Research in the Willett lab

Photo by G. Rouse

In the Willett lab we focus on the evolutionary genetics and genomics of aspects of speciation and adaptation. The basic ideas are how can one species split into two over time and what factors influence how organisms adapt to their environment. Current work centers on exploring questions in these areas using copepods and caterpillars as model systems.

 

Evolution of Reproductive Isolation

Much of the current work in our lab is focused on the intertidal copepod Tigriopus californicus. This copepod species inhabits rocky pools in geographically isolated rocky outcrops along the Pacific coast of North America. What is most interesting about this species from an evolutionary perspective is that these copepod populations have become genetically isolated and make an excellent system to study how recently isolated populations can begin to accumulate reproductive isolation and ecological adaptations (particularly now with the availability of our genome sequence) . Populations of this species then are an instructive model for the initial stages of speciation and adaptation.

 

Below we will highlight previous findings from our lab on this topic:

Tigriopus californicus female; photo by G. Rouse

Mitochondrial-Nuclear Coadaptation and Postzygotic Reproductive Isolation:

The proper function of the Electron Transport System requires coadaptation between nuclear-encoded and mtDNA-encoded proteins. When this coadaptation diverges between populations, hybridization between populations can result in compromised function of these complexes and potentially lowered hybrid fitness (hybrid breakdown). These fitness reductions would be examples of postzygotic reproductive isolation via a mechanism known as Dobzhansky-Muller (DM) incompatibilities. Few genes involved in DM-incompatibilities have been characterized to date so we do not know in general the types of genes that are involved and the evolutionary forces that have acted upon these genes.

 

We have studied to date: (1) the impacts of this genomic coadaptation upon the molecular evolution of the proteins and sequences involved in genomic coadaptation (MBE 2004), (2) how well these complexes function at the enzymatic level when they are composed of proteins from different copepod populations, and (3) how hybrid fitness is affected by mismatched combinations of interacting proteins from different populations (Evolution 2001, Evolution 2003, Genetics 2006, J. of Heredity 2008). A review of these studies can be found in Genetica 2011.

 

Recently we have extended this approach to the genome, first with markers spanning all chromosomes in the genome  (G3 2016) and later in whole genome sequencing work by Graduate Student Thiago Lima (and later in his postdoc in Ron Burton's Lab). This work highlights the large impact mito-nuclear interactions can have on reproductive isolation and contrasts with less prevalent impacts of nuclear-nuclear interactions (J. of Heredity 2018; Evolution 2019)

 

 

Complex III of the Electron Transport system displays the intimate coordination that occurs between mtDNA-encoded (CYTB) and nuclear encoded proteins (CYC, CYC1, and RISP) within these complexes of the mitochondrion (from Genetics 2006)

A more rapid rate of evolution of mtDNA may help drive the coevolutionary process with the nuclear genome. The mitochondrial genome of T. californicus is evolving at a much more rapid rate in this species in comparison to the nuclear genome, as much as 55-fold higher (JME 2012). This elevated rate stands in contrast to many other invertebrates that have only modestly higher mtDNA evoluitonary rates in comparison to nuclear genes.  The genome paper that we published  in collaboration with the Burton, Edmands, and now Barreto labs highlights the fact that this higher rate of mtDNA evolution is associated with faster evolution of proteins imported into the mitochondria that interact with mtDNA-encoded products but not those that do not interact (Nat. Ecol. Evol. 2018).

 

Current work  focuses on exploring the genetic basis of reproductive isolation in T. californicus using genomic approaches and exploring whether paternal inheritance of mtDNA could be slipping trhough in hybrids in contrast to the normal maternal inheritance pattern seen in most animals(work by graduate student Jeeyun Lee). Preview of Jeeyun's work here.

Figure from Barreto et al (2018) that compares the rate of evolution for nuclear-encoded proteins that are imported into the mitochondria and interact closely with mtDNA-encoded products versus those that do not interact as closely.

Return to Top

Potential Candidates Genes Underlying Reproductive Isolation

Many of the enzymes of central metabolism were surveyed using allozymes in a number of different species in the past. In some cases these studies yielded patterns consistent with selection acting upon these proteins. However, to date, little is known about whether divergence in these proteins could contribute to postzygotic reproductive isolation and speciation.

 

We have examined two of these proteins (Malic Enzyme-ME and Glutamate oxaloacetate transaminase-GOT2) that had previously been suggested to be involved in hybrid breakdown in these copepods (Burton 1987). Our work has shown that the ME2 gene is in a region of the genome with a large impact on hybrid viability and likely interacts with other genes to contribute to hybrid breakdown (JEB 2007). The region of the genome marked by the GOT2 locus interacts with the ME2 region in crosses of one population but not another suggesting that this incompatibility is likely to be complex and involves more than two partners (PLoS ONE 2011).

 

Finally, we have looked at the evolution of a set of gene duplicates in the T. californicus genome of cytoplasmic GOT1 genes. These genes show interesting patterns of gene conversion where regions of some duplicates are either highly divergent or highly similar to one another across copies in the same population (BMC Evol. Biol. 2013).

Reductions in hybrid fitness associated with genotypes at two different Malic Enzyme loci. ME2 is associated with the most dramatic departures from expected Mendelian inheritance patterns (from JEB 2007).

Return to Top

Genome-wide Explorations of Reproductive Isolation

The approaches outlined above focused on a relatively narrow set of candidate genes but did not look at patterns across the genome as a whole. Recent studies in our lab have taken started to explore the patterns of loci across the genome and how these contribute to the accumulation of postzygotic isolation in crosses of populations of T. californicus.

 

A study we recently published (G3 2016) looked a modest number of markers that spanned all twelve chromosomes of T. californicus in crosses between two southern California populations. Interestingly the largest impacts were centered on chromosome 2 (the chromosome that contains the ME2 locus described above). One other novel feature of this copepod system is the lack of recombination in female copepods. By genotying individuals in different backcrosses to the original populations we can determine the impact of entire chromosomes on hybrid viability. We found that multiple factors on chromosome 2 are likely to contribute to a drop in hybrid viability.

 

An alumnus of the lab, Thiago Lima, has completed a study examining patterns of skew across the entire genome in cross of two populations that shows vastly different patterns with highly skewed frequencies in adults but not larval copepods (Lima and Willett J. Herd. 2018). We have extended this to a larger set of crosses and reciprocal crosses to look at the genome-wide impacts of mito-nuclear  incomptabilities (Evolution 2019).

 

Speaking of genomes, in collaboration with the labs of Ron Burton at SIO, Suzanne Edmands at USC, and Felipe Barreto at OSU, we have finished a paper describing  genomes from a set of eight populations of T. californicus. Web resources for this genome are the genome sequence/annotation for the San Diego population of T. californicus and the annotation in a genome browser. We did a lot of work on the  San Diego assembly and now for this ~200Mb genome >99% of the sequence is in 12 scaffolds corresponding to the 12 chromosomes of this species.

The impact of regions of the genome on hybrid viability are reflected in skews in markers across the 12 chromosomes of T. californicus in crosses of two southern California copepod populations. Blue triangles give the viability for the AB alleles, while squares give the SD allele viability. See G3 (2016) paper for more details

Figure from Barreto et al. (2018) showing the 8 populations of Tigriopus californicus for which genomes were produced and a tree based on 150kb of sequence from 100 genes.

Return to Top

Evolution of Hybrid Sterility

In crosses between many species hybrid male sterility is much more common than hybrid female sterility suggesting that dominance effects are important (interactions involving sex chromosomes) and that males evolve faster than females (in reproductive traits). T. californicus are interesting because they lack differentiated sex chromosomes and may have lessened sexual selection acting on reproductive traits in males.

 

We looked at levels of hybrid male and female sterility in crosses between T. californicus populations. Although low levels of hybrid sterility were found, no difference was found between males and females suggesting there is not faster male evolution of hybrid sterility in this species (Genetica 2008). This has implications for the faster male evolution hypothesis of male hybrid sterility.

Return to Top

Evolution of Environmental Tolerance in Copepods

The genetically divergent populations of T. californicus are useful for studying how organisms adapt to their physical environment. The upper intertidal pools that this species inhabits can vary dramatically in both temperature and salinity when they are isolated from the more stable ocean water. We are studying both the patterns of thermal, hypoxia and salinity tolerance and how populations can become locally adapted to differences in environmental conditions. Current work is also exploring how this tolerance and adaptation interacts with a breakdown in mito-nuclear coadaptation in hybrid copepods.

 

Below we will highlight previous findings and current work from our lab on this topic:

Thermal Adaptation and its Evolution across Populations

Thiago Lima next to copepod pools on Bird Rock, La Jolla, CA. These pools are near or above the high tide line and vary substantially in temperature and salinity. Photo by C. Willett.

T. californicus can  experience dramatic swings in temperatures both seasonally and daily in its intertidal pools. Our work has shown that there is a latitudinal pattern of adaptation in high temperature tolerance across populations of this species with southern populations able to handle higher temperatures (Evolution 2010). Interestingly, there appears to be fitness tradeoffs with this adaptation such that southern populations have lower competitive fitness at moderate temperatures. We have also seen that hybrids between populations do not appear to be greatly compromised in thermal tolerance which may reflect a lowering of hybrid breakdown for this phenotype (Journal of Heredity 2011).

 

We have also recently completed a study of gene expression across the transcriptome using the sequencing technique of RNA-seq. This work aimed to understand what gene expression changes are associated with the competitive fitness differences seen across populations and different patterns of exposure to thermal variation. Interestingly, different populations with similar phenotypic responses to thermal variation showed very dissimilar gene expression responses suggesting that the underlying mechanisms of adaptation may differ for adaptation to the same conditions (Lima and Willett 2018 Ecol. Evol.)

Survival of copepod populations to acute heat stress decreases with increased latitude (Northern populations are on the right side). Evolution 2010.

Return to Top

Salinity Tolerance in Copepods and Variation across Populations

Tigriopus californicus can be exposed to dramatically changing salinity conditions in the upper intertidal pools that they inhabit. This ability to physiologically adapt to changing salinity has been shown to be compromised in hybrids between populations and this appears to be due in part to the inability to correctly synthesize what are known as free amino acids (Burton 1990). Uncovering the genes involved in this process could then reveal how this potential contributor to environmentally-determined reproductive isolation has evolved.

 

We have previously completed several studies looking at genes involved in the accumulation of one of these free amino acids (proline). This work has shown that key enzymes are not regulated at the mRNA stage and might be regulated via either translation control or protein modification ( CBP_B 2002, CBP_B 2003). Further studies are needed to elucidate the role of these proteins in breakdown in salinity tolerance.

 

We have also characterized the physiological response to copepods from different populations to stress caused by either increasing or decreasing salinities (PBZ 2021). In this same publication, graduate student Jeeyun Lee coupled this tolerance work with RNA-seq to examine how gene expression is changing in response to these salinities stresses in a pair of populations that differ in their response.

Pathway for the production of the free amino acid proline which allows copepods to adjust to changing salinity conditions. The enzymes P5CS and P5CR were characterized in CBP_B 2002

Return to Top

Copepod tolerance to low oxygen and low pH

In contrast to the nearby ocean, the tidepools that T. californicus inhabits vary considerably in both oxygen and pH driven largely by their small size and the processes of photosynthesis (of algae) and respiration (of all organisms). Graduate student Aimee Deconinck has been exploring how different populations of copepods handle both of these stressors. T. californicus is able to handle very low oxygen conditions (anoxia) for a surprisingly long period of time. Aimee has found a correlation between anoxia tolerance and latitude but not for pH (preview of results here). Currently, she is exploring how anoxia and hybrid breakdown interact (likely to be important given the central role of mtDNA in hybrid breakdown and the intimate connection of mitochondrial metabolism to oxygen levels.

Here is Aimee Deconinck working with the anoxia/pH culture box that she developed and built for this research on low oxygen and pH tolerance in T. californicus

Return to Top

Thermal Tolerance in Caterpillars and its Parasitoid Cotesia congregata

We have a collaboration with Joel Kingsolver's lab looking at how organisms respond to variation in temperature over time (i.e. warmer during the day and cooler at night) and how these results match up with the bulk of laboratory thermal biology studies that are done at constant temperatures. This work is focused on caterpillars of Manduca sexta, the tobacco hornworm (although you can also find them eating your tomato plants). We have also begun work on how thermal stress impacts the interactions between M. sexta and a parasitoid wasp (Cotesia congregata). Joel's lab has used M. sexta extensively in the past for thermal biology studies. Our lab is focused on the genetic end of the project.

Impact of fluctuating temperatures on gene expression in Manduca sexta

We are collaborating with Joel Kingsolver's lab  to examine how fluctuating temperatures impact the thermal biology of Manduca sexta. We have examined  wide-ranging sets of measures from growth rate and respiration to transcriptome-wide gene expression. We also have the opportunity to look at the evolution of thermal traits in this system by comparing lab and field populations of M. sexta that have been experiencing divergent selection for more than 300 generation and have diverged in their thermal performance.

 

Manduca sexta is an excellent system for studying many aspects of thermal physiology and it  has also become more amenable to genetic studies as well. A reference genome has now been released and a good deal of effort has made to annotate this sequence (link to genome paper). In a project led by graduate student Meggan Alston, and now published (J. Exp. Biol 2020) we used  RNA-seq to explore differences in gene expression associated with fluctuating temperatures and exposure to high temperatures. Large differences were seen between field and lab populations that appear to signal significant differences in how well these populations can acclimate to thermal stress.

Manduca sexta caterpillar on tomato plant (it isn't  just a tobacco hornworm). On-going work in collarboration with Joel Kingsolver is examining the impacts of fluctuating temperatures on thermal physiology in this species. Photo by EricM (licence info)

Parasitism by the wasp Cotesia congregata and how temperature impacts its interactions with Manduca

In a new proect, we are again collaborating with Joel Kingsolver's lab with  funding from NSF to examine how the interactions between Manduca and the parasitoid wasp C. congregata are impacted by thermal stress. Previous work from the Kingsolver lab has shown that a heat shock can kill the developing larvae of the wasp while not killing the caterpillar. We are delving into the  how this happens at multiple levels including , genetics, immune function, and the involvement of an endogenous virus that the female wasp injects along with her eggs that facilitate the normally successful hijacking of the Manduca caterpillar for wasp larvae development. Look for interesting results from graduate student Katherine Malinski on this work soon!

Manduca sexta caterpillar with cocoons from Cotesia congregata. Emerging wasp pictured. We are studied how temperuature impacts this parasitism at multiple levels including gene expression and immune function. Photo by Beatriz Moisset (licence info)

Return to Top

Other previous studies in the lab

We have done work only several other systems and welcome collaborations with researchers that are interested in topics related to topics discussed above. Below there is a brief discussion of studies on moths, crickets, and copepods looking at phylogenetic history, population genetics, and studies of pheromone evolution.

Phylogeography and Gene Flow

We are interested in how we can use genetic data to study  how historical isolation and current patterns of gene flow structure populations. One completed study focused on patterns of genetic differentiation at a fine scale in T. californicus populations and whether patterns of genetic differentiation are influenced by the process of genomic coadaptation (BMC Evol. 2009). Copepods can show high levels of divergenge in mtDNA even over scales as short as 10 km. Other past studies have included the use of mtDNA markers to examine patterns of hybridization in field crickets (Heredity 1997) and pheromone binding protein sequences (PBPs) to look at the evolution of pheromone races in corn borers (Genetics 1999).

Return to Top

Detecting Selection at the Molecular Level

Genes tress and recombination within PBPs between pheromone races suggests that they have recently differentiated and that the PBPs themselves are not crucial for differentiation in these races (Genetics 1999)

In addition to being able to use DNA sequences to help determine population history, inferences can also be made about the evolutionary forces that have acted upon genes from patterns of polymorphism and divergence. A number of statistical techniques have been developed to help tease out patterns caused by natural selection from those resulting from non-selective or neutral evolution.

 

Several of our previous studies have focused on determining whether selection has impacted patterns of variation at candidate genes. Pheromone binding proteins (PBPs) are involved in the ability of males to respond properly to species-specific pheromone signals from females  (an important form of reproductive isolation between many moth species); we looked at whether selection has influenced patterns of sequence variation in these PBPs. Indeed strong selection appears to have shaped patterns of differentiation in a number of cases and this selection is associated with changes in pheromones between species (MBE 2000, JME 2000).

 

In copepods, we have found evidence for selection acting upon proteins that may be involved in generating hybrid breakdown–cytochrome c (MBE 2004) interacts with mtDNA-encoded proteins and shows evidence for positive selection. In other systems strong selection has been found for a number of the genes that have been shown that are know to be involved in generating postzygotic reproductive isolation, but it is unclear at this point if this is a general phenomenon.

Choristoneura fumiferana (spruce budworm) on spruce. In the running for the most destructive insect in North America for the damage it causes in northern coniferous forests. We were looking at the evolution of PBPs in this species and related members of the genus (MBE 2000). Photo by Jerald E. Dewey (license info).

Return to Top

Role of Pheromone Binding Proteins in Pheromone Discrimination and their Evolution

Past work on pheromone systems focused on the role of pheromone binding proteins in male response and how these proteins have evolved between populations. We looked at PBPs in European corn borers (IBMB 1999, Genetics 1999) and Choristoneura (MBE 2000) and a diverse set of moths using similar pheromone components (JME 2000). Although PBPs are not the ultimate source of pheromone discrimination, selection is acting upon them to drive amino acid evolution between species that have also diverged in their pheromone blends

Male moth flying toward pheromone. Photo by K. Haynes.

Return to Top

Research Support

Current grants include National Science Foundation IOS-2029156 to myself and Joel Kingsolver and IOS-1155325. Previous grants include National Science Foundation grants IOS-1555959, IOS-1155325, DEB-0516139, and DEB-0821003. Previous work has also been supported by NSF by grants to R. Burton (DEB-98-15424) and C. Willett (NSF DDIG grant). Any opinions, findings and conclusions or recomendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation (NSF).

Return to Top

 

WILLETTLAB

@UNC

© Copyright 2016. All Rights Reserved. Rometheme

UNC BIOLOGY

Manduca sexta caterpillar with cocoons from Cotesia congregata. Emerging wasp pictured. We are studied how temperuature impacts this parasitism at multiple levels including gene expression and immune function. Photo by Beatriz Moisset (licence info)

Manduca sexta caterpillar on tomato plant (it isn't  just a tobacco hornworm). On-going work in collarboration with Joel Kingsolver is examining the impacts of fluctuating temperatures on thermal physiology in this species. Photo by EricM (licence info)

Much of the current work in our lab is focused on the intertidal copepod Tigriopus californicus. This copepod species inhabits rocky pools in geographically isolated rocky outcrops along the Pacific coast of North America. What is most interesting about this species from an evolutionary perspective is that these copepod populations have become genetically isolated and make an excellent system to study how recently isolated populations can begin to accumulate reproductive isolation and ecological adaptations (particularly now with the availability of our genome sequence) . Populations of this species then are an instructive model for the initial stages of speciation and adaptation.

 

Below we will highlight previous findings from our lab on this topic:

Complex III of the Electron Transport system displays the intimate coordination that occurs between mtDNA-encoded (CYTB) and nuclear encoded proteins (CYC, CYC1, and RISP) within these complexes of the mitochondrion (from Genetics 2006)

Reductions in hybrid fitness associated with genotypes at two different Malic Enzyme loci. ME2 is associated with the most dramatic departures from expected Mendelian inheritance patterns (from JEB 2007).

The approaches outlined above focused on a relatively narrow set of candidate genes but did not look at patterns across the genome as a whole. Recent studies in our lab have taken started to explore the patterns of loci across the genome and how these contribute to the accumulation of postzygotic isolation in crosses of populations of T. californicus.

 

A study we recently published (G3 2016) looked a modest number of markers that spanned all twelve chromosomes of T. californicus in crosses between two southern California populations. Interestingly the largest impacts were centered on chromosome 2 (the chromosome that contains the ME2 locus described above). One other novel feature of this copepod system is the lack of recombination in female copepods. By genotying individuals in different backcrosses to the original populations we can determine the impact of entire chromosomes on hybrid viability. We found that multiple factors on chromosome 2 are likely to contribute to a drop in hybrid viability.

 

An alumnus of the lab, Thiago Lima, has completed a study examining patterns of skew across the entire genome in cross of two populations that shows vastly different patterns with highly skewed frequencies in adults but not larval copepods (Lima and Willett J. Herd. 2018). We have extended this to a larger set of crosses and reciprocal crosses to look at the genome-wide impacts of mito-nuclear  incomptabilities (Evolution 2019).

 

Speaking of genomes, in collaboration with the labs of Ron Burton at SIO, Suzanne Edmands at USC, and Felipe Barreto at OSU, we have finished a paper describing  genomes from a set of eight populations of T. californicus. Web resources for this genome are the genome sequence/annotation for the San Diego population of T. californicus and the annotation in a genome browser. We did a lot of work on the  San Diego assembly and now for this ~200Mb genome >99% of the sequence is in 12 scaffolds corresponding to the 12 chromosomes of this species.

The impact of regions of the genome on hybrid viability are reflected in skews in markers across the 12 chromosomes of T. californicus in crosses of two southern California copepod populations. Blue triangles give the viability for the AB alleles, while squares give the SD allele viability. See G3 (2016) paper for more details

We looked at levels of hybrid male and female sterility in crosses between T. californicus populations. Although low levels of hybrid sterility were found, no difference was found between males and females suggesting there is not faster male evolution of hybrid sterility in this species (Genetica 2008). This has implications for the faster male evolution hypothesis of male hybrid sterility.

Survival of copepod populations to acute heat stress decreases with increased latitude (Northern populations are on the right side). Evolution 2010.

Pathway for the production of the free amino acid proline which allows copepods to adjust to changing salinity conditions. The enzymes P5CS and P5CR were characterized in CBP_B 2002

We are interested in how we can use genetic data to study  how historical isolation and current patterns of gene flow structure populations. One completed study focused on patterns of genetic differentiation at a fine scale in T. californicus populations and whether patterns of genetic differentiation are influenced by the process of genomic coadaptation (BMC Evol. 2009). Copepods can show high levels of divergenge in mtDNA even over scales as short as 10 km. Other past studies have included the use of mtDNA markers to examine patterns of hybridization in field crickets (Heredity 1997) and pheromone binding protein sequences (PBPs) to look at the evolution of pheromone races in corn borers (Genetics 1999).

Genes tress and recombination within PBPs between pheromone races suggests that they have recently differentiated and that the PBPs themselves are not crucial for differentiation in these races (Genetics 1999)

In addition to being able to use DNA sequences to help determine population history, inferences can also be made about the evolutionary forces that have acted upon genes from patterns of polymorphism and divergence. A number of statistical techniques have been developed to help tease out patterns caused by natural selection from those resulting from non-selective or neutral evolution.

 

Several of our previous studies have focused on determining whether selection has impacted patterns of variation at candidate genes. Pheromone binding proteins (PBPs) are involved in the ability of males to respond properly to species-specific pheromone signals from females  (an important form of reproductive isolation between many moth species); we looked at whether selection has influenced patterns of sequence variation in these PBPs. Indeed strong selection appears to have shaped patterns of differentiation in a number of cases and this selection is associated with changes in pheromones between species (MBE 2000, JME 2000).

 

In copepods, we have found evidence for selection acting upon proteins that may be involved in generating hybrid breakdown–cytochrome c (MBE 2004) interacts with mtDNA-encoded proteins and shows evidence for positive selection. In other systems strong selection has been found for a number of the genes that have been shown that are know to be involved in generating postzygotic reproductive isolation, but it is unclear at this point if this is a general phenomenon.

Choristoneura fumiferana (spruce budworm) on spruce. In the running for the most destructive insect in North America for the damage it causes in northern coniferous forests. We were looking at the evolution of PBPs in this species and related members of the genus (MBE 2000). Photo by Jerald E. Dewey (license info).

Manduca sexta caterpillar on tomato plant (it isn't  just a tobacco hornworm). On-going work in collarboration with Joel Kingsolver is examining the impacts of fluctuating temperatures on thermal physiology in this species. Photo by EricM (licence info)

Manduca sexta caterpillar with cocoons from Cotesia congregata. Emerging wasp pictured. We are studied how temperuature impacts this parasitism at multiple levels including gene expression and immune function. Photo by Beatriz Moisset (licence info)