i
S
AMMANFATTNING
Framtidens klimatförändringar förväntas leda till varmare yttemperaturer i Östersjön samt minskad salinitet som en följd av ökad nederbörd och avrinning. Förändringarna kan ha en allvarlig påverkan på dynamiken och funktionen hos brackvattenorganismer, särskilt växtplankton. Växtplankton producerar en betydande mängd organiskt material till andra trofiska nivåer, och vissa arter kan vara giftiga. Hur de kommer att påverkas i framtida klimatförhållanden är därför av stor vikt för hälsan hos människor och vattenekosystem. Syftet med denna avhandling var att undersöka huruvida klimatinducerade förändringar, såsom minskad salthalt, påverkar växtplanktons dynamik, fysiologi samt kemiska sammansättning i Östersjön.
För växtplankton i Östersjön sker en succession med en vårblomning bestående av kiselalger samt dinoflagellater och en sommarblomning som domineras av kolonibildande/ trådformiga cyanobakterier. Den rådande uppfattningen är att varmare klimatförhållanden kommer att gynna kolonibildande och trådformiga cyanobakterier. Denna avhandling visar att vårblomningens biomassa var lägre under år med mildare vintrar jämfört med kalla vintrar. Resultaten tyder på att den minskade årliga produktionen av organiskt material och därmed flöde till högre trofinivåer, i samband med en utebliven vårblomning inte kompenseras av sommarens cyanobakterieblomning. Genom hög provtagningsfrekvens, observerades en stark relation mellan dynamiken av pikocyanobakterier och trådformiga cyanobakterier i denna avhandling. Hög genetisk mångfald av cyanobakteriesamhället samt inom specifika arter och hög nischdifferentiering påvisades. De mest betydelsefulla faktorerna som påverkade artsammansättningen av cyanobakterier var hög temperatur samt låg salinitet under sommarmånaderna. Dessa förhållanden kan främja opportunistiska trådformiga cyanobakterier såsom Nodularia spumigena. Den här arten producerar olika bioaktiva föreningar, inklusive icke-ribosomala peptider som t.ex. nodularin som är ett hepatotoxin. Denna avhandling visar att subpopulationer av N. spumigena har utvecklat olika fysiologiska strategier, inklusive förändringar av den kemiska sammansättningen, för att klara av salthaltstress. Denna höga fenotypiska plasticitet säkerställer överlevnad i framtida klimatförhållanden. Dock orsakade salthaltstress en förkortning av filamenten hos vissa subpopulationer. Detta tyder på att en minskad salinitet i Östersjön kan främja ökad konsumtion av trådformiga cyanobakterier i framtiden vilket skulle medföra förändringar av kolflöden i ekosystemet. I denna avhandling, grupperades kemotyper och genotyper av N. spumigena i två huvudsakliga grupper som var oberoende av det geografiska ursprunget. Således kan den kemiska sammansättningen hos olika subpopulationer av N. spumigena användas för att studera mångfalden hos dessa närbesläktade organismer.
ii
Sammanfattningsvis har denna avhandling bidragit till ökad kunskap om hur växtplankton påverkas av kort- och långsiktiga miljöförändringar, vilket är av stor betydelse för att försöka förutse vilka konsekvenser framtida klimatförhållanden kommer att ha på Östersjön.
iii
R
ESUMEN
Los escenarios climáticos futuros en el Mar Báltico predicen un aumento de la temperatura del mar, además del incremento en la precipitación y la escorrentía fluvial lo que conllevará a una disminución de la salinidad. Estos cambios pueden afectar gravemente la dinámica y el funcionamiento de las comunidades salobres, en especial al fitoplancton. El fitoplancton es la mayor fuente de materia orgánica para niveles tróficos superiores, y algunas de sus especies pueden ser tóxicas. Conocer la respuesta de las especies tóxicas a condiciones climáticas futuras es de gran interés tanto para la salud de los seres humanos como para los ecosistemas acuáticos. El objetivo de esta tesis es evaluar el potencial de las alteraciones inducidas por el cambio climático, tales como la disminución de la salinidad, que pueden afectar la dinámica del fitoplancton, su fisiología y sus perfiles químicos en el Mar Báltico.
Los patrones de sucesión del fitoplancton en el Báltico Central consisten en una proliferación durante la primavera, cuando acontecen las diatomeas y los dinoflagelados, y otra en el verano, dominada por cianobacterias filamentosas/coloniales. El ascenso de la temperatura provocado por el cambio climático acentuará las proliferaciones de cianobacterias filamentosas/coloniales. En esta tesis se demuestra que la biomasa del fitoplancton durante las proliferaciones de primavera fue menor en años con inviernos más suaves comparado con inviernos fríos. Esto sugiere que en términos de exportación anual de carbono a niveles tróficos superiores, la pérdida de biomasa en la proliferación de primavera es improbable que se vea compensada por cianobacterias en verano. Además, la alta frecuencia de muestreo del fitoplancton realizada en esta tesis, reveló una fuerte relación entre la dinámica de las picocianobacterias y de las cianobacterias filamentosas. Por otro lado, se encontró una gran diversidad genética en poblaciones de cianobacterias que ocuparon distintos nichos ecológicos. A nivel de comunidad, la alta temperatura y baja salinidad fueron los principales factores que determinaron la composición de la comunidad de cianobacterias en verano. Estas condiciones climáticas pueden promover el predominio de cianobacterias filamentosas oportunistas tales como Nodularia spumigena. Esta especie produce una gran variedad de compuestos bioactivos, incluyendo péptidos no ribosomales como la hepatotoxina nodularina. En este trabajo se demuestra que diferentes subpoblaciones de N. spumigena desarrollaron diversas estrategias fisiológicas, incluyendo perfiles químicos, para hacer frente al estrés por salinidad. La destacada plasticidad fenotípica asegura un alto índice de supervivencia en condiciones climáticas futuras. Como compensación, bajo estrés por salinidad algunas subpoblaciones presentaron filamentos más cortos. Esto indica que la reducción en salinidad en el Mar Báltico podría estimular la herbivoría en cianobacterias filamentosas y modificar los flujos de carbono en el ecosistema. En esta tesis los quemotipos y los genotipos agruparon N. spumigena del Báltico en dos grupos principales sin influencia del origen
iv
geográfico. Por lo tanto, el perfil químico puede ser utilizado para explorar la diversidad intraespecífica en subpoblaciones de N. spumigena genéticamente relacionadas.
En general, esta tesis ha ampliado significativamente el conocimiento sobre las respuestas de comunidad y de población fitoplanctónica a cambios en las condiciones ambientales a corto y largo plazo, importantes para predecir los impactos de las futuras condiciones climáticas en el Mar Báltico.
v
A mi Familia.
A mi tío Rafa,
que siempre le fascinó
lo que hacía y creyó siempre en mí.
vii
“The sea, once it casts its spell, holds one in its net of wonder
forever”
– Jacques Cousteau “What we know is a drop, what we don’t know is an ocean”ix
L
IST OF
P
APERS
This thesis is based on the following papers, referred to in the text by their roman numbers.
I. Legrand, C., Fridolfsson, E., Bertos-Fortis, M., Lindehoff, E., Larsson, P., Pinhassi, J., and Andersson, A. (2015). Interannual variability of phyto-bacterioplankton biomass and production in coastal and offshore waters of the Baltic Sea. Ambio, 44, 427-438.
II. Bertos-Fortis, M., Farnelid, H. M., Lindh, M. V., Casini, M., Andersson, A., Pinhassi, J., and Legrand, C. (2016). Unscrambling cyanobacteria community dynamics related to environmental factors. Front. Microbiol. 7, 625.
III. Bertos-Fortis, M., Eriksson, F., Mazur-Marzec, H., and Legrand, C. Salinity-shock effects on growth, toxin levels and gene expression in Nodularia spumigena from the Baltic Sea. Manuscript
IV. Bertos-Fortis, M., Klotz, F., Mazur-Marzec, H., and Legrand, C. Phenotypic plasticity in brackish water cyanobacteria: survival at any cost under salinity changes. Submitted
V. Mazur-Marzec, H., Bertos-Fortis, M., Toruńska-Sitarz, A., Fidor, A., and Legrand, C. Chemical and genetic diversity of Nodularia spumigena from the Baltic Sea. In review, Marine Drugs
Reprint of published papers were made with the permission of the publishers. Supplementary material, tables and figures, for published papers can be found online.
x
Additional published work performed during the Ph.D period
but not included in the thesis
Godhe, A., Sjöqvist, C., Sildever, S., Sefbom, J., Harðardóttir, S., Bertos-Fortis, M., Bunse, C., Gross, S., Johansson, E., Jonsson, P. R., Khandan, S., Legrand, C., Lips, I., Lundholm, N., Rengefors, K. E., Sassenhagen, I., Suikkanen, S., Sundqvist, L., and Kremp, A. (2016). Physical barriers and environmental gradients cause spatial and temporal genetic differentiation of an extensive algal bloom. J. Biogeogr. 43, 1130-1142.
Bunse, C., Bertos-Fortis, M., Sassenhagen, I., Sildever, S., Sjöqvist, C., Godhe, A., Gross, S., Kremp, A., Lips, I., Lundholm, N., Rengefors, K., Sefbom, J., Pinhassi, J., and Legrand, C. (2016). Spatio-temporal interdependence of bacteria and phytoplankton during a Baltic Sea spring bloom. Front. Microbiol. 7, 517.
Calbet, A., Bertos, M., Fuentes-Grünewald, C., Alacid, E., Figueroa, R., Renom, B. and Garcés, E. (2011). Intraspecific variability in Karlodinium veneficum: Growth rates, mixotrophy, and lipid composition. Harmful Algae, 10, 654–667.
T
ABLE OF
C
ONTENTS
I
NTRODUCTION... 1
The Baltic Sea: a changing environment ... 1
The importance of phytoplankton on aquatic ecosystems ... 3
Phytoplankton in a changing environment ... 4
Phytoplankton blooms in the Baltic Sea ... 5
Nodularia spumigena: morphology, distribution and toxicity ... 6
A
IMS... 9
R
ESULTS ANDD
ISCUSSION... 10
Phytoplankton dynamics and successional patterns ... 10
Abiotic and biotic factors related to phytoplankton community dynamics ... 12
Phenotypic plasticity of N. spumigena to changing salinity: life strategies to ensure survival at any cost ... 15
Non-ribosomal peptides produced by N. spumigena: synthesis and regulatory mechanisms ... 16
C
ONCLUSIONS ANDF
UTUREP
ERSPECTIVES... 21
A
CKNOWLEDGEMENTS... 23
R
EFERENCES... 26
1
I
NTRODUCTION
The Baltic Sea: a changing environment
The Baltic Sea is a large, semi-enclosed sea with restricted water exchange and circulation through the narrow and shallow connection to the North Sea (Fig. 1). Because of this topography, water masses in the Baltic Sea have a long residence time (approx. 30 years). Characterized for being one of the largest brackish water bodies in the world, salinity decreases from 25 in the Kattegat to 6-8 in the Baltic Proper, and to 5-6 in the Bothnian Sea, down to 2-3 in the northern part of the Bothnian Bay (Matthäus, 2006). As many organisms are unable to cope with these salinity conditions, the Baltic Sea is a low-biodiversity ecosystem, exceptionally vulnerable to changes (Philippart et al., 2011). Today, more than 85 million people inhabit the drainage area of the Baltic Sea. In the last 150 years, the constant increase of human activities in open and coastal areas, especially industrialized agriculture, has promoted eutrophication in the Baltic Sea (Wulff et al., 2007). These high eutrophication levels have led to the expansion of hypoxic zones in the Baltic Sea (Zillén and Conley, 2010), compromising the endurance of many aquatic organisms.
Figure 1. Location map of the Baltic Sea.
2 Despite 40 years of effort to reduce eutrophication, 80% of the Baltic Proper has been classified as “bad ecological status” (HELCOM, 2009). Other environmental problems threatening the aquatic ecosystems in the Baltic region are related to the release of hazardous substances (pollutants, toxins, pharmaceuticals, sludge, oil spills), marine litter, overfishing, and the expansion of invasive species, e.g. round goby (Neogobius melanostomus) (HELCOM).
Climate change is an additional concern regarding the status of the Baltic Sea. Semi-enclosed seas appear to be warming up faster than other oceans (HELCOM, 2013). Seasonal patterns are likely to change, with a reduction of the ice season in northern areas during the next decades (HELCOM, 2013; IPCC, 2013). The increase in sea surface temperature is predicted to enhance stratification, increase evaporation and cloud formation. In this scenario, precipitation will be more frequent, giving rise to higher fresh-water inflow (20% more river run-off). As a result, salinity is estimated to decrease in the order of 2–2.5 units (Fig. 2; Meier et al., 2006; Neumann et al., 2012) and nutrient loading may enhance eutrophication (Meier et al., 2014). Due to the increase of human activities, additional release of carbon dioxide (CO2) to the
Figure 2. Map of the Baltic Sea (a) showing the present surface salinity gradient (BY15 is
the monitoring station at the Gotland Deep, SMHI monitoring programme) and (b) projection of salinity changes for the next century (Meier et al., 2006, reprinted with permission of John Wiley & Sons, Inc.).
RCM driven by different GCMs (and forced by one emission scenario, A2 or B2) are larger than the salinity differences between scenario simulations using different RCMs driven by the same GCM. The uncertainty caused by the emission scenario is also smaller than the uncertainty caused by GCM differences. These results are in line with the findings of De´que´ et al. [2006] for seasonal mean temperature and precipitation.
[14] In the projections with lateral boundary data from
HadAM3H, HadAM3P, and ARPEGE salinity changes are not statistically significant (Figure 2b). In the projections with lateral boundary data from ECHAM4 or ECHAM5 average salinities are between 29 and 45% lower than in the present climate (Table 1). Such large changes are caused by both increased westerly winds and increased river flow. The larger increase in runoff in the ECHAM4 driven simulations as compared to that in the HadAM3H driven simulations is a consequence of both an increased mean north-south SLP gradient over northern Europe and an enhanced transient activity [Ra¨isa¨nen et al., 2004]. The largest increased river flow of 26% was found in the regional scenario simulation forced with ECHAM5 (run no. 16). However, the salinity reduction in this projection is relatively modest due to a considerable decrease of P ! E over sea.
[15] In the consistent scenario simulations the ratio
be-tween net precipitation over sea and river flow varies between 12 and 19% which is similar to the ratio of 17% calculated for the present climate.
[16] In the projection with the largest salinity change
(RCAO-ECHAM4/A2, run no. 14), the 5 psu isoline is shifted at the sea surface from the northern Bothnian Sea to the western Bornholm Sea (Figure 3). The largest changes were found in the Belt Sea, due to northward shifted salinity fronts as a result mainly of the increased freshwater inflow, and in the northern Baltic proper. In most parts of the Baltic, salinity changes in coastal zones are smaller than in subba-sin centers. Due to increased wind induced mixing, the halocline is deeper than in the present climate and conse-quently the vertical flux of salt is smaller. We assume that this mechanism explains why salinity changes in deeper waters are larger than in coastal zones.
4.
Discussion
[17] The regional scenario simulations of this study rely
on a series of simplifications. Firstly, it is assumed that in future climate the interannual variability of hydrological and atmospheric forcing and of the sea level in Kattegat will Figure 2. Median profiles of salinity at monitoring station
BY15 for present climate 1961 – 1990 (black solid line, shaded areas indicate the ±2 standard deviation band calculated from two-daily values for 1903 – 1998) and in projections for 2071 – 2100 (colored lines). (a) Only effects from wind changes are considered. (b) Projections based upon wind and freshwater inflow changes are shown. Numbers in the legend correspond to the runs in Table 1.
Figure 3. Sea surface salinity (in psu). (a) Climatological data by Janssen et al. [1999] and (b) projection with the largest salinity change in 2071 – 2100 (RCAO-ECHAM4/A2). In Figure 3b projected changes were added to climatological data. Salinities larger than 13 psu are shown in black. In addition, the monitoring station at Gotland Deep (BY15) is depicted.
L15705 MEIER ET AL.: UNCERTAINTY OF SALINITY PROJECTIONS L15705
3
atmosphere is expected, contributing to acidification. For instance, projections for the Baltic Sea indicate a decrease in pH between 0.2–0.4 pH units by the end of this century (Havenhand, 2012). Projected environmental conditions are likely to affect the dynamics of brackish organisms and biogeochemical cycles in the Baltic Sea (HELCOM, 2013; Meier et al., 2014), yet the magnitude of these effects are largely unknown.
The importance of phytoplankton on aquatic
ecosystems
The oceans cover around 70% of the earth’s surface and the biological diversity is far greater than on land, with as many as 100 million different species. A large proportion of these aquatic organisms cannot be seen by the naked eye and are called microorganisms. These aquatic microorganisms play an essential role in biogeochemical cycles and climate regulation in the marine environment. Phytoplankton – derived from the Greek word phyton (plant) and planktos (wanderer or drifter) – is a large group of photosynthetic microorganisms responsible for 70% of the atmospheric oxygen on earth (Harris, 1986). In the process of photosynthesis, carbon dioxide (CO2) and
water are used to produce organic carbon using sunlight as the energy source; oxygen is generated as a by-product. In addition, phytoplankton are dependent on the availability of nutrients (e.g. nitrate, phosphate, silicate, iron) to grow. Overall, phytoplankton fix 40% of the global CO2 in the upper layer of the sea
(Falkowski, 1994) and are responsible for 50% of the global primary production (Falkowski and Raven, 2007). Half of the primary production is transferred to higher trophic levels through grazing; the remaining half is decomposed and remineralized by bacteria (Cole, 1982) and returned to the classical carbon and energy flow. Hence, phytoplankton play an essential role in the structure and the ecological functioning of aquatic food webs.
Phytoplankton species can be prokaryotes (cyanobacteria) but most are eukaryotes (single-celled algae). The most common phytoplankton groups are cyanobacteria, diatoms, dinoflagellates, and green algae. Some phytoplankton can sustain massive algal proliferations, called harmful algal blooms (HABs). The HABs formed by cyanobacteria, dinoflagellates and haptophytes can produce bioactive compounds (including toxins) making them a threat not only to the food web integrity (fish, zooplankton) and ecosystem functionality but also to cattle, house pets and even humans. HABs formed by cyanobacteria (CHABs) are natural phenomena in the Baltic Sea (Bianchi et al., 2000).
4 However, the extent of cyanobacterial blooms in the Baltic Sea have increased during the last decades due to eutrophication and the rise of hypoxia triggering the release of phosphorus from bottom waters (Zillén and Conley, 2010). Cyanobacterial blooms in the Baltic Sea have negative impacts on ecosystems, tourism and aquaculture. Consequently, the understanding of the ecology and ecophysiology of these organisms is of great relevance.
Phytoplankton in a changing environment
Phytoplankton response to the changing climate is vital to the future management of oceans and seas. Preliminary trends show changes in phytoplankton community dynamics such as the occurrence of earlier blooms (Kahru et al., 2011; Poloczanska et al., 2016) and higher diversity in the poles compared to equator zones is expected with an overall expansion of warm-water adapted species (Thomas et al., 2012; Poloczanska et al., 2016). In addition, slight environmental changes may reduce total phytoplankton biomass but increase primary productivity, changing the food web structure and metabolism (O’Connor et al., 2009). At cellular level, phytoplankton stoichiometry is predicted to have higher carbon to nutrient ratio, which is low quality food for zooplankton (Van De Waal et al., 2010; De Senerpont Domis et al., 2014); and to reduce cell size (Finkel et al., 2010; Guinder and Molinero, 2013).
A global decline of phytoplankton concentrations during the last century in response to increasing sea surface temperatures has already been reported (Boyce et al., 2010). However, harmful algal species response to climate change conditions may differ from that of the overall phytoplankton community. The increase in sea surface temperature, dissolved CO2 and HCO3
-is likely to promote HABs, but relevant long-term monitoring -is lacking to confirm these trends (Moore et al., 2008; Wells et al., 2015). From a climate change perspective, higher sea surface temperature could increase the formation of cyanobacterial blooms worldwide (Paerl and Huisman, 2008) and induce earlier and longer blooms at higher latitudes (Neumann et al., 2012; Paerl and Paul, 2012). It is still uncertain how the synergy of temperature and other climate-related changes i.e. salinity will affect cyanobacterial communities and to what extent these organisms will adapt. Few studies have addressed the long-term response of cyanobacteria to environmental disturbances, and most studies focus on freshwater species (Rouco et al., 2011; Low-Décarie et al., 2013).
5
Phytoplankton blooms in the Baltic Sea
In the Baltic Sea, the development of the spring blooms starts when the upper mixed layer is shallower than the euphotic zone due to increase in daily irradiance and thermal stratification (Wasmund et al., 1998). Waters are rich in dissolved inorganic nutrients due to prior turbulence and convective mixing during winter, triggering phytoplankton blooms. Diatoms and dinoflagellates dominate spring blooms in the Baltic Sea. In late spring, phytoplankton have utilized the vast majority of inorganic nutrients and these organisms are grazed, degraded by viral lysis or they sink to the bottom (Lignell et al., 1993). Surface waters become warmer (approximately 17°C) in early summer and a strong thermocline is formed, preventing mixing (Niemi, 1979; Kononen et al., 1996).
In summer, the Baltic Proper is characterized by low nitrogen (N) and high phosphorus (P) concentrations, i.e. low N:P ratio (Granéli et al., 1990). In addition, recurrent hypoxia events in deep sediments release additional P to the surface, enhancing lower N:P ratios (Vahtera et al., 2007). These conditions strongly favor diazotrophic cyanobacteria, organisms that can fix atmospheric N (N2) and convert it into ammonia (NH3), which is an advantageous trait to
outcompete other phytoplankton (Niemi, 1979). Past research suggests that P is the most limiting factor for cyanobacterial proliferations (Mohlin and Wulff, 2009; Andersson et al., 2015a).
In the Baltic Sea, cyanobacterial communities comprise filamentous, colonial and pico-cyanobacteria. Filamentous and colonial cyanobacteria have been mainly studied relying on morphological traits such as cell length, presence or
absence of akinetes and the shape of heterocysts cyanobacteria. Picocyanobacteria
(< 2 µm in size) can be counted by microscopy but cannot be identified at the species level. They are mainly studied together with heterotrophic bacteria using molecular approaches targeting the 16S rRNA gene (Andersson et al., 2010; Lindh et al., 2015). The study of the whole cyanobacterial community, both filamentous/colonial and pico-cyanobacteria, is rarely reported, and molecular techniques can be used for this purpose (Box 1).
6 Box 1
454-Pyrosequencing of 16S rRNA gene amplicons Genetic characterization through
sequencing of the 16S rRNA gene
is a common method to
taxonomically identify different bacteria, both heterotrophic and cyanobacteria, present in a sample. This molecular profiling method avoids the need for
isolation and culturing. Initially
seawater is filtered and DNA is extracted. (1) PCR amplification is performed with specific primers for targeting the 16S rRNA gene and the ligation of adaptors. (2) Prior to sequencing, amplification beads are coupled with the amplified DNA in an emulsion PCR to amplify fragments. (3) In the pyrosequencing reaction, each light signal corresponds to a nucleotide incorporation.
Nodularia spumigena: morphology, distribution and
toxicity
Nodularia spumigena (Fig. 3) is one of the most toxic cyanobacterium in the Baltic Sea. This species forms filaments or trichomes and contains three different morphological and functional cells: vegetative cells, heterocysts and akinetes. Vegetative cells perform the oxygenic photosynthesis. Heterocysts are differentiated cells specialized for the fixation of atmospheric nitrogen, covering nitrogen needs for biochemical processes of other cells in the filament. Akinetes are resting cells used as structure for survival and are formed under unfavourable conditions. They come from the enlargement of a vegetative cell with thick walls.
The first toxic bloom of N. spumigena was reported in Lake Alexandrina, Australia in 1878 (Francis, 1878). Today, N. spumigena is known to form massive algal blooms in lakes and estuaries in Australia, New Zealand and Turkey (Carmichael et al., 1988; John and Kemp, 2006; Akcaalan et al., 2009; McGregor et al., 2012; Sahindokuyucu Kocasari et al., 2015) and
PCR amplification and adaptor ligation 1
3 2
Pyrosequencingy g
p p
Samples pooled and emulsion PCR
Figure B1. Schematic procedure of gene library
preparation using 454 Pyrosequencing. Modified slide from Alex Sanchez.
7
brackish waters such as the Baltic Sea (Sivonen et al., 1989; Wasmund, 1997). Bloom formations in the Baltic Sea have been reported since the 19th century and fossil records reveal that these blooms have occurred for the last 7000 years (Bianchi et al., 2000). N. spumigena optimal temperature range oscillates between 20 and 25°C (Lehtimäki et al., 1994). In addition, this species is highly tolerant to strong irradiance and moderate salinity (Mazur-Marzec et al., 2005).
Nodularin, a cyclic pentapeptide, is the most well-known toxin produced by N. spumigena. Nodularin is a potent liver tumor promoter in both animals and humans (Moffitt and Neilan, 2004; Dittmann and Wiegand, 2006) by inhibiting protein phosphatases (Ohta et al., 1994). This toxin can accumulate in aquatic food webs as it is efficiently transferred to higher trophic levels like Box 2
Gene expression determined by RT-qPCR Gene expression is the process by which information from a gene (stored in the DNA) is used for the synthesis of a product, often proteins. Gene expression can change in response to environmental conditions, indicating stress patterns, and
can be quantified using reverse
transcriptase - quantitative polymerase chain reaction (RT-qPCR). It can for example be used to measure expression of non-ribosomal peptides (NRPs) genes in
Nodularia spumigena (e.g. ndaF gene).
This method reveals the potential of peptide synthesis under different conditions, e.g. salinity changes (Paper III).
Figure B2. Amplification plot of the
ndaF gene obtained by RT-qPCR
(raw data from Paper III).
Experiment:PCR efficiency_160915_2
Experiment Results Report Applied Biosystems StepOnePlus™ Instrument Amplification Plot (∆Rn vs. Cycle)
User: 8 Printed:2015 Sep 16 9:40:55 PM Figure 3. Nodularia spumigena (long filament in
the middle) and Dolichospermum spp. (pearl-necklace-like filaments), sampling location outside Kårehamn, east of Öland in July 2006. Photo courtesy of Elin Lindehoff.
8 zooplankton (Karjalainen et al., 2006), blue mussels and fish tissues (i.e. liver) (Sipiä et al., 2001). N. spumigena can, in addition to nodularin, produce other non-ribosomal peptides (NRPs) (Mazur-Marzec et al., 2013). The extent to which changes in environmental conditions impact the biosynthetic pathways and the synthesis of NRPs is largely unknown. Molecular and chemical techniques provide a wide range of possibilities to determine the potential of studying NRPs pathways and to screen new NRPs produced by cyanobacteria (Boxes 2 & 3).
Box 3 LC-MS/MS
In order to elucidate the structure of NRPs in cyanobacteria, liquid chromatography-tandem mass spectrometry is commonly used. This methodology was used in Papers III, IV and V and combines liquid chromatography (LC) and mass spectrometry (MS). The first technique (LC) separates the compounds present in a complex biological sample. In the MS system, structure characterization of the
compounds is performed, using different experiments (e.g. Enhanced Product ion
Mode). If tandem mass spectrometry is used, the precursor ion can be fragmented
to generate its fragmentation spectrum. MS/MS advantage is its high selectivity, as the collected spectrum represents a kind of fingerprint, unique for each structure.
Figure B3. Schematic simplification of LC-MS/MS methodology.
Nodularia cells Cell lysis
LC- MS/MS
Ionized parent peptide Primary fragmentation
9
A
IMS
To predict ecological consequences of climate-induced changes on phytoplankton, it is important to: firstly, identify the drivers and synergistic effects shaping phytoplankton community composition and dynamics; secondly, determine phytoplankton specific physiological and chemical responses to specific environmental factors.
Using field and laboratory experiments, the following aims were addressed in this thesis:
- To unravel phytoplankton dynamics and successional patterns in the Baltic Proper (Papers I and II).
- To study abiotic and biotic factors affecting phytoplankton dynamics (Papers I and II).
- To evaluate N. spumigena phenotypic plasticity in response to salinity stress in a climate change scenario perspective (Papers III and IV).
- To elucidate the structure of NRPs produced by Baltic N. spumigena (Papers III, IV and V) under stress factors (Papers III and IV).
- To investigate the gene expression of NRPs genes in N. spumigena under stress conditions (Paper III).
10
R
ESULTS AND
D
ISCUSSION
Phytoplankton dynamics and successional patterns
We investigated phytoplankton diversity and successional patterns in the Baltic Proper by performing spatial and temporal sampling (2-3 years). Phytoplankton were examined using traditional methods like microscopy and chlorophyll content (Chl a) (Papers I & II), and also by molecular techniques such as amplification of the 16S rRNA gene using 454 Pyrosequencing (Box 1, Paper II). Results showed that in the Baltic Proper, a clear successional pattern of phytoplankton could be identified (Fig. 4, Papers I & II). In spring, diatoms (Chaetoceros, Skeletonema, and Thalassiosira) and dinoflagellates (Peridiniella, Heterocapsa, Protoperidinium) bloomed and could account for more than 95% of the total phytoplankton biomass (Paper I). In early summer, filamentous cyanobacteria, e.g. Nodularia spumigena germinated from akinetes (Fig. 4, Suikkanen et al., 2010), and colonial cyanobacteria accounted up to 95% of the total phytoplankton community (Papers I & II). The most abundant species was Aphanizomenon (75-90% of total phytoplankton community), while toxic species such as N. spumigena and Dolichospermum were less abundant (15-50% of total phytoplankton community). In early autumn, with mixing conditions and lower temperatures, filamentous cyanobacteria formed akinetes and sank to the bottom, providing the seed bank for future blooms. In addition, picocyanobacteria appeared during the bloom of filamentous/colonial cyanobacteria in summer and persisted even after the bloom had decayed (Fig. 4, Paper II). Occurrence of picocyanobacteria may be associated to the highly N-rich microenvironment provided by diazotrophic cyanobacteria (Ploug et al., 2011). Furthermore, the increase in heterotrophic bacterial abundance during and after the cyanobacterial bloom (Lindh et al., 2015) indicates that bacteria can also benefit from cyanobacterial compounds that are rich in organic N and P.11
In Paper I and Paper II, many phytoplankton species co-occurred for weeks, competing for nutrients in low N:P conditions. This situation is called the “paradox of plankton” (Hutchinson, 1961), in which the competitive exclusion principle is not met. Instead, when two species compete for the same resources, both survive and there is no species extinction.
Figure 4. Conceptual model of seasonal succession patterns of phytoplankton in the Baltic
Proper. Curves indicate Chl a abundance. Adapted from Suikkanen et al. (2010) and Martin (2012).
Genetic characterization of the 16S rRNA sequences revealed highly diverse populations of filamentous, colonial and pico-cyanobacteria. Different patterns of occurrence were detected among populations within the same species, indicating both generalists and opportunists. We detected an Aphanizomenon/Dolichospermum (filamentous cyanobacteria) generalist cluster that occurred all year around and persisted when opportunistic filamentous cyanobacteria burst e.g. N. spumigena and Pseudanabaena (Fig. 4).
Chl a
Winter Spring Summer Autumn
Germination Time Akinete formation Sedimentation Bottom Diatoms Dinoflagellates Picocyanobacteria Aphanizomenon spp. Dolichospermum spp. Nodularia spumigena Phytoplankton groups/species
12 This pattern suggests that Aphanizomenon/Dolichospermum formed an “epidemic population structure” (dominance of a single cluster) similar to Planktothrix in a freshwater subalpine lake (D’Alelio et al., 2013). Picocyanobacterial sequences, Synechococcus/Cyanobium, formed two main clusters appearing all year around and the other clusters occurred occasionally (Paper II, Fig. 4). Similar occurrence patterns were detected in other Synechococcus subpopulations from Southern California Bight and Chesapeake Bay (Tai and Palenik, 2009; Cai et al., 2010). Overall, picocyanobacteria from the Baltic Sea consist of a highly genetically diverse group (Haverkamp et al., 2008; Ahlgren and Rocap, 2012; Paper II).
A recurrent phytoplankton succession, including picocyanobacteria, in the Baltic Proper is shown in this thesis. Temporal and spatial successional patterns of phytoplankton were affected by both seasonal environmental conditions and microbial community changes.
Abiotic and biotic factors related to phytoplankton
community dynamics
A common perception is that occurrence and seasonality of phytoplankton blooms is mainly related to light and nutrients (Legendre, 1990). In this thesis we considered both environmental conditions and hetetotrophic bacterial dynamics as potential indicators for changes in phytoplankton community composition in the Baltic Proper. The relationship between phytoplankton biomass and temperature showed in Paper I was not linear compared to results published for other temperate aquatic ecosystems (Li et al., 2006). High phytoplankton biomass levels (1200 mg m-3) were strongly linked with low temperatures (Fig. 5A, Paper I) and high nutrient levels, i.e. P, total P and N (Fig. 5B). This increase in biomass was related to the spring bloom of diatoms and dinoflagellates occurring after the ice breakup; this pattern is typical for cold-water ecosystems.
In the Baltic Proper, niches could be differentiated at subpopulation level of both filamentous/colonial and picocyanobacteria (Paper II). The occurrence of opportunistic filamentous cyanobacteria, e.g. the hepatotoxic N. spumigena, was linked to low salinity and high temperature. At community level, cyanobacteria that were prominent in summer were related to strong stratification, higher temperature, lower salinity (0.6 units decrease), and low nutrient concentrations (both P and N) (Paper II). In previous studies temperature, salinity (range between salinity 3 and 25), stability of the water
13
column, and nutrient availability were the reported environmental factors controlling cyanobacterial blooms in the Baltic Sea (Larsson et al., 1985; Wasmund et al., 2012; Andersson et al., 2015a). It has been shown that P availability in the water column stimulates growth of cyanobacteria (Paerl and Huisman, 2009; Andersson et al., 2015a). However in the Baltic Proper, since opportunistic cyanobacteria occurred during summer when nutrients were very low and were related to high bacterial abundance, we hypothesize that the nutrients were provided by bacteria through remineralization processes. However, the increase in nutrient concentration could not be detected. This was likely due to rapid uptake of remineralized nutrients during the low nutrient conditions in the Baltic Proper.
Figure 5. A) Relationship between phytoplankton biomass (mg m-3) and temperature (°C)
during 2010-2012 in the Baltic Proper (n=195) (modified from Paper I); B) 3D plot representing results from the linear model in which phytoplankton biomass is the dependent variable and Temperature and PC1 (representing P, total P and N) are the explanatory variables. PC1 is positively correlated with P, total P and N.
The interannual variability in phytoplankton biomass found for the Baltic Proper reflects changes in climate conditions. In cold winters, spring bloom represented the proliferations with highest biomass (7-8 mg Chl a m-3) while milder winters were followed by a decrease of spring bloom intensity (2-4 mg Chl a m-3). Future climate conditions, with warmer temperatures and milder winters, project a decrease of diatom concentration during spring but an increase of cyanobacteria abundance during summer in the Baltic region (HELCOM, 2007). Although cyanobacterial biomass may be promoted during
20 02 00 40 0 60 0 80 0 10 00 12 00 Phytoplankton biomass (mg m 3) 15 10 5 r2= 0.45 p< 0.001 Temperature (°C) 1.0 0.5 0 -0.5 -1.0 0 5 10 15 20 25 4 6 8 10 PC1 Temperature (°C)
Phytoplankton biomass (BoxCox transformation)
14 summer due to the rise in temperatures (Paerl and Huisman, 2008), it is unlikely that the cyanobacterial summer blooms can compensate the loss of spring biomass. In addition, the intensification of cyanobacterial primary production may promote remineralization processes through heterotrophic bacteria(Andersson et al., 2015b). Therefore in this scenario, less energy would be transferred to zooplankton and higher trophic levels; both because filamentous cyanobacteria are difficult to feed on and are poor quality food.
For the Baltic Sea, future climate scenarios predict higher temperature (2-4°C higher) and lower salinity (2-2.5 units lower) for the next century (Fig. 2, Meier et al., 2014). This thesis showed that those factors play an essential role in triggering CHABs (Fig. 6, Paper II). Therefore, the extensive monitoring performed in this study, both spatial and temporal, indicates that cyanobacteria have the capacity for range expansion in future climate conditions in the Baltic Proper.
Figure 6. Potential effects of changes in environmental conditions due to Climate change on
Cyanobacterial Harmful Algal bloom (CHAB) occurrence in the Baltic Proper (modified from O’Neil et al., 2012).
Less CHABs More CHABs
Environmental factors influencing CHABs in the Baltic Sea
Salinity River-run off Stratification Temperature
More diatoms and dinoflagellates
Nutrients
15
Phenotypic plasticity of N. spumigena to changing
salinity: life strategies to ensure survival at any cost
Long-term experiments on phytoplankton responses to climate change conditions have been mainly focused on CO2 and temperature increase (Table1). In the Baltic Proper, besides temperature, salinity has a key role in shaping the cyanobacterial community composition (Paper II). In this thesis, we focused primarily on salinity. Previous studies have shown that short-term shifts in salinity affect filamentous cyanobacteria and differences in tolerance depending on species, treatment and strain were detected (Lehtimäki et al., 1997; Hobson and Fallowfield, 2001; Moisander et al., 2002; Musial and Plinski, 2003; Mazur-Marzec et al., 2005; Engström-Öst et al., 2011; Möke et al., 2013; Rakko and Seppälä, 2014). We assessed phenotypic plasticity/adaptation of N. spumigena both to short and long-term salinity changes. Phenotypic plasticity can be defined as the variation in phenotype associated to a given genotype, with no requirement of genetic variation, in response to stress related to environmental changes (Pigliucci et al., 2006). The findings described in Papers III and IV emphasized the high phenotypic plasticity and rapid adaptation exhibited by N. spumigena strains (Fig. 7). In the short-term, some strains responded by lowering the growth rate (Paper III), but partially resumed growth during long-term experiments (Fig. 7, Paper IV). The results in Paper IV showed N. spumigena as a versatile organism, with a tendency to have shorter filaments (strain KAC11) at low salinity (Fig. 7). Shorter filaments are easier to be managed by zooplankton (Bednarska et al., 2014), which may enhance grazing pressure with subsequent changes in carbon flow in the food web. In addition, the low photosynthetic growth efficiency (PGE, defined as the production of cyanobacterial biomass per unit of inorganic carbon assimilated) in response to low salinity in some strains (strain KAC66) is likely to have large effects on biogeochemical cycles.
Existing models ensure the survival of phytoplankton in future climate conditions, but the extent of their adaptation capability is uncertain (Irwin et al., 2015). The new insights presented in this thesis regarding the phenotypic plasticity of N. spumigena represent crucial information to be added into community ecosystem models for the Baltic Sea. A more accurate assessment of cyanobacterial blooms including physiological and metabolic response to climate change conditions may help to unravel potential implications for the Baltic Proper ecosystem for the next century.
16
Figure 7. Example of two life strategies of Nodularia spumigena strains, resilience (KAC66)
versus versatility (KAC11), in response to decreased salinity. Abbreviations stand for environment (env.), growth rate (µ) and photosynthetic growth efficiency (PGE).
Non-ribosomal peptides produced by N. spumigena:
synthesis and regulatory mechanisms
Toxins are the most well-studied secondary metabolites produced by cyanobacteria (Repka et al., 2004; Dittmann and Wiegand, 2006; Sivonen and Böorner, 2008). The hepatotoxin nodularin was identified as a constitutive NRP produced by all strains of N. spumigena (Paper V). N. spumigena from the Gulf of Gdansk showed reduced nodularin biosynthesis at lower salinity (Mazur-Marzec et al., 2005). However, our results showed that the nodularin to Chl a ratio was constant in N. spumigena from the Baltic Proper at reduced salinity. Although some ecophysiological regulation must have taken place caused by short- and long-term salinity changes, it is possible that nodularin production was not down-regulated or up-regulated long enough to be able to detect changes on cellular toxin content (N. spumigena strains KAC66 and KAC11, Papers III & IV). As primary function, we suggest that nodularin is used for cellular reparation in stress conditions, similarly to microcystin (Zilliges et al., 2011). In fact, genes related to NRPs have existed prior to the evolution of eukaryotes (Rantala et al., 2004; Dittmann and Wiegand, 2006), indicating that these secondary metabolites are enrolled primarily in cellular
Model 1: KAC66 Model 2: KAC11 • short strain • = µ Final (300 gen.) salinity • = size • = µ • carbon uptake • = respiration • PGE Initial (0 gen.) salinity • size • µ • = carbon uptake • = respiration • = PGE Discussion
The aim of the study was to analyse, if any changes in growth rate of Nodularia spumigena has happened in the three salinities over the time period of two years. And veritably differences were detectable; already the biomass comparision showed significant differences. Nodularia growing in salinity 7 builds long filaments, which are three time longer than the ones from salinity 5 and eight times longer than at salinity 3. Even the naked eye can see, that the filmanents in 7 are quite long, why they got the name cat hair, which is not the case in salinity 5 and 3. In the original salinity (7) Nodularia was up to 5000 µm long, which are 5 mm. Smaller filaments were found in salinity 5 (max. 1800 µm) and 3 (max. 512 µm). The trichome length was comparable with other studies from Mazur-Marzec et al. (2005) with 1527 ± 791 µm (salinity 7), whereas the filaments in salinity 3 were shorter in this study than at theirs with of 804 ± 534 µm. The much shorter trichomes in the lower salinities are the first sign, that Nodularia still thrieves best in its original salinity of 7.
Figure 9: Figure 9: Pictures of Nodularia spumigena taken under the light microscope. As indicated the left column represents
Nodularia in salinity 7, middle column salinity 5 and right column salinity 3. The bars in the first two rows represent 20 µm
and the bars in the last row represent 200 µm.
And there are also other properties demonstrating this. Just in salinity 7 the filaments were buoyant and floated on the surface building vast piles (see Fig. 8, treatment 7(7)). Neither in salinity 5 or 3 buoancy could be observed after two years. So far, it was reported in the literature, that Nodularia spumigena can
20 µm
Discussion
The aim of the study was to analyse, if any changes in growth rate of Nodularia spumigena has happened in the three salinities over the time period of two years. And veritably differences were detectable; already the biomass comparision showed significant differences. Nodularia growing in salinity 7 builds long filaments, which are three time longer than the ones from salinity 5 and eight times longer than at salinity 3. Even the naked eye can see, that the filmanents in 7 are quite long, why they got the name cat hair, which is not the case in salinity 5 and 3. In the original salinity (7) Nodularia was up to 5000 µm long, which are 5 mm. Smaller filaments were found in salinity 5 (max. 1800 µm) and 3 (max. 512 µm). The trichome length was comparable with other studies from Mazur-Marzec et al. (2005) with 1527 ± 791 µm (salinity 7), whereas the filaments in salinity 3 were shorter in this study than at theirs with of 804 ± 534 µm. The much shorter trichomes in the lower salinities are the first sign, that Nodularia still thrieves best in its original salinity of 7.
Figure 9: Figure 9: Pictures of Nodularia spumigena taken under the light microscope. As indicated the left column represents
Nodularia in salinity 7, middle column salinity 5 and right column salinity 3. The bars in the first two rows represent 20 µm
and the bars in the last row represent 200 µm.
And there are also other properties demonstrating this. Just in salinity 7 the filaments were buoyant and floated on the surface building vast piles (see Fig. 8, treatment 7(7)). Neither in salinity 5 or 3 buoancy could be observed after two years. So far, it was reported in the literature, that Nodularia spumigena can
20 µm stable, low-salinity env.
dynamic, high-salinity env.
• long strain
17
functions. Furthermore, nodularin has an anti-grazing potential (Gorokhova and Engström-Ost, 2009) and together with other NRPs may be essential for microbial interactions.
Cyanobacterial can produce a wide array of NRPs and research on these compounds is an expanding field for potential drug candidates (Felczykowska et al., 2015). Diversity of NRPs has been screened in several cyanobacterial genera, including Aphanizomenon, Nodularia and Microcystis (Welker et al., 2004; Ferreira Ferreria, 2006; Fewer et al., 2009; Mazur-Marzec et al., 2013). A recent study highlighted that N. spumigena, besides nodularin, produces other linear and cyclic peptides such as spumigins, nodulapeptins, aeruginosins, and anabaenopeptins (Mazur-Marzec et al., 2013). In this thesis (Papers III, IV & V), we limited our analyses to the presence/absence of NRPs, because a quantitative determination requires standards of each of the peptides. Most standards of these compounds are not available, since many of the screened peptides were novel compounds (Box 3, Paper V). To facilitate the detection of individual peptides at low concentrations, Enhanced Product ion Mode was used because of its high sensitivity (Box 3). N. spumigena NRP profiles were specific at strain level (Fewer et al., 2009; Mazur-Marzec et al., 2013, Papers III, IV & V). Furthermore, the constitutive synthesis of NRPs shows the ability to cope with changes in salinity and nutrient conditions (Papers III & IV). Thus, chemical profiling (NRP profiles) can be used to classify cyanobacteria strains into metabolically different chemotypes, to distinguish and cluster cyanobacteria populations and to trace their dynamics in aquatic ecosystems (Rohrlack et al., 2009; Grabowska and Mazur-Marzec, 2016). Yet, why there is strain-specific production of NRPs and if there is a relation between specific chemotypes and specific environmental conditions are still open questions. Paper IV shows for the first time in N. spumigena that different chemotypes can represent distinct ecotypes.
NRPs produced by cyanobacteria showed high structural diversity (e.g. Paper V), indicating the existence of a wide range of different biosynthetic pathways. NRPs are synthesized by enzyme complexes that are encoded by gene clusters. Gene clusters for nodularin (nda), nodulapeptin or anabaenopeptin (apt), microcystin (mcy), aeruginosins (aer), and spumigin (spu) have been identified in cyanobacteria (Moffitt and Neilan, 2004; Voß et al., 2013). The potential of N. spumigena to produce NRPs can be assessed by measuring gene expression of NRPs genes (Box 2). Under any kind of stress situation, it is assumed that cells have the capability to control and self-regulate gene expression in order to adjust the amount and type of proteins produced. In
18 this thesis, we used three strains of N. spumigena to determine the response of NRP gene expression to salinity and nitrate stress (Paper III). Gene expression of the last gene in the nodularin gene cluster, ndaF, was not significantly different at early, late exponential or stationary phase regardless of decreased salinity. In addition, gene expression of other targets encoding NRPs (nda, spu, aer) was constant regardless of salinity and nutrient treatment. Overall, no stimulation was detected in any of the treatments and results in this thesis showed that expression of regions encoding NRPs persist during low growth rates due to saline shock.
19 Ta b le 1 . Su m m ar y ta bl e of p he no ty pi c pl as ti ci ty r es po ns es t o lo ng -te rm c lim ate -in du ce d ch an ge s in p hy to pla nk to n. C om m on ly , a da pta tio n in p hy to pla nk to n ha s be en as sessed u si ng g ro w th r at e as a pr ox y fo r fi tn ess. A ll r esp on ses ar e m easu red at th e po pu lat io n lev el ( bat ch cu lt ur es) . P hy to pl ankt on gr oup na m es a re ab br ev iat ed : C hl or o – Ch lo ro ph yt a, D in o – Di no fl ag el la te , C ya no – Cy an ob ac te ri a, D ia – Di at om, C oc co – Co cc ol ith op ho re . A bb re via tio ns in th e ta ble s ta nd fo r: N° g en . – num be r of ge ne ra ti ons , In cr. fi tn ess – in cr ea se in f itn es s, R ef . – re fe re nc es , P – phos pha te , P O C – pa rt ic ul at e or ga ni c ca rbon, P O N – pa rt ic ul at e or ga ni c ni tr oge n, N – ni tr at e, P IC – pa rt ic ul at e inor ga ni c ca rbon, N R P s – non -ri bo so m al p ep ti de s. Ta rg et s p ec ie s Gr ou p N° st ra in s N° gen . Tr ea tm ent s In cr. fi tn es s Ot h er t ra it s me as u re d Re f. Ch la m yd om on as r ei nh ar dt ii Ch lo ro 2 1000 In cre as ed C O2 (1 05 0 ppm ) Ye s Ce ll s iz e, p ho to sy nt he si s, re sp ira ti on Co ll in s & Be ll , 2004 Pr or oc en tr um tr ie st in um Di no 2 520 In cre as ed te m pe ra tur e (25° ) Ye s Ce ll s iz e Fl or es -Mo ya e t a l., 2008 Ch la m yd om on as r ei nh ar dt ii Ch lo ro 1 200 Di ve rs e P co nc en tr at io ns Ye s Fr eq ue nc y of p la sm id Co ll in s & M ea ux , 2009 Th al as si os ir a ps eu do na na Di a 1 100 In cre as ed C O2 (7 80 µ at m ), fl uct uat in g pH St ab le Ph ot os yn th et ic e ff ic ie nc y, PO C , PO N , ge ne e xpr es si on Cr aw fu rd e t a l., 2011 Al ex an dr iu m m in ut um Di no 2 180 -220 Re du ce p H ( 7. 5) , in cr ea se te m per at ur e (2 5° C ) Ye s In tra ce ll ul ar to xi n Fl or es -Mo ya , 2011 Mi cr oc ys ti s ae ru gi on os a Cy an o 3 87 Hi gh er N co nc en tr at io n an d te m pe ra tu re (3 0° C ) Ye s To xi n pr od uc ti on Ro uc o et a l., 2011 Em il ia ni a hu xl ey i Co cc o 1 500 In cre as ed C O2 ( 1100 µ at m an d 22 00 µ at m ) Ye s PI C , PO C , c el l s iz e Lo hb ec k et a l., 2012 Ge ph yt oc ap sa o ce an ic a Co cc o 1 670 In cre as ed C O2 (1 00 0 µ at m ), r ed uced p H ( 7. 8) Ye s Ce ll b io vo lu m e, p hot os ynt he si s ra te , P O C , PO N Ji n et al ., 2013 Sy ne choc oc cus , An ab ae na , Na vi cu la , Ni tz sc hi a, Ps eu do ki rc hn er ie ll a, Sc ene de sm us Cy an o, Di a, Ch lo ro 7 750 In cre as ed C O2 (1 00 0 ppm ), w ith a nd w ith ou t nut ri ent s Ye s NA Lo w -D écar ie et al ., 2013 Li ng ul od in iu m , Pr or oc en tr um , Al ex an dr iu m , Go ny au la x Di no 3 48 -126 In cre as ed C O2 (4 33 a nd 765 ppm ) St ab le NA Ta tt er s et a l., 2013 Em il ia ni a hu xl ey i Co cc o 1500 In cre as e tem per at ur e (2 6. 3 °C ) an d C O2 (1 00 0 an d 22 00 µ at m ) Ye s Ce ll s iz e, P O C, P IC, P O C Sc hl üt er e t a l., 2014
20 Os tr eo cc oc us ta ur i Ch lo ro 16 400 In cre as ed C O2 (1 00 0 µ at m ), fl uct uat in g C O2 (1 00 0 µ at m ) Ye s O xyge n ev ol ut io n rat es Sc ha um & C ol li ns , 2014 Em il ia ni a hu xl ey i Co cc o 1 2100 In cre as e CO 2 (1 00 0 an d 2200 µ at m ) Ye s Ce ll s iz e, P O C, P IC Sc hl üt er e t a l., 2016 Ch lo re ll a vu lg ar is Ch lo ro 1 100 In cre as ed te m p. (3 3° C ) Ye s Ph ot os yn th es is r at e, r es pi ra ti on , Pa df ie ld e t a l., 2016 No du la ri a sp um ig en a Cy an o 2 300 De cr ea se d sal in it y (7 , 5 , 3 ) St ab le / re su m ed Fi la m en t b io vo lu m e, N R P s, c ar bo n upt ake , r es pi ra ti on Pa pe r IV (t hi s th es is )
21
C
ONCLUSIONS AND
F
UTURE
P
ERSPECTIVES
Phytoplankton, including cyanobacteria, represent the base of the food web in the Baltic Sea and may play an even more vital role in future climate conditions. These organisms provide carbon and additional N (by diazotrophic cyanobacteria) to fuel the brackish ecosystem but they also contribute to the production of secondary metabolites and may promote oxygen deficiency during summer. Thus, it is important to investigate the niches occupied by different phytoplankton and how environmental disturbances might influence their physiological, metabolic and chemical responses. This thesis has contributed to the expansion of the current knowledge about phytoplankton and filamentous cyanobacteria dynamics, niche-specificity, genetic diversity, physiology, and metabolic aspects driven by environmental factors in the Baltic Sea.
In this thesis, results revealed recurrent successional patterns on phytoplankton and highlighted that environmental future conditions may promote changes favoring the occurrence of specific cyanobacterial populations e.g. the toxic Nodularia spumigena. Many studies base their research of microbial community dynamics on environmental factors. However, to provide a clearer picture of phytoplankton dynamics in a changing environment such as the Baltic Sea, it is also necessary to assess microbial dynamics and interactions.
At long-term, the loss of phytoplankton biomass during spring caused by milder winters is expected with an increase of cyanobacterial abundance. However, the cyanobacterial biomass in summer blooms will not compensate for the carbon loss during the spring bloom. These results suggest lower energy transfer to higher trophic levels and modifications of the food web functioning. Future climate conditions predict, besides higher temperatures, lower salinity in the Baltic Proper. Further investigations of synergistic effects of these environmental factors would help to assess the response of microbial food
22 webs. Therefore, experiments combining changes in salinity and temperature, not only in the short- but also in the long-term, will be key to unravel phytoplankton physiological responses to environmental changes.
As shown in this thesis, it is likely that Nodularia spumigena will rapidly adapt to the environmental conditions predicted in future climate scenarios. As N. spumigena occur every summer, a study assessing the fraction of adaptive N. spumigena could further highlight consequences of these blooms in the Baltic Sea ecosystem. In this context, determining the filament size can help to indicate the extent to which carbon flow may be modified. NRP profiling is a stable feature for each strain of N. spumigena isolated from the Baltic Sea. In addition, our studies support the potential of using NRPs profiles to study N. spumigena dynamics in strains that are closely genetically related. These studies are urgently warranted to forecast the responses of phytoplankton and notably cyanobacteria to different climate change scenario as well as their impact on the global ecosystem functioning.
23
A
CKNOWLEDGEMENTS
“Real knowledge is to know the extent of one’s ignorance” – Confucius My time as a PhD student has been enriching and has helped me to grow personally and professionally. It has been a life experience, at times frustrating, but extremely valuable and for that I am deeply grateful. I had the chance to travel and discover new places, to live in Sweden and Poland, meet super nice people with whom I worked hard and had loads of fun.
One person has come along with me for the entire journey and I would like, first and foremost, to thank my supervisor and mentor, Catherine. I feel extremely grateful for this time. You provided many opportunities to expand my knowledge inside and outside the Linnaeus walls, and your support and guidance has been crucial. Thanks for the useful critiques of the work that went into this research. We both know it has been a tough journey but we managed to make it together as a team! IEA! Thanks to Jarone Pinhassi, as my co-supervisor, for including me in Journal clubs and involving me in so many interesting discussions with the Microbial group. I would also like to thank Agneta and the Ecochange team in Umeå for inspiration and insightful discussions to broaden the perspective in this work.
Hanna Mazur-Marzec and her group Ania, Justyna, Ania T., Natalia and Agnieszka in the University of Gdansk (Poland) for their kind reception and help through the two periods I stayed in Gydnia. Hanna, you truly inspire me for your amusement regarding spectrum of peptides. I enjoyed working with all of you.
To my closest collaborators during this thesis Hanna Farnelid, Elin Lindehoff, Emil Fridolfsson and Carina Bunse; I really couldn’t have made it without you. Thanks for being patient with me, helping me throughout my time here, working together, and having fun too! I am also grateful to MPEA group:
24 Emmelie, KB, Martin, Frederik, Berner and Eva. Also to Caroline big, huge thanks for all the enthusiasm, support, chatting in the corridors and proofreading!
I am thankful for all the meetings and projects done with the Prodiversa network. We managed to do tremendous fieldwork in the Baltic Sea and I am glad to have participated. It made me realize that all together we can do better than alone; there is strength in numbers ;).
To the GENECO people, GENECO meetings and GENECO Mentor Program, thank you. To my mentor Helena, for taking her time to listen and care for me. Also special thanks to my great friends in the Mentor Program with whom I shared both good and bad moments. The Cesky Krumlov course in genomics was tough but allowed me to meet Eva, Ingrid, Jon, Gerard, Jordi and Lisa, which made this two-week course fantastic and full of enjoyment coding like crazy J.
Elena, I cannot thank you enough for all the things you brought into my life and the encouragement towards bettering myself along this journey. Te quiero Neni!
To my dearest friend Sandra, I am really, really happy you appeared in my life. You encouraged me, gave me fantastic ideas, comforted and helped me during the time you have been in Kalmar and also now that you are in Spain. Your support has been essential to me. Mil gracias Churri!
Special thanks to my former housemates and great friends Olga and Antonio. I very much enjoyed living with you all these years, spending great quality time together. I shared a lot with you and I will always have you in my heart. I really enjoyed all these years in and outside the university, and that has been because of all of you, my friends: Carina (for being there all the time and to listen to my bla bla bla stuff), Michelle (for the great nights having drinks), Jo (for caring about me and the good funny moments trying to talk “real English”), Felix (for great couples’ lunches and dinners), Domenico (for making life so funny), Anna S. (being the first Swedish friend J), big Håkan (for your really interesting talks), little Håkan (for the help during the Phd), Anu (your apple muffins, mmm), Stephan (for great lunches), Daniel Fernando (for telling me about the world), Kocken (fishing gear and tips), Andreas & Ullrika (for great moments together!). Thanks also to people that I shared more than one beer, fika break, lunch and corridor conversation: Joacim, Margarita, Carlos, Charlotte, Conny, Neus, Ben, Fabio, Karin H, Oscar, Peter, Devi, Emelie, Johanna, Daniel L., Daniel B., Furong, Xiaofen, Fatima, Anders, Henrik … and all those who were accidentally left out.
25
To my good and old friends in Spain and around the world; a mi grupo de amigos en Barcelona: Laura, Lujan, Dani, Raquel, Marc, Meri, Charlie, Ortega y Jordi de Vic. Quiero dar las gracias también a Ivan, por ser siempre una caja de sorpresas y hacerme reír. A Encarna por esos buenos meses aquí en Kalmar y en Cádiz, et trobo a faltar!. Merci Laurent por todos los ánimos. A mis amigos del Erasmus Niuris, Maria, Pablins, Ivanovich, Samu titi wapis que aunque nos veamos poquito se que siempre os tender ahí.
A mon amour Alexis. Merci d’avoir été présent dans les mauvais moments et de célébrer avec moi les bons. Ton réalisme m'a aidé à aller de l'avant au lieu de rester dans les nuages. Merci pour ton soutien et j’ai hâte de voir ce que la vie nous réserve. Je t’aime.
A mi familia, mis padres, mi hermana y peachy. Por estar siempre ahí y por vuestro soporte incondicional. Por todas las oportunidades que me habéis brindado y por creer en mí. A mi cousin Iolanda por todo el apoyo y ayuda en los últimos momentos de la tesis, gracias por la portada de esta tesis y por conseguir lo que quería. A ti peludeta, por ayudarme siempre a ver más allá de lo que tengo delante de la nariz, y por darme ese empujoncillo en momentos necesarios. Os quiero muchísimo.
