Research - LinnenLab

As evolutionary biologists, the core goal of our research is to understand how and why variation exists in nature. On the "how" side of things: how do changes at the DNA level give rise to changes in how organisms look and behave? On the "why" side of things: why do differences in the way organisms look and behave impact their ability to survive and reproduce? More generally, how do organisms adapt to new conditions and why do some changes lead to the formation of entirely new species? To answer these questions, we focus on a group of plant-feeding insects that serves as an excellent evolutionary model system (see "Critters" tab). We spend our time: collecting critters in the field, conducting experiments with live insects in the lab, extracting DNA and RNA at the bench, and analyzing large amounts of genomic data at our desks. Below are some recent and ongoing projects in the lab.

Adaptation to a new environment requires many different changes, including alterations in behavior, physiology, and morphology. To understand the number and types of genetic changes required to adapt to a new host plant, we are investigating the genetic basis of host-use differences between N. lecontei (left column) and N. pinetum (right column). Some time after diverging from a common ancestor, N. pinetum adapted to white pine, a thin-needled pine that N. lecontei tends to avoid. In addition to evolving a very strong preference for white pine, N. pinetum also evolved several traits that increase survival on the preferred host, including: a smaller ovipositor, a tendency to lay fewer eggs per needle, and increased larval performance. To uncover the genetic changes that produced these new traits, we use several complementary approaches, including: population genomics, gene-expression analysis, and quantitative trait locus (QTL) mapping. Together, these data will shed light on how organisms adapt to new environments. (Photos by: Robin Bagley, Melanie Hurst, and Ryan Ridenbaugh)

Color is an exceptionally useful trait for trying to understanding phenotypic variation in the wild. Not only is it easy to see and measure color traits, we also know a great deal about the underlying genetic pathways that produce different types of pigments. As color genes have been identified in a growing number of organisms, some interesting patterns have started to emerge: color variation tends to be genetically "simple" (small number of genes) and is often attributable to a handful of key color genes. To determine how genetically "predictable" color variation is, we are investigating the genetic underpinnings of color variation in Neodiprion larvae. Pine sawflies are a wonderful system for addressing this question because: (1) populations and species vary in many different pigmentation traits (body color, spotting/striping pattern, head color), (2) different color traits have evolved repeatedly in different species, and (3) we can make crosses between many different color morphs in the lab (figure above depicts a cross between two N. lecontei populations that differ in both body color and spotting pattern).

Like many organisms, sawflies experience dramatic changes in the way they interact with the abiotic and biotic environment as they develop. For example, larvae must consume pine needles and fend off predators and parasites to survive and grow, while adults do not feed at all and must locate suitable mates (males) and places to lay eggs (females). To understand how a single genome can produce such profound changes in morphology and ecology across an individual's life span, we are characterizing patterns of gene expression across development and between the sexes. We have found that whereas some genes are always highly expressed, others are very stage-specific (see Venn diagram above, which shows chemosensory genes that are highly expressed in at least one life stage). We hypothesize that genes with stage-specific expression will be frequent targets of natural selection. To test this hypothesis, we are characterizing differences in gene expression and genomic signatures of selection among many different sawfly species (see "Critters" tab).

Neodiprion females live for ~1-3 days in the wild. During that time, they must locate a suitable pine tree, select healthy needles, and deposit their eggs in needles in such a way that the eggs can survive for weeks to months. Laying eggs is a tricky business because pine needles are thin and packed full of sticky resin. Oviposition mistakes have dire consequences: selecting the wrong host or laying eggs improperly can cause all of a female's offspring to die. We therefore expect the senses of smell, taste, and touch to be under strong natural selection. To test this hypothesis- and to evaluate the relative importance of different sensory modalities to host adaptation- we are combining experimental manipulations (how does removing a particular sensory input impact oviposition?) with comparative genomics (do genes related to sensory functions show signatures of natural selection?). (left photo: ovipositing N. virginiana female, R. Bagley; right photo: N. lecontei eggs with a resin-draining slit, R. Ridenbaugh).

Like many other organisms that are distasteful or toxic to potential predators, some Neodiprion species advertise their unpleasantness via bright colors and high-contrast patterns. This is a common and effective anti-predator strategy because similarly colored individuals can share the cost of training naive predators, while reaping the benefits of low predation risk. However, this raises an evolutionary paradox: how can new color morphs ever persist if they get immediately attacked by naive predators? To resolve this paradox, we are working with Neodiprion species that vary in warning color within and between populations. To understand both the evolutionary processses and selective pressures that enabled the novel color morphs to spread, we are combining population genomics and experimental work. We are also working in collaboration with Dr. Carita Lindstedt at the University of Jyväskylä in Finland. (Photos: yellow and white N. lecontei morphs, by drphotomoto and Bea Leiberman [iNaturalist]).

Across species, Neodiprion larvae have very similar chemical defenses against predators. They all feed on conifer needles and sequester host resins in a specialized pouch. When disturbed, they rear back and regurgitate these resins in a charming "Hallelujah" display (see left column in photo above). These resins defend them against both arthropod and vertebrate predators. But despite highly similar larval chemical defenses, Neodiprion species vary in color (cryptic to conspicuous) and grouping strategy (solitary to large groups). To understand why these different strategies have evolved, we are taking a comparative phylogenetic approach in which we measure many different defensive traits and environmental variables from many different species. With these data, we can describe how different traits correlate with one another and with the environment to test specific hypotheses about the evolution of anti-predator defenses. (Photos by Robin Bagley and Kim Vertacnik).

Over 25% of all described species are plant feeding insects. In pine sawflies- and many other insects- changes in host use are associated with both population divergence and speciation. A goal of several ongoing projects in the lab is to figure out why. In other words, what is the mechanistic link between using different hosts and becoming different species (i.e., no longer able to produce viable, fertile offspring)? To address this question, we are taking an experimental approach to establish causal links between host-use traits and reproductive barriers between species. For example, we have found that while N. lecontei and N. pinetum are both very well adapted to laying eggs in their preferred pines (see "genetic basis of host adaptation" project), hybrid females have very low fitness because they prefer N. pinetum's host, but lay their eggs more like N. lecontei (top image: lecontei-like oviposition causes white pine needles to dry out and eggs fail to hatch; bottom image: the same oviposition pattern yields successful hatching on a non-white pine needle; both photos by Robin Bagley). The poor egg-laying performance of hybrid females directly reduces gene exchange in the wild. In addition to characterizing "postzygotic" barriers to gene exchange, we are also asking how different host preferences impact mating behaviors, thereby contributing to "prezygotic" isolation.

To reproduce successfully, a male and a female must: be present at the same time and same place, recognize one another as potential and acceptable mates, be physically capable of exchanging gametes, have gametes that can join to produce a viable zygote, and their offspring must themselves be viable and fertile in their natural habitat. As species diverge, these abilities break down until they can no longer interbreed at all. But what types of barriers to reproduction are most important to speciation? How early in the speciation process do particular barriers tend to arise and by how much do they reduce gene exchange? To address these questions, we are combining field and lab experiments to quantify the strength of different types of reproductive barriers in pairs of Neodiprion species. We are also examining how reproductive isolation varies across the geographic ranges of species pairs. One somewhat surprising result we've uncovered is that even "good" Neodiprion species (those that overlap in nature and manage to remain distinct) will mate in the lab and, often producing viable and fertile hybrid offspring. The lack of "intrinisic postzygotic isolation" (inviability or infertility caused by genetic incompatibilities) in this system is especially surprising (given current speciation theory) and we are currently exploring reasons why this may be. Photo: mating N. lecontei pair (female on left), by Robin Bagley.

As populations and species diverge, they build up genetic differences across the genome. When divergence is accompanied by gene flow, regions of the genome that contribute to divergent adaptation and reproductive isolation will be among the first to diverge, emerging as peaks of differentiation compared to the rest of the genome. Thus, genome scans provide a potentially powerful tool for finding genes that contribute to adaptation and speciation. However, genomic differentiation is also shaped by other evolutionary processes and local genomic features (e.g., gene content and local recombination rate). Moreover, when diverging populations and species experience different levels of gene flow and selection across their ranges, genomic differentiation will vary across space. To better understand the factors that shape genomic differentiation, we are characterizing genome-wide genetic differentiation at different time points in divergence and, for a give species pair, in different locations across their geographic range. For example, the figure above (unpublished data from Ashleigh Glover and Jeremy Davis) shows how genomic differentiation changes when we compare closely related populations (N. lecontei on different host plants) vs. when we compare closely related species (N. lecontei vs. N. pinetum).

Comparative genomic studies in fruit flies (Drosophila) suggest that ecological specialization is often accompanied by predictable changes at the genetic level. These changes include an accumulation of loss-of-function mutations in chemosensory genes and elevated rates of molecular evolution in intact genes. However, the generality of these patterns remains unclear. Also, an increasing number of studies suggest that particular genes are repeatedly involved in adaptive evolution. To test the extent to which changes in host use are "predictable," we are characterizing chemosensory genes (and other candidate host-use genes) in the genomes of many different Neodiprion species. We are then asking whether specialist species (feed on a single pine) show evidence of gene loss or positive selection relative to closely related generalist pine feeders (feed on multiple pines). We are also asking whether particular genes are repeatedly targeted in relation to gaining or losing a particular pine host. Together, these data will provide insight into the genetic predictability of host-use evolution.

Compared to other pine species that are in eastern North America, eastern white pine has incredibly thin needles. This poses a challenge for egg-laying females because they must embed their eggs in the needles without causing too much damage. Only a single North American species, N. pinetum, has managed to overcome this difficulty and enjoys essentially no competition from native Neodiprion species. Yet shortly after arriving in North America (~1914), the introduced pine sawfly (Diprion similis) managed to colonize this host and has now spread across much of the eastern US and Canada. This unintentional experiment provides a unique opportunity to investigate the predictability of white-pine adaptation. In particular, what phenotypic and genetic changes accompanied adaptation to white pine in North America? To address this question, NSF postdoc Jeremy Davis documenting changes in genotype and phenotype over time using museum collections from the native and historical ranges. Photo: D. similis damage on planted white pine in Lexington, KY (Emily Bendall).

Like all Hymenoptera, pine sawflies are haplodiploid: males develop from unfertilized eggs and are haploid; females develop from fertilized eggs and are diploid. The fact that males are haploid can have profound consequences for evolution. For example, because males are haploid, any new recessive mutation will immediately be visible to natural selection (by contrast, in diploids, recessive mutations are practically invisible if they are at low frequencies). In a nutshell, having haploid males should make natural selection much more efficient, which can have major consequences for the pace of adaptation and speciation. The fact that males make no genetic contribution to a female's male offspring can have other evolutionary consequences as well. We are broadly interested in how haplodiploidy impacts all sorts of evolutionary outcomes, from rates of evolution to outcomes of hybridization and patterns of genomic differentiation. We combine theoretical and empirical data and have been working in collaboration with Dr. Vitor Sousa at the University of Lisbon in Portugal. Photo: two gynandromorph N. lecontei that emerged from our lab colonies. Both have mostly female-like bodies, with one (right) or two (left) male antennae, by Robin Bagley.