Fungal and Oomycete cardinal temperatures (the Togashi dataset)

We collated and analysed temperature responses, specifically the minimum (Tmin), optimum (Topt) and maximum (Tmax) temperatures that comprise the ‘cardinal temperatures’, of various biological processes for 695 plant-associated microbes (631 fungi and 64 oomycetes) reported in "Togashi, K. (1949). Biological characters of plant pathogens: temperature relations. Meikundo". Cardinal temperatures can be used to derive temperature response functions, or thermal performance curves, using mathematical forms such as the beta function. The biological processes for which cardinal temperatures have been measured vary in their degree of host interaction. Experimental measurements for rates of growth in culture and often spore germination occur under axenic conditions, while infection and disease development occur as interactions with the host plant. Fruiting body formation, or fructification, and sporulation may or may not be measured in planta depending on experimental conditions.

Methods
The Togashi Dataset

Minimum (Tmin), optimum (Topt) and maximum (Tmax) temperatures (collectively ‘cardinal temperatures’) of five life-cycle processes (disease development (DD), fructification (FR), infection (IN), spore germination (SG) and sporulation (SP)), as well as growth in culture (GC) (collectively ‘biological processes’), were extracted and thence digitised from ref. 1 for fungi and oomycete species - henceforth the Togashi dataset. Biological processes such as wood decay, spore discharge, enzyme production, and saltation were excluded due to paucity of data. In brief, ref.1 is a compilation of published literature regarding plant pathogen temperature relations, published (in print) in 1949. Ref.1 contains over 300 pages of data from over 1000 publications (published in the 19th and 20th century).

The Index Fungorum (IF) and associated Species Fungorum (SF) databases (www.indexfungorum.org; www.speciesfungorum.org) were used to identify current synonymous names for each species recorded in the Togashi dataset [accessed between 8/5/2020 – 15/5/2020]. Where no current name was available, or species authorship name(s) were greatly inconsistent (i.e. no similarity to that cited in ref.15), the Mycobank database (www.mycobank.org) was used as an alternative [accessed between 8/5/2020 – 15/5/2020]. Species were recorded by their current name, according to the IF/SF or Mycobank databases (as above). Where no current name was explicitly provided, but the species could be identified, the species name searched and located was assumed to be current and correct. However, such records were treated as ambiguous. Similarly, if a species could be identified but species authorship name(s) cited in ref.1 showed no similarity to that on the IF/SF and/or Mycobank databases, or where species authorship name(s) were not provided by ref. 1, the species was included but treated as ambiguous. Where spelling of species names differed between ref. 1 and the IF/SF and/or Mycobank databases, but it was possible that the spelling in ref.1 was an error, species names were updated to reflect this, but treated as ambiguous. If a species cited in ref. 1 could not be identified at all on the IF/SF or Mycobank databases (i.e. where no synonymous names assigned by ref. 1 could be identified), it was excluded from the dataset, except in a few cases, where alternative resources were used to cross reference species names (see Togashi dataset for further details). In some cases, species were recorded in ref.1 under multiple, synonymous names. However, said species were at times found to be not synonymous, once identified in the IF/SF and/or Mycobank databases. Various methods were used to correct for this, detailed below.

If cardinal temperature data in ref.1 specifically referred to one of the nonsynonymous species, these were recorded under the currently designated name for that species on the IF/SF or Mycobank databases (as explained above). For example, GC cardinal temperature data were recorded for Fusarium sambucinum (Fuck.) [syn. Fusarium discolour var. sulphureum (App. et Wr.), Fusarium polymorphum (Mart.), Fusarium roseum (Lk.), Fusarium sulphureum (Schlecht)]. The IF/SF database classified F. sambucinum as F. roseum Link (1890) but F. sulphureum as F. sulphureum Schltdl. (1824). However, a subset of cardinal temperature data in ref.1 were recorded “as sulphureum”, and so were assigned to F. sulphureum, and not F. roseum in the Togashi dataset. In contrast, if cardinal temperature data associated with multiple, nonsynonymous species did not explicitly specify which species the data referred to, two alternative methods were used for clarification. First, the titles of publications cited in ref.1 for that species record were cross-referenced, to determine if a species name (or disease name that likely suggested a species) was provided in the title. If so, species names were corrected to match that of the publication title, for that data point. For example, GC cardinal temperature data were recorded by ref.1 for Corticium vagum (Berk. et Curt.) [syn. Rhizoctonia solani (Kühn)]. The IF/SF databases classified the former as Botryobasidium vagum ((Berk. & M.A. Curtis) D.P. Rogers (1935)), but the latter as R. solani (J.G. Kühn, (1858)). A subset of titles from publications used by ref.1 to extract cardinal temperature data contained the term “Rhizoctonia solani”, and so such data were here assigned to R. phaseoli, and not B. vagum. However, if no usable information was provided in publication titles, data were here recorded under the first species given by ref.1. This name was chosen as it was the bold, title name given to that species record in ref.1. For example, in ref.1 GC cardinal temperature data were recorded for Fusarium redolens (Wr.) [syn. Fusarium reticulatum (Mont.); Fusarium spinaciae (Sherb.)]. The IF/SF databases classified neither F. reticulatum nor F. spinaciae as synonymous with F. redolens. However, it was not clear in ref.1 which cardinal temperatures referred to which species, and titles of publications used to extract cardinal temperature data by ref.1 only stated “Fusarium”. Hence, all data were here recorded under the bold, title species name - F. redolens. Any records that underwent additional processing outlined here were also deemed ambiguous. Any synonymous species names assigned by ref.1 that could not be identified were deemed nonsynonymous. Where species in ref.1 were recorded under multiple species names, that were here found to be synonymous, we assumed all publications used by ref.1 to extract cardinal temperature data refer only to these species.

All cardinal temperature data concerning Fusarium oxysporum formae speciales were recorded under F. oxysporum, as well as their respective formae speciales.

The methods of determining fungi and oomycete species names resulted in cardinal temperature data for 695 species (631 fungi and 64 oomycetes, N = 8656) being recorded in the Togashi dataset. Previous analyses of thermal responses have considered only a handful of fungi and no oomycetes2–4. When data of ambiguous species records (explained above) were excluded, 568 species (514 fungi and 54 oomycetes, N = 6045) remained in the Togashi dataset. Excluding ambiguous species records had little influence on our results. All information regarding how species were named in the Togashi dataset, including species authorship name(s) cited in ref.1 and the various databases detailed above, changes to spelling of species names cited in ref.15, apparent synonymous and nonsynonymous species names cited in ref.15, cases where data were extracted from one species record and recorded as a different species, and species records treated as ambiguous, can be found in the Supplementary Data Togashi.

For each data point recorded in the Togashi dataset, where ref.1 recorded that the true value lies above or below the value provided, the value provided was recorded. For example, if Tmin was recorded as ‘below 8 °C’, 8 °C was recorded as Tmin; if Tmax was recorded as ‘above 25 °C’, 25 °C was recorded as Tmax. Where a range was provided, the mid-point was recorded. However, where a range was provided, but the true value was recorded to lie above or below this, the upper or lower limit was chosen, respectively. For example, if Tmin was quoted as ‘below 18 - 20 °C’, 18 °C was recorded as Tmin. Where a range was quoted for the entire biological process, the upper and lower bounds were recorded as Tmax and Tmin, respectively. For example, if IN was quoted as ‘occurring between 5 - 35 °C’, 5 °C was recorded as Tmin and 35 °C was recorded as Tmax. However, in cases where it was likely that the temperature range quoted referred to a range of optimal conditions, the mid-point was recorded as Topt. Cardinal temperatures were also estimated from prose in ref.1. Data under ‘Infection and disease development’ were independently recorded under IN and DD, unless the text specifically indicated one of these processes. Data recorded as ‘Specialization and resistance” were recorded under IN and/or DD, where appropriate. Data quoted in ref.1 that were the result of complex treatments and/or were not likely related to Tmin, Topt, or Tmax were excluded. Further information regarding how each cardinal temperature data point in the Togashi dataset was determined from information provided in ref.1 is reported in Supplementary Data Togashi. Where multiple references were provided for a single data point in ref.1, this was taken to represent independent observations, and so were individually included in the Togashi dataset. All data extraction was completed by the same researcher.

References

(1)       Togashi, K. Biological Characters of Plant Pathogens: Temperature Relations; Meibundo: Tokyo, 1949.

(2)       Alster, C. J.; Weller, Z. D.; Fischer, J. C. von. A Meta-Analysis of Temperature Sensitivity as a Microbial Trait. Global Change Biology 2018, 24 (9), 4211–4224. https://doi.org/10.1111/gcb.14342.

(3)       Dell, A. I.; Pawar, S.; Savage, V. M. The Thermal Dependence of Biological Traits. Ecology 2013, 94 (5), 1205–1206. https://doi.org/10.1890/12-2060.1.

(4)       Storch, D.; Menzel, L.; Frickenhaus, S.; Pörtner, H.-O. Climate Sensitivity across Marine Domains of Life: Limits to Evolutionary Adaptation Shape Species Interactions. Global Change Biology 2014, 20 (10), 3059–3067. https://doi.org/10.1111/gcb.12645.