Florida State University Libraries

Electronic Theses, Treatises and Dissertations The Graduate School

2018 Insights into Carbon Acquisition and Photosynthesis in Karenia Brevis under a Range of CO2 Concentrations Tristyn Lee Bercel

Follow this and additional works at the DigiNole: FSU's Digital Repository. For more information, please contact [email protected] FLORIDA STATE UNIVERSITY

COLLEGE OF ARTS AND SCIENCES

INSIGHTS INTO CARBON ACQUISITION AND PHOTOSYNTHESIS IN

KARENIA BREVIS UNDER A RANGE OF CO2 CONCENTRATIONS

TRISTYN LEE BERCEL

A Thesis submitted to the Department of Earth, Ocean, and Atmospheric Science in partial fulfillment of the requirements for the degree of Master of Science

2018 Tristyn Lee Bercel defended this thesis on September 17, 2018. The members of the supervisory committee were:

Sven Kranz Professor Directing Thesis

Angela Knapp Committee Member

Olivia Mason Committee Member

Michael Stukel Committee Member

Janie Wulff Committee Member

The Graduate School has verified and approved the above-named committee members, and certifies that the thesis has been approved in accordance with university requirements.

To my parents, Mark and Theresa, who may not believe in climate change, but believe in me.

iii ACKNOWLEDGMENTS

I wish to acknowledge the Florida Fish and Wildlife Commission for providing Karenia brevis (CCFWC-126), thank Thomas Kelly for providing his R-script which assisted in evaluation of the oxygen evolution data, and Raphael Richardson for his assistance in running the DIC samples. Special thanks to my committee, especially Sven Kranz, for their guidance with my work.

iv TABLE OF CONTENTS

List of Tables ...... vi List of Figures ...... vii Abstract ...... ix

1. GENERAL INTRODUCTION ...... 1

2. INSIGHTS INTO CARBON ACQUISITION AND PHOTOSYNTHESIS IN KARENIA BREVIS UNDER A RANGE OF CO2 CONCENTRATIONS ...... 10

3. GENERAL DISCUSSION ...... 37

References ...... 40

Biographical Sketch ...... 50

v LIST OF TABLES

Table 1. Carbonate chemistry measured and calculated in the experiment. Data are given for carbonate chemistry measured prior to the addition of cells, as well as during mid exponential growth phase ...... 20

Table 3. Measured net and gross primary productivity and respiration rates from 24 h experiments. Rates are average of three replicate measurements and averaged over the 12- hour light phase (NPP, GPP) or the dark period (RO, RC)...... 25

Table 4. Calculation parameters for NPPLosh. NPPLosh was based on equations outlined in (Losh et al., 2013) and was calculated with the following equation: NPPLosh = (((0.20 * (µ/R)) * 5 9 P)/(5.5 X 10 ) *8*3*12*(2/5)*2)* 3.6 x 10 (SE1) ...... 34

Table 5. Net and gross primary production and respiration rates normalized to cell counts ...... 34

Table 6. Calculation parameters for theoretical growth rates. Theoretical growth rates were calculated based on (Losh et al., 2013), using a calculated Cmin based on the following -7 equation: Cmin = ((NPPc/(8×3×12×(2/5)×2×3600×1000000))/6.06×10 )/M (Eq. 9) ...... 35

Table 7. Calculation of theoretical growth rates. Theoretical growth rates were calculated based on (Losh et al., 2013), using a calculated Cmin from eq. 9 substituted into the following -1 min equation: µtheoretical(d ) = (R×(C ×X))/Rubsat (Eq. 10) ...... 35

vi LIST OF FIGURES

Figure 1: Diagram showing expected changes in the marine environment with climate change. Compared to present conditions (Panel A), in the future ocean (Panel B), the mixed layer depth is expected to decrease, due to a weakening of mixing/strengthening of stratification due to increased temperature in the upper ocean. This will lead to a decreased upward mixing of remineralized nutrients, such as nitrate and phosphate, as well as an increase in the relative light intensity experienced by phytoplankton in the upper ocean ...... 2

Figure 2: Schematic showing the competing carbon fixation and photorespiration reactions which can be catalyzed by the Ribulose-1,5-bisphosphate carboxylase/oxygenase (RubisCO) enzyme ...... 5

Figure 3: Schematic showing CCM processes in a dinoflagellate cell containing a chloroplast with pyrenoid, where CO2 is concentrated in the proximity of RubisCO. External and internal - carbonic anhydrases (eCA and iCA) speed up the slow interconversion of CO2 and HCO3 and their activity is represented by long dashed lines. Short dashed lines indicate passive transport, as CO2 can freely diffuse through cell membranes. Solid lines with shaded boxes - indicate active transport, which can occur with both CO2 and HCO3 ...... 6

Figure 4. Growth rates of K. brevis based on cell count in the different pCO2 acclimations as well as the unbubbled control. Data shown are mean values (n > 20, ± SD). Statistical differences and groupings determined via ANOVA and Tukey’s HSD indicated using letters (p-values < 0.05) ...... 22

-1 Figure 5. Brevetoxin (pg cell ) for K. brevis in the different pCO2 treatments. Data shown are mean values (n > 3, ± SD). Statistical differences and groupings determined via ANOVA and Tukey’s HSD indicated using letters (p-values > 0.05) ...... 22

Figure 6: Photophysiology of K. brevis; Dark shaded areas indicate the night time. A) Diurnal trends of Fv/Fm values between the different pCO2 acclimations. Please note the y-axis magnification. B) GPPFRRf values -data during dark hours indicate photosynthetic potential. C) Nonphotochemical quenching (NPQ) for the different pCO2 acclimations. D) tES for the different pCO2 acclimations. Data shown are mean values (n ≥ 1, ± SD). Circles represent 150 µatm acclimation, Squares represent 400 µatm acclimation, and Triangles represent 780 µatm acclimation ...... 23

Figure 7: Nonphotochemical quenching (NPQ) at A) 9AM B) 12PM C) 3PM over the FLC (full light curve). Data shown are mean values (n ≥ 3, ± SD). Circles represent 150 µatm acclimation, Squares represent 400 µatm acclimation, and Triangles represent 780 µatm acclimation ...... 24

Figure 8: Net O2 evolution rates of K. brevis in different pCO2 acclimations measured continuously throughout a 24-hour period. Non-shaded areas indicate periods of light and data shown is net primary productivity (NPPO). Shaded areas indicate dark periods (night) and lines in these areas are representative of respiration rate (R). Respiration rates are shown as positive values for clarity. Data shown are mean values (n > 3, ± SD). Error bars are vii staggered by 30 min. for clarification, with the left most error bar representing 150 µatm, middle error bar representing 400 µatm, and right most error bar representing 780 µatm .....25

Figure 10: Net primary productivity measured and calculated using different parameterization. NPPC from oxygen evolution converted to C-fixation; NPPFRRf using FLC data from 12pm at -2 -1 121 µmol photons m s and equations from Lawrenz et al. (2013); NPPµ calculated using -1 -1 growth rate (µ (d )) and POC cell ; NPPLosh calculated based on Losh et al. (2013) using growth rate (µ (d-1)), R/GPP ratio, and protein cell-1 data ...... 30

viii ABSTRACT

Karenia brevis is a marine dinoflagellate commonly found in the Gulf of Mexico and important both ecologically and economically due to its production of the neurotoxin brevetoxin, which can cause respiratory illness in humans and widespread death of marine animals. K. brevis strains have previously shown to be sensitive to changes in CO2, in terms of growth and toxin production. Our study aimed to understand this sensitivity by measuring underlying mechanisms, such as photosynthesis, carbon acquisition, and photophysiology. K. brevis (CCFWC-126) did not show a significant response in growth, cellular composition of carbon and nitrogen, nor in photosynthetic rates between pCO2 concentrations of 150, 400 or 780 µatm. However, a strong response in its acquisition of inorganic carbon was found. Half saturation values for CO2 - increased from 1.5 to 3.3 µM, inorganic carbon preference switched from HCO3 to CO2 (14% to

56% CO2 usage), and external carbonic anhydrase activity was downregulated by 23% when comparing low and high pCO2. We conclude that K. brevis must employ an efficient and regulated carbon concentration mechanism (CCM) to maintain constant carbon fixation and growth across pCO2 levels. A positive correlation with pCO2, although not statistically significant, in cellular brevetoxin content was found. This study is the first explaining how this socioeconomically important species is able to efficiently supply inorganic carbon for photosynthesis which can potentially prolong bloom situations. This study also highlights that enhanced CO2, as projected for a future ocean, can affect underlying physiological processes of K. brevis, some of which could lead to increases in cellular brevetoxin production and therefore increased impacts on coastal ecosystems and economies.

ix CHAPTER 1

GENERAL INTRODUCTION

Atmospheric CO2 is predicted to rise to concentrations of 1000 µatm by the end of the 21st century (Solomon et al., 2007). Emissions from anthropogenic activities, including fossil fuel burning and land use change, are the main driver of the observed increase in atmospheric

CO2 (Pachauri et al., 2014). The global ocean has absorbed roughly 30% of the anthropogenically emitted CO2, land plants and soils have also absorbed 30%, and the remaining

40% remains in the atmosphere and contributes to the increase in atmospheric CO2 concentration

(Pachauri et al., 2014). The absorption of CO2 by the ocean through dissolution, known as the solubility pump, accounts for the majority of CO2 uptake into the ocean. When CO2 dissolves into water, it forms carbonate species and dissociates further, releasing hydrogen ions, which causes a concomitant decrease in pH, a process known as “ocean acidification”. The elevated

CO2 concentration in surface seawater has led to a decrease of 0.1 pH units since pre-industrial times, and this reduction is expected to reach 0.4 pH units by the end of the 21st century (Pachauri et al., 2014). Ocean acidification has wide reaching impacts on marine ecosystems, affecting many trophic levels, including marine pelagic photoautotrophic organisms, known as phytoplankton. Phytoplankton make up the base of the marine food web, conduct important ecosystem functions, and drive biogeochemical cycles (Winder and Sommer, 2012). While phytoplankton only make up around 0.2% of the biomass of global primary producers, they are responsible for roughly half of Earth’s primary production (Field et al., 1998). Ocean acidification has the ability to directly affect the health and physiology of phytoplankton species and changes to CO2 concentrations can directly affect photosynthesis. Phytoplankton are expected to respond to changes in pCO2 by altering growth rates, which will in turn effect competition and community structures (Mackey et al., 2015).

However, it is not only the effects stemming directly from increased atmospheric CO2 concentrations that have the ability to affect phytoplankton physiology and ecology. Since CO2 is a greenhouse gas, its increase has led to an increase in heat in both the atmosphere and upper ocean. In the upper 75 meters of the ocean, average temperatures have risen by 0.11 °C between 1971 and 2010 and are predicted to rise another 1.8-4 °C by the end of the 21st century (Meehl et al., 2007; Pachauri et al., 2014). This rise in temperature will result in a shoaling of the mixed 1 layer, which leads to a reduction in nutrient availability in the upper euphotic layer and changes to light regimes experienced by phytoplankton (Boyd et al., 2016). A schematic of expected environmental changes can be seen in Figure 1. In addition to changing nutrient and light dynamics, increased temperatures in surface seawater have a direct effect on phytoplankton physiological processes, namely growth and respiration rates.

Present Conditions Future Conditions A B

Euphotic Zone Euphotic Zone Relative Light Intensity Light Relative Relative Light Intensity Light Relative 3- 3- PO4 PO4 - - NO3 NO3 Mixed Layer Depth

3- - PO4 NO3 3- PO4 - Mixed Layer Depth NO3 - NO3

- 3- NO3 PO4 - 3- 3- NO3 PO4 PO4 PO 3- NO - 4 3- - 3 - PO4 NO3 3- NO3 PO4

It is predicted that continued anthropogenic emissions will lead to increases in both average global temperatures and the occurrence of arid regions (Pachauri et al., 2014). Currently, during the dry summer months in the northern hemisphere, large amounts of Saharan dust is carried via air streams from West Africa and transported across the Atlantic Ocean to its final destinations, such as the Caribbean and the eastern United States, including the Gulf of Mexico (GoM) (Lenes et al., 2012, 2001). Increases in temperature as well as in the size of arid regions will likely lead to increases of these dust deposition events, which is relevant to phytoplankton as aerosols may be an important source of iron in oligotrophic regions (Lenes et al., 2001). For

2 example, iron input from dust events from the Sahara in Northern Africa can spur blooms of phytoplankton in the Gulf of Mexico (Walsh et al., 2006).

Overall, increased atmospheric CO2 concentrations can lead to a wide range of climate change effects which will have profound impacts on phytoplankton physiology and ecology.

Thus, it is imperative to understand how a changing climate resulting from anthropogenic CO2 emissions will affect phytoplankton and their regulation of global climate and biogeochemical cycles. Over the last few decades, responses of phytoplankton to those predicted changes have been shown to be diverse and multifaceted. Responses to alterations in environmental conditions have been found to vary down to the species level, and furthermore, strains of the same species sometimes display different responses to similar stressors (Hutchins et al., 2013; Langer et al., 2011; Mackey et al., 2015; Petrou et al., 2016; Schaum et al., 2013). It has been speculated that the phytoplankton group of particular interest to this study, dinoflagellates, will become more abundant in the future ocean. Dinoflagellates are a diverse group of eukaryotic algae that thrive in marine and freshwater ecosystems and are estimated to make up around 40% of the total species of marine phytoplankton (Simon et al., 2009). While only relatively few dinoflagellate species are known to cause “harmful algal blooms” (HABs), the majority of HAB occurrences are caused by dinoflagellates as they are the major group producing toxins that impact humans (Glibert et al., 2005; Moore et al., 2008; Wang, 2008). Dinoflagellates are known to employee a variety of metabolisms and life styles which help this group’s success and ability to adapt to environmental change. Dinoflagellates are thought to become more abundant in the future as some members possess the ability to swim to access nutrients below a shallower mixed layer. Additionally, dinoflagellates span the full spectrum with regards to metabolic strategies, with strict photoautotrophs and strict heterotrophs, while some have mixotrophic metabolic strategies which can reduce their dependency on inorganic nutrients. Another hypothesis for their increased abundance is that dinoflagellates will directly benefit from enhanced CO2 by down-regulating the energetically expensive processes required to acquire inorganic carbon for photosynthesis. These energetic savings could be used by cells to increase growth rates, cellular toxin production, alter swimming abilities, and even enhance support for a host in a symbiotic relationship. Yet, despite the importance of dinoflagellates in the marine environment, relatively few studies have focused on the response of this group to changes in atmospheric CO2 (e.g. Hansen et al., 2007; Fu et al., 2010; Eberlein et al., 2014; Hardison et al., 2014; Magaña and Villareal, 2006; Maier Brown et al., 2006; Errera et al., 2010;

3 Lekan and Tomas, 2010; Errera and Campbell, 2011; Hardison et al., 2012, 2013; Errera et al., 2014; Hoins et al., 2016). Efficient photosynthesis is vital for marine photoautotrophic dinoflagellates to form and sustain large blooms. Pigments found in dinoflagellates include chlorophylls a and c2, beta- carotene, and a group of xanthophylls (including peridinin, dinoxanthin, and diadinoxanthin) which are responsible for the reddish-brown color of blooms (Hackett et al., 2004). However some dinoflagellates possess additional pigments, such as fucoxanthin, which were acquired through endosymbiosis (Hackett et al., 2004; Yoon et al., 2002, 2005). While pigments are important for the light reactions in photosynthesis, what makes dinoflagellates particularly interesting is their cellular machinery used in the photosynthetic dark reactions. The majority of carbon fixation done by phytoplankton is done using the C3 pathway (the photosynthetic carbon reduction cycle, or Calvin cycle) for inorganic carbon acquisition, with inorganic carbon being directly fixed by Ribulose-1,5-bisphosphate carboxylase/oxygenase (RubisCO) in the dark reactions (Badger et al., 1998; Giordano et al., 2005; Tortell, 2000). Most photoautotrophic dinoflagellates contain a very inefficient and unique form II RubisCO (Badger et al., 1998; Morse et al., 1995; Tortell, 2000). However, it has been found that some dinoflagellate species, including the species of interest in this study, Karenia brevis, possess the more efficient form I RubisCO, which is similar to that of haptophytes (Yoon et al., 2002, 2005). Phylogenies of K. brevis based on photosystem I and RubisCO genes have shown K. brevis to be closely related to haptophytes such as Emiliania huxleyi, Pavolva lutheri, or Prymnesium pervum (Yoon et al., 2005, 2002). Hence, it can be speculated that K. brevis might respond differently to changes in inorganic carbon concentrations compared to other dinoflagellates. Despite the increase in efficiency in form I RubisCO over form II RubisCO, they are both inherently inefficient regarding their affinity to CO2, with half saturation concentrations which are higher than the CO2 equilibrium concentrations in seawater (Giordano et al., 2005; Tortell, 2000). In addition to its low affinity for CO2, RubisCO’s inefficiency is also hindered by its oxygenase function, by which it reacts with O2 and photorespiration occurs, which is shown in Figure 2 (Giordano et al., 2005). The biochemical constraints imposed by the RubisCO enzyme thus demands for aquatic phytoplankton, which rely on the diffusive uptake of CO2 for carbon fixation, to be able to overcome these inefficiencies in order to be productive and capable of forming dense blooms. To do this, most phytoplankton species have so-called CO2 concentrating mechanisms (CCMs).

Figure 2: Schematic showing the competing carbon fixation and photorespiration reactions which can be catalyzed by the Ribulose-1,5-bisphosphate carboxylase/oxygenase (RubisCO) enzyme.

CCMs in phytoplankton can include physical compartmentalization and cellular processes that concentrate inorganic carbon around the enzyme RubisCO, and examples of these processes in dinoflagellates are shown in Figure 3. RubisCO in phytoplankton is typically concentrated in specialized compartments, known as pyrenoids in eukaryotic algae or carboxysomes in cyanobacteria. Previous studies on K. brevis have shown strong evidence for the existence of pyrenoids (Monroe et al., 2010). Cellular processes that constitute the CCM - include the active, energy dependent transport of CO2 and/or HCO3 into the cell, utilizing carbonic anhydrases (CAs) which speed up the slow interconversion of carbon species in aqueous solution, as well as reducing the efflux of CO2 from the cell (Giordano et al., 2005; Raven and Johnston, 1991). All of these processes require cellular energy and resource allocation. Consequently, it can be speculated that an increase in CO2 concentration could have the potential to decrease the need for cells to maintain a CCM. For example, cells could reduce - the active pumping of HCO3 into the cell, would have a reduced requirement to express CAs, as well as decreased RubisCO content relative to total proteins in a cell. This down-regulation of cellular CCMs would allow for energy and resources to be allocated to other physiological processes in phytoplankton, such as growth or toxin production (Giordano et al., 2005; Kranz et al., 2011; Mackey et al., 2015; Rost et al., 2008).

Figure 3: Schematic showing CCM processes in a dinoflagellate cell containing a chloroplast with pyrenoid, where CO2 is concentrated in the proximity of RubisCO. External and internal - carbonic anhydrases (eCA and iCA) speed up the slow interconversion of CO2 and HCO3 and their activity is represented by long dashed lines. Short dashed lines indicate passive transport, as CO2 can freely diffuse through cell membranes. Solid lines with shaded boxes indicate active - transport, which can occur with both CO2 and HCO3 .

In this study, we investigated the response of the dinoflagellate Karenia brevis to changes in CO2 concentration. K. brevis is an unarmored, photosynthetic marine dinoflagellate commonly found in GoM waters and is capable of forming dense blooms, known as red tides (Brand et al., 2012). K. brevis has been found in the field under a range of temperatures (7 to 33 °C), but optimal growth temperatures for laboratory cultures are between 22 and 28 °C (Brand et al., 2012). It has been found in laboratory studies that attempts to grow K. brevis in temperatures over 30 °C resulted in reduced growth, or even cell death (Errera et al., 2014; Magaña and Villareal, 2006). Optimal salinity conditions for K. brevis growth are between 30 and 35, but it has been found to grow between 18 and 45, with its blooms typically found in coastal waters but not estuaries (Brand et al., 2012). Field and laboratory studies have shown that K. brevis can utilize a range of nutrient sources, including urea, glutamate, and dissolved organic matter exuded from Trichodesmium spp. to support the nitrogen demand for growth and reproduction (Brand et al., 2012; Bronk et al., 2014; Killberg-Thoreson et al., 2014; Mulholland et al., 2014). It has additionally been found that K. brevis can utilize glycine, valine, and methionine, as well as organic phosphorus through the use of alkaline phosphatase (Brand et al., 2012). Lastly, cells of Synechococcus have been found within K. brevis cells, which supposedly has led to increased growth rates in laboratory cultures (Brand et al., 2012; Glibert et al., 2009). Overall, K. brevis is

6 able to utilize a wide range of nutrient sources to support growth, which allows this species to thrive in oligotrophic environments like the GoM. The ability of K. brevis to survive accumulated in bloom situations at the surface ocean in the GoM is also bolstered by its ability to adapt to a wide range of light environments. Measured compensation points for K. brevis clones have fallen between 6-8 µmol photon m-2 s-1 and saturation points in the range of 35-120 µmol photon m-2 s-1 in published studies (Brand et al., 2012 and references therein). A study by Schaeffer et al. (2007) using 10 different K. brevis isolates showed significant differences in the photosynthetic characteristics between their clones and found growth under a wide range of light intensities. The ability of K. brevis to adapt to higher light intensities is likely due to the use of the photoprotective xanthophyll cycle, in which diadinoxanthin and diaxanthin are used to dissipate energy (Brand et al., 2012). In addition, it has been hypothesized that brevetoxin may facilitate non-photochemical quenching and therefore allow K. brevis to mitigate light stress (Cassell et al., 2015). K. brevis typically reproduces asexually through binary fission once every 2-10 days, primarily at night, with the diel phased timing based on gene regulation (Brand et al., 2012; Van Dolah et al., 2009). While binary fission is their main form of reproduction, K. brevis has also been shown to produce planozygotes, which indicate the capability of this species for sexual reproduction, and the production of these gametes is sensitive to environmental controls (Brand et al., 2012). K. brevis has been shown to exhibit both geotaxis and phototaxis, and can swim at a speed of approximately 1 m/h using two flagella over a wide range of environmental conditions (McKay et al., 2006). Studies by Heil (1986) showed accumulation at the surface during the day, with upward swimming before the light period started, and a downward dispersal into the water column at night, with downward swimming beginning before the light period ended. The swimming observed in K. brevis has been hypothesized to synergize with hydrographic features which would allow for concentration of K. brevis cells independent of growth, and therefore serve as a mechanism for bloom initiation (Brand et al., 2012 and references therein). Studies looking at K. brevis in the GoM have found long-term increases in the occurrence of this dinoflagellate with K. brevis being roughly 15 fold more abundant in the years of 1994- 2002 than it was between 1954-1963 (Brand and Compton, 2007). Blooms of K. brevis occur annually in the GoM and are one of the most predictable harmful alga blooms on the planet (Heil et al., 2014). Blooms of K. brevis are particularly harmful due to their production of two types of lipid soluble toxins: hemolytic and neurotoxic (Wang, 2008). These toxins cause mortality of

7 marine mammals, fish, and other marine life, neurotoxic shellfish poisoning (NSP), and respiratory illness in humans (Flewelling et al., 2005; Landsberg, 2002; Wang, 2008) The neurotoxins are known as brevetoxins and are a class of ladder-like polycyclic ether toxins which are directly responsible for massive fish kills and nausea, loss of motor control, and sever muscular pain in humans (Wang, 2008). Hence, the occurrence of K. brevis blooms have sizeable impacts on the local environment and therefore have large ecological and economic effects (Fleming et al., 2005; Hoagland et al., 2009; Landsberg, 2002). During blooms, K. brevis can account for roughly 20% of total primary production on the West Florida Shelf, and is present yearlong, making it an important local carbon sink as well as an abundant food source for higher trophic levels (Vargo, 2009). Therefore, the impact of this species on the local economy and human health within the GoM necessitates the assessment of how environmental changes will affect this species. Additionally, identifying the mechanisms driving the responses can help to identify limitations and trade-offs occurring in physiological pathways, which is key for determining the competitive abilities of a specifies. While there have been a number of studies on K. brevis in the GoM done to understand how different environmental parameters can affect growth, productivity, and toxicity, only a few studies have investigated the response of this economically and ecologically important species to changes in CO2 concentration (Brand et al., 2012; Hardison et al., 2012, 2013). The studies which have looked at the CO2 sensitivity of K. brevis have found rather conflicting results. A substantial increase in growth, yet no changes in cellular toxin production were found at enhanced CO2 in Errera et al. (2014). Additionally, this study found a general trend of decreasing growth rates with increasing temperatures, with the exception of the high CO2 treatment. At high

CO2 concentrations, the positive CO2 effect on growth was able to compensate for the negative effect of increased temperature (Errera et al., 2014). In contrast, Hardison et al. (2014) found reduced growth rates and enhanced toxicity under reduced CO2, but did not test future CO2 scenarios. These results, in particular the opposing trends in brevetoxin production under low/high CO2, suggests the need for additional studies to gain a mechanistic understanding of the response of K. brevis to enhanced, as well as reduced, CO2 concentration. Despite the research done on this species and the high predictability of its bloom occurrences, remarkably little is known on some key features of the photoautotrophic species, including the modes of inorganic carbon acquisition, which can help explain and predict how this species is able to thrive during dense blooms as well as how it will respond to increased CO2 in a future ocean.

8 It is the goal of this study to characterize the response of K. brevis to CO2 levels that are representative of situations when CO2 is reduced as occurs during dense blooms in the environment, as well as the current and projected future atmospheric concentrations. To accomplish this, K. brevis was grown under three CO2 treatments and parameters such as growth rate, photosynthetic oxygen evolution, elemental composition, and brevetoxin production were measured. In order to understand the underlying mechanisms of these responses, the efficiency of inorganic carbon acquisition (kinetics and inorganic carbon source preferences, CA activity, and RubisCO saturation) and photophysiological processes were investigated. Additionally, a sensitivity analysis of the CCM requirements for this species was conducted.

9 CHAPTER 2

INSIGHTS INTO CARBON ACQUISITION AND PHOTOSYNTHESIS IN KARENIA BREVIS UNDER A RANGE OF CO2 CONCENTRATIONS

1. Introduction

st By the end of the 21 century, atmospheric CO2 is expected to increase to 1000 µatm

(Solomon et al., 2007). This increase in CO2 will result in a reduction of seawater pH of up to 0.4 units, a process called ‘ocean acidification’. In addition, a rise in temperature result has been predicted to result in a shoaling of the mixed layer, and a reduction in nutrient availability in the upper layer (Boyd and Doney, 2003). It has been speculated that dinoflagellates could become more abundant in the future ocean, as some species might benefit from projected changes in the marine environment. For example, dinoflagellates have the ability to swim and access nutrients below a shallower mixed layer (Glibert et al., 2005; Moore et al., 2008), use mixotrophic metabolic strategies to reduce dependency on inorganic nutrients (Stoecker et al., 2017), and could benefit from enhanced CO2 directly. Yet, despite the importance of pelagic dinoflagellates in the marine environment, relatively few studies have focused on the response of this group to future environmental changes (e.g. Hansen et al., 2007; Fu et al., 2010; Eberlein et al., 2014; Hardison et al., 2014; Magaña and Villareal, 2006; Maier Brown et al., 2006; Errera et al., 2010; Lekan and Tomas, 2010; Errera and Campbell, 2011; Hardison et al., 2012, 2013; Errera et al., 2014; Hoins et al., 2016). It is important to understand the response of marine pelagic photoautotrophic organisms to the environmental changes a future ocean will face, as these organisms make up the base of the marine food web, conduct important ecosystem functions, and drive biogeochemical cycles.

Over the last decades, responses of phytoplankton to temperature, nutrient availability, CO2, and ocean acidification have been characterized and found to be diverse and multifaceted, varying even between strains of the same species (Langer et al., 2011; Hutchins et al., 2013; Schaum et al., 2013; Petrou et al., 2016; Mackey et al., 2015). Obligate photoautotrophic organisms, such as many marine phytoplankton, require the - inorganic forms of carbon (CO2 or bicarbonate (HCO3 )) to build organic molecules such as sugars, amino acids, etc. The key enzyme for the conversion of CO2 into organic carbon within the Calvin-Benson cycle is Ribulose-1,5-bisphosphate carboxylase/oxygenase enzyme (RubisCO). Different forms of RubisCO have evolved over time, some of which are more or less 10 efficient in their ability to utilize CO2 (Raven et al., 2012; Tortell, 2000). Most photoautotrophic dinoflagellates contain a very inefficient type II RubisCO (Morse et al., 1995; Badger et al., 1998; Tortell, 2000). However, it has been found that some dinoflagellate species including the species of interest in this study, Karenia brevis, possess the more efficient type I RubisCO which is similar to that found in haptophytes (Yoon et al., 2005, 2002). K. brevis is thought to have acquired type I RubisCO via tertiary endosymbiosis of a haptophyte as phylogenetic studies based on RubisCO and photosystem I genes show a close relationship between K. brevis and the haptophytes Emiliania huxleyi, Pavolva lutheri, and Prymnesium pervum (Yoon et al., 2005, 2002). Hence, K. brevis might respond to changes in inorganic carbon availability more like haptophytes rather than other type II RubisCO containing dinoflagellates (Yoon et al., 2005). Nonetheless, both type I and II RubisCO are inherently inefficient as all RubisCO catalyze reactions for photorespiration using O2 in addition to carbon fixation using CO2. Its proclivity to catalyze either the carboxylation or oxygenation reaction represents a challenge for efficient carbon fixation in all photosynthetic organisms and is described as the substrate specificity factor

(Srel*). The substrate specificity factor for dinoflagellates with type II RubisCO is relatively low

(Srel* of 37 in Amphidinium carterae (Badger et al., 1998)), however, the type I RubisCO found in K. brevis potentially has a higher specificity factor (Srel* of 77-79 in Emiliania huxleyi

(Badger et al., 1998)). RubisCO also has a low affinity for CO2, with half saturation concentrations (Km) mostly higher than the CO2 concentration found in ambient marine environments (Km values of 6-185 µM in algae and between 14-114 µM in haptophytes and dinoflagellates) (Badger et al., 1998; Heureux et al., 2017; Tortell, 2000). To compensate for the poor efficiency of RubisCO, most species evolved so-called carbon concentrating mechanisms (CCMs). CCMs in general include pathways to increase the

CO2 concentration in the proximity of RubisCO compared to the concentration in the seawater. - Processes include the active, energy dependent transport of CO2 and/or HCO3 into the cell, the presence of carbonic anhydrases (CAs; which speed up the slow interconversion of carbon species), the reduction of efflux of CO2 from the cell back to the seawater, and specialized compartments in which RubisCO is concentrated, so called pyrenoids in eukaryotic algae or carboxysomes in cyanobacteria (Giordano et al., 2005; Raven and Johnston, 1991). Previous studies have found strong evidence for pyrenoids in K. brevis strains (Monroe et al., 2010), indicating that this species has structural components of a CCM. While a simple assumption for the existence of a CCM is that the half saturation concentration of photosynthesis to CO2 is lower

11 than the half saturation concentration of the RubisCO to CO2 (Giordano et al., 2005; Tortell, 2000), those mechanisms have not been investigated for K. brevis yet. The CCM can, in some instances, be a costly process since it requires active, energy driven uptake of inorganic carbon and the expression of a variety of enzymes (transporters, CAs) (Giordano et al., 2005). Despite a potentially large energy and resource requirement, the net benefit usually outweighs the costs, as cells would otherwise have high photorespiration and low CO2 fixation capability. It has been speculated that an increase in CO2 concentration could potentially decrease the need for CCMs - (e.g. acquire CO2 via diffusive uptake rather than active pumping of HCO3 , reduced requirement to express CAs) (Mackey et al., 2015 and references within). This down-regulation could potentially allow for energy and resources to be allocated to other physiological processes, such as growth, nitrogen acquisition (Giordano et al., 2005; Rost et al., 2008; Kranz et al., 2011; Mackey et al., 2015), and likely the production of secondary metabolites such as toxins. Despite the suggestion that dinoflagellates might benefit directly from enhanced dissolved CO2 concentrations in a future ocean, studies on growth or elemental composition responses in carefully conducted CO2 experiments indicate that responses are diverse. For example, Hoins et al. (2016) did not find any growth response to different low or high CO2 concentrations in the dinoflagellates Gonyaulax spinifera (grown under low light), Protoceratium reticulatum, Alexandrium fundyense, and Scrippsiella trochoidea (grown under nitrogen limitation), yet this study found a decreased growth rate in S. trochoidea with increased

CO2 under nitrogen-replete conditions. Eberlein et al. (2014) demonstrated that the dinoflagellates Alexandrium tamarense and S. trochoidea were unaffected by changes in pCO2 in their growth rates, yet S. trochoidea was sensitive to enhanced CO2 in its elemental composition. In our study, we investigated the responses and underlying key physiological processes of the dinoflagellate K. brevis to changes in CO2 concentration. K. brevis, a photosynthetic marine dinoflagellate commonly found in Gulf of Mexico (GoM) waters, is capable of forming dense blooms, also known as red tides, and able to produce brevetoxins, a type of neurotoxin (Brand et al., 2012). Blooms of K. brevis occur annually in the GoM and are one of the most predictable harmful algal blooms on our planet (Heil et al., 2014). The production of brevetoxin results in mortality of marine mammals, fish, and other marine life, and also causes respiratory illness in humans (Landsberg, 2002; Flewelling et al., 2005). Hence, the occurrence of K. brevis blooms have large impacts on the local environment and human health, and have sizeable environmental and economic effects (Landsberg, 2002; Fleming et al., 2005; Hoagland et al., 2009).

12 Due to the impact of this species on the local economy and human health within the GoM it is necessary to assess how environmental changes will affect this species. Many studies on K. brevis have been conducted in order to understand how different environmental parameters can effect growth, productivity, and toxicity (Steidinger, 2009; Brand et al., 2012; Hardison et al., 2012, 2013). Despite the research done, the high predictability of its occurrence, and impact on local economy, surprisingly little is known about some key features of this photoautotrophic species, e.g. the modes of inorganic carbon acquisition, which can explain and predict how this species is able to thrive during dense blooms or how it will respond to enhanced CO2 in a future ocean. Only a few studies have investigated the response of this important species to changes in

CO2 concentrations (Errera et al., 2014; Hardison et al., 2014). Briefly, Hardison et al. (2014) investigated three strains of K. brevis under ambient and reduced CO2 (CCMP 2228, CCMP 2229, and SP3) and found reduced growth and enhanced brevetoxin per cellular carbon under reduced CO2. Errera et al. (2014), using two different strains of K. brevis (SP1 and Wilson clone), tested low, ambient, and high CO2 concentrations and found a substantial increase in growth with elevated CO2, but no changes in toxin production were detected. While different strains and additional parameters, such as temperature, were tested in these studies, the opposing

CO2 dependent trends in brevetoxin production and an omitted mechanistic understanding of the responses calls for additional investigation of the response including studies on key physiological processes such as photosynthesis and carbon acquisition to a range of CO2 concentrations. Identifying the response of underlying physiological mechanisms can identify limitations and trade-offs in physiological pathways, which are key for determining the competitive abilities of this species. It is the goal of this study to provide a process-based investigation to better understand the previously measured physiological responses of K. brevis to CO2 levels representative of current (400 µatm), projected future atmospheric concentrations (780 µatm), and conditions with reduced CO2 (150 µatm, e.g. during dense blooms). Growth, elemental composition, and brevetoxin production responses to the changes in CO2 concentrations were measured. In order to understand those physiological responses, photosynthesis and photophysiology, and inorganic carbon acquisition mechanisms (kinetics and inorganic carbon source preferences, carbonic anhydrase activity, RubisCO saturation) were studied.

13 2. Materials and Methods

2.1 Culture conditions: Karenia brevis (CCFWC-126), isolated from the Gulf of Mexico (Tampa area) and provided to us by the Florida Fish and Wildlife Commission, was grown in semi-continuous batch approach at 26 °C in 1L polycarbonate bottles with a 12:12h light:dark cycle at 100 µmol photons m-2 s-1 (GE Daylight Ecolux® T12). Light intensities were sub-saturating (see section 3.4) but chosen in order to be able to compare data with existing literature. Cells were grown in 0.2 µm filtered unbuffered modified Aquil media with f/2 vitamin recipe (Guillard and Ryther, 1962; Price et al., 1989). Modifications of the media recipe included changes in concentration of Phosphate, Zinc, Manganese, and Copper, added to final concentrations of 1.5×10-5 M., 1×10-7 M, 2×10-8 M, 1.0×10-8 M, respectively. Vanadium and Chromium were added to final concentrations of 1×10-8 M. No silica was added to the media. The modifications were chosen as we determined that cells grew better in this modified medium compared to standard Aquil and L1-Si recipes. Air containing 150, 400, and 780 µmol CO2 was continuously sparged through the cultures, with care taken to maintain bubbling at low rates of ~4 mL min-1 as dinoflagellates have been found to be sensitive to fast bubbling in this experiment and others (van de Waal et al., 2014). CO2 concentrations were obtained by using mass flow controllers (Alicat Scientific) to mix CO2-free air (CO2-PG80, Pure Air) with pure

CO2 (Airgas), or by using ambient room air (~400 µatm). Mid-experiment, the ambient culture was switched to 400 µatm using an additional line in the gas-mixing system. Cultures were acclimated for at least 7 generations at the respective CO2 before being used for any experiments or before growth rates were determined (approximately 4 weeks of acclimation). Cultures were maintained in triplicate. A separate un-bubbled control culture kept at ambient pCO2 (~400 µatm) was used to check for a mechanical bubbling effect on growth rates. Regular dilutions with pre-acclimated media rigorously bubbled with target pCO2 for several days assured an equilibrated carbonate chemistry. Cells were able to grow exponentially up to 10 days before reaching stationary phase but, were diluted within 7 days as the carbonate chemistry started to shift (by 0.1 pH unit increase).

2.2 Carbonate chemistry: Air pCO2 was verified using a LI-820 CO2 analyzer (LI-COR, Lincoln, Nebraska, USA). Dissolved inorganic carbon (DIC) was measured in culture media following a modified protocol from Noguchi et al. (2013). Measured standards included water - with 3.4% NaCl bubbled with CO2-free air, freshly prepared HCO3 standards with known final 14 concentrations, preacclimated culture media prior to the addition of cells, as well as culture media from the mid-exponential growth phase. For the latter, cells were removed via filtration using an inline filter and a peristaltic pump. Exact methods for DIC analysis can be found in the supplemental materials (Method S1). Carbonate chemistry in the cultures was monitored daily by measuring culture pH (Zhang and Byrne, 1996). When cultures drifted by more than 0.1 pH, cells were immediately diluted with pre-acclimated medium and were not used for experiments for several generations. Carbonate chemistry was calculated using CO2 sys (Pierrot, et al., 2011)

3- with input parameters: Temp 26 °C, Salinity 35, PO4 50, Si 0, DIC (as measured), pH (as measured and converted to NBS scale). CO2 constants K1, K2 from Mehrbach et al. (1973) refit by Dickson and Millero (1987) were chosen, KSO4 source was by Dickson (1990), and total boron source was used as defined by Uppström (1974) .

2.3 Growth, chlorophyll a, elemental composition, brevetoxin analysis, protein concentration: Specific growth was measured via cell count (using a Coulter Counter Z2 (Beckman, Indiana, USA, size detection 12 and 30 µM)) and relative chlorophyll fluorescence (Relative Fluorescence Units) (Trilogy, Turner Design, California, USA) throughout the experimental phase. Chlorophyll a cell-1 (Chl a) was determined by filtering cells via gravity filtration onto a GF/F filter. In general, all filter samples were taken using gravity filtration to avoid breaking of the cells. Filters were stored immediately at -20°C or -80°C until further analysis. Chlorophyll a was extracted in 90% acetone for 24 hours with subsequent measurement using a UV/VIS spectrophotometer (Evolution 220, ThermoFisher, Massachusetts, USA) at the wavelengths 750, 663, 645 and 630 nm (following the ESS Method 150.1 (µg chl a L-1 = S [11.85 (Abs664) – 1.54 (Abs647)- 0.08 (Abs630)]/V; Where S = the volume of acetone used for the extraction (mL), V = The volume of water filtered, L = The cell path length (cm)). For particulate organic carbon (POC) and nitrogen (PON) analysis, 100 mL of culture were filtered onto precombusted GF/F filters (5 hours – 500°C), acidified, dried and subsequently analyzed using a continuous flow Isotope Ratio Mass Spectrometer by the UC Davis Stable Isotope Lab. Blanks were taken for each measurement using culture media prior to the addition of cells. Brevetoxin samples were extracted following Roth et al. (2007) and measured using a brevetoxin ELISA kit which specifically measures PbTx-2 and PbTx-3 (Abraxis Inc, Warminster, PA, USA). Exact methods for brevetoxin extraction can be found in the supplemental materials

15 (Method S2). Total cellular protein was determined using the BCA protein assay kit (Pierce, Thermo Scientific, Waltham, MA, USA).

2.4 Photophysiological parameters: Photosynthesis vs. irradiance fluorescence induction light curves (FLC) were measured using a Fast Repetition Rate Fluorometer (FRRf, FastOcean PTX, Chelsea Technologies Group) along with a FastAct Laboratory system (Chelsea Technologies). Each FLC measurement lasted 1.5 hrs and cultures were measured continuously over a 24-hour period. Between FLCs, cultures were exchanged using the peristaltic pump controlled by the FRRF. Cells were acclimated to each light intensity for 3 minutes. Additional details on the FLC settings can be found in the supplemental materials (Method S3).

Photophysiological data such as Fo, Fm, Fo’, Fm’, Ek (light saturation parameter) sLHII (a measure of the absorption cross section of PSII), tES (the time constant for the re-opening of a closed RCII with an empty Qb site), and NPQ (a measure of nonphotochemical quenching) were obtained. Derivation of these photochemical variables can be found in Oxborough et al. (2012).

To calculate gross primary production (GPP) from FRRf data (GPPFRRf), we used data from the 121 µmol photons m-2 s-1 (closest light intensity to acclimation light measured in the FRRf) and the following equation from Lawrenz et al. (2013) to calculate electron transport rates (ETR) in units of mol electron- (mg Chl a)-1 h-1:

-5 ETR = E × sLHII’ × nPSII × (Fq’/Fv’) × ΦRC × 2.43 × 10 (1)

2 Where E is light intensity, sLHII’ is the absorption cross section of PSII in the light (Å -1 quanta ), nPSII is a conversion factor for reaction centers to chl a (mol reaction centers mol Chl a-1) with a value of 0.002 based on values obtained from eukaryotic phytoplankton (Kolber and

Falkowski, 1993; Raateoja et al., 2004), (Fq’/Fv’) is the light adapted quantum yield, ΦRC is the electron yield from each RCII charge separation equaling 1 based on Lawrenz et al. (2013), and 2.43 x 10-5 is used as a unit conversion factor (Lawrenz et al., 2013). To further convert ETR into -1 -1 -1 GPP (mol C mg Chl a h ), we used the conversation factor Φe,C (mol electrons mol CO2 ) of 10 (Lawrenz et al., 2013).

GPPFRRf = ETR × Φe,C (2)

To calculate net primary production from the FRRf (NPPFRRf), we used the GPPFRRf data and the carbon-based GPP/Respiration (GPPC/R) ratio data (see below) in the following equation:

NPPFRRf = GPPFRRf – (GPPFRRf × R/GPPC) (3)

16 2.5 Photosynthetic oxygen evolution: Photosynthetic oxygen evolution was measured using a FirestingO2 optical oxygen meter (Pyro Science, Germany). Cultures in exponential growth phase were concentrated using gravity filtration over a 10.0 µm PTFE filter and subsequently resuspended into the measuring bottles using fresh CO2 equilibrated media. Care was taken to not let cells dry out (a thin film of media covered the cells at any point) and Fv/Fm measurements prior to and after filtration revealed no significant change (Data not shown). Cells were subsequently placed into gastight oxygen optode bottles (respiration bottles, Pyro Science, Germany), kept at 26°C and illuminated by LED lights (Bright White LED Strip Lights, Cool White) at 100 µmol photons m-2 s-1. Cultures were stirred gently to ensure homogenous gas distribution as well as to avoid any potential cell clumping or settling. Measurements were started in the dark period and lasted 24 hours. Light-dark timing was set to the same cycle as the cells in the different pCO2 acclimations were exposed to. Additional three 30-minute dark periods were set to quantify dark respiration during the light phase. Light-dependent respiration (e.g. photorespiration, chlororespiration) was not analyzed. Data was fit using linear regressions in ~30-minute increments throughout the light period to determine high resolution net primary productivity rates (NPPO) and throughout the dark periods to determine dark respiration (R) rates. A sample of the cell concentrate was taken before and after the measurements and the average cell number and/or chl a concentration was used to normalize rates of O2 production/respiration. Dark respiration (RO-light) during the light phase was determined by linear extrapolation of the three oxygen consumption measurements during the light phase. Oxygen- based gross primary productivity (GPPO) was calculated as follows:

GPPO= NPPO-light - RO-light (4)

Where NPPO-light equals the rate of O2 evolution in the light.

In order to calculate gross productivity on a per carbon basis (GPPC), we used

NPPO and applied a photosynthetic quotient of 1.4 (for nitrate usage) as well as a respiratory quotient of 1 (Wiliams and Robertson, 1991).

GPPC= NPPO/1.4 -RO-night (5) As a second estimate of net carbon production, the calculated specific growth rate µ (d-1) was multiplied by C cell-1. -1 NPPµ = C cell × µ (6)

17 Theoretical calculation of carbon fixation, based on growth rate and estimated cellular RubisCO concentration were conducted as shown in Losh et al. (2013) using the growth rates, C cell-1 and protein cell-1 data (Table 4).

2.6 Carbon acquisition measurements: Cultures were concentrated using the same gravity filtration method described in photosynthetic oxygen evolution rate experiments. Cells were washed and resuspended in CO2-free Aquil media buffered with 50 mM BICINE adjusted to the pH of growth conditions. Care was taken to not introduce CO2 through bubble injection and cells were illuminated for up to 20 minutes to reduce potential introduced CO2. The 14C disequilibrium method was used to determine the steady state fraction of 14 - 14 H CO3 and CO2 uptake in the cells after the relatively high pH measurement media (7.9 – 8.3) is spiked with a relatively low pH (5.75 – 6.7) 14C solution (Espie and Colman, 1986; Rost et al., 2007; Kottmeier et al., 2014). Data analysis was done using Graphpad Prism 7 (GraphPad Software, La Jolla California USA). Carbon uptake kinetics were determined using the 14C fixation method described by 14 - Tortell et al. (2010) where cells were incubated with H CO3 for 10 minutes over a range of inorganic carbon concentrations (16, 40, 116, 166, 333, 662, 1316, 2042, 3196, and 3881 µM DIC). Cells were prepared as described above. For both disequilibrium and kinetics measurements, reactions were stopped by transferring 600 µl of culture into 600 µl of 6M HCl with a subsequent degassing time of 24 hours. 6 mL of scintillation cocktail was added, and samples were counted in a Liquid Scintillation Counter (PerkinElmer TriCarb 5110 TR) to obtain disintegrations per minute. Curves were fit using Graphpad Prism 7 (GraphPad Software, La Jolla California USA) with a Michaelis-Menton equation:

V = Vmax[S] / Km + [S] (7)

Where V is reaction rate, Vmax is the maximum reaction rate, [S] is the substrate concentration, and Km is the half saturation concentration of the cell to DIC. Extracellular carbonic anhydrase (eCA) activity was measured according to Rost et al. (2007), based on Silverman (1982) with slight modifications. Cultures were concentrated by gravity filtration and growth media was exchanged stepwise with CO2-free Aquil media buffered with 50 mM BICINE adjusted to match pH of growth conditions as close as possible (7.9, 8.1, or

8.3). Analysis of cell size distribution and Fv/Fm prior to and after concentration showed no significant differences (Data not shown) and ensured the health and intactness of the cultures.

18 Cells were further concentrated using gentle centrifugation at 250 rpm for 1 minute before being measured with a membrane-inlet mass spectrometer (MIMS). A MIMS (Pfeiffer, QMC200, Germany) with a custom-made cuvette system (10 mL volume) was used. CA activity was determined by adding bicarbonate labeled with 13C and 18O to the media and the uncatalyzed rate 18 - of O loss, which is caused by the hydration and dehydration steps of CO2 and HCO3 in water. This was measured for 5 minutes after reaching equilibrium. Subsequently, 150-250 µL of the concentrated cells suspension was added to the media. The rate of 18O depletion with cells (S2) was compared to the uncatalyzed rate (S1). Rates were normalized to chl a. U = S2 / (S1 – 1) (8) Where U represents the enhancement factor, expressed as an x-fold increase in the - interconversion rate between CO2 and HCO3 .

3. Results

3.1 Carbonate chemistry: The carbonate chemistry in this experiment is reported in

Table 1. The measured values of DIC and pH between the different pCO2 acclimations and calculated dissolved CO2 concentration proved to be significantly different between all pCO2 acclimations. Calculated total alkalinity (calculated from DIC and pH) showed relative constant values (2309 – 2347 µmol/kg seawater in the pre-acclimated media) and slightly drifted values (2341 – 2428 µmol/kg seawater) in the media which contained cultures. Concentrations of calculated pCO2 and measured pCO2 (based on calculation in CO2sys using DIC and pH) were relatively similar within each of the different pCO2 acclimations with the largest difference seen in the 150 µatm culture between pre-acclimated media and media in which cells were grown (Table 1). Carbonate chemistry compared between pre-acclimated media and the media measured from mid exponentially grown cells (Table 1) showed that measured DIC was stable within the methodological accuracy. The drift in pH during cell growth nonetheless resulted in a drift in the target pCO2 and TA (see Table 1) indicating small drifts of the carbonate chemistry during cell growth. Dissolved CO2 concentrations of the medium in which cultures grew to mid exponential phase were calculated to be 2.8 ± 0.5, 9.8 ± 0.8, 20.8 ± 2.2 µmol/kg seawater and - HCO3 concentrations were calculated to be 1268 ± 43, 1734 ± 35, 1941 ± 13 µmol/kg seawater in the 150, 400, and 780 pCO2 acclimations, respectively.

19 Table 1. Carbonate chemistry measured and calculated in the experiment. Data are given for carbonate chemistry measured prior to the addition of cells, as well as during mid exponential growth phase.

Measured pCO2 (µatm) in Gas Line 150 435 780 Abiotic Carbonate DIC (µM) 1679 ± 2 2008 2089 ± 1 Chemistry pH media (NBS) 8.51 ± 0.04 8.12 ± 0.02 7.89 ± 0.01

Calculated pCO2 118.9 ± 0.1 446.6 ± 31.0 812.7 ± 0.5 (µatm) Calculated TA 2320 ± 3 2348 ± 18 2309 ± 1 (µmol/kg seawater)

CO2 (µmol/kg 3.3 ± 0.0 12.3 ± 0.9 22.4 ± 0.0 seawater)

- HCO3 (µmol/kg 1286 ± 2 1793 ± 11 1937 ± 1 seawater) Carbonate DIC (µM) 1714 ± 4 1982 ± 26 2103 ± 4. Chemistry after pH media (NBS) 8.63 ± 0.05 8.22 ± 0.03 7.95 ± 0.04

cells grown to mid Calculated pCO2 102.9 ± 16.8 355.6 ± 30.4 753.4 ± 78.8 exponential phase (µatm) Calculated TA 2428 ± 64 2376 ± 14 2342 ± 19 (µmol/kg seawater)

CO2 (µmol/kg 2.8 ± 0.5 9.8 ± 0.8 20.8 ± 2.2 seawater)

- HCO3 (µmol/kg 1268 ± 43 1734 ± 35 1941 ± 13 seawater)

3.2 Growth, chlorophyll a, elemental composition, brevetoxin analysis, protein concentration: Growth rates for the 150, 400, 780 µatm and the control cultures were: 0.21 ± 0.06 d-1, 0.22 ± 0.06 d-1, 0.20 ± 0.06 d-1, and 0.24 ± 0.05 d-1, respectively, and the bubbled acclimations were not statistically different (One-way ANOVA, p > 0.05, df =417) (Figure 4, Table 2). Additionally, no difference in growth was detected between the control (un-bubbled – open to atmosphere) and the 400 µatm bubbled culture (One-way ANVOA, p = 0.0786, df

=417). Average cell size for K. brevis also did not change across pCO2 concentrations with average cell diameters of 22.0 ± 0.4, 22.1 ± 1 and 22.0 ± 0.3 µm for the three pCO2 acclimations, respectively. Chlorophyll a cell-1 showed a similar response with values for the three acclimations of 13.63 ± 1.26, 14.20 ± 0.57, and 15.07 ± 0.13 pg Chl a cell-1 for the 150, 400, and 780 µatm cultures, respectively (Table 2). Cells/Relative Fluorescence Units ratios stayed constant (5.62 ± 0.75) throughout the exponential growth phase (Data not shown). Quotas for cellular C were 0.66 ± 0.20, 0.50 ± 0.06, and 0.51 ± 0.06 ng carbon cell-1 and cellular N average values were 0.12 ± 0.04, 0.09 ± 0.01, and 0.09 ± 0.01 ng nitrogen cell-1 for the 150, 400, and 780 µatm cultures, respectively (Table 2; n=3). The C/N ratio (mol:mol) was calculated as 6.57 ± 0.13, 6.83 ± 0.27, and 6.53 ± 0.22 for the 150, 400, and 780 µatm cultures, respectively (Table

20 2). One-way ANOVAs found no significant differences between Chl a cell-1, C cell-1, and N cell- 1 -1 -1 values for the different pCO2 acclimations. The larger standard error in C cell and N cell value in the 150 µatm cultures indicates a measurement error, likely an underestimation in cell counts in two of the replicate cultures of this acclimation from this day. Brevetoxin values (PbTx-2 and PbTx-3) in pg cell-1 were 3.43 ± 0.89, 3.91 ± 0.66, and 4.69 ± 0.37 for the 150, 400, and 780 µatm cultures, respectively (Figure 5, Table 2) with no significant differences (one-way ANOVA). Additionally, protein cell-1 in ng for the 150, 400, and 780 µatm cultures were 0.45 ± 0.02, 0.40 ± 0.05, and 0.43 ± 0.07, respectively (Table 2, no significant differences (one-way ANOVA). Protein per C was calculated from Protein cell-1 and C cell-1 and found to be 34%, 40%, and 42% for the 150, 400, and 780 µatm cultures, respectively.

Table 2. Average values for growth rate, cellular Chl a, POC, PON, C/N ratios, brevetoxin, and protein content for K. brevis cells under different pCO2 treatments. Values given represent mean ± SD $ Rates are averaged for each biological replicate. SD represents error between biological replicates. ^ Chl a, POC and PON, brevetoxin and protein per cell per cell per cell were taken during the mid of one exponential growth phase. Chl a pCO2 Growth rate µ POC ng PON ng C:N pg Brevetoxin ng Protein pg (ppmv) (d-1)$ cell-1^ cell-1^ (mol:mol) ^ cell-1^ cell-1^ cell-1^ 13.63 0.66 ± 0.12 ± 150 0.2087 ± 0.06 6.57 ± 0.13 3.43 ± 0.89 0.453 ± 0.020 ± 1.26 0.20 0.04

14.20 0.50 ± 0.09 ± 400 0.2154 ± 0.06 6.83 ± 0.27 3.91 ± 0.66 0.397 ± 0.047 ± 0.57 0.06 0.01 15.07 0.51 ± 0.09 ± 750 0.2047 ± 0.06 d 6.53 ± 0.22 4.69 ± 0.37 0.429 ± 0.070 ± 0.13 0.06 0.01

3.3 Photophysiology: No pCO2 effects were seen on the dark adapted Fv/Fm, or in the values for EK (light saturation point), sLHII (absorption cross section of PSII) and GPPFRRf (gross productivity as analyzed via FRRF as calculated in Eq. 6 and Eq. 7 at 121 µmol photons m-2 s-1 (the light intensity measured in the FRRf closest to acclimation light) for the K. brevis strain used in this study (Figure 6). EK values averaged over the light period were 483 ± 23, 489 ± 38, and 513 ± 11 µmol photons m-2 s-1 for the 150, 400, and 780 µatm acclimations, respectively (n ≥

5). Average values of sLHII measured over the light period were 4.25 ± 0.10, 3.99 ± 0.11, and 2 -1 4.10 ± 0.17 nm PSII for the 150, 400, and 780 µatm acclimations, respectively. A pCO2 effect on non-photochemical quenching (NPQ) as well as tES (time constant for the re-opening of a closed RCII with an empty Qb site) was apparent for some duration of the diurnal cycle (Figure 6 C, D). During the first three quarters of the light period (6 am until 3 pm) the 780 µatm cultures 21 had both higher NPQ and tES values compared to the 150 and 400 µatm cultures. Additionally, during these times, a consistently higher NPQ was observed for the 780 µatm cultures under all light intensities measured (Figure 7A, B). NPQ relaxed to values similar to the 150 and 400 µatm acclimation after 3pm (Figure 7C).

Fv/Fm values showed a pronounced diurnal cycle, increasing slightly from the onset of light until noon and decreasing towards the end of the light period (Figure 6A). GPPFRRf as 22 calculated according to eq. 6 and eq. 7 showed a slight increasing trend until midday and a subsequent slight decrease until the end of the light period (Figure 6B). Calculated GPPFRRf from 12 pm FLCs at 121 µmol photons m-2 s-1 (the light intensity measured in the FRRf closest to acclimation light) yielded rates of 111 ± 3, 102 ± 3, and 107 ± 7 µmol C mg Chl a-1 h-1 for the 150, 400, and 780 µatm acclimations, respectively.

3.4 Photosynthetic oxygen evolution: Net photosynthesis in K. brevis was largely -1 unaffected by the different pCO2 treatments averaging 56 ± 6, 67 ± 6, and 60 ± 5 µmol O2 Chl a h-1 over the light period for the 150, 400, and 780 µatm cultures respectively. Average photosynthetic rates during the light period are shown in Figure 8 and Table 3. Values of photosynthesis normalized per cell are given in the supplemental materials (Table 5). The -1 -1 measured rates of dark respiration averaged 37 ± 3, 26 ± 2, and 26 ± 2 µmol O2 Chl a h for the 150, 400, and 780 µatm cultures respectively and accounted for approximately 40-44% of the calculated gross photosynthesis (Figure 8, Table 3). In all three pCO2 treatments, photosynthetic rates decreased at the end of the light period while respiration rates increased over the light

23 period (data not shown) and decreased during the night. In the 150 µatm cultures, night time respiration rates were significantly higher compared to the 400 and 780 µatm cultures (One-way ANOVA, p < 0.0001, df=71). These diurnal trends in photosynthesis, respiration, and photophysiology (see section 3.3) indicate that K. brevis expresses a pronounced diurnal cycle, even in laboratory cultures under constant 12-hour light exposure.

24 Table 3. Measured net and gross primary productivity and respiration rates from 24 h experiments. Rates are average of three replicate measurements and averaged over the 12-hour light phase (NPP, GPP) or the dark period (RO, RC).

NPPO (µmol GPPO (µmol RO NPPC GPPC RC -1 -1 pCO2 (µatm) O2 mg Chl a O2 mg Chl a (µmol O2 mg (µmol C mg (µmol C mg (µmol C mg hr-1) hr-1) Chl a-1 hr-1) Chl a-1 hr-1) Chl a-1 hr-1) Chl a-1 hr-1) 150 56 ± 6 99 ± 12 37 ± 3 40 ± 5 83 ± 10 37 ± 3 400 67 ± 6 112 ± 8 26 ± 2 48 ± 5 93 ± 7 26 ± 2 750 60 ± 5 104 ± 6 26 ± 2 43 ± 3 87 ± 5 26 ± 2

Figure 8: Net O2 evolution rates of K. brevis in different pCO2 acclimations measured continuously throughout a 24-hour period. Non-shaded areas indicate periods of light and data shown is net primary productivity (NPPO). Shaded areas indicate dark periods (night) and lines in these areas are representative of respiration rate (R). Respiration rates are shown as positive values for clarity. Data shown are mean values (n > 3, ± SD). Error bars are staggered by 30 min. for clarification, with the left most error bar representing 150 µatm, middle error bar representing 400 µatm, and right most error bar representing 780 µatm.

3.5 Carbon acquisition:

3.5.1 Kinetics: The K1/2 values for CO2 were 1.36 ± 0.24, 1.64 ± 0.27, and 3.36 ± 0.49

µmol CO2 for the 150, 400, and 780 µatm cultures, respectively (Figure 9A). The 780 µatm cultures showed a significantly higher K1/2 (One-way ANOVA, p = 0.0008, df = 8) compared to

150 and 400 µatm cultures. The respective K1/2 values for DIC were 333.1 ± 58.1, 247.0 ± 40.3, and 294.6 ± 42.6 µmol DIC showing no statistically significant differences (One-way ANOVA, p = 0.94, df = 8). It should be noted that the measurements were conducted in pH adjusted media

(close to the respective acclimation pHs of 7.9, 8.1, 8.3, respectively). Hence dissolved CO2, - 2- HCO3 and CO3 concentration ratios varied in the respective assays.

25 3.5.2 C-source preference: Higher dissolved CO2 resulted in a significant increase in

CO2 utilization with 14% ± 1%, 26% ± 5%, 56% ± 8% CO2 uptake for the 150, 400, and 780 µatm cultures, respectively, (One-way ANOVA, p = 0.0002, df = 8) (Figure 9B).

3.5.3 External CA activity: Each pCO2 acclimation showed activity of eCA. Values for the 150, 400, and 780 µatm cultures were 6.69 ± 0.71, 5.43 ± 0.75, and 5.12 ± 0.67 U (µg Chl a-1). Notably, the 150 µatm cultures had significantly higher eCA activity (One-way ANOVA, p = 0.0016, df=2) (Figure 9C).

4. Discussion This study aimed to understand underlying mechanisms of a potential CO2 sensitivity in the dinoflagellate Karenia brevis (Errera et al., 2014; Hardison et al., 2014). Despite partially contradicting each other, both of those studies indicated that CO2 can affect the cellular physiology of K. brevis, leading to altered growth and changes in production of brevetoxin. As suggested by Errera et al. (2014), underlying processes such as insufficient inorganic carbon supply under “low” CO2 (55-245 µatm) to support the C-fixing enzyme RubisCO could have resulted in the reduced growth under the low and ambient CO2 concentrations measured in their study. Hardison et al. (2014) hypothesized, however, that K. brevis must have a relatively efficient CCM as cells grew optimally at low CO2 concentration down to 2.4 µM. In Hardison et al. (2014), low pCO2 was also found to induce cellular brevetoxin production while in Errera et al. (2014) toxin production did not change on the per cell level. Here we analyzed similar cellular responses, and additionally quantified photosynthesis and modes of carbon acquisition (the

CCM) under low, ambient, and high CO2 concentrations. Furthermore, we characterized diurnal processes of photosynthesis and photophysiology, e.g. light quenching mechanisms, which have been suggested to be affected by brevetoxin (Cassell et al., 2015). One of the challenges of growing non-armored dinoflagellates such as K. brevis under different CO2 concentrations is that mechanical shear stress, as implemented by bubbling, can harm the integrity of the cells (van de Waal et al., 2014). It has been shown that K. brevis can only cope with low amounts of shear stress (Martin et al., 2003). We used an undisturbed culture

(acclimated to atmospheric pCO2) as a control to identify potential physical shear stress on growth, morphology and photosynthetic yield response. The collected data indicate that the gentle bubbling performed to keep carbonate chemistry stable did not significantly alter rates of growth (comparing the control and the 400 µatm acclimation culture – Figure 1) nor did it change photosynthetic yield or cell size (Data not shown). In order to reach the target CO2 26 concentrations and keep them stable throughout the growth of the cultures, media was vigorously pre-bubbled with air at the acclimation pCO2 for at least 24 hours before adding the cells. Once cells were added, cultures were very gently bubbled while keeping the headspace of culture bottles filled with the appropriate pCO2 air through the use of an exhaust system. Using this approach and low cell densities (maximum ~6000 cells/ml) the cells were acclimated to the respective CO2 concentration over the duration of the exponential growth. While carbonate chemistry was not perfect and shifted slightly during cell growth, it was clearly different between the different acclimations.

Figure 9: CCM parameters measured in the experiment. A) Half saturation constants for CO2 B) - Fraction CO2 to HCO3 utilization. C) eCA activity. Data shown are mean values (n ≥ 3, ± SD). Statistical differences and groupings determined via ANOVA and Tukey’s HSD indicated using letters (p-values < 0.05). 27 Growth, cellular C and N quotas, and cellular chl a did not show a distinct CO2 response in this study and are in contrast to the two studies previously published on the effects of CO2 on K. brevis (Table 2, Figure 4, 5) (Errera et al., 2014; Hardison et al., 2014). While the measured average growth rates (0.2 – 0.24 d-1) were slightly lower compared to Errera et al. (2014) (growth rates of ~0.30 d-1) and Hardison et al. (2014) (growth rates of 0.30-0.55 d-1), these differences might be a strain specific trait response. However, other potential triggers such as differences in acclimation methods (carbonate chemistry modification methodology, growth medium) cannot be ruled out. Brevetoxin production (PbTx-2 and PbTx-3) of our strain of K. brevis (CCFWC-126) showed an increasing trend with increasing CO2 (Table 2, Figure 5), but, this trend is statistically not significant in the measured CO2 range. Results from this study agree with Errera et al. (2014) where no significant CO2 effect on cellular brevetoxin concentration was found. In contrast,

Hardison et al. (2014) found increasing cellular brevetoxin in K. brevis grown under low pCO2.

It should be noted that the CO2 concentration in the low pCO2 acclimation in Hardison et al. (2014) was much lower than in this study. In addition, while the ELISA assay used in this study only reacts with PbTx-2 and PbTx-3, it does not fully capture PbTx-1. However, the contributions of the different brevetoxins (PbTx-1, 2, and 3) and total brevetoxin shows that PbTx-1 only accounted for 3.69 ± 2.64% of total brevetoxin (Pierce and Henry, 2008) in K brevis found along the West Florida Shelf in the GoM. Therefore, despite not measuring all brevetoxin molecular structures, the data shown here should represent the majority of brevetoxin content in K. brevis. In general, responses in toxin production in dinoflagellates (and diatoms) to changes in CO2 are diverse (e.g. Fu et al., 2010; Sun et al., 2011; Tatters et al., 2012, 2013;

Hattenrath-Lehmann et al., 2015). If the response to CO2 is indeed strain specific, it would indicate that strains which respond positively in growth to increasing CO2 could increase in abundance in a future ocean. For toxin producing species, this could affect ecology as well as human health and economy if toxin production itself is additionally stimulated. While much of the brevetoxin synthesis pathway is unknown, it requires inorganic carbon and cellular energy (Calabro et al., 2014). Brevetoxin is thought to be synthesized from acetyl CoA and additional acetate groups (Van Dolah et al., 2009) coupled to glycolysis and the tricarboxylic acid cycle, which are clearly linked to cellular carbon metabolism and cellular energy and reductants are needed to fuel these processes. As shown in Hardison et al. (2012, 2013, 2014), brevetoxin content as a percent of cellular carbon is approximately 0.8-2.1% for

28 nutrient replete K. brevis cultures. Brevetoxin content as a percent of cellular C values measured here ranged from 0.32 to 0.62%. As brevetoxin is only a small percentage of cellular carbon, and

CCMs and glycolysis are not directly linked, the response is likely not directly triggered by CO2 availability. Nonetheless, it cannot be ruled out that a potential reallocation of cellular energy and reductants toward brevetoxin production could result in changes in cellular brevetoxin content.

The CO2 insensitivity of K. brevis strain CCFWC-126 indicates that this strain is capable of maintaining growth and cellular composition under a wide range of CO2 concentrations, indicative of an efficient CCM or a RubisCO half saturation concentration less than that of the ambient dissolved CO2 in the seawater. Rates of net and gross photosynthesis from the different pCO2 acclimations support both hypotheses (Figure 8, Table 3). NPPFRRf estimates showed strong agreement with the NPPC estimates obtained from oxygen evolution data (Figure 10), further supporting the idea that the range of CO2 tested in the acclimations here is not directly influencing photosynthesis. While other studies found enhanced respiration under elevated pCO2 (Wu et al., 2010; Gao et al., 2012; Yang and Gao, 2012; Eberlein et al., 2014), this study found an increase in night-time respiration under low pCO2, potentially alleviating external pH stress (Hansen, 2002). A diurnal cycle, similar to that observed in this study, has been seen in gene expression of photosynthetically relevant genes of K. brevis (Van Dolah et al., 2007) and in photosynthesis and respiration measurements of K. brevis in the field (Hitchcock et al., 2014). These diurnal cycles are seen in many organisms and, despite numerous ecophysiological implications, have some important implications for experimental approaches. For instance, it is important to sample at similar times of the day because cellular composition (such as C:N ratio, Chl a cell-1 or even brevetoxin cell-1) could be affected by this cycle. In order to gain a process-based understanding of responses of growth and photosynthesis, photophysiological data over a 24-hour cycle was analyzed. The measured diurnal pattern in dark adapted Fv/Fm (Figure 6A) indicate a diurnal regulation in photosynthetic machinery. In this study, light was sub-saturating (as indicated in the higher EK values compared to acclimation light), hence no photostress was initiated which would otherwise lead to a reduced

Fv/Fm. The enhanced NPQ and tES values in the 780 µatm acclimation (Figure 6C, D, 7) indicate that electron transport rate and energy conversion into photochemistry was reduced. Yet, differences under acclimation light were relatively small and did not result in reduced photosynthesis. An interesting, yet hypothetical, explanation of some of the measured data could

29 be that brevetoxin interacts with the light harvesting complex in PSII, facilitating NPQ, as suggested by Cassell et al. (2015). If this would be the case, cellular brevetoxin content might be enhanced during the morning hours in the 780 µatm culture compared to the other acclimations. Since brevetoxin (PbTx-2 and PbTx-3) was measured in the afternoon, this interesting connection might not have been detected. However, this study does not provide direct proof for such a connection.

Figure 10: Net primary productivity measured and calculated using different parameterization. NPPC from oxygen evolution converted to C-fixation; NPPFRRf using FLC data from 12pm at 121 -2 -1 µmol photons m s and equations from Lawrenz et al. (2013); NPPµ calculated using growth -1 -1 rate (µ (d )) and POC cell ; NPPLosh calculated based on Losh et al. (2013) using growth rate (µ (d-1)), R/GPP ratio, and protein cell-1 data.

In conclusion, despite small changes in photophysiology, net photosynthesis of K. brevis is not affected by CO2. Comparisons of growth and productivity (measured and calculated) show similar overall rates and patterns and support the observed CO2 insensitivity of K. brevis (Figure -1 10, FRRf (NPPFRRf), POC cell *µ (NPPµ), oxygen evolution (NPPC) and “NPPLosh” (based on -1 Protein cell , R/ GPPC ratio, growth rate, and estimated RubisCO turnover kinetics; see Table 4, 6, 7). Despite the large amount of assumptions included in some of those estimates, a very good agreement was observed. While it has been shown that K. brevis contains and maintains pyrenoids, one component of a CCM (Giordano et al., 2005; Monroe et al., 2010). K. brevis has shown a close phylogenetic relationship to the haptophyte Pavolva lutheri via analysis of both photosystem I and RubisCO genes (Yoon et al., 2005, 2002). P. lutheri’s RubisCO Km value was measured to be 17.6 µM (Heureux et al., 2017), which is one of the most efficient RubisCO carboxylation kinetics

30 measured in phytoplankton. Consequently, it should be asked if the cells require additional aspects of the CCM, a supposedly expensive mechanism, to acquire sufficient inorganic carbon for growth and reproduction. We therefore calculated the theoretical carbon demand, C-fixation potential, and growth without and with a potential CCM. Calculations assuming C-uptake in K. brevis lacking a CCM

(relying on diffusive CO2 uptake only) and assuming a similar Km value of RubisCO as found in

P. lutheri (Km = 17.6 µM) yield only a 16% RubisCO saturation under 150 µatm and a 56% CO2 saturation of RubisCO under 750 µatm (assuming a constant CO2 concentration at RubisCO similar to the equilibrated medium). RubisCO saturation is likely lower than those assumed above as internal CO2 has to be slightly lower compared to the external concentration in order for diffusive CO2 uptake to happen (Hopkinson et al., 2011). Since sufficient and similar growth under “non-CCM” conditions is unlikely, we calculated the theoretical growth rate using measured data on cellular composition and photosynthetic rates following Losh et al. (2013) (see Table 6, 7) with modifications based on measured respiration rates (Table 6, 7). These calculations, assuming an ~85 % CO2 saturation of RubisCO (Figure 9), yield in a theoretical specific growth rate of around 0.18 d-1 for 150 µatm, 0.28 d-1 for 400 µatm, and 0.26 d-1 for 750 µatm, which matches the growth rates measured in this study fairly well (Figure 4, Table 6, 7). Based on those calculations and assumptions, it is implicit that K. brevis must possess an CCM to saturate RubisCO and obtain the constant growth and productivity rates observed under the

CO2 concentrations tested here.

In general, any process enhancing the supply of CO2 to RubisCO is considered a part of - the CCM. The CCM in K. brevis was found to rely on CO2 as well as HCO3 uptake and external carbonic anhydrase. Using the Km (CO2) value for type I RubisCO of 17.6 µM CO2 (see above) and the measured K1/2 (CO2) values (Fig. 9), K. brevis must accumulate 13, 11, and 5 times the

CO2 concentration within the proximity of RubisCO compared to the media, leading to a saturation of 67%, 86%, and 86% in the 150, 400, and 780 pCO2 acclimations, respectively. While P. lutheri shows a close genetic relationship with K.brevis, P. lutheri does not contain pyrenoids and CCM parameters are unknown (Heureux et al., 2017). Consequently, the Km values in K. brevis might be more similar to those of Emiliania huxleyi and therefore all assumptions made here have to be taken with caution. Nonetheless, the measured data on K1/2 show that K. brevis employs a CCM, which under elevated CO2 is actively downregulated with

K1/2 values similar to those measured in the RubisCO type II containing Alexandrium tamarense

31 (Eberlein et al., 2014). Data measured in this study also fit well within the conceptual idea that phytoplankton aim to saturate RubisCO with CO2 between 80 and 90%, if energetically feasible.

In terms of carbon preference, K. brevis showed an increasing utilization of CO2 uptake with increasing pCO2 (Figure 9). Hence, this study suggests that K. brevis is able to I) overcome low

CO2 levels by switching to a more readily available but more expensive carbon source, which was seen by changes in the f-value in 14C disequilibrium experiments (Figure 9) and II) downregulate the energetic expense of the CCM once a higher external CO2 concentration is available. The measured eCA activity likely acts to maintain diffusive CO2 uptake and reduce the diffusive loss of internal CO2 as it supports a persistent CO2 concentration at the cell surface (at a given carbonate chemistry) (Hopkinson et al., 2013; Trimborn et al., 2008). Enhanced eCA activity has also been postulated to help to recover CO2 which leaks out of the cells as well as regulate the cell surface pH (Trimborn et al., 2008). As dinoflagellates have been shown to leak approximately 50% of the inorganic carbon taken up (Eberlein et al., 2014), eCA could play an important role in C-acquisition in K. brevis. With higher external CO2, less CA would be required as the gradient over the cellular membrane is reduced. Internal CA was not specifically measured in this study, but it is very likely that several internal CAs will support the supply of

CO2 for carbon fixation at RubisCO (Ratti et al., 2007). The measured regulation of K. brevis’s carbon acquisition as well the evidence of pyrenoids found by Monroe et al. (2010) support the conclusion that K. brevis maintains an active CCM.

The CO2 dependent changes in K1/2 values, inorganic carbon source preferences, and CA activity are likely reducing the energetic cost of the CCM under high CO2 (Hopkinson et al., 2011; Kranz et al., 2015). While this down-regulation of the CCM has a potential to increase growth and productivity purely by saving and reallocating energy between the processes involved, this response has not been shown here and is not always the case (Eberlein et al., 2014; Hopkinson et al., 2014). The slight increase in brevetoxin content cell-1 seen in the 780 µatm cultures could be the result of that energy saved from the CCM under high pCO2 levels, yet, this study lacks the molecular and mechanistic understanding to prove this hypothesis. The overall slow growth rate and low C requirement might indicate that the CCM regulation in K. brevis does not save a significant amount of energy, which could explain the moderate metabolic responses to high or low CO2. While CCMs have an energetic cost associated to the maintenance of the different mechanisms, the CCM overall might actually not be as costly as previously thought (Hopkinson et al., 2014).

32 5. Conclusions

Our study is the first measuring CCM activity in the ecologically and economically relevant dinoflagellate K. brevis. We found evidence for K. brevis having an active and efficient CCM, supporting growth, productivity, and brevetoxin production during bloom situations when

CO2 concentrations could be limiting. Additionally, the CCM is down-regulated under enhanced pCO2 conditions, which could result in energy reallocation from C-acquisition to other cellular processes. While this energy reallocation is speculative, it demonstrates the importance of investigating underlying processes such as CCMs when aiming to understand the impacts of environmental change on marine phytoplankton. The strong CO2 dependent regulation of the CCM and photophysiology indicates that K. brevis possesses mechanisms which can increase the resilience of this species under a range of CO2 concentrations, especially during bloom conditions and in a future ocean. Our results highlight the possibility that rising pCO2 levels could result in increased toxicity of K. brevis blooms if CO2 rises even further than the projected value of 780 µatm CO2. This result, along with evidence for increases in bloom occurrence in a future ocean (Brand and Compton, 2007), show the potential for increased impacts from K. brevis both ecologically and economically in a future high CO2/low pH ocean.

33 6. Supplemental

Table 4: Calculation parameters for NPPLosh NPPLosh was based on equations outlined in (Losh et al., 2013) and was calculated with the following equation: 5 9 NPPLosh = (((0.20 * (µ/R)) * P)/(5.5 X 10 ) *8*3*12*(2/5)*2)* 3.6 x 10 (SE1) Value Description Source µ Growth rate This study (Figure 4, Table 2) R Kcat of a Rubisco active site (Losh et al., 2013) (turnover rate, 3 C s-1 at 25°C) P Protein per cell This study, (Table 2) 0.20 Factor relating growth rate (Losh et al., 2013), modified and Rubisco as a percent of with respiration rate measured cellular protein (Cmin) in this study 5.5 X 105 Molecular weight of Type 1 (Losh et al., 2013) RubisCO – 550kDa 8 Assumption that all 8 (Losh et al., 2013) RubisCO active sites are active at once

3 Kcat of a Rubisco active site (Losh et al., 2013) (turnover rate, 3 C s-1 at 25°C) 12 Molecular weight of carbon 2/5 Amount of carbon lost to This study, Oxygen evolution respiration data (Figure 8, Table 3) 2 Assumption that half of (Losh et al., 2013) biomass is carbon 3.6 x 109 Conversion from seconds to hours, and mol to µmol

Table 5: Net and gross primary production and respiration rates normalized to cell counts pCO2 NPPO GPPO RO NPPC GPPC RC (µatm) (µmol O2 (µmol O2 (µmol O2 (µmol C (µmol C (µmol C cell-1 hr-1) cell-1 hr-1) cell-1 hr-1) cell-1 hr-1) cell-1 hr-1) cell-1 hr-1) 150 7.47×10-7 1.34×10-6 5.05×10-7 5.34×10-7 1.12×10-8 5.05×10-7 ± 5.54×10- ± 1.10×10- ± 2.44×10- ± 3.96×10- ± 9.26×10- ± 2.44×10- 8 7 8 8 8 8 400 9.52×10-7 1.59×10-6 3.71×10-7 6.80×10-7 1.32×10-6 3.71×10-7 ± 7.11×10- ± 9.04×10- ± 1.71×10- ± 5.08×10- ± 7.49×10- ± 1.71×10- 8 8 8 8 8 8 780 9.05×10-7 1.57×10-6 3.91×10-7 6.46×10-7 1.31×10-6 3.91×10-7 ± 6.55×10- ± 7.90×10- ± 2.33×10- ± 4.68×10- ± 6.60×10- ± 2.33×10- 8 8 8 8 8 8

Table 6: Calculation parameters for theoretical growth rates Theoretical growth rates were calculated based on (Losh et al., 2013), using a calculated Cmin based on the following equation: -7 Cmin = ((NPPc/(8×3×12×(2/5)×2×3600×1000000))/6.06×10 )/M (Eq. 9) Value Description Source NPPC Net Primary Productivity in This study, Oxygen evolution -1 -1 µmol O2 cell h data (Figure 8, Table 3) and equation 5 M Biomass carbon This study, POC cell-1 (Table 2) 8 Assumption that all 8 (Losh et al., 2013) RubisCO active sites are active at once 3 Kcat of a Rubisco active site (Losh et al., 2013) (turnover rate, 3 C s-1 at 25°C) 12 Molecular weight of carbon 2/5 Amount of carbon lost to This study, Oxygen evolution respiration data (Figure 8, Table 3) 2 Assumption that half of (Losh et al., 2013) biomass is carbon 3600 Conversion factor between seconds and hours 10000000 Conversion factor between mol and µmol 6.06×10-7 Relationship between (Losh et al., 2013) Rubisco content (mol Rubisco cell-1) and cell biomass

Table 7: Calculation of theoretical growth rates Theoretical growth rates were calculated based on (Losh et al., 2013), using a calculated Cmin from eq. 9 substituted into the following equation: -1 µtheoretical(d ) = (R×(Cmin×X))/Rubsat (Eq. 10) Variable Description Source R Kcat at RubisCO active site (3 (Losh et al., 2013) C s-1 at 25°C) Cmin RubisCO as a fraction of This study, eq. 9 cellular protein X (8×12×(2/5)×2×6.06×10- (Losh et al., 2013) with 7×86000) or 5 change in respiratory losses (see Table S3 for further explanation) Rubsat RubisCO saturation This study, calculated from K1/2 (DIC) values

35 Method S1: Dissolved inorganic carbon (DIC) was measured following a modified protocol from Noguchi et al. (2013). 20 mL of media was transferred from a culture vial through a GF/F syringe filter into a pre-evacuated 200 mL headspace bottle. The media was acidified using 200 µl of CO2-free 2N HCl (15 min incubation phase on a shaker, the -1 headspace was purged with CO2 free air at a flow of 0.2 L min and all CO2 from the headspace was measured using the LI-820 CO2 analyzer (LI-COR, Lincoln, Nebraska, USA).

Methods S2: Brevetoxin concentration was analyzed using an ELISA analysis kit and samples for measurement were prepared as follows. 100 mL of K. brevis culture was concentrated onto a 10 µM PFTE filters using gravity filtration. Samples were stored at -80°C until measurement. Five mL of 100% methanol (MeOH) was added to each sample and let extract over night at -20°C, then samples were probe sonicated (Q55, Qsonica LLC, Newtown, CT, USA) for 20 seconds at 50 amplitude on ice, centrifuged for 4 minutes at 4400 g, and supernatant was removed and stored in a 15 mL centrifuge tube at -20°C. This process was repeated two times with 5 mL and 2 mL of MeOH. All supernatants for the samples were pooled and stored at -20°C until measurement. Prior to brevetoxin ELISA analysis (Abraxis Inc, Warminster, PA, USA), samples were diluted (1:10 dilution) in sample diluents provided by Abraxis and subsequently measured using a multiplate reader (Synergy HTX; BioTek, Winooski, VT, USA). Methods S3: The FRRf used light emitting diodes (LEDs) with an emission wavelength of 450 nm to excite chl a. The FRRf was set to single turnover mode and had a saturation phase with 100 flashlets on a 2 µs pitch and a relaxation phase with 40 flashlets on a 60 µs pitch. Temperature was held constant at 26°C using a water bath. Cells were dark acclimated for 15 minutes before each FLC and acclimated to each light intensity for 3 minutes during the FLC. After each FLC, cells were replaced with fresh culture from the incubator using an automated peristaltic pump. Measured cells were disposed and not fed back into the culture bottle.

36 CHAPTER 3

GENERAL DISCUSSION AND OUTLOOK

Karenia brevis is a red tide dinoflagellate which has been described as having one of the most predictable bloom patterns in the Gulf of Mexico. This species has challenged scientists for decades and is currently a major threat for Florida’s coastal environment and economy. K. brevis is an organism which is capable of surviving in a wide range of light, temperature, salinity, and nutrient regimes. This adaptability and the economic and ecological importance of this species make it critical to investigate how it will respond to a changing ocean. Many studies have characterized its ecophysiological plasticity and quantified its responses to changes to environmental variables such as salinity, temperature, and nutrients, but this species has been understudied in terms of climate change responses.

The only studies to date, which have looked at the responses of K. brevis to CO2 are Hardison et al. (2014), Errera et al. (2014), and this study. Further, only Errera et al. (2014) and this study investigated the responses of K. brevis to elevated CO2 concentrations. Our results show a CO2 insensitivity in growth, cellular composition, and photosynthesis in K. brevis, which is in contrast to the previously mentioned studies done K. brevis. However, differences in the measured response patterns could be due to a multitude of factors, including differences in culture method, carbonate chemistry, light intensity, and strain used for study for example.

Overall, current studies on K. brevis indicate that the response of this species to increases in CO2 could be strain dependent. However, it should be noted that it is a common thread among the responses measured in K. brevis is that they are either unaffected by or benefit from increasing

CO2 concentrations. With the increase in atmospheric CO2 and the associated shifts in carbonate chemistry, temperature, and light regime that K. brevis will experience, it is important to consider interactive effects. Increasing air and sea surface temperatures are increasing bloom range and durations for some phytoplankton species, and it has been predicted that temperatures could increase 1.8 to 4 °C (Beardall et al., 2009; Hallegraeff, 2010; Meehl et al., 2007; Moore et al., 2008). While higher temperatures have typically been found to have negative consequences for K. brevis growth, this might not be the case in a future ocean with increased pCO2 as Errera et al. (2014) demonstrated the ability of some strains to overcome this negative effect when pCO2 was high. Increases in bloom concentration shown over the last 50 years and increases in water temperature 37 in the Gulf Stream could create a suitable environment for K. brevis blooms to extend further up the eastern US coast in the future (Brand and Compton, 2007; Errera et al., 2014). Additionally, it is possible that some strains might benefit from increased pCO2 and increase brevetoxin production, as was observed in this study. The ability of K. brevis to benefit from increased pCO2, even under higher temperatures, could lead to an expansion of their economic and ecological impact in a future ocean. While brevetoxins have significant impact on human health and animal kills, it is also interesting to note the potential for interactions of brevetoxin production and K. brevis’s ability to adapt to changes in the light environment. Cassell et al. (2015) found localization of brevetoxins in chloroplasts, where it binds to the light harvesting complex at PSII. This study furthermore found low toxin cultures to be deficient in non-photochemical quenching (NPQ). The light harvesting complex II functions to harvest light for use in photosynthesis and also to dissipate excess light energy as heat (NPQ) via the xanthophyll cycle. Consequently, brevetoxin production could facilitate NPQ and protect cells during exposure to high light over prolonged timescales. Additionally, a recent study by Chen et al. (2018) demonstrated differences in the redox states and the xanthophyll cycle in low and high toxin cultures. Interestingly, while our brevetoxin cell-1 trends were not statistically significant, we did observe a significant response in NPQ in our 780 µatm cultures (Fig. 7). The 780 µatm cultures had higher NPQ than the 150 and 400 µatm cultures during early hours of the day (Fig. 7). While brevetoxins have not been shown to have allelopathic effects, the ability for its interactions with K. brevis light harvesting and utilization could lead to competitive advantages in a future ocean (Brand et al., 2012 and references therein). In conclusion, it seems that the response of K. brevis to climate change might be strain specific. However, the responses of measured K. brevis strains to pCO2 in this study and others point to this species benefiting from increased DIC. The ability of K. brevis to regulate their CCM and reallocate energy to other processes (such as heat or light stress prevention) could enhance resilience of this species in a future ocean and allow it to expand in range. This, along with studies showing the increased occurrences of K. brevis blooms in the GoM, suggest that this species will likely benefit from climate change. This has implications for the local ecology in the GoM and the potential for K. brevis to become an even bigger environmental nuisance for Florida, and the USA as a whole, highlights the need for continued research into this economically important species.

38 Research done in the future on this organism should not only focus on CO2, but also on unraveling the underlying causes of CO2 sensitivities (or a lack thereof) as well as interactive effects of multiple environmental parameters. For example, K. brevis cultures will be used to isolate and test the effects of pH and DIC in order to better understand the results of this study. Additionally, future work with K. brevis (along with Thalassiosira sp., Synechococcus sp., and Prochlorococcus sp.) in the Kranz lab will be done to elucidate potential effects of light spectra on cellular processes, such as those measured in this study. Lastly, it will be important to investigate the potential for competitive advantages of K. brevis when cultured with other common GoM taxa, such as Pseudo-Nitzschia sp. cultures. Overall, while this thesis provides insights into the carbon acquisition and photosynthesis of K. brevis, there is much more to be learned about how its ecophysiology will change in a future ocean.

39 REFERENCES

Badger, M.R., Andrews, T.J., Whitney, S.M., Ludwig, M., Yellowlees, D.C., Leggat, W., Price, G.D., 1998. The diversity and coevolution of Rubisco, plastids, pyrenoids, and chloroplast-based CO2-concentrating mechanisms in algae. Can. J. Bot. 76, 1052–1071. https://doi.org/10.1139/b98-074

Beardall, J., Stojkovic, S., Larsen, S., 2009. Living in a high CO2 world: impacts of global climate change on marine phytoplankton. Plant Ecol. Divers. 2, 191–205. https://doi.org/10.1080/17550870903271363

Boyd, P.W., Dillingham, P.W., McGraw, C.M., Armstrong, E.A., Cornwall, C.E., Feng, Y. -y., Hurd, C.L., Gault-Ringold, M., Roleda, M.Y., Timmins-Schiffman, E., Nunn, B.L., 2016. Physiological responses of a Southern Ocean diatom to complex future ocean conditions. Nat. Clim. Change. https://doi.org/10.1038/nclimate2811

Boyd, P.W., Doney, S.C., 2003. The impact of climate change and feedback processes on the ocean carbon cycle, in: Ocean Biogeochemistry. Springer, pp. 157–193.

Brand, L.E., Campbell, L., Bresnan, E., 2012. Karenia: The biology and ecology of a toxic genus. Harmful Algae 14, 156–178. https://doi.org/10.1016/j.hal.2011.10.020

Brand, L.E., Compton, A., 2007. Long-term increase in Karenia brevis abundance along the Southwest Florida Coast. Harmful Algae 6, 232–252. https://doi.org/10.1016/j.hal.2006.08.005

Bronk, D.A., Killberg-Thoreson, L., Sipler, R.E., Mulholland, M.R., Roberts, Q.N., Bernhardt, P.W., Garrett, M., O’Neil, J.M., Heil, C.A., 2014. Nitrogen uptake and regeneration (ammonium regeneration, nitrification and photoproduction) in waters of the West Florida Shelf prone to blooms of Karenia brevis. Harmful Algae 38, 50–62. https://doi.org/10.1016/j.hal.2014.04.007

Calabro, K., Guigonis, J.-M., Teyssié, J.-L., Oberhänsli, F., Goudour, J.-P., Warnau, M., Bottein, M.-Y., Thomas, O., 2014. Further Insights into Brevetoxin Metabolism by de Novo Radiolabeling. Toxins 6, 1785–1798. https://doi.org/10.3390/toxins6061785

Cassell, R.T., Chen, W., Thomas, S., Liu, L., Rein, K.S., 2015. Brevetoxin, the Dinoflagellate Neurotoxin, Localizes to Thylakoid Membranes and Interacts with the Light-Harvesting Complex II (LHCII) of Photosystem II. ChemBioChem 16, 1060–1067. https://doi.org/10.1002/cbic.201402669

Chen, W., Colon, R., Louda, J.W., del Rey, F.R., Durham, M., Rein, K.S., 2018. Brevetoxin (PbTx-2) influences the redox status and NPQ of Karenia brevis by way of thioredoxin reductase. Harmful Algae 71, 29–39. https://doi.org/10.1016/j.hal.2017.11.004

Dickson, A., Millero, F., 1987. A comparison of the equilibrium constants for the dissociation of carbonic acid in seawater media. Deep Sea Res. Part Oceanogr. Res. Pap. 34, 1733–1743.

40 Dickson, A.G., 1990. Thermodynamics of the dissociation of boric acid in synthetic seawater from 273.15 to 318.15 K. Deep Sea Res. Part Oceanogr. Res. Pap. 37, 755–766.

Eberlein, T., Van de Waal, D.B., Rost, B., 2014. Differential effects of ocean acidification on carbon acquisition in two bloom-forming dinoflagellate species. Physiol. Plant. 151, 468– 479. https://doi.org/10.1111/ppl.12137

Errera, R.M., Bourdelais, A., Drennan, M.A., Dodd, E.B., Henrichs, D.W., Campbell, L., 2010. Variation in brevetoxin and brevenal content among clonal cultures of Karenia brevis may influence bloom toxicity. Toxicon 55, 195–203. https://doi.org/10.1016/j.toxicon.2009.07.013

Errera, R.M., Campbell, L., 2011. Osmotic stress triggers toxin production by the dinoflagellate Karenia brevis. Proc. Natl. Acad. Sci. 108, 10597–10601. https://doi.org/10.1073/pnas.1104247108

Errera, R.M., Yvon-Lewis, S., Kessler, J.D., Campbell, L., 2014. Reponses of the dinoflagellate Karenia brevis to climate change: pCO2 and sea surface temperatures. Harmful Algae 37, 110–116. https://doi.org/10.1016/j.hal.2014.05.012

Espie, G.S., Colman, B., 1986. Inorganic Carbon Uptake during Photosynthesis: I. A Theoretical Analysis Using the Isotopic Disequilibrium Technique. PLANT Physiol. 80, 863–869. https://doi.org/10.1104/pp.80.4.863

Field, C.B., Behrenfeld, M.J., Randerson, J.T., Falkowski, P., 1998. Primary Production of the Biosphere: Integrating Terrestrial and Oceanic Components. Science 281, 237–240. https://doi.org/10.1126/science.281.5374.237

Fleming, L.E., Backer, L.C., Baden, D.G., 2005. Overview of aerosolized Florida red tide toxins: exposures and effects. Environ. Health Perspect. 618–620.

Flewelling, L.J., Naar, J.P., Abbott, J.P., Baden, D.G., Barros, N.B., Bossart, G.D., Bottein, M.- Y.D., Hammond, D.G., Haubold, E.M., Heil, C.A., others, 2005. Brevetoxicosis: red tides and marine mammal mortalities. Nature 435, 755–756.

Fu, F., Place, A., Garcia, N., Hutchins, D., 2010. CO2 and phosphate availability control the toxicity of the harmful bloom dinoflagellate Karlodinium veneficum. Aquat. Microb. Ecol. 59, 55–65. https://doi.org/10.3354/ame01396

Gao, K., Xu, J., Gao, G., Li, Y., Hutchins, D.A., Huang, B., Wang, L., Zheng, Y., Jin, P., Cai, X., Häder, D.-P., Li, W., Xu, K., Liu, N., Riebesell, U., 2012. Rising CO2 and increased light exposure synergistically reduce marine primary productivity. Nat. Clim. Change. https://doi.org/10.1038/nclimate1507

Giordano, M., Beardall, J., Raven, J.A., 2005. CO2 Concentrating Mechanisms in Algae:Mechanisms, Environmental Modulation, and Evolution. Annu. Rev. Plant Biol. 56, 99–131. https://doi.org/10.1146/annurev.arplant.56.032604.144052

41 Glibert, P., Anderson, D., Gentien, P., Granéli, E., Sellner, K., 2005. The Global, Complex Phenomena of Harmful Algal Blooms. Oceanography 18, 136–147. https://doi.org/10.5670/oceanog.2005.49

Glibert, P., Burkholder, J., Kana, T., Alexander, J., Skelton, H., Shilling, C., 2009. Grazing by Karenia brevis on Synechococcus enhances its growth rate and may help to sustain blooms. Aquat. Microb. Ecol. 55, 17–30. https://doi.org/10.3354/ame01279

Guillard, R.R.L., Ryther, J.H., 1962. Studies of marine planktonic diatoms: I. Cyclotella Nana Hustedt, and Detonula Confervacea (Cleve) Gran. Can. J. Microbiol. 8, 229–239. https://doi.org/10.1139/m62-029

Hackett, J.D., Anderson, D.M., Erdner, D.L., Bhattacharya, D., 2004. Dinoflagellates: a remarkable evolutionary experiment. Am. J. Bot. 91, 1523–1534. https://doi.org/10.3732/ajb.91.10.1523

Hallegraeff, G.M., 2010. Ocean climate change, phytoplankton community responses, and harmful algal blooms: a formidable predictive challenge. J. Phycol. 46, 220–235. https://doi.org/10.1111/j.1529-8817.2010.00815.x

Hansen, P., Lundholm, N., Rost, B., 2007. Growth limitation in marine red-tide dinoflagellates: effects of pH versus inorganic carbon availability. Mar. Ecol. Prog. Ser. 334, 63–71. https://doi.org/10.3354/meps334063

Hansen, P.J., 2002. Effect of high pH on the growth and survival of marine phytoplankton: implications for species succession. Aquat. Microb. Ecol. 28, 279–288.

Hardison, D.R., Sunda, W.G., Shea, D., Litaker, R.W., 2013. Increased Toxicity of Karenia brevis during Phosphate Limited Growth: Ecological and Evolutionary Implications. PLoS ONE 8, e58545. https://doi.org/10.1371/journal.pone.0058545

Hardison, D.R., Sunda, W.G., Tester, P.A., Shea, D., Litaker, W.R., 2014. Increased cellular brevetoxins in the red tide dinoflagellate Karenia brevis under CO2 limitation of growth rate: Evolutionary implications and potential effects on bloom toxicity. Limnol. Oceanogr. 59, 560–577. https://doi.org/10.4319/lo.2014.59.2.0560

Hardison, D.R., Sunda, W.G., Wayne Litaker, R., Shea, D., Tester, P.A., 2012. Nitrogen limitation increases brevetoxins in Karenia brevis (Dinophyceae): Implications for bloom toxicity. J. Phycol. 48, 844–858. https://doi.org/10.1111/j.1529-8817.2012.01186.x

Hattenrath-Lehmann, T.K., Smith, J.L., Wallace, R.B., Merlo, L.R., Koch, F., Mittelsdorf, H., Goleski, J.A., Anderson, D.M., Gobler, C.J., 2015. The effects of elevated CO2 on the growth and toxicity of field populations and cultures of the saxitoxin-producing dinoflagellate, Alexandrium fundyense: Effects of elevated CO2 on A. fundyense. Limnol. Oceanogr. 60, 198–214. https://doi.org/10.1002/lno.10012

Heil, C.A., Dixon, L.K., Hall, E., Garrett, M., Lenes, J.M., O’Neil, J.M., Walsh, B.M., Bronk, D.A., Killberg-Thoreson, L., Hitchcock, G.L., Meyer, K.A., Mulholland, M.R., Procise, L., Kirkpatrick, G.J., Walsh, J.J., Weisberg, R.W., 2014. Blooms of Karenia brevis

42 (Davis) G. Hansen & Ø. Moestrup on the West Florida Shelf: Nutrient sources and potential management strategies based on a multi-year regional study. Harmful Algae 38, 127–140. https://doi.org/10.1016/j.hal.2014.07.016

Heureux, A.M.C., Young, J.N., Whitney, S.M., Eason-Hubbard, M.R., Lee, R.B.Y., Sharwood, R.E., Rickaby, R.E.M., 2017. The role of Rubisco kinetics and pyrenoid morphology in shaping the CCM of haptophyte microalgae. J. Exp. Bot. 68, 3959–3969. https://doi.org/10.1093/jxb/erx179

Hitchcock, G.L., Kirkpatrick, G., Lane, P.V.Z., Langdon, C., 2014. Comparative diel oxygen cycles preceding and during a Karenia bloom in Sarasota Bay, Florida, USA. Harmful Algae 38, 95–100. https://doi.org/10.1016/j.hal.2014.05.010

Hoagland, P., Jin, D., Polansky, L.Y., Kirkpatrick, B., Kirkpatrick, G., Fleming, L.E., Reich, A., Watkins, S.M., Ullmann, S.G., Backer, L.C., 2009. The costs of respiratory illnesses arising from Florida Gulf Coast Karenia brevis blooms. Environ. Health Perspect. 117, 1239.

Hoins, M., Eberlein, T., Groβmann, C.H., Brandenburg, K., Reichart, G.-J., Rost, B., Sluijs, A., Van de Waal, D.B., 2016. Combined Effects of Ocean Acidification and Light or Nitrogen Availabilities on 13C Fractionation in Marine Dinoflagellates. PLOS ONE 11, e0154370. https://doi.org/10.1371/journal.pone.0154370

Hopkinson, B.M., Dupont, C.L., Allen, A.E., Morel, F.M.M., 2011. Efficiency of the CO2- concentrating mechanism of diatoms. Proc. Natl. Acad. Sci. 108, 3830–3837. https://doi.org/10.1073/pnas.1018062108

Hopkinson, B.M., Meile, C., Shen, C., 2013. Quantification of Extracellular Carbonic Anhydrase Activity in Two Marine Diatoms and Investigation of Its Role. PLANT Physiol. 162, 1142–1152. https://doi.org/10.1104/pp.113.217737

Hopkinson, B.M., Young, J.N., Tansik, A.L., Binder, B.J., 2014. The Minimal CO2- Concentrating Mechanism of Prochlorococcus spp. MED4 Is Effective and Efficient. PLANT Physiol. 166, 2205–2217. https://doi.org/10.1104/pp.114.247049

Hutchins, D.A., Fu, F.-X., Webb, E.A., Walworth, N., Tagliabue, A., 2013. Taxon-specific response of marine nitrogen fixers to elevated carbon dioxide concentrations. Nat. Geosci. 6, 790–795. https://doi.org/10.1038/ngeo1858

Killberg-Thoreson, L., Mulholland, M.R., Heil, C.A., Sanderson, M.P., O’Neil, J.M., Bronk, D.A., 2014. Nitrogen uptake kinetics in field populations and cultured strains of Karenia brevis. Harmful Algae 38, 73–85. https://doi.org/10.1016/j.hal.2014.04.008

Kolber, Z., Falkowski, P.G., 1993. Use of active fluorescence to estimate phytoplankton photosynthesis in situ. Limnol. Oceanogr. 38, 1646–1665.

− Kottmeier, D.M., Rokitta, S.D., Tortell, P.D., Rost, B., 2014. Strong shift from HCO3 to CO2 uptake in Emiliania huxleyi with acidification: new approach unravels acclimation versus

43 short-term pH effects. Photosynth. Res. 121, 265–275. https://doi.org/10.1007/s11120- 014-9984-9

Kranz, S.A., Eichner, M., Rost, B., 2011. Interactions between CCM and N2 fixation in Trichodesmium. Photosynth. Res. 109, 73–84. https://doi.org/10.1007/s11120-010-9611- 3

Kranz, S.A., Young, J.N., Hopkinson, B.M., Goldman, J.A.L., Tortell, P.D., Morel, F.M.M., 2015. Low temperature reduces the energetic requirement for the CO2 concentrating mechanism in diatoms. New Phytol. 205, 192–201. https://doi.org/10.1111/nph.12976

Landsberg, J.H., 2002. The effects of harmful algal blooms on aquatic organisms. Rev. Fish. Sci. 10, 113–390.

Langer, G., Probert, I., Nehrke, G., Ziveri, P., 2011. The morphological response of Emiliania huxleyi to seawater carbonate chemistry changes: an inter-strain comparison. J. Nannoplankton Res. 32, 29–34.

Lawrenz, E., Silsbe, G., Capuzzo, E., Ylöstalo, P., Forster, R.M., Simis, S.G.H., Prášil, O., Kromkamp, J.C., Hickman, A.E., Moore, C.M., Forget, M.-H., Geider, R.J., Suggett, D.J., 2013. Predicting the Electron Requirement for Carbon Fixation in Seas and Oceans. PLoS ONE 8, e58137. https://doi.org/10.1371/journal.pone.0058137

Lekan, D.K., Tomas, C.R., 2010. The brevetoxin and brevenal composition of three Karenia brevis clones at different salinities and nutrient conditions. Harmful Algae 9, 39–47. https://doi.org/10.1016/j.hal.2009.07.004

Lenes, J.M., Darrow, B.P., Cattrall, C., Heil, C.A., Callahan, M., Vargo, G.A., Byrne, R.H., Prospero, J.M., Bates, D.E., Fanning, K.A., Walsh, J.J., 2001. Iron fertilization and the Trichodesmium response on the West Florida shelf. Limnol. Oceanogr. 46, 1261–1277.

Lenes, J.M., Prospero, J.M., Landing, W.M., Virmani, J.I., Walsh, J.J., 2012. A model of Saharan dust deposition to the eastern Gulf of Mexico. Mar. Chem. 134–135, 1–9. https://doi.org/10.1016/j.marchem.2012.02.007

Losh, J.L., Young, J.N., Morel, F.M.M., 2013. Rubisco is a small fraction of total protein in marine phytoplankton. New Phytol. 198, 52–58. https://doi.org/10.1111/nph.12143

Mackey, K., Morris, J.J., Morel, F., Kranz, S., 2015. Response of Photosynthesis to Ocean Acidification. Oceanography 25, 74–91. https://doi.org/10.5670/oceanog.2015.33

Magaña, H.A., Villareal, T.A., 2006. The effect of environmental factors on the growth rate of Karenia brevis (Davis) G. Hansen and Moestrup. Harmful Algae 5, 192–198. https://doi.org/10.1016/j.hal.2005.07.003

Maier Brown, A.F., Dortch, Q., Dolah, F.M.V., Leighfield, T.A., Morrison, W., Thessen, A.E., Steidinger, K., Richardson, B., Moncreiff, C.A., Pennock, J.R., 2006. Effect of salinity on the distribution, growth, and toxicity of Karenia spp. Harmful Algae 5, 199–212. https://doi.org/10.1016/j.hal.2005.07.004

44 Martin, D.F., Carnahan, R.P., Krzanowski, J.J., 2003. Effect of Shear Forces on The Release of Brevetoxins From Karenia brevis. Fla. Sci. 109–112.

Meehl, G.A., Stocker, T.F., Collins, W.D., Friedlingstein, P., Gaye, T., Gregory, J.M., Kitoh, A., Knutti, R., Murphy, J.M., Noda, A., others, 2007. Global climate projections.

Mehrbach, C., Culberson, C., Hawley, J., Pytkowicx, R., 1973. Measurement of the apparent dissociation constants of carbonic acid in seawater at atmospheric pressure. Limnol. Oceanogr. 18, 897–907.

Monroe, E.A., Johnson, J.G., Wang, Z., Pierce, R.K., Van Dolah, F.M., 2010. Characterization and expression of nuclear-encoded polyketide synthases in the brevetoxin-producing dinoflagellate Karenia brevis. J. Phycol. 46, 541–552. https://doi.org/10.1111/j.1529- 8817.2010.00837.x

Moore, S.K., Trainer, V.L., Mantua, N.J., Parker, M.S., Laws, E.A., Backer, L.C., Fleming, L.E., 2008. Impacts of climate variability and future climate change on harmful algal blooms and human health. Environ. Health 7, S4. https://doi.org/10.1186/1476-069X-7-S2-S4

Morse, D., Salois, P., Markovic, P., Hastings, J., 1995. A nuclear-encoded form II RuBisCO in dinoflagellates. Science 268, 1622–1624. https://doi.org/10.1126/science.7777861

Mulholland, M.R., Bernhardt, P.W., Ozmon, I., Procise, L.A., Garrett, M., O’Neil, J.M., Heil, C.A., Bronk, D.A., 2014. Contribution of diazotrophy to nitrogen inputs supporting Karenia brevis blooms in the Gulf of Mexico. Harmful Algae 38, 20–29. https://doi.org/10.1016/j.hal.2014.04.004

Noguchi, T., Hatta, M., Yamanaka, T., Okamura, K., 2013. Fast Measurement of Dissolved Inorganic Carbon Concentration for Small-Volume Interstitial Water by Acid Extraction and Nondispersive Infrared Gas Analysis. Anal. Sci. 29, 9–13. https://doi.org/10.2116/analsci.29.9

Oxborough, K., Moore, C.M., Suggett, D.J., Lawson, T., Chan, H.G., Geider, R.J., 2012. Direct estimation of functional PSII reaction center concentration and PSII electron flux on a volume basis: a new approach to the analysis of Fast Repetition Rate fluorometry (FRRf) data: Analysis of FRRf data: a new approach. Limnol. Oceanogr. Methods 10, 142–154. https://doi.org/10.4319/lom.2012.10.142

Pachauri, R.K., Allen, M.R., Barros, V., Broome, J., Cramer, W., Christ, R., Church, J., Clarke, L., Dahe, Q., Dasgupta, P., others, 2014. Climate change 2014: synthesis Report. Contribution of working groups I, II and III to the fifth assessment report of the intergovernmental panel on climate change. IPCC.

Petrou, K., Kranz, S.A., Trimborn, S., Hassler, C.S., Ameijeiras, S.B., Sackett, O., Ralph, P.J., Davidson, A.T., 2016. Southern Ocean phytoplankton physiology in a changing climate. J. Plant Physiol. 203, 135–150. https://doi.org/10.1016/j.jplph.2016.05.004

45 Pierce, R.H., Henry, M.S., 2008. Harmful algal toxins of the Florida red tide (Karenia brevis): natural chemical stressors in South Florida coastal ecosystems. Ecotoxicology 17, 623– 631. https://doi.org/10.1007/s10646-008-0241-x

Pierrot, D.E., Wallace, D.W.R., Lewis, E., 2011. MS Excel Program Developed for CO2 System Calculations. https://doi.org/10.3334/CDIAC/otg.CO2SYS_XLS_CDIAC105a

Price, N.M., Harrison, G.I., Hering, J.G., Hudson, R.J., Nirel, P.M.V., Palenik, B., Morel, F.M.M., 1989. Preparation and Chemistry of the Artificial Algal Culture Medium Aquil. Biol. Oceanogr. 6, 443–461. https://doi.org/10.1080/01965581.1988.10749544

Raateoja, M., Seppälä, J., Kuosa, H., 2004. Bio-optical modelling of primary production in the SW Finnish coastal zone, Baltic Sea: fast repetition rate fluorometry in Case 2 waters. Mar. Ecol. Prog. Ser. 267, 9–26.

Ratti, S., Giordano, M., Morse, D., 2007. CO2-concentrating mechanisms of the potentially toxic dinoflagellate Protoceratium Reticulatum (Dinophyceae, Gonyaulacales). J. Phycol. 43, 693–701. https://doi.org/10.1111/j.1529-8817.2007.00368.x

Raven, J.A., Giordano, M., Beardall, J., Maberly, S.C., 2012. Algal evolution in relation to atmospheric CO2: carboxylases, carbon-concentrating mechanisms and carbon oxidation cycles. Philos. Trans. R. Soc. B Biol. Sci. 367, 493–507. https://doi.org/10.1098/rstb.2011.0212

Raven, J.A., Johnston, A.M., 1991. Mechanisms of inorganic-carbon acquisition in marine phytoplankton and their implications for the use of other resources. Limnol. Oceanogr. 36, 1701–1714. https://doi.org/10.4319/lo.1991.36.8.1701

Rost, B., Kranz, S., Richter, K.-U., Tortell, P.D., 2007. Isotope disequilibrium and mass spectrometric studies of inorganic carbon acquisition by phytoplankton. Limnol. Oceanogr. Methods 5, 328–337.

Rost, B., Zondervan, I., Wolf-Gladrow, D., 2008. Sensitivity of phytoplankton to future changes in ocean carbonate chemistry: current knowledge, contradictions and research directions. Mar. Ecol. Prog. Ser. 373, 227–237. https://doi.org/10.3354/meps07776

Roth, P.B., Twiner, M.J., Wang, Z., Bottein Dechraoui, M.-Y., Doucette, G.J., 2007. Fate and distribution of brevetoxin (PbTx) following lysis of Karenia brevis by algicidal bacteria, including analysis of open A-ring derivatives. Toxicon 50, 1175–1191. https://doi.org/10.1016/j.toxicon.2007.08.003

Schaeffer, B.A., Kamykowski, D., McKay, L., Sinclair, G., Milligan, E.J., 2007. A comparison of photoresponse among ten different Karenia brevis (Dinophyceae) isolates. J. Phycol. 43, 702–714.

Schaum, E., Rost, B., Millar, A.J., Collins, S., 2013. Variation in plastic responses of a globally distributed picoplankton species to ocean acidification. Nat. Clim. Change 3, 298–302. https://doi.org/10.1038/nclimate1774

46 Silverman, D.N., 1982. Carbonic anhydrase: Oxygen-18 exchange catalyzed by an enzyme with rate-contributing Proton-transfer steps. Methods Enzymol. 87, 732–752.

Simon, N., Cras, A.-L., Foulon, E., Lemée, R., 2009. Diversity and evolution of marine phytoplankton. C. R. Biol. 332, 159–170. https://doi.org/10.1016/j.crvi.2008.09.009

Solomon, S., Qin, D., Manning, M., Averyt, K., Marquis, M., 2007. Climate change 2007-the physical science basis: Working group I contribution to the fourth assessment report of the IPCC. Cambridge university press.

Steidinger, K.A., 2009. Historical perspective on Karenia brevis red tide research in the Gulf of Mexico. Harmful Algae 8, 549–561.

Stoecker, D.K., Hansen, P.J., Caron, D.A., Mitra, A., 2017. Mixotrophy in the Marine Plankton. Annu. Rev. Mar. Sci. 9, 311–335. https://doi.org/10.1146/annurev-marine-010816- 060617

Sun, J., Hutchins, D.A., Feng, Y., Seubert, E.L., Caron, D.A., Fu, F.-X., 2011. Effects of changing pCO2 and phosphate availability on domoic acid production and physiology of the marine harmful bloom diatom Pseudo-nitzschia multiseries. Limnol. Oceanogr. 56, 829–840. https://doi.org/10.4319/lo.2011.56.3.0829

Tatters, A.O., Flewelling, L.J., Fu, F., Granholm, A.A., Hutchins, D.A., 2013. High CO2 promotes the production of paralytic shellfish poisoning toxins by Alexandrium catenella from Southern California waters. Harmful Algae 30, 37–43. https://doi.org/10.1016/j.hal.2013.08.007

Tatters, A.O., Fu, F.-X., Hutchins, D.A., 2012. High CO2 and Silicate Limitation Synergistically Increase the Toxicity of Pseudo-nitzschia fraudulenta. PLoS ONE 7, e32116. https://doi.org/10.1371/journal.pone.0032116

Tortell, P.D., 2000. Evolutionary and ecological perspectives on carbon acquisition in phytoplankton. Limnol. Oceanogr. 45, 744–750. https://doi.org/10.4319/lo.2000.45.3.0744

Tortell, P.D., Trimborn, S., Li, Y., Rost, B., Payne, C.D., 2010. Inorganic carbon utilization by Ross Sea phytoplankton across natural and experimental CO2 gradients. J. Phycol. 46, 433–443.

Trimborn, S., Lundholm, N., Thoms, S., Richter, K.-U., Krock, B., Hansen, P.J., Rost, B., 2008. Inorganic carbon acquisition in potentially toxic and non-toxic diatoms: the effect of pH- induced changes in seawater carbonate chemistry. Physiol. Plant. 133, 92–105. https://doi.org/10.1111/j.1399-3054.2007.01038.x

Uppström, L.R., 1974. The boron/chlorinity ratio of deep-sea water from the Pacific Ocean, in: Deep Sea Research and Oceanographic Abstracts. Elsevier, pp. 161–162.

47 van de Waal, D.B., Eberlein, T., Bublitz, Y., John, U., Rost, B., 2014. Shake it easy: a gently mixed continuous culture system for dinoflagellates. J. Plankton Res. 36, 889–894. https://doi.org/10.1093/plankt/fbt138

Van Dolah, F.M., Lidie, K.B., Monroe, E.A., Bhattacharya, D., Campbell, L., Doucette, G.J., Kamykowski, D., 2009. The Florida red tide dinoflagellate Karenia brevis: New insights into cellular and molecular processes underlying bloom dynamics. Harmful Algae 8, 562–572. https://doi.org/10.1016/j.hal.2008.11.004

Van Dolah, F.M., Lidie, K.B., Morey, J.S., Brunelle, S.A., Ryan, J.C., Monroe, E.A., Haynes, B.L., 2007. Microarray analysis of diurnal- and circadian-regulated genes in the Florida red-tide dinoflagellate Karenia brevis (Dinophyceae). J. Phycol. 43, 741–752. https://doi.org/10.1111/j.1529-8817.2007.00354.x

Vargo, G.A., 2009. A brief summary of the physiology and ecology of Karenia brevis Davis (G. Hansen and Moestrup comb. nov.) red tides on the West Florida Shelf and of hypotheses posed for their initiation, growth, maintenance, and termination. Harmful Algae 8, 573– 584. https://doi.org/10.1016/j.hal.2008.11.002

Walsh, J.J., Jolliff, J.K., Darrow, B.P., Lenes, J.M., Milroy, S.P., Remsen, A., Dieterle, D.A., Carder, K.L., Chen, F.R., Vargo, G.A., Weisberg, R.H., Fanning, K.A., Muller-Karger, F.E., Shinn, E., Steidinger, K.A., Heil, C.A., Tomas, C.R., Prospero, J.S., Lee, T.N., Kirkpatrick, G.J., Whitledge, T.E., Stockwell, D.A., Villareal, T.A., Jochens, A.E., Bontempi, P.S., 2006. Red tides in the Gulf of Mexico: Where, when, and why? J. Geophys. Res. 111. https://doi.org/10.1029/2004JC002813

Wang, D.-Z., 2008. Neurotoxins from Marine Dinoflagellates: A Brief Review. Mar. Drugs 6, 349–371. https://doi.org/10.3390/md20080016

Wiliams, P.J.L., Robertson, J.E., 1991. Overall planktonic oxygen and carbon dioxide metabolisms: the problem of reconciling observations and calculations of photosynthetic quotients. J. Plankton Res. https://doi.org/10.1093/oxfordjournals.plankt.a042366

Winder, M., Sommer, U., 2012. Phytoplankton response to a changing climate. Hydrobiologia 698, 5–16. https://doi.org/10.1007/s10750-012-1149-2

Wu, Y., Gao, K., Riebesell, U., 2010. CO2-induced seawater acidification affects physiological performance of the marine diatom Phaeodactylum tricornutum. Biogeosciences 7, 2915– 2923. https://doi.org/10.5194/bg-7-2915-2010

Yang, G., Gao, K., 2012. Physiological responses of the marine diatom Thalassiosira pseudonana to increased pCO2 and seawater acidity. Mar. Environ. Res. 79, 142–151. https://doi.org/10.1016/j.marenvres.2012.06.002

Yoon, H.S., Hackett, J.D., Bhattacharya, D., 2002. A single origin of the peridinin- and fucoxanthin-containing plastids in dinoflagellates through tertiary endosymbiosis. Proc. Natl. Acad. Sci. 99, 11724–11729. https://doi.org/10.1073/pnas.172234799

48 Yoon, H.S., Hackett, J.D., Van Dolah, F.M., Nosenko, T., Lidie, K.L., Bhattacharya, D., 2005. Tertiary endosymbiosis driven genome evolution in dinoflagellate algae. Mol. Biol. Evol. 22, 1299–1308.

Zhang, H., Byrne, R.H., 1996. Spectrophotometric pH measurements of surface seawater at in- situ conditions: absorbance and protonation behavior of thymol blue. Mar. Chem. 52, 17– 25. https://doi.org/10.1016/0304-4203(95)00076-3

49 BIOGRAPHICAL SKETCH

Tristyn Bercel

Education Florida State University, Tallahassee, FL August 2015 – Current Degree: Biogeochemical Oceanography – PhD Supervisor: Dr. Sven Kranz Florida State University, Tallahassee, FL August 2015 – September 2018 Degree: Biogeochemical Oceanography – MSc Supervisor: Dr. Sven Kranz Arizona State University, Tempe, AZ August 2011 - May 2015 Degree: Bachelor of Science Degree honors: Summa Cum Laude Degree date: May 11, 2015 Major: Earth and Space Exploration Concentration: Astrobiology and Biogeosciences Minor: Biological Sciences Minor: Sustainability Thesis: A 16S rRNA gene sequencing approach to investigating microbial community composition of “mysterious lakes” in the Badain Jaran Desert

Publications/Presentations Elser, J.J.; Bercel, T.; Learned, J.; Poret-Peterson, A.; Raymond, J.; Ze, R.; Niu, D.; Fu, H, (2015), “MYSTERIOUS LAKES” AMID MEGADUNES: A LIMNOLOGICAL EXPLORATION OF THE GROUNDWATER-FED PONDS AND LAKES OF BADAIN JARAN, NORTH-CENTRAL CHINA.", Association of the Sciences of Limnology and Oceanography 2015 Conference Talk

Kranz, S. A., & Bercel, T, (2016, August), “Photosynthesis and carbon acquisition of the red-tide dinoflagellate Karenia brevis under ambient and elevated CO2 concentrations.” Poster presentation at The IXth International Symposium on Inorganic Carbon Utilization by Aquatic Photosynthetic Organisms, CCM International Scientific Committee, Cambridge, UK. (International)

Kranz, S. A., & Bercel, T. (2017, January). “CO2 effects on the marine dinoflagellate Karenia brevis - carbon acquisition and photophysiology”. Presentation at Xiamen Symposium on Marine Environmental Sciences (XMAS) - The Changing Ocean Environment: From a Multidisciplinary Perspective, State Key Laboratory of Marine Environmental Science (MEL) of Xiamen University, Xiamen, China. (International) Retrieved from http://mel.xmu.edu.cn/conference/3XMAS/

A. T. Poret-Peterson1a*, T. Bercel1b, J. Learned2c, Z. Ren2d, N. Decao3, F. Hua3, J. Raymond1, and J. J. Elser2d*, (2018). “Mysterious lakes” amid megadunes: a limnological

50 and microbiological exploration of groundwater-fed lakes and ponds of Badain Jaran, China (In review)

T. Bercel, S. Kranz, (2018), “Advancing our understanding of light quality on phytoplankton productivity using LEDs - implications for modeling based on laboratory experiments” Ocean Sciences Meeting 2018 Poster

Kranz, S. A., & Bercel, T., (presented 2018, September), “The light and dark side of red tides: A physiological perspective on how cellular photosynthesis and carbon acquisition can lead to sustained blooms of Karenia brevis.” Oral presentation at the Ecology and Evolution Seminar, FSU-BIO

Experience Graduate Research Assistant August 2015-Current Florida State University, Tallahassee, FL

Graduate Teaching Assistant EVR 1001 – Introduction to Environmental Sciences January 2017-Current Florida State University, Tallahassee, FL

Graduate Lecturer / Teaching Assistant OCE 1001 – Elementary Oceanography August 2016-December 2016 Florida State University, Tallahassee, FL

Undergraduate Lab Research June 2014 - July 2015 Senior Thesis Research under Jason Raymond Arizona State University, Tempe, AZ

SI Leader for General Biology and General Genetics January 2013 - May 2015 Arizona State University, Tempe, AZ

Undergraduate Teaching Assistant January 2015 - March 2015 Arizona State University, Tempe, AZ

Undergraduate Lab Research May 2013 - May 2014 Ariel Anbar’s Astrobiology Lab Arizona State University, Tempe, AZ

Undergraduate Teaching Assistant August 2013 - May 2014 Arizona State University, Tempe, AZ

Geochemical and Biological Field Work in Yellowstone National Park July 2013 Everett Shock’s GEOPIG Group Arizona State University, Tempe, AZ