Plant secondary metabolites and low temperature are the major limiting factors for Beauveria bassiana (Bals.-Criv.) Vuill. (Ascomycota: Hypocreales) growth and virulence in a bark beetle system

Journal Pre-proofs Plant secondary metabolites and low temperature are the major limiting factors for Beauveria bassiana (Bals.-Criv.) Vuill. (Ascomycota: Hypocreales) growth and virulence in a bark beetle system Andrew J. Mann, Thomas Seth Davis PII: DOI: Reference:

S1049-9644(19)30463-3 https://doi.org/10.1016/j.biocontrol.2019.104130 YBCON 104130

To appear in:

Biological Control

Received Date: Revised Date: Accepted Date:

6 July 2019 8 October 2019 22 October 2019

Please cite this article as: Mann, A.J., Davis, T.S., Plant secondary metabolites and low temperature are the major limiting factors for Beauveria bassiana (Bals.-Criv.) Vuill. (Ascomycota: Hypocreales) growth and virulence in a bark beetle system, Biological Control (2019), doi: https://doi.org/10.1016/j.biocontrol.2019.104130

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

© 2019 Published by Elsevier Inc.

Original Research article for Biological Control

Plant secondary metabolites and low temperature are the major limiting factors for Beauveria bassiana (Bals.-Criv.) Vuill. (Ascomycota: Hypocreales) growth and virulence in a bark beetle system

Andrew J. Mann*, Thomas Seth Davis

Department of Forest and Rangeland Stewardship, Colorado State University, 1472 Campus Delivery, Fort Collins, CO 80523-1472, USA

*Correspondence:

email: [email protected]; phone: +1-512-914-5145

ORCID: AJM: 0000-0002-0976-9647; TSD: 0000-0003-2303-7554

1

Abstract Beauveria bassiana is a ubiquitous entomopathogen and widely used as a biological control agent for a variety of arthropod pests, including bark beetles. The North American spruce beetle (Dendroctonus rufipennis) is a major pest in forest landscapes and recent studies show that B. bassiana is pathogenic to beetles in the lab but successful field applications have been limited by abiotic and biotic factors within the study system. To understand how habitat conditions impact fungal ecology, we evaluated variation in B. bassiana radial growth across conditions representative of the spruce beetle habitat, incorporating a range of temperatures, competition with spruce beetle symbiotic fungi, as well as exposure to Engelmann spruce tree secondary metabolites, nutrient limitations, osmotic potentials, and sunlight. Pathogenicity to spruce beetle was subsequently quantified in tests varying beetle origin, temperature, and bioassay arena substrate. Three major findings emerged: (1) growth of genetically similar B. bassiana isolates varied considerably in response to abiotic conditions, suggesting significant within-haplotype phenotypic diversity; (2) low ambient temperatures and exposure to Engelmann spruce tree secondary metabolites, two conditions which are prevalent in spruce beetle habitat, strongly inhibit B. bassiana growth; and (3) pathogenicity varied across environments: all isolates appeared pathogenic under room conditions, but when beetles inhabited in planta bioassays under in natura conditions most isolates were not pathogenic. This is the first study to evaluate the inhibitory effects that a series of tree secondary metabolites have on B. bassiana growth, virulence and pathogenicity. These collective findings have implications for field applications of B. bassiana but also the interpretation of entomopathogenicity, and could help to explain discrepancies between laboratory and field tests.

2

Keywords: biological control; Dendroctonus rufipennis; entomopathogenic fungi; environmental stress; in planta; monoterpenes

3

1. Introduction The ubiquitous insect-killing fungus Beauveria bassiana (Balsamo-Crivelli) Vuillemin (Hypocreales: Cordycipitaceae) is isolated from an incredible diversity of sources including soil, phylloplane habitats, and a wide variety of insect species (St. Leger et al., 1992; Bidochka et al., 2002; Rehner and Buckley, 2005). Accordingly, a great deal of phenotypic variation exists among isolates of this species. Commercial development of nontoxic mycologically-based pest management technologies (“mycoinsecticides”) has coincided with public recognition of the safety issues associated with traditional chemical control methods; B. bassiana has been widely tested as a biological control agent for arthropod pests in recent years as an alternative to these methods (Faria and Wraight, 2007). Additionally, B. bassiana is a desirable biological control agent because it can penetrate the insect exoskeleton and does not need to be ingested by its host (Khan et al., 2012). However, promising laboratory assessments often result in unsuccessful field applications of B. bassiana for insect population control (Edgington et al., 2000; Faria et al., 2001; Hajek and Goettel, 2007; Vega et al., 2012; Lacey et al., 2015; Davis et al., 2018b). This discrepancy could exist as a result of poor isolate selection due to a mismatch between laboratory and field conditions during evaluation; for instance, highly virulent strains may not tolerate a wide range of environmental conditions representative of a target pest’s habitat. The spruce beetle, Dendroctonus rufipennis Kirby (Coleoptera: Curculionidae: Scolytinae), colonizes Engelmann spruce (Picea engelmannii Parry ex Engelm.; Pinales: Pinaceae) and is one of the most significant forest pests in western North America (Jenkins et al., 2014; O’Connor et al., 2014; Colorado State Forest Service, 2017); beetle population activity is associated with mortality of at least 17 million P. engelmannii over the past two decades. Due to these impacts, recent work has focused on assessing the potential for B. bassiana to control D.

4

rufipennis population growth, but field tests have had limited success due in part to the extremely variable and inhibitory abiotic and biotic conditions that occur in high elevation forests characteristic of D. rufipennis habitat (Davis et al., 2018b). This is despite the apparent ubiquity of B. bassiana in forest soils (Reay et al., 2008; Yao et al., 2012; Niemczyk et al., 2019), bark (Yao et al., 2012), bark beetle oral secretions (Cardoza et al., 2009), insect integuments (Yao et al., 2012), fecal material (frass, Reay et al., 2008; Takov et al., 2012; Yao et al., 2012), and as an endophyte of coniferous trees (Reay et al., 2010; Brownbridge et al., 2012). Accordingly, a better understanding of how abiotic and biotic factors of North American spruce forests impact growth, pathogenicity, and virulence of B. bassiana can improve its use as a biological control agent of D. rufipennis. Spruce forests of the southern Rocky Mountain region are typically characterized by dense overstory vegetation that allows little sun penetration to the forest floor (Johnson et al., 2004). Overall, these areas experience considerable annual temperature fluctuations where summer temperatures range between 10 and 20 °C (Six and Bracewell, 2015; Dell and Davis, 2019). Dendroctonus species have a cryptic life cycle primarily spent in subcortical phloem environments that are rich with tree secondary metabolites, especially monoterpene hydrocarbons (Davis et al., 2018a). During the process of tree death following bark beetle attack, tree water potentials gradually decline over time and generally range between -0.5 and -2.0 MPa (Klepzig et al., 2004), which may limit fungal growth. In addition, D. rufipennis is associated with a symbiotic fungus, Leptographium abietinum (Peck) Wingfield (Ophiostomatales: Ophiostomataceae; Six and Bentz, 2003), which inhibits growth of antagonistic microorganisms (Davis et al., 2019). Consequently, B. bassiana parasitizing D. rufipennis must contend with a range of abiotic and biotic factors including cool temperatures, competing microbial species,

5

exposure to tree secondary metabolites, significant water deficits, and low-intensity ultraviolet light. Despite this habitat complexity, most studies evaluate B. bassiana isolate selection to target insects, including bark beetles, under laboratory conditions which do not take into account the phenotypic differences among many B. bassiana isolates, limitations due to insect behavioral response, or represent the complex abiotic and biotic factors that may limit growth and pathogenicity in natural environments (Moore, 1970; Moore, 1973; Pabst and Sikorowski, 1980; Hunt et al., 1984; Hunt, 1986; Sevim et al., 2010; Tanyeli et al., 2010; Zhang et al., 2011; Kocacevik et al., 2015; Srei et al., 2017; Davis et al., 2018b; Xu et al., 2018). The goal of this study was to characterize the phenotypic variation among isolates of B. bassiana collected across the Rocky Mountain region of the western United States through measurements of radial growth rate and pathogenicity to D. rufipennis under different abiotic and biotic conditions. Our specific objectives were to (1) evaluate the impacts of abiotic and biotic factors characteristic of the D. rufipennis habitat on B. bassiana growth performance and (2) to experimentally test the pathogenicity of regional B. bassiana isolates to D. rufipennis across a range of bioassay conditions. These studies contribute new insights into which abiotic and biotic factors limit B. bassiana success in forest habitats and can help to inform subsequent development of mycoinsecticides. Our results also indicate that experimental conditions greatly impact the virulence of B. bassiana with important consequences for the interpretation of entomopathogenicity of biocontrol agents to tree-killing bark beetles.

2. Materials and methods 2.1. Isolation and molecular identification of Beauveria bassiana strains

6

The B. bassiana strains used in this study were isolated on selective media from conifer trees, insect cadavers, and soil across a range of forest habitats in the Rocky Mountain region (Table 1). Spore powders were produced through industrial-scale solid substrate culture methods as described by Davis et al. (2018b), and a total of 14 isolates were chosen for use in the present study. In addition, six isolates of L. abietinum were collected from spruce forests in Colorado for evaluation of B. bassiana direct competition with L. abietinum as described by Davis et al. (2019). All growth rate and pathogenicity evaluations used material from the same generation of stock. For species identification and comparison of genetic diversity among B. bassiana isolates, DNA was extracted from 100 mg of fresh mycelia using ZR Fungal/Bacterial DNA MiniPrep Kit (Zymo Research Corporation, Irvine, CA, USA). Concentration of extracted DNA was measured using a nanodrop (ThermoScientific, Waltham, MA) to ensure a 260 nm/280 nm ratio of ~1.8 before sequencing of the Internal Transcribed Spacer (ITS) and Elongation Factor 1-α (EF1-α) regions following the methods of Rehner and Buckley (2005). The ITS region was amplified and sequenced using primers ITS 5 (5’-GGAAGTAAAAGTCGTAACAAGG-3’) and ITS 4 (5’-TCCTCCGCTTATTGATATGC-3’). The EF1-α region was amplified and sequenced using primers EF1T (5′-ATGGGTAAGGARGACAAGAC-3′) and 1567R (5′ACHGTRCCRATACCACCSATC-3′). Polymerase chain reaction mixtures (total 25 μl) contained 10 ng of template DNA (or no DNA template for negative control) and used the following conditions for 36 amplification cycles: 30 second denaturation at 94 °C, 30 second annealing at 56 °C, 1 minute extension at 72 °C, and 10 minute incubation at 72 °C. Sequence data for each isolate was aligned in Geneious Software (Biomatters, Auckland, New Zealand) and compared to the NCBI database using the BLAST procedure (Altschul et al.,

7

1990) for putative species identification. Sequencing of the ITS region, the universal DNA barcode for fungi (Schoch et al., 2012), yielded a sequence length of 596 base pairs and EF1α sequences yielded 1175 base pairs. All sequences were matched to sequences of Beauveria bassiana (Genbank accession numbers: ITS ≥ 99% match with B. bassiana; EF1-α ≥ 80% match with B. bassiana) with 99% coverage and an e-value of 0.0 (Table 2). The ITS and EF1-α regions were concatenated and a maximum likelihood phylogeny was created in MEGA7 using 200 bootstraps (Kumar et al., 2016). Isolates were all one haplotype based on ≥ 70% node support.

2.2. Beauveria bassiana radial growth in response to abiotic and biotic conditions representative of Dendroctonus rufipennis habitat We tested the hypothesis that there is phenotypic variation among isolates grown in different abiotic and biotic conditions representative of D. rufipennis habitats. Mycelial radial growth rate strongly positively correlates with biomass production in many fungal species (Ogidi et al., 2016) and is an indicator of environmental tolerance in entomopathogenic fungi (Jaronski, 2010). Accordingly, this trait is analyzed as the primary response variable in studies evaluating fungal response to abiotic and biotic factors. Beauveria bassiana growth was tested in vitro in response to six environmental factors. These environmental factors included (1) a range of temperatures, (2) competition with the spruce beetle symbiont L. abietinum, (3) a range of identities and concentrations of P. engelmannii phloem monoterpenes, (4) a range of concentrations of chitin as a sole nutrient source, (5) a range of osmotic water potentials, and (6) effects of sunlight exposure.

8

Except where noted below, the following is true for all tests: each test was simultaneously replicated 3-6 times for each isolate in 60 × 15 mm Petri dishes (VWR International, Radnor, Pennsylvania) containing 2% malt extract agar (MEA, pH 5.3, SigmaAldrich, St. Louis, MO; Davis et al., 2019). Petri dishes were inoculated with 0.1 mg B. bassiana conidial powder (22,499 ± 4,748 SE viable conidia) by aseptic transfer using a sterile probe dipped into a surfactant solution (0.01% Silwet L77, Helena Agri-Enterprises, Collierville, TN). Dishes were inverted after 48 hours and tests occurred in dark growth chambers (Thermo Fisher Scientific, Waltham, MA) kept at a constant 23 °C. Mycelial growth was traced and measured every 24 to 48 hours for 10 to 13 days while fungi were in the exponential radial growth phase. Mycelial growth rate (mm/d) was determined by dividing the total distance of radial growth (measured in two places on each dish and averaged together) by the total period of the assay. 2.2.1. Beauveria bassiana radial growth in response to temperature. Fungal colony radial growth in response to representative temperatures was tested simultaneously for three replicates of each isolate in isothermal growth chambers maintained at 5, 10, 15, 20, 23, 25, 30, and 35 °C (N = 292 total experimental units). 2.2.2. Beauveria bassiana competition with Leptographium abietinum. All B. bassiana isolates were evaluated in a dual culture setting for their ability to compete for resource space against six unique L. abietinum isolates (CF4, CF6, CF9, CF11, CF17, and CF22 described in Davis et al. (2019)) to test the hypothesis that the bark beetle symbiont, L. abietinum, inhibits growth of B. bassiana (N = 84 total experimental units). Each fungus was inoculated 8 mm from the edge of 95 × 15 mm Petri dishes (Fisherbrand, Waltham, MA). Plates were scanned (Epson V600, Suwa, Japan) after 20 d of growth to test the ability of each fungus to capture and maintain its occupied space. These images were analyzed using ImageJ software (National

9

Institutes of Health, Washington, D.C). Total area (% of dish) occupied by B. bassiana for each replicate was treated as the response variable. 2.2.3. Beauveria bassiana radial growth in response to Engelmann spruce tree secondary metabolites. Petri dishes were amended with one of five monoterpenes found in the phloem of all Engelmann spruce trees in Colorado (Davis et al., 2018a) including (+)-α-pinene (98% purity, Sigma-Aldrich), (-)-β-pinene (99% purity, Sigma-Aldrich), (+)-3-carene ( > 90% purity, SigmaAldrich), myrcene ( > 95% purity, Sigma-Aldrich), and terpinolene ( > 90% purity, SigmaAldrich) at 0.1, 1.0, and 5.0% (v/v) concentrations, consistent with constitutive (0.1 and 1%) and induced (5%) monoterpene concentrations (N = 907 total experimental units). Colony radial growth rates under monoterpene amendments were standardized to a percentage of the mean radial growth rate of each colony on non-amended media (2% MEA) for statistical analysis. 2.2.4. Beauveria bassiana radial growth on media containing chitin as a nutrient. To evaluate the ability of isolates to grow in minimal media containing only arthropod exoskeleton contents, a potential virulence factor, Petri dishes containing water agar were amended with either 0.1, 1.0, or 5.0% (v/v) shrimp chitin (Sigma-Aldrich) to represent potential growth on the bark beetle exoskeleton (N = 127 total experimental units). Colony growth rates under chitin amendments were standardized to a percentage of the mean growth rate of each colony on nonamended media (2% MEA) for statistical analysis. 2.2.5. Beauveria bassiana radial growth on media with limited water availability. To assess effects of water limitation on colony radial growth, B. bassiana isolates were transferred to Petri dishes containing 1% MEA amended with KCl (EMD Chemicals Gibbstown, NJ) and sucrose (Sigma-Aldrich) to generate osmotic potentials of -0.5 MPa, -1.0 MPa, -2.0 MPa (N = 160 total experimental units) using ratios described by Whiting and Rizzo (1999); these levels

10

are representative of phloem water potential in drought-stressed conifers or trees in decline following bark beetle attack (Klepzig et al., 2004). Colony radial growth rates under limited water availability were standardized to a percentage of the mean growth rate of each colony on non-amended media (1% MEA) for statistical analysis. 2.2.6. Beauveria bassiana radial growth response to ultraviolet light. The response of B. bassiana isolates to low-intensity light conditions was evaluated in non-inverted Petri dishes that were placed in a windowsill and exposed to 13 d of indirect sunlight at an intensity of 4.8 ± 0.2 SE µmol/m2/sec for 12 h 58 min ± 9 SE min per day (N = 142 total experimental units). Sunlight intensity was measured in three locations on the windowsill using a quantum light meter (Apogee Instruments, Logan, UT) covered by a Petri dish lid.

2.3. Beauveria bassiana pathogenicity to Dendroctonus rufipennis and variation in isolate virulence Four experiments were performed to test the hypothesis that Rocky Mountain isolates of B. bassiana vary in their pathogenicity and virulence to D. rufipennis across a range of conditions. Experiments used sterile 95 × 15 mm Petri dishes as the test arena, and replicates consisted of six adult spruce beetles; each isolate was replicated ten times in each experiment (N = 140 experimental units and 840 adult beetles per experiment). Pathogenicity (the ability of B. bassiana to cause mortality) of isolates in each experiment was determined by comparison against a negative control treatment. The negative control treatments (N = 10 experimental units and 60 adult beetles per experiment) were treated with distilled water containing a surfactant solution (0.01% Silwet L77). The B. bassiana concentration added to each replicate was standardized using serial dilution to 1.0 × 106 viable conidia/ml suspension for each test isolate,

11

and 1 ml of spore suspension was administered to the tops of substrate surfaces contacted by beetles rather than directly to beetle integuments (i.e., spore suspension was applied to either filter papers or to the surface of bark discs). Experiments spanned a range of conditions, varying test temperatures, bioassay substrates, and beetle source (Table 3). Experiments were designed to allow for beetle attraction, avoidance, and grooming behaviors by allowing test individuals space to move around within test arenas. Insect behavior is an important factor to consider in bioassays as detection and avoidance of B. bassiana by target insect species is previously reported (Meyling and Pell, 2006; Mburu et al., 2013) and may be associated with host resistance. To evaluate infectivity following beetle mortality in bioassays, a mycosis test was performed on all dead beetles according to methods of Bugeme et al. (2008). A 1 ml aliquot of distilled water was added to filter paper and maintained at a constant 30 °C for 48 hours in the dark; beetles colonized by B. bassiana readily sporulate under these conditions and appear to ‘mummify,’ confirming adults were infected by B. bassiana in each replicate. 2.3.1. Experimental conditions. Experiment 1 took place at 23 °C with D. rufipennis reared from logs collected from five infested P. engelmannii trees at Cameron Pass, Colorado (coordinates: 40.52058 N, 105.89283 W, elevation: 3100 meters). To incite colonization of selected trees, trees were baited with an attractant containing frontalin, 1-methylcyclohex-2-en-1ol, and host tree kairomones (monoterpenes; Synergy Semiochemicals, Burnaby, Canada) during May 2017. During September and October 2017 following D. rufipennis colonization, baited trees were felled, cut into billets of ~0.6-meter length, and placed into rearing containers ventilated with a 1 × 1 mm mesh in a laboratory at 23 °C with a relative humidity of ~30%. Billets in rearing chambers accumulated approximately 800 degree-days in the laboratory, after

12

which point new adult beetles were harvested from logs for testing. The spruce beetle has individuals within the same population that exhibit either a 1- or 2-year life cycle (Holsten et al., 1999), and this experiment was designed to control for beetle age by ensuring that all beetles were new adults and not a mix of 1- and 2- year-old beetles. An aliquot of 1 ml of standardized spore suspension (1.0 × 106 viable conidia/ml in distilled water amended with 0.01% Silwet L77) was applied to filter papers contacted by beetles (Whatman Grade 2, 4.25 cm diameter, Maidstone, United Kingdom), as described by Davis et al. (2018b). Subsequent experiments (experiments 2-4) used beetles actively responding to semiochemicals collected during their dispersal period, as actively flying beetle populations are likely the most vulnerable to B. bassiana applications. To capture dispersing beetles a total of 10 Lindgren funnel traps (Synergy Semiochemicals) baited with standard attractants (enhanced spruce beetle lure, Synergy Semiochemicals) were deployed to collect adults during peak flight season (Dell and Davis, 2019) at Monarch Pass, Colorado (coordinates: 38.49666 N, 106.32558 W, elevation: 3448 m). Moist single-ply paper towels were placed in collection cups to provide a surface for beetles to adhere to and beetles were collected and returned to the lab within 48 hours of capture. Collections were made twice weekly from mid-June until July 2018. Prior to use in experiments, beetles were subjected to a simple fitness test using the approach described by Chiu et al. (2017). Experiments 2 and 3 were identical to experiment 1 in all parameters, with the condition that the source of beetles differed (experiment 1 tested effects of B. bassiana on lab-reared beetles) and experiment 3 was performed at a constant 10 °C rather than at 23 °C. This temperature was chosen to reflect mean temperatures in the field during dispersal (Dell and Davis, 2019). Experiment 4 was an in planta test that supplied beetles with phloem and bark,

13

which could affect survival rates of D. rufipennis and pathogenicity of B. bassiana. Phloem also contains monoterpenes that may reduce efficacy of B. bassiana. Methods for creating ‘phloem sandwiches’ followed Aflitto et al. (2014) with slight modifications: one hundred fifty 8 × 8 cm pieces of phloem were excised from standing P. engelmannii with outer bark still intact and later cut into circles to fit firmly on the bottom of 6 cm diameter Petri dishes. A 1 ml aliquot of B. bassiana spore suspension at a cell density of 1.0 × 106 viable conidia/ml was applied to the top of the bark surface. Replicates in all experiments were checked regularly and all beetles in each Petri dish were scored as ‘alive’ or ‘dead’ at each recording. Experiments 1-3 were scored daily in this way until all beetles had died. In experiment 4, replicates were scored for survival and mycosis every three days for 90 d, which is the approximate length of the spruce beetle flight season in Colorado (Dell and Davis, 2019).

2.4. Data analysis All analyses were performed using the R statistical programming language (R Core Team, 2017). To test the hypothesis that isolates vary in their radial growth in response to temperatures, a twoway ANOVA was used to analyze the fixed effects of isolate identity (N = 14), temperature (5, 10, 15, 20, 23, 25, 30, and 35 °C), and the isolate × temperature interaction. No fungal growth was observed by any isolate at 5 and 35 °C; these factor levels were omitted from subsequent statistical analysis since they showed no variance. To analyze resource capture ability of B. bassiana isolates when paired with spruce beetle symbiotic fungi (L. abietinum), a one-sample Student’s t-test was used to test the hypothesis that the mean percent of the Petri dish (%) occupied by B. bassiana isolates differed from 50% (the mean value that would be expected if

14

both fungal species captured equal areas of resource space). The radial growth of B. bassiana in response to exposure to tree secondary metabolites (monoterpenes) was analyzed using threeway ANOVA to test the fixed effects of monoterpene identity (α-pinene, β-pinene, 3-carene, myrcene, and terpinolene), monoterpene concentration (0.1, 1, and 5%), isolate identity, and all two- and three-way interactions on the response of B. bassiana mean colony relative radial growth (i.e., as compared to radial growth rate in the absence of monoterpenes). The response of isolates to chitin as the sole nutrient source was analyzed using two-way ANOVA to test the effects of chitin concentration (0.1, 1, and 5%), isolate identity, and the isolate × chitin concentration interaction on mean colony relative radial growth. The effects of varying water potentials (0, -0.5, -1.0, and -2.0 MPa) on colony radial growth were analyzed using an identical model. To test the effects of exposure to low-intensity sunlight on radial growth, a two-way ANOVA used to analyze the fixed effects of light treatment (dark vs. sunlight); isolate identity, and the isolate identity × light treatment interaction. In experiments testing pathogenicity and virulence of B. bassiana isolates to D. rufipennis, the median survival time (MST) of test beetles was the primary response variable analyzed. MST was analyzed using Kaplan-Meier survival analysis and a log-rank test implemented using R packages ‘survminer’ (Kassambara and Kosinski, 2018), and ‘survival’ (Therneau and Grambsch, 2000) for calculations and ‘ggplot2’ (Wickham, 2009), ‘ggpubr’ (Kassambara, 2018), ‘gridExtra’ (Augie, 2017), and ‘cowplot’ (Wilke, 2019) for visualization. Every isolate was compared to the negative control treatment through log-rank tests to evaluate differences in pathogenicity and virulence across isolates and experiments. In all statistical analyses, a Type I error rate of α = 0.05 was used for assigning statistical significance to treatment effects.

15

3. Results 3.1. Beauveria bassiana radial growth in response to abiotic and biotic conditions representative of Dendroctonus rufipennis habitat There was considerable isolate-to-isolate variability in radial growth as a response to the six abiotic and biotic factors. Isolates exhibited statistically significant phenotypic variation in every abiotic condition but not the biotic condition, competition against the spruce beetle symbiont, L. abietinum. 3.1.1. Beauveria bassiana radial growth in response to temperature. There was significant variation in the mean radial growth rate of isolates due to the effect of temperature (F5, 165 = 79.251; P < 0.001) and isolate identity (F13, 165 = 2.174; P = 0.012), but there was no evidence of an isolate × temperature interaction (F65, 165 = 0.896; P = 0.689). No growth of any B. bassiana isolate occurred at 5 °C and 35 °C; maximum radial growth occurred at 23 °C, but similar radial growth rates were generally observed between 20 and 30 °C with the greatest variation in radial growth at 25 °C. Radial growth was reduced by 88% at 15 °C and by 97% at 10 °C relative to the radial growth rate at 23 °C (Figure 1). 3.1.2. Beauveria bassiana competition with Leptographium abietinum. Both B. bassiana and L. abietinum grew until touching and showed no signs of forming an inhibition zone. Each fungus maintained the captured resource space in competition assays for at least 20 d. On average, B. bassiana captured only 44% of the available space and proved to be a significantly weaker (t83 = -2.128; P = 0.036) competitor for resource space than L. abietinum which captured 56% of the resource space on average. Beauveria bassiana isolates did not differ in their ability to compete with L. abietinum for resource space (F13, 70 = 0.664; P = 0.790) as shown in Figure 2.

16

3.1.3. Beauveria bassiana radial growth in response to Engelmann spruce tree secondary metabolites. Mean radial growth of B. bassiana isolates varied due to the effects of isolate identity (F13, 697 = 8.081; P < 0.001), monoterpene identity (F4, 697 = 36.735; P < 0.001), monoterpene concentration (F2, 697 = 471.472; P < 0.001), an isolate × monoterpene identity interaction (F52, 697 = 1.777; P < 0.001), an isolate × monoterpene concentration interaction (F26, 697

= 2.210; P = 0.001), and a monoterpene identity × monoterpene concentration interaction (F8,

697

= 20.455; P < 0.001); however, there was no evidence for a three-way isolate × monoterpene

identity × monoterpene concentration interaction (F104, 697 = 0.955; P = 0.605). The ANOVA model explained 70% of the variation in radial growth rate (R2 = 0.709; whole model: F209, 697 = 8.157; P < 0.001) and the main effect of monoterpene concentration had the greatest effect size on colony radial growth and explained 55% of the modeled variance; the interaction between monoterpene identity and concentration explained an additional 9% of the modeled variance. Variability due to isolate effects only described 6% of the variance in colony radial growth rate. All tested monoterpenes and concentrations were inhibitory to all B. bassiana isolates, with a couple exceptions for a single isolate (429DA). However, all isolates were able to grow in the presence of every monoterpene at low concentrations. Increasing monoterpene concentrations generally slowed fungal radial growth, with total inhibition of many isolates at 5% monoterpene concentration. Average isolate radial growth rate was reduced by 41% and 71% at monoterpene concentrations found constitutively in Engelmann spruce phloem (0.1 % and 1.0% v/v, respectively) and by 94% on average when exposed to inducible (5.0% v/v) monoterpene concentrations. 3-carene and terpinolene were the most inhibitory monoterpenes overall and each caused ~76% reduction in mean colony radial growth, though nearly all isolates were inhibited by terpinolene at 5% concentration. α-pinene was intermediate in its inhibitory effects and

17

reduced mean colony radial growth by 71% and myrcene and β-pinene were the least inhibitory monoterpenes and reduced mean colony radial growth by 58 and 57%, respectively (Figure 3). 3.1.4. Beauveria bassiana radial growth on media containing chitin as a nutrient. There was significant variation in the radial growth of B. bassiana isolates due to the effect of isolate identity (F13, 84 = 49.473; P < 0.001), chitin concentration (F2, 84 = 12.672; P < 0.001), and isolate × chitin concentration in radial growth media (F26, 84 = 2.538; P = 0.001). All isolates were able to grow on minimal media containing chitin as the sole nutrient source, but isolates experienced a 55, 51, and 58% reduction when chitin concentrations were 0.1, 1, and 5%, respectively (Figure 4). 3.1.5. Beauveria bassiana radial growth on media with limited water availability. There was significant variation in the mean radial growth rates of B. bassiana isolates due to the effects of isolate identity (F13, 112 = 3.149; P < 0.001), media water potential (F3, 112 = 59.959; P < 0.001), and an isolate × water potential interaction (F39, 112 = 1.955; P = 0.003). Radial growth rate of control treatment (no water deficit; 1.208 ± 0.038 SE mm/day) did not differ from that of radial growth rate under minor water deficit (-0.5 MPa; 1.208 ± 0.044 SE mm/day) and radial growth was slightly enhanced under modest water deficits (-1.0 MPa; 1.418 ± 0.010 SE mm/day). However, B. bassiana isolates growing under an increased water deficit (-2.0 MPa) exhibited a ~23% and ~34% reduction in mean radial growth rates (0.934 ± 0.011 SE mm/day) as compared to radial growth under control and moderate water deficit treatments; indicating that fungal growth in a tree tissue microhabitat is likely to increase during initial tree dry-down but then decline over time as water deficit increases beyond -2.0 MPa (Figure 5). 3.1.6. Beauveria bassiana radial growth response to ultraviolet light. There was significant variation in colony radial growth rates due to the main effects of isolate identity (F13,

18

55

= 2.354; P = 0.014) and light treatment (F1, 55 = 82.270; P < 0.001), but there was no evidence

for an isolate × light treatment interaction (F13, 55 = 1.352; P = 0.212). Nearly all (13 of 14, or ~93%) isolates exhibited reduced radial growth rates when exposed to low-intensity sunlight. The majority of variance in radial growth rate was due to the effect of light treatment (63% of the modeled variance), however, one isolate (429DA) appeared tolerant of sunlight exposure (Figure 6).

3.2. Beauveria bassiana pathogenicity to Dendroctonus rufipennis and variation in isolate virulence The range of MST differed considerably among experiments and ranged from 6-10 days in experiment 1 (Table 4, Figure S1), 5 days for every isolate in experiment 2 (Table 4, Figure S2), 8-11 days in experiment 3 (Table 4, Figure S3), and 19-62 days in experiment 4 (Table 4, Figure S4). Phenotypic variation in virulence among isolates was significant according to log-rank tests based on Kaplan-Meier assumptions in experiment 1 (P < 0.001), experiment 2 (P < 0.001), and experiment 4 (P < 0.001), but not in experiment 3 (P = 0.660). A group of isolates (429BTF, GHA, D900, L429, L447, and 90(1)MPB) all had the lowest MST times in experiment 1 (indicating low beetle survival time, consistent with a high degree of virulence), but isolate 50C caused the lowest MST (consistent with most rapid D. rufipennis mortality) in experiments 3 and 4. The relative pathogenicity of isolates differed from experiment to experiment based on MST relative to negative control treatments (χ2 = 58.300, df = 3, P < 0.001). Log-rank tests indicate that every isolate was pathogenic in experiments performed under warm (23 °C) conditions on new adult beetles (experiment 1); similarly, all isolates were pathogenic in

19

experiments performed under warm conditions with dispersing beetles of unknown age (experiment 2). However, only isolate 50C was pathogenic in experiments tested under relatively cooler conditions (10 °C) with dispersing beetles (experiment 3). Under the most system-specific conditions that provisioned dispersing beetles with phloem under cool conditions (10 °C), only isolates 50C and 14B were pathogenic (Table 4).

4. Discussion Beauveria bassiana strains isolated from various sources throughout the Rocky Mountain region showed considerable phenotypic variation in terms of the environmental factors that affected radial growth rates and their relative ability to reduce survival of D. rufipennis under a range of experimental conditions. Forest systems introduce new factors to overcome in the application of B. bassiana as a biological control agent (Hesketh et al., 2010; Popa et al., 2012), and aspects of habitat complexity are often not accounted for in studies which only examine effects of temperature, relative humidity, and ultraviolet light. However, the ability of isolates to grow under a range of more complex environmental conditions, as well as their relative ability to impact insect populations across those conditions, are key for the development of successful mycologically-based biocontrol technologies. Our results demonstrate several important issues related to this point: (1) B. bassiana isolates vary widely in their radial growth response to abiotic factors representative of the D. rufipennis habitat, even among genetically similar strains isolated from similar habitats and sources; (2) conifer tree secondary metabolites are highly inhibitory to B. bassiana radial growth, especially at induced concentrations, though low temperatures also greatly reduced radial growth; and (3) the interpretation of isolate pathogenicity and virulence differs substantially depending on experimental conditions – with

20

many isolates exhibiting pathogenicity at room temperature and on filter paper, but few isolates exhibiting pathogenicity when tests are performed on actual tree tissues (phloem) at a systemspecific temperature. These collective findings have implications for the application of B. bassiana as a biological control agent of bark beetles. Our results confirm that B. bassiana is highly affected by temperature and, consistent with the literature (Yeo et al., 2003; Bugeme et al., 2008), nearly every isolate maximizes radial growth rates at or near 23-25 °C. The thermal growth threshold of 5 °C is a potential problem for B. bassiana application in the D. rufipennis habitat; though the low temperatures did not cause the fungus to die, but rather freeze because growth resumed when Petri dishes were brought into room temperature following completion of the 5 °C growth assays. The results of experiments to test fungal competition corroborated recent studies indicating that both B. bassiana and L. abietinum are able to capture and maintain space (Davis et al., 2018b; Davis et al., 2019), and that the fungi apparently compete with one another for resource space but neither fungus is able to overtake its competitor. The concentration of chitin in media also had a significant overall effect on the radial growth of B. bassiana, though radial growth of isolates was lowest in media with the highest concentration of chitin – potentially indicating a reduced need for radial growth under high nutrient conditions. This also indicates that contact with target insects may be more likely in low-chitin conditions (i.e., in the absence of an arthropod host), as B. bassiana more rapidly expands surface area when chitin concentrations are low. Isolates exhibited enhanced growth in media containing osmotic water potentials of -0.5 and -1.0 MPa and only moderate reductions in overall radial growth as media water deficits increased. Beauveria bassiana varies greatly in its ability to tolerate dry environments. In an earlier study, an osmotic potential of -1.76 MPa caused complete growth inhibition for some

21

isolates while just a 4% decrease in others (Devi et al., 2005). Dendroctonus beetles must also contend with arid conditions as their host trees die but spend most of their lives in the phloem of conifer trees – which is a relatively humid environment. Our results suggest that practitioners should prioritize environmental factors other than osmotic potential during B. bassiana strain selection for control of bark beetles, especially when choosing among this group of isolates. Additionally, exposure to sunlight can completely inhibit fungal growth due to a lack of melanin in mycelial tissues and is often cited as the most limiting abiotic factor of B. bassiana efficacy in many systems (Jaronski, 2010; Fernandes et al., 2015). We found significant phenotypic variation among isolates in their ability to grow in an environment with low-intensity sunlight exposure, and inhibition of most isolates following sunlight exposure, though no isolates were completely inhibited by exposure to sunlight. Furthermore, as with osmotic potential, sunlight exposure may not be a continuous challenge for B. bassiana in this system because bark beetles spend most of their lives below the bark surface of host trees. Covered traps that contain B. bassiana in dissemination chambers have been tested for the control of emerald ash borer (Agrilus planipennis Fairmaire Coleoptera: Cuprestidae; Lyons et al., 2012), D. simplex LeConte (Srei et al., 2017), Hypothenemus hampei Ferrari (Coleoptera: Curculionidae: Scolytinae; Mota et al., 2017), and Ips typographus Linnaeus (Coleoptera: Curculionidae: Scolytinae; Grodski and Kosibowicz, 2015) and can further alleviate the need for a selected B. bassiana isolate to tolerate sunlight in application against bark and woodboring beetles, as these developing technologies allow for B. bassiana to be inoculated into relatively protected habitats. Conifer secondary metabolites, including monoterpenes, are a central aspect of tree defense in response to bark beetles and other herbivores (Raffa, 2014). While monoterpenes are always produced by conifer trees at low (constitutive) levels, the composition and concentration

22

of these monoterpenes are often upregulated (i.e., induced) when trees are challenged by insects or pathogens (Litvak and Monson, 1998). Inoculation with B. bassiana can also induce a plant defense response (Shrivastava et al., 2015) which suggests that defensively induced trees are likely to inhibit B. bassiana growth even though the presence of entomopathogens could benefit tree survival during a bark beetle attack (Elliot et al. 2000; Hay et al., 2004; Cory and Ericsson, 2010). Beauveria bassiana inhibition in response to plant secondary metabolite exposure has been well-studied in crop systems such as tomato (Lycopersicon esculentum L.; Hare and Andreadis, 1983; Costa and Gaugler, 1989; Poprawski et al., 2000; Santiago-Álvarez et al., 2006), cotton (Gossypium hirsutum L.; Poprawski and Jones, 2001; Santiago-Álvarez et al., 2006), maize (Zea mays L., Ramoska and Todd, 1985), and wasabi (Wasabia japonica Matsumura; Atsumi and Saito, 2015); but only once with tree secondary metabolites (Davis et al., 2018b) and never before at the scale of this study. Monoterpenes were extremely inhibitory in this study and aside from low temperature, are the single most-limiting factor to successful biological control of bark beetles in forest environments. Plant secondary metabolite identity played a critical role in limiting B. bassiana growth; especially terpinolene and 3-carene, which both reduced growth by over 92% compared to growth rates in the absence of monoterpenes. These secondary metabolites are present in virtually all P. engelmannii in the southern Rocky Mountains (Davis et al., 2018a). As concentrations of monoterpenes increased, so did the degree of growth inhibition. Average colony growth was reduced by over 90% in media amended with 1% monoterpenes and over 98% in media containing 5% monoterpenes. The induced level (5% v/v concentration) used in this study is conservative, as monoterpene concentrations can increase by over 30 times when conifers are challenged by bark beetles and their symbiotic fungi (Raffa and Smalley, 1995).

23

Inhibition of fungal growth caused by monoterpenes may functionally eliminate the possibility of B. bassiana entomopathogenicity towards bark beetles colonizing host tree tissues, or application of the fungus as an endophyte, and should be one of the primary environmental factors that practitioners consider in future isolate selection processes and field applications. Our results also show that the interpretation of isolate pathogenicity can differ substantially depending on the experimental design, which is problematic as most studies rely on short tests under relatively unrealistic laboratory conditions to assign isolate virulence. In our experiments, virulence of B. bassiana towards dispersing beetles differed from mortality rates of adults reared and collected from Engelmann spruce logs (Table 3; Figures S1 and S2); indicating the importance of evaluating future biological control application against the appropriate life stage. While every isolate was pathogenic to D. rufipennis in filter paper bioassays under room temperature (23 °C), only two isolates were pathogenic to D. rufipennis in bioassays that took place in planta and at a mean summer temperature representative of southern Rocky Mountain spruce forests (10 °C). If tests of B. bassiana pathogenicity and virulence are not done in planta under in natura conditions, they are likely to be misleading and misrepresent the efficacy of isolates at reducing insect population densities under field conditions. Interestingly, neither of the pathogenic isolates in the in planta phloem experiment were among the most virulent at 23 °C. Hence, a key reason why so many promising laboratory studies lead to ineffective field application may be the lack of consideration for system-specific environmental conditions during strain evaluation. The results in this study support findings by Kreutz et al. (2004) where filter paper was deemed an unsuitable bioassay substrate for I. typographus because filter paper does not provide nutrients to the beetles. Furthermore, neither of the pathogenic isolates in the phloem bioassay were top growers in 10 °C or when exposed to monoterpenes. Thus, eliminating isolates

24

from the screening process based solely on radial growth rate could also lead to incorrect conclusions about virulence as positive correlations between pathogenicity and radial growth rate do not always occur with B. bassiana. The studies reported here have several implications for the future development and application of B. bassiana as a mycologically-based method of pest control. First, future studies on the biology and potential application of B. bassiana should take a multivariate approach and include complex environmental factors unique to the desired application habitat. Second, in planta bioassays under representative environmental conditions are vital; bioassays performed under simplified conditions are misleading because B. bassiana isolates are highly phenotypically variable and radial growth responses do not necessarily translate to the expression of pathogenecity in the presence of host plant material. Finally, building a stronger foundational understanding of the mechanisms that cause tree secondary metabolites to limit B. bassiana efficacy will allow for more successful utilization of entomopathogenic fungi as biocontrol agents of bark and woodboring beetles.

CRediT authorship contribution statement Andrew J. Mann: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Writing – original draft, Writing – review & editing. Thomas Seth Davis: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Supervision, Writing – review & editing

Declarations of Competing Interest None

25

Acknowledgements For assistance with fungal DNA extraction and genetic analyses, we are grateful to Kris Otto and John Dobbs. We appreciate Drs. Javier Mercado and José Negrón for providing access to microscopes and growth chambers. Thank you to Isaac Dell, Katie Jones, Jimmy Mattila, and Khum Thapa-Magar for assistance with laboratory and field work.

Funding This work was supported by Colorado State University

26

References Aflitto, N.C., Hofstetter, R.W., McGuire, R., Dunn, D.D., Potter, K.A., 2014. Technique for studying arthropod and microbial communities within tree tissues. JOVE-J. Vis. Exp. e50793. https://doi.org/10.3791/50793 Altschul, S.F., Gish, W., Miller, W., Myers, E.W., Lipman, D.J., 1990. Basic local alignment search tool. J. Mol. Biol. 215, 403-410. https://doi.org/10.1016/S0022-2836(05)80360-2 Atsumi, A., Saito, T., 2015. Volatiles from wasabi inhibit entomopathogenic fungi: implications for tritrophic interactions and biological control. J. Plant Interact. 10, 152-157. https://doi.org/10.1080/17429145.2015.1039613 Augie, B., 2017. gridExtra: Miscellaneous functions for “grid” graphics. R package version 2.3. https://www.CRAN.R-project.org/package=gridExtra Bidochka, M.J., Menzies, F.V., Kamp, A.M., 2002. Genetic groups of the insect-pathogenic fungus Beauveria bassiana are associated with habitat and thermal growth preferences. Arch. Microbiol. 178, 531-537. https://doi.org/10.1007/s00203-002-0490-7 Brownbridge, M., Reay, S.D., Nelson, T.L., Glare, T.R., 2012. Persistence of Beauveria bassiana (Ascomycota: Hypocreales) as an endophyte following inoculation of radiata pine seed and seedlings. Biol. Control 61, 194-200. https://doi.org/10.1016/j.biocontrol.2012.01.002 Bugeme, D.M., Maniania, N.K., Knapp, M., Boga, H.I., 2008. Effect of temperature on virulence of Beauveria bassiana and Metarhizium anisopliae isolates to Tetranychus evansi. Exp. Appl. Acarol. 46, 275–285. https://doi.org/10.1007/s10493-008-9179-1 Cardoza, Y.J., Vasanthakumar, A., Suazo, A., Raffa, K.F., 2009. Survey and phylogenetic

27

analysis of culturable microbes in the oral secretions of three bark beetle species. Entomol. Exp. Appl. 131, 138-147. https://doi.org/10.1111/j.1570-7458.2009.00844.x Chiu, C.C., Keeling, C.I., Bohlmann, J., 2017. Toxicity of pine monoterpenes to mountain pine beetle. Sci. Rep.-U.K. 7, 8858. https://doi.org/10.1038/s41598-017-08983-y Colorado State Forest Service, 2017. 2017 Report on the health of Colorado’s forests: Meeting the challenge of dead and at-risk trees. Colorado State University, Fort Collins, pp. 1-23. Cory and Ericsson, 2010. Fungal entomopathogens in a tritrophic context. BioControl. 55, 75-85. https://doi.org/10.1007/s10526-009-9247-4 Costa, S.D., Gaugler, R.R., 1989. Sensitivity of Beauveria bassiana to solanine and tomatine: Plant defensive chemicals inhibit an insect pathogen. J. Chem. Ecol. 15, 697-706. https://doi.org/10.1007/BF01014712 Davis, T.S., Horne, F.B., Yetter, J.C., Stewart, J.E., 2018a. Engelmann spruce chemotypes in Colorado and their effects on symbiotic fungi associated with North American spruce beetle. J. Chem. Ecol. 44, 601-610. https://doi.org/10.1007/s10886-018-0961-1 Davis, T.S., Mann, A.J., Malesky, D., Jankowski, E., Bradley, C., 2018b. Laboratory and field evaluation of the entomopathogenic fungus Beauveria bassiana (Deuteromycotina: Hyphomycetes) for population management of spruce beetle, Dendroctonus rufipennis (Coleoptera: Scolytinae), in felled trees and factors limiting pathogen success. Environ. Entomol. 47, 594-602. https://doi.org/10.1093/ee/nvy036 Davis, T.S., Stewart, J.E., Mann, A., Bradley, C., Hofstetter, R.W., 2019. Evidence for multiple ecological roles of Leptographium abietinum, a symbiotic fungus associated with North American spruce beetle. Fungal Ecol. 38, 62-70. https://doi.org/10.1016/j.funeco.2018.04.008

28

Dell, I.H., Davis, T.S., 2019. Effects of site thermal variation and physiography on flight synchrony and phenology of the North American spruce beetle (Coleoptera: Curculionidae, Scolytinae) and associated species in Colorado. Environ. Entomol. 48, 998-1011. https://doi: 10.1093/ee/nvz067 Devi, K.U., Sridevi, V., Mohan, C.M., Padmavathi, J., 2005. Effect of high temperature and water stress on in vitro germination and growth in isolates of the entomopathogenic fungus Beauveria bassiana (Bals.) Vuillemin. J. Invertebr. Pathol. 88, 181-189. https://doi.org/10.1016/j.jip.2005.02.001 Edgington, S., Segura, H., De la Rosa, W., Williams, T., 2000. Photoprotection of Beauveria bassiana: Testing simple formulations for control of the coffee berry borer. Int. J. Pest Manage. 46, 169-176. https://doi.org/10.1080/096708700415490 Elliot, S.L., Sabelis, M.W., Janssen, A., van der Geest, L.P.S., Beerling, E.A.M., Fransen, J., 2000. Can plants use entomopathogens as bodyguards? Ecol. Lett. 3, 228-235. https://doi.org/10.1046/j.1461-0248.2000.00137.x Faria, M. de, Wraight, S.P., 2001. Biological control of Bemisia tabaci with fungi. Crop Prot. 20, 767–778. https://doi.org/10.1016/S0261-2194(01)00110-7 Faria, M.R. de, Wraight, S.P., 2007. Mycoinsecticides and mycoacaricides: A comprehensive list with worldwide coverage and international classification of formulation types. Biol. Control 43, 237-256. https://doi.org/10.1016/j.biocontrol.2007.08.001 Fernandes, É.K., Rangel, D.E., Braga, G.U., Roberts, D.W., 2015. Tolerance of entomopathogenic fungi to ultraviolet radiation: A review on screening of strains and their formulation. Curr. Genet. 61, 427-440. https://doi.org/10.1007/s00294-015-0492-z Grodski, W., Kosibowicz, M., 2015. An attempt to use the fungus Beauveria bassiana (Bals.)

29

Vuill. in forest protection against the bark beetle Ips typographus (L.) in the field. Leśne Prace Badawcze 76, 5-17. Hajek, A.E., Goettel, M.S. 2007. Guidelines for evaluating effects of entomopathogens on nontarget organisms. In: Lacey, L.A., Kaya, H.K. (Eds), Field Manual of Techniques in Invertebrate Pathology. Springer, Dordrecht, pp. 816-833. https://doi.org/10.1007/978-14020-5933-9_40 Hare, J.D., Andreadis, T.G., 1983. Variation in susceptibility of Leptinotarsa decemlineata (Coleoptera: Chrysomelidae) when reared on different host plants to the fungal pathogen, Beauveria bassiana in the field and laboratory. Environ. Entomol. 12, 1892-1897. https://doi.org/10.1093/ee/12.6.1892 Hay, M.E., Parker, J.D., Burkepile, D.E., Caudill, C.C., Wilson, A.E., Hallinan, Z.P., Chequer, A.D., 2004. Mutualisms and aquatic community structure: The enemy of my enemy is my friend. Annu. Rev. Ecol. Evol. S. 35, 175-197. https://doi.org/10.1146/annurev.ecolsys.34.011802.132357 Hesketh, H., Roy, H.E., Eilenberg, J., Pell, J.K., Hails, R.S., 2010. Challenges in modelling complexity of fungal entomopathogens in semi-natural populations of insects. BioControl 55, 55-73. https://doi.org/10.1007/s10526-009-9249-2 Holsten, E.H., Their, R.W., Munson, A.S., Gibson, K.E., 1999. The spruce beetle. USDA Forest Service, Forest Insect and Disease Leaflet 127, pp. 11. Hunt, D.W.A., 1986. Absence of fatty acid germination inhibitors for conidia of Beauveria bassiana on the integument of the bark beetle Dendroctonus ponderosae (Coleoptera: Scolytidae). Can. Entomol. 118, 837-838. Hunt, D.W.A., Borden, J.H., Rahe, J.E., Whitney, H.S., 1984. Nutrient-mediated germination of

30

Beauveria bassiana conidia on the integument of the bark beetle Dendroctonus ponderosae (Coleoptera: Scolytidae). J. Invertebr. Pathol. 44, 304-314. Jaronski, S.T., 2010. Ecological factors in the inundative use of fungal entomopathogens. BioControl 55, 159–185. https://doi.org/10.1007/s10526-009-9248-3 Jenkins, M.J., Hebertson, E.G., Munson, S.A., 2014. Spruce beetle biology, ecology, and management in the Rocky Mountains: An addendum to spruce beetle in the Rockies. Forests 5, 21-71. https://doi.org/10.3390/f5010021 Johnson, D.M., Germino, M.J., Smith, W.K., 2004. Abiotic factors limiting photosynthesis in Abies lasiocarpa and Picea engelmannii seedlings below and above the alpine timberline. Tree Physiol. 24, 377-386. https://doi.org/10.1093/treephys/24.4.377 Kassambara, A., 2018. ggpbubr: ‘ggplot2’ Based publication ready plots. R package version 0.2. https://www.CRAN.R-project.org/package=ggpubr Kassambara, A., Kosinski, M., 2018. survminer: Drawing survival curves using ‘ggplot2.’ R package version 0.4.3. URL https://www.CRAN.R-project.org/package=survminer Khan, S., Guo, L., Maimaiti, Y., Mijit, M., Qiu, D., 2012. Entomopathogenic fungi as microbial biocontrol agent. Mol. Plant. Breed. 3, 63-79. https://doi.org/10.5376/mpb.2012.03.0007 Klepzig, K.D., Flores-Otero, J., Hofstetter, R.W., Ayres, M.P., 2004. Effects of available water on growth and competition of southern pine beetle associated fungi. Mycol. Res. 108, 183-188. https://doi.org/10.1017/S0953756203009055 Kocacevik, S., Sevim, A., Eroglu, M., Demirbag, Z., Demir, I., 2015. Molecular characterization, virulence and horizontal transmission of Beauveria pseudobassiana from Dendroctonus micans (Kug.) (Coleoptera: Curculionidae). J. Appl. Entomol. 139, 381-389. doi:10.1111/jen.12181

31

Kreutz, J., Vaupel, O., Zimmermann, G., 2004. Efficacy of Beauveria bassiana (Bals.) Vuill. against the spruce bark beetle, Ips typographus L., in the laboratory under various conditions. J. Appl. Entomol. 128, 384-389. Kumar, S., Strecher, G., Tamura, K., 2016. MEGA7: Molecular evolutionary genetics analysis version 7.0 for bigger datasets. Mol. Biol. Evol. 33, 1870-1874. Lacey, L.A., Grzywacz, D., Shapiro-Ilan, D.I., Frutos, R., Brownbridge, M., Goettel, M.S., 2015. Insect pathogens as biological control agents: Back to the future. J. Invertebr. Pathol. 132, 1-41. https://doi.org/10.1016/j.jip.2015.07.009 Litvak, M.E., Monson, R.K., 1998. Patterns of induced and constitutive monoterpene production in conifer needles in relation to herbivory. Oecologia 114, 531-540. https://doi.org/10.1007/s004420050477 Lyons, D.B., Lavallée, R., Kyei-Poku, G., Van Frankenhuyzen, K., Johny, S., Guertin, C., Francese, J.A., Jones, G.C., Blais, M., 2012. Towards the development of an autocontamination trap system to manage populations of emerald ash borer (Coleoptera: Burprestidae) with the native entomopathogenic fungus, Beauveria bassiana. J. Econ. Entomol. 105, 1929-1939. Mburu, D.M., Maniania, N.K., Hassanali, A., 2013. Comparison of volatile blends and nucleotide sequences of two Beauveria bassiana isolates of different virulence and repellency towards the termite Macrotermes michealseni. J. Chem. Ecol. 39, 101–108. https://doi.org/10.1007/s10886-012-0207-6 Meyling, N.V., Pell, J.K., 2006. Detection and avoidance of an entomopathogenic fungus by a generalist insect predator. Ecol. Entomol. 31, 162–171. https://doi.org/10.1111/j.03076946.2006.00781.x

32

Moore, G.E., 1970. Isolating entomogenous fungi and bacteria, and tests of fungal isolates against the southern pine beetle. J. Econ. Entomol. 63, 1702-1704. Moore, G.E., 1973. Pathogenicity of three entomogenous fungi to the southern pine beetle at various temperatures and humidities. Environ. Entomol. 2, 54-57. Mota, L.H.C., Silva, W.D., Sermarini, R.A., Demétrio, C.G.B., Bento, J.M.S., Delalibera, I., 2017. Autoinoculation trap for management of Hypothenemus hampei (Ferrari) with Beauveria bassiana (Bals.) in coffee crops. Biol. Control 111, 32-39. https://doi.org/10.1016/j.biocontrol.2017.05.007 Niemczyk, M., Sierpińska, A., Tereba, A., Sokołowski, K., Przybylski, P., 2019. Natural occurrence of Beauveria spp. in outbreak areas of cockchafers (Melolontha spp.) in forest soils from Poland. BioControl 64, 159-172. https://doi.org/10.1007/s10526-019-09927-3 O’Connor, C.D., Lynch, A.M., Falk, D.A., Swetnam, T.W., 2014. Post-fire forest dynamics and climate variability affect spatial and temporal properties of spruce beetle outbreaks on a sky island mountain range. Forest Ecol. Manag. 336, 148-162. https://doi.org/10.1016/j.foreco.2014.10.021 Ogidi, O.C., Nunes, M.D., Oyetayo, V.O., Akinyele, B.J., Kasuya, M.C.M., 2016. Mycelial growth, biomass production and iron uptake by mushrooms of Pleurotus species cultivated on Urochloa decumbens (Stapf) R. D. Webster. J. Food Res. 5, 13-19. https://doi.org/10.5539/jfr.v5n3p13 Pabst, G.S., Sikorowski, P.P., 1980. Susceptibility of southern pine beetle (Dendroctonus frontalis) on oligidic medium to Paecilomyces viridis and also Beauveria bassiana, and Metarhizium anisopliae. J. Georgia Entomol. So. 15, 235-241. Popa, V., Deziel, E., Lavallee, R., Bauce, E., Guertin, C., 2012. The complex symbiotic

33

relationships of bark beetles with microorganisms: A potential practical approach for biological control in forestry. Pest Manag. Sci. 68, 963-975. https://doi.org/10.1002/ps.3307 Poprawski, T.J., Greenberg, S.M., Ciomperlik, M.A., 2000. Effect of host plant on Beauveria bassiana- and Paecilomyces fumosoroseus-induced mortality of Trialeurodes vaporariorum (Homoptera: Aleyrodidae). Environ. Entomol. 29, 1048-1053. https://doi.org/10.1603/0046-225X-29.5.1048 Poprawski, T.J., Jones, W.J., 2001. Host plant effects on activity of the mitosporic fungi Beauveria bassiana and Paecilomyces fumosoroseus against two populations of Bemisia whiteflies (Homoptera: Aleyrodidae). Mycopathologia 151, 11-20. https://doi.org/10.1023/A:1010835224472 R Core Team, 2017. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/ Raffa, K., 2014. Terpenes tell different tales at different scales: Glimpses into the chemical ecology of conifer - bark beetle - microbial interactions. J. Chem. Ecol. 40, 1-20. https://doi.org/10.1007/s10886-013-0368-y Raffa, K.F., Smalley, E.B., 1995. Interaction of pre-attack and induced monoterpene concentrations in host conifer defense against bark beetle-fungal complexes. Oecologia 102, 285-295. Ramoska, W.A., Todd, T., 1985. Variation in efficacy and viability of Beauveria bassiana in the chinch bug (Hempitera: Lygaeidae) as a result of feeding activity on selected host plants. Environ. Entomol. 14, 146-148. https://doi.org/10.1093/ee/14.2.146 Reay, S.D., Brownbridge, M., Cummings, N.J., Nelson, T.L., Souffre, B., Lignon, C., Glare,

34

T.R., 2008. Isolation and characterization of Beauveria spp. associated with exotic bark beetles in New Zealand Pinus radiata plantation forests. Biol. Control 46, 484-494. https://doi.org/10.1016/j.biocontrol.2008.05.006 Reay, S.D., Brownbridge, M., Gicquel, B., Cummings, N.J., Nelson, T.L., 2010. Isolation and characterization of endophytic Beauveria spp. (Ascomycota: Hypocreales) from Pinus radiata in New Zealand forests. Biol. Control 54, 52-60. https://doi.org/10.1016/j.biocontrol.2010.03.002 Rehner, S.A., Buckley, E., 2005. A Beauveria phylogeny inferred from nuclear ITS and EF1-α sequences: evidence for cryptic diversification and links to Cordyceps teleomorphs. Mycologia 97, 84-98. https://doi.org/10.1080/15572536.2006.11832842 Santiago-Álvarez, C., Maranhão, E.A., Maranhão, E., Quesada-Moraga, E., 2006. Host plant influences pathogenicity of Beauveria bassiana to Bemisia tabaci and its sporulation on cadavers. BioControl 51, 519-532. https://doi.org/10.1007/s10526-005-5737-1 Schoch, C.L., Seifert, K.A., Huhndorf, S., Robert, V., Spouge, J.L., Levesque, C.A., Chen, W., 2012. Nuclear ribosomal internal transcribed spacer (ITS) region as a universal DNA barcode marker for fungi. P. Natl. Acad. Sci. USA 109, 6241-6246. https://doi.org/10.1073/pnas.1117018109 Sevim, A., Demir, I., Tanyeli, E., Demirbağ, Z., 2010. Screening of entomopathogenic fungi against the European spruce bark beetle, Dendroctonus micans (Coleoptera: Scolytidae). Biocontrol Sci. Techn. 20, 3-11. doi:10.1080/09583150903305737 Shrivastava, G., Ownley, B.H., Augé, R.M., Toler, H., Dee, M., Vu, A., Köllner, T.G., Chen, F.,

35

2015. Colonization by arbuscular mycorrhizal and endophytic fungi enhanced terpene production in tomato plants and their defense against a herbivorous insect. Symbiosis 65, 65-74. https://doi.org/10.1007/s13199-015-0319-1 Six, D.L., Bentz, B.J., 2003. Fungi associated with the North American spruce beetle, Dendroctonus rufipennis. Can. J. Forest Res. 33, 1815-1820. https://doi.org/10.1139/X03-107 Six, D.L., Bracewell, R., 2015. Dendroctonus. In Vega, F.E., Hofstetter, R.W. (Eds.), Bark Beetles: Biology and Ecology of Native and Invasive Species. Academic Press, Cambridge, pp. 305-350. Srei, N., Lavallée, R., Guertin, C., 2017. Susceptibility of Dendroctonus simplex to Hypocreales fungi: Towards the development of a biological control strategy. J. Appl. Entomol. 141, 487-495. https://doi.org/10.1111/jen.12361 St. Leger, R.J., Allee, L.L., May, B., Staples, R.C., Roberts, D.W., 1992. World-wide distribution of genetic variation among isolates of Beauveria spp. Mycol. Res. 96, 10071015. https://doi.org/10.1016/S0953-7562(09)80108-1 Takov, D.I., Doychev, D.D., Linde, A., Atanasova Draganova, S., Kirilova Pilarska, D., 2012. Pathogens of bark beetles (Curculionidae: Scolytinae) and other beetles in Bulgaria. Biologia 67, 966-972. https://doi.org/10.2478/s11756-012-0086-x Tanyeli, E., Sevim, A., Demirbag, Z., Eroglu, M., Demir, I., 2010. Isolation and virulence of entomopathogenic fungi against the great spruce bark beetle, Dendroctonus micans (Kugelann) (Coleoptera: Scolytidae). Biocontrol Sci. Techn. 20, 695-701. doi:10.1080/09583151003717219 Therneau, T.M., Grambsch, P.M., 2000. Modeling survival data: Extending the Cox model.

36

Springer, New York. ISBN 0-387-98784-3. Vega, F.E., Meyling, N.V., Luangsa-Ard, J.J., Blackwell, M., 2012. Chapter 6 – Fungal entomopathogens. In: Vega, F.E., Kaya, H.K. (Eds.), Insect Pathology, 2nd ed. Academic Press, San Diego, pp 171-220. https://doi.org/10.1016/B978-0-12-384984-7.00006-3 Whitting, E.C., Rizzo, D.M., 1999. Effect of water potential on radial colony growth of Armillaria mellea and A. gallica isolates in culture. Mycologia 91, 627-635. Wickham, H., 2009. ggplot2: Elegant graphics for data analysis. Springer-Verlag New York. Wilke, C.O., 2019. cowplot: Streamlined plot theme and plot annotations for ‘ggplot2.’ R package version 0.9.4. https://www.CRAN.R-project.org/package=cowplot Xu, L., Zhang, Y., Zhang, S., Deng, J., Lu M., Zhang, L., Zhang, J., 2018. Comparative analysis of the immune system of an invasive bark beetle, Dendroctonus valens, infected by an entomopathogenic fungus. Dev. Comp. Immunol. 88, 65-69. doi:10.1016/j.dci.2018.07.002 Yao, J., Zhang, L., Chen, X., Li, Z., 2012. A population survey of Beauveria bassiana in the microhabitat of the red turpentine beetle, Dendroctonus valens, in Chinese pine forests. Biocontrol Sci. Techn. 22, 1254-1267. https://doi.org/10.1080/09583157.2012.724382 Yeo, H., Pell, J.K., Alderson, P.G., Clark, S.J., Pye, B.J., 2003. Laboratory evaluation of temperature effects on the germination and growth of entomopathogenic fungi and on their pathogenicity to two aphid species. Pest Manag. Sci. 59, 156–165. https://doi.org/10.1002/ps.622 Zhang, L.-W., Liu, Y.-J., Yao, J., Wang, B., Huang, B., Li, Z.-Z., Fan, M.-Z., Sun, J.-H., 2011. Evaluation of Beauveria bassiana (Hyphomycetes) isolates as potential agents for control of Dendroctonus valens. Insect Sci. 18, 209–216. doi:10.1111/j.1744-7917.2010.01361.x

37

Table 1. Geographic origin (state) and source of Beauveria bassiana isolates used in this study. Isolate Name 14B 34C 429BTF 429DA 50C 90(1)MPB AZ5 AZ6 D900 ES12(1) GHA L429 L447 Spruce1

Geographic origin Montana Montana Wyoming Wyoming Montana Montana Arizona Arizona Montana Montana/Idaho Proprietary Wyoming Utah Utah

38

Isolate source Pinus ponderosa bark Pinus ponderosa bark Picea engelmannii bark Dendroctonus rufipennis adult Pinus ponderosa bark Dendroctonus ponderosae adult Pinus ponderosa stand soil Pinus ponderosa stand soil Pinus ponderosa stand soil Picea engelmannii stand soil Registered strain of B. bassiana Picea engelmannii bark Dendroctonus rufipennis larva Picea engelmannii bark

Table 2. Beauveria bassiana isolates used in this study and associated GenBank accession matches based on percent identity (shown in parentheses). For three isolates, EF1-α was not characterized. Closest GenBank Accession Isolate Name 14B 34C 429BTF 429DA 50C 90(1)MPB AZ5 AZ6 D900 ES12(1) GHA L429 L447 Spruce1

ITS

EF1-α

B. bassiana MF802497.1 (100.00%) B. bassiana MF872398.1 (100.00%) B. bassiana MF802492.1 (100.00%) B. bassiana MF802491.1 (100.00%) B. bassiana MF872398.1 (100.00%) B. bassiana EU272501.1 (100.00%) B. bassiana MF872398.1 (100.00%) B. bassiana MF872398.1 (100.00%) B. bassiana EU272501.1 (99.30%) B. bassiana MF802497.1 (100.00%) B. bassiana MF802497.1 (100.00%) B. bassiana KX901310.1 (100.00%) B. bassiana MF802497.1 (100.00%) B. bassiana EU721494.1 (100.00%)

B. bassiana AY531925.1 (93.81%) B. bassiana JQ043233.1 (80.21%) B. bassiana JQ043233.1 (97.43%) B. bassiana JQ043233.1 (99.27%) B. bassiana JQ043233.1 (98.76%) B. bassiana AY531925.1 (95.61%) B. bassiana JQ043233.1 (92.93%) B. bassiana JQ043233.1 (94.13%) B. bassiana AY531893.1 (96.54%) B. bassiana AY531925.1 (97.57%) B. bassiana AY531925.1 (99.23%)

39

Table 3. Conditions for bioassays testing Beauveria bassiana isolate pathogenicity to Dendroctonus rufipennis, arranged from least (1) to most (4) representative of the Dendroctonus rufipennis habitat. Experiment number (1) (2) (3) (4)

Beetle source

Temperature

Location of beetle collection

Reared from logs Flight capture Flight capture Flight capture

23 °C 23 °C 10 °C 10 °C

40.52058 N, 105.89283 W 38.49666 N, 106.32558 W 38.49666 N, 106.32558 W 38.49666 N, 106.32558 W

40

Arena substrate Filter paper Filter paper Filter paper Spruce phloem

Table 4. Summary of Median Survival Time for each isolate across experimental conditions with 95% confidence intervals in parentheses. Asterisks denote whether treatment (isolate) significantly differed from negative control treatment (water) based on a log-rank test. Bioassay

Isolate

Experiment (1) 23 °C, reared beetles

Experiment (2) 23 °C, dispersing beetles

10 (8-10) ** 5 (5-5) *** 8 (6-8) *** 5 (4-5) *** 6 (6-6) *** 5 (5-5) *** 7 (5-7) *** 5 (5-5) *** 8 (8-8) *** 5 (4-5) *** 6 (6-8) *** 5 (5-5) *** 8 (8-10) *** 5 (5-5) *** 8 (8-8) *** 5 (5-5) ** 6 (6-8) *** 5 (5-5) *** 8 (8-8) *** 5 (5-5) *** 6 (6-6) *** 5 (5-5) *** 6 (6-8) *** 5 (5-5) *** 6 (6-8) *** 5 (4-5) *** 10 (8-10) * 5 (5-5)*** 10 (10-12) 5.5 (5-6) *P < 0.05; **P < 0.01; ***P < 0.001 14B 34C 429BTF 429DA 50C 90(1)MPB AZ5 AZ6 D900 ES12(1) GHA L429 L447 Spruce1 Control

41

Experiment (3) 10 °C, dispersing beetles

Experiment (4) 10 °C, dispersing beetles, provisioned with phloem

11 (8-11) 11 (8-11) 11 (8-11) 11 (8-11) 8 (8-11) * 11 (11-11) 11 (8-11) 11 (8-11) 11 (8-11) 11 (11-11) 11 (8-11) 11 (8-11) 11 (11-11) 11 (8-11) 11 (11-11)

19 (19-26) ** 38 (31-38) 35 (27-35) 48 (42-64) 19 (15-33) * 34 (26-56) 33 (33-33) 34 (34-52) 34 (27-57) 25 (19-32) 33 (33-44) 40 (28-53) 62 (42-78) 35 (35-54) 31 (20-45)

Figure legends Figure 1. The effect of temperature variability on radial growth rate (mm/day) of Beauveria bassiana isolates. Black circles show the mean growth rate of all isolates at each temperature. Bars show ± 1 SD.

Figure 2. The mean proportion (percent) of area captured by Beauveria bassiana isolates (dark gray bars) when paired against multiple isolates of the spruce beetle symbiotic fungus Leptographium abietinum (light gray bars) in a resource-limited environment. The dashed line denotes 50%, the proportion that would be expected if each fungus captured equivalent areas on average.

Figure 3. The effects of the isolate identity × monoterpene concentration interaction on relative radial growth of fourteen Beauveria bassiana isolates for five monoterpenes found in the phloem of all Picea engelmannii. Lack of a bar indicates total growth inhibition (i.e., no growth) at that concentration. Error bars show ± 1 SE, and shading denotes monoterpene concentration.

Figure 4. Relative radial growth rate of fourteen Beauveria bassiana isolates in minimal chitin media (water agar + chitin) at three concentrations. Error bars show ± 1 SE, and shading denotes chitin concentration in media (v/v).

Figure 5. Relative effect of media osmotic water potential (MPa) on radial growth of Beauveria bassiana isolates; increasingly negative MPa values are associated with a higher degree of water deficit. Error bars show ± 1 SE.

42

Figure 6. Relative impacts of exposure to low-intensity sunlight (photoperiod: 13 L/11 D; 4.8 ± 0.2 SE µmol/m2/sec) on radial growth of Beauveria bassiana isolates, as compared to growth rates under dark conditions (photoperiod: 0 L/24 D). Error bars show ± 1 SE.

43

Figure 1.

44

Figure 2.

45

Figure 3.

46

Figure 3, continued.

47

Figure 3, continued.

48

Figure 4.

49

Figure 5.

50

Figure 6.

51

52

Figure S1. Kaplan-Meier survivorship curves for an in vitro bioassay that applied fungal spores to filter papers at 23 °C in the dark. Solid black lines indicate survival time of lab reared Dendroctonus rufipennis after exposure to the indicated isolate of Beauveria bassiana. Solid gray lines show sham treatments (no B. bassiana), solid color fills denote 95% confidence intervals, and dashed black lines show median survival time (50% mortality) for each treatment and sham treatment. Asterisks indicate P-value (*P < 0.05; **P < 0.01; ***P < 0.001) based on a log-rank test comparing the median beetle survival time when exposed to sham treatment to median beetle survival time when exposed to each isolate.

53

54

Figure S2. Kaplan-Meier survivorship curves for an in vitro bioassay that applied fungal spores to filter papers at 23 °C in the dark. Solid black lines indicate survival time of flight captured Dendroctonus rufipennis after exposure to the indicated isolate of Beauveria bassiana. Solid gray lines show sham treatments (no B. bassiana), solid color fills denote 95% confidence intervals, and dashed black lines show median survival time (50% mortality) for each isolate and the sham treatment. Asterisks indicate P-value (*P < 0.05; **P < 0.01; ***P < 0.001) based on a log-rank test comparing the median beetle survival time when exposed to sham treatment to median beetle survival time when exposed to each isolate.

55

56

Figure S3. Kaplan-Meier survivorship curves for an in vitro bioassay that applied fungal spores to filter papers at 10 °C in the dark. Solid black lines indicate survival probability of flight captured Dendroctonus rufipennis after exposure to the indicated isolate of Beauveria bassiana. Solid gray lines show sham treatments (no B. bassiana), solid color fills denote 95% confidence intervals, and dashed black lines show median survival time (50% mortality) for each isolate and the sham treatment. Asterisks indicate P-value (*P < 0.05; **P < 0.01; ***P < 0.001) based on a log-rank test comparing the median beetle survival time when exposed to sham treatment to median beetle survival time when exposed to each isolate.

57

58

Figure S4. Kaplan-Meier survivorship curves for an in planta bioassay that applied fungal spores to phloem sandwiches at 10 °C in the dark. Solid black lines indicate survival time of Dendroctonus rufipennis after exposure to the indicated isolate of Beauveria bassiana. Solid gray lines show control treatments (no B. bassiana), solid color fills denote 95% confidence intervals, and dashed black lines show median survival time (50% mortality) for each isolate and the sham treatment. Asterisks indicate P-value (*P < 0.05; **P < 0.01; ***P < 0.001) based on a log-rank test comparing the median beetle survival time when exposed to sham treatment to median beetle survival time when exposed to each isolate.

59

   

Beauveria bassiana isolates vary in their response to abiotic and biotic factors Tree secondary metabolites and low temperatures inhibit Beauveria bassiana growth Many isolates are pathogenic in the laboratory but not under in natura conditions Natural phloem substrate greatly influences Beauveria bassiana virulence

60

CRediT authorship contribution statement Andrew J. Mann: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Writing – original draft, Writing – review & editing. Thomas Seth Davis: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Supervision, Writing – review & editing

61

62