Ceratocystis eucalypticola sp. nov. from Eucalyptus in South Africa and comparison to global isolates from this tree.

Eucalyptus trees, mostly native to Australia, are widely planted in the tropics and Southern Hemisphere for the production of wood and pulp. Worldwide surveys of diseases on these trees have yielded a large collection of Ceratocystis isolates from dying trees or from wounds on their stems. The aim of this study was to characterise these isolates and to consider their relatedness to each other. Culture appearance, morphological features and a distinctive fruity odour in all cultures were typical of species in the Ceratocystis fimbriatasensu lato (s. lat.) complex. Phylogenetic analyses of sequences for the combined ITS, βt-1 and TEF1-α gene regions revealed a genetically diverse group of isolates residing in a single large clade, that were distinct from all other species in the C. fimbriatas. lat. complex. Based on morphology and phylogenetic inference, the Eucalyptus isolates are recognised as closely related. The South African isolates are described here as a new species, C. eucalypticola.

INTRODUCTION

Eucalyptus species are mostly native to Australia, but have been widely planted in the tropics and Southern Hemisphere. This is because they are adapted to a wide range of different environments and are typically fast growing. It has further been suggested that the success of these trees as non–natives is due to the separation from their natural enemies (Wingfield , Roux & Wingfield 2009). The potential threat of pests and pathogens to the sustainability of eucalypt plantations in areas where they are not native is consequently great and of substantial concern to forestry industries globally (Old , Wingfield ).

In order to understand and manage the threat of pests and pathogens to Eucalyptus species grown as non-natives and in plantations, tree health surveys are undertaken regularly. Amongst the pathogens that have been found on these trees, a Ceratocystis sp. in the C. fimbriata s. lat. complex causes serious disease problems in Brazil, the Republic of Congo, Uganda, and Uruguay (Laia , Roux , 2001, 2004, Barnes ). Various other Ceratocystis species in the C. fimbriata s. lat. complex have also been found on naturally occurring or artificially induced wounds on the stems of trees, in various parts of the world. Some of these have been shown to be cryptic taxa that have been provided with names (van Wyk , 2008, 2010a, Rodas , Heath , Kamgan Nkuekam ). Several species are thought to be pathogens, while the role of others in tree health is not known.

The genus Ceratocystis comprises a diverse group of fungi, including saprophytes causing blue-stain of lumber and serious pathogens that cause mortality (Kile 1993). The genus is typified by C. fimbriata s. str. that is a pathogen restricted to root crops, specifically sweet potato (Engelbrecht & Harrington 2005). Ceratocystis fimbriata s. lat. represents a diverse assemblage of isolates, some of which have been treated as distinct taxa defined based on phylogenetic inference, morphological differences, and mating behaviour (Barnes , Engelbrecht & Harrington 2005, Johnson , van Wyk , 2008, Heath ). However, Ferreira treated some isolates of the C. fimbriata s. lat. complex from Brazil as representing a particular population of C. fimbriata s. str., rather than as discrete taxa.

Global surveys of the health of Eucalyptus species in plantations have yielded a large collection of isolates that can loosely be accommodated in the C. fimbriata s. lat. complex. The aim of this study was to characterise these isolates and to consider patterns in their distribution on Eucalyptus species worldwide.

MATERIALS AND METHODS

Isolates

Isolates used in this study were obtained from: (1) artificially induced wounds on the stems of Eucalyptus trees in South Africa, Thailand, and Indonesia (Table 1). The isolates were obtained by directly transferring spore masses from the apices of ascomata produced on the wounded inner bark and wood to agar plates. When sporulating structures were absent, the wood samples were placed in moist chambers to enhance sporulation. Spore masses were transferred to 2 % Malt Extract Agar (MEA) in Petri dishes and incubated at room temperature. Additionally, the carrot baiting technique was used to obtain isolates (Moller & DeVay 1968). (2) cultures were sourced from the culture collection (CMW) of the Forestry and Agricultural Biotechnology Institute (FABI) at the University of Pretoria, South Africa. These isolates had previously been identified as representing the C. fimbriata s. lat. complex and were from diseased Eucalyptus trees in various parts of the world including Brazil, Uganda, Congo, and Uruguay (Table 1).

Table 1. Isolates of Ceratocystis fimbriata s. lat. spp. used in this study.

Species	Isolate no.	GenBank accession no.	Host	Area
C. albifundus	CMW4068	DQ520638, EF070429, EF070400	Acacia mearnsii	South Africa
C. albifundus	CMW5329	AF388947, DQ371649, EF070401	Acacia mearnsii	Uganda
C. atrox	CMW19383, CBS120517	EF070414, EF070430, EF070402	Eucalyptus grandis	Australia
C. atrox	CMW19385, CBS120518	EF070415, EF070431, EF070403	Eucalyptus grandis	Australia
C. cacaofunesta	CMW15051, CBS152.62	DQ520636, EF070427, EF070398	Theobroma cacao	Costa Rica
C. cacaofunesta	CMW14809, CBS115169	DQ520637, EF070428, EF070399	Theobroma cacao	Ecuador
C. caraye	CMW14793, CBS114716	EF070424, EF070439, EF070412	Carya cordiformis	USA
C. caraye	CMW14808, CBS115168	EF070423, EF070440, EF070411	Carya ovata	USA
C. colombiana	CMW9565, CBS121790	AY233864, AY233870, EU241487	Soil	Colombia
C. colombiana	CMW5751, CBS121792	AY177233, AY177225, EU241493	Coffea arabica	Colombia
C. colombiana	CMW9572	AY233863, AY233871, EU241488	Mandarin	Colombia
C. eucalypticola	CMW9998, CBS124017	FJ236721, FJ236781, FJ236751	Eucalyptus sp.	South Africa
C. eucalypticola	CMW10000, CBS124019	FJ236722, FJ236782, FJ236752	Eucalyptus sp.	South Africa
C. eucalypticola	CMW11536, CBS124016	FJ236723, FJ236783, FJ236753	Eucalyptus sp.	South Africa
C. eucalypticola	CMW12663	FJ236724, FJ236784, FJ236754	Eucalyptus sp.	South Africa
C. eucalypticola	CMW15054, CBS124018	FJ236725, FJ236785, FJ236755	Eucalyptus sp.	South Africa
C. fimbriata s. str.	CMW15049, CBS141.37	DQ520629, EF070442, EF070394	Ipomaea batatas	USA
C. fimbriata s. str.	CMW1547	AF264904, EF070443, EF070395	Ipomaea batatas	Papua New Guinea
C. fimbriatomima	CMW24174, CBS121786	EF190963, EF190951, EF190957	Eucalyptus sp.	Venezuela
C. fimbriatomima	CMW24176, CBS121787	EF190964, EF190952, EF190958	Eucalyptus sp.	Venezuela
C. larium	CMW25434, CBS122512	EU881906, EU881894, EU881900	Styrax benzoin	Indonesia
C. larium	CMW25435, CBS122606	EU881907, EU881895, EU881901	Styrax benzoin	Indonesia
C. manginecans	CMW13851, CBS121659	AY953383, EF433308, EF433317	Mangifera indica	Oman
C. manginecans	CMW13852, CBS121660	AY953384, EF433309, EF433318	Hypocryphalus mangifera	Oman
C. neglecta	CMW17808, CBS121789	EF127990, EU881898, EU881904	Eucalyptus sp.	Colombia
C. neglecta	CMW18194, CBS121017	EF127991, EU881899, EU881905	Eucalyptus sp.	Colombia
C. obpyriformis	CMW23807, CBS122608	EU245004, EU244976, EU244936	Acacia mearnsii	South Africa
C. obpyriformis	CMW23808, CBS122511	EU245003, EU244975, EU244935	Acacia mearnsii	South Africa
C. papillata	CMW8857	AY233868, AY233878, EU241483	Annona muricata	Colombia
C. papillata	CMW8856, CBS121793	AY233867, AY233874, EU241484	Citrus lemon	Colombia
C. papillata	CMW10844	AY177238, AY177229, EU241481	Coffea arabica	Colombia
C. pirilliformis	CMW6569	AF427104, DQ371652, AY528982	Eucalyptus nitens	Australia
C. pirilliformis	CMW6579, CBS118128	AF427105, DQ371653, AY528983	Eucalyptus nitens	Australia
C. platani	CMW14802, CBS115162	DQ520630, EF070425, EF070396	Platanus occidentalis	USA
C. platani	CMW23918	EF070426, EF070397, EU426554	Platanus sp.	Greece
C. polychroma	CMW11424, CBS115778	AY528970, AY528966, AY528978	Syzygium aromaticum	Indonesia
C. polychroma	CMW11436, CBS115777	AY528971, AY528967, AY528979	Syzygium aromaticum	Indonesia
C. polyconidia	CMW23809, CBS122289	EU245006, EU244978, EU244938	Acacia mearnsii	South Africa
C. polyconidia	CMW23818, CBS122290	EU245007, EU244979, EU244939	Acacia mearnsii	South Africa
C. populicola	CMW14789, CBS119.78	EF070418, EF070434, EF070406	Populus sp.	Poland
C. populicola	CMW14819, CBS114725	EF070419, EF070435, EF070407	Populus sp.	USA
C. smalleyi	CMW14800, CBS114724	EF070420, EF070436, EF070408	Carya cordiformis	USA
C. smalleyi	CMW26383, CBS114724	EU426553, EU426555, EU426556	Carya cordiformis	USA
C. tanganyicensis	CMW15991, CBS122295	EU244997, EU244969, EU244929	Acacia mearnsii	Tanzania
C. tanganyicensis	CMW15999, CBS122294	EU244998, EU244970, EU244939	Acacia mearnsii	Tanzania
C. tsitsikammensis	CMW14276, CBS121018	EF408555, EF408569, EF408576	Rapanea melanophloeos	South Africa
C. tsitsikammensis	CMW14278, CBS121019	EF408556, EF408570, EF408577	Rapanea melanophloeos	South Africa
C. variospora	CMW20935, CBS114715	EF070421, EF070437, EF070409	Quercus alba	USA
C. variospora	CMW20936, CBS114714	EF070422, EF070438, EF070410	Quercus robur	USA
C. virescens	CMW11164	DQ520639, EF070441, EF070413	Fagus americanum	USA
C. virescens	CMW3276	AY528984, AY528990, AY529011	Quercus robur	USA
C. zombamontana	CMW15235	EU245002, EU244974, EU244934	Eucalyptus sp.	Malawi
C. zombamontana	CMW15236	EU245000, EU244972, EU244932	Eucalyptus sp.	Malawi
Ceratocystis sp.	CMW4797	FJ236733, FJ236793, FJ236763	Eucalyptus sp.	Congo
Ceratocystis sp.	CMW4799	FJ236734, FJ236794, FJ236764	Eucalyptus sp.	Congo
Ceratocystis sp.	CMW4902	FJ236715, FJ236775, FJ236745	Eucalyptus sp.	Brazil
Ceratocystis sp.	CMW5312	FJ236731, FJ236791, FJ236761	Eucalyptus sp.	Uganda
Ceratocystis sp.	CMW5313	FJ236732, FJ236792, FJ236762	Eucalyptus sp.	Uganda
Ceratocystis sp.	CMW7764	FJ236726, FJ236786, FJ236756	Eucalyptus sp.	Uruguay
Ceratocystis sp.	CMW7765	FJ236727, FJ236787, FJ236757	Eucalyptus sp.	Uruguay
Ceratocystis sp.	CMW7766	FJ236728, FJ236788, FJ236758	Eucalyptus sp.	Uruguay
Ceratocystis sp.	CMW7767	FJ236729, FJ236789, FJ236759	Eucalyptus sp.	Uruguay
Ceratocystis sp.	CMW7768	FJ236730, FJ236790, FJ236760	Eucalyptus sp.	Uruguay
Ceratocystis sp.	CMW14631	FJ236744, FJ236804, FJ236774	Eucalyptus sp.	Indonesia
Ceratocystis sp.	CMW14632	FJ236743, FJ236803, FJ236773	Eucalyptus sp.	Indonesia
Ceratocystis sp.	CMW16008	FJ236735, FJ236795, FJ236765	Eucalyptus sp.	Thailand
Ceratocystis sp.	CMW16009	FJ236736, FJ236796, FJ236766	Eucalyptus sp.	Thailand
Ceratocystis sp.	CMW16010	FJ236737, FJ236797, FJ236767	Eucalyptus sp.	Thailand
Ceratocystis sp.	CMW16034	FJ236739, FJ236799, FJ236769	Eucalyptus sp.	Thailand
Ceratocystis sp.	CMW16035	FJ236738, FJ236798, FJ236768	Eucalyptus sp.	Thailand
Ceratocystis sp.	CMW18572	FJ236740, FJ236800, FJ236770	Eucalyptus sp.	Indonesia
Ceratocystis sp.	CMW18577	FJ236742, FJ236802, FJ236772	Eucalyptus sp.	Indonesia
Ceratocystis sp.	CMW18591	FJ236741, FJ236801, FJ236771	Eucalyptus sp.	Indonesia

PCR and sequencing reactions

DNA was extracted from all isolates as described by van Wyk . Three gene regions were selected for PCR amplification, including ITS1 and ITS2, including the 5.8S rDNA operon, part of the beta-tubulin (βt-1) gene, and part of the Transcription Elongation Factor-1 alpha (TEF1-α) gene region. The reactions and programme for amplification were as described by van Wyk . The primers utilized were ITS1 and ITS4 (White ), βt1a and βt1b (Glass & Donaldson 1995), and EF1F and EF1R (Jacobs ).

Sequencing reactions were set up and run as described by van Wyk . Sequences of the isolates from Eucalyptus were analysed with Chromas Lite 2.01 (http://www.technelysium.com.au). These sequences as well as those for all species in the C. fimbriata s. lat. species complex (Table 1) were aligned using MAFFT (http://timpani.genome.ad.jp/%7emafft/server/) (Katoh ). All sequences derived from this study have been deposited in GenBank (Table 1).

Combined gene tree for all described species in the C. fimbriata s. lat. complex

Representative isolates of all described species in the C. fimbriata s. lat. complex were included in this dataset, including those obtained for this study from CMW. The sequences of three gene regions (ITS, βt-1 and TEF1-α) were combined and a partition homogeneity test (PHT) was used to determine if the data from the three regions could be combined, using the software programme PAUP v. 4.0b10 (Swofford 2002). Settings in PAUP were as described in van Wyk . Ceratocystis virescens was selected as the outgroup taxon.

MrModeltest2 (Nylander 2004) was used to determine the most appropriate model of nucleotide substitution for each of the three gene regions, respectively. These models were then included in the Bayesian analyses using MrBayes (Ronquist & Huelsenbeck 2003). The Bayesian analyses were run as described in van Wyk .

Combined and separate gene trees of un-named Ceratocystis fimbriata s. lat. isolates obtained from Eucalyptus

This dataset consisted only of Ceratocystis fimbriata s. lat. isolates from Eucalyptus trees and that have not yet been described as separate species. A closely related and previously described species, C. colombiana, also obtained from Eucalyptus, was included as an outgroup. This was done to determine whether these isolates represent one group with no separate grouping or whether geographical grouping exists, as has been documented in C. fimbriata s. lat. (Engelbrecht & Harrington 2005, Ferreira ).

Models were obtained for each of the ITS, βt-1 and TEF1-α gene regions with the use of MrModeltest2 (Nylander 2004). Consistent with both the first datasets, these models were incorporated into MrBayes (Ronquist & Huelsenbeck 2003) in order to run Bayesian analyses.

Utilising the C. fimbriata s. lat. isolates from Eucalyptus trees obtained from the CMW culture collection, the Molecular Evolutionary Genetics Analysis software (MEGA) 4 (Tamura ) was used to determine the amount of variation for each gene region. The three gene regions were inspected to determine the number of fixed alleles between them. Allele trees were drawn using the software TCS (Clement ) from the combined dataset for the Eucalyptus isolates, including the closely related species C. colombiana, known only from Eucalyptus.

Culture characteristics and morphology

Two isolates of Ceratocystis fimbriata s. lat. from Eucalyptus were selected from each country, other than Brazil, for which only one Eucalyptus isolate was available. These were used to describe morphological characteristics. Isolates were transferred to each of five 2 % Malt Extract Agar (MEA) plates and incubated in the dark. The isolates were incubated at 30 °C for 7 d, after which the growth was assessed.

Microscopic examinations were made of isolates from Indonesia, Uruguay, Thailand, and South Africa. Isolates from other countries were excluded because the cultures did not produce ascomata. All taxonomically informative structures were measured from 10 d old cultures on 2 % MEA, mounted in lactic acid. Ten measurements were made for each of the two isolates from Indonesia, Uruguay, Thailand, and South Africa.

A preliminary study of isolates representing the larger collection of C. fimbriata s. lat. isolates from Eucalyptus, and nested together in the same phylogenetic clade, showed that they are morphologically very similar. Consequently, four isolates (CMW 9998, CMW 15054, CMW 10000 and CMW 11536) from Eucalyptus in South Africa were selected for more detailed study. These South African isolates were transferred to five 2 % MEA plates each and incubated at seven different temperatures. These temperatures included 4 °C and six temperatures between 10 °C and 35 °C at 5 °C intervals. Growth was assessed after 7 d of incubation in the dark. Colony colour was assessed for the same isolates used as in the growth studies, grown on 2 % MEA for seven to 10 d at room temperature (25 °C). The colour charts of Rayner (1970) were used for descriptions of colony colour.

Fifty measurements were made of all taxonomically informative characters for isolate CMW 11536 from Eucalyptus in South Africa. An additional ten measurements were made of these structures for isolates CMW 9998 and CMW 10000 and CMW 15054. The minimum, maximum, average and standard deviation (stdv) was calculated for the measurements of each structure and these are presented in this study as; (minimum-) stdv minus the mean – stdv plus the mean (-maximum).

RESULTS

Twenty-five isolates obtained from CMW that had been isolated from Eucalyptus trees were included in this study (Table 1). Fifteen of these originated from natural or artificially induced wounds on trees in three countries, South Africa, Thailand, and Indonesia. In addition, ten of the isolates were from trees that are believed to have been killed by the fungus. The latter isolates were from Brazil, Congo, Uganda, and Uruguay.

Results were obtained for three separate datasets. The first provided a broad phylogenetic placement (i.e. Latin American or North American, Asian, and African clade) of the C. fimbriata s. lat. isolates from Eucalyptus. A more focussed analysis determined whether these isolates could be linked to any of the previously described species in the C. fimbriata s. lat. complex that were obtained from Eucalyptus. Thereafter, the isolates from Eucalyptus apparently representing undescribed species were considered in combined as well as single gene trees generated from the sequence data for these isolates. This was to determine whether they could be grouped based on geographical origin.

Combined gene tree for all described species in the Ceratocystis fimbriata s. lat. complex

Amplicons for the three gene regions were on average 500 bp for the ITS and βt-1 gene regions and 800 bp for the TEF1-α region (Table 1). The PHT for the data set including all described species in the C. fimbriata s. lat. complex, had a low value (P=0.01), but could be combined (Cunningham 1997).

Of the 1 989 characters in this dataset, 1 102 were constant, 45 were parsimony uninformative while 842 were parsimony informative. One hundred and forty two most parsimonious trees were obtained, of which one was selected for presentation (Fig. 1). The tree topology was as follows: Tree length (TL) = 2054 steps, Consistency Index (CI) = 0.7, Retention Index (RI) = 0.9 and Rescaled Consistency (RC) = 0.6. Phylogenetic analyses revealed a clade specific for the isolates from Eucalyptus (Fig. 1). Isolates in this large clade had high bootstrap (88 %) and Bayesian (88 %) support and included some substructure (Fig. 1). The substructure in the large clade for the isolates from Eucalyptus was not strongly supported and these isolates were treated as reflecting a single group of genetically related, but not identical isolates. The closest phylogenetic relative of the isolates in the Eucalyptus clade was C. colombiana (van Wyk ).

Fig. 1.

Phylogenetic tree based on the combined sequences of the ITS, βt and TEF1-α gene regions for isolates from Eucalyptus including those provided the name C. eucalypticola and other described species in the C. fimbriata s. lat. complex. Ceratocystis virescens represents the out-group taxon. Bootstrap values are indicated at the branch nodes and Bayesian values in brackets.

The models obtained using MrModeltest2 were the HKY+I+G model for both the ITS and the TEF1-α genes and the GTR+G model for the βt-1 gene region. Including these models in the Bayesian analyses resulted in a burnin of 7000. These 7000 trees were discarded from the final analyses. The posterior probabilities obtained with the Bayesian analyses supported the bootstrap values obtained in PAUP (Fig. 1).

Combined and separate gene trees for undescribed Ceratocystis fimbriata s. lat. isolates from Eucalyptus

In the dataset for the combined gene regions, there were 1 765 characters of which 1 680 were constant, 31 were parsimony uninformative while 54 were parsimony informative. Twenty-four most parsimonious trees were obtained, one of which was selected for presentation (Fig. 2). The tree topology was as follows: TL = 107 steps, CI = 0.8, RI = 0.9 and RC = 0.7. One well-supported clade (100 % bootstrap, 100 % Bayesian) was observed with high variation. Three clades that were supported within this large clade were also observed (Fig. 2). The models obtained for this dataset were the HKY model for the ITS gene, the F81 model for the βt-1 gene region and the HKY+I model for the TEF1-α gene region. A burn-in of 1000 was obtained and these 1000 trees were discarded from the final analyses. The posterior probabilities obtained with the Bayesian analyses supported the bootstrap values obtained with PAUP (Fig. 2).

Fig. 2.

A phylogenetic tree for the combined sequences of the ITS, βt and EF1-α gene regions, including only the undescribed C. fimbriata s. lat. isolates with Eucalyptus as their host. The closely related species, C. colombiana, is included as outgroup. Bootstrap support is indicated at the branch nodes while Bayesian support is indicated in brackets.

Three well-supported clades were observed; the first included Asian (Indonesia and Thailand) and South American (Brazil and Uruguay) isolates; the second clade included African (Republic of Congo and South Africa) isolates while the third clade included African (Uganda) and Asian (Thailand) isolates. The previously described species, C. colombiana, grouped apart from these three clades (Fig. 2).

Where the data were treated separately, the trees for the ITS, βT and TEF1-α gene regions had a different topology when compared with those for the combined gene regions (Fig. 3). For the ITS gene tree, the same three clades emerged as in the combined dataset and included those for Asian and South American isolates, the African isolates and the African together with Asian isolates. However, only the African and Asian clade had strong support (97 %), the other two clades, African (63 %) and the Asian/ South American (55 %) clades had weak support (Fig. 3). In the case of the βt-1 gene tree, there was no support and all branches collapsed (Fig. 3). For the TEF1-α gene tree, there were two small clades encompassing the South African isolates that had high and medium support (85 % and 65 % respectively), while the rest of the isolates grouped in a single clade with strong (85 %) support (Fig. 3).

Fig. 3.

Three phylograms each representing a single gene region (ITS, βt and TEF-1α, top to bottom) for the undescribed isolates from Eucalyptus representing C. fimbriata s. lat. showing low variation in the three separate gene regions as well as no support for the sub-clades observed in the combined gene trees. No outgroup was assigned to this dataset.

Where data for the C. fimbriata s. lat. isolates were analysed in MEGA, the results showed that in the ITS gene region, the C. fimbriata s. lat. isolates obtained from Eucalyptus were separated from C. colombiana by an average of 23 nucleotide differences (Table 2). Where isolates from different countries were compared, there was also variation in the ITS with a maximum of 13 bp and average of 5 bp differences (Table 2).

Table 2. The number of differences observed between the sequences of the isolates from Eucalyptus (C. fimbriata s. lat.) from Brazil, South Africa, Uruguay, Uganda, Congo, Thailand, Indonesia, and C. colombiana.

Country	Brazil	South Africa	Uruguay	Uganda	Congo	Thailand	Indonesia	C. colombiana
Gene region
ITS
Brazil	–	9	0	6	13	0	0	23
South Africa	9	8	6	6	0	4	9	21
Uruguay	0	6	4	7	9	0	0	21
Uganda	6	6	7	0	9	0	7	28
Congo	13	0	9	9	0	6	11	25
Thailand	0	4	0	0	6	7	0	20
Indonesia	0	9	0	7	11	0	1	22
C. colombiana	23	21	21	28	25	20	22	1
βt
Brazil	–	0	0	0	0	0	0	3
South Africa	0	1	0	0	0	0	0	3
Uruguay	0	0	0	0	0	0	0	3
Uganda	0	0	0	0	0	0	0	3
Congo	0	0	0	0	0	0	0	3
Thailand	0	0	0	0	0	0	0	3
Indonesia	0	0	0	0	0	0	0	3
C. colombiana	3	3	3	3	3	3	3	0
TEF
Brazil	–	13	9	12	12	12	12	21
South Africa	13	7	0	0	0	0	0	8
Uruguay	9	9	9	0	0	0	0	7
Uganda	12	0	0	7	0	0	0	6
Congo	12	0	0	0	0	0	0	8
Thailand	12	0	0	0	0	1	0	8
Indonesia	12	0	0	0	0	0	5	8
C. colombiana	21	8	7	6	8	8	8	0

Where isolates of C. fimbriata s. lat. from Eucalyptus were compared with C. colombiana in the βt-1 gene region, there were only 3bp differences between them (Table 2). Within the clade representing the C. fimbriata s. lat. group from Eucalyptus, there was only one base pair difference observed in the South African group and no differences between isolates from different countries (Table 2).

For the TEF1-α gene region, there were 21bp differences between the isolate from Brazil and C. colombiana and an average of 8 bp differences between C. colombiana and the other isolates from Eucalyptus. Only the single isolate from Brazil differed from the other isolates while no differences were observed between the isolates from the other countries. The allele networks drawn from the combined gene regions (ITS, βt-1 and TEF1-α) for the C. fimbriata s. lat. obtained from Eucalyptus revealed a single tree with high variation (Fig. 4). There was no obvious geographic structure with regards to the origin of the eucalypt isolates. The previously described species, C. colombiana, formed a separate allele tree (Fig. 4).

Fig. 4.

Allele networks obtained from the three combined gene regions (ITS, βt and TEF1-α for all isolates from Eucalyptus as well as C. colombiana. The species C. colombiana is represented as highly different to the Eucalyptus isolates due to the fact that it formed a separate allele tree. The C. fimbriata s.lat. isolates from Eucalyptus all formed one allele tree with high variation observed within the tree.

All isolates from Eucalyptus had a similar greenish olivaceous (33′′′f) (Rayner 1970) colony colour. The cultures had a banana odour similar to that of many Ceratocystis species.

The cultures all grew optimally at 30 °C. No clear morphological differences could be observed between isolates from different countries (Table 3).

Table 3. Morphological comparison of two representative isolates from Indonesia, South Africa, Thailand, and Uruguay. Ten measurements were taken of each structure and the (minimum-) average minus standard deviation – average plus standard deviation and (-maximum) given below.

Characteristic / Country	Indonesia	South Africa	Thailand	Uruguay
Ascomatal bases
Shape	Globose	Globose	Globose	Globose
Length	(125−)162–199(−200)	(120−)142–190(−202)	(188−)190–197(−200)	(144−)170–197(−200)
Width	(143−)173–193(−200)	(132−)143–193(−216)	(154−)177–199(−212)	(141−)164–184(−197)
Ascomatal necks
Length	(390−)400–450(−470)	(372−)392–460(−486)	(354−)370–400(−424)	(354−)368–386(−409)
Width (bases)	(24−)25–35(−40)	(24−)25–35(−42)	(24−)25–35(−39)	(23−)26–32(−38)
Width (apices)	(15−)16–18(−20)	(15−)16–20(−22)	(16−)17–19(−20)	(15−)16–22(−25)
Ostiolar hyphae
Shape	Divergent	Divergent	Divergent	Divergent
Length	(36−)43–53(−63)	(39−)40–52(−62)	(33−)35–39(−41)	(38−)41–51(−53)
Ascospores
Length	3–5	3–5	3–4	3–4
Width (excluding sheath)	4–6	4–6	4–6	4–6
Width (including sheath)	5–8	5–7(−8)	5–7	6–7
Primary phialides
Length	(69−)70–100(−134)	(73−)76–114(−131)	(67−)76–96(−100)	(73−)75–83(−88)
Width (bases)	4–6	4–6	4–6	2–4
Width (broadest point)	4–6	4–6	6–8	4–5
Width (apices)	3–5	3–5	3–5	3–4
Secondary phialides
Length	(60−)70–100(−143)	(64−)69–109(−143)	(63−)68–77(−99)	(69−)72–96(−109)
Width (bases)	3–6	3–6	5–6	3–6
Width (apices)	5–7	5–7	4–8	6–8
Primary conidia
Length	(13−)19–20(−24)	(15−)18–24(−25)	(10−)13–17(−18)	(10−)11–15(−18)
Width	4–5	4–5	3–4	2–3
Secondary conidia
Length	6–8	6–8	6–8	(7−)9–11
Width	5–8	5–7	5–8	6–8
Chlamydospores
Shape	Globose/Subglobose	Globose/Subglobose	Globose/Subglobose	Globose/Subglobose
Length	10–15	10–13	12–15	(6−)7–11(−13)
Width	8–13	8–10	10–13	(5−)7–11(−12)

Isolate CMW 11536 from Eucalyptus in South Africa was chosen to represent the global collection of isolates obtained from Eucalyptus. Three additional isolates (CMW 9998, CMW 10000 and CMW 15054), also from South Africa, were chosen as additional specimens for description. Cultures of these isolates were grown on 2 % MEA, dried down and have been deposited with the National Collection of Fungi (PREM), Pretoria, South Africa. Living cultures are maintained in the culture collection (CMW) of the Forestry and Agricultural Biotechnology Institute (FABI) at the University of Pretoria, South Africa and the Centraalbureau voor Schimmelcultures (CBS) in Utrecht, The Netherlands.

Where growth in culture was characterised based on the average colony diameter (from the five inoculated plates) for the four selected Eucalyptus isolates from South Africa, after 7 d, limited growth was observed at 4 °C (8 mm), 10 °C (7 mm), 15 °C (19 mm) and 35 °C (10 mm). Intermediate growth was observed after 7 d at 20 °C (34 mm) and 25 °C (35 mm), while the optimum temperature for growth in culture was 30 °C at which isolates reached an average of 39 mm diam after 7 d.

TAXONOMY

Isolates of the Ceratocystis from Eucalyptus, originating from many different countries, were phylogenetically distinct from all other Ceratocystis species residing in the C. fimbriata s. lat. clade. They also formed distinct phylogenetic groups based on geographic origin and might be found to represent distinct taxa in the future. For the present, those isolates from South Africa, which also had a morphology different to all described species from Eucalyptus (Table 4) are described as representing a novel taxon.

Table 4. Morphological comparison of previously described species in the C. fimbriata s. lat. species complex obtained from Eucalyptus trees compared to C. eucalypticola.

Character / Species	C. atrox	C. eucalypticola	C. fimbriatomima	C. neglecta	C. colombiana	C. pirilliformis
Ascomatal bases
Shape	Globose	Globose	Globose	Globose	Globose	Obpyriform
Length	(120−)140–180 (−222)	(105−)140–186 (−222)	(142−)173–215 (−234)	(173−)202–244 (−281)	(140−)177–237 (−294)	145–216(−279)
Width	(120−)150–178 (−200)	(118−)146–184 (−216)	(145−)178–225 (−255)	(153−)178–228 (−250)	(140−)177–237 (−294)	115–186(−206)
Ascomatal necks
Length	(270−)310–400 (−460)	(274−)376–464 (−499)	(446−)660–890 (−1070)	(691−)745–840 (−889)	(375−)448–560 (−676)	372–683(−778)
Width (bases)	(21−)26–34(−40)	(19−)25–33(−42)	(28−)32–42(−47)	(27−)31–39(−46)	(24−)27–35(−43)	18–33(−40)
Width (apices)	(13−)14–16(−19)	(14−)16–20(−22)	(16−)18–24(−28)	(14−)16–20(−22)	(12−)14–18(−19)	12–21(−25)
Ostiolar hyphae
Shape	Divergent	Divergent	Divergent	Divergent	Divergent	Convergent
Length	(18−)20–26(−28)	(39−)45–59(−66)	(40−)49–61(−68)	(35−)41–49(−54)	(28−)38–46(−52)	N/A
Ascospores
Length	3–4	3–5	2–4	3–6	3–4	4–6
Width (excluding sheath)	3–4	4–6	4–6	4–7	(3−)4–6(−7)	3–5
Width (including sheath)	4–6	5–7(−)8	5–7	5–8	6–8(−11)	3–5
Primary phialides
Length	(78−)87–151(−218)	(58−)77–113(−131)	(49−)60–94(−122)	(75−)80–114(−152)	(58−)65–83(−106)	62–147(−216)
Width (bases)	5–7(−13)	(3−)4–6(−7)	4–7	(4−)5–7(−8)	4–6(−8)	N/A
Width (broadest point)	4–7	4–6(−7)	5–9	5–9	(3−)6–8(−9)	N/A
Width (apices)	4–9	3–5	3–5	(3−)4–6(−7)	3–5(−6)	N/A
Secondary phialides
Length	(39−)43–57(−66)	(43−)60–100(−143)	Absent	(38−)48–76(−89)	(42−)49–71(−85)	N/A
Width (bases)	5–7(−9)	(3−)4–6(−7)	Absent	(3−)5–7(−8)	(4−)5–7	N/A
Width (apices)	4–6)–7)	(4−)5–7(−8)	Absent	(3−)5–7(−8)	(5−)6–8	N/A
Primary conidia
Length	(9−)11–15(−17)	(14−)16–22(−25)	(14−)20–28(−31)	(11−)15–27(−30)	(12−)16–24(−29)	12–25(−33)
Width	3–5	3–5	3–5	(3−)5–6	4–6	2–5
Secondary conidia
Length	(7−)8–12(−14)	(6−)7–9(−12)	Absent	(6−)10–11	9–14	4–6
Width	(5−)6–8(−9)	4–6(−7)	Absent	(4−)5–7(−9)	6–8(−11)	3–5
Chlamydospores
Shape	Absent	Globose/Subglobose	Subglobose	Globose	Globose	Oval
Length	Absent	(10−)11–13(−15)	(6−)10–14(−15)	(8−)10–12(−13)	11–14	8–12(−13)
Width	Absent	8–10(−11)	(6−)7–11(−12)	(9−)10–14(−16)	11–15(−17)	5–8(−10)
Reference	Van Wyk et. al. 2007	This study	Van Wyk et. al. 2008	Rodas et. al. 2008	Van Wyk et. al. 2010a	Barnes et. al. 2003

Ceratocystis eucalypticola M. van Wyk & M.J. Wingf., sp. nov.

MycoBank MB512397

(Fig. 5)

Fig. 5.

Morphological characteristics of Ceratocystis eucalypticola. a. Ascomata with globose base. b. Hat-shaped (in side view) and cucullate (in top view) ascospores. c. Divergent ostiolar hyphae d. Dark, globose to sub-globose chlamydospore. e. Primary conidiophore, flask-shaped phialide, producing cylindrical conidia. f. Tubular shaped secondary conidiophore, producing a chain of barrel-shaped conidia. g. Chain of cylindrical conidia. h. Chain of barrel-shaped conidia. i. A chain of barrel-shaped conidia, two hat-shaped ascospores and a cylindrical conidium. Bars: a. = 100 μm, b, f–i = 5 μm, c–e = 10 μm.

Etymology: The name refers to Eucalyptus on which the fungus occurs.

All species of Ceratocystis from Eucalyptus are phylogenetically distinct. Colonies of C. eucalypticola are typically green colonies, relatively slow growing, and have a fruity banana odour.

Type: South Africa: Kwa-Zulu Natal: KwaMbonambi, isolated from artificially wounded Eucalyptus, 15 Dec. 2002, M. van Wyk & J. Roux (PREM 60168 – holotype; cultures ex-holotype CMW 11536 = CBS 124016)

Description: Ascomatal bases dark brown to black, globose, un-ornamented (105−)140–186(−222) μm wide, (118−)146–184(−216) μm high. Ascomatal necks dark brown to black at bases becoming lighter towards the apices, (274−)376–464(−499) μm long, apices (14−)16–20(−22) μm wide, bases (19−)25–33(−42) μm wide. Ostiolar hyphae divergent, (39−)45–59(−66) μm long. Ascospores hyaline, hat-shaped in side view, invested in sheath, 3–5 μm long, 4–6 μm wide without sheath, 5–7(−8) μm wide including sheath. Anamorph thielaviopsis-like, conidiophores of two types: Primary conidiophores phialidic, flask-shaped, (58−)77–113(−131) μm long, (3−)4–6 μm wide at the bases, 4–6(−7) μm wide at broadest points and 3–5 μm wide at apices. Secondary conidiophores flaring or wide mouthed, (43−)60–100(−143) μm long, (3−)4–6(−7) μm wide at bases and (4−)5–7(−8) μm wide at apices. Primary conidia cylindrical in shape (14−)16–22(−25) μm long, 3–5 μm wide. Secondary conidia, barrel-shaped, abundant, (6−)7–9(−12) μm long, 4–6(−7) μm wide. Chlamydospores, scarce, hair brown (17′′′′i), globose to sub-globose (10−)11–13(−15) μm long, 8–10(−11) μm wide.

Habitat: Wounded and diseased Eucalyptus.

Known distribution: South Africa.

Other material examined: South Africa: Mpumalanga, Sabie, isolated from artificially wounded Eucalyptus trees, 14 July 2002, M. van Wyk & J. Roux (PREM 60169; living cultures CMW 9998 = CBS 124017); loc. cit., isolated from artificially wounded Eucalyptus trees, 14 July 2002, M. van Wyk & J. Roux (PREM 60170; living cultures CMW 10000 = CBS 124019).

DISCUSSION

Isolates of Ceratocystis fimbriata s. lat. collected from Eucalyptus in Brazil, Indonesia, Republic of Congo, South Africa, Thailand, Uganda, and Uruguay were shown to be phylogenetically related. These included isolates taken from wounds on trees and also those that were associated with trees dying as result of infection by the fungus. Although all isolates from Eucalyptus resided in a single large clade, there was a high degree of diversity among them. It is thus possible that they represent a number of different cryptic species that cannot be resolved. For the present, those isolates from South Africa are provided with the name C. eucalypticola here. Future studies should seek to include additional isolates from Eucalyptus as well as to include sequences for gene regions not considered in this study, and that might discriminate more clearly between species in the C. fimbriata s. lat. complex. Currently, the group is unified based on a specific host and relatively strong phylogenetic similarity. In this respect, it also provides the foundation for further studies including a suite of isolates that would be difficult to obtain.

The species of Ceratocystis most closely related to C. eucalypticola is C. colombiana. Ceratocystis colombiana is a pathogen of coffee trees (Marin ) as well as numerous other hosts including indigenous crops in Colombia. Although the two species are phylogenetically related, they are ecologically distinct and are not likely to be confused.

Ceratocystis eucalypticola is one of a number of species in the C. fimbriata s. lat. complex to be described from Eucalyptus trees. Other species from this host include; C. atrox (van Wyk ) and C. corymbiicola (Kamgan Nkuekam ) from Australia, C. pirilliformis (Barnes ) from Australia and South Africa, C. neglecta (Rodas ) from Colombia, C. fimbriatomima (van Wyk ) from Venezuela, and C. zombamontana (Heath ) from Malawi. All of these species from Eucalyptus can be distinguished from each other based on phylogenetic inference and they have some morphological features that can be used to recognise them.

Morphologically, the specimens of C. eucalypticola cited here resemble species in the C. fimbriata s. lat. complex. The fungus has the typical green colony colour, is relatively slow growing, and has a fruity banana odour. Ceratocystis eucalypticola can be distinguished from other species in the C. fimbriata s. lat. complex in that they occur on Eucalyptus and based on differences in size of some diagnostic characters for this group of fungi.

Ceratocystis eucalypticola includes isolates only from wounds on trees in South Africa in the absence of disease, but is very closely related to isolates that originated from dying trees and that have been shown to be pathogenic (Laia ; Roux , 2001, 2004). The species is also closely related to isolates that were collected from wounds on trees in countries other than South Africa where a Ceratocystis disease on Eucalyptus has not been seen. Eucalyptus death associated with C. eucalypticola has never been found in South Africa although trees dying of unknown causes are thought to have died due to infection by this fungus, which can be difficult to isolate. The fungus collected from wounds on trees has also been shown to be pathogenic in greenhouse inoculation trials (Roux , van Wyk ).

Isolates of C. eucalypticola from South Africa represent a clonal population (van Wyk ) and it was most likely introduced into the country. It is thus intriguing that Eucalyptus death associated with this fungus has not been seen. This might be due to planting stock susceptible to C. eucalypticola not having occurred in the country, or that conditions for infection were not suitable. Alternatively, it is possible that trees dying of unexplained causes might have been killed by C. eucalypticola, even though the fungus was not isolated from them. This is a question that is currently being pursued, particularly linked to unexplained Eucalyptus death in South Africa and where Ceratocystis cultures emerge from isolations.