analyses of the ACT gene pointed to the presence of two distinct, but closely related, groups of Ch. cubensis isolates. The first one, characterized by COAD 3400–3401 and COAD 3421–3423, represented the isolates obtained from samples collected at the clonal mini-garden in the state of Minas Gerais. The second group, within which the first group is nested, contains the isolates COAD3478-3486, representing the isolates collected in two different locations in the states of São Paulo and Mato Grosso do Sul. TUB and ITS could, however, not differentiate these two clades of Ch. cubensis. Although it may seem interesting, a recent study on the structure and genetic variability of Ch. cubensis concluded that it was not strongly influenced by geographical origin of the isolates in Brazil (Oliveira et al. 2022). Individual and multi-locus trees showed a typical topology delineating known Chrysoporthe species.

With regards to the pathogenicity test, the greater susceptibility of clones CNB 007 to the wilt followed by CNB 010 was also observed in a survey carried out by the Company where these Eucalyptus clones have been propagated (C. S. Abreu, personal communication). This is despite the fact that susceptibility at the clonal mini-garden it was evaluated using the incidence of wilting plants, whereas our findings were based on the length of the xylem lesion. It is noteworthy that the wilt on Eucalyptus mini-stumps arises just after the outset of collections of minicuts, which sheds light on the role played by this mechanical damage in the development of the disease.

The fact that the aggressiveness of the Ch. cubensis isolate differs depending on the Eucalyptus clone demonstrates that the clones present different levels of susceptibility. Similar results were obtained when isolates of Ch. cubensis were inoculated on Eucalyptus trees belonging to different genotypes since there is a high inter and intra-specific genetic variability among Eucalyptus genotypes (Guimarães et al. 2010; Van Heerden et al. 2005).

The aggressiveness of Ch. cubensis on Eucalyptus plants was directly related to temperature: the higher the temperature, the greater the susceptibility to the pathogen. However, the higher temperature had a greater influence on lesion development in CNB 032 than in CNB 010, which indicates that less susceptible clones may be more influenced by higher temperatures. The positive effect of temperature on disease development is supported by the geographic distribution of the pathogen in tropical and subtropical areas between latitudes 30° North and South, and its predominance in areas with high temperatures and rainfall (Hodges et al. 1979).

This work revealed the causal agent of a new disease in mini-stumps of Eucalyptus that has negatively impacted their longevity and caused significant losses, leading to implications such as the need to replace the dead mini-stumps with new healthy ones, subsequently downgrading productivity and increasing costs. However, a question that has not yet been answered is what the possible inoculum source of Ch. cubensis would be considering that the Eucalyptus canker disease does not occur on the extensive Eucalyptus plantations that surround the clonal mini-garden where this study was performed. Thus, a hypothesis that needs to be investigated is whether Ch. cubensis is present as an endophytic fungus on Eucalyptus mini-stumps. Chrysoporthe cubensis and other Cryphonectriaceae have been found as asymptomatic fungal endophytes in Melastomataceae trees and shrubs that grow alongside commercial Eucalyptus plantations in Colombia (Granados et al. 2020). If this is also the case on Eucalyptus mini-stumps, the stress that results from often collection of minicuts may lead to the development of the wilt disease.