Four new Geosmithia species from bark beetles infesting indigenous South African trees
Abstract
Over the past two decades, numerous Geosmithia fungi have been isolated from the bodies and galleries of wood–boring beetles. However, this genus of asexual Sordariomycetes remains taxonomically and ecologically understudied, especially in the Southern Hemisphere. In South Africa, two prior surveys reported Geosmithia species from bark beetles, but neither thoroughly investigated species identities. In this study, we collected bark beetles from native trees in the Western Cape Province of South Africa and isolated, identified and described their associated Geosmithia species. Geosmithia spp. previously collected in South Africa were also re-examined. The ITS sequences of Geosmithia isolates from 13 beetle taxa infesting 10 host species were considered. Additional gene regions, BT, EF1a and RPB2, were sequenced for a subset of isolates. Four previously described species, G. flava, G. langdonii, G. omnicola and G. pumila were identified by phylogenetic analyses. Additionally, four novel taxa were identified and are here described as G. capensis, G. multisociorum, G. oroboidis and G. stellenboschiana. Geosmithia multisociorum appears to be a generalist associated with multiple beetle–host combinations in the Northern and Southern Hemisphere, whereas G. oroboidis is currently known only from a single beetle and tree host in South Africa. South African isolates of G. capensis and G. stellenboschiana appeared to be restricted to Lanurgus spp. and Hypothenemus sp. beetles, respectively, but both species are also known from beetles and hosts in other countries.
1. Introduction
Geosmithia (Sordariomycetes: Hypocreales) is an ecologically and taxonomically understudied genus of globally distributed filamentous fungi (Kolařík and Hulcr, 2023). The genus was historically considered part of Penicillium (Eurotiomycetes: Eurotiales), due to the similarities of their conidiophores (Pitt, 1979). Species are asexual, characterised by hydrophobic spores and conidia produced in long chains (Kolarík et al., 2004, 2005). The dry spores of Geosmithia species enable airborne dispersal and represent an anomaly for a genus commonly isolated from wood–boring beetles and their galleries. Their apparent lack of entomochoric adaptations implies that these fungi do not rely on insects for dispersal. However, the consistency with which certain assemblages of Geosmithia species are isolated from specific beetle populations suggests a strong link (Kolařík and Jankowiak, 2013; Kolařík and Kirkendall, 2010; Kolařík et al., 2008; Kubátová et al., 2004).
Fungal-bark beetle symbioses with detrimental impacts on tree health are well known. These include the well-studied bark beetle symbioses with ophiostomatoid fungi in the Orders Ophiostomatales and Microascales, which are responsible for diseases such as Dutch Elm Disease (De Beer et al., 2022; Santini and Faccoli, 2015) and Ceratocystis wilt of various hosts (De Beer et al., 2014; Roux and Wingfield, 2013) respectively. In contrast, only one detrimental Geosmithia-beetle symbiosis is known; that of the walnut twig beetle, Pityophthorus juglandis and G. morbida causing Thousand Cankers Disease of black walnut, Juglans nigra (Kolařík et al., 2011; Tisserat et al., 2009). Other than a small number of Geosmithia species that have been described as primary and auxiliary associates of ambrosia beetles (Kolařík et al., 2015; Kolařík and Kirkendall, 2010), the role that most Geosmithia species play in the nutrition and other aspects of the ecology of their beetle associates remains unclear (Kolařík and Hulcr, 2023).
Numerous studies have considered Geosmithia species in the Northern Hemisphere (Kolařík and Hulcr, 2023) and several of these have included extensive surveys (e.g. Huang et al., 2019; Kolařík et al., 2017; Kolařík and Jankowiak, 2013; Kolařík et al., 2007, 2008). In contrast, knowledge of these fungi and their vectors in the Southern Hemisphere is limited. Single records are available from Australia (Kolařík and Hulcr, 2023; Pitt, 1979; Sakalidis et al., 2011), Brazil (Crous et al., 2018), New Zealand, Peru, and the Seychelles (Kolarík et al., 2004). Of these, only the Seychelles isolate was from a bark beetle. In South Africa, a single targeted survey identified five Geosmithia species from four different bark beetles infesting unhealthy or dead Virgilia trees (Machingambi et al., 2014). A recent South African survey of fungi from bark beetles in the genus Lanurgus (Coleoptera: Scolytinae) infesting Widdringtonia trees yielded numerous Geosmithia isolates representing different taxa (Basson et al., 2024), but these were not identified at the species level.
Given the number of Geosmithia taxa that have been isolated in surveys of only two native tree genera in South Africa (Basson et al., 2024; Machingambi et al., 2014), it seems likely that a great diversity of Geosmithia species has yet to be discovered in the country. Furthermore, the rate of Geosmithia species discovery, both in the Northern Hemisphere and in South African surveys has surpassed species descriptions. According to Kolařík and Hulcr (2023), at least 99 Geosmithia species have been documented, but only 32 have been described. The remaining ≥67 species are recognised phylogenetically with only a numerical identity. This study was prompted by the unstudied diversity and undescribed species of Geosmithia in South Africa as well as the need for species descriptions in this genus. The overall aim was to isolate, identify, and describe Geosmithia species from native bark beetle-infested trees in the Western Cape Province of South Africa, and also to re-examine the isolates collected by Basson et al. (2024) and Machingambi et al. (2014).
2. Materials and methods
2.1. Sampling and fungal isolation
Geosmithia isolates previously collected from beetles on two Widdringtonia (Basson et al., 2024) and three Virgilia species (Machingambi et al., 2014) were obtained from the culture collection (CMW) of the Forestry and Agricultural Biotechnology Institute (FABI), University of Pretoria, South Africa. Bark beetle-infested tissues of five additional native tree species, representing five different families, were also collected (Table 1). Collection of bark beetles and isolation of Geosmithia species was performed as described by Basson et al. (2024). Briefly, plant material was placed in emergence cages, from which emerging bark beetles were collected. Additionally, the bark was removed from plant material to expose and directly remove beetles from their galleries. A subset of the collected beetles was stored in 98 % ethanol at −80 °C for identification and reference purposes. Beetle identifications followed those presented in the previous publications (Basson et al., 2024; Machingambi et al., 2014). Those collected from additional hosts in this study were identified by J. Hulcr, School of Forest, Fisheries, and Geomatics Sciences, University of Florida.
Table 1. Species of South African native host trees and beetles from which the Geosmithia isolates considered in this study were obtained.
Betty’s Bay; Kirstenbosch Botanical Garden;
Silvermine Nature Reserve, Cape Town
34°20′51.0″S
5
“
18°55′29.4″E;
33°59′11.3″S
18°25′34.4″E;
34°05′27.9″S
18°25′17.6″E
Fungi were isolated from living bark beetles, by (i) rolling individual beetles across the surface of 2 % Malt Extract Agar (MEA, Biolab, South Africa) or (ii) shaking a single beetle in 100 μl sterile deionised water before spreading the solution onto Potato Dextrose Agar (PDA, Biolab, South Africa). After incubation at room temperature, fungal colonies resembling Geosmithia were purified by transferring hyphal tips to clean agar plates. Cultures were subsequently assigned to phenotypic groups based on texture, colour and pattern.
2.2. Fungal identification
Representative isolates from each Geosmithia phenotypic group, representing all 10 tree hosts considered in this study, were used for species identification. DNA was isolated from young (3-5-day old) fungal cultures using the Prepman Ultra Sample Preparation Reagent (Thermo Fisher Scientific, MA, USA) and the ITS region was amplified with primers ITS1F (Gardes and Bruns, 1993) and ITS4 (White et al., 1990). The 20 μl PCR reaction contained 10 μl Ampliqon Taq DNA Polymerase Master Mix RED (Biomol, Germany), 0.4 μM of each primer, and 1.5 μl of the Prepman DNA extraction. Reaction conditions were 5 min initial denaturation at 95 °C, 35 cycles of 30 s denaturation at 95 °C, 30 s annealing at 55 °C and 90 s elongation at 72 °C. This was followed by a final elongation step of 10 min at 72 °C. Amplicons were sequenced at the Central Analytical Facilities (CAF), Stellenbosch University, South Africa.
Incorporating Geosmithia reference sequences obtained from GenBank (Table S1), a Maximum Likelihood (ML) ITS tree was computed using MAFFT v7.525 (Katoh and Standley, 2013) for sequence alignment and trimAl v1.4. rev22 for alignment curation (Capella-Gutiérrez et al., 2009). Appropriate substitution models were identified with Modeltest-NG 0.1.7 (Darriba et al., 2020) and the phylogeny was determined with RAxML-NG 1.2.2 (Kozlov et al., 2019), applying 1000 bootstrap replicates. Based on the ITS phylogeny, additional gene regions were sequenced for a subset of isolates, using the same PCR mixture as for ITS and adjusting only the annealing temperature in the PCR reaction. These were Beta-tubulin (BT) with primers T1/Bt2b (Glass and Donaldson, 1995; O’Donnell and Cigelnik, 1997) at Ta = 53 °C, the Elongation Factor (EF)1a large intron with primers EF1-728 F/EF2 (Carbone and Kohn, 1999) at Ta = 53 °C, the EF1a large exon with primers EF1-983 F/EF1-2218 R (Rehner and Buckley, 2005) at Ta = 55 °C and RNA Polymerase II (RPB2) with primers RPB2-5F2/fRPB2-7cR (Liu et al., 1999; Sung et al., 2007) at Ta = 57 °C.
ML phylogenies for individual regions and the partitioned concatenated dataset were constructed as described above. According to the AIC, nucleotide substitution models GTR + I + G4, TPM1uf + I + G4, TIM3+I + G4 and GTR + G4 were applied to the ITS, BT, EF1a exon and RPB2 partitions, respectively. Bayesian Inference was performed on the cancatenated dataset in MrBayes 3.2.7a (Ronquist et al., 2012) from a random starting tree, sampling across GTR model space (nst = mixed) for each of the four partitions. Two MCMC runs of four chains were run for 10 million generations. Runs were sampled every 5000 generations and the default 25 % was discarded as burn-in. The GTR substitution models with the highest posterior probabilities for the ITS, BT, EF1a exon and RPB2 partitions were M134 = 123141 (PP = 0.304), M45 = 111231 (PP = 0.329), M36 = 122232 (PP = 0.175) and M15 = 121121 (Kimura, 1980; PP = 0.304).
2.3. Morphological examinations
Isolates grown on 2 % MEA and incubated at optimum temperature for 14 days were used to study morphological characteristics. Fruiting structures were mounted in water, which was replaced with 85 % lactic acid for further observation. Microscopes (Nikon Eclipse Ni, SMZ18), mounted with a Nikon DS-Ri2 camera, and imaging software (Nikon NIS-Elements) were used to study the structures. Twenty-five structures were measured for the conidiogenous apparatus: penicilli (between stipes and conidia), stipes, rami (first branch), metulae (branch-bearing phialides) and conidiogenous cells (phialides), 50 structures were measured for conidia, and 10 structures were counted for the number of whorls. The dimensions of the structures were presented in min–max (average ± SD).
For the growth study, mycelial plugs (5 mm diam.) were obtained from 7-day-old cultures. The plugs were placed at the centre of 90 mm Petri dishes containing 2 % MEA. Three replicates per strain were kept in incubators with temperatures ranging from 5 to 35 °C, in 5 °C intervals. For each Petri dish, two diameters were measured at 8-, 14- and 28-days incubation or just before mycelia growing at the optimum temperature reached the edge of the Petri dish. Average daily growth was calculated for each temperature. The plates were left in incubators for 28 days for characterization. Strains kept at 5 °C and 35 °C that did not show any growth after 28 days were re-incubated at their optimal temperatures for another 14 days to assess viability. Colours were described according to the ISCC-NBS system. Type specimens were lodged at the H.G.W.J. Schweickerdt herbarium (PRU) and ex-type cultures were deposited at the culture collection (CMW-IA) of Innovation Africa in the University of Pretoria, Hatfield, Pretoria, South Africa.
3. Results
3.1. Beetle identification and Geosmithia isolations
Each of the five native tree hosts collected in this study yielded a single bark beetle species from which fungal isolations were made (Table 1). The beetles included a Ctonoxylon sp., a Hypothenemus sp., two species of Lanurgus and Xyloctonus maculatus. The Geosmithia isolates from the studies of Basson et al. (2024) and Machingambi et al. (2014) represented an additional eight bark beetle species, three on Cupressaceae and five on Fabaceae hosts. No overlap in beetle species was found among the different genera of hosts, although the same beetle taxa infested the three Virgilia species investigated by Machingambi et al. (2014).
Eight Geosmithia species were identified (Table 2), of which seven were isolated from the five tree hosts collected during this study. Geosmithia sp. A (Machingambi et al., 2014), described here as G. oroboidis, was the only previously collected species that was not re-isolated. The greatest diversity of Geosmithia species was encountered on the Cryphalini sp. 1 beetle from Virgilia oroboides subsp. oroboides and Hypothenemus sp. collected from Searsia angustifolia. Both yielded four Geosmithia taxa from a single beetle species (Table 1). A maximum of two Geosmithia taxa were isolated from each of the remaining beetle species. Of the eight species identified, only G. oroboidis was seemingly beetle- and host-specific. Six others were isolated from at least two different beetle species and host trees in this study, whereas, G. stellenboschiana has also been found outside of South Africa.
Table 2. Identity and metadata of the Geosmithia strains isolated in this study.
BB = Betty’s Bay; CB = Cederberg; FP = Franschhoek Pass; GE = George; KB = Kalk Bay, Cape Town; KBG = Kirstenbosch Botanical Garden; SMNR = Silver Mine Nature Reserve, Cape Town; SB = Stellenbosch; SR = Storm’s River. See Table 1 for more information.
b
Asterisks (∗) indicate ITS sequences generated by Machingambi et al. (2014). All other sequences were generated in this study.
3.2. Phylogenetic analyses
The ITS phylogeny (Fig. S1A) grouped the Geosmithia isolates into seven clades that were further explored. As previously noted (Kolařík and Hulcr, 2023), phylogenetic resolution of closely related Geosmithia species with ITS data alone was poor, but most relationships were resolved by considering sequence data for the additional gene regions. The individual gene trees (Figs. S1B–E) and the concatenated phylogeny (Fig. 1) identified three previously described species, G. langdonii, G. omnicola, and G. pumila as well as isolates belonging to the G. microcorthyli species complex. Four novel species were resolved. Of these, G. capensis and G. oroboidis, were sufficiently distinct from other taxa to be resolved using only ITS sequence data, whereas G. multisociorum and G. stellenboschiana required data for additional gene regions to delineate the species boundaries.
Fig. 1. Maximum Likelihood phylogeny of the concatenated ITS, BT, EF1a exon and RPB2 regions of a subset of South African isolates (blue). Clades containing species considered in this study are highlighted and labelled C1-C7, according to the clades on the phylogenies of individual regions (Fig. S1). Geosmithia type strains (T), including those assigned as types in this study, are shown in bold. Isolates of Machingambi et al. (2014) are labelled according to their original identifications. Branch support values before the slash represent Transfer Bootstrap Expectation (TBE) support values ≥ 70 %. Bayesian Inference posterior probabilities (PP) above 0.9 are indicated on main branches after the slash, with asterisks denoting a PP of 1.0. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)
None of the phylogenies could confidently distinguish between G. microcorthyli and the two phylogenetic species, Geosmithia sp. 8 and Geosmithia sp. 48. Investigation of the individual alignments did not reveal any differences between Geosmithia sp. 8 and Geosmithia sp. 48 in the ITS or EF1a exon regions available for the latter species. Compared to G. microcorthyli, the portions of the ITS and RPB2 regions that were investigated were not variable among reference sequences. Polymorphism was detected at one position in the BT alignment and three positions in the EF1a exon alignment. None of the South African isolates nor other Northern Hemisphere isolates shared these polymorphisms. We, therefore, conclude that the South African isolates in this clade are distinct from G. microcorthyli and likely represent Geosmithia sp. 8, which may be synonymous with Geosmithia sp. 48.
For the concatenated phylogenetic tree, bootstrapping converged after 400 replicates. The LogLikelihood of the final ML tree was −16499.706459. The Bayesian analysis reached a maximum standard deviation of split frequencies <0.01, with an average Estimated Sample Size <1000 and a Potential Scale Reduction Factor value of 1.00, indicating convergence of the runs.
Fig. 2. Micrographs of Geosmithia capensis sp. nov (ex-holotype, CMW-IA 6985, CMW 64002). A–C. Conidiophores bearing chained conidia, showing pigmented exudates. D. Conidiophores borne on aerial funiculose mycelia. E–I. Conidiophores and penicilli. J. Conidiogenous cells (phialides). K. Conidia. L. Cultures on 2 % MEA at different temperatures for 14 days (left, white background) and 28 days (right, black background). Scale bars: B–D = 50 μm; A = 25 μm; G–K = 10 μm, E, F = 5 μm.
MycoBank: MB855500.
Etymology: “Capensis” (from the cape), refers to its common occurrence in the Western Cape Province of South Africa.
Diagnosis: Geosmithiacapensis is phylogenetically distinct from other known species based on the ITS or any single gene region.
Type: South Africa: Western Cape Province, Stellenbosch, banks of the Eerste River. Isolated from Lanurgus jubatus infesting Olea europaea subsp. africana. June 2023. W. Rippon, WR01 (Holotype PRU(M) 4605, stored in a metabolically inactive state; ex-holotype culture CMW-IA 6985, CMW 64002). GenBank: PQ032691 (ITS); PQ048054 (BT); PQ045591 (RPB2).
Description: Sexual morph not observed. Asexual morph penicillium-like. Conidiophores upright from vegetative hyphae on surface of media or on funiculose aerial mycelia, simple to branched; stipes hyaline to pale yellow with age, verruculose, simple or often branched irregularly, 22–174 (81.8 ± 22.44) μm long; penicilli terminal, appressed or somewhat loosely packed, getting pigmented with age, branched in 1–4 tiers including conidiogenous cells, 20–55 (36.9 ± 10.16) μm long; rami hyaline to pale yellow with age, verruculose, 11–20 (15.5 ± 2.33) μm long, in whorls of 2–4; metulae verruculose, 7–14 (10.4 ± 1.71) μm long, in whorls of 3–5. Conidiogenous cells phialidic, cylindrical, narrowed at apex, smooth or verruculose, 8–11 (9.2 ± 0.95) μm long, in whorls of 4–6. Conidia hyaline, aseptate, mostly cylindrical with round apex and truncate base, straight or occasionally slightly curved, occasionally ellipsoidal, 4–6 × 1.5–3 (4.8 ± 0.40 × 1.9 ± 0.26) μm, in basipetal chain. Funiculose aerial mycelia becoming yellowish with age. Pigmented exudate droplets on penicilli and stipes.
Culture characteristics: Optimum growth on 2 % MEA in 8 d at 25 °C reaching 1.52 mm/d, followed by 20 °C (1.36 mm/d), 15 °C (0.82 mm/d), 30 °C (0.65 mm/d) and 10 °C (0.23 mm/d). No growth at 5 °C and 35 °C; 5 °C cultures and two out of three 35 °C cultures grew back after re-incubated at 25 °C. Cultures (28 day-old) fertile; shape circular; margin entire (10–25 °C), lobate (30 °C); elevation flat (20–30 °C) to raised (10–15 °C); texture fluffy (10–15 °C), velvety becoming fluffy towards centre with patches of yellowish white aerial mycelia (20 °C), velvety with zonation or sectoring (25–30 °C); colour above yellowish white becoming paler towards edges (10 °C), dark orange yellow becoming paler towards edges (15 °C), strong yellowish brown becoming paler towards edges (20 °C), light yellowish brown becoming paler towards edges (25 °C), yellowish white with light yellowish brown sectors (30 °C); density medium (10–30 °C); pigmentation on media vivid yellow to brilliant yellow (15–30 °C).
Notes: Geosmithiacapensis is described from isolates in the Western Cape Province of South Africa but forms a monophyletic clade (clade C3) with two Hungarian isolates of “Geosmithia sp. 11” that were well-supported (≥96 %) in all individual trees. The two isolates of “Geosmithia sp. 11” were reported from Scolytusintricatus beetles on Quercus spp. (Kolařík et al., 2008). An isolate collected from Adansoniigregorii bark in the Kimberley region of Western Australia also resolved within this clade using ITS data (Fig. S1A; Sakalidis et al., 2011). Additional Northern Hemisphere isolates are not available, but this taxon reportedly also occurs in the Mediterranean (Kolařík et al., 2008) and southeastern USA (Huang et al., 2017).
Fig. 3. Micrographs of Geosmithia multisociorum sp. nov (ex-holotype, CMW-IA 6981, CMW 40739). A, B. Conidiophores born on funiculose mycelia. C. Single conidiophore with chained conidia. D. Simple conidiophores on surface of media. E. Verruculose conidiophore and conidial apparatus. F. Conidiogenous cells (phialides). G. Conidia. H. Conidiophores on aerial funiculose mycelia. I. Upright conidiophore. J. Conidiophore with stipe attenuating towards the base. K. Colony on 2 % MEA in 14 days (left, white background) and 28 days (right, black background). Scale bars: A = 500 μm; B = 100 μm; C, D, H = 50 μm; I, J = 10 μm; E–G = 5 μm.
MycoBank: MB855501.
Etymology: “Multisociorum”, referring to its wide range of tree hosts and beetle associates.
Diagnosis: Geosmithiamultisociorum is closely related to G. microcorthyli. They can be differentiated by a few SNPs in the BT (1) and EF1a exon (3) regions and by conidial shape: G. multisociorum (ellipsoidal to cylindrical) and G. microcorthyli (globose).
Type: South Africa: Western Cape Province, Betty’s Bay, Harold Porter National Botanical Garden. Isolated from Cryphalini sp. 1 infesting V. oroboides. 2012. N.M. Machingambi, NM93 (Holotype PRU(M) 4604, stored in a metabolically inactive state; ex-holotype culture CMW-IA 6981, CMW 40739). GenBank: KJ513226
(ITS); PQ048047 (BT); PQ048070 (EF1a exon).
Description: Sexual morph not observed. Asexual morph penicillium-like. Conidiophores upright borne on vegetative hyphae on surface of media or on funiculose aerial mycelia, simple or branched; stipes verruculose, straight or infrequently curved, occasionally attenuated towards base, 9–150 (60 ± 33.5) μm long; penicilli often loosely appressed, asymmetric, in 1–3 tiers including conidiogenous cells, 15.5–44 (26.8 ± 9.19) μm long; rami verruculose, 9–19 (12.3 ± 2.77) μm long, in whorls of 2–3; metulae verruculose, 7–12 (9.6 ± 1.16) μm long, whorls of 2–5. Conidiogenous cells phialidic, verruculose, 7–13 (9.1 ± 1.42) μm, in whorls of 3–6. Conidia hyaline, aseptate, ellipsoidal to cylindrical with round ends, smooth, 3–5 × 2–3 (3.66 ± 0.35 × 2.03 ± 0.23) μm, in basipetal chains.
Culture characteristics: Optimum growth on 2 % MEA in 8 d at 25 °C reaching 1.82 mm/d, followed by 20–30 °C (1.44 mm/d), 15 °C (0.9 mm/d), 10 °C (0.37 mm/d). No growth at 5 °C and 35 °C; 5 °C and 35 °C cultures grew back after re-incubation at 25 °C. Cultures (28 day-old) fertile; shape circular; margin entire (10–30 °C); elevation flat (10–30 °C); texture velvety to powdery (10–30 °C) with some fluffy sectors (15–25 °C); colour above yellowish white (10 °C), yellowish grey becoming greyish yellow towards with some sectors yellowish white (15–25 °C), yellowish white becoming paler towards edges (30 °C); density dense (10 °C), sparse to medium towards centre (15–30 °C); pigmentation on media absent.
Notes: Geosmithiamultisociorum was recovered from five different hosts and six different beetles across the Western Cape (Table 2). It is closely related to the primary ambrosial fungus G. microcorthyli, but conidial shape differentiates them. It was found to be con-specific with two Geosmithia spp. From the Northern Hemisphere, “Geosmithia sp. 8” from Scolytusintricatus infesting Quercus spp. in Europe (Kolařík et al., 2008) and “Geosmithia sp. 48” from Phloeosinusdentatus infesting Juniperusvirginiana in the USA (Huang et al., 2019).
Fig. 4. Micrographs of Geosmithia oroboidis sp. Nov (ex-holotype, CMW-IA 6980, CMW 40732) (A, B) Conidiophores and chained conidia directly borne on 2 % MEA (C) Conidiophore cluster (D) Verruculose conidiogenous cells (phialides) (E) Conidia (F) Conidiophore originating from an inflated basal cell (black arrow) and undulate vegetative hyphae (white arrow) (G) Upright conidiophore (H) Colonies on 2 % MEA at different temperatures at 14 days (left, white background) and 28 days (right, black background). Scale bars: A, B = 100 μm; C = 50 μm; D–G = 10 μm.
MycoBank: MB855502.
Etymology: Referring to its host tree, Virgiliaoroboides.
Diagnosis: Geosmithiaoroboidis is morphologically distinguishable from closely related taxa by its larger conidial dimensions (3–6 × 2–3 μm), the presence of soluble pigmentation at all temperatures and slower growth (18–26 mm/8 d, 20–30 mm/14 d) at optimum temperature: G. funiculosa (1.6–5 × 0.9–2.5 μm, 39–48 mm/14 d), G. pulverea (2.1–5.1 × 1.1–2.0 μm, 30–37 mm/8 d), G. pumila (1.5–2.9 × 0.9–1.9 μm, 25–33 mm/8 d) and G. subfulva (1.1–2.2 × 1.0–1.7 μm, 24–36 mm/8 d). Phylogenetically, it is distinguished from its closest relatives G. subfulva and G. pulverea with either ITS or EF1a exon sequences. Five SNPs in the ITS region and 19 SNPs and one indel in the EF1a exon are unique to G. oroboidis.
Type: South Africa: Western Cape Province, Betty’s Bay, Harold Porter National Botanical Garden; 34°20′51.0″S, 18°55′29.4″E. 2012. Isolated from Scolytoplatypus fasciatus infesting Virgiliaoroboides subsp. oroboides. N.M. Machingambi NM105 (Holotype PRU(M) 4603, stored in a metabolically inactive state; ex-holotype culture CMW-IA 6980, CMW 40732). GenBank: KJ533336
(ITS); PQ048068 (EF1a exon).
Description: Sexual morph not observed. Asexual morph penicillium-like. Conidiophores upright from vegetative hyphae mostly on surface of media, occasionally on inflated basal cells, verruculose, mostly simple; stipes verruculose, straight, 15–73 (45.6 ± 16.18) μm long; penicilli terminal, branched in 1–3 tiers including conidiogenous cells, 13–41 (27.8 ± 7.02) μm long, laterally appressed, asymmetric or symmetric; rami verruculose, 10–15 (12.7 ± 1.21) μm long, 2–4 in wholrs; metulae smooth or verruculose, 7–16 (10.1 ± 1.82) μm long, 2–5 in whorls. Conidiogenous cells phialidic, hyaline, smooth or verruculose, cylindrical with narrowed tip, 9–15 (11.1 ± 1.40) μm long, 3–7 in whorls. Conidia hyaline, pale yellow-shade in mass, cylindrical with round apex and pointed base to ellipsoidal, straight or slightly curved, 3–6 × 2–3 (4.9 ± 0.53 × 2.1 ± 0.23) μm, in basipetal chain. Some vegetative hyphae undulate. Funiculose mycelia scarce.
Culture characteristics: Optimum growth on 2 % MEA in 8 d at 25 °C reaching 1.05 mm/d, followed by 20 °C (0.6 mm/d), 30 °C (0.58 mm/d), 15 °C (0.54 mm/d) and 10 °C (0.39 mm/d). No growth at 5 °C and 35 °C; 5 °C cultures grew back when re-incubated at 25 °C but 35°C-cultures did not grow back. Daily growth rate declined by 76, 69, 80 % after 8 days at 20 °C, 25 °C and 30 °C, respectively. Cultures (28 day-old) fertile; shape circular; margin entire (10 °C), irregular with lobate (15–30 °C), no submerged hyphae at margins; elevation raised (10–30 °C); texture velutinous (20–30 °C) to fluffy (10–15 °C); colour above light yellowish brown (10 °C), yellowish grey with greyish yellowish brown edges (15 °C), yellowish grey with dark greyish yellow edges or patches (20 °C), greyish yellow (25 °C), yellowish grey (30 °C); density dense becoming furrowed (30 °C); pigmentation on media deep orange yellow (15–30 °C), inconspicuous.
Host: Ellatoma sp. on Virgiliaoroboides subsp. ferruginea; Scolytoplatypusfasciatus infesting V. oroboides subsp. oroboides.
Distribution: Western Cape Province, South Africa.
Additional specimen examined: South Africa: Western Cape Province, Betty’s Bay, Harold Porter Botanical Garden; −34.333581, 18.916811.2011. Isolated from Scolytoplatypusfasciatus infesting Virgiliaoroboides. N.M. Machingambi, NM005 (culture CMW-IA 6982, CMW 40742). GenBank: KJ533338
(ITS); PQ048071 (EF1a exon).
Notes: Geosmithiaoroboidis was isolated by Machingambi et al. (2014) as “Geosmithia sp. A″ from Scolytoplatypusfasciatus on V. oroboides subsp. oroboides in the Harold Porter National Botanic Garden, Betty’s Bay, South Africa. Additional collections were reported from an Ellatoma sp. on V. oroboides subsp. ferruginea in George. This taxon is distinct from all previously described Geosmithia species, being resolved with 100 % support with ITS and EF1a exon sequences (clade C1 in Fig. 1 and S1). Geosmithiaoroboidis showed a close affinity to G. funiculosa (Crous et al., 2022), G. pulverea, G. pumila, and G. subfulva (Zhang et al., 2022) in both morphological and phylogenetic analyses. Their cultures are mostly velvety and have yellow to brown shades. Geosmithiafuniculosa is widely distributed in Europe on various bark beetle species, whereas the other three species were collected in China from the galleries of Scolytus semenovi, Dinoderus sp., and Ernoporus japonicus, respectively. The larger conidial dimensions (3–6 × 2–3 μm), no growth at 5 °C or 35 °C, and the presence of pigmentation on 2 % MEA differentiates G. oroboidis from its close phylogenetic relatives. The CMW-IA 6982 (CMW 40742) culture showed less pigmentation and fruiting structures than the ex-holotype (CMW-IA 6980, CMW 40732).
Fig. 5. Micrographs of Geosmithia stellenboschiana sp. nov (ex-holotype, CMW-IA 6987, CMW 64012). A. B. Conidiophores with chained conidia borne on aerial funiculose mycelia or single hypha. C. Conidiogenous cells (phialides). D. Conidia. E, F. Verruculose conidiophores. G. Culture on 2 % MEA at different temperatures for 14 days (left, white background) and 28 days (right, black background). Scale bars: A, B = 100 μ; C, E, F = 10 μm; D = 5 μm.
MycoBank: MB855503.
Etymology: Referring to the town of Stellenbosch where this the holotype of this species was collected.
Diagnosis: Geosmithiastellenboschiana is a sister taxon to G. langdonii and can be distinguished by multiple SNPs in the BT (12 SNPs), EF1a exon (13) and RPB2 (31) gene regions.
Type: South Africa: Western Cape Province, Stellenbosch, Banks of the Eerste River. Isolated from Hypothenemus sp. infesting Searsiaangustifolia. June 2023. W. Rippon, WR28 (Holotype PRU(M) 4606, stored in a metabolically inactive state; ex-holotype culture CMW-IA 6987, CMW 64012). GenBank: PQ032681 (ITS); PQ048061 (BT); PQ045598 (RPB2); PQ048081 (EF1a exon).
Description: Sexual morph not observed. Asexual morph penicillium-like. Conidiophores upright borne on vegetative hyphae on surface of media or on funiculose aerial mycelia, mostly simple, occasionally branched; stipes verruculose, 0–3-septate, 10–117 (36.4 ± 25.5) μm long; penicilli terminal, mostly branched in 2–3 tiers including conidiogenous cells, occasionally in 4 tiers, 14–37 (22.4 ± 6.64) μm long, mostly appressed, asymmetric, occasionally irregularly branched; rami verruculose, infrequently absent, 9–15 (11.1 ± 1.46) μm long, in whorls of 2–4; metulae verruculose, 7–11 (9.1 ± 1.03) μm long, in whorls of 2–5. Conidiogenous cells phialidic, verruculose, cylindrical, narrowed at the tip, 7–12 (9.2 ± 1.42) μm long, in whorls of 2–6. Conidia hyaline, aseptate, oblong to ellipsoidal, cylindrical with narrowed tip, borne in basipetal chains, 4–7 × 2–3 (4.8 ± .0.78 × 2.5 ± 0.28) μm, in basipetal chains.
Culture characteristics: Optimum growth on 2 % MEA in 8 d at 20–25 °C reaching 1.9 mm/d, followed by 15 °C (1.09 mm/d), 30 °C (0.78 mm/d), 10 °C (0.51 mm/d) and 5 °C (0.17 mm/d in 14 d). No growth at 35 °C, one of three 35 °C cultures grew back after re-incubated at 25 °C. Cultures (28 day-old) fertile; shape circular; margin (10–30 °C); elevation umbonate (10–30 °C); texture fluffy becoming woolly towards centre (10–25 °C), velvety becoming fluffy towards centre with cottony patches (30 °C); colour above yellowish white with white sectors (10–30 °C); density medium, dense in some sectors; pigmentation on media absent; sectoring common at all temperatures.
Distribution: California, USA; Czech Republic; Spain; Western Cape Province, South Africa.
Additional specimen examined: South Africa: Western Cape Province, Stellenbosch, Banks of the Eerste River, June 2023. Isolated from Hypothenemus sp. infesting Searsiaangustifolia. W. Rippon, WR20 (culture CMW-IA 6986, CMW 64011). GenBank: PQ032678 (ITS); PQ048060 (BT); PQ045597 (RPB2); PQ048078 (EF1a exon).
Notes: Geosmithiastellenboschiana formed a sister clade with G. langdonii, which was isolated from Scolytusintricatus in the Czech Republic (Kolarík et al., 2005). The two “Geosmithia sp. 32” isolates from Phloeosinus spp. on Cupressaceae in the Czech Republic (Kolařík et al., 2008) and California (Kolařík et al., 2017) were found to be con-specific with G. stellenboschiana. This species, therefore, occurs on at least two genera of bark beetles on both gymnosperm and angiosperm hosts. Geosmithiastellenboschiana cultures did not show yellow shades nor pigmentation on 2 % MEA like G. langdonii and its in vitro conidial dimensions (4–7 × 2–3 μm) were larger than G. langdonii (3–5.5 × 1.5–3.5 μm).
4. Discussion
This is only the second study specifically considering Geosmithia species in the Southern Hemisphere. It also represents the first Geosmithia species descriptions from South Africa. Only one other species has been described from the Southern Hemisphere (Crous et al., 2018). That species, G. carolliae, was isolated from the wings of a bat in Brazil, but based on ITS sequences, it likely also occurs on Hypoborus ficus beetles infesting Ficus carica in the Mediterranean region (Kolařík et al., 2007). “Geosmithia sp. A″ in the study of Machingambi et al. (2014), described as G. oroboidis in the present study, therefore, remains the only Geosmithia species known exclusively from the Southern Hemisphere, as noted by Kolařík and Hulcr (2023).
The Geosmithia species described in this study were obtained from beetles infesting various native South African trees, representing two orders of gymnosperms and four angiosperm orders. The three species known from the Northern Hemisphere, G. langdonii, G. omnicola and G. pumila, were obtained from multiple beetle–host combinations, consistent with their previously established generalist nature (Kolařík and Hulcr, 2023; Pepori et al., 2015; Zhang et al., 2022). Geosmithiamultisociorum, described in this study, was also isolated from a variety of beetles and plant hosts and was the most widely occurring of all taxa thus far encountered in South Africa. This is even though the closely related Northern Hemisphere isolates, referred to as Geosmithia sp. 8 and Geosmithia sp. 48, are known from single beetle species, Scolytusintricatus and Phloeosinusdentatus, on Quercus spp. and Juniperusvirginiana, respectively (Huang et al., 2019; Kolařík et al., 2008). Along with the multiple beetle vectors and tree hosts, the global distribution of this taxon, which includes South Africa, Europe, and the USA, points to its status as a generalist Geosmithia species.
In contrast to G. capensis and G. multisociorum, G. oroboidis appeared highly specialised, associating with a specific beetle and host tree. The G. capensis isolates collected in this study were all associated with only Lanurgus spp. beetles, although occurring on both an angiosperm and coniferous host tree. Other specialist Geosmithia species are known from specific host genera, e.g. G. morbida from Jugulans spp. (Kolařík et al., 2011) and G. ulmaceae from Ulmus spp. hosts (Kolařík and Hulcr, 2023; Pepori et al., 2015) or restricted to specific geographic areas, such as G. funiculosa in Europe (Crous et al., 2022). Further surveys will be required to determine whether the species described in this study, specifically G. oroboidis, represent host-specific or geographically restricted taxa.
Geosmithiamultisociorum showed only minor differences to G. microcorthyli in the BT and EF1a exon sequences. Geosmithiamicrocorthyli is a primary ambrosial symbiont of a Microcorthylus sp. on Cassia grandis (Kolařík and Kirkendall, 2010) and the distinction between bark and ambrosial beetle symbionts presents a strong argument to distinguish between these species. Although ambrosial fungi may associate with more than one related beetle species (Saucedo-Carabez et al., 2018), we know of no examples where the same ambrosial fungus is associated with phylogenetically different beetles. Along with the gene sequences, this also suggests that the isolates obtained from Phloeosinus sp. in
Israel and deposited on GenBank as G. microcorthyli (Meshram et al., 2022) are the same as G. multisociorum.
The nature of the association between the Geosmithia species and bark beetles investigated in this study is largely unknown. Basson et al. (2024) considered the occurrence frequency of various fungal taxa associated with Lanurgus spp. beetles on Widdringtonia trees. They found that the Geosmithia species from the Lanurgus sp. 4 on W. nodiflora stems, described here as G. multisociorum, was isolated from the beetle almost 90 % of the time. In contrast, G. capensis, was isolated from only 36 % of the galleries of the W. cedarbergensis stem beetle Lanurgus sp. 3 and from 8 % of the galleries of Lanurgus sp. 1 on the same host. Both Geosmithia species were also isolated from other native South African host trees and beetles considered in this study.
Similar to other surveys that have preceded it, this study illustrates that collections from previously unexplored regions, hosts, and beetles will continue to yield novel Geosmithia taxa. The interspecific diversity of Geosmithia species in the Northern Hemisphere has received focused attention in the past 20 years (Kolařík and Hulcr, 2023) and the diversity found in South Africa appears to be similarly vast. South Africa’s rich diversity of vascular plants and understudied insect diversity, especially in the Cape Floristic Region of the Western Cape (Manning and Goldblatt, 2012), suggests that many other Geosmithia species will likely be found on native trees throughout the region. Although knowledge regarding Geosmithia species diversity continues to expand, the ecological role of these beetle-associated fungi, specifically of the non-ambrosial species, remains uncertain. Focused attention is required to understand the nature of the association between these fungi and their beetle associates, their impact on host trees and their global distribution patterns.
We use cookies to personalize content and ads, to provide social media features and to analyze our traffic. We also share information about your use of our site with our social media, advertising and analytics partners who may combine it with other information that you have provided to them or that they have collected from your use of their services. You consent to our cookies if you continue to use our website.
We use cookies to personalize content and ads, to provide social media features and to analyze our traffic. We also share information about your use of our site with our social media, advertising and analytics partners who may combine it with other information that you have provided to them or that they have collected from your use of their services. You consent to our cookies if you continue to use our website.
Necessary cookies help make a website usable by enabling basic functions like page navigation and access to secure areas of the website. The website cannot function properly without these cookies.
Preference cookies enable a website to remember information that changes the way the website behaves or looks, like your preferred language or the region that you are in.
Statistics
Statistic cookies help website owners to understand how visitors interact with websites by collecting and reporting information anonymously.
Marketing
Marketing cookies are used to track visitors across websites. The intention is to display ads that are relevant and engaging for the individual user and thereby more valuable for publishers and third-party advertisers.
Unclassified
Unclassified cookies are cookies that we are in the process of classifying, together with the providers of individual cookies.
Business Directory Registration Enquiry
Subscribe To Our Newsletter
Four new Geosmithia species from bark beetles infesting indigenous South African trees