Scientific Background of the Proposal

Two predicted effects of global climate change for the Iberian Peninsula are an increase in the annual mean temperature and a decrease in the annual mean precipitation, but also an increase in the interannual variance of these climate parameters (AEMET, 2017). These changes are likely to affect the water regime, especially in small water bodies with low water inertia.

The water bodies of the Mediterranean region already differ in the length of the flooded periods (hereafter, hydroperiod), ranging from ephemeral to permanent (Camacho et al., 2009). Furthermore, depending on various factors such as catchment size or the characteristics of their inflow (rain, surface or groundwater feeding), they differ in the inter-annual variance of the duration of their hydroperiods. Thus, global climate change, besides increasing water temperatures, will most likely result in (1) a shortening of the typical hydroperiod of each water body and (2) an increase in the variation of the length of the hydroperiod, a variation that will be largely irregular.

Both (1) and (2), in particular because of the occurrence of random bursts of extremely short hydroperiods and the inability of organisms to anticipate the end of the hydroperiod (i.e., the uncertainty involved), may pose a very significant environmental challenge, in particular for species that carry out the active phases of their life cycles entirely in the water (so-called hololimnetic organisms). In response to this challenge, some ‘evolutionary rescue’ (Ferriere & Legendre, 2013) can be expected by the local adaptation of these hololimnetic populations to the new conditions, as well as a demographic rescue by immigration to some local populations from other populations, rather than by adaptation of the former (Bell, 2012). In any case, it is foreseeable that the fate of hololimnetic populations will depend critically on how well they adapt to the new hydroperiod regimes.

The adaptive capacity of populations depends on their population sizes, since the larger the population size, the greater the genetic diversity – the fuel of natural selection – and the smaller the relative effect of genetic drift (Gomulkiewicz & Houle, 2009). Shorter hydroperiods will result in shorter periods of population proliferation, and in a possible mismatch of phenology in the production and exiting of the resistance (dormant) stages with which hololimnetic populations cope with the seasonality of their growing periods (i.e., the part of the hydroperiod in use by the active population) in the aquatic environment. As a consequence, local populations may decrease in size, and eventually, some may be lost. This is aggravated under decreased population connectivity by increasing habitat isolation due to higher frequency of droughts. In addition, local adaptation tends to reduce genetic variation, both at selected loci and in those parts of the genome hitchhiking with them (Via & West, 2008), dropping the chances to react to future selective challenges (Pauls et al. 2013). Overall, evolutionary potential at local and meta-population scales would be diminished, as would the possibility of demographic rescue.

Laboratory evolution experiments have repeatedly shown that aquatic invertebrates display an amazing capacity for evolutionary adaptation to environmental novelty with respect to a wide range of factors (Declerck et al. 2002, 2015; Decaestecker et al. 2007; Geerts et al. 2015; Tarazona et al. 2017). Significantly, this adaptation includes the intensification, in populations of monogont rotifers, of compensatory bet-hedging strategies (García-Roger et al. 2014, 2017) in response to unpredictable fluctuations in the length of the growing season (i.e., Franch Grass et al. 2017; Tarazona et al. 2017). These studies focused on the production and hatching patterns of diapausing eggs of monogonont rotifers. These aquatic microinvertebrates proliferate asexually by parthenogenesis until sexual reproduction is induced, resulting in the production of diapausing eggs. These eggs are the means of survival of the lineages during adverse periods (the dormant stage), which are recurrent in the waters inhabited by rotifers. Obviously, the period in which a water body dries up is a period of maximum adversity. The number of diapausing eggs is therefore indicative of biological fitness (Serra et al. 2019).

Optimisation of diapausing egg production and hatching patterns depends critically on the environmental information that rotifers may have incorporated into their genotypes through the effects of natural selection, due to the conditions that operated in their evolutionary past. The studies mentioned in the previous paragraph are related to bet-hedging strategies between years due to the unpredictability in the hydroperiod length. An additional aspect deserving deeper attention is the bet-hedging within a year associated to the uncertainty at the onset of the growing season, the so-called ‘false starts’ of the growing season. A false start is characterised by the presence of signals that misindicate the arrival of a new growing season. The error means that the hatched eggs give clones that fail to proliferate sufficiently as to produce a new cohort of diapausing eggs. Unlike the notion of a non-false, but also insufficient, hydroperiod, the false start is strongly correlated temporally with a real start, i.e. the false precedes the real one. To cope with variability in the onset of favourable conditions, immediate versus short-term delayed hatching represents a trade-off between the advantages of quick niche filling and the unpredictability of reproductive success under unstable environmental conditions (Vanoverbeke & De Meester, 2009).

Within monogonont rotifers, Brachionus plicatilis is the best-known species and among the most used in evolutionary ecology (Declerck & Papakostas 2017; Serra et al. 2019), with extensive information on genetic structure, demographic features, and local adaptation patterns (Ortells et al., 2000; Campillo et al., 2009, 2011; Montero-Pau et al., 2011; Franch-Gras et al., 2017a, 2019). B. plicatilis sensu stricto (Müller 1786) is a taxonomically well-defined species belonging to the B. plicatilis cryptic species complex (Mills et al., 2017), known to occur in brackish continental waters of different degree of unpredictability (Franch-Gras et al. 2017a). Previous studies on B. plicatlis adaptation have evidenced that the timing of hatching and the hatching fraction are two diapause-related traits influenced by habitat unpredictability. On one hand, hatching fractions of diapausing eggs have shown intermediate values in populations across a gradient of habitat unpredictability, showing some adaptation to unpredictable conditions (Franch-Gras et al. 2017a). On the other, experimental evolution in the lab has achieved divergent adaptive responses for both traits in populations under divergent predictability regimes, with earlier timing of diapausing egg production and lower hatching fractions in populations subjected to unpredictable regimes (Tarazona et al. 2017). Some members of the research team expert in evolutionary ecology are benefiting on a project (UNPREDSCALES: PID2020-114153GB-I00) to carry on experimental evolution under four regimes related to unpredictability between and within seasons. Interestingly, some questions can be added for which there are no answers yet and which could be empirically tested under the frame of this proposal. We will complement that experiment by adding an extra set of replicates to compare the strongest selection regime (unpredictability both between and within seasons) in two starting conditions: multiclonal experimental populations from 9 natural ponds (control) vs. equally-diverse multiclonal populations but originating from a subset of the previous (after excluding those ponds that have a long history of recent droughts). Specifically, because of the above, it is important, to know to what extent adaptive capacity is compromised by the gene pool in a metapopulation. The combined effect of climate change and its consequent reduced genetic variability is expected to affect life history traits such as diapause emergence patterns and leave genomic signatures that can be detected. We will explore such signatures in the laboratory experimental populations using a whole genome approach and correlate these changes in the natural populations of the selected sites along a gradient of environmental unpredictability. As in previous studies, this research takes advantage of samples obtained from well-characterised natural populations, and of an experimental evolutionary approach combined with genomic analysis methods.

Local adaptation of specific populations under selection to the new conditions posed by climate change will influence community assemblages by means of species turn-over, invasions and biotic interactions among the selected specific genotypes. The organization of biological communities of inland waters are also strongly influenced by several climate-related factors, such as the hydroperiod of the water body or its thermal regime (Wetzel, 2001; Williams, 2006). Actually, recent global warming has already impacted aquatic ecosystems and is considered a major threat to their biodiversity (Moss, 2010; Dudgeon, 2020), together with other drivers of biodiversity loss, including human modification, exploitation, pollution and biological invasions (IPBES 2019). Among these, there is an increasing concern on the impact of freshwater invaders. Indeed, the number of exotic species invading continental waters are increasing at alarming rates and may pose major threats to native species (Clavero & García-Berthou, 2006; Gherardi, 2007; Ricciardi & MacIsaac, 2011). This is also true for the Iberian Peninsula, where not only well-known invasive fishes and plants are becoming widely distributed, but also numerous small invertebrates whose potential effects on the ecosystem functions and native biodiversity remain mostly unknown (García-Berthou et al., 2007; Zamora-Marín et al., 2023; Bisquert-Ribes et al., in press). In addition, the combination of climate change and biological invasions may have higher combined impact on biodiversity than their sole contribution; new invasive species may appear, or the already present invaders may change their distribution, with unforeseen impacts on native communities (Hellman et al., 2008). It is therefore imperative for freshwater conservation planning to know the potential responses of aquatic communities to global warming, for instance using ecological niche models to predict geographical expansions or contractions, and especially considering the interactions of climate change with biological invasions. One of such interactions, of special concern to humans is the expansion of emerging pathogens in the temperate zone (WHO, 2023a), some of which may originate in warmer areas such as the tropics. The emergence and re-emergence of arboviral diseases such as Flaviviruses are favoured by their great genetic plasticity, their diversity and ability to adapt to different scenarios and the involvement of various species in their transmission cycles, causing a great impact on humans and animals. Mosquitoes are responsible for the largest number of annual deaths caused by diseases worldwide, as they are the transmitters of viruses such as dengue or Japanese encephalitis virus (WHO, 2023a). Although the ecology and distribution of mosquito vectors of diseases with high impact on human and animal health has been extensively studied, the study of vectors involved in the transmission of pathogens with lower health impact is more limited, especially in areas of new geographic expansion. West Nile fever virus (WNFV) is an expansive, emergent, and virulent model flavivirus, considered to be the most widely distributed worldwide (Chancey et al., 2015). Given its feeding plasticity (ornithophilic and mammophilic), this virus is transmitted by mosquitoes of the genus Culex (Cx.) spp. especially those belonging to the Cx. pipiens s.l. complex (Becker et al., 2010), among others. Since 2010, the occurrence of outbreaks in equine farms in the Iberian Peninsula, have been recurrent. The largest outbreak in humans took place in the southwestern Iberian Peninsula in 2020, when77 cases were reported, 8 of which resulted in death. Interspecific susceptibility to WNV is diverse. Corvids and raptors have been described as particularly susceptible birds with a high capacity for replication and development of viremias with high titers in blood (Pérez-Ramírez et al., 2014). However, other bird species capable of developing chronic viremia such as finches or sparrows (Wheeler et al., 2012a, b) or other species not yet studied, could favour the persistence of the WNV life cycle even in times of vector hibernation (Gamino & Höfle, 2013), and could be serving as another pathway by which the disease is maintained in the absence of mosquitoes. Besides pathogens of human interest, there is consensus of other pathogens affecting biodiversity; of special concern is the spread of chytridiomycosis, caused by Batrachochytrium dendrobatidis (Bd), causing significant declines in amphibian populations worldwide and even extinction of some populations (Berger et al., 1998; Bosch et al., 2001; Lips et al., 2006; Rachowicz et al., 2006; Skerratt et al., 2007). Bd has infected more than 500 species on all continents except Antarctica (Global Bd Mapping Group, 2018). In Europe, Spain is the most affected country. Host diversity can affect Bd dynamics, with dilution of effects in diverse communities (Searle et al., 2011), but with possible amplification effects in the presence of invasive species (Miaud et al., 2016). Resistant species are refractory to infection or rapidly eliminate it, having limited potential to maintain a disease in the system (Brannelly et al., 2018). It is therefore important to keep track of the spread of this plague and if there are reservoir populations in water bodies of interest to conservation of biodiversity.

At inland water ecosystems, such as wetlands, favourable conditions exist for biogeochemical transformations, therefore being closely linked to the global carbon cycle (Battin et al., 2009). Due to their humid conditions and the predominance of anoxic conditions in their sediments, which slow down the degradation of organic matter, wetlands are ideal areas to fix and store carbon from the atmosphere (Mitsch et al., 2013). They occupy fifth place in the global carbon sequestration of terrestrial ecosystems, behind boreal and temperate forests, savannahs and temperate grasslands, despite the small area they occupy (Mitra et al, 2005). In general terms, it is estimated that they can accumulate almost 7% of the world’s soil carbon (Bolin and Sukumar, 2000). In wetlands, the carbon balance is closely linked to the production and emission of greenhouse gases (GHG), and particularly methane, with estimates that they emit between 24% and 40% of global atmospheric methane (Whalen, 2005), being the most important natural source of this GHG with high greenhouse effect (radiative forcing) potential (GWP). The balance between photosynthetic carbon fixation and its emissions of carbon dioxide (CO2) and methane (CH4) allows estimating the contribution of a wetland to the atmospheric pool of GHG. All of this is susceptible to being modified if temperature patterns and hydroperiods change as a consequence of climate change itself (Camacho et al., 2017).

Microorganisms in wetlands are key players for biogeochemical processes of the different element’s cycles (Barton & Northup, 2011). Classically, microbes and their metabolic capacities were characterized using traditional culture-dependent approaches. However, most microbes cannot be easily cultured due to their complex habitats, such as unreproducible environmental, temporal, physical, biochemical, and genetic complexities (Wu et al., 2019). Classical methods are limited to identifying a narrow spectrum of microorganisms, leaving many of them, and their possible role in the ecosystem, uncharacterized. Furthermore, for specific issues, such as human and environmental health, the identification of potential pathogens is also of concern.

The components of a biological community may be identified by molecular methods, which enables the retrieval of DNA sequences from a mixture of a complex community in a culture-independent way (Choubey et al., 2021). Many recent studies have adopted amplicon 16S Illumina MiSeq sequencing to explore the diversity of microbial community in different types of ecosystems. This approach allows a deeper understanding of the ecology and dynamics of bacterial communities in nature, in order to explain how they interact and shape the biogeochemical functioning of an ecosystem. Such molecular approaches to study bacterial diversity that do not require microbiological culture techniques have shown that most of the historically suggested model prokaryotic organisms are of lesser importance “in situ” than one might expect, whilst other microorganisms (Degnan & Ochman, 2012), often not yet cultured, are responsible for most of the key processes in nature. High-throughput sequencing targeted at regions conserved, the so-called “metabarcoding”, like 16S rRNA in prokaryotic assemblages, or other taxonomically conservative genes for eukaryotic organism (rcbl, COI, etc.) extracted from the so-called environmental-DNA (e-DNA), is currently considered as the most reliable and cost-effective method of species composition and diversity analysis of environmental samples from aquatic environments (Fukuda et al., 2016).

However, an advanced step to go deeper into the composition and, especially, the metabolic potential of a microbial community, is the use of metagenomics, this, is, the study of the structure and function of entire nucleotide sequences isolated and analysed from all the organisms (typically microbes) in a bulk sample containing a specific community of microorganisms (https://www.genome.gov/genetics-glossary/Metagenomics), such as those residing on aquatic ecosystems. The most relevant taxa within a microbial community, and their metabolic capacities, can be determined using Metagenome Assembled Genomes (MAGs), unveiling the metabolic potential of all taxa present and helping in the determination of their biogeochemical key roles in the ecosystem, thus providing a model for the adaptations to specific conditions (Cabello Yeves et al., 2023).

A bacterial pathogen is any bacteria that is capable of causing a disease to a living being. Taking this definition into account, the list of pathogens increases as our knowledge increases, and even the concept of indicator and pathogen is sometimes confusing, since some indicator organisms are capable of causing disease. Several databases collect and compile all potentially pathogenic organisms, such as PATRIC or RISC group Database. As an example, restricting to the human pathogens, the total number of bacterial genera with potentially pathogenic species today amounts to about 210, which implies that classical individual detection methods are limited for the identification and quantification of pathogens. In recent years, the characterization of microbial communities through molecular approaches has been gaining acceptance compared to classical culture methods, with the next generation sequencing (NGS) technique emerging as a great tool for microbiome environmental research (Sogin et al. 2006; Jones et al. 2009; Mueller-Spitz et al. 2009; Hassan et al., 2018). NGS technology, based on microbial 16S rRNA sequences, allows the characterization of microbial communities (Salipante et al., 2014; Sala-Comorera et al., 2020) and has enough sensitivity to detect minority microorganisms, being very useful for the detection of pathogenic microorganisms.

Apart from the previously described approaches, based on DNA analysis, the study of gene expression of prokaryotes in the environment can yield interesting results at ecological level, since changes in expression are strongly influenced by environmental factors (Gionchetta et al. al. 2019). Therefore, different environmental conditions, such as differences in the temperature patterns and in the hydroperiod can be reflected in the gene expression patterns of microorganisms. Gene expression of prokaryotic communities can be studied by RT-qPCR. For this, the total RNA is extracted from the entire community and retro-transcribed (RT) to copy DNA (cDNA), which is the DNA sequence complementary to the extracted RNA (Porter and Hajibabaei 2018). From this cDNA, quantitative PCR (qPCR) of marker genes of certain metabolisms or microbial groups can be performed, so that the degree of participation of the different microbial taxa in specific metabolisms can be quantified (He et al. 2018), as well as the effect of environmental differences or changes on the expression of these marker genes (Vigneron et al.2019).