Will shellfish persist in a changing ocean?

Summary of this post: in a podcast!

In the course that I am taking with the Alan Alda Center for Communicating Science at Stony Brook University, we are learning about making podcasts. Here is my first attempt at one- I have no experience with sound editing, so please excuse sound quality issues, or lack of fancy background music. The goal of this exercise was to distill my message into a short summary. Let me know your thoughts in the comments, and stay tuned for improved versions!

Introduction (skip to get to the literature review!)

The following post will discuss some of the foundational research for what I plan to study in the final chapter of my dissertation. While I can’t cover all of the really cool studies that I’ve been reading about to prepare for this next phase, I will present some highlights and links to more information. If you think that I’ve missed a critical study, please tell me about it in the comments below! My literature search is ongoing and I welcome feedback and questions.

My research so far has been on the copepod Tigriopus californicus (see last week’s post to learn about them!) in the San Juan Islands. Now I want to explore similar questions in a different system: populations of the blue mussel, Mytilus edulis, throughout the Long Island Sound.

Over the past two years, when I was not conducting my own experiments, I was working in the Padilla Lab and at the NOAA Northeast Fisheries Science Center in Milford, CT as part of a Sea Grant-funded project (NY Sea Grant press release here) to test for acclimation and adaptation to ocean acidification in populations of mussels spanning a gradient of water quality conditions in the Long Island Sound. The project is ongoing: we are currently conducting experiments on F1 adults and F2 larvae that have been reared and maintained in a range of carbonate chemistry conditions (including reciprocal transplants of animals among treatments to separate phenotypic plasticity and genetic adaptation). Across generations, we are quantifying life history, shell growth and structure, and physiological responses, while also measuring carbonate chemistry in the field at our mussel collection sites.

I plan to build on this project by exploring the molecular underpinnings of interesting differences in population performance that we have observed, both within and between generations. I recently completed an NSF Graduate Research Internship (GRIP) at the NOAA Northwest Fisheries Science Center in Seattle, WA with Dr. Krista Nichols, where I learned to conduct population genomic studies on non-model organisms using RAD-seq and Pool-seq methods (to be discussed in future posts!). I will use these methods to analyze tissue, embryo, and larvae samples that we have preserved throughout this project to characterize genetic differences among populations and generations and to test for signatures of selection.

Literature Review: Evolutionary responses of shellfish to ocean acidification (OA)

If you need background information about OA, here are some resources:

NOAA PMEL

NOAA Ocean & Coasts Education Resources

Shellfish and ocean acidification: What is going on?

In 2007, the Whiskey Creek Shellfish Hatchery in Netarts Bay, Oregon had an oyster larvae mortality event unlike anything they had ever seen before. After some initial investigation into the pathogenic bacterium Vibrio tubiashii, hatchery managers finally determined that the actual cause was a change in the chemistry of the seawater. Coastal waters of the Pacific Northwest periodically go through lower pH events due to upwelling of deep water that is naturally more carbon dioxide-rich and undersaturated with respect to aragonite. In recent years however, these changes in seawater chemistry have been more intense, leading shellfish growers and scientists to take action.

After these larval supply shortages, the industry, local governments, and researchers created strategies for dealing with the effects of ocean acidification (OA), including rigorous water quality monitoring, buffering seawater in hatcheries, changing the timing of spawning and larval rearing to avoid upwelling events, and breeding of OA-tolerant lines (often built off of already existing breeding programs to increase output and disease resistance) [1]. While these efforts have been very successful- they facilitated rapid industry growth despite the challenges of OA- there is still work to be done. We don’t know if these immediate solutions will continue to protect the industry as OA intensifies. We know even less about natural populations, which are of critical economic, ecological, and cultural importance to coastal communities around the world [2].

In the past decade, there has been an explosion of studies on biological responses to OA and it is clear that shellfish survival and growth can be negatively impacted, especially during early life stages [2]. There is now great interest in predicting whether populations of bivalve molluscs will be able to acclimate and/or adapt to rapid changes in seawater chemistry.

How do we study evolution in a changing ocean?

The review Sunday et al. 2014 [3] made a call for research to understand the evolutionary responses of populations to OA, as most studies prior to this publication focused on short-term responses of individual organisms. They outlined different methods for studying evolution in a changing ocean, which I will address below by discussing key studies that have used each approach.

Measuring standing genetic variation

For adaptation to occur, there must be genetic variation in traits involved with responses to OA within a population [3]. In Sunday et al. 2011, researchers measured the potential for evolution to OA in the mussel Mytilus trossulus using a quantitative genetics approach. They conducted crosses to form 40 families, exposed the larvae from each family to ambient and OA seawater treatments, and quantified larval growth, which is strongly related to fitness. Using crosses allowed them to compare relatives, so that the variance in the larval growth phenotype could be partitioned into genetic and environmental components, which was then used to calculate heritability and estimate the potential for evolution in downstream simulations. The researchers found some additive genetic variance for larval size in the ambient seawater treatment, but none in the OA treatment, which suggested that the study population had limited potential for adaptation to OA. [4]

While quantitative genetic studies are very useful because they can be conducted on a range of organisms with no need for prior genetic knowledge, multigenerational studies are needed for a more complete understanding of evolutionary potential. Studying populations exposed to OA over ontogeny and across generations allows for the direct measurement of population dynamics and exploration of evolutionary mechanisms underlying resilience [3].

Experimental evolution- multigenerational studies

In OA research, most multigenerational studies to date have been conducted on single-celled organisms with quick generation times, such as coccolithophores, diatoms, and dinoflagellates. In recent years however, more studies on multicellular and longer-lived organisms are being done, including cross-generational research on bivalves [3].

Pioneering work on transgenerational responses to OA in the Sydney rock oyster, Saccostrea glomerata, has been conducted by Parker, Ross, O’Connor, and colleagues. Before this work, most studies measured responses of organisms exposed to acute OA stress, over short periods, often within a single life stage. These scientists were interested in carry-over effects, or whether the experience of parents influences the performance of their offspring through transgenerational plasticity, which can confer resilience under long-term stressful conditions.

The researchers took adult oysters that were conditioning for reproduction (forming their gonads) and exposed them to ambient and OA seawater treatments. These oysters were spawned, and gametes were divided into the two different treatments for fertilization, in a factorial design including all combinations of adult and offspring pH exposure. Larvae were raised in these treatments, and their survival, development, and shell length were quantified. OA reduced growth, development, and survival in the larvae, however, there were also positive carry-over effects from the parents. Larvae that were spawned from adults that had been exposed to OA during their reproductive conditioning had faster growth and development than larvae that were spawned from adults exposed to ambient conditions [5].

To take this study even further, the researchers asked if the positive carry-over effects they found could persist into adulthood of the F1 offspring, or even to F2 larvae. They focused on two lines of oysters from the first phase of the research: a control line (parents and F1 offspring were exposed to ambient conditions) and a transgen line (parents and F1 offspring were exposed to OA) and raised the F1 larvae until they were old enough to settle. The settled larvae were then transferred into ambient conditions in the field for 14 months, until they reached reproductive maturity. The adult F1 oysters were brought back into the lab to conduct the previously described experiment all over again, this time on F1 adults and their F2 larvae and spat (juveniles) [6]. See their figure below for details about the experimental design:

The positive carry-over effects did persist into adulthood for the F1 generation (F1 adults in the transgen line had better physiological performance in OA than adults from the control line),  as well as into the F2 generation. Transgen F2 larvae had faster development and growth and fewer shell abnormalities in OA compared to control F2 larvae [6]. This oyster may be more resilient to future OA than we previously expected,  based on studies that measured responses only within a single generation.

An exciting new study, Thomsen et al. 2017, was the first to conduct an OA selection experiment on a bivalve across three generations under continuous exposure to OA treatments. They collected blue mussels, Mytilus edulis, from a population in the Baltic Sea and set up family lines by crossing specific F0 dams and sires (moms and dads) and tested different groups of F1 siblings separately in three pH treatments, to identify tolerant and sensitive lines. F1 juveniles were maintained in their respective seawater treatments until maturity. F1 adults were spawned (in two separate years) and crosses were made within and between tolerant and sensitive lines (see figure below for details), using animals that were reared in each of the pH treatments. F2 larvae from each cross (and historic pH exposure) were divided into the three treatments so that there was a full factorial design with different combinations of historic and current pH exposure [7].

For the F1 generation, there was large variability in larval survival among families, but low survival in the most extreme OA treatment across the board. In the F2 generation, under the most extreme OA treatment, larvae with parents that were raised under OA were larger than larvae whose parents were raised in ambient conditions. However, the opposite trend was seen for F2 larval survival. Larvae from mothers that were raised in ambient conditions had higher survival than larvae from mothers that were raised under OA. While the selection of OA-tolerant F1 mussels led to higher calcification in OA exposed F2 larvae early in development, the ultimate high mortality of these larvae was not improved by selection, or through transgenerational acclimation [7].

From the studies described above, it is clear that there is high variation in responses to OA among populations and species. Therefore, we cannot generalize findings from few studies to all shellfish. More long-term research will be critical for accurate predictions of shellfish susceptibility to OA.

Infer past adaptation by comparing populations 

An increasingly common method of inferring capacity for adaptation to OA is to compare different populations or family lines that have historically been exposed to different environments. Using a common garden approach, scientists can quantify responses to OA among different populations and infer local adaptation, or past selection, if populations have higher performance in treatments that are similar to their historic, native environment. Variation in performance among populations suggests evolutionary potential within a species, and understanding how evolution has occurred in the past can help inform how organisms may respond in the future.

In the Saccostrea glomerata studies by Parker and colleagues, they also compared a wild population to a family line that was bred for fast growth and disease-resistance in shellfish farms [8]. In Thomsen et al. 2017, they compared two populations of mussels: one from the Kiel Fjord of the Baltic Sea, which is characterized by highly variable and often low pH conditions, and the other from the North Sea, which has higher and more stable pH, due to higher salinity and buffering capacity [7]. Wright et al. 2014 quantified growth, shell strength, standard metabolic rate, and predation by a whelk under OA in 8 genetically distinct family lines of the Pacific oyster, Crassostrea gigas [9]. All of these studies found interesting differences in performance under OA conditions among populations. In S. glomerata, the family line that was bred for robustness in an aquaculture setting was also more resilient to OA [8], and in Mytilus, the population from the harsh Kiel Fjord was more resilient to OA than the population from the benign North Sea, providing evidence for local adaptation and potential for selection in response to future OA [7].

Harnessing the power of molecular tools

With recent advances in next-generation sequencing technology, it is becoming feasible for more labs to study the molecular biology of non-model organisms. Understanding the population dynamics of bivalves and identifying the molecular mechanisms underlying resilience to OA will be critical for the selective breeding of shellfish for aquaculture and predicting the vulnerability of natural populations.

Transcriptomics can help us understand changes in gene expression that occur when organisms are exposed to different seawater chemistries, and reveal the mechanisms underlying phenotypic plasticity [3]. Using population genomics, we can study the genetic diversity within and among populations, and observe changes in allele frequencies in populations undergoing selection overtime. Kingston et al. 2018 used a genome-wide association approach (GWAS) to link genetic variation with variance in calcification rates in response to combined OA, warming, and food limitation across 7 populations of Mytilus spanning a hybrid zone in the Gulf of Maine. This approach allowed them to identify genomic regions that are involved with phenotypic variation in calcification (a trait that is strongly related to fitness in bivalves) and thus could be targets of selection under OA and climate change [10]. Identifying candidate loci for OA resilience could become important for genetic engineering, which may be necessary to prevent the extinction of shellfish species that ultimately cannot adapt to OA.

For more information:

The Roberts Lab at the University of Washington and the Allam Lab at Stony Brook University are two of multiple labs around the world that are working to understand the molecular mechanisms of resilience to OA in bivalve molluscs.

References

[1]  Barton A., G.G. Waldbusser, R.A. Feely, S.B. Weisberg, J.A. Newton, B. Hales, S. Cudd, B. Eudeline, C.J. Langdon, I. Jefferds, T. King, A. Suhrbier, & K. McLaughlin 2015. Impacts of coastal acidification on the Pacific Northwest shellfish industry and adaptation strategies implemented in response. Oceanography. 28(2):146–159.

[3]  Sunday J.M., P. Calosi, S. Dupont, P.L. Munday, J.H. Stillman, & T.B.H. Reusch 2014. Evolution in an acidifying ocean. Trends in Ecology and Evolution. 29:117–125.

[4]  Sunday J.M., R.N. Crim, C.D.G. Harley, & M.W. Hart 2011. Quantifying Rates of Evolutionary Adaptation in Response to Ocean Acidification. PLoS ONE. 6(8): e22881. doi:10.1371/journal.pone.0022881

[6]  Parker L.M., W.A. O’Connor, D.A. Raftos, H.O. Pörtner, & P.M. Ross 2015. Persistence of positive carryover effects in the oyster, Saccostrea glomerata, following transgenerational exposure to ocean acidification. PLoS ONE. 10(7): e0132276. doi:10.1371/journal.pone.0132276

[7]  Thomsen J., L.S. Stapp, K. Haynert, H. Schade, M. Danelli, G. Lannig, K.M. Wegner, & F. Melzner 2017. Naturally acidified habitat selects for ocean acidification–tolerant mussels. Science Advances. 3, e1602411.

[8]  Parker L.M., P.M. Ross, & W.A. O’Connor 2011. Populations of the Sydney rock oyster, Saccostrea glomerata, vary in response to ocean acidification. Marine Biology. 158:689–697.

[9]  Wright J.M., L.M. Parker, W.A. O’Connor, M. Williams, P. Kube, & P.M. Ross 2014. Populations of Pacific oysters Crassostrea gigas respond variably to elevated CO₂ and predation by Morula marginalba. Biological Bulletin. 226:269-281.

[10]  Kingston S.E., P. Martino, M. Melendy, F.A. Reed, & D.B. Carlon 2018. Linking genotype to phenotype in a changing ocean: inferring the genomic architecture of a blue mussel stress response with genome-wide association. Journal of Evolutionary Biology. 31:346–361.