1 Supporting Information
2 Article title: : Host-generalist fungal pathogens of seedlings may maintain forest diversity via 3 host-specific impacts and differential susceptibility among tree species 4 Authors: Erin R. Spear and Kirk D. Broders 5
6 The following Supporting Information is available for this article:
7 Fig. S1 Examples of disease symptoms in the forests of Panama.
8 Fig. S2 Details of shadehouse-based inoculation experiments.
9 Fig. S3 Rank abundance plot and OTU accumulation curve.
10 Fig. S4 Overlap in fungal OTUs among sampling years, methods used to obtain symptomatic
11 seedlings, isolation media, and tissue sampled.
12 Fig. S5 Correlation between OTU host range and isolation frequency.
13 Table S1 Taxonomic assignments, traits, sampling effort, and observed OTUs for tree species
14 evaluated in our survey and experimental approaches.
15 Table S2 Methodological details pertaining to the multi-year collection of symptomatic
16 seedlings, and microbial isolation and sequencing.
17 Table S3 Average light levels, air temperatures, and relative humidities of the shadehouses
18 used for inoculation experiments versus ambient conditions.
19 Table S4 Estimated taxonomic placement, isolation frequency, number of observed hosts,
20 estimated host specialization, and phylogenetic pattern of host use of the OTUs.
21 Table S5 Overlap in seedling-associated OTUs among tree species.
22 Table S6 Results of the beta-binomial generalized linear regression with the proportion of
23 diseased seedlings as a function of seed size and shade tolerance.
24 Table S7 Average estimates based on the best-ranked beta-binomial generalized linear
1
25 regressions with the proportion of diseased seedlings as a function of seed size and spatial
26 distribution relative to annual rainfall.
27 Methods S1 Methods used to estimate the taxonomic placement of the 66 OTUs and assign
28 nomenclature.
2
29
30 Fig. S1 Disease symptoms on the (a,g-j,m,o,p-t,v) stems, (b-f,k,l,n,o) leaves, and (u) root of
31 seedlings in the forests of Panama. In some panels, arrows direct the viewer’s attention to
32 disease.
3
33 34 Fig. S2 (a) Inoculation experiments were conducted in Smithsonian Tropical Research Institute
35 shadehouses in Gamboa, Panama. (b) Surface-sterilized seeds were germinated in flats of
36 autoclave-sterilized commercial soil. (c,f) Seedlings were transplanted to individual pots
37 containing autoclaved commercial soil and (d,e) either rice visibly colonized by one of the
38 fungal isolates or inoculum-free, autoclave-sterilized rice. Disease symptoms were documented
39 every three days and were categorized as seedling mortality, (g-o) stem damage, (p,q) wilted
40 tissue, and (r,s) stunted seedling growth.
4
41 42 Fig. S3 (a) The 66 observed OTUs, defined by 99% ITS sequence similarity, are ranked from most
43 to least abundant on the horizontal axis, with the total number of isolates per OTU plotted on
44 the vertical axis (full dataset: n = 211 isolates) (BiodiversityR package; Kindt & Coe, 2005). Most
45 of the OTUs are rare (50% singletons), indicated by the steep shape of the curve. The four
46 common OTUs (observed ≥10 times and comprising 35% of the isolates) are named. (b) Non-
47 asymptotic accumulation of OTUs isolated from 124 symptomatic seedlings (vegan package;
48 Oksanen et al., 2019). The curve, derived from the observed richness and representing the
49 mean accumulation of OTUs over 999 randomizations of seedling order, indicates incomplete
50 sampling and a diverse community.
5
51 References: 52 Kindt R, Coe R. 2005. Tree diversity analysis. A manual and software for common statistical 53 methods for ecological and biodiversity studies. Nairobi, Kenya: World Agroforestry 54 Centre (ICRAF). 55 Oksanen J, Blanchet FG, Friendly M, Kindt R, Legendre P, McGlinn D, Minchin PR, O'Hara RB, 56 Simpson GL, Solymos P et al. 2019. vegan: Community Ecology Package. R package 57 version 2.5-6. [WWW document] URL https://CRAN.R-project.org/package=vegan 58 [accessed 2 June 2020].
6
59 60 Fig. S4 Venn diagrams depicting the overlap in non-singleton fungal operational taxonomic
61 units (OTUs; based on 99% sequence similarity) among the five sampling years, three methods
62 used to obtain seedlings with disease symptoms, four media used for isolation, and three
63 tissues sampled (data subset A: n = 178 isolates, from 110 seedlings). (a) Fungi, and two
64 oomycetes, were isolated from symptomatic seedlings in Panama over five years (2007, 2010-
65 2012, 2019). (b) Symptomatic seedlings were obtained in three ways: (i) opportunistic
66 collection of naturally occurring seedlings, (ii) seedlings germinated in a shadehouse and then
67 transplanted to forest sites, and (iii) surface-sterilized seeds planted directly in forest sites. (c,
7
68 d) The advancing margin(s) of diseased area(s) was/were excised, and the excised tissue
69 piece(s) (leaf, stem, and/or root) was/were surface sterilized (Gilbert & Webb, 2007) and plated
70 on (i) Water agar, (ii) Pimaricin, Ampicillin, Rifampicin, and Pentachloronitrobenzene (PARP);
71 and/or Malt Extract Agar (MEA) amended with antibiotic to prevent bacterial growth, either (iii)
72 chloramphenicol or (iv) rifampicin. See Table S2 and Spear (2007) for additional methodological
73 details. (a) Of the 33 non-singleton OTUs, 19 were observed in more than one year. While no
74 OTUs were observed across all five years, three OTUs were observed across four of the
75 sampling years. The greatest number of unique, non-singleton OTUs were observed in 2019,
76 the year we collected the greatest number of seedlings. (b) Eighteen non-singleton OTUs were
77 isolated from seedlings obtained using more than one method. Four non-singleton OTUs were
78 isolated from seedlings obtained using all three methods. We isolated the greatest number of
79 unique, non-singleton OTUs from naturally occurring seedlings, the most common sampling
80 method. (c) Nineteen non-singleton OTUs were isolated on multiple media. One non-singleton
81 OTU was isolated from tissue pieces plated on all four media. We isolated the greatest number
82 of unique, non-singleton OTUs on MEA + rifampicin, the media used for the greatest number of
83 seedlings and tissue pieces. (d) Twenty-two non-singleton OTUs were isolated from multiple
84 tissues. Four non-singleton OTUs were isolated from all three tissues. We isolated the greatest
85 number of unique, non-singleton OTUs from leaves, the best-sampled tissue.
86 References: 87 Gilbert GS, Webb CO. 2007. Phylogenetic signal in plant pathogen-host range. Proceedings of 88 the National Academy of Sciences, USA 104: 4979-4983. 89 Spear ER. 2017. Phylogenetic relationships and spatial distributions of putative fungal 90 pathogens of seedlings across a rainfall gradient in Panama. Fungal Ecology 26: 65-73
8
91 92 Fig. S5 The observed host range of an OTU is positively correlated with isolation frequency
93 (survey-based assessment of host range: blue points, one-tailed Spearman's rank correlation
94 rho = 0.963, P < 0.001; host range observed during the inoculation experiments: green points,
95 one-tailed Spearman's rank correlation rho = 0.721, P = 0.053), suggesting that host-generalized
96 fungi may be more common in this system.
9
97 Table S1 Tree species from which putative pathogens were isolated (original host = OH, 26 tree species) and/or for which
98 vulnerability to pathogens was assessed (target = T, 35 tree species) via inoculation experiments. For each tree species, the following
99 is listed: a two- or three-letter code (for Fig. 3 and Tables 1, S2, and S5), taxonomic assignments, and the number of seedlings
100 collected, sites from which seedlings were collected, unique isolates observed, and OTUs observed. Average seed dry mass (mg),
101 shade tolerance, and spatial distribution relative to annual rainfall are listed for the tree species used to explore the relationship
102 between disease susceptibility and plant life history traits. The imperfect match between original hosts and targets was driven by
103 limited seed availability, long seed dormancy periods, and space and time constraints. Additionally, several tree species that were
104 not original hosts were included in the inoculation experiments as phytometers (measures of the pathogenicity of the isolates)
105 because of their previously observed disease susceptibility (e.g., L. seemannii, Augspurger & Wilkinson, 2007).
Role Species Code Family Order Seed Shade Dist.3 Seedlings Sites Isolates OTUs mass tol.2 collected (mg)1 OH, T Anacardium excelsum ANE Anacardiaceae Sapindales 1507 dry 20 5 41 24 OH, T Dalbergia retusa DR Fabaceae Fabales 130 tol dry 18 2 20 10 OH Pouteria reticulata PR Sapotaceae Ericales 11 1 30 18 OH, T Virola surinamensis VS Myristicaceae Magnoliales 11 3 15 11 OH Faramea occidentalis FO Rubiaceae Gentianales 7 1 13 7 OH Protium panamense PP Burseraceae Sapindales 7 1 18 11 OH Protium tenuifolium PT Burseraceae Sapindales 6 2 10 9 OH, T Calophyllum longifolium CL Calophyllaceae Malpighiales 5 1 14 9 OH, T Castilla elastica CE Moraceae Rosales 203.4 dry 5 2 6 4 OH, T Hymenaea courbaril HC Fabaceae Fabales 5 3 6 5 OH Cassia moschata CAM Fabaceae Fabales 4 3 5 4 OH, T Lacmellea panamensis LAP Apocynaceae Gentianales 237.4 tol wet 4 3 7 5 OH Nectandra cuspidata NC Lauraceae Laurales 4 2 5 5 OH, T Cochlospermum vitifolium CV Bixaceae Malvales 26 intol 3 1 4 4 OH Swietenia macrophylla SWM Meliaceae Sapindales 2 1 2 2 OH, T Trichilia tuberculata TT Meliaceae Sapindales 151 tol 2 2 2 2 OH, T Brosimum utile BU Moraceae Rosales 1763.5 wet 1 1 2 2 OH, T Cojoba rufescens CR Fabaceae Fabales 236.2 tol dry 1 1 2 2 OH Dipteryx oleifera DO Fabaceae Fabales 1 1 1 1 10
OH, T Genipa americana GA Rubiaceae Gentianales 123 tol dry 1 1 1 1 OH Mouriri myrtilloides MM Melastomataceae Myrtales 1 1 1 1 OH Ormosia coccinea OC Fabaceae Fabales 1 1 1 1 OH, T Ormosia macrocalyx OM Fabaceae Fabales 379 tol dry 1 1 1 1 OH Randia armata RA Rubiaceae Gentianales 1 1 1 1 OH Spondias mombin SPM Anacardiaceae Sapindales 1 1 2 2 OH, T Tetragastris panamensis TEP Burseraceae Sapindales 295 tol 1 1 1 1 T Annona glabra AG Annonaceae Magnoliales 229 intol wet T Coccoloba manzinellensis COM Polygonaceae Caryophyllales T Copaifera aromatica CA Fabaceae Fabales 893.6 tol dry T Eugenia nesiotica EN Myrtaceae Myrtales 346.5 tol T Garcinia intermedia GI Clusiaceae Malpighiales 541 tol T Guapira standleyana GS Nyctaginaceae Caryophyllales 61.2 tol T Inga goldmanii IG Fabaceae Fabales T Inga sapindoides IS Fabaceae Fabales 406.4 intol T Jacaranda copaia JC Bignoniaceae Lamiales 5 intol T Lacistema aggregatum LA Lacistemataceae Malpighiales 10.7 intol T Licania platypus LIP Chrysobalanaceae Malpighiales T Luehea seemannii LS Malvaceae Malvales 3 intol dry T Pachira quinata PQ Malvaceae Malvales 40 intol dry T Posoqueria latifolia POL Rubiaceae Gentianales 196.5 tol T Psychotria limonensis PSL Rubiaceae Gentianales 6.6 tol T Psychotria marginata PM Rubiaceae Gentianales 7.5 tol T Quararibea asterolepis QA Malvaceae Malvales 335 tol T Siparuna pauciflora SP Siparunaceae Laurales T Swartzia simplex SS Fabaceae Fabales 1025 tol T Symphonia globulifera SG Clusiaceae Malpighiales 2334 tol wet T Tocoyena pittieri TOP Rubiaceae Gentianales 956.2 tol wet 106 1Seed mass sources: (1) Daws MI, Garwood NC, Pritchard HW. 2005. Traits of recalcitrant seeds in a semi‐deciduous tropical forest in Panama: some ecological implications. Functional Ecology 19: 107 874-885. (2) Myers JA, Kitajima K. 2007. Carbohydrate storage enhances seedling shade and stress tolerance in a neotropical forest. Journal of Ecology 95: 383–395. (3) E. R. Spear, unpublished data. 108 (4) Svenning JC, Wright SJ. 2005. Seed limitation in a Panamanian forest. Journal of Ecology 93: 853-862. (5) Wright SJ, Kitajima K, Kraft NJB, Reich PB, Wright IJ, Bunker DE, Condit R, Dalling JW, 109 Davies SJ, Díaz S et al. 2010. Functional traits and the growth–mortality trade‐off in tropical trees. Ecology 91: 3664-3674. 110 2Shade tolerance sources: (1) Augspurger CK. 1984. Light requirements of neotropical tree seedlings: a comparative study of growth and survival. Journal of Ecology 72: 777-795. (2) Brown SH, Mark 111 S. 2013. Fact sheet: Annona glabra. USDA, Cooperative Extension Service, University of Florida, IFAS, Florida A. & M. [WWW document] URL http://www.doc-developpement-durable.org/file/Arbres- 112 Fruitiers/FICHES_ARBRES/Cachiman-cochon-mammier-Annona-glabra/Pond%20Apple%20-%20Lee%20County%20Extension%20-%20University%20of%20Florida.pdf. [accessed 21 May 2020]. (3) 113 Comita LS, Aguilar S, Pérez R, Lao S, Hubbell SP. 2007. Patterns of woody plant species abundance and diversity in the seedling layer of a tropical forest. Journal of Vegetation Science 18: 163-174. (4) 114 Hall JS, Ashton MS. 2016. Guide to early growth and survival in plantations of 64 tree species native to Panama and the Neotropics. Balboa, Ancón, República de Panamá: Smithsonian Tropical 115 Research Institute. (5) Kitajima K, Llorens AM, Stefanescu C, Timchenko MV, Lucas PW, Wright SJ. 2012. How cellulose‐based leaf toughness and lamina density contribute to long leaf lifespans of 116 shade‐tolerant species. New Phytologist 195: 640-652. (6) Martin WA, Flores EM. 2002. Copaifera aromatica Dwyer. In Vozzo JA, ed. Tropical tree seed manual. Washington DC, USA: USDA Forest 117 Service, 405-407. (7) Molofsky J, Augspurger CK. 1992. The effect of leaf litter on early seedling establishment in a tropical forest. Ecology 73: 68-77. (8) Paul GS, Montagnini F, Berlyn G P, Craven DJ,
11
118 van Breugel M, Hall JS. 2012. Foliar herbivory and leaf traits of five native tree species in a young plantation of Central Panama. New Forests 43: 69-87. (9) Pearcy RW, Valladares F, Wright SJ, De 119 Paulis EL. 2004. A functional analysis of the crown architecture of tropical forest Psychotria species: do species vary in light capture efficiency and consequently in carbon gain and growth? Oecologia 120 139: 163-177. 121 3Distribution sources: (1) Engelbrecht BM, Comita LS, Condit R, Kursar TA, Tyree MT, Turner BL, Hubbell SP. 2007. Drought sensitivity shapes species distribution patterns in tropical 122 forests. Nature 447: 80-82. (2) Condit R, Pérez R, Daguerre N. 2010. Trees of Panama and Costa Rica. Princeton, NJ: University Press. (3) Perez R, Condit R. 2020. Tree Atlas of Panama. [WWW 123 document] URL http://ctfs.si.edu/webatlas/maintreeatlas.php. [accessed 21 May 2020]
12
124 Table S2 Methodological details pertaining to the multi-year collection of symptomatic seedlings, and microbial isolation and
125 sequencing. Seedlings were collected from the lowland tropical forests of Panama in 2007, 2010-2012, and 2019 by E. R. Spear and
126 T. Brenes-Arguedas (column 1). Symptomatic seedlings were obtained by opportunistically collecting naturally occurring seedlings
127 and by baiting pathogens from the soil by planting seedlings or surface-sterilized seeds directly in the forest sites (col. 2). The forest
128 sites from which seedlings were collected included: Buena Vista Peninsula (BV), Barro Colorado Nature Monument (BCI), Gunn Hill in
129 Ciudad del Saber (formerly Fort Clayton; FC), private property on Santa Rita Ridge (SRR), Parque Natural Metropolitano (PNM),
130 Sendero Camino de Cruces in Parque Nacional Soberanía (CC), and Sendero del Charco in Parque Nacional Soberanía (SC) (col. 3; see
131 Fig. S1 in Spear (2017) for a map with approximate locations and additional site details). In total, 124 seedlings of 26 tree species
132 were collected, but the tree species collected varied by site and year (col. 4; see Table S1 for the full species names). Seedlings were
133 collected during the rainy season (col. 5 provides date ranges). For all 124 collected seedlings, the advancing margin(s) of diseased
134 area(s) was/were excised, and the excised symptomatic tissue piece(s) was/were surface sterilized following Gilbert & Webb (2007)
135 prior to plating. Fungi (and two oomycetes) were isolated on four types of media Water Agar (WA; 20 isolates); Pimaricin, Ampicillin,
136 Rifampicin, and Pentachloronitrobenzene (PARP†; 11 isolates); and Malt Extract Agar (MEA) amended with antibiotic to prevent
137 bacterial growth, either chloramphenicol (chloram; 59 isolates) or rifampicin (rifamp; 121 isolates) (col. 6). Morphologically unique
138 isolates were subcultured into pure culture and a piece of mycelium was excised from each pure culture for molecular analysis (col.
139 6). The DNA extractions, PCR amplifications, and bidirectional Sanger sequencing of either the nuclear ribosomal internal transcribed
140 spacer (ITS) region (149 isolates) or the ITS plus an adjacent portion of the large subunit (ITS+LSU)‡ (62 isolates) were completed
141 over multiple years and by multiple labs: the lab of A. E. Arnold at the University of Arizona (methods followed lab's specific
142 protocols; e.g., Sandberg et al., 2014); the International Cooperative Biodiversity Groups (ICBG) lab at the Smithsonian Tropical
143 Research (STRI) (methods followed lab's specific protocols; e.g., Higginbotham et al., 2014); the molecular research lab at STRI Naos
144 Marine Laboratories; the lab of K. D. Broders at the STRI; and Macrogen, Inc. (col. 7). The primers ITS1F, ITS5, ITS4, and LR3 (Vilgalys
13
145 & Hester, 1990; White et al., 1990; Gardes & Bruns, 1993) were used for amplification and sequencing (col. 8). Edited DNA
146 sequences for all, but one, of the fungal isolates collected from 2007-2012 have been deposited in NCBI GenBank (col. 9). Edited
147 DNA sequences for one fungal isolate collected in 2011, and 119 fungal isolates and two oomyectes collected in 2019 will be
148 deposited upon manuscript acceptance (col. 9). All 211 isolates (collected from 2007-2011) were used for our survey-based
149 assessment of the host associations and ranges of putative pathogens (col. 10). Twenty-seven of the isolates collected in 2010 and
150 2011 were used for our experimental assessments of pathogenicity and host range (seedling inoculation experiments) conducted in
151 2011 and 2012 (col. 11).
Year & Source of Forest No. of Date No. of unique Facilities where Primers used NCBI Isolates used for Isolates used collector symptomatic site(s) seedlings range of isolates, media extractions, GenBank survey-based for seedlings, & & tree seedling used, plant tissue amplifications, & Accession assessment of the host experimental number of spp. collection sampled sequencing were numbers associations and assessments of tree spp. collected completed ranges of putative pathogenicity collected pathogens and host range 2007§ Seeds (1) BV 22 Jul. 21- 28 fungal isolates (1) Arnold Lab ITS5, ITS1F, ITS4, KY413686- Yes No T. Brenes- germinated in (2) FC Tree spp: Nov.13, Media: LR3‡ KY413775 Arguedas a STRI (3) SRR (1) BU, 2007 (1) WA, shadehouse (2) CAM, (2) PARP† and then (3) CV, Plant tissue: planted (4) HC, (1) stem, (5) LAP, (2) root (6) NC, (7) OC, (8) OM (9) PT, (10) SWM 2010§ (1) Surface- (1) SRR 28 Jul 20- 34 fungal isolates (1) Arnold Lab, ITS1F, ITS4, LR3‡ Yes Yes E. R. Spear sterilized (2) PNM Tree spp: Nov. 16, Media: (2) STRI ICBG lab, seeds planted (1) ANE, 2010 (1) MEA+chloram (3) STRI mol. directly in (2) CE, (2) PARP† research lab forest sites (3) CR, Plant tissue: (see Spear et (4) GA, (1) stem, al. 2015 for (5) HC, (2) root, additional (6) PT, (3) leaf details) (7) RA, (2) Naturally (8) TEP, occurring (9) TT, seedings (10) VS
14
2011§ Naturally (1) PNM 8 May 21- 8 fungal isolates (1) STRI ICBG lab, ITS5, ITS1F, ITS4, Yes Yes E. R. Spear occurring (2) BCI Tree spp: Jun. 9, Media: (2) STRI mol. LR3‡ seedings (3) CC (1) ANE, 2011 (1) MEA+chloram research lab, (4) SC (2) CL, Plant tissue: (3) Broders Lab, (3) DO (1) stem, (4) Macrogen, Inc. (2) root, (3) leaf 2012§ Surface- (1) SRR 18 Jul. 5 -Jul. 20 fungal isolates (1) Arnold Lab ITS5, LR3‡ Yes No E. R. Spear sterilized (2) PNM 19, 2012 seeds were Tree sp: Media: planted in (1) DR (1) MEA+chloram forest sites Plant tissue: (1) stem, (2) root, (3) leaf 2019 Naturally (1) BCI 48 Sept. 27- 119 fungal (1) STRI mol. ITS5, ITS4 Will be Yes No E. R. Spear occurring Tree spp: Oct. 14, isolates & 2 research lab, deposited seedings (1) ANE, 2019 oomycetes (2) Broders Lab¶, upon (2) CL, Media: (3) Macrogen, Inc. manuscript (3) FO, (1) MEA+rifamp acceptance (4) LP, Plant tissue: . (5) MM, (1) stem, (6) PP, (2) root, (7) PR, (3) leaf (8) PT, (9) SPM, (10) VS 152 †While PARP (the isolation medium for 11 isolates) contains antifungals and was used with the intention of isolating oomycetes, only fungi were cultivated. We 153 believe there was a problem with the antifungals or medium preparation. All non-singleton fungal operational taxonomic units isolated on PARP were also 154 isolated on MEA+rifamp, MEA +chloram, and/or water agar (Fig. S4c). 155 ‡ For 17 isolates, paired-end reads were not possible due to a low quality read in one direction. For 62 isolates (amplified & sequenced in 2007, 2010-2012), 156 the reverse primer was LR3 and amplicons consisted of the ITS region and an adjacent portion of the large subunit (ITS+LSU) following U’ren et al., (2010). 157 § Methods and sequences previously published in Spear (2007). 158 ¶ Amplified DNA was generated by either direct colony polymerase chain reaction (DC-PCR; following Walch et al., 2016 ) or PCR of DNA extracted in TE (Tris- 159 EDTA) buffer. A T100™ Thermal Cycler (Bio-Rad Laboratories, Inc, Hercules, CA, USA) was used for amplification: 3 min of initial denaturation at 95 °C, followed 160 by 36 cycles of 95°C for 30 s, 54°C for 30 s, and 72 °C for 1 min, and a final extension step of 72°C for 10 min (modified from U’ren et al., 2010). Amplification 161 was verified with gel eletrophoresis and GelRed® Nucleic Acid Gel Stain (Biotium, Inc., Fremont, CA, USA). 162 163 References: 164 Gardes M, Bruns TD. 1993. ITS primers with enhanced specificity for basidiomycetes – application to the identification of 165 mycorrhizae and rusts. Molecular Ecology 2: 113-118.
15
166 Gilbert GS, Webb CO. 2007. Phylogenetic signal in plant pathogen-host range. Proceedings of the National Academy of Sciences, USA 167 104: 4979-4983. 168 Higginbotham S, Wong WR, Linington RG, Spadafora C, Iturrado L, Arnold AE. 2014. Sloth hair as a novel source of fungi with 169 potent antiparasitic, anti-cancer and anti-bacterial bioactivity. PloS ONE 9: e84549. 170 Vilgalys R, Hester M. 1990. Rapid genetic identification and mapping of enzymatically amplified ribosomal DNA from several 171 Cryptococcus species. Journal of Bacteriology 172: 4238–4246. 172 Spear ER. 2017. Phylogenetic relationships and spatial distributions of putative fungal pathogens of seedlings across a rainfall 173 gradient in Panama. Fungal Ecology 26: 65-73. 174 U’Ren, J. M., F. Lutzoni, J. Miadlikowska, and A. E. Arnold. 2010. Community analysis reveals close affinities between endophytic 175 and endolichenic fungi in mosses and lichens. Microbial Ecology 60: 340–353. 176 Walch G, Knapp M, Rainer G, Peintner U. 2016. Colony-PCR is a rapid method for DNA amplification of Hyphomycetes. Journal of 177 Fungi 2: 12. 178 White TJ, Bruns T, Lee S, Taylor J. 1990. Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. In: 179 Innis N, Gelfand D, Sninsky J, White T, eds. PCR protocols: A guide to methods and applications. New York, NY, USA: Academic 180 Press, 315-322.
16
181 Table S3 Average light levels, air temperatures, and relative humidities of the two shadehouses
182 used for the inoculation experiments (Fig. S2) versus the forest understory. The screened-in
183 shadehouses are located in Gamboa, Panama (9°7'9.87"N, 79°42'4.96"W). The
184 photosynthetically active radiation (PAR, in µmol of photons s-1 m-2) reaching shadehouse
185 seedlings was measured during the afternoon of a uniformly overcast day (Oct. 12, 2011).
186 Measurements were taken inside and directly outside the shadehouses with a LI-250 light
187 meter, a LI-190 quantum sensor, and a one-meter LI-191 line quantum sensor (LI-COR, Lincoln,
188 NE, USA). The mean (± SD) wet season forest understory light value was obtained from Brenes-
189 Arguedas et al., (2011). In 2006 and 2007, Brenes-Arguedas et al., (2011) measured
190 instantaneous light 0.5 m above the forest floor with LI-190 quantum sensors (LI-COR), and QSO
191 sensors (Apogee Instruments, Logan, UT, USA) and CR200 and CR1000 data-loggers (Campbell
192 Scientific) in a nearby (within ca. 16 km) forest in Barro Colorado Nature Monument (BCNM),
193 Buena Vista Peninsula (9°11'N, 79°49'W). In each shadehouse, air temperature and relative
194 humidity (RH) were measured at 10-mimute intervals (CS500 probe, Campbell Scientific, Inc.,
195 Logan, UT, USA). Hourly mean temperature and minimum and maximum RH were recorded on
196 a CR200 datalogger (Campbell Scientific). Measurements were taken November 3-5 and 5-11,
197 2011 for shadehouses 2 and 1, respectively. We obtained forest understory air temperature
198 and RH data for the same time period (Nov. 3-11, 2011) from the Physical Monitoring Program
199 of the Smithsonian Tropical Research Institute (Paton, 2019a; Paton, 2019b) Forest understory
200 air temperature and RH data were measured at 15-minute intervals with a CS215 Temperature
201 and Relative Humidity Probe (Campbell Scientific) at a nearby (within ca. 16 km) forest in
202 BCNM, the Lutz Tower (sensor height 1m) on Barro Colorado Island (9°9'42.06"N,
203 79°50'15.83"W). We calculated hourly means from STRI’s timestamped 15-minute interval data
204 to allow for comparison with our air temperature and RH data. Both shadehouses were used for
205 the inoculation experiments in 2011 and only shadehouse 1 was used in 2012.
Location Mean ± SD Mean ± SD air Mean min. ± SD Mean max. ± SD % of full PAR temperature‡ RH‡ RH‡
Shadehouse 1 1.4% † 26.1 ± 1.9ºCns 85.1 ± 8.6%*** 88.1 ± 6.4%***
17
Forest understory 1.3 ± 0.8% 25.4 ± 0.9ºC 95.6 ± 0.5% 98.3 ± 1.2%
Shadehouse 2 1.7% † 25.6 ± 1.3ºCns 87.6 ± 6.9%*** 90.5 ± 4.6%***
Forest understory Same as above 25.3 ± 0.5ºC 96.7 ± 0.6% 99.0 ± 0.4% 206 †Because inside and outside PAR measurements were not taken simultaneously, we calculated mean % full PAR as 207 the average of the PAR values recorded inside (shadehouse 1: n = 13, shadehouse 2: n = 8) divided by the average 208 of the PAR values recorded outside (shadehouse 1: n = 9, shadehouse 2: n = 8) and we could not calculate SD. For 209 this reason, and because we did not have access to the raw data summarized in Brenes-Arguedas et al., (2011), we 210 did not use statistical tests to compare shadehouse and forest understory light levels. 211 ‡ The air temperature and relative humidity data are not normally distributed (assessed by variable and 212 shadehouse with Shapiro-Wilk tests of normality [stats package; R Development Core Team, 2020], all P < 0.01). 213 Therefore, two-tailed Wilcoxon rank-sum tests (stats package; R Core Team, 2020) were used to compare the air 214 temperature and RH conditions of the two shadehouses to those of the forest understory (ns = not significant, 215 ***P <0.001; Quinn & Keough, 2002). n = 150 per group for shadehouse 1 and its corresponding understory data. n 216 = 46 per group for shadehouse 2 and its corresponding understory data. 217 218 References: 219 Brenes-Arguedas T, Roddy AB, Coley PD, Kursar TA. 2011. Do differences in understory light 220 contribute to species distributions along a tropical rainfall gradient? Oecologia 166: 443- 221 456. 222 Paton S. 2019a. Barro Colorado Island, Lutz tower 1m_Air Temperature. The Smithsonian 223 Institution. Dataset. https://doi.org/10.25573/data.10042394.v7 224 Paton S. 2019b. Barro Colorado Island, Lutz tower 1m_Relative Humidity. The Smithsonian 225 Institution. Dataset. https://doi.org/10.25573/data.10042400.v7 226 Quinn GP, Keough MJ. 2002. Experimental design and data analysis for biologists. Cambridge, 227 UK: Cambridge University Press. 228 R Core Team. 2020. R: a language and environment for statistical computing. R Foundation for 229 Statistical Computing, Vienna, Austria.
18
230 Table S4 For each of the 66 OTUs, we report its estimated taxonomic placement and the UNITE database accession code(s)
231 associated with the reference sequence(s) used to assign nomenclature (Kõljalg et al., 2013), the number of times it was isolated,
232 and the number of tree species from which it was isolated. For the 33 non-singleton OTUs (data subset A: n = 178 isolates), we
233 report estimated host specialization based on the d' index (value and category; low (L): 0–0.33, moderate (M): 0.34–0.67). To
234 determine if there is a phylogenetic signal to the host range of the 31 OTUs isolated from more than one host species (data subset C:
235 n = 174 isolates), we used the ses.mpd function (‘picante’ package; Kembel et al., 2010). In that analysis, observed mean
236 phylogenetic distances (MPD) between host tree species infected by a given OTU were compared to mean phylogenetic distances
237 generated in a null model with random host-OTU associations (expected mean and standard deviation). Negative and positive
238 standardized effect sizes indicate phylogenetic clustering and phylogenetic evenness of host use, respectively. (i.e., a given OTU
239 associates with hosts more closely [Obs. MPD < Exp. MPD] or more distantly [Obs. MPD > Exp. MPD] related than expected by
240 chance) (Kembel et al., 2010). The P value indicates probability that the deviation from the expected MPD is due to chance. R-
241 Diaporthe fraxini-angustifoliae and Q-Diaporthe cf. columnaris are the only OTUs that exhibit marginally significant phylogenetic
242 clustering (bolded).
No. of Host Null Null obs. specialization: model model No. of host d' index & Obs. mean SD of Standardized OTU - Estimated taxonomic placement - UNITE code† isolates spp. category MPD MPD MPD effect size P A - Colletotrichum xanthorrhoeae - SH1543705 (C. 35 9 0.225 - L 219.3 233.0 21.2 -0.645 0.34 xanthorrhoeae), SH1543739 (unidentified) B - Nectriaceae sp. - SH1610517 (Nectriaceae), SH1610162 15 8 0.292 - L 278.8 233.0 23.3 1.969 0.97 (Cylindrocladium buxicola) C - Mycoleptodiscus suttonii - SH1562616 13 7 0.538 - M 262.4 233.8 26.8 1.068 0.85 F - Diaporthe tulliensis - SH1540611 11 5 0.347 - M 223.3 234.7 34.7 -0.328 0.49 I - Sordariomycetes sp. - SH1541118 7 6 0.402 - M 230.6 235.0 29.2 -0.149 0.56 G - Clonostachys rosea - SH1522825 6 4 0.274 - L 280.0 232.4 39.4 1.209 0.81
19
D - Cylindrocladiella variabilis - SH1610166 6 3 0.350 - M 307.8 230.5 48.6 1.591 0.85 H - Xylariaceae sp. - SH1541124 6 3 0.242 - L 234.4 235.0 49.2 -0.013 0.60 AAD - Mycoleptodiscus suttonii - SH1562616 5 4 0.272 - L 233.9 232.5 40.0 0.035 0.66 AU - Beltrania pseudorhombica - SH1563660 5 4 0.061 - L 299.2 232.2 38.4 1.744 0.95 J - Pseudopestalotiopsis theae - SH1552673 5 4 0.206 - L 279.8 233.4 39.9 1.162 0.78 L - Neopestalotiopsis foedans - SH1552672 (N. foedans), 5 4 0.031 - L 217.5 234.1 40.0 -0.416 0.41 SH2700365 (Pezizomycotina) M - Lasiodiplodia gonubiensis - SH1507365 5 3 0.126 - L 320.3 232.1 47.3 1.866 0.94 S - Colletotrichum citricola - SH1543707 4 4 0.198 - L 233.9 233.1 38.6 0.021 0.66 N - Gliocladiopsis elghollii - SH1546330 4 3 0.669 - M 204.0 235.6 51.0 -0.619 0.30 E - Beltrania pseudorhombica - SH1563660 4 2 0.350 - M 238.3 232.2 68.4 0.090 0.56 O - Fusarium pseudensiforme - SH1546322 (F. 4 2 0.593 - M 235.1 233.9 67.4 0.017 0.36 pseudensiforme), SH1546498 (Hypocreales) AV - Cylindrocladiella variabilis - SH1212036 3 3 0.122 - L 320.3 232.9 48.1 1.817 0.97 AW - Calonectria pseudonaviculata - SH1610162 3 3 0.118 - L 237.2 232.4 47.3 0.103 0.64 AX - Nectriaceae sp. - SH1212162 (Neonectria sp.), 3 3 0.511 - M 237.2 234.6 49.4 0.053 0.67 SH1610423 (Cylindrocladiella sp.), SH1212036 (C. variabilis) AY - Beltraniella cf. endiandrae - SH1563659 3 3 0.383 - M 196.6 233.7 50.1 -0.741 0.17 P - Colletotrichum thailandicum - SH1543711 3 3 0.083 - L 237.2 234.8 50.2 0.048 0.76 Q - Diaporthe cf. columnaris - SH1540633 (Diaporthales), 3 2 0.226 - L 119.7 232.1 68.4 -1.644 0.09 SH1540609 (D. columnaris) AAA - Diaporthe fraxini-angustifoliae - SH1540607 2 2 0.481 - M 235.1 235.3 71.9 -0.003 0.49 AAB - Trichoderma spirale - SH1567965 (T. spirale), 2 2 0.156 - L 215.2 232.8 67.1 -0.262 0.24 SH1552633 (Pezizomycotina) AAC - Xylariaceae sp. - SH1541166 2 2 0.380 - M 361.3 234.9 68.6 1.841 0.87 AZ - Clonostachys cf. miodochialis - SH1522826 2 2 0.530 - M 214.0 233.4 69.5 -0.280 0.18 (Hypocreales), SH1522827 (C. miodochialis) K - Trichoderma spirale - SH1567965 2 2 0.570 - M 238.3 234.1 71.1 0.059 0.84 R - Diaporthe fraxini-angustifoliae - SH1540607 2 2 0.086 - L 119.7 232.1 68.4 -1.644 0.09 U - Colletotrichum magnisporum - SH1543718 2 2 0.118 - L 238.3 232.6 70.1 0.082 0.67 Xylaria multiplex V - - SH1541132 2 2 0.134 - L 238.3 233.6 65.6 0.073 0.73 T‡ - Macrophomina - SH1507375 (Botryosphaeriaceae), 2 1 0.343 - M SH1507369 (M. phaseolina) X‡ - Ceratobasidium sp. - SH1551758 2 1 0.164 - L
20
Diaporthe endophytica AA - - SH1540603 1 1 Colletotrichum thailandicum AAE - - SH1543711 1 1 AAF - Nectriaceae sp. - SH1610517 1 1 Colletotrichum brevisporum AAG - - SH1543708 1 1 Beltraniella endiandrae AAH - - SH1563659 1 1 Diaporthe columnaris AAI - cf. - SH1540609 1 1 Gliocladiopsis elghollii AAJ - - SH1546330 1 1 Cylindrocladiella variabilis AAK - - SH1212036 1 1 AAL - Sordariomycetes sp. - SH1198320 1 1 Talaromyces AAM - - SH1516144 1 1 Cylindrocladiella variabilis AAN - - SH1212036 1 1 Diaporthe siamensis AB - - SH1540610 1 1 AC - Nigrospora cf. oryzae - SH1549605 (unidentified fungus), 1 1 SH1549606 (N. oryzae) Gliocladiopsis AD - sp. - SH1546383 1 1 Pestalotiopsis AE - sp. - SH1563667 1 1 Fusarium equiseti AF - - SH1610158 1 1 Pestalotiopsis rhododendri AG - - SH1563658 1 1 AH - Oomycota sp. 1 1 Gliocephalotrichum cylindrosporum AI - - SH1546353 1 1 Beltraniopsis neolitseae AJ - - SH1563663 1 1 Diaporthe endophytica AK - - SH1540603 1 1 Digitiseta multidigitata AL - - SH1546391 1 1 Endomelanconiopsis endophytica AM - - SH1507376 1 1 Phytophthora palmivora AN - 1 1 Xylaria AO - sp. - SH1554119 1 1 Diaporthe endophytica AP - - SH1540603 1 1 Ramularia AQ - sp. - SH1209934 1 1 AR - Mycosphaerellaceae sp. - SH1606644 1 1 Fusarium AS - - SH1610159 1 1 AT - Chaetosphaeriaceae sp. - SH1168551 1 1 W - Sordariomycetes sp. - SH1198320 1 1 Y - Talaromyces marneffei - SH1516144 1 1
21
Z - Colletotrichum brevisporum - SH1543708 1 1 †All UNITE Species Hypothesis accession codes, beginning with acronym SH, end in ".08FU" (not shown), denoting the version number and the acronym for Fungi (Kõljalg et al., 2013). Because our OTU grouping strategy (99% sequence similarity) often differed from that of the UNITE database (97-100% similarity), multiple, distinct OTUs in our study sometimes share the same nomenclature based on the species hypothesis (SH) of the best-matching reference sequence in UNITE (e.g., OTUs D, AV, AAK, and AAN). ‡There are no ses.mpd results for T - Macrophomina sp. and X - Ceratobasidium sp. because those OTUs were only isolated from a single host species. 243 244 References 245 Kembel SW, Cowan PD, Helmus MR, Cornwell WK, Morlon H, Ackerly DD, Blomberg SP, Webb CO. 2010. Picante: R tools for 246 integrating phylogenies and ecology. Bioinformatics 26: 1463-1464. 247 Kõljalg U, Nilsson RH, Abarenkov K, Tedersoo L, Taylor AFS, Bahram M, Bates ST, Bruns TD, Bengtsson-Palme J, Callaghan TM et al. 248 2013. Towards a unified paradigm for sequence-based identification of fungi. Molecular Ecology 22: 5271–5277. 249
22
250 Table S5 Overlap in OTUs among tree species, considering only the 13 tree species and 22 OTUs with 3 or more observations (data
251 subset B n = 144 isolates). Gray cells contain the number of observed OTUs and, in parentheses, isolates collected for each tree
252 species (original hosts in Table S1). White cells contain the shared richness of OTUs between pairs of tree species and, in
253 parentheses, the estimated OTU community similarity (1 – Chao index, which is abundance-based and adjusted for unseen species),
254 ranging from zero (no similarity) to one (identical communities). Black cells contain the pooled number of OTUs for each pair of tree
255 species. ANE CL CM CE CV DR FO HC LP PR PP PT VS ANE 13 (29) 3 (0.30) 0 (0) 1 (0.03) 2 (0.35) 4 (0.47) 3 (0.34) 0 (0) 2 (0.36) 7 (0.81) 4 (0.36) 2 (0.23) 7 (1) CL 17 7 (12) 0 (0) 0 (0) 1 (0.10) 2 (0.24) 2 (0.35) 0 (0) 2 (0.33) 7 (1) 3 (0.42) 2 (0.32) 2 (0.17) CM 15 9 2 (3) 1 (0.24) 0 (0) 2 (0.54) 0 (0) 1 (0.24) 0 (0) 1 (0.04) 0 (0) 0 (0) 2 (0.18) CE 14 9 3 2 (4) 0 (0) 1 (0.11) 0 (0) 1 (0.21) 0 (0) 0 (0) 0 (0) 0 (0) 1 (0.09) CV 15 10 6 6 4 (4) 1 (0.08) 0 (0) 1 (0.21) 3 (1) 1 (0.08) 0 (0) 0 (0) 1 (0.16) DR 16 12 7 8 10 7 (16) 2 (0.32) 1 (0.11) 1 (0.07) 4 (0.63) 2 (0.23) 1 (0.18) 2 (0.15) FO 14 9 6 6 8 9 4 (10) 0 (0) 0 (0) 2 (0.32) 1 (0.27) 2 (0.61) 1 (0.07) HC 16 10 4 4 6 9 7 3 (4) 1 (0.16) 0 (0) 0 (0) 1 (0.18) 1 (0.09) LP 15 9 6 6 5 10 8 6 4 (6) 2 (0.21) 1 (0.13) 0 (0) 1 (0.08) PR 19 13 14 15 16 16 15 16 15 13 (25) 5 (0.59) 3 (0.42) 6 (1) PP 15 10 8 8 10 11 9 9 9 14 6 (13) 1 (0.21) 1 (0.06) PT 15 9 6 6 8 10 6 6 8 14 9 4 (5) 1 (0.14) VS 15 14 9 10 12 14 12 11 12 16 14 12 9 (13) 256
23
257 Table S6 Results of the beta-binomial (logit link) generalized linear regression with the
258 proportion of seedlings with disease as a function of seed size and shade tolerance (26 tree
259 species, 179 observations: 42 for intolerant spp., 137 for tolerant spp.). Estimates and standard
260 errors (SE) are log odds of disease for a one-unit increase in the variable. The intercept
261 represents shade-intolerant tree species at a hypothetical seed dry mass (mg) of zero. Model
262 averaging was not done because no other model had a ΔAICc ≤ 2. P-values indicating statistical
263 significance are in bold (P < 0.05). Variable Estimat SE z value P e intercept -0.812 0.342 -2.377 0.018 seed size -0.004 0.001 -3.057 0.002 shade tolerance -1.131 0.385 -2.938 0.003 seed size:shade tolerance 0.004 0.001 -2.457 0.014 264
265 Table S7 Average estimates based on the best-ranked (ΔAICc ≤ 2) beta-binomial (logit link)
266 generalized linear regressions with the proportion of seedlings with disease as a function of
267 seed size and spatial distribution relative to annual rainfall (15 tree species, 154 observations:
268 108 for dry-site spp., 46 for wet-site spp.). Estimates and standard errors are log odds of
269 disease for a one-unit increase in the variable. The intercept represents a dry-site tree species
270 at a hypothetical seed dry mass (mg) of zero. P-values indicating statistical significance are in
271 bold (P < 0.05). Variable Estimate Std. error z value P intercept -1.517 0.222 6.829 <0.001 seed size -0.0005 0.0003 1.593 0.111 distribution -1.255 0.619 2.028 0.042 seed size:distribution 0.0008 0.0006 1.286 0.198 272
24
273 Methods S1 Methods used to estimate the taxonomic placement of the 66 OTUs and assign
274 nomenclature.
275 We estimated the taxonomic placement of the fungi isolated from symptomatic
276 seedlings by querying all 209 fungal sequences in our study against the well-curated and
277 annotated Full UNITE+INSD dataset for Fungi (v. 8.2, released 2020-02-04; Abarenkov et al.,
278 2020). Fungal sequences accessioned in the UNITE database have been clustered into what are
279 hypothesized to be species-level groups (Species Hypotheses, SHs), each with a unique
280 accession code. We implemented BLASTN (Altschul et al., 1990) in Python with default
281 parameters (blastn -query Spear.fasta -task blastn -db uniteDB -out
282 Python_blastout_1Oct20.txt -evalue 10 -outfmt ‘6 qseqid pident qcovs evalue bitscore length
283 sallseqid’ -max_target_seqs 5 -num_threads 16). For each query, we considered the top five
284 hits and an OTU was assigned nomenclature based on the reference sequence meeting the
285 following criteria: an alignment length ≥75 bp (range 374-1148 bp, mean 759 bp, median 618
286 bp), an E-value ≤ 10-36 (all <10-150) (as in Radujković et al., 2019), query cover ≥90% (range 91-
287 100%, mean 99.5%, median 100%), and the highest percent identity for the OTU (range 93-
288 100%, mean 99.7%, median 100%) (as suggested by Lücking et al., 2020).
289 Following the aforementioned steps of our initial, Python-based approach, there was
290 taxonomic ambiguity for 18 of the 64 fungal OTUs. These OTUs had multiple (2-3) different SH
291 accession codes among the hits with a percent identity of 100% and/or they were not identified
292 to the genus level based on the SH of their “best” hit. In an attempt to resolve ambiguities and
293 achieve the greatest taxonomic resolution possible, we queried one to two representative
294 sequence(s) from each of those 18 OTUs against the web-based UNITE database (v. 8.2,
295 accessed 9-Oct and 16-Nov-2020; Nilsson et al., 2018), reviewed the top 30 hits for each
296 sequence, and, when possible and appropriate (e.g., no taxonomic conflicts among the named
297 matches), revised their nomenclature based on the criteria specified above (e.g., OTU AC was
298 revised from “unidentified fungus” to Nigrospora cf. oryzae). As a result, we manually assigned
299 the nomenclature for 10 of the 18 OTUs with taxonomic ambiguity following our Python-based
300 approach. For those 10 OTUs, we list all relevant SH accession codes (i.e., the SH of the “best”
301 hit and the SH(s) of the reference sequence(s) used to assign nomenclature) in Table S4.
25
302 For the two oomycetes (OTUs AH and AN), their sequences were queried against
303 GenBank (accessed 8-Oct-2020; Benson et al., 2012) and the curated database Phytopthora-ID
304 (v. 2.0, accessed 8-Oct-2020; Grünwald et al., 2011).
305 When annotating OTUs based on reference sequences named to the species level, we
306 considered: (i) 99-100% percent identity a positive match between the query and reference
307 sequence; (ii) 97-98.99% percent identity a close match, the observed differences may fall
308 within the variability of the species (5 OTUs; annotated as cf.); and 90-96.99% percent identity
309 to be a positive match at the genus, but not species, level (3 OTUs: T, AS, AAM).
310 Dual nomenclature for pleomorphic fungi (i.e., different scientific names for the asexual
311 and sexual forms of single species) has been replaced with a single name for a fungal species
312 (Hawksworth et al., 2011). For each OTU, we verified the current accepted name by reviewing
313 Index Fungorum (www.indexfungorum.org), Mycobank (Crous et al., 2004), and published
314 literature, and we revised the nomenclature for OTU AF (from Gibberella intricans to Fusarium
315 equiseti; Xia et al., 2019), OTU AQ (from Mycosphaerella to Ramularia; Wijayawardene et al.,
316 2014), and OTU AW (from Cylindrocladium buxicola to Calonectria pseudonaviculata; Lombard
317 et al., 2010). Additionally, we followed the nomenclature of Hernández-Restrepo et al., (2019)
318 for the taxonomic ranks of Mycoleptodiscus.
319 All nomenclature should be interpreted as an estimate due to several limitations
320 associated with our sequence-based approach (Kang et al., 2010; Hofstetter et al., 2019;
321 Lücking et al., 2020). While the nuclear ribosomal internal transcribed spacer (ITS) region is the
322 formal barcode for the molecular identification of fungi (Schoch et al., 2012). It is well
323 established that the ITS region is an unreliable barcode for species discrimination for certain
324 taxa (Lücking et al., 2020); for example, Calonectria (Liu et al., 2020), Diaporthe (Santos et al.,
325 2017), Colletotrichum (Marin-Felix et al., 2017), and Penicillium (Seifert et al., 2007). A multi-
326 locus phylogenetic evaluation is required for accurate species identification (sensu Santos et al.,
327 2017). Additionally, precision is limited by the incomplete taxonomic and geographic coverage
328 of existing databases, even well-curated ones like UNITE (Kõljalg et al., 2013; Lücking et al.,
329 2020). Additionally, fungal sequences accessioned in the UNITE database have been clustered
330 into what are hypothesized to be species-level groups (Species Hypotheses, SHs) based on
26
331 sequence similarity thresholds ranging from 97–100% (Kõljalg et al., 2013; Robbertse et al.,
332 2017; Nilsson et al., 2018). While there is no single threshold that appropriately addresses
333 lineage-specific intra- versus interspecific variability for all fungi (Nilsson et al., 2008), we used a
334 threshold value of 99% sequence similarity to designate OTUs because: (1) 97-98.5% sequence
335 similarity is too relaxed for species delimitation for some taxonomic groups (e.g., Garnica et al.,
336 2016) and (2) we are making statements about host specificity so we adopted a stringent
337 similarity threshold that splits rather than lumps, but that accounts for a small amount of
338 sequencing error . Because our grouping strategy (99% sequence similarity) often differed from
339 that of the UNITE database (97-99% similarity), multiple, distinct OTUs in our study sometimes
340 share the same nomenclature based on the SH of the best-matching reference sequence in
341 UNITE (e.g., OTUs D, AV, AAK, and AAN are all annotated as Cylindrocladiella variabilis). It
342 should also be noted that, for certain groups of fungi (e.g., Trichoderma), UNITE species
343 hypotheses (SHs) may erroneously include distinct species (Robbertse et al., 2017; Lücking et
344 al., 2020).
345 References: 346 Abarenkov K, Zirk A, Piirmann T, Pöhönen R, Ivanov F, Nilsson RH, Kõljalg U. 2020. Full 347 UNITE+INSD dataset for Fungi. UNITE Community. 10.15156/BIO/786372 348 Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ. 1990. Basic local alignment search tool. 349 Journal of Molecular Biology 215: 403-410. 350 Benson DA, Cavanaugh M, Clark K, Karsch-Mizrachi I, Lipman DJ, Ostell J, Sayers EW. 2012. 351 GenBank. Nucleic Acids Research 41: D36-D42. 352 Crous PW, Gams W, Stalpers JA, Robert V, Stegehuis G. 2004. MycoBank: an online initiative to 353 launch mycology into the 21st century. Studies in Mycology 50: 19-22. 354 Garnica S, Schön ME, Abarenkov K, Riess K, Liimatainen K, Niskanen T, Dima B, Soop K, 355 Frøslev TG, Jeppesen TS et al. 2016. Determining threshold values for barcoding fungi: 356 lessons from Cortinarius (Basidiomycota), a highly diverse and widespread 357 ectomycorrhizal genus. FEMS Microbiology Ecology 92: fiw045. 358 Grünwald NJ, Martin FN, Larsen MM, Sullivan CM, Coffey MD, Hansen EM, Parke JL. 2011. 359 Phytophthora-ID.org: A sequence-based Phytophthora identification tool. Plant Disease 360 95: 337-342. 361 Hawksworth DL. 2011. A new dawn for the naming of fungi: impacts of decisions made in 362 Melbourne in July 2011 on the future publication and regulation of fungal names. IMA 363 Fungus 2: 155-162. 364 Hernández-Restrepo M, Bezerra JDP, Tan YP, Wiederhold N, Crous PW, Guarro J, Gené J. 365 2019. Re-evaluation of Mycoleptodiscus species and morphologically similar fungi. 366 Persoonia 42: 205-227.
27
367 Hofstetter V, Buyck B, Eyssartier G, Schnee S, Gindro K. 2019. The unbearable lightness of 368 sequenced-based identification. Fungal Diversity 96: 243-284. 369 Kang S, Mansfield MA, Park B, Geiser DM, Ivors KL, Coffey MD, Grünwald NJ, Martin FN, 370 Lévesque CA, Blair JE. 2010. The promise and pitfalls of sequence-based identification 371 of plant-pathogenic fungi and oomycetes. Phytopathology 100: 732-737. 372 Kõljalg U, Nilsson RH, Abarenkov K, Tedersoo L, Taylor AFS, Bahram M, Bates 373 ST, Bruns TD, Bengtsson-Palme J, Callaghan TM et al. 2013. Towards a unified paradigm 374 for sequence-based identification of fungi. Molecular Ecology 22: 5271–5277. 375 Liu QL, Li JQ, Wingfield MJ, Duong, TA, Wingfield BD, Crous PW, Chen SF. 2020. 376 Reconsideration of species boundaries and proposed DNA barcodes for Calonectria. 377 Studies in Mycology 100106. 378 Lombard L, Crous PW, Wingfield BD, Wingfield MJ. 2010. Phylogeny and systematics of the 379 genus Calonectria. Studies in Mycology 66: 31-69. 380 Marin-Felix Y, Groenewald JZ, Cai L, Chen Q, Marincowitz S, Barnes I, Bensch K, Braun U, 381 Camporesi E, Damm U et al. 2017. Genera of phytopathogenic fungi: GOPHY 1. Studies 382 in Mycology 86: 99-216. 383 Lücking R, Aime MC, Robbertse B, Miller AN, Ariyawansa HA, Aoki T, Cardinali G, Crous PW, 384 Druzhinina IS, Geiser DM. 2020. Unambiguous identification of fungi: where do we 385 stand and how accurate and precise is fungal DNA barcoding? IMA Fungus 11: 1-32. 386 Nilsson RH, Kristiansson E, Ryberg M, Hallenberg N, Larsson K-H. 2008. Intraspecific ITS 387 variability in the kingdom Fungi as expressed in the international sequence databases 388 and its implications for molecular species identification. Evolutionary Bioinformatics 389 Online 4: 193-201. 390 Nilsson RH, Larsson K-H, Taylor AFS, Bengtsson-Palme J, Jeppesen TS, Schigel D, Kennedy P, 391 Picard K, Gl€ockner FO, Tedersoo L et al. 2018. The UNITE database for molecular 392 identification of fungi: handling dark taxa and parallel taxonomic classifications. Nucleic 393 Acids Research 47: D259–D264. 394 Radujković D, van Diggelen R, Bobbink R, Weijters M, Harris J, Pawlett M, Vicca S, Verbruggen 395 E. 2020. Initial soil community drives heathland fungal community trajectory over 396 multiple years through altered plant–soil interactions. New Phytologist 225: 2140-2151. 397 Robbertse B, Strope PK, Chaverri P, Gazis R, Ciufo S, Domrachev M, Schoch CL. 2017. 398 Improving taxonomic accuracy for fungi in public sequence databases: applying ‘one 399 name one species’ in well-defined genera with Trichoderma/Hypocrea as a test case. 400 Database 2017 401 Rossman AY, Seifert KA, Samuels GJ, Minnis AM, Schroers HJ, Lombard L, Crous PD, Põldmaa 402 K, Cannon PF, Summerbell RC et al. 2013. Genera in Bionectriaceae, Hypocreaceae, and 403 Nectriaceae (Hypocreales) proposed for acceptance or rejection. IMA Fungus 4: 41-51. 404 Santos L, Alves A, Alves R. 2017. Evaluating multi-locus phylogenies for species boundaries 405 determination in the genus Diaporthe. PeerJ 5: e3120. 406 Schoch CL, Seifert KA, Huhndorf S, Robert V, Spouge JL, Levesque CA, Chen W. 2012. Nuclear 407 ribosomal internal transcribed spacer (ITS) region as a universal DNA barcode marker for 408 Fungi. Proceedings of the National Academy of Sciences, USA 109: 6241–6246. 409 Seifert KA, Samson RA, Dewaard JR, Houbraken J, Lévesque CA, Moncalvo JM, Louis-Seize G, 410 Hebert PD. 2007. Prospects for fungus identification using CO1 DNA barcodes, with
28
411 Penicillium as a test case. Proceedings of the National Academy of Sciences, USA 104: 412 3901-3906. 413 Wijayawardene NN, Crous PW, Kirk PM, Hawksworth DL, Boonmee S, Braun U, Dai D-Q, 414 D’souza MJ. Diederich P, Dissanayake A et al. 2014. Naming and outline of 415 Dothideomycetes–2014 including proposals for the protection or suppression of generic 416 names. Fungal Diversity 69: 1-55. 417 Xia JW, Sandoval-Denis M, Crous PW, Zhang XG, Lombard L. 2019. Numbers to names– 418 restyling the Fusarium incarnatum-equiseti species complex. Persoonia 43: 186-221.
29