Screening of Antibacterial Activity of Some Resupinate Fungi, Reveal Gloeocystidiellum lojanense sp. nov. (Russulales) against E. coli from Ecuador

Bacterial resistance to antibiotics is a serious public health problem that needs new antibacterial compounds for control. Fungi, including resupinated fungi, are a potential source to discover new bioactive compounds efficient again to bacteria resistant to antibiotics. The inhibitory capacity against the bacterial species was statistically evaluated. All the species (basidiomata and strains) were molecularly characterized with the ITS1-5.8S-ITS2 barcoding marker. The strains Ceraceomyces sp., Fuscoporia sp., Gloeocystidiellum sp., Oliveonia sp., Phanerochaete sp., and Xenasmatella sp. correspond to resupinate Basidiomycetes, and only the strain Hypocrea sp. is an Ascomycete, suggesting contamination to the basidiome of Tulasnella sp. According to the antagonistic test, only the Gloeocystidiellum sp. strain had antibacterial activity against the bacterial species Escherichia coli of clinical interest. Statistically, Gloeocystidiellum sp. was significantly (<0.001) active against two E. coli pathotypes (O157:H7 and ATCC 25922). Contrarily, the antibacterial activity of fungi against other pathotypes of E. coli and other strains such as Serratia sp. was not significant. The antibacterial activity between 48 and 72 h increased according to the measurement of the inhibition halos. Because of this antibacterial activity, Gloeocystidiellum sp. was taxonomically studied in deep combined morphological and molecular characterization (ITS1-5.8S-ITS2; partial LSU D1/D2 of nrDNA). A new species Gloeocystidiellum lojanense, a resupinate and corticioid fungus from a tropical montane rainforest of southern Ecuador, with antibacterial potential against E. coli, is proposed to the science.

The Enterobacteriaceae family is the largest and most heterogeneous group of Gramnegative bacteria of clinical importance [6]. Within this group, genera such as Citrobacter, Enterobacter, Escherichia, Klebsiella, Proteus, Serratia, Shigella, and Salmonella are the most frequent causes of human infections [7][8][9]. About 80% of infections, including urinary tract

Morphological Analysis
Only Gloeocystidiellum sp., after positive antibacterial activity, was analyzed morphologically in detail. In order to perform the analysis, freehand sections were made with a razor blade under a stereomicroscope (Stemi Carl Zeiss). The microscopic procedure followed Xing et al. [49]. The preparations of the sections were with phloxine 1% and decolorization with 10% potassium hydroxide (KOH) solution and Congo Red 1%. The amyloid reaction was evaluated with Melzer's reagent ( Figure 2). A sulfoaldehyde (SA) reaction to detect a sulfuric reaction of gloecystidia was performed with sulfuric acid + vanillin (Sigma-Aldrich). Observations were under a light microscope (CX31, Olympus) at 100X magnification. A detailed illustration (Figure 3) of the specimen was performed by hand using a scale (1 × 1 cm 2 = 5 × 5 µm 2 ) and later revision of the taxonomic key for the genus available in Wu, Larsson, and Ryvarden [50,51]. Color codes are based on the online server https://encycolorpedia.es/ (11 July 2022) [52].

Morphological Analysis
Only Gloeocystidiellum sp., after positive antibacterial activity, was analyzed morphologically in detail. In order to perform the analysis, freehand sections were made with a razor blade under a stereomicroscope (Stemi Carl Zeiss). The microscopic procedure followed Xing et al. [49]. The preparations of the sections were with phloxine 1% and decolorization with 10% potassium hydroxide (KOH) solution and Congo Red 1%. The amyloid reaction was evaluated with Melzer's reagent ( Figure 2). A sulfoaldehyde (SA) reaction to detect a sulfuric reaction of gloecystidia was performed with sulfuric acid + vanillin (Sigma-Aldrich). Observations were under a light microscope (CX31, Olympus) at 100× magnification. A detailed illustration (Figure 3) of the specimen was performed by hand using a scale (1 × 1 cm 2 = 5 × 5 µm 2 ) and later revision of the taxonomic key for the genus available in Wu, Larsson, and Ryvarden [50,51]. Color codes are based on the online server https://encycolorpedia.es/ (11 July 2022) [52].

Molecular Analysis
DNA was isolated from fresh basidiomata using the Phire Plant Direct Master Mix PCR Kit (Thermo Scientific™) and subsequently from pure fungal strains with the In-nuPREP DNA Kit (Analytik-jena™) according to the manufacturer's instructions. For the polymerase chain reaction (PCR), the ITS1 (5 CCGTAGGTGAACCTGCGG3 ) and NL4 (5 GGTCCGTGTTTCAAGACGG3 ) primers were used [53] to amplify the internal transcribed spacer (ITS) region and a partial sequence of nuclear large subunits (LSUs). The reactions were carried out under the following conditions: initial denaturation (98 • C, J. Fungi 2023, 9, 54 4 of 17 5 min), followed by 40 cycles with denaturation (98 • C, 10 s), hybridization (55 • C, 10 s), extension (72 • C, 30 s), and final extension (72 • C, 5 min). Final reaction volume was 20 µL, including 1 µL of the extracted DNA for each reaction. The PCR product was purified with the PureLink™ PCR Purification Kit (Thermo Scientific™) and sequenced at Macrogen Inc. (Seoul, Korea), with the same set of primers used for PCR amplification. All sequences corresponding to the basidiomata, and strains were subjected to a BLAST search against the GenBank database (https://www.ncbi.nlm.nih.gov) as well as the UNITE database (https://unite.ut.ee).  Only the sequences corresponding to Gloeocystidiellum sp. were assembled and edited using Lasergene 7 (DNAStar, Madison, WI, USA). The sequences obtained in this study were submitted to the NCBI nucleotide database under the accession numbers presented in Table 1. The alignment of the sequences obtained and reference sequences downloaded from GenBank (Table 1) was performed in the program MAFFT 7 [54] under the GINSI algorithm. Alignments were visualized in PhyDE software [55] in order to check for ambiguities, especially at the tails, for a manual adjustment if required. The alignments were analyzed by means of a neighbor-joining (NJ) approach using a Kimura two-parameter (K2P) with 1000 bootstrap repetitions (BS) and a maximum likelihood using the general time-reversible (GTR) method with 1000 bootstrap repetitions. Both analyses were performed with MEGA 11 software [56].
Finally, four phylogenetic trees were calculated: the first includes 28 sequences for the region (ITS1-5.8S-ITS2) ( Figure S1), the second tree includes 36 sequences for the partial LSU region (D1/D2) ( Figure S2), the third tree corresponds to 31 concatenated sequences (ITS1-5.8S-ITS2 and partial LSU D1/D2) (Figure 4), and the fourth tree includes 34 sequences that correspond only to the ITS1 region ( Figure 5) due to the requirement of comparing three short sequences (JQ7345551, JQ7169401, JQ7345541) described in Chile by Gorjón and Hallenberg [57]. All the phylogenetic trees presented in this study were inferred using a maximum likelihood approach.        Interspecific variation between sequences from specimen HUTPL(F)2181 and close phylogenetic species was evaluated by pairwise distances for the ITS1-5.8S-ITS2 regions and partial LSU D1/D2 in the MEGA 11 software with a Kimura two-parameter distance [58] by the "partial deletion" and "complete deletion" method.

Antibacterial Activity 2.4.1. Tested Bacterias
Ten Gram-negative bacterial strains of clinical interest were used in this study to evaluate the antibacterial activity of seven strains from resupinated fungi. These strains correspond to six different species: Enterobacter aerogenes; Enterobacter cloacae; Escherichia coli, including two pathotypes (uropathogenic Escherichia coli (UPEC), phylogenetic group (GF) A; GF B2; O157:H7) and the certified strain E. coli ATCC 25922 GF: B2; Klebsiella pneumoniae (ATCC BAA-1706); Pseudomonas aeruginosa; and Serratia sp. All the strains are available in the strain collection of the Laboratory of Cultures and Conservation of Microorganisms-UTPL, and were previously characterized with standard biochemical tests and API20E tests.

Antibacterial Activity Assay
Antibacterial activity was determined using the agar disc diffusion method by the Kirby-Bauer method as described in CLSI M100 [59]. All bacterial strains were grown on Trypticase Soy Agar (TSA, DIFCO) for 24 h at 37 • C. Subcultured colonies were suspended in 0.85% saline, followed by 0.5 McFarland density adjustment (1.5 × 10 8 CFU/mL) by spectrophotometry.
The antibacterial tests were carried out by antagonism in a Petri dish (9 cm diameter) with Mueller-Hinton agar. Discs (0.5 cm diameter) of PDA medium colonized with the strains replaced similar space (three replicates) in the medium with bacterial growth. All plates were prepared and incubated at 37 • C for 24-72 h. Antibacterial activity was recorded as diameter (mm) of the zone of inhibition formed around the fungal disc ( Figure S1). Commercial antibiotic discs for susceptibility testing based on CLSI reports [59][60][61] were also used as positive controls. Antibiotics correspond to 30 µg of cefepime for all strains, except for K. pneumoniae and uropathogenic E. coli phylogenetic group B2, in which 10 µg of imipenem was used as a positive control due to its bacterial resistance. Each Petri dish with bacterial growth is considered as one test, carried out in triplicate.

Data Analysis
Statistical analysis was performed by calculating the average (X) of the three repetitions of each test. Prior to the analysis, the dependent variable "inhibition" was transformed to a logarithm in base 2 to fulfill the assumption of normality (p > 0.05). Two-way analysis of variance to test the effects of Gloeocystidiellum sp. strain (the only fungus with antibacterial activity) and event on bacterial inhibition was used. Finally, a post-hoc Tukey test to check differences in inhibition between pairs of bacteria was performed. All analyses were carried out in R Studio software (RStudio, Inc., Boston, MA, USA) [62].

Results
Six of seven strains identify as five Basidiomycetes, and one ascomycete under six orders ( Table 2) was not characterized taxonomically because they did not present antibacterial activity. On the other hand, Gloeocystidiellum sp., which presents positive antibacterial activity against E. coli, is taxonomically described as a new species for science. The BLAST identity percent found between 87 and 97 suggests a possibility to find some new species for the science in this tropical forest.

Taxonomy
Gloeocystidiellum lojanense A. Jaramillo, D. Cruz & C. Decock., sp. nov. (Figures 2 and 3). ETYMOLOGY: The specific epithet refers to Loja province (Ecuador) where the species was found for the first time.
Gloeocystidiellum lojanense is morphologically related to Gloeocystidiellum formosanum, G. compactum, G. aspellum, and G. rajchenbergii [50,57]. These species are closely similar as far as their morphology is concerned, especially the basidiospores and gloeocystidia features (Table 3); additionally, all of them present an amyloid reaction in Meltzer's reagent. Microscopically, the shape and size (Table 3) of basidiospores and gloeocystidia in G. lojanense are indistinguishable from those of its allied species listed above. Although the size of the basidiospores has been considered as the main characteristic to define species in this group [50,57,[63][64][65], it is evident that they can overlap each other, causing taxonomic confusion, as has already been indicated in other groups of fungi (e.g., Tulasnella spp.), where molecular data allowed revealing some cryptic morphological species [32].
Another reason to propose G. lojanense as a new neotropical species is their biogeographic and ecological location, as has been described in other Gloeocystidiellum species [64,65]. G. lojanense was found in a tropical montane rainforest in southern Ecuador at 2900 m asl; different to the places in eastern Asia (Taiwan, China) or Patagonia Andean of Chile, about 600 to 1250 m asl, reported to the closest morphological species [50,57,66].

Phylogenetic Hypothesis for Gloeocystidiellum lojanense
The phylogenetic results, based on the ITS or the concatenated data sets (Figures 4, 5, S2 and S3) showed that G. lojanense form an independent well-supported clade (respectively, 96 and 100% bootstrap) that is here interpreted as a new phylogenetic species. Gloeocystidiellum lojanense forms a sister clade ( Figure 5, BS 98/99% for ML/NJ) to the clade gathering G. aspellum, G. compactum, and G. formosatum from several countries ( Table 1). A similar topology if observed is when the ITS1 region (63/56% BS for ML/NJ), including the sequence from the species G. rajchenbergii, is analyzed.
The resulting interspecific groups are totally supported when the sequences were analyzed by complete deletion and partial deletion; for example, the range for the ITS1-5.8-ITS2 region is between 6.45% and 6.90% (Table 4), ITS1 is between 2.94 and 3.69%, and partial LSU D1/D2 is 0.81% (Table 4).  Table 3. Comparison of size measurements between representative structures of Gloeocystidiellum lojanense versus the most morphologically related species into the genus Gloeocystidiellum, according the data from Donk [50].

Antibacterial Activity
Gloeocystidiellum lojanense HUTPL(F)550 was the unique strains tested by inhibiting the bacterial strains of Escherichia coli and Serratia sp. (Table 5). The other six strains (Table 2) were not further charactrized because they were totally inactive against the bacterias tested. The symbol "-" = without activity; C = positive control; SD standard deviation is indicated by ± (n = 9).
Gloeocystidiellum lojanense was observed to be more effective at inhibiting mainly two pathotypes of E. coli (i.e., Escherichia coli ATCC 25922 and Escherichia coli O157:H7). The inhibition halos generated for these strains were larger in size (between 22.78 and 24.33 mm in diameter (Table 5) and statistically significant (<0.001) versus the other tests ( Figure 6, Table 6). However, the inhibition halos generated by G. lojanense did not exceed the halos generated from the positive controls (30.33-34.33  The symbol "-" = without activity; C = positive control; SD standard deviation is indicated by ± (n = 9).
The box plot clearly indicates that Gloeocystidiellum lojanense presents greater inhibition in the bacteria Escherichia coli ATCC 25922 (22.44/22.78 mm) and Escherichia coli O157:H7 (24.33 mm), increasing during 48 and 72 hours, compared to Escherichia coli. UPEC FG: A, Escherichia coli UPEC FG: B2 and Eb, EGA, and Serratia sp., which were more stable over time ( Figure 6). The analysis of variance indicated that the fungal species, time, as well as their interaction between the two factors, significantly affect the inhibition of the bacteria, showing significant values of each factor (<0.001) ( Table 6).  The bacteria Enterobacter aerogenes, Enterobacter cloacae, Klebsiella pneumoniae, Klebsiella pneumoniae ATCC BAA-1706, and Pseudomonas aeruginosa did not exhibit inhibitory activity by G. lojanense.
The analysis of variance indicated that the fungal species, time, as well as their interaction between the two factors, significantly affect the inhibition of the bacteria, showing significant values of each factor (<0.001) ( Table 6).

Discussion
Gloeocystidiellum lojanense represents a species new to science from a tropical montane rainforest in southern Ecuador. This species is phylogenetically supported, forming an independent clade with BS values greater than 96% for the ITS1-5.8S-ITS2 and partial LSU regions, analyzed independently or concatenated. This species presents an interspecific genetic divergence for the ITS1-5.8S-ITS2 region greater than 6.45% with respect to the other species. This interspecific divergence of 6.45% would generate a barcode gap with respect to the 3% or 4% thresholds of intraspecific variability studied for fungi [32,67].
Gloeocystidiellum lojanense exhibited antibacterial activity against the four strains of the E. coli species, with E. coli ATCC 25922 and E. coli O157:H7 ( Figure 6) showing greater inhibition than the other E. coli strains. This inhibitory action may be due to the fact that these E. coli pathotypes are less virulent and resistant [68]. Other pathotypes (e.g., Escherichia coli UPEC FG: A, Escherichia coli UPEC FG: B2) are reported to be resistant to antibiotics [69], observed in slight size changes in the inhibition halos. Likewise, G. lojanense did not significantly inhibit the growth of Serratia sp. This species is also known to have a high bacterial resistance to antibiotics [70], as evidenced in the positive control with a smaller size of the inhibition halo. Suay et al. [71] present a study indicating that the species Gloeocystidiellum porosum has antibacterial activity against Pseudomonas aeruginosa, Serratia marcescens, and Staphylococcus aureus, which is why a further exploration of this group of fungi is required.
Antibacterial inhibition increased between 48 and 72 h, probably because the fungal metabolites exhibiting these antibacterial properties were generated secondarily by nutrient depletion, or simply by recognition of the bacterial foreign agent [72]. Six different strains of resupinated fungi within (Amylocorticiales, Cantharellales, Hymenochaetales, Polyporales, Russulales) and one Hypocrea sp. (Hypocreales) did not exhibit inhibition activity against any bacterial strain tested. The Gloeocystidiellum sp. (active against bacteria) and Xenasmatella sp. (inactive against bacteria) share the same order Russulales but differ in antibacterial activity. Many studies report different fungi species in the same order such as this study (e.g., Polyporales, Russulales) but positive for bacteria inhibitions [25,26]. Probably, the antibacterial activity of fungi is restricted at the species level by specific genetic adaptations [73] requiring further analysis of the gene expression of these fungi.
Statistically, the antibacterial activity of G. lojanense is significant mainly for two pathotypes of E. coli (i.e., E. coli ATCC 25922 and E. coli O157:H7). However, our analysis requires an evaluation of the minimum inhibitory concentration by obtaining its extracts, as has already been evaluated for other fungi [74].
Ecuador, considered within the megadiverse countries worldwide [75][76][77], has as consequence a great diversity of fungi [27,29,33,34] that require integrative taxonomic research to discover new species such as G. lojanense with bioactive potential and applications in several sectors such as human health.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/jof9010054/s1, Figure S1: Antibacterial activity assay flowchart; Figure S2: Maximum Likelihood phylogenetic tree for the ITS1-5.8-ITS2 region for sequence positioning corresponding to Gloeocystidiellum lojanense sp. nov. Bar = number of expected substitutions per position. * Sequences generated in this study; Figure S3: Maximum likelihood phylogenetic tree for the LSU region for sequence positioning corresponding to Gloeocystidiellum lojanense sp. nov. Bar = number of expected substitutions per position. * Sequences generated in this study.