Terfezia solaris-libera sp. Nov., A New Mycorrhizal Species within the Spiny-Spored Lineages

Fungal Collections and Collecting Sites

Terfezia ascocarps were collected in different years and from different locations in Centre and Southern Portugal. Throughout the collection period (from February to April), fresh specimens were brought to the laboratory for morphological and molecular characterization. The putative plant host was registered and soil samples were collected nearby each specimen. Fragments of each specimen were frozen at-20ºC for DNA amplification and the remaining specimen were dried at 40°C and stored in sealed plastic bags, labeled with collection details. All samples are deposited at the Herbarium of the Evora University Herbarium (UEVH- FUNGI), Portugal. Soil samples (50mm diam., 150mm depth) were collected in each collection site. A compose sample of 6 soil samples replicas per site was made and analyzed at the Laboratório Químico Agrícola Rebelo da Silva (INIAV/LQARS) for particle size and subsequent soil textural classification and water pH measurements [15].

Morphological Study

External ascocarp characteristics (shape, colour, appearance) were in detail recorded from fresh specimens. Ascocarps were then cut and the morphology of the peridium and gleba was described. Microscopic observations were performed in distilled water, KOH 5% and Melzer’s reagent. Spores dimensions are based on a minimum of 100 randomly selected spores outside asci. Peridium of dried specimens were rehydrated and examined in KOH 5%. Melzer’s reagent was used to test the amyloidity of asci, spores and tissues. Asci and ascospores were examined using a Leica DM750 microscope equipped with a digital camera (Leica ICC50 W). For identification, ascomata were compared with the descriptions from Bordallo, et al. [6, 7, 14] and Mattirolo, et al. [13].

DNA Amplification and Sequencing

DNA extraction from the analyzed specimen was performed by CTAB method, following the protocol described in Nobre, et al. [16]. All extraction products were stored at-20˚C and later used directly in the PCR. The Internal Transcribed Spacer (ITS) region of the rDNA, including the 5.8S ribosomal gene, was amplified using the ITS5 and ITS4 primers [17]. PCR reactions were conducted using 1μl of the extracted DNA in a standard 25μl reaction, with 0.5pmol/μl of each primer, 1.5mM MgCl2, 0.5mM dNTPs and 0.04U/ml Taq DNA polymerase. PCR reactions were performed using a Mastercycler Gradient thermocycler (Eppendorf, Hamburg, Germany) with the following cycling parameters: an initial denaturalization step for 3min at 95°C, followed by 35 cycles consisting of: 30s at 95°C, 30s at 95°C (annealing temp.), 1min at 72°C, and a final extension at 72°C for 10min. All the PCR products were purified using the NZYGelpure kit (from NZYTech, Lda) and sequencing was done commercially (STAB VIDA, Lda.).

Phylogenetic Analysis

The final dataset comprised 78 Terfezia ITS sequences (Table 1, Supplemental Table 1) and a Tirmania nivea specimen as out group (FN395015). Additional to the sequences pertaining the collected specimens, the dataset comprised 3 representative sequences from each Terfezia clade previously identified and described [11]. Care was taken to always include the Terfezia species type sequence as deposited in the comprehensive database Gen Bank All sequences were assembled, edited and aligned with online MAFFT version 7, using the E-INS-i strategy [18]. The phylogenetic reconstruction analysis based on the above ITS sequences was performed in BEAST v.4.2.8 software [19], allowing the software to estimate the evolutionary model. All other settings were left as default. Three independent Markov Chain Monte Carlo (MCMC) runs, starting from randomly chosen topologies were performed as a safeguard against spurious results. The MCMC were run for 10,000,000 generations with data sampled every 1,000 generations. Log- file outputs were inspected in Tracer v.1.6 to determine chain convergence and burnin. The first 10% of the generations were discarded as the burn-in, the tree-file outputs from individual MCMC runs were combined in Log Combiner v.2.4.8. And the maximum clade credibility tree and corresponding posterior probabilities were obtained using Tree Annotator v.2.4.8 [19]. To test the congruence between methods, reconstructions with Maximum Likelihood (ML), Minimum Evolution (ME) and Neighbor-Joining (NJ) methods were performed in Mega 10.0.5 [20]. The best-fit nucleotide substitution model was selected with the program Model Test [21] applying the Akike information criterion (AIC) and this information was used to calculate the distances in the ML analysis.

Supplemental Material 1

To test the congruence between methods, reconstructions with Maximum Likelihood (ML), Minimum Evolution (ME) and Neighbor-Joining (NJ) methods were performed in Mega 10.0.5 Kumar, et al. [20]. Each analysis involved 79 nucleotide sequences. All ambiguous positions were removed for each sequence pair (pairwise deletion option). There were a total of 738 positions in the final dataset.

The best-fit nucleotide substitution model selected with the program jModel Test Posada, et al. [21] applying the Akaike information criterion (AIC) was GTR+I+G. This model was used to calculate the distances for the ML analysis (Figure S1).

Table S1: Terfezia collections analysed.

As to the NJ (Figure S2) and ME (Figure S3) analysis, the most simple model was assumed where evolutionary distances were computed using the p-distance method, showing the robustness of the monophyly of the groups considered as taxonomic units.

Click to enlarge

Figure S1: The relation between specimens was inferred using the Maximum Likelihood method and General Time Reversible model Nei M, et al. [32]. The tree with the highest log likelihood (-1968.39) is shown. The percentage of trees in which the associated taxa clustered together is shown next to the branches. Initial tree(s) for the heuristic search were obtained automatically by applying Neighbor-Join and BioNJ algorithms to a matrix of pairwise distances estimated using the Maximum Composite Likelihood (MCL) approach, and then selecting the topology with superior log likelihood value. The rate variation model allowed for some sites to be evolutionarily invariable ([+I], 37.13% sites). The tree is drawn to scale, with branch lengths measured in the number of substitutions per site. All positions with less than 95% site coverage were eliminated, i.e., fewer than 5% alignment gaps, missing data, and ambiguous bases were allowed at any position (partial deletion option).

Click to enlarge

Figure S2: The relation between specimens was inferred using the Neighbor-Joining method Saitou N, et al. [33]. The optimal tree with the sum of branch length = 0.69421768 is shown. The percentage of replicate trees in which the associated taxa clustered together in the bootstrap test (1000 replicates) are shown next to the branches Felsenstein J, et al. [34]. The tree is drawn to scale, with branch lengths in the same units as those of the evolutionary distances used to infer the phylogenetic tree. The evolutionary distances were computed using the p-distance method Nei M, et al. [32] and are in the units of the number of base differences per site.

Click to enlarge

Figure S3: The relation between specimens was inferred using the Minimum Evolution method Rzhetsky A, et al. [35]. The optimal tree with the sum of branch length = 0.69543083 is shown. The percentage of replicate trees in which the associated taxa clustered together in the bootstrap test (1000 replicates) are shown next to the branches Felsenstein J, et al. [34]. The tree is drawn to scale, with branch lengths in the same units as those of the evolutionary distances used to infer the phylogenetic tree. The evolutionary distances were computed using the p-distance method Nei M, et al. [32] and are in the units of the number of base differences per site. The ME tree was searched using the Close-Neighbor-Interchange (CNI) algorithm Nei M, et al. [32] at a search level of 1.

Taxonomy

Name: Terfezia solaris-libera Louro, Nobre, Santos-Silva, sp. nov. Type: PORTUGAL, Alentejo, Arraiolos, in roadside verges near grasslands areas without trees, associated exclusively with Tuberaria guttata; 25 April 2018; leg. C. Santos-Silva, Holotype: UEVH-FUNGI 2004089.

Diagnosis: Ascomata 1.5-2.5cm in diam., hypogeous, subglobose and light coloured to brownish. Peridium 500- 700μm thick, poorly delimited and pseudoparenchymatous composed of subglobose cells of variable size. Gleba with brownish pockets of fertile tissue surrounded by whitish, sterile, veins, becoming uniformly brownish coloured at maturity. Ascospores globose, (20-)21.6(-23) μm diam. (median=22μm) including ornamentation; (15- )15.7(-17)μm (median=16μm) without ornamentation; light brown and ornamented with conical, blunt spines, sometimes cylindrical, generally straight, but sometimes slightly curved, separated, 2-3μm long, 1-1.5μm wide at the base. Terfezia solaris-libera sp. nov. Differs from other spiny-spored Terfezia species associated with Tuberaria guttata by its poor delimited thicker peridium and different sporal ornamentation, and from all Terfezia spp. in its ITS nrDNA sequence. T. fanfani usually reaches larger ascocarp dimensions, has prismatic peridium cells, olive green tinges in mature gleba and different spore ornamentation constituted by sharp thin elongated conic spines (2-)3-4(-5)μm long; T. lusitanica has a lighter yellowish and thinner peridium and a blackish gleba upon maturity and finally T. extremadurensis has a thinner well delimited peridium and tuber like gleba. Exclusively associated with Cistus, Terfezia cistophila shares a similar habitat as the aforementioned species, differing from all in its spermatic odour. Etymology: “solaris” refers to the sun, due to its particular spore ornamentation and ecology, and “libera” due to the collection date of the type (25 April, the Freedom day in Portugal which commemorates the Carnation Revolution). Description: Ascomata hypogeous, subglobose, 1.5-2.5cm in diameter, light ochre colour at first, becoming darker in maturity, smooth (Figures 1A & 1B). Faint odour, not distinctive. Mild taste. Peridium 500-700μm thick, not separable from gleba, poorly delimited, whitish in cross section, pseudoparenchymatous, composed of subglobose cells of variable size, hyaline and thin-walled in the innermost layers (max. 40μm diam.), yellowish and with thicker walls in the outermost layers (max. 8μm diam.) (Figure 1C). Gleba solid, fleshy, succulent, whitish at first, then with brownish pockets of fertile tissue surrounded by whitish sterile veins, finally becoming uniformly brownish coloured at maturity (Figures 1A & 1B). Asci no amyloid, subglobose, sessile, 65- 80 x 50-65=μm diam., walls 1μm thick, with 6-8 irregularly disposed spores (Figures 1F & 1G), randomly arranged in the gleba. Ascospores. globose, (20-) 21.6(-23)μm diam. (median=22=μm) including ornamentation; (15-)15.7(- 17)μm (median = 16μm) without ornamentation; hyaline, smooth and uniguttulated at first, by maturity light brown and ornamented with conical, blunt spines, sometimes cylindrical, generally straight, but sometimes slightly curved, separated, 2-3μm long, 1-1.5μm wide at the base (Figures 1D & 1E).

Habitat and Distribution: Hypogeous mycorrhizal fungi, in sandy loam (80-83% sand material), acid soils (pH 6), appearing solitary or in small groups, from February to April, in roadside verges near grasslands areas without trees, associated exclusively with Tuberaria guttata. Occurring in Southwestern Iberian Peninsula.

Specimens Examined: PORTUGAL: Alentejo: Arraiolos, 30 March 2017, C. Santos-Silva leg. (UEVH-FUNGI 2003820, GenBank: MN338729; UEVH-FUNGI 2003821, GenBank: MN338730; UEVH-FUNGI 2003840, GenBank: MN338727; UEVH-FUNGI 2003846, GenBank: MN338728); Idem, idem, Arraiolos, 14 February 2019, R. Louro leg. (UEVH-FUNGI 2004746; GenBank: MN338732); Idem, idem, Évora, 27 February 2019, R. Louro leg. (UEVH-FUNGI 2004593, GenBank: MN338733).

Remarks: Terfezia solaris-libera sp. nov. differs from other spiny-spored Terfezia species associated with Tuberaria guttata by its poor delimited thicker peridium and different sporal ornamentation, and from all Terfezia spp. in its ITS nrDNA sequence. T. fanfani usually reachs large ascocarp dimensions, often with prismatic peridium cells, with olive green tinges in mature gleba and a different sporal ornamentation. T. lusitanica has a lighter yellowish and thinner peridium and a blackish gleba upon maturity, T. extremadurensis has a thinner well delimited peridium and Tuber-like gleba. Exclusively associated with Cistus, Terfezia cistophila shares a similar habitat as the aforementioned species, differing from all in its spermatic odour. It can be separated from T. solaris- libera sp. nov. By its thinner peridium and smaller spores.

Click to enlarge

Figure 1: Terfezia solaris-libera, sp. nov. A: Fresh Mature Ascocarp Sectioned (50mm), B: Dry Ascocarp External Peridium and Gleba (50mm), C: Detail of the Pseudoparenchymatous Peridium (50μm), D & E: Details of Mature and Immature Spore Ornamentation Under Different Light Conditions (15μm & 10μm); F & G: Asci and Mature and Immature Ascopores (45μm & 20μm).

Key to Examined Species

1a. Associated exclusively with Cistus and with spermatic odor.. T. cistophila 1b. Associated, mainly, with Tuberaria guttata without spermatic odor............................................................................. 2 2a. Peridium with reddish colour and gleba with olive-green tinges................................................................................................................ T. fanfani 2b. Peridium with no reddish colours and gleba with no green tinges................................................................................................... 3 3a. Gleba with blackish-grey pockets at maturit.......................... T. lusitanica 3b. Gleba with brownish pockets at maturit................................ T. solaris-libera sp. nov.

Phylogenetic Analysis

The reconstructed phylogeny ample supports the newly described species T. solaris-libera sp. nov. and corroborates the existence of 17 distinct clades representing well supported monophyletic groups (Figure 2). The sequence originally named T. leptoderma (GenBank accession no. AF396864) remains isolated and does not nested inside of any clade. All reconstructed phylogenies were congruent, regardless the method used (Supplemental Material 1). The clade comprising the T. solaris-libera sp. nov. Sequences harbors also the Genbank sequences HM056215, HQ698132 and HQ698149 (Figure 2). The sequences referred to above belong to ascocarps collected in Spain, initially identified as Terfezia sp_., _T. aff. Olbiensis and T. aff. Olbiensi respectively. The other spiny-spored Terfezia species were also well supported in monophyletic groups, separating specimens of T. lusitanica, T. cistophila and T. fanfani.