Introduction
Climate change-induced dieback is a global problem affecting forest ecosystems worldwide (Allen et al. 2010; Anderegg et al.2012; Young et al. 2017; Hartmann et al. 2018). The causal factors underlying forest dieback have been commonly attributed to complex interactions between abiotic and biotic factors (Simler-Williamson et al. 2019). For instance, drought is often viewed as a predisposing factor responsible for promoting outbreaks by some insect species such as bark beetles (Coleoptera: Curculionidae, Scolytinae) that attack the main stems of mature trees (Gaylord et al. 2013; Anderegg et al. 2015; Netherer et al. 2019; Gely et al. 2020; Öhrn et al . 2021). Although there are proposed mechanisms underlying drought-induced tree mortality, i.e., a failure of the plant’s water transport-hydraulic-system and carbon starvation due to prolonged stomatal closure and reduced photosynthesis (McDowell et al. 2008; Anderegg et al. 2012; Meir et al. 2015; Adams et al. 2017; Choat et al. 2018; Hartmannet al. 2018), we have less understanding of the combined effects of drought and insect attacks in the field (Kolb et al. 2016; Stephenson et al. 2019; Huang et al. 2020). Significant challenges in conducting field experiments include experimental control of both water stress and bark beetle attacks on the mature trees, and measurement of the cascading physiological changes through tree death (Kolb et al. 2016; Choat et al. 2018).
Most studies focusing on tree mortality were either conducted during or after observed mortality (retrospectively) (e.g., Camarero et al.2015; Gaylord et al. 2013; 2015; Kolb et al. 2016; Öhrnet al . 2021). Moreover, those studies primarily examined the role of a single stress agent such as drought or herbivory (e.g., Meddenset al. 2014; Erbilgin et al. 2017) although their effects can be interacting. Likewise, some studies utilized seedlings under controlled conditions to improve standardization (i.e., Turtola et al. 2003; Lusebrink et al. 2011) but it is well recognized that plant ontogeny strongly affects many responses and thus results obtained with seedlings may not be applicable to mature trees (Boege & Marquis, 2005; Erbilgin & Colgan, 2012; Moreira et al. 2017). As such, experimental demonstrations of both predisposing and biotic factors leading toand measuring the physiological mechanisms responsible fortree mortality in the field are sparse (Kolb et al. 2016; Huang et al. 2020). This knowledge gap hinders our ability to accurately understand the mechanism of tree mortality from multiple and sometimes synergistic factors.
Drought can have profound impacts on tree responses to subsequent insect attacks. These responses vary by feeding guild and are most impactful with bark and wood boring insects (reviewed by Kolb et al. 2016). Briefly, prolonged droughts induce stomatal closure, which in turn reduces photosynthesis, potentially leading to depletion of carbohydrate reserves, i.e., non-structural carbohydrates (NSCs), comprised of soluble sugars and starch (McDowell et al. 2008; Anderegget al. 2012; Mitchell et al. 2013; Choat et al.2018). Hydraulic failure occurs due to disruption of water movement in the xylem due to formation of air bubbles (embolism) during desiccation (Adams et al. 2017). Furthermore, NSC depletion may also reduce water movement and retention, promoting cellular dehydration (Salaet al. 2012; Deans et al. 2020). Reductions in carbohydrate production and depletion of NSCs due to drought can influence many functions in plants including biosynthesis of carbon-dependent terpenes that serve antidessication and antiherbivory functions (McDowell et al. 2011; Huang et al. 2020; Hussain et al. 2020).
Ponderosa pine (Pinus ponderosa ) is one of the most widely distributed conifer species in western North America, ranging from southern Canada to Mexico, and from the Plains States to the Pacific Coast. Two major threats to ponderosa pines, and other co-occurring pine species, are drought and bark beetles (Negrón et al. 2009; Bentzet al. 2010; Raffa et al. 2013; Creeden et al.2014). Over recent decades, several million ponderosa pine trees were killed by a combination of these two stressors in various parts of the species’ range (Savage et al. 1994; USDA Forest Service 2002; Breshears et al. 2005; Allen et al. 2010; Pile et al. 2019).
Bark beetles’ reproduction involves offspring completing their development within the phloem layer beneath the outer bark of their host trees. Pine phloem contains toxic carbon-dependent secondary metabolites including monoterpenes, diterpene resin acids, and phenolics (Francheshiet al. 2005; Keeling & Bohlmann 2006; Celedon & Bohlmann 2019; Erbilgin, 2019). Studies have reported that monoterpenes and diterpenes are toxic to the bark beetles and their mutualistic fungi (Kopperet al. 2005; Chiu et al. 2017; Ullah et al. 2021). Numerous studies with simulated or actual herbivore attacks have extensively shown mobilization and transport of stored NSCs in support of the synthesis of defense metabolites in plants including conifers (Goodsman et al . 2013; Dietze et al. 2014; Hartmann & Trumbore 2016; Adams et al . 2017; Li et al . 2018; Rothet al . 2018; Hussain et al . 2020). Effects of drought on bark beetles can be positive or negative depending on drought intensity and duration (Lombardero et al. 2000; Gaylord et al. 2013, 2015). Currently, we lack a clear understanding about the importance of combined drought and bark beetle attacks on NSCs, carbon-based defense metabolites, and their interactions in tree mortality (Kolb et al. 2016; Huang et al. 2020).
An earlier study suggested a relationship between drought and ponderosa pine terpene defenses (Kolb et al. 2019), but how drought and biotic challenges interact to influence NSCs and their relationships with carbon-dependent defense metabolites (terpenes) has not been jointly investigated and remains poorly understood. Thus, our primary objective was to characterize the changes in NSC and terpene concentrations in ponderosa pine trees that were experimentally exposed to a factorial combination of drought stress (induced via root trenching), pheromone-induced bark beetle attacks, and crushing beetles onto phloem/xylem interface in the experiment described in Kolb et al . (2019). We hypothesized that the combination of drought and bark beetle attacks more rapidly deplete NSCs in tree stems that either stress agent alone, thereby eliciting tree mortality. We addressed the following five research questions: (1) Do drought stress, bark beetle attacks, and the beetle crushing influence NSCs of ponderosa pine trees? (2) Does drought impact the effect of biotic stressors on NSCs and terpenes? (3) Do changes in NSCs result in changes in defense metabolites? (4) Do concentrations of NSCs and defense metabolites vary between live and dying ponderosa pine trees? (5) Can interactions between NSCs and defense metabolites in dying ponderosa pine trees explain their mortality?