Introduction
All organisms require trace amounts of metals to function and produce new biomass. However, elevated metal concentrations quickly becomes toxic (Lemire et al., 2013; Waldron & Robinson, 2009), with structural effects on polluted ecosystems (Blanck, 2002; Blanck & Dahl, 1996; Pesce et al., 2010; Schmitt et al., 2005). Field studies on plant communities growing on polluted mine tailings have shown that dominant species rely primarily on migration of tolerant individuals from adjacent populations, rather than de-novo mutations (Macnair, 1987). More recently, the critical role of standing genetic variability in evolution of metal tolerance has been shown also for invertebrates (Janssens et al., 2009), yeast (Grangeteau et al., 2017), mycorrhizal fungi (Bazzicalupo et al., 2020), and bacteria (Carlson et al., 2019). With their rapid generation turn over and huge population sizes, phytoplankton might evolve metal tolerance from either standing genetic variation or new mutations. Metal tolerant species of phytoplankton have been reported to dominate polluted aquatic environments (Foster, 1982; Kalinowska & Pawlik-Skowrońska, 2008) but, it is not know to what extent these tolerant species relies on adaptation through selection from standing genetic variation, or novel mutations (Xu et al., 2018; Zhao et al., 2017).
It has been argued that selection from standing genetic variation could enable phytoplankton populations to rapidly adapt to changing environmental conditions (Godhe & Rynearson, 2017; Rengefors et al., 2017). These selection arguments are based on indirect observations such as large population sizes (Sassenhagen et al., 2021), high dispersal rates (Hutchinson, 1961), and high genetic diversity in phytoplankton (Flowers et al., 2015; Kashtan et al., 2014; Osuna-Cruz et al., 2020). However, empirical evidence of phytoplankton phenotypic variability and capacity of adaptation is limited, with knowledge limited to a few key climate change traits and model species (Lohbeck et al., 2012; Schaum et al., 2017; Schaum et al., 2018; Wolf et al., 2019). The evolutionary potential of phytoplankton has mostly been assayed through artificial evolution experiments using a low-diverse genetic background. In such artificial evolution experiments, a single strain or, in some cases, a mixture of a few strains – often a random assembly of strains from different culture collections and geographical origins – is subjected to strong directional selection pressure, and the change in growth rate is recorded as a proxy for evolutionary potential (Collins & Bell, 2004; Lohbeck et al., 2012; Reusch & Boyd, 2013; Schaum et al., 2018). Such experiments reveal how unexposed populations may respond to novel selection pressures, like metal stress, through de-novo mutations and they can also explore how plastic responses contribute to phenotypic changes over time (Schaum & Collins, 2014; Xu et al., 2018). However, they contain insufficient diversity to account for selection from the standing genetic diversity already present in natural populations composed of thousands of unique clones (Sassenhagen et al., 2021).
We have identified the local populations of the diatom Skeletonema marinoi Sarno & Zingone, in and around the Baltic Sea, as a system to study intraspecific diversity and adaptation in phytoplankton. The genetic structure of S. marinoi in the Baltic Sea is primarily linked to the strong salinity gradient (Godhe et al., 2016; Pinseel et al., 2022; Sjöqvist et al., 2015). However, like many other coastal phytoplankton (McQuoid et al., 2002; Montresor et al., 2013),S. marinoi produces benthic resting stages that can anchor it to geographical locations (Sundqvist et al., 2018). Consequently, population structure can develop over distances as short as a few kilometers (Härnström et al., 2011; Sefbom et al., 2018), although the drivers of such differentiations have not been identified. The Baltic Sea also has an extensive and well-documented history of human pollution (HELCOM, 2010; Lehtonen et al., 2017; Reusch et al., 2018), which could be one driving force of local adaptation, as has been shown forDaphnia populations (Kerfoot et al., 1999). At a small (ca. 5 km2) copper mining polluted inlet in the Baltic Sea, we have previously observed that strains of S. marinoi appear overly tolerant to several metals present in the mining ore (Andersson et al., 2020) and that toxic metal stress can affect interspecific competition between diatoms (Andersson et al., 2022). The mine is located on the shoreline and was active from the 17thuntil the early 20th century (Söderhielm & Sundblad, 1996). The sediment in the inlet is polluted with several metals whose concentrations across depth layers correlate negatively with the abundance of S. marinoi micro-fossils (Ning et al., 2018), suggesting that metal pollution has had an adverse effect on this species.
The present study aimed to test if copper tolerance has evolved in the mining-exposed S. marinoi population and to what extent intraspecific trait variation enables the population to adapt rapidly to a toxic environment. Our primary hypotheses were that centuries of mining exposure have caused copper tolerant strains to evolve and that during renewed exposure to toxic stress, the copper-exposed population has an evolutionary advantage over an unexposed reference population. To test these hypotheses, we isolated a large number S. marinoistrains from the mining-exposed inlet (30 strains) and an unexposed reference inlet (28 strains) and quantified their copper tolerance in mono-clonal experiments. We then assembled the populations back together and performed a 42-day long artificial evolution experiment (Fig. S1) where we tracked the strain selection process using amplicon sequencing of a recently developed nuclear locus with exceptional intraspecific diversity (Pinder et al., 2023). This metabarcoding approach allowed us to observe the selection process with unprecedented resolution, and to quantify how selection from intraspecific diversity allow phytoplankton populations to adapt to adverse environmental conditions.