Susceptibility of Eucalyptus hybrid clones to Botryosphaeria canker in Uganda

ABSTRACT The study assessed the susceptibility of the nine commonly grown Eucalyptus clones to Neofusicoccum species associated with Botryosphaeria canker in Uganda. The inoculation trials indicated that susceptibility of Eucalyptus hybrids differed significantly (p = .000), clones GU609, GU7, GC578, and GC796 exhibiting a higher tolerance than GC784, GC550, GU8, GC514, and GC540. The results further revealed that N. parvum was more pathogenic than N. kwambonambiense. The generated information can be exploited in sustainable forest management by expanding the growing of tolerant hybrids in areas with high Botryosphaeria canker disease pressure.


Introduction
Botryosphaeria canker is associated with fungi in the family Botryosphaeriaceae which include species that are most commonly known to cause die-back and canker diseases on twigs, branches and trunks of trees including Eucalyptus species and hybrids . Neofusicoccum species (formerly known as Botryosphaeria species) belong to the Botryosphaeriaceae and have been reported worldwide in countries such as Uganda, Australia, Chile, China, Ethiopia, Indonesia, South Africa, Uruguay, Venezuela to mention but few (Iturritxa, Slippers, Mesanza, & Wingfield, 2011;Mohali, Slippers, & Wingfield, 2009;. Neofusicoccum species are saprophytic or endophytic organisms which infect a wide range of monocotyledonous, dicotyledonous and gymnosperm hosts (Burgess, Barber, & Hardy, 2005;Burgess & Wingfield, 2002;. Symptom development in the hosts normally occurs when exposed to unsuitable environmental conditions such as drought, freezing, extreme temperatures, defoliation, hail, and wounds caused by insects or other pathogens (Roux et al., 2005).
Clonal propagation is known to be of great value in plantation forestry worldwide as it facilitates the development of uniformly fast-growing genotypes with homogeneous physical properties and disease resistance (Dehon, Rezende, Resende, & Assis, 2014;Wingfield et al., 2013). Eucalyptus clonal hybrids were introduced to E. Africa and Uganda in 1997, from Mondi Forests Ltd, in South Africa, suitable for a range of wood products and for planting by smallholder farmers (Epila-Otara & Ndhokero 2010;Kilimo Trust, 2011). The clones were generated by cross-breeding Eucalyptus species and produced three main hybrid clone groups. GC clones (grandis-camaldulensis) that are a cross between Eucalyptus grandis and Eucalyptus camaldulensis; GU clones (grandis-uroplylla) that are a cross between Eucalyptus grandis and Eucalyptus uroplylla; and GT clones (grandistereticornis) that are a cross between Eucalyptus grandis and Eucalyptus tereticornis. The most popular clones grown in Uganda are GC and GUs that are grown throughout the country (Epila-Otara andNdhokero 2010, Kilimo Trust, 2011).
It's anticipated that the successful use of Eucalyptus hybrid clones in plantation forestry is capable of reducing timber and wood shortage worldwide (Dehon et al., 2014;Wingfield et al., 2013). However, diseases such as Botryosphaeria canker, may threaten their successful establishment. The genetic uniformity of clonal hybrids, exposes them more to disease if susceptible ones are grown under favorable environmental conditions as compared to Eucalyptus species (Guimarães et al., 2010). Previous nationwide surveys in Uganda reported Botryosphaeria canker being the most widely distributed disease (Roux and Slippers 2007;Nyeko & Nakabonge, 2008). Neofusicoccum parvum (Pennycook & Samuels) Crous, Slippers & A.J.L. Phillips) and Neofusicoccum kwambonambiense Pavlic, Slippers, M.J. Wingfield, have similarly been reported as the most predominant and pathogenic Botryosphaeriaceae on Eucalyptus species in many parts of the world (Pillay, Slippers, Wingfield, & Gryzenhout, 2013) and previous studies reported their presence in Uganda (Nakabonge, 2002;Toljander, Nyeko, Stenström, Ihrmark, & Barklund, 2007). In spite of this, no studies have been conducted to determine the susceptibility of the grown Eucalyptus hybrid clones to the disease. In this study, we assessed the susceptibility of nine Eucalyptus hybrid clones to Botryosphaeria canker caused by Neofusicoccum species. Understanding the susceptibility of Eucalyptus hybrids to Botryosphaeria canker is essential in an effort to identify resistant hybrid clones, since the alternative mechanisms such as the use of fungicides have not recorded success in management of the disease .

Isolation of fungi
Samples were collected from symptomatic Eucalyptus hybrid clones GU and GCs from Serere in Serere district Eastern Uganda, Mukungwe, Masaka District central Uganda, Ntungamo sub-country Ntungamo District in western Uganda and Kifu, Mukono District Central Uganda (Figure 1). The areas were selected based on previous reports of occurrence of Botryosphaeria canker (Nyeko & Nakabonge, 2008). For isolations, perithecia or pycnidia were picked from the Eucalyptus twigs and plated directly onto sterile, 2% malt extract agar (MEA), Fisher Scientific UK Limited. The MEA cultures were incubated for 10 days at 36°C until mycelia with fruiting structures grew. Single conidia from isolates resembling Botryosphaeriaceae (e.g., brownish white and fluffy mycelia) were sub-cultured until pure cultures were obtained.

Molecular phylogenetic characterization
For each single conidial isolate culture, actively growing mycelium were scraped off the surface a MEA plate and used for DNA extraction using the method described by Roux, Coutinho, Byabashaija, and Wingfield (2001). DNA concentrations were estimated visually on a 1% agarose gel using known concentrations of lambda (λ) DNA after dying with ethidium bromide and photographed under UV illumination.
Amplification of the ITS region was performed with the primers ITS1F (Gardes & Bruns, 1993) and ITS4 (Farris, Kallersjo, Kluge, & Bult, 1995). The PCR reaction mixture (25 μL), PCR conditions and visualization of products were as described by Slippers et al. (2004). Five μL of the PCR reaction mixture was loaded onto a 2% agarose gel, also containing 1% ethidium bromide. This was exposed to UV light to visualize the PCR products. Sanger sequencing was performed using both forward and reverse primers at INQABA Biotech, Pretoria, South Africa.
Homology searches were done from the GenBank/EMBL databases using the BLAST program (National Center for Biotechnology Information, U. S. National Institute of Health, Bethesda. http://www.ncbi.nlm.nih.gov/BLAST). Eight sequences with homologies >80%, known to be of Botryosphaeria spp. (Table 1), were selected and co-aligned with those from the Ugandan sequences obtained in this study (Table 1) using the program ClustalW (Thompson, Higgins, & Gibson, 1994) in MEGA Version X (Kumar, Stecher, Li, Knyaz, & Tamura, 2018).
Phylogenetic analysis was performed in MEGA Version X software (Kumar et al., 2018). The evolutionary history was inferred using the UPGMA (Unweighted Pair Group Method with Arithmetic mean) method (Sneath & Sokal, 1973). The analysis involved 14 nucleotide sequences. The evolutionary distances were computed using the Maximum Composite Likelihood method (Tamura, Nei, & Kumar, 2004).The percentage of replicate trees in which the associated taxa clustered together in the bootstrap test (500 bootstraps) was generated (Felsenstein, 1985). The tree was drawn to scale, with branch lengths in the same units as those of the evolutionary distances used to infer the phylogenetic tree. The fungus Guignardia mangiferae A.J. Roy, known to be related to Botryosphaeria spp. as a Botryosphaeriaceae was used as the out-group to root the trees   (Table 1). The identified species where used for pathogenicity tests.

Pathogenicity trial
A total of eight (GU 7, GU 8, GU 609, GC 540, GC 796, GC 784, GC 578, and GC 550) 7-months old clones were used in the pathogenicity tests. The clonal cuttings were grown in pots under screen house conditions at Makerere University, Uganda with day/night temperature of approximately 20-25°C. Each clone consisted of 35 trees and these included 7 replicates (trees) for each of the 4 isolates. In the same clone were included seven replicates to represent the control. This made the total number of trees per clone 35 (5x7) and the number of trees in the experiment 315 (35x9).
Using a 5 mm cork borer, agar discs were made from 2 week old MEA plate cultures of each isolate and from sterile MEA plates for the controls. The same size wounds were made in the bark of the~2 m tall clones to expose the cambium at a height of about 40 cm from the soil level. Trees were immediately inoculated by placing an agar disk, with the mycelium side facing the cambium into each wound and the inoculation sites were rapidly sealed with parafilm (Pechiney, Chicago, USA) to prevent desiccation and contamination. Controls were inoculated in the same way using non-colonized agar plugs. Six weeks following inoculation the lengths and widths of the resulting lesions were taken. The fungus was re-isolated from the lesions by cutting small pieces of wood from the leading edges of lesion margins that were surface-sterilized in 70% ethanol and plated directly onto MEA. The whole experiment was repeated to improve the precision and reliability of the results.
Data from the two experiments were merged. One way ANOVA was used to determine the variation in the effects of isolates on individual clones. Isolates were grouped into species and data were re-analyzed using one way ANOVA to determine the differences in the effects of Botryosphaeria species on individual clones. An overall analysis of data was done to determine the overall pathogenicity of isolates using two-way ANOVA.

Isolation and molecular characterization
A total of 58 Botrosphariaceae samples were collected and isolated from symptomatic Eucalytpus trees. From the 58 isolates, 5 were selected based on cultural characteristics, and used for DNA sequencing. Alignment of the 14 total isolates including those from the genebank were successfully conducted. Phylogenetic analysis of the aligned 14 sequences generated an optimal tree with the sum of branch length = 0.19079930. The percentage of replicate trees in which the associated taxa clustered together in the bootstrap test (500 replicates) are indicated next to the branches (Figure 2). The tree was drawn to scale, with branch lengths in the same units as those of the evolutionary distances used to infer the phylogenetic tree. All ambiguous positions were removed for each sequence pair. There were a total of 624 positions in the final dataset. Phylogenetic analysis revealed three groups representing three species. Where, isolates GN1, GN2 and GN3 represented N. parvum, and GN5 represented N. kwambonambiense ( Figure 2) and GN4 represented Neofusicoccum species an unresolved isolate that will require further efforts to identify to species level.

Pathogenicity of Botryosphaeria isolates
Nine weeks after inoculation, all the inoculated trees had developed lesions. Clear brown discolorations stretching from the point of inoculation were observed and all controls were overgrown with callus tissue (Figures 3 and 4). Trees inoculated with GN2 had the longest lesions followed by GN3, GN1, GN4, and GN5, respectively. The widths varied slightly although appeared to be significantly different (F = 18.73; df = 5; P = .000). Lesion lengths differed significantly among isolates (F = 45.76; df = 5; P = .000). There was also a significant isolate x clone interaction (F = 2.35; df = 40; P = .000). Neofusicoccum parvum (GN1, GN2, and GN3) had the highest lesion dimensions, followed by Neofusicoccum species (GN4) and lastly N. kwambonambiense (GN5). The mean widths varied slightly among the isolates.

Susceptibility of eucalyptus clones to Neofusicoccum sp
Eucalyptus clones GC 784, GC 550, GU 8, GC 514, and GC 540 exhibited higher lesion lengths than clones GU 609, GU 7, GC 578 and GC 796. However, the width of lesions indicated GU8 and GU7 as the highest while the rest of the clones varied slightly. Lesion dimensions differed significantly (F = 4.79; df = 8; P = 000) and (F = 5.83; df = 8; P = .05) for length and width, respectively. Length of lesions indicated no significant difference in clone by species interaction (P = .000). On the contrary, the lesion width showed a significant (P = .000) interaction between clones and species.

Discussion
We evaluated the relative susceptibility of Eucalyptus hybrid clones that are currently grown as commercial plantation forest trees in Uganda to Neofusicoccum species and assessed the pathogenicity of the fungal species to the hybrid clones. There was varying susceptibility within the Eucalyptus hybrid clones and pathogenicity between the two Neofusicoccum species. The results indicate that Eucalyptus hybrid clones could be used in the sustainable management of Botryosphaeria canker in areas where the disease is prevalent.
DNA sequences of the ITS regions of five (5) Botryosphaeriaceae isolates yielded Neofusicoccum parvum and Neofusicoccum kwambonambiense species. Neofusicoccum parvum was first described from Kiwifruit and Populus spp. in New Zealand as B. parva (Pennycook, & Samuels, 1985). Other studies, identified Neofusicoccum parvum causing disease on Eucalyptus and other woody species in Uganda, South Africa and Venezuela (Heath et al. 2011, Nakabonge, 2002Mohali, Slippers, & Wingfield, 2007;Mohali et al., 2009;Pavlic, Slippers, Coutinho, & Wingfield, 2009). N. kwambonambiense is a closely related species to N. parvum and was isolated from Eucalyptus and from symptomless branches and leaves, dying branches and pulp of ripe fruits of Syzygium cordatum in Kwambonambi South Africa thus its name (Pavlic, Slippers, Coutinho, Gryzenhout, & Wingfield, 2007;Pavlic et al., 2009). The isolation of N. parvum and N. kwambonambiense from samples collected during the current study confirms earlier reports of the occurrence of the pathogens in Uganda and their association to clonal Eucalyptus hybrids (Nakabonge, 2002).
Trees inoculated with sterile media were able to recover from wounds by developing callus tissue unlike fungal-inoculated trees which had growing lesions, confirming that Neofusicoccum species which were used are pathogenic to all the inoculated Eucalyptus clones. The endophytic nature of Neofusicoccum species enables them to live within healthy plant tissues (Smith, Wingfield, & Petrini, 1996) and this can make their introduction into new environments on germplasm easily possible (Slippers et al., 2004). Although Eucalyptus hybrid clones grown in Uganda have their origin in South Africa, the disease had already been reported in the country on E. grandis as far back as 2001 (Nakabonge, 2002), before the introduction of hybrid clones and their nationwide planting.
A significant difference in the levels of pathogenicity among Neofusicoccum species were recorded. N. parvum was the most aggressive to all Eucalyptus hybrid clones used in the study (Figure 3). N. parvum is known as an aggressive canker pathogen, possessing a high potential of causing large lesions, cracks in the bark and black kino exudation on Eucalyptus hybrid clones (Mohali et al., 2009). The fungus has a wide host range and global distribution and it has been recorded as the most pathogenic on woody plants of all Botryosphaeriales (Jami, Wingfield, Gryzenhout, & Slippers, 2017;Li et al., 2014;. Neofusicoccum kwambonambiense was the less pathogenic, in this study but significant differences were observed from the control (P = .000) (Figure 3). In a study conducted by Pavlic et al. (2009), however, N. kwambonambiense was reported to be more pathogenic, than N. parvum, to Syzygium cordatum in South Africa. Earlier studies though had reported N. parvum and N. kwambonambiense being more pathogenic to Eucalyptus than S. cordatum (Pavlic et al., 2007). The presence of these species in Uganda and their confirmed ability to cause disease on Eucalyptus clones grown in the region is an indication that monitoring campaigns should be established to avert any losses to the commercial forestry sector. Exploring the varying resistance to diseases in the Eucalyptus planting stock, could contribute to the sustainable management of plantation forests.
Generally, lesions which developed on inoculated trees were relatively small. This may give an impression that all clones tested in this study have a certain degree of tolerance to Neofusicoccum infection. This tolerance however differed significantly among the clones, with GU609, GU7, GC578, and GC796 exhibiting a higher tolerance than GC784, GC550 GU8, GC514 and GC540 (Figure 4). The variation in susceptibility of Eucalyptus hybrid clones corroborates previous studies by van Heerden, Amerson, Preisig, Wingfield, and Wingfield (2005) and Guimarães et al. (2010) toward Chryphonetria cubensis.
In conclusion, the varying responses reported in this study indicate that there is an opportunity to sustainably manage the disease by growing clones that are tolerant to Botryosphaeria canker in places where the disease is most prevalent. Since, Neofusicoccum species are known to be stress-related and opportunistic pathogens of Eucalyptus spp. (Smith et al., 1996), there is a need to avoid the situation that may trigger development of disease in rather tolerant clones. The study could also be repeated under field conditions and in various agro-ecological zones to identify genotypes suitable for various conditions. Meanwhile, the more tolerant clones (i.e., GU609, GU7, GC578, and GC796) can be planted in the most affected areas.

Limitations
The study has generated information which is the first of its kind toward understanding the resistance of the grown Eucalyptus clones to the most prevalent disease in Uganda. However, it could have been interesting to evaluate the pathogenicity of Botryosphaeriaceae from other hosts on Eucalyptus clones grown in Uganda. Further studies should address this important aspect that will contribute to ecological sustainability in plantation forestry disease management.