Influence of Water Quality and Climate Variables on Growth of the Harmful Alga, Prymnesium Parvum by Md Rakib Hasan Rashel, B.S

Influence of Water Quality and Climate Variables on Growth of the Harmful Alga, Prymnesium parvum

Md Rakib Hasan Rashel, B.S., M.S.

A Dissertation

Biology

Submitted to the Graduate Faculty of Texas Tech University in Partial Fulfillment of the Requirements for the Degree of

DOCTOR OF PHILOSOPHY

Approved

Dr. Reynaldo Patiño Chair of Committee

Dr. Timothy B. Grabowski

Dr. Randall M. Jeter

Dr. Michael J. San Francisco

Dr. John C. Zak

Dr. Mark Sheridan Dean of the Graduate School

May 2020

ACKNOWLEDGEMENTS

I would like to express my heartfelt gratitude to my advisor and mentor, Dr.

Reynaldo Patiño, for giving me the opportunity to work in his laboratory, for his continuous support and guidance, and great patience on every aspect of my doctoral program.

I would also like to extend my gratitude to my other committee members, Dr.

Timothy B. Grabowski, Dr. Randall M. Jeter, Dr. Michael J. San Francisco, Dr. John C.

Zak for their time, support, valuable advice and critical assessments I needed during this program.

I would like to thank Dr. Matthew VanLandeghem for his time and effort to teach me useful laboratory methods relevant to this project. A special thanks go to my Dr.

Tirhas Hailu for her help, friendly suggestions during my doctoral program. I would like to thank Rita Jones and Amanda Garcia from the Texas Cooperative Fish and Wildlife

Unit (TCFWRU) for their administrative support. I would also like to thank my former and colleagues and lab mates, Seydou Toe, Emily Richardson, Brittanie Dabney,

Mousumi Mary, Shisbeth Tabora, and Lindsay Williams for their help, support, and shared resources.

I would also thank Texas Cooperative Fish and Wildlife Unit, Department of

Biological Sciences, for providing the opportunity to work and financial support in the form of a Teaching/Research Assistantship for all these years. My research was

Texas Tech University, Rakib H. Rashel, May 2020 supported by Texas Tech University intramural funding. I would like to thank TTUAB and TechASM for providing me travel funds and Grants-in-Aid for research.

I am grateful to my family, especially my parents and elder brothers, for all their love and support. Last but not least, I am thankful to my wife, Eshrat Labony, for her tremendous support throughout my doctoral degree.

iii

Texas Tech University, Rakib H. Rashel, May 2020

TABLE OF CONTENTS ACKNOWLEDGEMENTS ...... ii ABSTRACT ...... viii LIST OF TABLES ...... xiv LIST OF FIGURES ...... xv LIST OF ABBREVIATIONS ...... xvii I. INTRODUCTION ...... 1 Biology of harmful algae ...... 1 Cyanobacteria and blue green algal bloom ...... 3 Dinoflagellates and Red tide ...... 5 Brown Tide ...... 6 Golden alga ...... 6 Global occurrence of golden alga ...... 7 Environmental factors associated with golden alga blooms ...... 9 Salinity ...... 9 Inoculum size ...... 10 Sulfate ...... 11 Fluoride ...... 12 Water hardness ...... 12 Temperature ...... 13 pH ...... 13 Nutrients ...... 14 Carbon dioxide ...... 16 Significance of the present study ...... 17 Research objectives ...... 18 Literature Cited ...... 19 Ⅱ. INFLUENCE OF GENETIC BACKGROUND, SALINITY, AND INOCULUM SIZE ON GROWTH OF THE ICHTHYOTOXIC GOLDEN ALGA (PRYMNESIUM PARVUM) ...... 29 Abstract ...... 29 1. Introduction ...... 30 2. Materials and methods ...... 33 iv

Texas Tech University, Rakib H. Rashel, May 2020

2.1. Basic culture procedures ...... 33 2.2. Experimental cultures ...... 35 2.3. Analytical procedures ...... 36 3. Results ...... 38 3.1. Influence of inoculum size and genetic strain ...... 38 3.2. Influence of salinity, genetic strain, and temperature ...... 39 4. Discussion ...... 42 4.1. Inoculum size and genetic strain ...... 42 4.2. Salinity, genetic strain, and temperature ...... 43 4.3. Early cell density ...... 45 4.4. Ecological considerations and conclusions ...... 46 Literature Cited ...... 49 Tables ...... 53 Figures ...... 54 Ⅲ. GROWTH RESPONSE OF THE ICHTHYOTOXIC HAPTOPHYTE, PRYMNESIUM PARVUM CARTER, TO CHANGES IN SULFATE AND FLUORIDE CONCENTRATIONS ...... 62 Abstract ...... 62 1. Introduction ...... 63 2. Materials and methods ...... 66 2.1. Basic culture procedures ...... 66 2.2. Experimental design ...... 67 2.3. Analytical procedures ...... 70 3. Results ...... 72 3.1. Effects of changes in sulfate concentration at low salinity ...... 72 3.2. Effects of salinity under different major salt scenarios ...... 74 3.3. Effects of differences in sulfate concentration at high salinity ...... 75 3.4. Effects of changes in fluoride concentration at low salinity ...... 76 4. Discussion ...... 77 4.1. Effects of sulfate and other ions at low salinity ...... 78 4.2. Effects of sulfate and other ions at high salinity ...... 79 4.3. Effects of fluoride ...... 81

Texas Tech University, Rakib H. Rashel, May 2020

5. Summary and Conclusions ...... 82 Literature Cited ...... 85 Tables ...... 90 Figures ...... 94 Ⅳ. RATIO OF INORGANIC TO ORGANIC NITROGEN INFLUENCES GROWTH OF THE HAPTOPHYTE, PRYMNESIUM PARVUM ...... 100 Abstract ...... 100 1. Introduction ...... 101 2. Materials and Methods ...... 103 2.1. Basic culture procedures ...... 103 2.2. Experimental cultures ...... 104 2.3. Analytical procedures ...... 105 3. Results ...... 106 4. Discussion ...... 108 Literature Cited ...... 111 Tables ...... 114 Figures ...... 115 Ⅴ. GROWTH OF THE HARMFUL HAPTOPHYTE, PRYMNESIUM PARVUM, UNDER PAST, PRESENT AND PROJECTED FUTURE AIR CONCENTRATIONS OF CARBON DIOXIDE ...... 121 Abstract ...... 121 1. Introduction ...... 122 2. Materials and Methods ...... 124 2.1. Basic culture procedures ...... 124 2.2. Experimental design ...... 125 2.3. Analytical procedures ...... 126 3. Results ...... 128 3.1. Effects of carbon dioxide at low salinity ...... 128 3.2. Effects of carbon dioxide at high salinity ...... 129 4. Discussion ...... 130 Literature Cited ...... 134 Tables ...... 138

Texas Tech University, Rakib H. Rashel, May 2020

Figures ...... 139 Ⅵ. CONCLUSION ...... 143 APPENDIX ...... 148

vii

Texas Tech University, Rakib H. Rashel, May 2020

ABSTRACT

Golden alga Prymnesium parvum Carter is a euryhaline, ichthyotoxic haptophyte

(Chromista) that typically inhabits marine and estuarine habitats, but it has recently

invaded brackish inland waters where it is capable of forming fish-killing blooms and

cause major ecological damages. In recent years, some field studies reported new and sometimes unexpected information concerning environmental variables that associate with golden alga presence and abundance in inland waters of the USA. Important examples include (1) a biphasic association between golden alga abundance and salinity, where the association is hypothesized to be positive at low salinity and negative at high salinity; (2) a positive association between abundance and sulfate concentration; and (3) a positive association between abundance and organic nitrogen concentrations. These working hypotheses, however, are based on descriptive observations in the field and cannot be used as conclusive evidence of causal associations. There is a need, therefore, to test these field-generated hypotheses under the controlled environment of a laboratory, where only the independent variable or variables of interest are allowed to vary. A fourth question being addressed in this dissertation is whether changes in air CO2 concentration

can influence growth of golden alga by providing additional carbon for fixation. Results

with other algae have been inconsistent and this question has never been addressed for

golden alga.

To address the first working hypothesis, in Chapter 2 salinity (5–30 psu) effects

on golden alga growth were determined at a standard laboratory temperature (22 °C) and

one associated with natural blooms (13 °C). Studies reported a minimum initial cell

viii

Texas Tech University, Rakib H. Rashel, May 2020 density (inoculum size) requires for golden algal survival in field conditions, however, information regarding the inoculum-size effects in laboratory conditions is not available for this algal species. Inoculum-size effects were determined over a wide size range

(100–100,000 cells ml-1), as it may influence golden algal growth.

A strain widely distributed in the USA, UTEX-2797 was the primary study subject but another of limited distribution, UTEX-995 was used to evaluate growth responses in relation to genetic background. Variables examined were exponential growth rate (r), maximum cell density and, when inoculum size was held constant (100 cells ml-1), density at onset of exponential growth (early cell density). In UTEX-2797, maximum cell density increased as salinity increased from 5 to ~10–15 psu and declined thereafter regardless of temperature but r remained generally stable and only declined at salinity of 25–30 psu. In addition, maximum cell density correlated positively with r and early cell density, the latter also being numerically highest at salinity of 15 psu. In

UTEX-995, maximum cell density and r responded similarly to changes in salinity – they remained stable at salinity of 5–10 psu and 5–15 psu, respectively, and declined at higher salinity. Also, maximum cell density correlated with r but not early cell density.

Inoculum size positively and negatively influenced maximum cell density and r, respectively, in both strains and these effects were significant even when the absolute size difference was small (100 versus 1000 cells ml-1). When cultured under similar conditions, UTEX-2797 grew faster and to far higher density than UTEX-995. In conclusion, (1) UTEX-2797’s superior growth performance may explain its relatively wide distribution in the USA, (2) the biphasic growth response of UTEX-2797 to salinity variation, with peak abundance at salinity of 10–15 psu, generally mirrors golden alga

Texas Tech University, Rakib H. Rashel, May 2020 abundance-salinity associations in US inland waters, and (3) early cell density – whether artificially manipulated or naturally attained – can influence UTEX-2797 bloom potential.

2- Because of its presumed coastal/marine origin where SO4 levels are high, the

2- relatively high SO4 concentration of its brackish inland habitats, and the sensitivity of marine chromists to sulfur deficiency, Chapter 3 examined whether golden alga growth is

2- sensitive to SO4 concentration. Fluoride is a ubiquitous ion that has been reported at higher levels in golden alga habitat; thus, the influence of F- on growth also was

2- examined. In low-salinity (5 psu) artificial seawater medium, overall growth was SO4 -

-1 dependent up to 1000 mg l using MgSO4 or Na2SO4 as source; the influence on growth rate, however, was more evident with MgSO4. Transfer from 5 to 30 psu inhibited growth

2- when salinity was raised with NaCl but in the presence of seawater levels of SO4 , these effects were fully reversed with MgSO4 as source and only partially reversed with

Na2SO4. Growth inhibition was not observed after acute transfer to 30 psu in a commercial sea salt mixture. In 5-psu medium, F- inhibited growth at all concentrations

2- tested. These observations support the hypothesis that spatial differences in SO4 – but not F- – concentration help drive the inland distribution and growth of golden alga and also provide physiological relevance to reports of relatively high Mg2+ concentrations in golden alga habitat. At high salinity, however, the ability of sulfate to maintain growth under osmotic stress was weak and overshadowed by the importance of Mg2+. A

2- 2+ mechanistic understanding of growth responses of golden alga to SO4 , Mg and other ions at environmentally relevant levels and under different salinity scenarios will be necessary to clarify their ecophysiological and evolutionary relevance.

Texas Tech University, Rakib H. Rashel, May 2020

As a mixotroph, P. parvum can utilize organic and inorganic nitrogen (N) for growth but the relative importance of each fraction when present in combination is uncertain. Some field studies have suggested there is a positive association between organic N and golden alga distribution or abundance, but experimental evidence supporting this observation is unavailable. The objective of this study is to determine if different molar ratios of inorganic to organic N affect growth of golden alga in a standard culture medium (5 psu). Sodium nitrate was used as source of inorganic N and urea and glycine as sources of organic N. Concentrations of total N (880 μM) and phosphorus (36

μM) were kept constant at F/2 levels. The inorganic:organic N ratios tested were 1:0,

0.75:0.25, 0.5:0.5, 0.75:0.25, and 0:1. Cultures were conducted under standard conditions

(initial density, 100 cells ml-1; 22°C, ~6500 lux), and endpoints measured were early cell density (day 3, cells ml-1), exponential growth rate (r, day-1) and maximum cell density

(cells ml-1). Growth rate was unaffected by changes in inorganic:organic N ratio. Early and maximum cell density, however, increased gradually as the fraction of organic N increased from 0 to 0.75 followed by a decrease at 1, and this pattern was stronger when using glycine as source. In conclusion, while golden alga can grow in cultures supplemented exclusively with organic or inorganic N, optimal growth occurs when both are present and the organic fraction is predominant. These findings are consistent with field observations and provide context for a better understanding of the association between nutrient stoichiometry and golden alga growth.

Carbon dioxide is the primary source of carbon for photosynthetic fixation by plants and algae. Because it is highly soluble in water, changes in air CO2 concentration can lead to corresponding changes in dissolved CO2 concentration of surface waters. This

Texas Tech University, Rakib H. Rashel, May 2020

has led to concerns over the potential effects of the rising air CO2 concentration on

growth of harmful algae. Current knowledge suggests that these effects may be species-

specific and no information is available for Prymnesium parvum, a euryhaline haptophyte

that forms toxic blooms. This study determined and compared the effects on P. parvum

growth by air CO2 at concentrations of 280 (pre-industrial era), 400 (current), and 670 ppm (a projected scenario by 2100). Batch cultures were conducted in two different media, Artificial Seawater Medium at salinity of 5 and Instant Ocean® at salinity of 30.

Treatments were done in triplicate and experiments were conducted twice. Early (pre-

exponential) growth was not affected by air CO2. Exponential growth rates were

positively stimulated by air CO2 concentration in the higher but not the lower salinity

medium. In both media, maximum population density was strongly and positively

associated with CO2 concentration. Medium pH increased during incubation but to a

similar level in both media regardless of CO2 concentration. This increase in pH is likely

due to the onset of carbon limitation as cell populations neared their maximum density

which could not be overcome at even the highest air CO2 concentration tested. Our

results suggest that relative to pre-industrial times, current concentrations of atmospheric

CO2 may already be enhancing growth of P. parvum in the field and as CO2 levels continue to rise, so may the magnitude of this effect.

In conclusion, this current study results are consistent with field observations and confirmed all working hypotheses (field-generated) tested in this study, except the effects of fluoride on golden algal growth. In this study, results showed F- negatively associated

with golden alga abundance, however, field study hypothesized a positive association.

xii

Texas Tech University, Rakib H. Rashel, May 2020

The results of this study are expected to provide basic and comparative information to a better understanding of the effects of environmental factors on golden alga growth.

xiii

Texas Tech University, Rakib H. Rashel, May 2020

LIST OF TABLES 2.1. Pearson partial correlation of maximum cell density with either exponential growth rate (r) or early cell density (early-D) in Prymnesium parvum as a function of genetic strain and temperature over a range of salinities………………………………………………………………………… 53 3.1. The nominal concentration of major ions or constituents of Artificial Seawater Medium (ASM), Instant Ocean® (IO), and Seawater (SW)………..…90 3.2. Output of one-way and two-way ANOVA of data collected in this study. Growth indices examined in Prymnesium parvum cultures included exponential growth rate (r), maximum cell density, and early cell density………………………………………………………………...….………91 3.3. Non-parametric Spearman's correlation of maximum cell density, exponential growth rate (r), and early cell density in Prymnesium parvum cultures versus sulfate concentration (0–1000 mg l-1) in modified ASM at a salinity of 5 psu………………………………………………………….….……92 3.4. Pearson partial correlation of maximum cell density with exponential growth rate (r) or early cell density in Prymnesium parvum cultures as a function of sulfate concentration (0–1000 mg l-1) in modified ASM at a salinity of 5 psu………………………………………………………….….……93 4.1. Output of two-way ANOVA of data collected in this study. Growth indices examined in Prymnesium parvum cultures included exponential growth rate (r), maximum cell density, and early cell density…………………………………………………………………...…..….114 4.2. Pearson partial correlation of maximum cell density with exponential growth rate (r) or early cell density in Prymnesium parvum cultures as a function of the relative concentration of inorganic and organic N……………..115 5.1. Output of two-way ANOVA of data collected in this study. Growth indices examined in Prymnesium parvum cultures included exponential growth rate (r), maximum cell density, and early cell density……….....……...138 5.2. Pearson partial correlation of maximum cell density with exponential growth rate (r) or early cell density in Prymnesium parvum cultures……….....139 A. 1. Ion composition of ASM and customized medium……………………..…...…153

xiv

Texas Tech University, Rakib H. Rashel, May 2020

LIST OF FIGURES 2.1. Growth curves of Prymnesium parvum as a function of inoculum size (100–100,000 cells ml-1) and genetic strain (UTEX-2797 and UTEX-995) in standard base medium of salinity 5 psu at 22 °C...... 54 2.2. Exponential growth rate (r) and maximum cell density of Prymnesium parvum as a function of inoculum size and genetic strain in a standard base medium of salinity 5 psu at 22 °C...... 55 2.3. Growth curves of Prymnesium parvum as a function of salinity, genetic strain, and temperature in cultures inoculated with 100 cells ml-1...... 56 2.4. Exponential growth rate (r) and maximum cell density of Prymnesium parvum as a function of salinity, genetic strain, and temperature in cultures inoculated with 100 cells ml-1...... 57 3.1. Growth indices of Prymnesium parvum as a function of sulfate -1 concentration (0–1000 mg l , as MgSO4) in modified ASM at a salinity of 5 psu...... 94 3.2. Growth indices of Prymnesium parvum as a function of sulfate -1 concentration (0–1000 mg l , as Na2SO4) in modified ASM at a salinity of 5 psu...... 95 3.3. Growth indices of Prymnesium parvum as a function of salinity in modified ASM or Instant Ocean with salinities of 5 and 30 psu...... 96 3.4. Growth indices of Prymnesium parvum in modified ASM as a function of salinity and MgSO4 concentration...... 97 3.5. Growth indices of Prymnesium parvum in modified ASM as a function of salinity and Na2SO4 concentration...... 98 3.6. Growth indices of Prymnesium parvum as a function of fluoride concentration (0–56.50 mg l-1) in modified ASM at salinity of 5 psu...... 99 4.1. Growth indices of Prymnesium parvum as a function of different molar ratios of inorganic to organic (urea) nitrogen in modified ASM...... 116 4.2. Growth indices of Prymnesium parvum as a function of different molar ratios of inorganic to organic (glycine) nitrogen in modified ASM...... 117

5.1. Growth indices of Prymnesium parvum as a function of air CO2 concentration in modified ASM (low salinity)...... 140

5.2. Growth indices of Prymnesium parvum as a function of air CO2 concentration in Instant Ocean (IO; high salinity)...... 141 5.3. Changes in culture medium pH during the incubation period at different air CO2 concentrations (A) ASM at salinity of 5, (B) IO at salinity of 30...... 142 S.2.1. Maximum cell density of Prymnesium parvum (UTEX-2797) as a function of salinity at a temperature of 22 °C following inoculation xv

Texas Tech University, Rakib H. Rashel, May 2020

with 100 cells ml-1...... 60 S.2.2. Exponential growth rate (r) and maximum cell density as a function of salinity in Prymnesium parvum (UTEX-2797) cultures inoculated with 100,000 cells ml-1 and maintained at 22 °C...... 61 S.4.1. Maximum cell density of Prymnesium parvum as a function of different molar ratios of inorganic to organic (urea) nitrogen in modified ASM...... 118 S.4.2. Transitional phase in log-transformed growth curve of Prymnesium parvum as a function of different molar ratios of inorganic to organic (urea) nitrogen in modified ASM...... 119 S.4.3. Maximum cell density of Prymnesium parvum as a function of different molar ratios of inorganic to organic (glycine) nitrogen in modified ASM...... 120

xvi

Texas Tech University, Rakib H. Rashel, May 2020

LIST OF ABBREVIATIONS

ANOVA Analysis of Variance ASM Artificial Seawater Medium °C Degree Celsius

CO2 Carbon dioxide F- Fluoride g l-1 Gram per liter HAB Harmful Algal Bloom h Hour IO Instant Ocean LOD Limit of detection l Liter Mg2+ Magnesium mg l-1 Milligrams per liter mg Milligram N Nitrogen P Phosphorus psu Practical Salinity Unit ppm Parts per million SW Seawater SEM Standard Error of the Mean 2- SO4 Sulfate TN Total Nitrogen Tukey’s HSD Tukey’s Honest Significant Difference μM Micromolar

xvii

Texas Tech University, Rakib H. Rashel, May 2020

CHAPTER I

INTRODUCTION

Biology of harmful algae

Algae are a very important component of freshwater, estuarine, and marine ecosystems. Some of the algal species are harmful, but most species are not. Over the last several decades, many countries throughout the world have experienced an increased incidence of Harmful Algal Blooms (HAB) (Anderson, 2009). Harmful algal blooms cause harm to aquatic organisms either by producing toxins, which may lead to harmful consequence even at low cell concentrations (Sellner et al., 2003) or by affecting the ecological integrity of ecosystems due to excessive growth (Anderson et al., 2002;

Glibert et al., 2001, 2005; Heisler et al., 2008). Of the known ~5000 phytoplankton species (Sournia et al., 1991), only ~300 species can display massive blooms along with water discoloration while ~40 species can produce toxins that can accumulate in fish and shellfish and cause human poisoning indirectly by seafood consumption (Hallegraeff,

1993). Some toxic HAB species appear to be involved primarily in allelopathy, causing deleterious effects on competitor phytoplankton species (Fistarol et al., 2003, 2004;

Legrand et al., 2003). Non-toxic HAB cause damage to aquatic ecosystems by displacing indigenous species, altering habitats, or depressing water oxygen levels. Globally, the incidence and geographic scope of HAB have increased significantly over the last few decades. As recently as a few decades ago, relatively few countries were affected by

HAB, but now most countries are affected by more than one harmful or toxic algal species (Anderson, 1989; Hallegraeff, 1993). Algal blooms can be harmful to other algae

Texas Tech University, Rakib H. Rashel, May 2020 and aquatic organisms via multiple mechanisms but producing toxins and creating hypoxic or anoxic conditions are some of the most common mechanisms.

Algal species that produce toxins need not develop a large accumulation of cells, as their harmful impacts can occur at low abundances (Anderson et al., 2012a). The subject of this study, golden alga (Prymnesium parvum), produces toxins that can be lethal to gilled aquatic animals (Sager et al., 2008). Ulitzur and Shilo (1966) concluded that the gill membrane is the primary site affected by golden alga toxins.

Photosynthetic algae produce oxygen in the presence of sunlight, but at night they consume dissolved oxygen from their surrounding environment. At high density, they can deplete dissolved oxygen and create hypoxic or anoxic conditions (Landsberg, 2002).

Decomposing algal cells can further increase oxygen demand and exacerbate oxygen- depleted conditions (Landsberg, 2002). Hypoxic and anoxic conditions can result in impaired health or death of aquatic organisms, including fishes. A prominent example of this phenomenon is the “dead zones” in the Gulf of Mexico (Rabalais et al., 2002). High algal cell density can also cause fish mortality by damaging delicate structures of gills, causing suffocation due to excess mucous secretion (Landsberg, 2002).

Most major marine and freshwater algal phyla include HAB species. Harmful algal phyla include prokaryotic blue green algae or cyanobacteria, and eukaryotic algae.

Eukaryotic algae include dinoflagellates, chrysophytes, and haptophytes (Reynolds,

1984; Round, 1965), which may produce red tide, brown tide, and golden algal bloom, respectively.

Texas Tech University, Rakib H. Rashel, May 2020

Cyanobacteria and blue green algal bloom

Cyanobacteria (blue green algae) are considered the most ancient (~2.5 billion years) phytoplankton on the planet and are capable of forming harmful blooms in different aquatic ecosystems (Paerl et al., 2001; O'Neil et al., 2012). Cyanobacteria are prokaryotic organisms but have historically been grouped with "algae" and also have been referred to as blue-green algae, Cyanophyceae, Cyanophyta, or Myxophyceae

(Carmichael, 2008). In freshwater ecosystems, cyanobacteria can be found in three different morphological groups; (I) unicellular, which may be either solitary or aggregated in colonies, (II) undifferentiated cells with non-heterocystous filaments, and

(III) differentiated cells with heterocystous filaments (Pearl et al., 2001). Each of these groups includes HAB species. Notorious cyanobacteria include the genera Anabaena,

Aphanizomenon, Cylindrospermopsis, Lyngbya, Microcystis, Nodularia, Oscillator,

Synechococcus, Trichodesmium and others which are generally confined to nutrient- enriched impoundments (Fogg, 1969; Reynolds and Walsby, 1975; Paerl, 1988; Paerl and

Tucker, 1995). Recently, HAB forming cyanobacteria have been termed either

"CyanoHABs" (Carmichael, 2008; Paerl, 2008) or "Cyanobacterial blooms" (Hudnell et al., 2008).

CyanoHABs have been reported in the scientific literature for more than

130 years, but in recent decades the incidence and intensity of CyanoHABs as well as their associated ecological and economic impacts have increased in freshwater, estuarine and marine ecosystems (Carmichael, 2001, 2008; Heisler et al., 2008; Paerl, 2008; Paul,

2008). The incidence of CyanoHABs events is increasing in the USA and have been

Texas Tech University, Rakib H. Rashel, May 2020 reported in at least 35 states (Lopes et al, 2008). CyanoHABs are complex events and multiple environmental factors may be responsible for blooms (Heisler et al., 2008).

However, known drivers include increased nutrient levels, transport of cells or cysts due to anthropogenic activities, overfishing, and others (Heisler et al., 2008). Other physicochemical factors, such as changing salinity levels in estuarine and freshwater bodies may cause shifts in phytoplankton species composition (Bordalo and Vieira,

2005).

Freshwater bodies enriched with phosphorus (P) typically experience changes in the phytoplankton community towards dominance by cyanobacteria (Smith, 1986;

Watson et al., 1997; Paerl and Huisman, 2008). CyanoHABs are considered as indicators of over-enrichment of nutrients (Paerl and Fulton, 2006), and appropriate watershed management to reduce excess nutrients can decrease CyanoHABs events in some areas

(Edmondson and Lehman, 1981).

Most of the CyanoHABs species produce "cyanotoxins" with neurotoxic, hepatotoxic or dermatotoxic activities (Fristachi et al., 2008), and they can also impair drinking water by changing taste and odor. However, the harmful effects of CyanoHABs can occur even in the absence of visible indications of a bloom (Lopes et al., 2008).

Cyanotoxins also affect aquaculture resources (Tucker, 2000), and the presence of large blooms in recreational water bodies can reduce recreational opportunities (Walker et al.,

2008).

Texas Tech University, Rakib H. Rashel, May 2020

Dinoflagellates and Red tide

Dinoflagellates are another algal group that causes HAB commonly known as

"red tide" (Vila et al., 2005; Anderson et al., 2012b). They are a very large and diverse

group of eukaryotic algae present mainly in marine ecosystems but also found in

freshwater. They produce various toxins (Anderson, 1994; Wang, 2008) capable of

creating negative health effects on shellfish, animals, and humans (Trainer and Baden,

1999; Van Dolah, 2000; Ciminiello and Fattorusso, 2006; Wang, 2008). Nuisance and

toxic dinoflagellates can flourish in some brackish conditions, where blooms can cause

large ecological and economic losses (Paerl et al., 2001). Some species can form cysts

under unfavorable conditions, which are considered as seeds for the development of red

tides, because they resume motility when environmental conditions become favorable for

growth (Anderson, 1994; Baden et al., 1995).

Prorocentrum species form red tides in many parts of the world (Taylor and

Seliger, 1979; Anderson et al., 1985; Granéli et al., 1990). Karenia brevis was originally

restricted to the Gulf of Mexico and the Caribbean Sea, but now has been carried by

ocean or brackish water current around the west coast of Florida as far as North Carolina,

where it has caused red tides (Kirkpatrick et al., 2004).

Toxic red tides of K. brevis have been observed in Florida since the 1840s

(Kirkpatrick et al., 2004). Recently, red tides appear increasingly in incidence, duration and geographic spread (Viviani, 1992; Van Dolah, 2000).

Texas Tech University, Rakib H. Rashel, May 2020

Brown Tide

Brown tides were first observed in 1985 in several locations along the northeast coast of the United States (Sieburth et al., 1988; Olsen, 1989). A chrysophyte,

Aureococcus anophagefferens produces brown tides. Increasing nutrients inputs may produce HAB in some areas (Anderson et al., 2002; Hecky and Kilham, 1988), but brown tide associated with A. anophagefferens seem to be the result not of nutrient enrichment, but variations in the ratio of organic and inorganic nutrients (Laroche et al., 1997). Brown tide blooms have impacted mainly fishes and seagrasses (Gastrich and Wazniak, 2002).

Brown tide blooms have significant negative impacts on the productivity of seagrasses

(Cosper et al., 1989; Dennison et al., 1989).

Golden alga

Prymnesium spp. belong to the kingdom Chromista, class Prymnesiophyceae, order Prymnesiales, and family Prymnesiaceae. Members of genus Prymnesium are single-celled, biflagellate and possess a single haptonema. Flagella helps golden alga to swim and haptonema helps it to attach to surfaces or to capture food particles (Kawachi et al., 1991; Green and Jordan, 1994). Most Prymnesium spp. are cosmopolitan and the best-known species is P. parvum, golden alga.

Golden alga is a toxigenic species typically inhabiting marine and estuarine habitats, but it has also invaded inland brackish waters, where it is capable of forming fish-killing blooms and causing major ecological damage (Baker et al., 2007, 2009; Sager

Texas Tech University, Rakib H. Rashel, May 2020

et al., 2008; Lutz-Carrillo et al., 2010; Southard et al., 2010). Golden alga produces toxic

chemical substances that affect primarily gilled aquatic organisms. These substances are

also allelopathic and deter grazers (Granéli and Johansson, 2003; Granéli et al., 2012).

Golden alga is a mixotrophic species that can ingest other microalgae or zooplankton by the process of phagotrophy and is also able to utilize dissolved organic matter by osmotrophy (Granéli et al., 2012).

Global occurrence of golden alga

Massive fish kills associated with golden alga have been associated with

significant ecological and monetary losses. Golden alga has been identified on all

continents except for Antarctica (Larsen and Bryant, 1998; Lutz-Carillo et al., 2010).

This species was originally found in coastal and estuarine systems. However, reports of golden alga in inland waters were relatively rare until recently (Lutz-Carillo et al., 2010).

In North America, golden alga has mostly affected inland water systems. A genetic study of golden alga strains in North America determined that most derive primarily from

Scotland, but several strains also were identified that are related to those from England,

Denmark, and Norway (Lutz-Carillo et al., 2010).

The first recorded event of fish kills caused by golden alga occurred in Holland in

1920 (Liebert and Deerns, 1920, as cited in Guo et al., 1996). Golden alga was documented as the culprit of fish-killing in the brackish waters of Denmark in 1938

(Reichenbach-Klinke, 1973). In 1947, mass mortalities in carp ponds occurred in Israel and since then this has been a recurring event (Shilo and Shilo, 1953). Multiple fish

Texas Tech University, Rakib H. Rashel, May 2020

mortalities associated with golden alga were documented in the River Thurne system in

England starting in 1969 (Bales et al., 1993). Mixed blooms of golden alga occurred

every summer during 1989–1996 in the Sandefjord system of southwest Norway (Larsen

and Bryant, 1998; Johnsen et al., 2010). Since the 1970s, golden alga blooms have caused

large-scale fish mortalities in the Vasse-Wonnerup estuary in Australia (Hallegraeff,

1992). Blooms were reported in aquaculture systems in Israel (Reisch and Aschner, 1947) and China (Guo et al., 1996). Although, recurring fish kill due to golden alga blooms in

China have been reported since 1963 (Guo et al., 1996).

In the last few decades, golden alga blooms have appeared in previously

unaffected areas (Genitsaris et al., 2009; Vasas et al., 2012, Roelke et al., 2016),

including the United States. In 1982 an estimated 2300 fish were killed in California

Creek, Brazos River basin, Texas, and a retrospective analysis suggested golden alga was

the killer (Glass et al., 1991). The first confirmed fish kill due golden alga occurred in

1985 in the Pecos River (Rio Grande Basin), Texas, with approximately 110,000

estimated fish deaths (James and De La Cruz, 1989; Rhodes and Hubbs, 1992). In

November and December 1986, additional fish kills occurred with an estimated 500,000

fish deaths in the same areas of the Pecos River. During November and December 1988,

48,000 fish were killed in Paint Creek, a tributary of Brazos River, Texas (James and Da

La Cruz, 1989). In the 2000s, golden alga blooms expanded their range rapidly (Southard

et al., 2010). Golden alga blooms continue their spreading across the landscape and now

are a common occurrence in five river basins in west and central Texas (Sager et al.,

2008; Southard et al., 2010), and the number of fish deaths as of 2009 was estimated to

be in the tens of millions (Southard et al., 2010). The expansion of this algal bloom range

Texas Tech University, Rakib H. Rashel, May 2020

was not only limited to the southcentral USA as it quickly spread all over the USA inland

waters and to date presence of this species reported in at least 23 states (Roelke et al.,

2016).

Environmental factors associated with golden alga blooms

Salinity

Golden alga is a euryhaline species and can tolerate a wide range of salinities in

both laboratory and field conditions. Previous studies reported golden alga can grow in

the laboratory culture at salinities from ~3–4 psu (practical salinity unit) to full-strength

seawater, but the highest growth rates were observed at the upper end of the salinity range (≥ 15 psu) (Padilla, 1970; Larsen and Bryant, 1998; Baker et al., 2007, 2009;

Hambright et al., 2015). In the field, however, the positive association between golden alga abundance and salinity is normally observed at a lower range of salinities

(Hambright et al., 2010, 2015; Roelke et al., 2011, 2012; Israël et al., 2014; Patiño et al.,

2014; VanLandeghem et al., 2015a).

In the Brazos River, Texas, golden alga blooms occur at reported salinity threshold of 0.5 psu at Lake Granbury and Whitney and 1.5 psu at Lake Possum

Kingdom (Roelke et al., 2011). Golden alga blooms were observed in Colorado River reservoirs with salinity levels above 1.0 psu (Patiño et al., 2014). In Lake Texoma, which is the largest reservoir in the Red River basin (Hambright et al., 2010), salinity above 1.7 psu was the best predictor of golden alga blooms (Hambright et al., 2015).

Texas Tech University, Rakib H. Rashel, May 2020

A recent study of golden alga in the Pecos River basin, considered one of the saltiest river systems in North America, reported an inverse relation between alga abundance and salinity at high level (Israël et al., 2014). The authors of the Pecos River

study concluded that the association between golden alga abundance and salinity may

follow a biphasic pattern; namely, positive and negative associations at lower and higher

salinities, respectively (Israël et al., 2014). A study conducted in saline lakes in China

reported that golden alga cell density is positively correlated with salinity up to 8 psu and

above that level golden alga cell density is negatively correlated (Gou, 1983; cited in Guo

et al., 1996). This biphasic pattern of salinity effects suggests an intermediate range of

salinity for optimal growth.

Inoculum size

The influence of inoculum size on golden algal growth in the laboratory has never

been studied. Inoculum size is, in fact, often not reported in the methodology of

published experimental studies. Recent mesocosm studies, however, suggest that golden

alga survival and growth may be affected by early cell density. For example, an in-lake

microcosm study concluded that immigration (as inoculum) can be an important factor

determining whether a subsequent bloom will form (Errera et al., 2008). More

specifically, Errera et al. (2008) found that the addition of ~500 golden alga cells ml-1

over background levels promoted winter growth under certain conditions. Another

microcosm study reported that a minimum initial abundance (propagule pressure) of

Texas Tech University, Rakib H. Rashel, May 2020

~6400 cells ml-1 was required for survival at one-week post-inoculation (Acosta et al.,

2015).

Sulfate

Sulfate is an important source of sulfur for marine and freshwater phytoplankton.

When phosphorus limitation is relieved, sulfur can become the next nutrient to limit productivity in some freshwater systems (Giordano et al., 2005). Sulfate is a major anion in west Texas surface waters and can contribute significantly to salinity. The specific sulfur requirement for golden alga is unknown. However, golden alga is believed to have originated from marine and brackish water habitats, where sulfur concentration is much higher than in most inland freshwaters. A recent analysis of archived data indicated that

sulfate levels in reservoirs of the southcentral USA with a history of blooms (average,

~600 mg l-1) were >8-fold higher than in naive reservoirs (~70 mg l-1) and were nearly

equal those of chloride (~800 mg l-1) (Patiño et al., 2014). Similar pattern also has been

observed in a few other places; for example, blooms of this species in association with high sulfate levels were recorded in a freshwater pond in Germany (Moestrup, 1994).

Therefore, sulfate deserves closer attention as a potential driver for golden algal bloom

formation.

Texas Tech University, Rakib H. Rashel, May 2020

Fluoride

At high levels, fluoride can inhibit algae and cyanobacteria growth by inhibiting

photosynthesis (Bhatnagar and Bhatnagar, 2000). However, fluoride can either inhibit,

enhance, or not affect algal growth depending upon the species and exposure

concentrations (Camargo, 2003). Recent studies have reported relatively high fluoride

levels in bloom-impacted (>1 mg l-1) compared to non-impacted (<1 mg l-1) water bodies

of the southcentral USA (VanLandeghem et al., 2012, 2015a). Algae generally tolerate a

higher level of fluoride than cyanobacteria (Bhatnagar and Bhatnagar, 2000), and

cyanobacteria may be allelopathic to golden alga (James et al., 2011; Roelke et al., 2012).

Golden alga growth may, therefore, benefit from relatively high fluoride concentration.

Water hardness

Water hardness may also influence golden alga growth. In laboratory assays, the

potency of golden alga toxin extracts is increased in the presence of hardness cations

(Yariv and Hestrin, 1961; Ulitzur and Shilo, 1964, 1966). Moreover, field studies have

shown a positive association between golden alga toxicity and hardness levels.

VanLandeghem et al. (2012) found that water toxicity was higher at water hardness levels

-1 above 700 mg l CaCO3.

Texas Tech University, Rakib H. Rashel, May 2020

Temperature

Golden alga abundance is strongly associated with water temperature. In Texas

inland waters, golden alga blooms usually occur at seasonally lower temperatures in fall

or late winter or early spring (Sager et al., 2008). In Lake Texoma, peak golden alga

abundance was documented between 10–15 °C and above 20 °C, blooms generally did

not occur (Hambright et al., 2010). A similar observation was made in Colorado River

reservoirs, where peak golden alga abundance occurred between 10–13 °C

(VanLandeghem et al., 2015b).

In laboratory cultures, optimum growth is observed at relatively warm

temperatures (20–30 °C) (Edvardsen and Paasche, 1998). Temperatures greater than 30

°C were inhibitory to golden alga growth and 35 °C resulted in lysis; however, golden alga cells can survive at 2 °C for many days (Shilo and Aschner, 1953). Danish,

Norwegian and English strains of golden alga exhibited a maximum growth rate at 15 °C, but Danish and Norwegian strain can tolerate a wide temperature range of 5–30 °C

(Larsen and Bryant, 1998). Temperature seems to interact with salinity in affecting golden alga growth and toxicity. For example, optimal growth rates in laboratory cultures occur at 25–28 °C and salinity of 20–25 psu (Baker et al., 2007).

Earlier studies concluded that water pH might be an important factor governing

the occurrence of toxic golden alga blooms (Valenti et al., 2010; Prosser et al., 2012;

Texas Tech University, Rakib H. Rashel, May 2020

Roelke et al., 2012). A laboratory study reported that the potency of golden alga toxins

increases as pH increases above 7, while toxins become ineffective at pH 7 and below

(Ulitzur and Shilo, 1964). However, a retrospective analysis of Brazos and Colorado

River reservoirs did not find associations between water pH and toxic bloom events

(Patiño et al., 2014). A more sampling-intensive study in Colorado River reservoirs also indicated water pH and golden alga toxic bloom occurrence were also not related

(VanLandeghem et al., 2015a). Previous studies of P. parvum examined the effects of pH manipulation on growth rate in the laboratory (Berge et al., 2010; Lysgaard et al., 2018) and maximum cell density in field mesocosms (Prosser et al., 2012). Overall, studies of

the association between pH and golden alga have provided inconsistent results.

Nutrients

Golden alga can grow in a wide variety of habitats; however, most toxic events in

inland systems have occurred in brackish, nutrient-rich water bodies (Guo et al., 1996;

Edvardsen and Paasche, 1998; Granéli et al., 2012; Roelke et al., 2007, 2016). Most

experimental studies have evaluated the effects of nitrogen (inorganic), phosphorus

(inorganic), and their relative amounts on toxin production of this algal species but only

few have addressed the effects on growth (Johansson and Granéli, 1999, Lundgren et al.,

2016). There are some inconsistencies, however, regarding the influence of nutrients on

golden algal growth. Namely, some mesocosm studies reported that low N:P ratios are

strongly associated with bloom formation (Hambright et al., 2010), while others reported

that high N:P ratios favor blooms in microcosm experiments (Errera et al., 2008).

Texas Tech University, Rakib H. Rashel, May 2020

Golden alga abundance in the field is positively associated with nutrient

concentrations (Hambright et al., 2010, 2014; VanLandeghem et al., 2015b). Phosphorus

(P) is limited in most inland surface waters and consequently, P is generally the limiting

nutrient for algal growth (Dillon and Rigler, 1974; Schindler, 1977, 1978). Nitrogen (N)

and P imbalance enhance toxicity. When N:P is low or high, golden alga cell undergoes

physiological stress that can stimulate increase production of toxins (Dafni et al., 1972;

Granéli and Johansson, 2003; Granéli et al., 2008; Uronen et al., 2005; Valenti et al.,

2010). Recently, however, Israël et al. (2014) found a positive association between

organic P and N and golden alga (presence and abundance, respectively) and negative

association with inorganic N in the Pecos River. They also found that golden alga occurs

at a wide range of nutrient concentrations when salinity is relatively low, but only at mid-

to-high nutrient concentrations when salinity is relatively high (Israël et al., 2014). This

was the first report of the interaction effects of salinity and nutrients in the field. Another

recent field study of reservoirs in the upper Colorado River Basin (Texas, USA) found

that golden alga declined seasonally as levels of inorganic N increased (VanLandeghem

et al., 2015b). The same study found that golden alga-impacted reservoirs have higher concentrations of organic N than non-impacted reservoirs (VanLandeghem et al., 2015b).

A few studies have evaluated the use of inorganic N to control golden alga in small water bodies, where it has been shown that a high concentration of ammonium is toxic

(Grover et al., 2007).

Texas Tech University, Rakib H. Rashel, May 2020

Carbon dioxide

Carbon dioxide is an important element for carbon fixation by photosynthesis.

Changes in atmospheric concentrations of carbon dioxide can, therefore, influence micro

and macroalgal growth (Gao et al., 2012). Photosynthetic carbon fixation in seawater

leads to a decrease of carbon dioxide in seawater and promotes the absorption of

atmospheric carbon dioxide (Arrigo, 2007). Addition of carbon dioxide in water produces

more carbonic acid and lowers the water pH by an acidification process.

Due to climate change and anthropogenic activities, the lower of two projected high-end scenarios of future conditions considers that air CO2 concentration may reach

670 ppm by 2100 (Jones et al., 2013). Since the industrial revolution, the pH of surface

seawater has declined by about 0.1 pH units and is expected to decrease an additional 0.3

units by 2100 (Feely et al., 2009). Rising atmospheric carbon dioxide concentration may

lead to further acidification in water by lowering pH, which is one of the important

factors that influences aquatic ecosystems. Growth responses of cyanobacteria and other

algae due to ocean acidification are species-dependent. Several studies were conducted to examine the influence of carbon dioxide and pH on phytoplankton growth. The combined

effects of increased CO2 (available as dissolved inorganic carbon) and reduced pH

(acidification) on phytoplankton are in fact complex and, depending on the species and

suite of environmental conditions, the outcome can be stimulation, inhibition or no

effects on growth (Gao et al., 2012). A few studies examined the effects of pH

manipulation on golden alga growth rate in the laboratory (Berge et al., 2010; Lysgaard et

al., 2018) and maximum cell density in field mesocosms (Prosser et al., 2012); however,

Texas Tech University, Rakib H. Rashel, May 2020

no studies available regarding the effects of changes in air CO2 concentration on golden alga growth.

Significance of the present study

In recent years, our laboratory and others have gathered new and sometimes unexpected information concerning environmental variable associations with golden alga presence and abundance in inland waters in the USA. Notable examples include the biphasic association between golden alga abundance and salinity, the positive association with sulfate and fluoride concentration, and the positive association with organic nitrogen concentration. This information, however, is based on descriptive observations in the field and cannot be used as conclusive evidence of causal associations. There is a need, therefore, to test field-generated hypotheses under the controlled environment of a laboratory, where only the independent variable or variables of interest are allowed to vary. This will be the basic strategy underlying laboratory experiments of my dissertation research. I will not be addressing the interaction effects of multiple variables on golden alga growth, because my objectives are based on previous field findings, and those findings led to hypotheses about the influence of specific single variables.

Texas Tech University, Rakib H. Rashel, May 2020

Research objectives

The goal of this proposed research is to characterize the influence of water quality

and climate variables on golden alga growth. The following specific objectives will be

addressed:

1. Determine the influence of genetic background, salinity, and inoculum size on growth

of the ichthyotoxic golden alga (chapter 2)

2. Determine the growth response of the ichthyotoxic haptophyte, Prymnesium parvum

Carter, to changes in sulfate and fluoride concentrations (chapter 3)

3. Determine the association between the ratio of inorganic to organic nitrogen and

growth of the harmful alga (chapter 4)

4. Determine the growth of Prymnesium parvum under past, present and projected future

concentrations of air carbon dioxide (chapter 5)

5. Develop a customized medium that will favor golden alga growth under conditions of

low salinity and temperature (at salinity 1.6 psu and temperature 13 °C)

(Appendix A).

Texas Tech University, Rakib H. Rashel, May 2020

Literature Cited

Acosta, F., Zamor, R.M., Najar, F.Z., Roe, B.A., Hambright, K.D., 2015. Dynamics of an experimental microbial invasion. Proc. Natl. Acad. Sci. USA., 112, 11594–11599.

Anderson, D. M., 1989. Toxic algal blooms and red tides: a global perspective. Red Tides: Biology, Environ. Sci. and Toxic. 11–16. Anderson, D. M., 1994. Red Tides. Scientific American 271(2), 62–68.

Anderson, D. M., 2009. Approaches to monitoring, control and management of harmful algal blooms (HABs). Oce. and Coas. Manage. 52(7), 342–347.

Anderson, D. M., White, A.W., and Baden, D.G., 1985. Toxic dinoflagellates, Elsevier, New York. 561.

Anderson, D. M., Glibert, P. M., Burkholder, J. M., 2002. Harmful algal blooms and eutrophication: Nutrient sources, composition, and consequences. Estuaries 25(4), 704–726.

Anderson, D. M., Cembella, A. D., Hallegraeff, G. M., 2012a. Progress in understanding harmful algal blooms: paradigm shifts and new technologies for research, monitoring, and management. Ann. Rev. of Mar. Sci. 4, 143–176.

Anderson, D. M., Alpermann, T. J., Cembella, A. D., Collos, Y., Masseret, E., Montresor, M., 2012b. The globally distributed genus Alexandrium: multifaceted roles in marine ecosystems and impacts on human health. Harmful Algae 14, 10–35.

Arrigo, K. R., 2007. Carbon cycle: Marine manipulations. Nature 450(7169), 491–492.

Baden, D. G., Melinek, R., Sechet, V., Trainer, V. L., Schultz, D. R., Rein, K. S., Hale, L., 1995. Modified immunoassays for polyether toxins: implications of biological matrixes, metabolic states, and epitope recognition. J. of A. Inter. 78(2), 499–508.

Baker, J. W., Grover, J. P., Brooks, B. W., Ureña-Boeck, F., Roelke, D. L., Errera, R., Kiesling, R. L., 2007. Growth and toxicity of Prymnesium parvum (haptophyta) as a function of salinity, light, and temperature. J. of Phyco. 43(2), 219–227.

Baker, J. W., Grover, J. P., Ramachandrannair, R., Black, C., Valenti, T. W., Brooks, B. W., Roelke, D. L., 2009. Growth at the edge of the niche: An experimental study of the harmful alga Prymnesium parvum. Limnol. Oceanogr. 54(5), 1679–1687.

Bales, M., Moss, B., Phillips, G., Irvine, K., Stansfield, J., 1993. The changing ecosystem of a shallow, brackish lake, Hickling Broad, Norfolk, U.K. II. Long-term trends in water chemistry and ecology and their implications for restoration of the lake. Fresh. Bio. 29(1), 141–165.

Berge, T., Daugbjerg, N., Andersen, B.B., Hansen, P.J., 2010. Effect of lowered pH on marine phytoplankton growth rates. Mar. Ecol. Prog. Ser. 416, 79–91.

Texas Tech University, Rakib H. Rashel, May 2020

Bhatnagar, M., Bhatnagar, A., 2000. Algal and cyanobacterial responses to fluoride. Fluoride 33(2), 55–65.

Bordalo, A. A., Vieira, M. E. C., 2005. Spatial variability of phytoplankton, bacteria and viruses in the mesotidal salt wedge Douro Estuary (Portugal). Est. Coast. and Shelf Sci. 63, 143–154.

Camargo, J.A., 2003. Fluoride toxicity to aquatic organisms: a review. Chemosphere 50(3), 251–264.

Carmichael, W. W., 2001. Health Effects of Toxin-Producing Cyanobacteria: “The CyanoHABs.” Human and ecological risk assessment. An Inter. J. 7(5), 1393– 1407.

Carmichael, W., 2008. A world overview – One-hundred-twenty-seven years of research on toxic cyanobacteria – Where do we go from here? In H. K. Hudnell (Ed.), Cyanobacterial Harmful Algal Blooms: State of the Science and Research Needs. 105–125.

Ciminiello, P., Fattorusso, E., 2006. Bivalve molluscs as vectors of marine biotoxins involved in seafood poisoning. Prog. in Mole. and Sub. Bio. 43, 53–82.

Cosper, E. M., Dennison, W., Milligan, A., Carpenter, E. J., Lee, C., Holzapfel, J., Milanese, L., 1989. An examination of the environmental factors important to initiating and sustaining “Brown Tide” blooms. In E. M. Cosper, V. M. Bricelj, E. J. Carpenter (Eds.), Novel phytoplankton blooms, 317–340.

Dafni, Z., Ulitzur, S., Shilo, M., 1972. Influence of light and phosphate on toxin production and Ggrowth of Prymnesium parvum. J. of Gen. Micro. 70(2), 199– 207.

Dennison, W. C., Marshall, G. J., Wigand, C., 1989. Effect of “Brown Tide” shading on eelgrass (Zostera marina L.) distributions. In E. M. Cosper, V. M. Bricelj, E. J. Carpenter (Eds.), Novel Phytoplankton Blooms. 675–692.

Dillon, P. J., Rigler, F. H., 1974. The phosphorus-chlorophyll relationship in lakes1,2. Limnol. Oceanogr. 19(5), 767–773.

Edmondson, W. T., Lehman, J. T., 1981. The effect of changes in the nutrient income on the condition of Lake Washington. Limnol. Oceanogr. 26(1), 1–29.

Edvardsen, B., Paasche, E., 1998. Bloom dynamics and physiology of Prymnesium and Chrysochromulina. In: Anderson, D.M., Cembella, A.D., Hallegraff, G.M. (Eds.), The physiological ecology of harmful algal blooms. Springer-Verlag, Heidelberg. 193–208.

Texas Tech University, Rakib H. Rashel, May 2020

Errera, R.M., Roelke D.L., Kiesling R., Brooks, B.W., Grover, J.P., Schwierzke, L., Ureña-Boeck, F., Baker, J.W., Pinckney, J.L., 2008. Effect of imbalanced nutrients and immigration on Prymnesium parvum community dominance and toxicity: results from in-lake microcosm experiments. Aquat. Microb. Ecol. 52, 33–44.

Feely, R., Doney, S.,Cooley, S., 2009. Ocean acidification: Present conditions and future changes in a high-CO2 world. Oceanography 22(4), 36–47.

Fistarol, G., Legrand, C., Granéli, E, 2003. Allelopathic effect of Prymnesium parvum on a natural plankton community. Mar. Eco. Prog. Serie. 255, 115–125.

Fistarol, G. O., Legrand, C., Selander, E., Hummert, C., Stolte, W., and Granéli, E., 2004. Allelopathy in Alexandrium spp.: effect on a natural plankton community and on algal monocultures. Aqua. Micro. Eco. 35(1), 45–56.

Fogg, G. E., 1969. The Leeuwenhoek Lecture, 1968: The Physiology of an algal nuisance. Proceedings of the Royal Society of London. Series B, Bio. Sci. 173(1031), 175–189.

Fristachi, A., Sinclair, J. L., Hall, S., Berkman, J. A. H., Boyer, G., Burkholder, J., Walker, S., 2008. Occurrence of cyanobacterial harmful algal blooms workgroup report. In H. K. Hudnell (Ed.), Cyanobacterial harmful algal blooms: State of the science and research needs. 45–103.

Gao, K., Helbling, E. W., Häder, D. P., Hutchins, D. A., 2012. Responses of marine primary producers to interactions between ocean acidification, solar radiation, and warming. Mar. Eco., Progress Series, 470, 167–189.

Gastrich, M. D., Wazniak, C. E., 2002. A Brown tide bloom index based on the potential harmful effects of the brown tide alga, Aureococcus anophagefferens. Aquat. Eco. Hea. And Mange. 5(4), 435–441.

Genitsaris, S., Kormas, K. A., Moustaka-Gouni, M., 2009. Microscopic eukaryotes living in a dying lake (Lake Koronia, Greece). FEMS Micro. Eco. 69(1), 75–83.

Giordano, M., Norici, A., Hell, R., 2005. Sulfur and phytoplankton: acquisition, metabolism and impact on the environment. New Phytologist, 166(2), 371–382.

Glass, J., Linam, G., Ralph, J., 1991. The association of Prymnesium parvum with fish kills in Texas. Texas Parks and Wildlife Document, 1–8.

Glibert, P. M., Magnien, R., Lomas, M. W., Alexander, J., Fan, C., Haramoto, E. Kana, T. M., 2001. Harmful algal blooms in the chesapeake and coastal bays of Maryland, USA: Comparison of 1997, 1998, and 1999 events. Estuaries (6), 875.

Glibert, P., Anderson, D., Gentien, P., Granéli, E., Sellner, K.., 2005. The global, complex phenomena of harmful algal blooms. Oceanography 18(2), 136–147.

Texas Tech University, Rakib H. Rashel, May 2020

Gou, W., 1983. The reason for the fish death at aquacultural experimental station at Ningxia and the distribution of Prymnesium parvum in Ningxia. J. Dalian Fish. Coll. 1, 43–48 (in Chinese).

Granéli, E., Johansson, N., 2003. Increase in the production of allelopathic substances by Prymnesium parvum cells grown under N- or P-deficient conditions. Harmful Algae 2(2), 135–145.

Graneli, E., Sundstrom, B., Edler, L., Anderson, D.M., 1990. Toxic marine phytoplankton, Elsevier, New York. pp. 554.

Granéli, E., Weberg, M., Salomon, P. S., 2008. Harmful algal blooms of allelopathic microalgal species: The role of eutrophication. Harmful Algae 8(1), 94–102.

Granéli, E., Edvardsen, B., Roelke, D. L., Hagström, J. A., 2012. The ecophysiology and bloom dynamics of Prymnesium spp. Harmful Algae 14, 260–270.

Green, J.C., Jordan, R.W., 1994. Systematic history and taxonomy. In: The Haptophyte Algae (eds Green, J.C., Leadbeater, B.S.C.), Oxford University Press, Oxford, UK, 1–21.

Grover, J.P., Baker, J.W., Ureña-Boeck, F., Brook, B.W., Errera, R.M., Roelke, D.L., Kiesling, R.L., 2007. Laboratory tests of ammonium and barley straw extract as agents to suppress abundance of the harmful alga Prymnesium parvum and its toxicity to fish. Water Res. 41, 2503–2512.

Guo, M., Harrison, P. J., Taylor, F. J. R., 1996. Fish kills related to Prymnesium parvum N. Carter (Haptophyta) in the People’s Republic of China. J. of App. Phy. 8(2), 111–117.

Hallegraeff, G. M., 1992. Harmful algal blooms in the Australian region. Mar. Poll. Bull. 25(5–8), 186–190.

Hallegraeff, G. M., 1993. A review of harmful algal blooms and their apparent global increase. Phycologia 32(2), 79–99.

Hambright, K. D., Zamor, R. M., Easton, J. D., Glenn, K. L., Remmel, E. J., Easton, A. C., 2010. Temporal and spatial variability of an invasive toxigenic protist in a North American subtropical reservoir. Harmful Algae 9(6), 568–577.

Hambright, K. D., Easton, J. D., Zamor, R. M., Beyer, J., Easton, A. C., Allison, B., 2014. Regulation of growth and toxicity of a mixotrophic microbe: implications for understanding range expansion in Prymnesium parvum. Freshwater Science 33(3), 745–754.

Hambright, K.D., Beyer, J.E., Easton, J.D., Zamor, R.M., Easton, A.C., Hallidayschult, T.C., 2015. The niche of an invasive marine microbe in a subtropical freshwater impoundment. ISME Journal 9: 256–264.

Texas Tech University, Rakib H. Rashel, May 2020

Hecky, R. E., Kilham, P., 1988. Nutrient limitation of phytoplankton in freshwater and marine environments: A review of recent evidence on the effects of enrichment. Limnol. Oceanogr. 33, 796–822.

Heisler, J., Glibert, P. m., Burkholder, J. m., Anderson, D. M., Cochlan, W., Dennison, W. C., Suddleson, M., 2008. Eutrophication and harmful algal blooms: A scientific consensus. Harmful Algae 8, 3–13.

Hudnell, H. K., Dortch, Q., Zenick, H., 2008. An overview of the interagency, International symposium on cyanobacterial harmful algal blooms (ISOC-HAB): advancing the scientific understanding of freshwater harmful algal blooms. Adv. in Expe. Med. and Bio. 619, 1–16.

Israël, N. M. D., VanLandeghem, M. M., Denny, S., Ingle, J., Patiño, R., 2014. Golden alga presence and abundance are inversely related to salinity in a high-salinity river ecosystem, Pecos River, USA. Harmful Algae 39, 81–91.

James, T., De La Cruz, A., 1989. Prymnesium parvum Carter (Chrysophyceae) as a suspect of mass mortalities of fish and shellfish communities in western Texas. Texas J. of Sci. 41, 429–430.

James, S. V., Valenti, T. W., Roelke, D. L., Grover, J. P., Brooks, B. W., 2011. Probabilistic ecological hazard assessment of microcystin-LR allelopathy to Prymnesium parvum. J. of Plan. Res. 33(2), 319–332.

Johansson, N., Granéli, E., 1999. Influence of different nutrient conditions on cell density, chemical composition and toxicity of Prymnesium parvum (Haptophyta) in semi-continuous cultures. J. of Exp. Mar. Bio. and Eco. 239(2), 243–258.

Johnsen, T. M., Eikrem, W., Olseng, C. D., Tollefsen, K. E., Bjerknes, V., 2010. Prymnesium parvum: The Norwegian Experience. JAWRA J. of the Ame. Wat. Res. Asso. 46(1), 6–13.

Jones, C., Robertson, E., Arora, V., Friedlingstein, P., Shevliakova, E., Bopp, L., Brovkin, V., Hajima, T., Kato, E., Kawamiya, M., Liddicoat, S., Lindsay, K., Reick, C.H., Roelandt, C., Segschneider, J., Tjiputra, J., 2013. Twenty-first- century compatible CO2 emissions and airborne fraction simulated by cmip5 earth system models under four representative concentration pathways. J. Clim. 26, 4398–4413.

Kawachi, M., Inouye, I., Maeda, O., Chihara, M., 1991. The haptonema as a food- capturing device: observations on Chrysochromulina hirta (Prymnesiophyceae). Phycologia 30, 563–57.

Kirkpatrick, B., Fleming, L. E., Squicciarini, D., Backer, L. C., Clark, R., Abraham, W., Baden, D. G., 2004. Literature review of Florida red tide: Implications for human health effects. Harmful Algae 3(2), 99–115.

Texas Tech University, Rakib H. Rashel, May 2020

Landsberg, J. H., 2002. The effects of harmful algal blooms on aquatic organisms. Rev. in Fish. Sci. 10(2), 113–390.

Laroche, J., Nuzzi, R., Waters, R., Wyman, K., Falkowski, P., Wallace, D., 1997. Brown tide blooms in Long Island’s coastal waters linked to interannual variability in groundwater flow. Glo. Cha. Bio. 3(5), 397–410.

Larsen, A., Bryant, S., 1998. Growth rate and toxicity of Prymnesium parvum and Prymnesium patelliferum (Haptophyta) in response to changes in salinity, light and temperature. Sarsia 83, 409-418.

Legrand, C., Rengefors, K., Fistarol, G. O., Granéli, E., 2003. Allelopathy in phytoplankton - biochemical, ecological and evolutionary aspects. Phycologia 42(4), 406–419.

Liebert F, Deerns, W.M., 1920. Onderzoek naar de oorzak van een Vischsterfte in den Polder Workumer Nieuwland, nabij Workum. Verhandungen en Rapporten uitgegeven door Rijkinstituten voor Visscherijonderzoek, 81–93.

Lopez, C.B., Jewett, E.B., Dortch, Q., Walton, B.T. Hudnell, H.K., 2008. Scientific assessment of freshwater harmful algal blooms. Washington, D.C., Interagency working group on harmful algal blooms, hypoxia, and human health of the joint subcommittee on ocean science and technology, 65.

Lundgren, V., Glibert, P.M., Granéli, E., Vidyarathna, N.K., Fiori, E., Ou, L., Flynn, K.J., Mitra, A., Stoecker, D.K., Hansen, P.J., 2016. Metabolic and physiological changes in Prymnesium parvum when grown under, and grazing on prey of, variable nitrogen:phosphorus stoichiometry. Harmful Algae 55, 1–12.

Lutz-Carrillo, D. J., Southard, G. M., Fries, L. T., 2010. Global genetic relationships among isolates of golden alga (Prymnesium parvum). J. of the Ame. Wat. Res. Asso. 46(1), 24–32.

Lysgaard, M.L., Eckford-Soper, L., Daugbjerg, N., 2018. Growth rates of three geographically separated strains of the ichthyotoxic Prymnesium parvum (Prymnesiophyceae) in response to six different pH levels. Estuar. Coast. Shelf Sci. 204, 98–102.

Moestrup, Ø., 1994. Economic aspects: ‘blooms’, nuisance species and toxins. In: Green J.C., Leadbeater B.S.C., editors. The Haptophyte Algae. Oxford: Clarendon Press, 265–285.

Olsen, P. S., 1989. Development and distribution of a brown-water algal bloom in Barnegat Bay, New Jersey with perspective on resources and other red tides in the region. In E. M. Cosper, V. M. Bricelj, E. J. Carpenter (Eds.), Novel Phytoplankton Blooms, 189–212.

Texas Tech University, Rakib H. Rashel, May 2020

O’Neil, J. M., Davis, T. W., Burford, M. A., Gobler, C. J., 2012. The rise of harmful cyanobacteria blooms: The potential roles of eutrophication and climate change. Harmful Algae 14, 313–334.

Padilla, G.M., 1970. Growth and toxigenesis of the chrysomonad Prymnesium parvum as a function of salinity. J. Protozool. 17, 456–462.

Paerl, H. W., 1988. Nuisance phytoplankton blooms in coastal, estuarine, and inland waters. Limnol. Oceanogr. 33, 823–843.

Paerl, H.W., 2008. Nutrient and other environmental controls of harmful cyanobacterial blooms along the freshwater-marine continuum. Adv. in Exp. Med. and Bio. 619, 217–237.

Paerl, H. W., Tucker, C. S., 1995. Ecology of blue-green algae in aquaculture ponds. J. of the Wor. Aqua. Soci. 26(2), 109–131.

Paerl, H. W., Fulton, R. S., 2006. Ecology of harmful cyanobacteria. In P. D. E. Granéli and P. D. J. T. Turner (Eds.), Ecology of Harmful Algae, 95–109.

Paerl, H. W., Huisman, J., 2008. Blooms like it hot. Science, 320(5872), 57–58.

Paerl, H. W., Fulton, R. S., Moisander, P. H., Dyble, J., 2001. Harmful Freshwater algal blooms, with an emphasis on cyanobacteria. The Sci. Wor. J. 1, 76–113.

Patiño, R., Dawson, D., VanLandeghem, M. M., 2014. Retrospective analysis of associations between water quality and toxic blooms of golden alga (Prymnesium parvum) in Texas reservoirs: Implications for understanding dispersal mechanisms and impacts of climate change. Harmful Algae 33, 1–11.

Paul, V. J., 2008. Global warming and cyanobacterial harmful algal blooms. Adv. in Expe. Med. and Bio. 619, 239–257.

Prosser, K. N., Valenti Jr., T. W., Hayden, N. J., Neisch, M. T., Hewitt, N. C., Umphres, G. D., Brooks, B. W., 2012. Low pH preempts bloom development of a toxic haptophyte. Harmful Algae 20, 156–164.

Rabalais, N. N., Turner, R. E., Wiseman, W. J., 2002. Gulf of Mexico Hypoxia, A.K.A. “The Dead Zone.” Annual Review of Ecology and Systematics, 33(1), 235–263.

Reich, K. Aschner, M., 1947. Mass development and control of the phytoflagellate Prymnesium parvum in fishponds in Palestine. Pal. J. of Bot. Jerusalem Series 4, 14–23.

Reichenbach-Klinke, H., 1973. Fish pathology. T. F. H. Publications, Neptune City, NJ. 512.

Texas Tech University, Rakib H. Rashel, May 2020

Reynolds, C. S., 1984. The ecology of freshwater phytoplankton. Cambridge University Press.

Reynolds, C. S., Walsby, A. E., 1975. Water-Blooms. Biological Reviews 50(4), 437– 481.

Rhodes, K., Hubbs, C., 1992. Recovery of the Pecos River from a red tide fish kill. Southwestern Naturalist 37:178-187.

Roelke, D. L., Errera, R. M., Kiesling, R., Brooks, B. W., Grover, J. P., Schwierzke, L., Pinckney, J. L., 2007. Effects of nutrient enrichment on Prymnesium parvum population dynamics and toxicity: results from field experiments, Lake Possum Kingdom, USA. Aqua. Micro. Eco. 46(2), 125–140.

Roelke, D. L., Grover, J. P., Brooks, B. W., Glass, J., Buzan, D., Southard, G. M., Nelson, J., 2011. A decade of fish-killing Prymnesium parvum blooms in Texas: roles of inflow and salinity. J. of Plank. Res. 33(2), 243–253.

Roelke, D. L., Brooks, B. W., Grover, J. P., Gable, G. M., Schwierzke-Wade, L., Hewitt, N. C., Weyhenmeyer, G., 2012. Anticipated human population and climate change effects on algal blooms of a toxic haptophyte in the south-central USA. Can. J. of Fish. Aquat. Sci. 69, 1389–1404.

Roelke, D.L., Barkoh, A., Brooks, B.W., Grover, J.P., Hambright, K.D., La Claire, J.W. II, Moeller, P. D.R., Patiño, R., 2016. A chronicle of a killer alga in the west: ecology, assessment, and management of Prymnesium parvum blooms. Hydrobiologia 764, 29–50

Round, F. E., 1965. The epipsammon; a relatively unknown freshwater algal association. Bri. Phy. Bull. 2(6), 456–462.

Sager, D. R., Barkoh, A., Buzan, D. L., Fries, L. T., Glass, J. A., Kurten, G. L., Swanson, E., 2008. Toxic Prymnesium parvum: A potential threat to US reservoirs. In M. S. Allen, S. Sammons, M. J. Maceina (Eds.), Balancing Fisheries Management and Water Uses for Impounded River Systems, 62, 261–273.

Schindler, D. W., 1977. Evolution of phosphorus limitation in lakes. Science 195(4275), 260–262.

Schindler, D. W., 1978. Factors regulating phytoplankton production and standing crop in the world's freshwaters. Limnol. Oceanogr. 23(3).

Sellner, K. G., Doucette, G. J., Kirkpatrick, G. J., 2003. Harmful algal blooms: causes, impacts and detection. J. of Ind. Micro. and Biotech. 30(7), 383–406.

Shilo, M., Aschner, M., 1953. Factors governing the toxicity of cultures containing the phytoflagellate Prymnesium parvum Carter. J. of Gen. Micro. 8(3), 333–343.

Texas Tech University, Rakib H. Rashel, May 2020

Shilo, M., Shilo, M., 1953. Conditions which determine the efficiency of ammonium sulphate in the control of Prymnesium parvum in fish breeding ponds. App. Micro. 1(6), 330–333.

Sieburth, J. M., Johnson, P. W., Hargraves, P. E., 1988. Ultrastructure and ecology of Aureococcus Anophageferens Gen. Et Sp. Nov. (chrysophyceae): The dominant picoplankter during a bloom in Narragansett Bay, Rhode Island, Summer 1985. J. of Phyco. 24(3), 416–425.

Smith, V. H., 1986. Light and nutrient effects on the relative biomass of blue-green algae in lake phytoplankton. Can. J. of Fish. and Aqu. Sci. 43(1), 148–153.

Sournia, A., Chrdtiennot-Dinet, M. J., Ricard, M., 1991. Marine phytoplankton: how many species in the world ocean? J. of Plank. Res. 13(5), 1093–1099.

Southard, G. M., Fries, L. T., Barkoh, A., 2010. Prymnesium parvum: The Texas Experience. JAWRA J. of the Ame. Wat. Res. Asso. 46(1), 14–23.

Taylor, D.L., Seliger, H.H., 1979. Toxic dinoflagellate blooms. Elsevier/North Holland, New York. 505.

Trainer, V. L., Baden, D. G., 1999. High affinity binding of red tide neurotoxins to marine mammal brain. Aqu. Toxi. 46(2), 139–148.

Tucker, C. S., 2000. Off-flavor problems in aquaculture. Rev. Fish. Sci. 8(1), 45–88.

Ulitzur, S., Shilo, M., 1964. A sensitive assay system for determination of the ichthyotoxicity of Prymnesium parvum. J. of Gen. Micro. 36, 161–169.

Ulitzur, S., Shilo, M., 1966. Mode of action of Prymnesium parvum ichthyotoxin. The J. of Protozo. 13(2), 332–336.

Uronen, P., Lehtinen, S., Legrand, C., Kuuppo, P., Tamminen, T., 2005. Haemolytic activity and allelopathy of the haptophyte Prymnesium parvum in nutrient-limited and balanced growth conditions. Mar. Eco. Pro. Ser. 299, 137–148.

Valenti, T. W., James, S. V., Lahousse, M. J., Schug, K. A., Roelke, D. L., Grover, J. P., Brooks, B. W., 2010. A mechanistic explanation for pH-dependent ambient aquatic toxicity of Prymnesium parvum carter. Toxicon: Official J. of the Inter. Soci. on Toxi. 55(5), 990–998.

Van Dolah, F. M., 2000. Marine algal toxins: origins, health effects, and their increased occurrence. Environ. Hea. Perspe. 108, 133–141.

VanLandeghem, M. M., Meyer, M. D., Cox, S. B., Sharma, B., Patiño, R., 2012. Spatial and temporal patterns of surface water quality and ichthyotoxicity in urban and rural river basins in Texas. Water Research 46(20), 6638–6651.

Texas Tech University, Rakib H. Rashel, May 2020

VanLandeghem, M.M., Farooqi, M., Southard, G.M., Patiño, R., 2015a. Associations between water physicochemistry and Prymnesium parvum presence, abundance, and toxicity in west texas reservoirs. J. Am. Water Resour. Asso. 51, 471–486.

VanLandeghem, M.M., Farooqi, M., Southard, G., Patiño, R., 2015b. Spatiotemporal associations of reservoir nutrient characteristics and the invasive, harmful alga Prymnesium parvum in west Texas. J. of the Ame. Wat. Res. Asso. 51, 487–501.

Vasas, G., M-Hamvas, M., Borics, G., Gonda, S., Máthé, C., Jámbrik, K., Nagy, Z. L., 2012. Occurrence of toxic Prymnesium parvum blooms with high protease activity is related to fish mortality in Hungarian ponds. Harmful Algae 17, 102– 110.

Vila, M., Giacobbe, M. G., Masó, M., Gangemi, E., Penna, A., Sampedro, N., Galluzzi, L., 2005. A comparative study on recurrent blooms of Alexandrium minutum in two Mediterranean coastal areas. Harmful Algae 4(4), 673–695.

Viviani, R., 1992. Eutrophication, marine biotoxins, human health. The Sci. of the T. Envir. 631–662.

Walker, S. R., Lund, J. C., Schumacher, D. G., Brakhage, P. A., McManus, B. C., Miller, J. D., Gitelson, A. A., 2008. Nebraska experience. Adv. in Expe. Med. and Bio. 619, 139–152.

Wang, D. Z., 2008. Neurotoxins from marine dinoflagellates: A Brief Review. Marine Drugs 6(2), 349–371.

Watson, S. B., McCauley, E., Downing, J. A., 1997. Patterns in phytoplankton taxonomic composition across temperate lakes of differing nutrient status. Limnol. Oceanogr. 42(3), 487–495.

Yariv, J., Hestrin, S., 1961. Toxicity of the extracellular phase of Prymnesium parvum cultures. J. of Gen. Micro. 24(2), 165–175.

Texas Tech University, Rakib H. Rashel, May 2020

CHAPTER Ⅱ

INFLUENCE OF GENETIC BACKGROUND, SALINITY, AND INOCULUM SIZE ON GROWTH OF THE ICHTHYOTOXIC GOLDEN ALGA (PRYMNESIUM PARVUM)

Abstract

Salinity (5–30 psu) effects on golden alga growth were determined at a standard laboratory temperature (22 °C) and one associated with natural blooms (13 °C).

Inoculum-size effects were determined over a wide size range (100–100,000 cells ml-1).

UTEX-2797, maximum cell density increased as salinity increased from 5 to ~10–15 psu and declined thereafter regardless of temperature but r remained generally stable and only declined at salinity of 25–30 psu. In addition, maximum cell density correlated positively with r and early cell density, the latter also being numerically highest at salinity of 15 psu. In UTEX-995, maximum cell density and r responded similarly to changes in salinity – they remained stable at salinity of 5–10 psu and 5–15 psu, respectively, and declined at higher salinity. Also, maximum cell density correlated with r but not early cell density. Inoculum size positively and negatively influenced maximum cell density and r, respectively, in both strains and these effects were significant even when the absolute size difference was small (100 versus 1000 cells ml-1). When cultured under

Texas Tech University, Rakib H. Rashel, May 2020 similar conditions, UTEX-2797 grew faster and to far higher density than UTEX-995. In conclusion, (1) UTEX-2797’s superior growth performance may explain its relatively wide distribution in the USA, (2) the biphasic growth response of UTEX-2797 to salinity variation, with peak abundance at salinity of 10–15 psu, generally mirrors golden alga abundance-salinity associations in US inland waters, and (3) early cell density – whether artificially manipulated or naturally attained – can influence UTEX-2797 bloom potential.

*This chapter published as: Rashel, R.H, Patiño, R., 2017. Influence of genetic background, salinity, and inoculum size on growth of the ichthyotoxic golden alga

(Prymnesium parvum). Harmful Algae 66, 97–104.

1. Introduction

The worldwide incidence and ecological impacts of harmful algal blooms (HABs) have been on the rise in the last few decades (Hallegraeff, 1993; Sellner et al., 2003;

Anderson, 2009; Roelke et al., 2016). One such HAB species is the unicellular haptophyte Prymnesium parvum, commonly known as golden alga. Golden alga is present in all continents except Antartica and, as with other HAB species, their geographic range has expanded in recent years particularly in inland aquatic habitats

(Roelke et al., 2016). The first documented toxic bloom of golden alga in North America occurred in the Pecos River (USA) in 1985 (James and De La Cruz, 1989), but records of its presence or toxic bloom incidence now exist for a total of 23 US states (Roelke et al.,

2016). A variety of biotic and abiotic factors can influence the growth and toxicity of

Texas Tech University, Rakib H. Rashel, May 2020

golden alga (Hambright et al., 2010, 2015; Granéli et al., 2012; Israël et al., 2014;

Manning and La Claire, 2010; Patiño et al., 2014; VanLandeghem et al., 2015a, 2015b,

2015c; Roelke et al., 2016). A retrospective study of reservoir water quality in two major

river basins of the USA concluded that salinity is one of the most important variables

associated with golden alga distribution and bloom formation at the landscape scale

(Patiño et al., 2014). Also, logistic regression modeling of field data collected from

surface waters in Texas and New Mexico (USA) revealed that golden alga abundance is

generally a good predictor of ichthyotoxicity (VanLandeghem et al., 2015c).

Golden alga is a euryhaline species believed to have originated in coastal or

marine environments (Nicholls, 2003), and its first appearance in the continental USA

occurred in inland waters of relatively high salinity (James and De La Cruz, 1989;

Rhodes and Hubbs, 1992). While golden alga can grow in the laboratory at salinities from

~3–4 psu to full-strength seawater, growth rates are typically highest at the upper end of

the salinity range (≥ 15 psu) (Padilla, 1970; Larsen and Bryant, 1998; Baker et al., 2007,

2009; Hambright et al., 2014). In the field, however, the positive association between

golden alga abundance and salinity is normally restricted to a lower range of salinities,

from ~0.5–1.0 psu (Roelke et al., 2011; Patiño et al., 2014) to ~8–10 psu [Gou, 1983

(cited in Guo et al., 1996); Hambright et al., 2010, 2014; Roelke et al., 2012; Israël et al.,

2014; VanLandeghem et al., 2015a]. At higher salinity, the association becomes negative

[Gou, 1983; cited in Guo et al. (1996); Israël et al., 2014]. It appears, therefore, that the salinity range for optimal growth of golden alga in natural inland waters does not match observations in the laboratory. Additional experimental studies that focus on salinity as

Texas Tech University, Rakib H. Rashel, May 2020

driver of golden alga growth are necessary to understand and reconcile differences

between laboratory and field observations.

The influence of “inoculum size” on survival and growth of golden alga has only

recently began to garner attention. An in-lake microcosm study of a reservoir in the

southern Great Plains (USA) found that the addition of ~500 golden alga cells ml-1 over

background levels promoted winter growth under certain conditions (Errera et al., 2008).

Another study reported that survival of golden alga following its introduction to an

experimental microcosm depended on inoculum size, with a minimum initial abundance

of ~6400 cells ml-1 required for survival at one-week post-inoculation (Acosta et al.,

2015). To the authors’ knowledge, the influence of inoculum size on growth of golden alga has not been addressed by laboratory studies and, in fact, inoculum size is often not reported. A better understanding of how inoculum size affects growth is necessary for proper evaluation of laboratory-generated growth indices and of the value of those indices to an understanding of field observations.

The objectives of this study were to characterize the effects of genetic background

(strain), salinity, and inoculum size on golden alga growth in laboratory batch cultures.

The effects of salinity were assessed at two temperatures, one within the optimal range for growth in the laboratory (22 °C; Baker et al, 2007) and another favorable to blooms in the field (13 °C; VanLandeghem et al., 2015a). The primary strain of golden alga used in this study is a Texas isolate of relatively widespread distribution in the continental USA but for comparative purposes, the effects of inoculum size and salinity were also investigated on a genetic strain of limited distribution in US inland waters.

Texas Tech University, Rakib H. Rashel, May 2020

2. Materials and methods

2.1. Basic culture procedures

2.1.1. Algal strains

The two strains of P. parvum used in this study were obtained from the UTEX

Culture Collection of Algae (The University of Texas at Austin, Texas, USA). The

primary strain used, UTEX LB 2797 (UTEX-2797), is an isolate from the Colorado River

(Texas) that is genetically similar to a strain from Scotland and seems to be the most widespread strain in US inland waters (Lutz-Carrillo et al., 2010). The second strain,

UTEX LB 995 (UTEX-995), is an isolate from England (River Blackwater, Essex) and, within the USA, the geographic distribution of a genetically related strain seems to be restricted to the northeast coast (Lutz-Carrillo et al., 2010). Stock cultures of both strains were maintained in a standard base medium at salinity of 5 psu.

2.1.2. Stock cultures

The standard base medium for this study consisted of a modified UTEX artificial sea water medium (https://utex.org/products/artificial-seawater-medium) containing NaCl

(Fisher S271), MgSO4∙7H2O (Sigma 230391), KCl (Fisher P217) and CaCl2∙2H2O

(Sigma C3881) at final concentrations of 84 mM, 2.84 mM, 2.16 mM, 0.54 mM,

respectively. The final salinity and pH were 5 psu and ~ 8.1, respectively. The salinity of culture media was confirmed by measurement with a pre-calibrated YSI 85 probe

(Yellow Springs Incorporated, Yellow Springs, OH, USA). Ferrous ammonium sulfate in

Texas Tech University, Rakib H. Rashel, May 2020

the original recipe for trace metals was replaced with an equimolar amount of ferric chloride (Sigma 236489) because of reported toxic effects of ammonium on golden alga growth (Grover et al., 2007). Base medium was enriched with F/2 levels of nutrients and vitamins (Guillard, 1975). All ingredients were added before autoclaving except vitamins. After autoclaving, vitamin B12 (Fisher BP862), biotin (Sigma B4639), and

thiamine (Sigma T1270) were added aseptically by filter-sterilization (Nalgene, 0.45 μm,

sterile analytical filter unit, Thermo Fisher Scientific Inc., Waltham, MA, USA). Filter-

sterilized (Nalgene, 0.45 μm) sodium bicarbonate (Sigma S5761) was then added as an

inorganic carbon source at a concentration of 93 μM. A salinity of 5 psu was chosen for

the base medium because it is within the typical range of salinities associated with golden

alga habitat in inland waters of the continental USA (Israël et al., 2014; VanLandeghem

et al., 2015a), and also because golden alga growth in laboratory cultures is relatively

reduced at lower salinities (Baker et al., 2007).

Stock cultures of both strains were grown in 250 ml Erlenmeyer flasks filled with

of 100 ml of standard base medium. Non-axenic cultures were maintained in an incubator

(I36LLVLC9; Percival Scientific Inc. Perry, IA, USA) at 22 °C and 12:12 h light:dark

photoperiod with a light intensity of ~6500 lux. Regular fluorescent light was used and its

intensity was confirmed with a Traceable light meter pen (Traceable S 98187, Control

Company, Friendswood, Texas, USA). Culture flasks were gently swirled once daily. As

the first stock cultures (from original UTEX inocula) reached late exponential growth,

they were used to inoculate secondary cultures at a density of 10,000 cells ml-1 under the

same conditions. This process was repeated at least 3 times before any experimentation

Texas Tech University, Rakib H. Rashel, May 2020

and indefinitely thereafter to maintain stock cultures ready as a source of inoculum for

experimental cultures.

2.2. Experimental cultures

2.2.1. Influence of inoculum size and genetic strain

The influence of inoculum size on growth of both strains of golden alga was

examined over a wide range of sizes, from the lowest reported in earlier studies (100 cells

ml-1; Baker et al., 2007, 2009) to sizes well above those considered to represent “bloom”

conditions (≥ 10,000 cells ml-1; Roelke et al., 2007) – up to 100,000 cells ml-1. Cell abundance was determined every 3 days and cultures were grown until they reached stationary phase. All inoculum-size treatments were conducted in triplicate. Other basic culture procedures and culture media for these experiments were as described for stock cultures (salinity of 5 psu, 22 °C).

2.2.2. Influence of salinity, genetic strain, and temperature

Basic culture procedures and culture media for these experiments were the same as for stock cultures. Salinity was manipulated by adding NaCl (Fisher S271) to the base culture medium (salinity of 5 psu) to achieve salinities of 5 to 30 psu. Salinity of experimental media was confirmed by measurement with a pre-calibrated YSI 85 probe.

Unless otherwise noted, inoculum size for these experiments was 100 cells ml-1 and

temperature was maintained at 22 °C. As golden alga blooms usually occur during the

cooler months of the year, the salinity experiment with UTEX-2797 (Texas isolate) was

Texas Tech University, Rakib H. Rashel, May 2020

also conducted at 13 °C, which is within the range of temperatures at which peak

abundances occur in the field (VanLandeghem et al. 2015a). Cell abundance was

determined every 3 days and cultures were grown until they reached stationary phase.

Two complete, independent trials were conducted for each strain-temperature

combination, and all treatments in each trial were conducted in triplicate.

2.3. Analytical procedures

2.3.1. Cell counts

Aliquots of 500 µl of experimental media were taken from each replicate flask

and used to determine cell numbers using a hemocytometer (Hausser Scientific

Company, Horsham, PA, USA) under a compound microscope (Olympus BH2, Olympus

America Inc., Center Valley, PA, USA). Sub-samples were diluted with fresh medium if

needed to maintain the total cell count below 50,000 cells ml-1, and the counting method

used was as described by Southard (2005). A total of three counts per replicate flask were

taken and the average cell number was reported for each replicate. The limit of detection

(LOD) for this procedure is 333 cells ml-1.

2.3.2. Estimation of growth indices and statistical analyses

The primary growth indices estimated were exponential growth rate (r, day-1) and

maximum cell density (cells ml-1). Growth rate was calculated using the following

equation (Guillard, 1973; Wood et al., 2005):

Texas Tech University, Rakib H. Rashel, May 2020

ln N ln N = t t 2 − 1 𝑟𝑟 2 1 −

where N1 and N2 are cell densities at times t1 and t2 (t2 > t1). For each individual replicate,

times were chosen so that they bracket the linear portion of the ln-transformed growth

curve. Exponential growth rate would equal specific growth rate (μ) in the absence of

mortality (Wood et al., 2005). No assumptions about mortality rate were made in this

study and estimated growth rates are reported as r. The highest cell count was considered

to be the maximum cell density achieved by each replicate. Mean values (± SEM) are

reported for all growth indices.

Two-way ANOVA was used to assess the effects of inoculum size and genetic

strain on growth indices, and pairwise differences between inoculum-size levels were

assessed separately for each strain using Tukey’s Honest Significant Difference tests

(Tukey’s HSD). Effects of salinity were analyzed separately for each strain using 2-way

ANOVA, with salinity and trial as factors (two full-treatment trials were conducted for each strain). If trial had a significant effect or there was significant interaction between salinity and trial, pairwise differences between salinity levels were determined within individual trials using Tukey's HSD. Otherwise, values from both trials were pooled for multiple comparison analysis.

Results from the inoculum-size experiment indicated that variation in initial cell density greatly affects both r and maximum cell density under otherwise constant culture conditions (see section 3.1). This observation prompted an interest in examining possible associations between maximum cell density (as index of bloom potential) and other growth parameters under a constant inoculum size but variable culture conditions. More

Texas Tech University, Rakib H. Rashel, May 2020

specifically, the post hoc question of interest is whether maximum cell density is a

function of not only r but also “early cell density” under varying levels of salinity (but

constant inoculums size) for each strain-temperature combination. Abundance values at day 3 for UTEX-2797 at 22 °C, day 9 for UTEX-2797 at 13 °C, and day 9 for UTEX-995

at 22 °C were used as indices of early cell density. These times correspond to or are near

the onset of exponential growth for each strain-temperature combination at most salinity levels (see section 3.2). A few replicates had undetectable abundance values (< 333 cells ml-1) at these early times. Undetectable values were replaced with √(LOD/2), a replacement approach with a low error rate when < 50 percent of the values are censored

(Croghan and Egeghy, 2003). Partial correlations were used for this analysis, which allow an evaluation of bivariate associations while removing the potential influence on the association by a third variable, also known as the control variable (Zar, 2010). The rationale for this analytical approach is that results under section 3.1 suggested the existence of associations among all three variables. A Bonferroni adjustment to α was applied to account for the family-wise error rate of this analysis.

All statistical analyses were conducted with Dell Statistica™ (version 13,

StatSoft., Inc., Tulsa, OK, USA), and graphics were made using GraphPad Prism 6

(GraphPad Software, Inc., La Jolla, CA, USA).

3. Results

3.1. Influence of inoculum size and genetic strain

Texas Tech University, Rakib H. Rashel, May 2020

Stationary growth phase for UTEX-2797 was achieved on day 21 at inoculum size

of 100–10,000 cells ml-1 and on day 18 at 100,000 cells ml-1. Stationary phase for UTEX-

995 was reached on day 36 at 100, day 33 at 1000, and day 30 at 10,000–100,000 cells ml-1 (Fig. 1). On a semi-ln plot, the linear portion of the exponential growth phase

generally bracketed 3–12 days of culture at all inoculum sizes for UTEX-2797, and 9–18

days at inoculum size of 100–1000 cells ml-1 or 9–21 days at 10,000–100,000 cells ml-1

for UTEX-995 (data not shown).

Results of 2-way ANOVA indicated that r and maximum cell density differed

between strains (F1,16 = 737, p < 0.0001 and F1,16 = 13,940, p < 0.0001) and among

inoculum sizes (F3,16 = 578, p < 0.0001 and F3,16 = 261, p < 0.0001). An interaction effect between strain and inoculum size was noted for maximum cell density (F3,16 = 104, p <

0.0001) but not r (p > 0.05). When cultured under the same conditions, r was higher for

UTEX-2797 than UTEX-995 (Fig. 2A, C). As inoculum size increased from 100 to

100,000 cells ml-1, r gradually decreased from 0.64 to 0.22 day-1 in UTEX-2797 (Fig. 2A;

Tukey’s HSD, p < 0.05) and from 0.43 to 0.04 day-1 in UTEX-995 (Fig. 2C; Tukey’s

HSD, p < 0.05). Maximum cell density was much higher for UTEX-2797 than UTEX-

995 (Fig. 2B, D) and unlike r, as inoculum size increased maximum cell density also increased in both strains (Figs. 2B, D; Tukey’s HSD, p < 0.05 for both strains).

3.2. Influence of salinity, genetic strain, and temperature

Unless otherwise noted, inoculum size was 100 cells ml-1 for experiments under

this section. Stationary growth phase at 22 °C was generally reached at day 21 by UTEX-

2797 and day 36 by UTEX-995 at all salinity levels (Fig. 3A, B). On a semi-ln plot, the

Texas Tech University, Rakib H. Rashel, May 2020

linear portion of the exponential growth phase in UTEX-2797 generally bracketed 3–12 days at all salinity levels, and in UTEX-995 it bracketed 9–18 days at salinity of 5–20 psu, and 15–24 days at salinity of 25 psu (data not shown). The UTEX-995 grew very poorly at salinity of 30 psu and growth parameters could not be estimated.

In UTEX-2797 at 13 °C, stationary growth phase was generally reached by day

36 of culture (Fig. 3C) and, on a semi-ln plot, the linear portion of the exponential growth phase generally bracketed 9–33 days at salinity of 5–20 psu and 15–27 days at salinity of

25 psu (data not shown). Growth of UTEX-2797 was poor under the combination of 13

°C and salinity of 30 psu, and r and early cell density could not be estimated.

In UTEX-2797 at 22 °C, 2-way ANOVA showed no differences between trials for

r (F1,24 = 3.529, p = 0.07) but a significant between-trial effect was noted for maximum cell density (F1,24 = 17.8, p = 0.0003). Salinity had a significant effect on r (F5,24 = 13.09,

p < 0.0001) and on maximum cell density (F5,24 = 316.6, p < 0.0001). Interaction effects were observed between salinity and trial for maximum cell density (F5,24 = 5.06, p =

0.0026) but not for r (p > 0.05). Exponential growth rate did not vary significantly as

salinity increased from 5 to 25 psu but was significantly reduced at 30 (Tukey’s HSD, p <

0.05; Fig. 4A). Maximum cell density increased in both trials as salinity increased from 5

to ~10–15 psu and generally decreased at higher salinity, reaching its lowest values at

salinity of 30 psu (Tukey’s HSD, p < 0.05; Fig. 4B and Supplemental Fig. 1). Significant

inhibition of both growth parameters at salinity of 30 psu relative to 5 psu was also

observed when inoculum size was 100,000 cells ml-1 (Supplemental Fig. 2).

In UTEX-995 at 22 °C, 2-way ANOVA showed no differences between trials for r (F1,20 = 3.657, p = 0.50) or maximum cell density (F1,20 = 0.218, p = 0.64). Salinity had

Texas Tech University, Rakib H. Rashel, May 2020

a significant effect on r (F4,20 = 11.6, p < 0.0001) and maximum cell density (F4,20 =

144.8, p < 0.0001). No interaction effects were observed for either r (p > 0.05) or maximum cell density (p > 0.05). Exponential growth rate did not vary significantly as salinity increased from 5 to 15 psu but was reduced at 20–25 psu (Fig. 4C; Tukey’s HSD, p = 0.05) and became negligible (unable to be determined) at 30 psu. Maximum cell density showed a response pattern to salinity similar to r – it did not vary significantly from salinity of 5 to 10 psu but gradually declined as salinity increased thereafter

(Tukey’s HSD, p < 0.05) and reached negligible levels at 30 psu (Fig. 4D).

In UTEX-2797 at 13 °C, 2-way ANOVA showed no differences between trials for

r (F1,20 = 0.462, p = 0.07) or maximum cell density (F1,20 = 0.453, p = 0.51). Salinity had

a significant effect on r (F4,20 = 26.0, p < 0.0001) and maximum cell density (F5,24 =

553.0, p < 0.0001). No interaction effects were observed for either r (p > 0.05) or maximum cell density (p > 0.05). Exponential growth rate did not vary significantly

(relative to control) as salinity increased from 5 to 20 psu but declined at 25 psu (Tukey’s

HSD, p < 0.05) and became negligible at 30 psu (Fig. 4E). Maximum cell density, however, increased as salinity increased from 5 to ~10–15 psu and decreased at higher salinities (Tukey’s HSD, p < 0.05) (Fig. 4F).

Partial correlation analysis showed that, for UTEX-2797, maximum cell density was positively associated with r and with early cell density at both incubation temperatures (Table 1). For UTEX-995, however, only the partial correlation between maximum cell density and r was significant (Table 1). Simple correlation analysis yielded similar results (Supplemental Table 1). Visual examination of early cell density

Texas Tech University, Rakib H. Rashel, May 2020

summary statistics suggested that the highest values occurred at salinity of 15 psu

(Supplemental Table 2).

4. Discussion

4.1. Inoculum size and genetic strain

The Texas strain of golden alga, UTEX-2797 greatly outperformed UTEX-995 in terms of growth potential when cultured under the same conditions. In both genetic strains, however, r decreased as inoculum size increased. Exponential growth rates in cultures inoculated with 100,000 cells ml-1 were 33 and 11 percent of those in cultures

inoculated with 100 cells ml-1 for UTEX-2797 and UTEX-995, respectively. These observations are consistent with reports for other microalgal species (Pratt, 1940; Lu et al., 2012; Gani et al., 2016). While the reason for this inverse association is uncertain, there is evidence indicating that toxins produced by golden alga can inhibit its own growth (Olli and Trunov, 2007) and as inoculum size increases, presumably so would the concentration of growth-inhibiting compounds. This particular scenario has been previously used to explain the negative association between inoculum size and r observed in Chlorella spp. (Pratt, 1940). In addition, decreased effective irradiance (due to increased turbidity) may have contributed to the reduction in r, especially at the higher inoculum sizes. Inoculum size can influence the metabolic profile of some microalgae

(Lu et al., 2012, 2013), and inoculum size-dependent changes in growth rate could be due to or associated with inoculum size-dependent changes in metabolism.

Texas Tech University, Rakib H. Rashel, May 2020

In contrast to its inverse association with r, inoculum size was positively

associated with maximum cell abundance in both strains of golden alga. Maximum cell

density was 53 and 81 percent higher in UTEX-2797 and UTEX-995, respectively, when

inoculated at an initial density of 100,000 cells ml-1 compared to 100 cells ml-1. These

findings are consistent with the results of some (Gani et al., 2016; Heidari et al., 2016)

but not all (Pratt, 1940; Lu et al., 2012) previous studies of microalgae. Some of the

earlier studies did not analyze their data quantitatively, however, and their conclusions

are difficult to evaluate. For example, although Pratt (1940) concluded that initial and

maximum cell densities are not associated in Chlorella vulgaris, the time-series plots

provided (Figs. 1–3 in Pratt, 1940) appear to show a small positive association as the

cultures reached stationary phase. Also, a more recent study of the same species (C.

vulgaris) reported a positive association between initial and final cell densities under

conditions of nutrient deprivation (Heidari et al., 2016). Curiously, the difference in

maximum cell density between the 1000 and 100 cells ml-1 inoculum-size treatments was

55,000 cells ml-1 for UTEX-2797, which is more than 50-fold greater than the difference

in initial density (= 900 cells ml-1). Thus, differences in nutrient availability in relation to

respective absolute increases in biomass cannot explain the association observed between

inoculum size and maximum cell density. An examination of cause-effect relationships is

outside the scope of the present study and an explanation for these observations will

require further investigation.

4.2. Salinity, genetic strain, and temperature

Texas Tech University, Rakib H. Rashel, May 2020

In this series of experiments, UTEX-2797 also grew faster and to considerably higher maximum cell density than UTEX-995 when cultured under the same conditions.

Notably, maximum cell abundance for UTEX-2797 cultured at 13 °C was still higher – over 3-fold greater – than for UTEX-995 grown at 22 °C (Fig. 4).

Exponential growth rate remained generally stable in the salinity range of 5 to 25 psu in UTEX-2797 and 5 to 15 in UTEX-995. At respective higher salinities, r decreased as salinity increased in both strains. It is difficult to compare these findings with results of previous studies of golden alga growth because they either used unspecified inoculum sizes (e.g., Padilla, 1970; Larsen and Bryant, 1998; Hambright et al., 2014), which can strongly influence growth indices (see preceding discussion), or were conducted in a multivariate context where other culture conditions were simultaneously manipulated

(e.g., Baker et al., 2007). To the authors’ knowledge, there is only one earlier study of golden alga that is based on comparable experimental designs (20 °C, inoculum size of

100 cells ml-1), and this study reported a positive association between r and salinity in the

range of 0.5 to 4 psu for UTEX-2797 (Baker et al., 2009). These observations suggest

that the response of r to changes in salinity can be described as a unimodal function but

with a remarkably flat or extended “peak.” Namely, for UTEX-2797 at 20–22 °C, the

association is positive at salinity below 4–5 psu (Baker et al., 2009), non-existent between 5 and ~25 psu, and negative at > 25 psu (present study). In addition, a field study reported a positive association between growth rate and salinity levels below 3 psu

(Hambright et al., 2015). Thus, under otherwise constant environmental conditions, r

seems to be negatively impacted by salinity only at the extremes of the natural salinity

range. The same general pattern was observed at 13 °C for UTEX-2797 [relatively stable

Texas Tech University, Rakib H. Rashel, May 2020

(no association) at salinity of 5–20 psu, and negative association > 20 psu]. Thus, the

unimodal response of r to salinity variation seems to occur independently of temperature.

The response of maximum cell density to changes in salinity differed between the

two strains of golden alga. In UTEX-2797, maximum cell density significantly increased

as salinity increased from 5 to ~10–15 psu and decreased at higher salinity regardless of temperature. In contrast, maximum cell density in UTEX-995 did not change when

salinity increased from 5 to 10 psu but gradually decreased as salinity increased further.

Because r remained relatively stable at a wide range of salinity, especially for UTEX-

2797, it appears that salinity-dependent changes in maximum cell density are not an

exclusive function of r. Results of partial correlation analysis supported this conclusion

for UTEX-2797, and also pointed to cell density prior to exponential growth as a

potentially important contributing variable.

4.3. Early cell density

Partial (and simple) correlation analyses showed that maximum cell density is

positively and strongly associated with r over a wide salinity range in both strains of

golden alga. This result was anticipated as it is consistent with a previous report by Baker

et al. (2007) indicating that maximum cell density and r (or μ) are strongly correlated in

UTEX-2797 cultured under varying levels of salinity, temperature, and light intensity

[Pearson r = 0.85; simple correlation analysis of data from Table 2 in Baker et al.

(2007)]. A novel finding of the present study, however, is that maximum cell density is

also positively correlated with early cell density in UTEX-2797 (but not UTEX-995).

This observation may explain why maximum cell density in UTEX-2797 increased as

Texas Tech University, Rakib H. Rashel, May 2020

salinity increased from 5 to ~10–15 psu despite an unchanging r over the same salinity

range – namely, because early cell density values were also numerically higher at salinity

of 10–15 psu regardless of temperature (Supplemental Table 1). [Note: the relationship

between initial and final cell densities during the period of exponential growth is fixed,

that is, linear (see section 2.3.2) and therefore, differences in maximum cell densities will

occur under conditions of constant r when initial cell densities differ.] These observations suggest that, under certain conditions, pre-exponential growth (during transition from lag to exponential phase) can be an important determinant of bloom size in some genetic strains of golden alga.

4.4. Ecological considerations and conclusions

Stock cultures for this study were maintained at a salinity of 5 psu for many generations before experimentation. It is therefore reasonable to assume they were well adapted to a salinity level typical of golden alga habitat in inland waters of North

American (Israël et al., 2014; Patiño et al., 2014; Roelke et al., 2016). Curiously, growth potential (maximum cell density) increased as salinity was raised from 5 to 10–15 psu in

UTEX-2797 but not UTEX-995, and declined at higher salinity in both strains. These observations indicate a preference for mid-range salinity (10–15 psu) by UTEX-2797 and for lower-range salinity (~5–10 psu), closer to the typical salinity of golden alga habitat, by UTEX-995. When cultured under the same conditions, however, UTEX-2797 outgrew

UTEX-995 by large margins suggesting that the wider distribution of UTEX-2797 in the

USA (Lutz-Carrillo et al., 2010) is a function of its superior growth performance and correspondingly higher invasive potential. This conclusion is based on results of

Texas Tech University, Rakib H. Rashel, May 2020

monoculture studies. Co-culture of UTEX-2797 with UTEX-995 as well as other golden alga strains would be required to confirm the relative growth advantages of UTEX-2797.

The pattern of growth responses to salinity variation observed for UTEX-2797 is

generally similar to golden alga abundance-salinity associations reported in US inland waters. Namely, the highest growth potential for UTEX-2797 in the laboratory was observed at salinity of 10–15 psu (present study), and abundance-salinity associations in the field seem to be positive up to ~8–10 psu and become negative at higher salinity

(Hambright et al., 2010, 2014; Roelke et al., 2012; Israël et al., 2014; Vanlandeghem et

al., 2015a; Roelke et al., 2016). In the Pecos River, one of the saltiest river systems in

North America, the incidence of golden alga markedly decreases when salinity exceeds

~12 psu and becomes undetectable at ~19 psu (Israël et al., 2014). Field studies in the

USA, especially the south-central region, are likely to have sampled or mostly sampled

UTEX-2797 (Lutz-Carrillo et al., 2010). Thus, one reason for this similarity in

abundance-salinity associations may be that field (USA) and laboratory (present) studies

have shared the same study subject, UTEX-2797. A biphasic association between golden

alga abundance and salinity, with peak abundance levels at ~8 psu, has been also reported

for a series of saline lakes in the People’s Republic of China [Gou (1983), cited in Guo et

al. (1996)]. The identity of the genetic strain(s) sampled in the PRC study is unknown.

Because growth responses to salinity variation differ among strains (Larsen and Bryant,

1998; Hambright et al., 2014; present study), characterization of the genetic background

of sampled populations would broaden the utility of future field studies of golden alga.

The finding that early cell abundance – whether artificially manipulated (i.e., by

inoculum size) or naturally attained – can influence golden alga growth has ecological

Texas Tech University, Rakib H. Rashel, May 2020 and practical implications. Maximum abundance levels were significantly enhanced by even a small absolute increase in inoculum size (+900 cells ml-1) in both strains of golden alga. Also, naturally-attained cell densities under varying levels of salinity (but constant inoculum size) were positively associated with maximum abundance levels in UTEX-

2797. These observations are consistent with results of an in-lake microcosm study where the addition of ~500 cells ml-1 over background levels significantly enhanced winter abundance of golden alga under certain conditions (Errera et al., 2008). The earlier study thus concluded that immigration can be an important factor determining whether a bloom will form (Errera et al., 2008). Another study reported that post-inoculation survival in a golden alga-free microcosm system is influenced by the size of the inoculum, also termed propagule pressure in the context of colonization and dispersal mechanisms (Acosta et al., 2015). These previous (Errera et al., 2008; Acosta et al., 2015) and present observations together highlight the importance of early cell density as driver of not only golden alga survival and invasion potential, but also growth and bloom potential. The ecological relevance of the negative association between inoculum size and r is more difficult to ascertain but, as already mentioned (section 4.1) this phenomenon could be interpreted in the context of inoculum size-dependent changes in levels of self-toxicity or effective irradiance. Lastly, from a practical perspective, the present results also suggest the need for reporting inoculum sizes by laboratory studies of golden alga growth so that the data generated can be contextually evaluated.

Texas Tech University, Rakib H. Rashel, May 2020

Literature Cited

Acosta, F., Zamor, R.M., Najar, F.Z., Roe, B.A., Hambright, K.D., 2015. Dynamics of an experimental microbial invasion. Proc. Natl. Acad. Sci. USA. 112, 11594–11599.

Anderson, D.M., 2009. Approaches to monitoring, control and management of harmful algal blooms (HABs). Ocean Coast. Manage. 52, 342–347.

Baker, J.W., Grover, J.P., Brooks, B.W., Ureña-Boeck, F., Roelke, D.L., Errera, R., Kiesling, R.L., 2007. Growth and toxicity of Prymnesium Parvum (Haptophyta) as a function of salinity, light, and temperature. J. Phycol. 43, 219–227.

Baker, J.W., Grover, J.P., Ramachandrannair, R., Black, C., Valenti, T.W., Brooks, B.W., Roelke, D.L., 2009. Growth at the edge of the niche: an experimental study of the harmful alga Prymnesium parvum. Limnol. Oceanogr. 54, 1679–1687.

Croghan, C.W., Egeghy, P.P., 2003. Methods of dealing with values below the limit of detection using SAS. U.S. Environmental Protection Agency, Washington, D.C. . (Last accessed, 23.8.2014).

Gani, P., Sunar, N.M., Matias-Peralta, H., Abdul Latiff, A.A., Abdul Razak, A.R., 2016. Influence of initial cell concentrations on the growth and biomass productivity of microalgae in domestic wastewater. Appl. Ecol. Environ. Res. 14, 399–409.

Granéli, E., Edvardsen, B., Roelke, D.L., Hagström, J.A., 2012. The ecophysiology and bloom dynamics of Prymnesium spp. Harmful Algae 14, 260–270.

Gou, W., 1983. The reason for the fish death at aquacultural experimental station at Ningxia and the distribution of Prymnesium parvum in Ningxia. J. Dalian Fish. Coll. 1, 43–48 (in Chinese).

Guillard, R.R.L., 1973. Division rates. In: Stein, J.R. (Ed.), Handbook of Phycological Methods: Culture methods and growth measurements. Cambridge University Press, Cambridge, pp. 289–311.

Texas Tech University, Rakib H. Rashel, May 2020

Guillard, R.R.L., 1975. Culture of phytoplankton for feeding marine invertebrates. In: Smith, W.L., Chanley, M.H. (Eds.), Culture of marine invertebrate animals. Plenum Press, New York, pp. 29–60.

Guo, M., Harrison, P.J., Taylor, F.J.R., 1996. Fish kills related to Prymnesium parvum N. Carter (Haptophyta) in the People’s Republic of China. J. of Appl. Phycol. 8, 111–117.

HACH, 2013. DR/890 Colorimeter Procedures Manual. HACH Company, 10th edition, P/N 48470-22. Available at: www.hach.com/asset- get.download.jsa?id=7639982259. Accessed August 15, 2016.

Hallegraeff, G.M., 1993. A review of harmful algal blooms and their apparent global increase. Phycologia 32, 79–99.

Hambright, K.D., Zamor, R.M., Easton, J.D., Glenn, K.L., Remmel, E.J., Easton, A.C., 2010. Temporal and spatial variability of an invasive toxigenic protist in a North American subtropical reservoir. Harmful Algae 9, 568–577.

Hambright, K.D., Easton, J.D., Zamor, R.M., Beyer, J., Easton, A.C., Allison, B., 2014. Regulation of growth and toxicity of a mixotrophic microbe: implications for understanding range expansion in Prymnesium parvum. Freshw. Sci. 33, 745–754.

Heidari, M., Kariminia, H.-R., Shayegan, J., 2016. Effect of culture age and initial inoculum size on lipid accumulation and productivity in a hybrid cultivation system of Chlorella vulgaris. Process Saf. Environ. 104, 111–122.

Israël, N.M.D., VanLandeghem, M.M., Denny, S., Ingle, J., Patiño, R., 2014. Golden alga presence and abundance are inversely related to salinity in a high-salinity river ecosystem, Pecos River, USA. Harmful Algae, 39, 81–91.

James, T.L., De La Cruz A., 1989. Prymnesium parvum Carter (Chrysophyceae) as a suspect of mass mortalities of fish and shellfish communities in western Texas. Texas J. Sci. 41, 429–430.

Larsen, A., Bryant, S., 1998. Growth rate and toxicity of Prymnesium parvum and Prymnesium patelliferum (Haptophyta) in response to changes in salinity, light and temperature. Sarsia 83, 409–418.

Lu, S., Wang, J., Niu, Y., Yang, J., Zhou, J., Yuan, Y., 2012. Metabolic profiling reveals growth related FAME productivity and quality of Chlorella sorokiniana with different inoculum sizes. Biotechnol. Bioeng. 109, 1651–1662.

Texas Tech University, Rakib H. Rashel, May 2020

Lu, S., Wang, J., Ma, Q., Yang, J., Li, X., Yuan, Y., 2013. Phospholipid metabolism in an industry microalga Chlorella sorokiniana: the impact of inoculum sizes. PLoS ONE 8, e70827.

Lutz-Carrillo, D.J., Southard, G.M., Fries, L.T., 2010. Global genetic relationships among isolates of golden alga (Prymnesium parvum). J. Am. Water Resour. Assoc. 46, 24–32.

Manning, S.R., La Claire J.W., II, 2010. Prymnesins: toxic metabolites of the Golden Alga, Prymnesium parvum Carter (Haptophyta). Mar. Drugs 8, 678–704.

Nicholls, K.H., 2003. Haptophyte algae. In: Wehr, D., Smith, R.G. (Eds.), Freshwater Algae of North America. Academic Press, Boston, pp. 511–521.

Olli, K., Trunov, K., 2007. Self-toxicity of Prymnesium parvum (Prymnesiophyceae). Phycologia 46, 109–112.

Padilla, G.M., 1970. Growth and toxigenesis of the chrysomonad Prymnesium parvum as a function of salinity. J. Protozool. 17, 456–462.

Patiño, R., Dawson, D., VanLandeghem, M.M., 2014. Retrospective analysis of associations between water quality and toxic blooms of golden alga (Prymnesium parvum) in Texas reservoirs: Implications for understanding dispersal mechanisms and impacts of climate change. Harmful Algae 33, 1–11.

Pratt, R., 1940. Influence of the size of the inoculum on the growth of Chlorella vulgaris in freshly prepared culture medium. Am. J. Bot. 27, 52–56.

Rhodes, K., Hubbs, C., 1992. Recovery of the Pecos River from a red tide fish kill. Southwest. Nat., 37, 178–187.

Roelke, D.L., Errera, R.M., Kiesling, R., Brooks, B.W., Grover, J.P., Schwierzke, L., Pinckney, J.L., 2007. Effects of nutrient enrichment on Prymnesium parvum population dynamics and toxicity: results from field experiments, Lake Possum Kingdom, USA. Aquat. Micro. Ecol. 46, 125–140.

Roelke, D.L., Grover, J.P., Brooks, B.W., Glass, J., Buzan, D., Southard, G.M., Nelson, J., 2011. A decade of fish-killing Prymnesium parvum blooms in Texas: roles of inflow and salinity. J. Plankton Res. 33, 243–253.

Roelke, D.L., Brooks, B.W., Grover, J.P., Gable, G.M., Schwierzke-Wade, L., Hewitt, N.C., 2012. Anticipated human population and climate change effects on algal blooms of a toxic haptophyte in the south-central USA. Can. J. Fish. Aquat. Sci. 69, 1389–1404.

Texas Tech University, Rakib H. Rashel, May 2020

Sellner, K.G., Doucette, G.J., Kirkpatrick, G.J., 2003. Harmful algal blooms: causes, impacts and detection. J. Ind. Microb. Biot. 30, 383–406.

Southard, G.M., 2005. Appendix A: Identification and enumeration of Prymnesium parvum cells, Version AEW-IDE 1.1. In: Barkoh, A., Fries, L.T. (Eds.), Management of Prymnesium parvum at Texas State Fish Hatcheries. Texas Parks and Wildlife Department, Management Data Series 236, PWD RP T3200-1138 (1/06), Austin, Texas, pp. 97-98.

VanLandeghem, M.M., Farooqi, M., Southard, G.M., Patiño, R., 2015a. Associations between water physicochemistry and Prymnesium parvum presence, abundance, and toxicity in west Texas reservoirs. J. Am. Water Resour. Assoc. 51, 471–486.

VanLandeghem, M.M., Farooqi, M., Southard, G.M., Patiño, R., 2015b. Spatiotemporal associations of reservoir nutrient characteristics and the invasive, harmful alga Prymnesium parvum in west Texas. J. Am. Water Resour. Assoc. 51, 487–501.

VanLandeghem, M.M., Denny, S., Patiño, R., 2015c. Predicting the risk of toxic blooms of golden alga from cell abundance and environmental covariates. Limnol. Oceanogr.: Methods 13, 568–586.

Wood, A.M., Everroad, R.C., Wingard, L.M., 2005. Measuring growth rates in microalgal cultures. In: Andersen, R.A (Ed.), Algal Culturing Techniques. Elsevier Academic Press, 598 pp.

Zar, J.H., 2010. Biostatistical Analysis, 5th edition. Pearson Prentice Hall, New Jersey, USA, 944 pp.

Texas Tech University, Rakib H. Rashel, May 2020

Tables

Table 2.1. Pearson partial correlation of maximum cell density with either exponential growth rate (r) or early cell density (early-D) in Prymnesium parvum as a function of genetic strain and temperature over a range of salinities. Partial Correlation with Maximum Cell Density Salinity Temperature Correlated Control Strain range (°C) Variable variable Pearson r p* n r Early-D 0.92 <0.0001 36 UTEX-2797 5-30 22 Early-D r 0.81 <0.0001 36 r Early-D 0.82 <0.0001 30 UTEX-995 5-25 22 Early-D r 0.48 0.009 30 r Early-D 0.75 <0.0001 30 UTEX-2797 5-25 13 Early-D r 0.59 0.0007 30 *Bonferroni-corrected critical α = 0.008. Cultures were inoculated with 100 cells ml-1, and early cell density was determined on day 3 (UTEX-2797, 22 °C) or 9 (UTEX- 2797, 13 °C; UTEX-995, 22 °C). When r was the correlation variable, early cell density was used as control variable, and vice versa. Data used for these analyses are those reported in Figure 4. Exponential growth rate could not be estimated at a salinity of 30 psu for UTEX-2797 at 13 °C and for UTEX-995. Significant correlations and their p-values are bolded. n, sample size.

Texas Tech University, Rakib H. Rashel, May 2020

Figures

Figure 2.1. Growth curves of Prymnesium parvum as a function of inoculum size (100– 100,000 cells ml-1) and genetic strain (UTEX-2797 and UTEX-995) in standard base medium of salinity 5 psu at 22 °C. Each time point represents the mean (± SEM) of 3 replicates.

Texas Tech University, Rakib H. Rashel, May 2020

Figure 2.2. Exponential growth rate (r) and maximum cell density of Prymnesium parvum as a function of inoculum size and genetic strain in a standard base medium of salinity 5 psu at 22 °C. A-B, UTEX-2797; C-D, UTEX-995. Bars with the same letter codes do not differ significantly (Tukey’s HSD, p ˂ 0.05). Each bar represents the mean (+ SEM) of 3 replicates.

Texas Tech University, Rakib H. Rashel, May 2020

Figure 2.3. Growth curves of Prymnesium parvum as a function of salinity, genetic strain, and temperature in cultures inoculated with 100 cells ml-1. A, UTEX-2797 at 22 °C; B, UTEX-995 at 22 °C; C, UTEX-2797 at 13 °C. These plots show the results of 1 of 2 trials conducted. Each bar represents the mean (± SEM) of 3 replicates.

Texas Tech University, Rakib H. Rashel, May 2020

Figure 2.4. Exponential growth rate (r) and maximum cell density of Prymnesium parvum as a function of salinity, genetic strain, and temperature in cultures inoculated with 100 cells ml-1. A-B, UTEX-2797 at 22 °C; C-D, UTEX-995 at 22 °C; E-F, UTEX- 2797 at 13 °C. Bars with the same letter codes do not differ significantly (Tukey’s HSD, p ˂ 0.05). Each bar represents the mean (± SEM) of 6 replicates. Results of multiple comparisons shown in B were conducted separately for each trial, and statistical significances are separately shown for each in large and small case letters, respectively – for more detailed trial results see Supplemental Figure 1. NC, not calculated.

Texas Tech University, Rakib H. Rashel, May 2020

Supplemental Table 1. Simple correlations among maximum cell density (max-D), exponential growth rate (r), and early cell density (early-D) as a function of genetic strain and temperature in Prymnesium parvumcultures over a range of salinities. Cultures were inoculated with 100 cells ml-1. See Figure 4 and Supplemental Table 2 for additional details about the data used. Significant correlations are bolded; n, sample size. Simple Correlation Coefficients (Pearson r)* Temperature Strain Salinity range (°C) max-D,r max-D,early-D early-D,r n UTEX-2797 5-30 22 0.75 0.30 -0.28 36 UTEX-995 5-25 22 0.78 0.29 -0.01 30 UTEX-2797 5-25 13 0.80 0.69 0.47 30 *Bolded coefficients are significant at Bonferroni-corrected critical α = 0.006

Note: for UTEX-2797 at 22 °C, the partial correlation coefficients for max-D and r, and for max-D and early-D were considerably larger (see Table 1) than their respective simple correlation coefficients shown in this table. This is because the simple correlation between r and early cell density is negative (even if non-significantly) and when each of these variables serves as control variable for their partial correlation with max-D, their negative influence is removed yielding larger partial coefficients and sometimes also switching from non-significant simple correlation to significant partial correlations (e.g., max-D and early-D). Control variables that exert a negative influence are sometimes termed "suppressor" variables.

Texas Tech University, Rakib H. Rashel, May 2020

Supplemental Table 2. Early cell density (mean count and standard error of the mean, n = 6) as a function of genetic strain and temperature in Prymnesium parvumcultures over a range of salinities . Cultures were inoculated with 100 cells ml-1. Counts were taken at day 3 for UTEX-2797 at 22 °C, day 9 for UTEX-2797 at 13 °C, and day 9 for UTEX-995 at 22 °C. For the latter two strain-temperature combinations, counts were highly erratic at 30 psu at (and after) day 9 and could not be reliably determined. Early cell density (cells ml-1) in cultures grown at different salinity* 5 10 15 20 25 30 Strain Temperature (°C) Count day Mean SEM Mean SEM Mean SEM Mean SEM Mean SEM Mean SEM 22 3 834 75 1056 134 1278 102 1056 159 834 188 834 206 UTEX-2797 13 9 333 0 445 70 556 70 256 49 178 49 – – UTEX-995 22 9 336 84 503 141 556 165 392 56 227 68 – – -1 *Undetectable values were replaced with √(LOD/2), or 13 cells ml

Texas Tech University, Rakib H. Rashel, May 2020

Supplemental Figure 2.1. Maximum cell density of Prymnesium parvum (UTEX-2797) as a function of salinity at a temperature of 22 °C following inoculation with 100 cells ml-1. Each panel represents the results of two separate trials and each bar the mean (+ SEM) of 3 replicates. Bars with the same letter codes do not differ significantly (Tukey’s HSD, p ˂ 0.05).

Texas Tech University, Rakib H. Rashel, May 2020

Supplemental Figure 2.2. Exponential growth rate (r) and maximum cell density as a function of salinity in Prymnesium parvum (UTEX-2797) cultures inoculated with 100,000 cells ml-1 and maintained at 22 °C. Results of Student’s t-tests (two-tailed) indicated that r and maximum cell density were both significantly suppressed at a salinity of 30 psu relative to 5 psu (n =3 per treatment). For comparison purposes, relevant data from Figure 4 is shown on the right side of the panels – culture conditions were the same except inoculum size was 100 cells ml-1, and differences in parameter values between salinity of 5 and 30 psu were also significant (see Fig. 4).

Texas Tech University, Rakib H. Rashel, May 2020

CHAPTER Ⅲ

GROWTH RESPONSE OF THE ICHTHYOTOXIC HAPTOPHYTE, PRYMNESIUM PARVUM CARTER, TO CHANGES IN SULFATE AND FLUORIDE CONCENTRATIONS

Abstract

Golden alga Prymnesium parvum Carter is a euryhaline, ichthyotoxic haptophyte

2- (Chromista). Because of its presumed coastal/marine origin where SO4 levels are high,

2- the relatively high SO4 concentration of its brackish inland habitats, and the sensitivity of marine chromists to sulfur deficiency, this study examined whether golden alga growth

2- is sensitive to SO4 concentration. Fluoride is a ubiquitous ion that has been reported at higher levels in golden alga habitat; thus, the influence of F- on growth also was

2- examined. In low-salinity (5 psu) artificial seawater medium, overall growth was SO4 -

-1 dependent up to 1000 mg l using MgSO4 or Na2SO4 as source; the influence on growth rate, however, was more evident with MgSO4. Transfer from 5 to 30 psu inhibited growth

2- when salinity was raised with NaCl but in the presence of seawater levels of SO4 , these effects were fully reversed with MgSO4 as source and only partially reversed with

Na2SO4. Growth inhibition was not observed after acute transfer to 30 psu in a commercial sea salt mixture. In 5-psu medium, F- inhibited growth at all concentrations

Texas Tech University, Rakib H. Rashel, May 2020

*This chapter published as: Rashel, R.H., Patiño, R., 2019. Growth response of the ichthyotoxic haptophyte, Prymnesium parvum Carter, to changes in sulfate and fluoride concentrations. PLoS ONE 14 (9), e0223266.

1. Introduction

Known in North America as golden alga, Prymnesium parvum Carter is a euryhaline, ichthyotoxic haptophyte believed to be of marine or coastal origin but with the ability to grow in estuaries, brackish embayments (Nicholls, 2003), and inland brackish waters (Edvardsen and Imai, 2006; Roelke et al., 2016). Curiously, however, toxic blooms of this species normally occur within the relatively narrow salinity range of

0.5 to 12 psu (Edvardsen and Imai, 2006; Israël et al., 2014; Roelke et al., 2016) and abundance is negatively associated with salinity at levels exceeding ~8–12 psu (Gou,

1983 (cited in Guo et al. 1996); Israël et al., 2014). A recent experimental study confirmed the biphasic response of an inland strain of golden alga (UTEX LB 2797) to salinity and reported that the highest growth potential is observed at ~10–15 psu (Rashel and Patiño, 2017). While multiple abiotic and biotic factors may interact to influence golden alga growth in nature (Hambright et al., 2010; Granéli et al., 2012; Roelke et al.,

2012; Israël et al., 2014; Patiño et al., 2014; VanLandeghem et al., 2015a,b), the

Texas Tech University, Rakib H. Rashel, May 2020

relatively low salinity range for bloom development seems counterintuitive given its

highly euryhaline potential.

The chemical composition of inland surface waters is driven by a complex

interaction among the geochemistry of watersheds and aquifers, topography, climate

(Dillon and Kirchner, 1975), and anthropogenic changes in land cover (Foley et al. 2005;

Kaushal et al. 2005; VanLandeghem et al., 2012). Little is known about the influence of

specific ions on golden alga growth and the role they may play in determining the spatial

distribution of this species in inland waters. A retrospective study of reservoir water

quality and golden alga in the southcentral USA found that spatial variability in sulfate

concentration is associated with the intra- and inter-basin distribution of toxic blooms

(Patiño et al., 2014). Levels of sulfate in reservoirs of the southcentral USA with a history

of blooms (average, ~600 mg l-1) were >8-fold higher than in naive reservoirs (~70 mg l-

1) and were nearly equal those of chloride (~800 mg l-1) (Patiño et al., 2014). Blooms of

this species in association with high sulfate levels also have been recorded in a few other

instances, e.g., in a freshwater pond in Germany (Moestrup, 1994). In addition, some

studies have reported relatively high levels of fluoride in bloom-impacted (>1 mg l-1)

compared to non-impacted (<1 mg l-1) water bodies of the southcentral USA

(VanLandeghem et al., 2012, 2015a). These findings led to the hypotheses that sulfate

(Patiño et al., 2014) and fluoride (VanLandeghem et al., 2015a) may influence the

distribution and growth of golden alga at least in part via mechanisms independent of

salinity.

Sulfate, the primary source of sulfur for algae and plants (Takahashi et al., 2011;

Giordano and Raven, 2014), is the second most abundant anion in aquatic habitats after

Texas Tech University, Rakib H. Rashel, May 2020

chloride. Its concentration ranges from ~10 to 1300 mg l-1 in inland waters (Holmer and

Storkholm, 2001; VanLandeghem et al., 2015a; Patiño et al., 2014) and 2800 to 3000 mg

l-1 in seawater (Prioretti and Giordano, 2016). Eukaryotic phytoplankton communities in

inland waters are typically dominated by green algae (Takahashi et al., 2011) but in

marine habitats they consist primarily of chromist (chlorophyll a+c) algae, which include the Haptophyta (Falkowski et al., 2004; Takahashi et al., 2011; Cavalier-Smith, 2018).

Growth of certain green and chromist algae can be reduced under conditions of sulfur

deficiency (Yildiz et al., 1994; Giordano et al., 2005; Ratti et al., 2011; Bochenek et al.,

2013).

Fluoride is another ubiquitous anion in freshwater with concentration typically

ranging from 0.01 to 0.30 mg l-1, and in seawater, its concentration ranges from 1.2 to 1.5

mg l-1 (Dobbs, 1974; Camargo, 1996, 2003). Considerably higher concentrations have

been recorded in inland waters impacted by geothermal or volcanic activity (Camargo,

2003). Fluoride can either inhibit, enhance or not affect algal growth depending upon the

species and exposure concentrations (Camargo, 2003). Given the wide variation in

responses to fluoride among algal species and the relatively high levels of this ion in

golden alga-impacted habitats, VanLandeghem et al. (2015a) hypothesized that tolerance

to fluoride relative to other phytoplankton may confer a growth advantage to golden alga.

The specific objective of this study is to test the hypotheses put forward by earlier

field studies that sulfate (Patiño et al., 2014) and fluoride (VanLandeghem et al., 2015a)

positively influence golden alga growth independently of salinity. While the earlier

studies used a multivariate approach to evaluate general associations between

environmental variables and golden alga, cause-effect associations could not be

Texas Tech University, Rakib H. Rashel, May 2020 addressed because of the complexity of the natural environment and the strong collinearity among several of the variables (e.g., between sulfate and salinity (Patiño et al., 2014)). The present study used an experimental approach where the variables of interest (sulfate and fluoride) were manipulated while maintaining salinity and other ambient conditions constant. To our knowledge, salinity-independent effects of sulfate on growth have not been examined before in inland algae and this study is the first to evaluate salinity-independent effects of sulfate and fluoride on growth of golden alga.

2. Materials and methods

2.1. Basic culture procedures

The strain of P. parvum used in this study, UTEX LB 2797, is the most widespread strain found in US inland waters (Lutz-Carrillo et al., 2010) and was obtained from the UTEX Culture Collection of Algae (The University of Texas at Austin, Texas,

USA). Stock cultures were grown in UTEX Artificial Seawater Medium (ASM; https://utex.org/products/artificial-seawater-medium) modified (diluted) to a salinity of 5 psu and with pH of ~ 8.1. Modified ASM, also referred to as base medium in this study, was enriched with F/2 levels of nutrients and vitamins, and ferrous ammonium sulfate in the original trace metal recipe was replaced with an equimolar amount of ferric chloride

(see Rashel and Patiño, 2017 for additional details of medium preparation). Cultures were maintained non-axenically in 250-ml Erlenmeyer flasks filled with 100 ml of modified

ASM in an incubator (I36LLVLC9; Percival Scientific Inc. Perry, IA, USA).

Temperature and photoperiod were set at 22 °C and 12:12 h light:dark, and light intensity

Texas Tech University, Rakib H. Rashel, May 2020

was ~6500 lux. Cultures were gently swirled once daily. Late stationary phase cells were

used as inocula to maintain stock cultures and as seed for experimental cultures.

2.2. Experimental design

2.2.1. Effects of changes in sulfate concentration at low salinity

Earlier field studies reported that inland surface waters with a history of golden alga blooms have relatively high concentrations of sulfate (Patiño et al., 2014;

VanLandeghem et al., 2015a). The present experiments were designed to determine whether this association is causal and independent of salinity. Basic culture procedures were the same as for stock cultures. The salinity of modified ASM is 5 psu and its sulfate concentration is 250 mg l-1 (Table 1), values which are well within the range observed in

golden alga habitat of the Southern Great Plains and the southwestern USA (Patiño et al.,

2014; Israël et al., 2014; VanLandeghem et al., 2015a).

In one series of experiments, sulfate concentration was manipulated by replacing

MgSO4∙7H2O (original recipe) with MgCl2∙6H2O (Sigma M9272) and/or adding different

-1 amounts of MgSO4∙7H2O to achieve concentrations of 0, 50, 250, and 1000 mg l while maintaining salinity constant at 5 psu. Salinity of experimental media was confirmed by measurement with a pre-calibrated YSI 85 multiparameter probe (Yellow Springs

Incorporated, Yellow Springs, OH, USA). Trace metal compounds containing sulfate in the original recipe were also replaced with alternatives without sulfate; namely,

ZnSO4∙7H2O, MnSO4∙H2O, and CoSO4∙7H2O were replaced with equimolar amounts of

Texas Tech University, Rakib H. Rashel, May 2020

ZnCl2 (Sigma Z0152), MnCl2∙4H2O (Sigma 221279), and CoCl2∙6H2O (Sigma 255599), respectively.

Magnesium is a key component of chlorophyll and earlier studies have shown that changes in its concentration can affect algal growth (Finkle and Appleman, 1953). In a separate experiment, the ability of Mg-free sulfate (as Na2SO4) to influence golden alga

growth was examined. Media with sulfate concentrations of 0, 50, 250, and 1000 mg l-1

were prepared by replacing MgSO4∙7H2O (in original base medium) with, and adding

different amounts of, Na2SO4 (Fisher S421-500) while keeping salinity constant (5 psu).

As previously noted, trace metal compounds in the original recipe that contain sulfate

were replaced with alternatives without sulfate. Magnesium concentration in this

experiment was kept constant at the base medium value of 60 mg l-1 (Table 1) by using

MgCl2∙6H2O.

Initial cell density (inoculum size) in the culture flasks was 100 cells ml-1 and each treatment concentration was conducted in triplicate. Two complete, independent trials were conducted for the experiment with MgSO4 and one trial was conducted with

Na2SO4. Cell density in each flask was determined every 3 days until batch cultures

reached late stationary phase (see Analytical procedures section).

2.2.2. Effects of salinity under different major salt scenarios

An earlier study showed that abruptly increasing the salinity of modified ASM

from 5 to 30 psu by adding NaCl strongly suppresses golden alga growth (Rashel and

Patiño, 2017). The present experiment was designed to determine if growth suppression

at high salinity also occurs under a relatively complex salt scenario similar to that of

Texas Tech University, Rakib H. Rashel, May 2020

seawater (Table 1). Modified ASM (see preceding section) and Instant Ocean® sea salts

(IO) were used to prepare respective experimental media at salinities of 5 and 30.

Modified ASM has a salinity of 5 psu, and salinity of 30 psu in this medium was achieved by simply adding NaCl. In IO, salinities of 5 and 30 psu were prepared by direct addition of the appropriate amounts of commercial salt preparation to deionized water.

Salinity of experimental media was confirmed by measurement with a multiparameter

probe. Nominal concentrations of major ions in culture media for this experiment and in

seawater are shown in Table 1 [the elemental composition of IO and seawater is based on

Atkinson and Bingman (1998)].

Culture media for these experiments were filter-sterilized (Nalgene, 0.45 μm,

sterile analytical filter unit, Thermo Fisher Scientific Inc., Waltham, MA, USA) after

addition of nutrients and trace metals. Filter-sterilization was used instead of autoclaving

because the latter procedure caused salts in IO media to precipitate. Other media and

culture procedures were as described earlier. All treatments were conducted in triplicate.

2.2.3. Effects of differences in sulfate concentration at high salinity

Results of experiments under the preceding sections led to the ad hoc question,

can the inhibitory effect of NaCl-dependent high salinity on golden alga growth at 30 psu

(Rashel and Patiño, 2017) be reduced or eliminated by a sulfate concentration corresponding to that of seawater at 30 psu? Two trials were conducted in this experiment. The following media were prepared for trial 1: (1) modified ASM at salinity of 5 psu, (2) modified ASM with additional NaCl to raise the salinity to 30 psu, and (3)

-1 modified ASM with 2.4 g l sulfate (as MgSO4) and the appropriate amount of NaCl to

Texas Tech University, Rakib H. Rashel, May 2020 achieve a salinity of 30 psu. Media for the second trial were prepared in the same manner

-1 except that the source of sulfate to achieve the concentration of 2.4 g l was Na2SO4. The concentration of sulfate in the third treatment of each trial corresponds to its concentration in seawater adjusted to a salinity of 30 psu (Table 1). Media, culture and cell enumeration procedures were as described previously except that treatments for the first trial were conducted in quadruplicate.

2.2.4. Effects of changes in fluoride concentration at low salinity

An earlier field study reported that surface waters with a history of golden alga blooms have relatively high concentrations of fluoride (VanLandeghem et al., 2015a).

This experiment was designed to determine whether this association is causal and independent of salinity. The formula for modified ASM does not include fluoride. Four nominal fluoride concentrations were prepared for this experiment by adding NaF (Sigma

S6776) to base medium: 0, 2.25, 11.30, and 56.50 mg l-1. Cell culture and enumeration procedures were as described earlier and all treatments within a trial were conducted in triplicate. Two independent, complete trials were conducted.

2.3. Analytical procedures

2.3.1. Cell counts

Cell density was determined as described by Rashel and Patiño (2017). Briefly, aliquots of 500 µl of experimental cultures were taken from each replicate flask at 3-day intervals and used to determine abundance using a hemocytometer under a compound

Texas Tech University, Rakib H. Rashel, May 2020

microscope. To maintain the total cell counts below 50,000 cells ml-1, sub-samples were

diluted with fresh medium when needed. A total of three counts per replicate flask were

taken and the average cell number was reported for each replicate.

2.3.2. Estimation of growth parameters and statistical analyses

Growth parameters estimated for all experiments were exponential growth rate (r,

day-1) and maximum cell density (cells ml-1). In addition, early cell density (cells ml-1)

was estimated for the sulfate experiments at low salinity (section Effects of changes in sulfate concentration at low salinity) to assess correlations between growth indices with the full range of sulfate concentrations tested as well as global relationships among growth indices. Maximum cell density was the highest cell count achieved by each replicate. Early cell density is an index of cell status and growth during the transition between the lag and exponential phases and, under certain conditions, can influence maximum growth potential even in the absence of changes in exponential growth rate

(Rashel and Patiño, 2017).

Growth rate for golden alga was calculated using the following equation (Wood et al., 2005):

ln N ln N = t t 2 − 1 𝑟𝑟 2 1 −

Texas Tech University, Rakib H. Rashel, May 2020

where N1 and N2 are cell densities at times t1 and t2 (t2 > t1). For each individual replicate,

times were chosen so that they bracket the linear portion of the ln-transformed growth

curve. Mean values are reported for all growth parameters.

Most data were analyzed by one-way ANOVA followed by pairwise comparisons

using Tukey’s Honest Significant Difference (Tukey’s HSD) test. Two-way ANOVA

was used in experiments where two full trials were conducted [sulfate (as MgSO4) and

fluoride]; if no trial or interaction effects were observed, values from both trials were pooled and mean separations assessed by Tukey’s HSD. Spearman correlation analysis was used to assess the association between growth indices and sulfate concentration, and partial correlation analysis was used to determine associations between maximum cell

density (i.e., growth potential) and the other two growth indices (r and early cell density)

over the full range of sulfate concentrations. Family-wise error rate was controlled by

adjusting p-values according to the Holm-Šídák method (Holm, 1979). Analysis of

variance and mean separations were conducted with SPSS 17.0 (SPSS Inc., Chicago,

USA), correlation analyses with Statistica version 13.3 (Tibco Software, Inc., Palo Alto,

CA, USA), and Holm-Šídák p-value adjustments and graphics with GraphPad Prism 6

(GraphPad Software, Inc., La Jolla, CA, USA).

3. Results

3.1. Effects of changes in sulfate concentration at low salinity

3.1.1. Sulfate source: MgSO4

Texas Tech University, Rakib H. Rashel, May 2020

Stationary growth phase was generally achieved on day 21 at all sulfate

concentrations (Fig. 1A). On a semi-ln plot, the linear portion of the exponential growth phase generally bracketed 3–12 days. Results of 2-way ANOVA showed no differences between trials for r, maximum cell density or early cell density (Table 2). Sulfate

concentration affected r and maximum cell density, but not early cell density (Table 2).

No interaction effects were observed between sulfate concentrations and trial for r,

maximum cell density or early cell density (Table 2). Replicates from both trials were

therefore pooled for mean separations. As sulfate concentration increased, r (Fig. 1C;

Tukey's HSD, p < 0.05) and maximum cell density (Fig. 1D; Tukey's HSD, p < 0.05)

both increased. Early cell density was 1.34 times higher at 1000 mg l-1 compared to 0 mg

l-1 but ANOVA yielded no significant differences (Fig. 1B). Compared to the lowest

sulfate concentration, r and maximum cell density at a nominal sulfate concentration of

1000 mg l-1 were 1.21 and 1.38 times higher, respectively. Results of Spearman

correlation analysis showed that maximum cell density and r, but not early cell density,

were correlated with sulfate concentration (Table 3). Result of partial correlation analysis

of all data combined indicated that maximum cell density was associated with r but not

early cell density (Table 4).

3.1.2. Sulfate source: Na2SO4

2+ When Na2SO4 was used as source of sulfate while keeping Mg concentration constant, stationary growth phase was generally achieved on day 21–24 at all sulfate concentrations (Fig. 2A). On a semi-ln plot, the linear portion of the exponential growth phase generally bracketed 3–12 days. Results of one-way ANOVA showed sulfate

Texas Tech University, Rakib H. Rashel, May 2020

concentration had no significant effect on r or early cell density but had a significant

effect on maximum cell density (Table 2). As sulfate concentration increased, maximum

cell density increased (Fig. 2D; Tukey's HSD, p < 0.05). Compared to the lowest sulfate

concentration (nominal 0 mg l-1), maximum cell density in cultures grown at a nominal

sulfate concentration of 1000 mg l-1 was 1.61 times higher. Results of Spearman

correlation analysis showed that maximum and early cell density, but not r, were

correlated with sulfate concentration (Table 3). Results of partial correlation analysis,

however, indicated that maximum cell density was significantly associated with both, r and early cell density (Table 4).

3.2. Effects of salinity under different major salt scenarios

Stationary growth phase was generally achieved on day 21 under all treatment

conditions (Fig. 3A). On a semi-ln plot, the linear portion of the exponential growth phase generally bracketed 3–12 days. Results of one-way ANOVA showed that treatment had significant effects on r and maximum cell density (Table 2). When cells were grown at a salinity of 30 psu in modified ASM (salinity adjusted by adding NaCl), r and

maximum cell density were generally lower than in all other treatments (Fig. 3; Tukey’s

HSD, p < 0.05); more specifically, their average values were reduced to 85 and 38

percent of values recorded for r and maximum cell density at 5 psu in modified ASM.

Growth parameters at 5 or 30 in IO did not differ between themselves or from cells

grown at 5 psu in modified ASM (Fig. 3B, 3C; Tukey’s HSD, p < 0.05).

Texas Tech University, Rakib H. Rashel, May 2020

3.3. Effects of differences in sulfate concentration at high salinity

One notable result of the preceding experiment was that unlike cells cultured in

modified ASM at 30 psu, cells cultured in IO at 30 psu did not exhibit growth

suppression relative to cells cultured in modified ASM at 5 psu. The primary differences

in major ion composition between modified ASM and IO media (at 30 psu) are the higher

concentrations in IO media of hardness cations (Ca2+ and Mg2+), sulfate, and potassium

(Table 1). The present experiments evaluated the ability of sulfate to influence growth of

golden alga when cells are transferred from 5 to 30 psu (in modified ASM) under

-1 -1 relatively high (0.61 g l ; sulfate source, MgSO4) or low (0.06 g l ; sulfate source,

2+ Na2SO4) levels of Mg .

3.3.1. Sulfate source: MgSO4

Stationary growth phase was generally achieved on day 24 for all treatments (Fig.

4A). On a semi-ln plot, the linear portion of the exponential growth phase generally bracketed 3–12 days. Results of one-way ANOVA showed that treatment conditions had significant effects on r and maximum cell density (Table 2). When cells were grown in modified ASM at 30 without the additional sulfate, r and maximum cell density were markedly reduced compared to the other treatments (Fig. 4B, 4C; Tukey’s HSD, p <

0.05); more specifically, r and maximum cell density in these cultures were reduced to 71 and 35 percent of the values observed in modified ASM at 5 psu. The addition of sulfate

-1 (2.4 g l ) as MgSO4, however, completely nullified the inhibitory effect of high salinity

on golden alga growth (Fig. 4B, 4C).

Texas Tech University, Rakib H. Rashel, May 2020

3.3.2. Sulfate source: Na2SO4

Stationary growth phase was generally achieved on day 21–24 under all treatment

conditions (Fig. 5A). On a semi-ln plot, the linear portion of the exponential growth phase generally bracketed 3–12 days. Results of one-way ANOVA showed treatment conditions had a significant effect on r and maximum cell density (Table 2). When cells

were grown in modified ASM at 30 without additional sulfate, r (Fig. 5B) and maximum

cell density (Fig. 5C) were significantly reduced compared to control at 5 psu. The

-1 addition of a seawater level of sulfate (2.4 g l ) as Na2SO4 only partially restored growth potential (Tukey's HSD, p < 0.05).

3.4. Effects of changes in fluoride concentration at low salinity

Stationary growth phase was generally achieved on day 21 at all fluoride concentrations (Fig. 6A). On a semi-ln plot, the linear portion of the exponential growth phase generally bracketed 3–12 days. Results of 2-way ANOVA showed no differences between trials for r or maximum cell density (Table 2). Fluoride concentration affected r

and maximum cell density (Table 2). No interaction effects were observed between

fluoride and trial for r and maximum cell density (Table 2). Replicates from both trials

were therefore pooled for mean separations. Exponential growth rate decreased as

fluoride concentration increased up to 11.30 mg l-1, and then increased at the highest

concentration tested (relative to growth at 11.30 mg l-1) although it was still lower than in

the control treatment (Fig. 6B; Tukey's HSD, p < 0.05). A very similar pattern was

Texas Tech University, Rakib H. Rashel, May 2020 observed for maximum cell density; namely, maximum cell density decreased as fluoride concentration increased from 0 to 11.30 mg l-1 and then increased at the highest concentration while still remaining below values observed in the control (Fig. 6C;

Tukey's HSD, p < 0.05).

4. Discussion

It is well established that growth of golden alga is influenced by salinity under either natural or laboratory conditions (Padilla, 1970; Larsen and Bryant, 1998; Baker et al., 2007; Hambright et al., 2010, 2014; Israël et al., 2014; Patiño et al., 2014;

VanLandeghem et al, 2015a; Roelke et al., 2016; Rashel and Patiño, 2017). Inland field studies of golden alga that included high-salinity sites in their sampling designs, however, reported biphasic associations between abundance and salinity with peak abundance at

~8–12 psu [Gou (1983), cited in Guo et al, (1996); Israël et al., 2014]. A recent laboratory study where salinity was manipulated by adding NaCl to modified (pre- diluted) artificial seawater experimentally confirmed this biphasic association, with suppression of golden alga growth being particularly notable at > 20 psu (Rashel and

Patiño, 2017). In contrast to the preceding observations, laboratory studies where salinity was manipulated by diluting artificial or natural seawater – which results in the proportional dilution of all ionic constituents – reported what seemed to be log-linear associations between salinity and growth without an apparent inhibition at high salinity

(Hambright et al., 2014) or with relatively attenuated biphasic associations (Padilla,

1970). Results of the present study seem to have resolved these conflicting observations by showing that the association between salinity and growth cannot be interpreted

Texas Tech University, Rakib H. Rashel, May 2020 without considering the ionic composition of natural waters or artificial media. The present findings also confirmed the working hypothesis that environmentally relevant

2- concentrations of SO4 (in inland habitats) enhance growth of golden alga independently of salinity. Contrary to initial expectations, however, algal growth was negatively influenced by F- at all concentrations tested.

4.1. Effects of sulfate and other ions at low salinity

Maximum cell density achieved by golden alga cultures at a constant salinity of 5

2- psu was positively associated with SO4 concentration. While the association was clear

2- regardless of SO4 source, its strength seemed to differ between the two sources; namely,

2- -1 as SO4 concentration increased from nominal 0 to 1000 mg l , maximum cell density increased by 61% when the source was Na2SO4 compared to 28% when the source was

2- MgSO4. On the other hand, r was positively associated with SO4 only when the source

2- was MgSO4, and early cell density was associated with SO4 only when the source was

Na2SO4. Results of partial correlation analysis of all data combined, however, revealed

2- that maximum cell density was strongly associated with r regardless of SO4 source, indicating the existence of confounding (masking) effects of early cell density on this association. Overall, these observations suggest that while Mg2+ enhances growth

2- performance and particularly in regard to r, SO4 stimulates growth of golden alga in a concentration-dependent manner and independently of salinity. The mechanisms of this association are at present unknown. Sulfate is the primary source of sulfur for algae

(Takahashi et al., 2011; Giordano and Raven, 2014), however, and under some conditions can limit productivity in freshwater systems (Giordano et al., 2005). Thus, the positive

Texas Tech University, Rakib H. Rashel, May 2020

2- association between SO4 concentration and golden alga growth could be interpreted, at

least partly, in a nutritional context.

The positive association between Mg2+ and golden alga growth revealed by this study may also be of ecological relevance. In brackish waters of the southcentral USA

2- where golden alga blooms have occurred, relatively high levels of SO4 are typically

accompanied by correspondingly high levels of hardness cations, including Mg2+ (Israël

et al. 2014; VanLandeghem et al., 2015a). Magnesium is a key micronutrient known to

influence algal growth, lipid content and composition, and chlorophyll content (Finkle

and Appleman, 1953; Hilt et al., 1987; Weiss et al., 2001; Esakkimuthu et al., 2016;

Srivastava et al., 2017). Thus, the relatively high levels of Mg2+ present in golden alga

inland habitats may also serve to facilitate growth of this species.

4.2. Effects of sulfate and other ions at high salinity

Growth of golden alga was significantly reduced when salinity was abruptly

increased from 5 to 30 psu by adding NaCl to modified ASM. This observation confirms

an earlier study of golden alga that reported growth inhibition in modified ASM at

salinities > 20 psu (Rashel and Patiño, 2017). When algal cells were transferred to 30 psu

in a more complex salt solution (IO) resembling the ionic composition of seawater,

however, growth was unaffected. Because Na+ and Cl- concentrations are similar in ASM

and IO at 30 psu (Table 1), it appears that growth suppression in ASM is the direct

consequence of increased salinity and that major ions in IO other than Na+ or Cl- prevent

or counteract the acute effects of osmotic stress on growth. Sulfate is the only major

anion that is present at a higher concentration in IO than ASM at 30 psu (~ 9-fold higher)

Texas Tech University, Rakib H. Rashel, May 2020

and among the major cations, Mg2+ and Ca2+ have the highest relative concentration in IO

2- 2+ 2+ (~20-fold higher) (Table 1). These three ions – SO4 , Mg , and Ca – individually or in

combination are likely candidates for the role of osmotic-stress rescue factors.

2- -1 Adding seawater levels of SO4 (2.4 g l ) using Na2SO4 as source did not fully

prevent growth inhibition caused by raising the salinity from 5 to 30 psu in modified

ASM. Maximum cell density and r were measurably higher in the presence of the extra

2- SO4 than in its absence but in both cases, growth was still greatly reduced compared to

control cultures at 5 psu. When MgSO4 was used as sulfate source, however, growth in

the high salinity medium was fully restored to control levels. These observations indicate

2- 2+ that although SO4 has a minor positive influence on algal growth at high salinity, Mg

is required for full growth restoration. The nominal concentration of Mg2+ in ASM

-1 -1 without additional MgSO4 is 0.06 g l , in ASM with additional MgSO4 is 0.61 g l , and

in IO (at 30 psu) is 1.26 g l-1 (Table 1). Because the growth performance in ASM with

additional MgSO4 and in IO were both similar to their respective control values, a

concentration of ~0.61 g l-1 Mg2+ seems sufficient to nullify the inhibitory effect of high

salinity. The mechanism by which Mg2+ rescues golden alga growth from high-salinity

stress was not addressed in this study and is unknown. It should be noted, however, that

the addition of seawater concentrations of Mg2+ to modified ASM at 30 psu simply restored growth to the same level observed in ASM at 5 psu, and that the latter medium contains a 10-fold lower concentration of Mg2+ (0.06 g l-1, Table1). Thus, the rescue

effect of Mg2+ at high salinity is unlikely to represent a classical nutritional function.

Talarski et al. (2016) reported that golden alga transferred from 5-psu medium to

30-psu natural seawater underwent major changes in gene expression, with at least 1507

Texas Tech University, Rakib H. Rashel, May 2020

and 1000 transcripts showing up and down regulation, respectively. Some of the

differentially expressed transcripts were associated with salinity stress, osmolyte

production, or ion transport. Golden alga in the study of Talarski et al. (2016) was

maintained in 30-psu seawater for several rounds of culture over several months before

analysis. Thus, interpretation of the acute effects of high salinity observed in the present

study in the context of the results of Talarski et al. (2016) should be made with caution.

In plants, however, Mg2+ is necessary for the proper functioning of hundreds of enzymes

(Kobayashi and Tanoi, 2015) and exposure to environmental stress often results in higher

demand for this ion in order to maintain energy balance (Igamberdiev and Kleczkowski,

2011). Elevated Mg2+ may thus facilitate the major metabolic readjustments necessary to

cope with osmotic stress (Talarski et al., 2016), which in turn may have served to prevent

growth inhibition at high salinity in the present study.

4.3. Effects of fluoride

Fluoride concentrations in inland waters typically range from 0.01 to 0.30 mg l-1 but natural or anthropogenic inputs can result in much higher concentrations (Dobbs,

1974; Camargo, 2003; Ochoa-Herrera et al., 2009). Exposure of algae to high F- levels

can result in growth suppression although the effective concentration varies widely

among species (Hekman et al., 1984; Bhatnagar and Bhatnagar, 2000; Camargo, 2003).

Growth suppression may be due, at least partly, to impaired photosynthesis (Wu et al.,

2007). Growth enhancement after exposure to fluoride also has been observed in some

species (Camargo, 2003), and enhancement at low concentration followed by suppression

at higher concentration (hormesis) has been reported as well (Wu et al., 2007). In the few

Texas Tech University, Rakib H. Rashel, May 2020

field studies of golden alga where F- concentration was measured, blooms seemed to

occur in water bodies with concentrations >1 mg l-1 (VanLandeghem et al., 2012, 2015a).

Given the wide variation in algal responses to F-, VanLandeghem et al. (2015a) hypothesized that putative tolerance to F- could favor growth of golden alga, a mixotrophic species, by providing a competitive advantage over phototrophic algae.

Results of the present study rejected the hypothesis; namely, growth inhibition occurred at the lowest nominal concentration of F- tested (2.25 mg l-1) making golden alga one of

the least fluoride-tolerant algal species studied to date. This finding raises the question of why, despite their intolerance to this ion, is golden alga able to thrive in water bodies with relatively high F- concentration. The relatively high levels of F- in golden alga

habitat, however, are accompanied by relatively high levels of water hardness

(VanLandeghem et al., 2012, 2015a) and as hardness increases, the bioavailability of

fluoride ion decreases (Camargo, 2003). Thus, the positive association between F- levels

and golden alga presence in the field could simply be a spurious observation.

5. Summary and Conclusions

This study showed that growth of golden alga at low (brackish) salinity responds

2- 2+ positively to elevated SO4 and Mg . These observations support the hypothesis that

2- spatial differences in SO4 concentration are partly responsible for determining the

inland distribution of golden alga (Patiño et al., 2014), and also provide physiological

relevance to reports of relatively high Mg2+ concentrations in water bodies with a history

of golden alga blooms (VanLandeghem et al., 2015a). At high salinity, however, the

influence of sulfate on golden alga growth was minor while elevated Mg2+ appeared to be

Texas Tech University, Rakib H. Rashel, May 2020

necessary to maintain growth under osmotic stress. Growth of golden alga was negatively

associated with F- concentration; consequently, positive associations between F- and algal

abundance observed in the field (VanLandeghem et al., 2015a) may simply be

coincidental. Overall, results of the present study highlight the importance of considering

the ionic composition of natural waters or experimental media in studies of the

association between salinity and golden alga distribution or growth.

Haptophytes belong to the red lineage of algae (Chromista), which traces its origin to the marine environment and which currently dominates marine phytoplankton communities (Cavalier-Smith, 2018). Sulfur is required for algal growth and can be limiting in freshwater systems but not in the ocean, where its levels are much higher

(Ratti et al., 2011; Giordano et al., 2005). Below normal seawater concentrations, growth of marine chromists (including the haptophyte, Emiliania huxleyi Lohmann) associates

2- positively with SO4 concentration while growth of marine cyanobacteria and green

lineage species seems to be largely unaffected (Ratti et al., 2011; Prioretti and Giordano,

2- 2016). It has been proposed that the growth response of extant marine chromists to SO4

is a vestigial trait that allowed their ancestral species an advantage over cyanobacteria

2- and green algae as oceanic SO4 concentrations increased during the Mesozoic Era

(Norici et al., 2005; Ratti et al., 2011; Prioretti and Giordano, 2016). In this context, the

2- ability of golden alga to increase its growth in the presence of elevated SO4 (up to at least 1000 mg l-1) in brackish environments is intriguing and suggests this response may

share the same evolutionary path as that of its marine relatives. Curiously, growth of the

2- freshwater cyanobacterium Microcystis aeruginosa Kützing is suppressed at SO4

concentrations as low as 40 mg l-1 (Chen et al., 2016), suggesting elevated sulfate has the

Texas Tech University, Rakib H. Rashel, May 2020 opposite effect than in golden alga. In contrast to marine chromists, however, the present

2- results with golden alga indicate that the ability of SO4 to maintain or stimulate growth at high salinity is limited and overshadowed by the importance of Mg2+. A mechanistic

2- 2+ 2+ understanding of the growth responses to SO4 , Mg and other ions (e.g., Ca ) at environmentally relevant levels and under different salinity scenarios will be necessary to clarify their ecophysiological and evolutionary relevance to golden alga.

Texas Tech University, Rakib H. Rashel, May 2020

Literature Cited

Atkinson, M.J., 1998. Bingman C. Elemental composition of commercial seasalts. J Aquaric Aquat Sci. 8(2), 39–43.

Baker, J.W., Grover, J.P., Brooks, B.W., Ureña-Boeck, F., Roelke, D.L., Errera, R., Kiesling, R.L., 2007. Growth and toxicity of Prymnesium parvum (Haptophyta) as a function of salinity, light, and temperature. J Phycol. 43(2), 219–227.

Bhatnagar, M., Bhatnagar, A., 2000. Algal and cyanobacterial responses to fluoride. Fluoride 33(2), 55–65.

Bochenek, M., Etherington, G.J., Koprivova, A., Mugford, S.T., Bell, T.G., Malin, G., Kopriva, S., 2013. Transcriptome analysis of the sulfate deficiency response in the marine microalga Emiliania huxleyi. New Phytol. 199(3), 650–662.

Camargo, J.A., 1996. Estimating safe concentrations of fluoride for three species of nearctic freshwater invertebrates: Multifactor Probit Analysis. Bull. Environ. Contam. Toxicol. 56(4), 643–648.

Camargo, J.A., 2003. Fluoride toxicity to aquatic organisms: a review. Chemosphere 50(3), 251–264.

Cavalier-Smith, T., 2018. Kingdom Chromista and its eight phyla: a new synthesis emphasising periplastid protein targeting, cytoskeletal and periplastid evolution, and ancient divergences. Protoplasma. 255(1), 297–357.

Chen, L., Gin, K.Y.H., He, Y., 2016. Effects of sulfate on microcystin production, photosynthesis, and oxidative stress in Microcystis aeruginosa. Environ. Sci. Pollu. Res. 23(4), 3586–3595.

Dillon, P.J., Kirchner, W.B., 1975. The effects of geology and land use on the export of phosphorus from watersheds. Water Res. 9(2), 135–148.

Dobbs, C.G., 1974. Fluoride and the environment. Fluoride 7(3), 123–135.

Edvardsen, B., Imai, I., 2006. The Ecology of Harmful Flagellates Within Prymnesiophyceae and Raphidophyceae. In: Granéli E, Turner JT, editors. Ecology of Harmful Algae. Ecological Studies, vol. 189. Springer, Berlin, pp. 67– 79.

Esakkimuthu, S., Krishnamurthy, V., Govindarajan, R., Swaminathan, K., 2016. Augmentation and starvation of calcium, magnesium, phosphate on lipid production of Scenedesmus obliquus. Biomass and Bioenergy 88, 126–134.

Falkowski, P.G., Katz, M.E., Knoll, A.H., Quigg, A., Raven, J.A., Schofield, O., Taylor, F.J.R., 2004. The evolution of modern eukaryotic phytoplankton. Science 305(5682), 354–360.

Texas Tech University, Rakib H. Rashel, May 2020

Finkle, B.J., Appleman, D., 1953. The effect of magnesium concentration on growth of Chlorella. Plant Physiol. 28(4), 664–673.

Foley, J.A., Defries, R., Asner, G.P., Barford, C., Bonan, G., Carpenter, S.R., Chapin, F.S., Coe, M.T., Daily, G.C., Gibbs, H.K., et al. 2005. Global consequences of land use. Science 309(5734), 570–574.

Giordano, M., Norici, A., Hell, R., 2005. Sulfur and phytoplankton: acquisition, metabolism and impact on the environment. New Phytol. 166(2), 371–382.

Giordano, M., Raven, J.A., 2014. Nitrogen and sulfur assimilation in plants and algae. Aqua. Bot. 118, 45–61.

Granéli, E., Edvardsen, B., Roelke, D.L., Hagström, J.A., 2012. The ecophysiology and bloom dynamics of Prymnesium spp. Harmful Algae 14, 260–270.

Guo, M, Harrison, P.J., Taylor, F.J.R., 1996. Fish kills related to Prymnesium parvum N. Carter (Haptophyta) in the People’s Republic of China. J. App. Phycol. 8(2), 111– 117.

Hambright, K.D., Zamor, R.M., Easton, J.D., Glenn, K.L., Remmel, E.J., Easton, A.C., 2010. Temporal and spatial variability of an invasive toxigenic protist in a North American subtropical reservoir. Harmful Algae 9(6), 568–577.

Hambright, K.D., Easton, J.D., Zamor, R.M., Beyer, J., Easton, A.C., Allison, B., 2014. Regulation of growth and toxicity of a mixotrophic microbe: implications for understanding range expansion in Prymnesium parvum. Fresh. Sci. 33(3), 745– 754.

Hekman, W.E., Budd, K., Palmer, G.R., MacArthur, J.D., 1984. Responses of certain freshwater planktonic algae to fluoride. J. Phycol. 20(2), 243–249.

Hilt, K.L., Gordon, P.R., Hein, A., Caulfield, J.P., Falchuk, K.H., 1987. Effects of Iron-, Manganese-, or Magnesium-deficiency on the growth and morphology of Euglena gracilis. J. Protozool. 34(2), 192–198.

Holm, S., 1979. A simple sequentially rejective multiple test procedure. Scand. J. of Statis. 6 (2), 65–70.

Holmer, M., Storkholm, P., 2001. Sulphate reduction and sulphur cycling in lake sediments: a review. Freshw. Biol. 46(4), 431–451.

Igamberdiev, A.U., Kleczkowski, L.A., 2011. Magnesium and cell energetics in plants under anoxia. Biochem. J. 437(3), 373–379.

Israël, N.M.D., VanLandeghem, M.M., Denny, S., Ingle, J, Patiño, R., 2014. Golden alga presence and abundance are inversely related to salinity in a high-salinity river ecosystem, Pecos River, USA. Harmful Algae 39, 81–91.

Texas Tech University, Rakib H. Rashel, May 2020

Kaushal, S.S., Groffman, P.M., Likens, G.E., Belt, K.T., Stack, W.P., Kelly, V.R., Band, L.E., Fisher, G.T. 2005. Increased salinization of fresh water in the northeastern United States. Proc Natl Acad Sci USA. 102(38), 13517–13520.

Kobayashi, N.I., Tanoi, K., 2015. Critical issues in the study of magnesium transport systems and magnesium deficiency symptoms in plants. Int. J. Mol. Sci. 16(9), 23076–23093.

Larsen, A, Bryant, S., 1998. Growth rate and toxicity of Prymnesium parvum and Prymnesium patelliferum (Haptophyta) in response to changes in salinity, light and temperature. Sarsia 83(5), 409–418.

Lutz-Carrillo, D.J., Southard, G.M., Fries, L.T., 2010. Global genetic relationships among isolates of golden alga (Prymnesium parvum). J. Am. Water Resour. Assoc. 46(1), 24–32.

Moestrup, Ø., 1994. Economic aspects: ‘blooms’, nuisance species and toxins. In: Green J.C., Leadbeater B.S.C., editors. The Haptophyte Algae. Oxford: Clarendon Press. 265–285.

Nicholls, K.H., 2003. Haptophyte Algae. In: Wehr JD, Sheath RG, editors. Freshwater Algae of North America. Boston: Academic Press. 511–521.

Norici, A., Hell, R., 2005. Giordano M. Sulfur and primary production in aquatic environments: an ecological perspective. Photosynth. Res. 86(3), 409–417.

Ochoa-Herrera, V., Banihani, Q., León, G., Khatri, C., Field, J.A., Sierra-Alvarez, R., 2009. Toxicity of fluoride to microorganisms in biological wastewater treatment systems. Water Res. 43(13), 3177–3186.

Padilla, G.M., 1970. Growth and toxigenesis of the chrysomonad Prymnesium parvum as a function of salinity. The J. of Protozoo. 177, 456–462.

Patiño, R., Dawson, D., VanLandeghem, M.M., 2014. Retrospective analysis of associations between water quality and toxic blooms of golden alga (Prymnesium parvum) in Texas reservoirs: Implications for understanding dispersal mechanisms and impacts of climate change. Harmful Algae 33, 1–11.

Prioretti, L., Giordano, M., 2016. Direct and indirect influence of sulfur availability on phytoplankton evolutionary trajectories. J. Phycol. 52(6), 1094–1102.

Rashel, R.H., Patiño, R., 2017. Influence of genetic background, salinity, and inoculum size on growth of the ichthyotoxic golden alga (Prymnesium parvum). Harmful Algae 66, 97–104.

Ratti, S., Knoll, A.H., Giordano, M., 2011. Did sulfate availability facilitate the evolutionary expansion of chlorophyll a+c phytoplankton in the oceans? Geobiology 9(4), 301–312.

Texas Tech University, Rakib H. Rashel, May 2020

Roelke, D.L., Barkoh, A., Brooks, B.W., Grover, J.P., Hambright, K.D., La Claire, J.W. Ⅱ., Moeller, P. D.R., Patiño, R., 2016. A chronicle of a killer alga in the west: ecology, assessment, and management of Prymnesium parvum blooms. Hydrobiologia 764(1), 29–50.

Srivastava, G., Nishchal, Goud, V.V., 2017. Salinity induced lipid production in microalgae and cluster analysis (ICCB 16-BR_047). Bioresource Technology 242, 244–252.

Takahashi, H., Kopriva, S., Giordano, M., Saito, K., Hell, R., 2011. Sulfur assimilation in photosynthetic organisms: molecular functions and regulations of transporters and assimilatory enzymes. Annu. Rev. Plant Biol. 62(1), 157–184.

Talarski, A., Manning, S.R., La Claire, J.W. Ⅱ., 2016. Transcriptome analysis of the euryhaline alga, Prymnesium parvum (Prymnesiophyceae): effects of salinity on differential gene expression. Phycologia 55(1), 33–44.

VanLandeghem, M.M., Farooqi, M., Southard, G.M., Patiño, R., 2015a. Associations between Water Physicochemistry and Prymnesium parvum Presence, Abundance, and Toxicity in West Texas Reservoirs. J. Am. Water Resour. Assoc. 51(2), 471– 486.

Vanlandeghem, M.M., Farooqi, M., Southard, G.M., Patiño, R., 2015b. Spatiotemporal associations of reservoir nutrient characteristics and the invasive, harmful alga Prymnesium parvum in West Texas. J. Am. Water Resour. Assoc. 51(2), 487– 501.

VanLandeghem, M.M., Meyer, M.D., Cox, S.B., Sharma, B., Patiño, R., 2012. Spatial and temporal patterns of surface water quality and ichthyotoxicity in urban and rural river basins in Texas. Water Res. 46(20), 6638–6651.

Weimin, G., 1983. The reason for the fish death at aquacultural experimental station at Ningxia and the distribution of Prymnesium parvum in Ningxia. J. Dalian Fish. Coll. 1, 43–48 (in Chinese).

Weiss, M., Haimovich, G., Pick, U., 2001. Phosphate and sulfate uptake in the halotolerant alga Dunaliella are driven by Na+-symport mechanism. J. Plant Physiol. 158(12), 1519–1525.

Wood, A.M., Everroad, R.C., Wingard, L., 2005. Measuring growth rates in microalgal cultures. In: Andersen RA., editor. Algal Culturing Techniques. Elsevier Academic Press; p. 598.

Texas Tech University, Rakib H. Rashel, May 2020

Wu, Y., Li, P., Zhao, X., 2007. Effect of fluoride on carbonic anhydrase activity and photosynthetic oxygen evolution of the algae Chlamydomonas reinhardtii. Fluoride 40(1), 51–64.

Yildiz, F.H., Davies, J.P., Grossman, A.R., 1994. Characterization of sulfate transport in Chlamydomonas reinhardtii during sulfur-limited and sulfur-sufficient growth. Plant Physiol. 104(3), 981–987.

Texas Tech University, Rakib H. Rashel, May 2020

Tables

Table 3.1. The nominal concentration of major ions or constituents of Artificial Seawater Medium (ASM), Instant Ocean® (IO), and Seawater (SW). Major ions (g l-1) in experimental media and seawater (SW) Instant Ocean Artificial Seawater Medium (ASM) (IO) SW 30 psu + 30 psu + 30 Ion 5 psu 30 psu 5 psu 30 psu MgSO4 Na2SO4 psu* + Na 1.77 11.60 11.60 11.60 1.77 10.62 9.27 2+ Mg 0.06 0.06 0.61 0.06 0.21 1.26 1.10 + K 0.08 0.08 0.08 0.08 0.06 0.37 0.34 2+ Ca 0.02 0.02 0.02 0.02 0.06 0.38 0.35 - Cl 2.73 17.90 17.90 17.90 3.08 18.50 16.74 2- SO4 0.25 0.25 2.40 2.40 0.37 2.21 2.31 *Values are adjusted from concentrations reported for full-strength seawater (35 psu)

Texas Tech University, Rakib H. Rashel, May 2020

Table 3.2. Output of one-way and two-way ANOVA of data collected in this study. Growth indices examined in Prymnesium parvum cultures included exponential growth rate (r), maximum density, and early density. Response Effects DFn, DFd F-value p-value Effects of gradual changes in sulfate r trial 1, 16 0.205 0.657 concentration at low salinity (sulfate maximum density 1, 16 0.624 0.441 source: MgSO4) early density 1, 16 0.000 0.999 r interaction 3, 16 0.912 0.457 maximum density 3, 16 0.759 0.533 early density 3, 16 0.888 0.468 r concentration 3, 16 9.678 0.001 maximum density 3, 16 30.94 < 0.0001 early density 3, 16 0.762 0.532 Effects of gradual changes in sulfate r concentration 3, 8 1.681 0.248 concentration at low salinity (sulfate maximum density 3, 8 0.109 < 0.0001 source: Na2SO4) early density 3, 8 2.801 0.109 Effects of salinity under different major salt r concentration 3, 8 6.006 0.019 scenarios maximum density 3, 8 38.668 < 0.0001 Effects of differences in sulfate r concentration 2, 9 15.893 < 0.0001 concentration at high salinity (Sulfate maximum density 2, 9 81.41 < 0.0001 source: MgSO4) Effects of differences in sulfate r concentration 2, 6 25.995 < 0.0001 concentration at high salinity (Sulfate maximum density 2, 6 367.39 < 0.0001 source: Na2SO4) Effects of gradual changes in fluoride r trial 1, 16 0.008 0.93 concentration at low salinity maximum density 1, 16 2.099 0.167

r interaction 3, 16 0.846 0.489 maximum density 3, 16 0.66 0.588 r concentration 3, 16 20.888 < 0.0001 maximum density 3, 16 53.882 < 0.0001 DFn, degrees of freedom numerator; DFd, degrees of freedom denominator.

Texas Tech University, Rakib H. Rashel, May 2020

Table 3.3. Non-parametric Spearman's correlation of maximum cell density, exponential growth rate (r), and early cell density in Prymnesium parvum cultures versus sulfate concentration (0–1000 mg l-1) in modified ASM at a salinity of 5 psu. Matrix of Simple Correlations Source Maximum Early 2- of SO4 density p-value* r p-value* density p-value* n

MgSO4 0.96 0.0005 0.78 0.0005 0.25 0.24 24 Na2SO4 0.95 0.0005 0.52 0.09 0.68 0.0334 12 2- *Holm-Šídák adjusted p-values for multiple tests within each SO4 source; the adjustment includes tests described in Table 4. Data used for these analyses are those reported in figures 1 and 2 and Table 4. n, sample size.

Texas Tech University, Rakib H. Rashel, May 2020

Table 3.4. Pearson partial correlation of maximum cell density with exponential growth rate (r) or early cell density in Prymnesium parvum cultures as a function of sulfate concentration (0–1000 mg l-1) in modified ASM at a salinity of 5 psu. Partial Correlation with Maximum Cell Density Source of Correlated Control 2- SO4 Variable variable Pearson r p-value* n Early r 0.76 0.0005 24 density MgSO4 Early r 0.36 0.17 24 density Early r 0.85 0.0024 12 density Na2SO4 Early r 0.94 0.0005 12 density 2- *Holm-Šídák adjusted p-values for multiple tests within each SO4 source; the adjustment includes tests described in Table 3. When r was the correlation variable, early density was used as a control variable, and vice versa. Data used for these analyses are those reported in figures 1 and 2 and Table 3. n, sample size.

Texas Tech University, Rakib H. Rashel, May 2020

Figures

Figure 3.1. Growth indices of Prymnesium parvum as a function of sulfate concentration -1 (0–1000 mg l , as MgSO4) in modified ASM at a salinity of 5 psu. (A) Growth curves, (B) early density, (C) exponential growth rate (r), and (D) maximum cell density. Each time point or bar represents the mean (± SEM) of 6 replicates. Bars with the same letter codes do not differ significantly (Tukey’s HSD, p ˂ 0.05).

Texas Tech University, Rakib H. Rashel, May 2020

Figure 3.2. Growth indices of Prymnesium parvum as a function of sulfate concentration -1 (0–1000 mg l , as Na2SO4) in modified ASM at a salinity of 5 psu. (A) Growth curves, (B) early density, (C) exponential growth rate (r), and (D) maximum cell density. Each time point or bar represents the mean (± SEM) of 3 replicates. Bars with the same letter codes do not differ significantly (Tukey’s HSD, p ˂ 0.05).

Texas Tech University, Rakib H. Rashel, May 2020

Figure 3.3. Growth indices of Prymnesium parvum as a function of salinity in modified ASM or Instant Ocean with salinities of 5 and 30 psu. Salinity of 30 psu in modified ASM was achieved by the addition of NaCl to 5-psu medium. In IO medium, different salinities were achieved by direct addition of the appropriate amounts of salt mixture. (A) Growth curves, (B) exponential growth rate (r), and (C) maximum cell density. Each time point or bar represents the mean (± SEM) of 3 replicates. Bars with the same letter codes do not differ significantly (Tukey’s HSD, p ˂ 0.05).

Texas Tech University, Rakib H. Rashel, May 2020

Figure 3.4. Growth indices of Prymnesium parvum in modified ASM as a function of salinity and MgSO4 concentration. Salinity of 30 psu in modified ASM was achieved by -1 2- adding NaCl to 5-psu medium. The treatment receiving MgSO4 (2.4 g l SO4 ) received a correspondingly lower amount of NaCl to maintain salinity at 30 psu. (A) Growth curves, (B) exponential growth rate (r), and (C) maximum cell density. Each time point or bar represents the mean (± SEM) of 4 replicates. Bars with the same letter codes do not differ significantly (Tukey’s HSD, p ˂ 0.05).

Texas Tech University, Rakib H. Rashel, May 2020

Figure 3.5. Growth indices of Prymnesium parvum in modified ASM as a function of salinity and Na2SO4 concentration. Salinity of 30 psu in modified ASM was achieved by -1 2- adding NaCl to 5-psu medium. The treatment receiving Na2SO4 (2.4 g l SO4 ) received a correspondingly lower amount of NaCl to maintain salinity at 30 psu. (A) Growth curves, (B) exponential growth rate (r), and (C) maximum cell density. Each time point or bar represents the mean (± SEM) of 3 replicates. Bars with the same letter codes do not differ significantly (Tukey’s HSD, p ˂ 0.05).

Texas Tech University, Rakib H. Rashel, May 2020

Figure 3.6. Growth indices of Prymnesium parvum as a function of fluoride concentration (0–56.50 mg l-1) in modified ASM at salinity of 5 psu. (A) Growth curves, (B) exponential growth rate (r), and (C) maximum cell density. Each time point or bar represents the mean (± SEM) of 6 replicates. Bars with the same letter codes do not differ significantly (Tukey’s HSD, p ˂ 0.05).

Texas Tech University, Rakib H. Rashel, May 2020

CHAPTER Ⅳ

RATIO OF INORGANIC TO ORGANIC NITROGEN INFLUENCES GROWTH OF THE HAPTOPHYTE, PRYMNESIUM PARVUM

Abstract

Prymnesium parvum is a toxic bloom-producing haptophyte which in inland ecosystems is typically found in eutrophic brackish waters. As a mixotroph, P. parvum can utilize organic and inorganic nitrogen (N) for growth but the relative importance of each fraction when present in combination is uncertain. Some field studies have suggested there is a positive association between organic N and golden alga distribution or abundance, but experimental evidence supporting this observation is unavailable. The objective of this study is to determine if different molar ratios of inorganic to organic N affect growth of golden alga in a standard culture medium (5 psu). Sodium nitrate was used as source of inorganic N and urea and glycine as sources of organic N.

Concentrations of total N (880 μM) and phosphorus (36 μM) were kept constant at F/2 levels. The inorganic:organic N ratios tested were 1:0, 0.75:0.25, 0.5:0.5, 0.75:0.25, and

0:1. Cultures were conducted under standard conditions (initial density, 100 cells ml-1;

22°C, ~6500 lux), and endpoints measured were early cell density (day 3, cells ml-1),

exponential growth rate (r, day-1) and maximum cell density (cells ml-1). Growth rate was

unaffected by changes in inorganic:organic N ratio. Early and maximum cell density,

however, increased gradually as the fraction of organic N increased from 0 to 0.75

followed by a decrease at 1, and this pattern was stronger when using glycine as source.

In conclusion, while golden alga can grow in cultures supplemented exclusively with

100

Texas Tech University, Rakib H. Rashel, May 2020

organic or inorganic N, optimal growth occurs when both are present and the organic

fraction is predominant. These findings are consistent with field observations and provide

context for a better understanding of the association between nutrient stoichiometry and

golden alga growth.

1. Introduction

Harmful algal blooms (HAB) are a global problem driven to a considerable degree by cultural eutrophication of inland and coastal waters worldwide (Hallegraeff,

1993; Anderson, 2009; Gobler et al., 2016; Pearl et al., 2016). Impacts of HAB can be measured in terms of biological degradation of aquatic ecosystems, losses in ecosystem services, and financial costs associated with mitigation and management efforts.

Prymnesium parvum, known in North America as golden alga, is a HAB-forming haptophyte of worldwide distribution (Granéli et al., 2012). Toxic blooms of this species in North America are relatively recent phenomena, but they have spread widely since first reported in the mid-1980s (Roelke et al., 2016). Golden alga can grow in a wide variety of habitats; however, most bloom events in inland systems have occurred in brackish, nutrient-rich water bodies (Guo et al., 1996; Edvardsen and Paasche, 1998; Granéli et al.,

2012; Roelke et al., 2016).

Previous field and laboratory studies reported that environmental conditions such as salinity, levels of salinity-associated ions (i.e., sulfate, magnesium), and total and relative amounts of nitrogen (N) and phosphorus (P) can influence bloom formation or toxicity of this species (Baker et al. 2007; Hambright et al., 2010; Brooks et al. 2011;

Patiño et al., 2014; VanLandeghem et al., 2015; Roelke et al., 2016; Rashel and Patiño,

101

Texas Tech University, Rakib H. Rashel, May 2020

2017, 2019). There are some inconsistencies, however, regarding the influence of

nutrients beyond the consensus that eutrophic conditions are the norm for bloom

development. Namely, some field studies reported that low N:P ratios are strongly

associated with bloom formation (Hambright et al., 2010), while others reported that high

N:P ratios favor blooms in microcosm experiments (Errera et al., 2008). In addition,

while it is known that golden alga can utilize organic and inorganic nutrients

(McLaughlin, 1958; Lindehoff et al., 2010, 2011; Granéli et al., 2012), the relative

importance of each fraction to support growth when present in combination is uncertain.

Ammonium (a form of inorganic N) at high concentrations is toxic to golden alga

(McLaughlin, 1958; Grover et al., 2007). A field study of reservoirs in the upper

Colorado River Basin (Texas, USA) found that seasonal declines in golden alga

abundance following peak bloom densities occurred as levels of inorganic N increased

(VanLandeghem et al., 2015). The same study found that golden alga-impacted reservoirs

had higher concentrations of organic N than non-impacted reservoirs (VanLandeghem et al., 2015). In the Pecos River Basin (New Mexico and Texas, USA), golden alga abundance was positively associated with organic N and negatively with inorganic N

(Israël et al., 2014). These observations are also consistent with inferential conclusions that toxic blooms of P. parvum in other locations were initiated when inorganic nutrients were limited and organic nutrients sources seemed plentiful (Granéli et al., 2012).

The objective of this study is to determine if changes in the ratio of inorganic to

organic N affect growth of golden alga. Two sources of organic N were used: urea, which

is an important source of N for phytoplankton communities (Solomon et al., 2010); and

glycine, which is among the most prominent dissolved free amino acids found in

102

Texas Tech University, Rakib H. Rashel, May 2020

freshwater, estuarine, and marine habitats (Crawford et al., 1974; Suttle et al., 1991;

Hubberten et al., 1994; Horňák et al., 2016; Horňák and Pernthaler, 2019). Based on

findings of previous field studies (Israël et al., 2014; VanLandeghem et al., 2015), the

working hypothesis of the present study is that as the relative concentration of organic N

increases (with a concominant decrease in inorganic N) while keeping total N constant,

maximum population densities will also increase.

2. Materials and Methods

2.1. Basic culture procedures

The strain of golden alga used in this study (UTEX LB2797) was obtained from

the UTEX Culture Collection of Algae (The University of Texas at Austin, Texas, USA).

Stock cultures were grown in Artificial Seawater Medium (ASM;

https://utex.org/products/artificial-seawater-medium) modified (diluted) to a salinity of 5

psu as described by Rashel and Patiño (2017). For stock cultures, modified ASM was

enriched with F/2 levels of nutrients and vitamins (Guillard, 1975); the source of N in this

medium is sodium nitrate. In the original trace metal solution, an equimolar amount of

ferric chloride was added as a replacement for ferrous ammonium sulfate to avoid toxic

effects of ammonium on golden alga (Grover et al., 2007). Major salts and nutrients were added before autoclaving, and vitamins were added after autoclaving. Stock and experimental cultures were maintained in 250-ml Erlenmeyer flasks with a working volume of 100 ml ASM and kept in an incubator (I36LLVLC9; Percival Scientific Inc.

Perry, IA, USA) with a pre-set temperature and photoperiod of 22 °C and 12:12 h

103

Texas Tech University, Rakib H. Rashel, May 2020

light:dark, respectively. Light intensity was ~6500 lux. Regular fluorescent lights were

used to achieve the desired light intensity. Cultures were not axenic but aseptic

procedures were followed to minimize bacterial contamination. Late exponential phase

cells were used as inocula to maintain stock cultures and as seed for all experimental

cultures. All cultures were swirled once daily.

2.2. Experimental cultures

The two organic N sources used in this study, urea and the amino acid glycine,

have been previously shown capable of supporting golden alga growth (McLaughlin,

1958; Lindehoff et al., 2010, 2011). Modified ASM for these experiments was prepared

following the same procedures used for stock cultures except that inorganic N (sodium

nitrate) was incrementally replaced with the appropriate amounts of urea (Sigma U5378,

Sigma-Aldrich, St. Louis, MO, USA) or glycine (Sigma G8790, Sigma-Aldrich, St.

Louis, MO, USA) to achieve the desired ratio of inorganic to organic N while

maintaining total N constant at F/2 level (880 μM). The following molar ratios were

prepared: 1:0, 0.75:0.25, 0.5:0.5, 0.25:0.75, and 0:1 (inorganic:organic N). All treatment

media contained standard F/2 levels of P (36 μM) and vitamins.

A working volume of 100 ml of appropriate medium was inoculated to an initial

density of 100 cells ml-1. All treatments were conducted in triplicate, and two full

independent trials for each source of organic N were conducted. Experimental cultures were maintained under the same ambient conditions as stock cultures.

104

Texas Tech University, Rakib H. Rashel, May 2020

2.3. Analytical procedures

Cell densities were determined by hemocytometry as described by Rashel and

Patiño (2017). Briefly, an aliquot of 500 μL of culture medium from each replicate flask

was taken every 3 days and used to determine the cell numbers. If needed to facilitate

counting, the aliquot was diluted with fresh medium to maintain the cell number below

50,000 cells ml-1. For each replicate flask, a total of three counts were taken and the

average cell number was reported.

The primary growth indices estimated for all experiments were early cell density

(cells ml-1), exponential growth rate (r, day-1), and maximum cell density (cells ml-1).

Exponential growth rate (r) was calculated by using the following equation (Wood et al.,

2005):

ln N ln N = t t 2 − 1 𝑟𝑟 2 1 −

where N1 and N2 represent cell densities at day t1 and t2, respectively (t2 > t1). For each

replicate, t1 and t2 were chosen so that they bracket the linear portion of the ln-

transformed growth curve. Cell counts on day 3 were considered as early cell density and

represent growth near or at the beginning of the exponential growth phase, and maximum

cell density was the highest cell count observed during the culture period.

Two-way ANOVA was used to determine the effects of inorganic:organic N

ratios on growth. Main factors were ratio level and trial, and outcomes were assessed

separately for each organic N source. If trial had a significant main or interaction effect,

105

Texas Tech University, Rakib H. Rashel, May 2020

pairwise differences between means were determined separately for each trial; otherwise,

values from both trials were pooled. Mean separations were conducted using Tukey’s

HSD test. To determine the associations between maximum population density and the

other two growth indices (early cell density and r), partial correlation analysis was used where early cell density and r are used either control or correlated variable. All analyses were done in SPSS 25.0 (SPSS Inc., Chicago, USA). Graphics were prepared with

GraphPad Prism 8 (GraphPad Software, Inc., La Jolla, CA, USA).

3. Results

Regardless of ratio level, the stationary growth phase was reached by days 21–24 and 18–24 when urea (Fig. 1A) or glycine (Fig. 2A) were used as organic N source, respectively. On semi-ln plots, the linear portion of the exponential growth phase generally bracketed days 3–12.

When using urea as organic N source, significant trial effects were observed for r and maximum cell density, but not early cell density (Table 1). No interaction effects were observed between treatment and trial for any of the endpoints measured (Table 1).

Changes in inorganic:organic N ratio had significant effects on early and maximum cell density but not r (Table 1). A trend for increasing values of early cell density was observed as the fraction of organic N increased from 0 to 0.75, followed by a decline at 1; however, significant differences were recorded only between the organic N fractions of

0.75 and 1 (Fig. 1B). For maximum cell density, pairwise differences between ratio levels were assessed separately for each trial (Fig. 1D, Supplemental Fig. 1). A significant trend of increasing maximum density was noted as the organic N fraction increased from 0 to

106

Texas Tech University, Rakib H. Rashel, May 2020

0.75 in one of the trials; the second trial showed the same trend, but it was not significant

(Fig. 1D). In both trials, a sharp decline in maximum abundance was observed when urea was the sole source of N (Fig. 1D). Maximum population densities when urea was the sole N source were much lower than for any of the other treatments, including inorganic

N as sole source (Fig. 1D). While r was not affected by treatment (Table 1, Fig 1C), the sharp decline in maximum population densities when urea was the exclusive source of exogenous N was associated with a shortened transition period between the exponential and stationary phases (Supplemental Fig. 2). In additions, results of partial correlation analysis of all data combined showed that maximum cell density was significantly associated with r (Pearson coefficient, 0.65) and early cell density (Pearson coefficient,

0.65) (Table 2).

When using glycine as organic N source, significant trial and interaction effects were observed only for maximum cell density (Table 1). Treatment effects were observed for early and maximum cell density, but not r (Table 1). A trend for increasing values of early cell density was observed as the fraction of organic N increased from 0 to 0.75, followed by a moderate decline at 1; however, significant differences were recorded only between the organic N fractions of 0 and 0.75, and 0.75 and 1 (Fig. 2B). Pairwise differences between ratio levels for maximum cell density were assessed separately for each trial (Fig. 2D, Supplemental Fig. 3). A significant trend for increasing maximum cell density was noted as the organic N fraction increased from 0 to 0.75, followed by a moderate decline at 1 (Fig. 2D). Maximum cell densities did not differ when either sodium nitrate or glycine served as sole source of N (Fig. 2D). While r was not affected by changes in inorganic:organic N ratio (Table 1, Fig. 2C), results of partial correlation

107

Texas Tech University, Rakib H. Rashel, May 2020 analysis of all data combined showed that maximum cell density was significantly associated with r (Pearson coefficient, 0.66) and early cell density (Pearson coefficient,

0.79) (Table 2).

4. Discussion

Consistent with earlier studies (McLaughlin, 1958; Lindehoff et al., 2010, 2011), golden alga grew in the exclusive presence of exogenous inorganic or organic N.

Maximum population densities during culture were the same when nitrate or glycine were used as sole sources of N although they were relatively reduced in the presence of urea alone. Maximum densities were higher, however, when inorganic and organic N were both present (even in the case of urea) and highest when organic N was predominant

(75% of total N). The same pattern of responses was generally observed for early cell density, suggesting that treatment effects began early during the culture period. While r was not affected based on ANOVA results, partial correlation analysis indicated that maximum population density was significantly associated with r when the effects of early cell density were removed. Therefore, under nutrient-sufficient conditions, results of this study showed that golden alga growth generally increases as organic N increases from 0 to 75% of total available N – in other words, as inorganic N decreases from 100 to 25% of total N.

The present findings are consistent with field observations of Israël et al. (2014), who found that golden alga abundance in the Pecos River Basin (Texas and New Mexico,

USA) is positively associated with organic N and negatively with inorganic N. In addition, VanLandeghem et al. (2015) observed an inverse association between golden

108

Texas Tech University, Rakib H. Rashel, May 2020 alga abundance and inorganic N as cell abundance declined following peak bloom densities in reservoirs of the upper Colorado River (Texas). They also found that golden alga-impacted reservoirs have higher levels of organic N than non-impacted reservoirs

(VanLandeghem et al., 2015). The analyses of Israël et al. (2014) and VanLandeghem et al. (2015) were done on unfiltered samples and thus represent total digestable N, which likely includes not only dissolved N but also particulate N associated with living matter; namely, they represent total inorganic and organic N fractions. Therefore, while the sources of N (especially organic N) used in the present laboratory study may differ from those available to golden alga in the field, they seem to be a reasonable proxy given the similarity in results. Based on the results of the same field studies (Israël et al., 2014;

VanLandeghem et al., 2015) that led to the working hpothesis of the presnt study, Roelke et al. (2016) also suggested the possibility that organic nutrient availability more strongly controls intracellular nutrient limitation than inorganic nutrient availability.

Lindehoff et al. (2010) found that uptake relative to availability (relative preference index) generally indicated a preference of golden alga for amino acids

(including glycine) over nitrate, and of nitrate over urea in golden alga. Their study was conducted under nutrient-limiting conditions and did not report treatment effects on cell abundance; however, their results are generally consistent with the present findings of the relatively poor performance of urea and of similar performance of nitrate and glycine as sources of N for golden alga growth. In addition, Lindehoff et al. (2011) reported that P. parvum has a low affinity for urea and considered this compound of lesser importance as source of N. As mentioned already, however, the combination of urea and nitrate, especially when urea was predominant, yielded better growth in the present study than

109

Texas Tech University, Rakib H. Rashel, May 2020 each compound individually. The presence and nutrient composition of prey may also influence golden alga’s preference for dissolved N (Lundgren et al., 2016).

The mechanisms by which the combined presence of organic and inorganic N results in better growth compared to the individual fractions alone were not addressed in this study. It is known, however, that inorganic N can influence the uptake of organic N, and vice versa, in various phytoplankton species (e.g, McCarthy and Eppley, 1972; Antia et al., 1991; Glibert and Terlizzi, 1999; Solomon et al., 2010). These interactions may result in differential availability of the two fractions when present in combination but would not fully explain the consequent physiological effects; namely, the metabolic processes necessary for growth. Additional research is clearly necessary to understand mechanisms.

In conclusion, while golden alga could grow in the exclusive presence of either inorganic or organic N as shown by previous studies, growth was optimal when both were present and organic N was predominant. In addition to total or inorganic nutrient concentration or N:P ratios, which are commonly used indices of nutrient sources and composition, organic nutrient fractions should also be empirically evaluated in field studies of golden alga for their potential to drive or influence bloom formation.

110

Texas Tech University, Rakib H. Rashel, May 2020

Literature Cited

Anderson, D.M., 2009. Approaches to monitoring, control and management of harmful algal blooms (HABs). Ocean Coast. Manage. 52 (7), 342.

Antia, N.J., Harrison, P.J., Oliveira, L., 1991. The role of dissolved organic nitrogen in phytoplankton nutrition, cell biology and ecology. Phycologia 30,1–89.

Baker, J.W., Grover, J.P., Brooks, B.W., Ureña-Boeck, F., Roelke, D.L., Errera, R.M., Kiesling, R., 2007. Growth and toxicity of Prymnesium parvum (Haptophyta) as a function of salinity, light and temperature. J. of Phycol. 43, 219–227.

Brooks, B.W., Grover, J.P., Roelke, D.L., 2011. Prymnesium parvum, An emerging threat to inland waters. Environ. Toxi. and Chem. 30, 1955–1964.

Crawford, C.C., Hobbie, J.E., Webb, K.L., 1974. The utilization of dissolved free amino acids by estuarine microorganisms. Ecology 55, 551–563.

Edvarsen, B., Paasche, E., 1998. Bloom dynamics and physiology of Prymnesium and Chrysochromulina. Physiol. Ecol. Harmful Algal Bloom. In: Anderson, D.M., Cembella, A.D., Hallegraeff, G.M., (eds.). Physiological Ecology of Harmful Algal Blooms: NATO ASI Series. G41. Springer-Verlag; Berlin, Germany. pp. 193–208.

Errera, R.M., Roelke, D.L., Kiesling, R., Brooks, B.W., Grover, J.P., Schwierzke, L., Ureña-Boeck, F., Baker, J.W., Pinckney, J.L., 2008. The effect of imbalanced nutrients and immigration on Prymnesium parvum community dominance and toxicity: Results from in-lake microcosm experiments, Texas, USA. Aqua. Micro. Eco. 52, 33–44.

Glibert, P.M., Terlizzi, D.E., 1999. Cooccurrence of elevated urea levels and dinoflagellate blooms in temperate estuarine aquaculture ponds. Appl. Environ. Micro. 65, 5594–5596.

Gobler, C.J., Burkholder, J.M., Davis, T.W., Harke, M.J., Johengen, T., Stow, C.A., Van de Waal, D.B., 2016. The dual role of nitrogen supply in controlling the growth and toxicity of cyanobacterial blooms. Harmful Algae 54, 87–97.

Granéli, E., Edvardsen, B., Roelke, D.L., Hagström, J.A., 2012. The ecophysiology and bloom dynamics of Prymnesium spp. Harmful Algae 14, 260–270.

Grover, J.P., Baker, J.W., Ureña-Boeck, F., Brooks, B.W., Errera, R.M., Roelke, D.L., Kiesling, R.L., 2007. Laboratory tests of ammonium and barley straw extract as agents to suppress abundance of the harmful alga Prymnesium parvum and its toxicity to fish. Water Res. 41, 2503–2512.

111

Texas Tech University, Rakib H. Rashel, May 2020

Guillard, R.L., 1975. Culture of phytoplankton for feeding marine invertebrates. In Smith, W.L. and M.H. Charley (Eds). Culture of Marine Invertebrate Animals. Plenum Press, New York. 29–60.

Guo, M., Harrison, P.J., Taylor, F.J.R., 1996. Fish kills related to Prymnesium parvum N. Carter (Haptophyta) in the People’s Republic of China. J. Appl. Phycol. 8, 111– 117.

Hallegraeff, G.M., 1993. A review of harmful algal blooms and their apparent global increase. Phycologia 32, 79–99.

Hambright, K.D., Zamor, R.M., Easton, J.D., Glenn, K.L., Remmel, E.J., Easton, A.C., 2010. Temporal and spatial variability of an invasive toxigenic protist in a North American subtropical reservoir. Harmful Algae 9, 568–577.

Horňák, K., Pernthaler, J., 2019. Homeostatic regulation of dissolved labile organic substrates by consumption and release processes in a freshwater lake. Limnol. Oceanogr. 9999, 1–12.

Horňák, K., Schmidheiny, H., Pernthaler, J., 2016. High-throughput determination of dissolved free amino acids in unconcentrated freshwater by ion-pairing liquid chromatography and mass spectrometry. J. Chromatogr. A 1440, 85–93.

Hubberten, U., Lara, R.J., Kattner, G., 1994. Amino acid composition of seawater and dissolved humic substances in the Greenland Sea. Mar. Chem. 45, 121–128.

Israël, N.M.D., VanLandeghem, M.M., Denny, S., Ingle, J., Patiño, R., 2014. Golden alga presence and abundance are inversely related to salinity in a high-salinity river ecosystem, Pecos River, USA. Harmful Algae 39, 81–91.

Lindehoff, E., Granéli, E., Glibert, P.M., 2010. Influence of prey and nutritional status on the rate of nitrogen uptake by Prymnesium parvum (haptophyte). J. Am. Water Resour. Assoc. 46, 121–132.

Lindehoff, E., Granéli, E., Glibert, P.M., 2011. Nitrogen uptake kinetics of Prymnesium parvum (Haptophyte). Harmful Algae 12, 70–76.

McLaughlin, J.J.A., 1958. Euryhaline chrysomonads: Nutrition and toxigenesis in Prymnesium parvum, with notes on Isochrysis galbana and Monochrysis lutheri. J. Protozool. 5, 75–81.

112

Texas Tech University, Rakib H. Rashel, May 2020

McCarthy, J.J., Eppley, R.W., 1972. A comparison of chemical isotopic and enzymatic method for measuring nitrogen assimilation of marine phytoplankton. Limnol. Oceanogr. 17(3), 371–382.

Patiño, R., Dawson, D., VanLandeghem, M.M., 2014. Retrospective analysis of associations between water quality and toxic blooms of golden alga (Prymnesium parvum) in Texas reservoirs: Implications for understanding dispersal mechanisms and impacts of climate change. Harmful Algae 33, 1–11.

Pearl, H.W., Gardner, W.S., Havens, K.E., Joyner, A.R., McCarthy, M.J., Newell, S.E., Qin, B., Scott, J.T., 2016. Mitigating cyanobacterial algal blooms in aquatic ecosystems impacted by climate change and anthropogenic nutrients. Harmful Algae 54, 213–222.

Rashel, R.H., Patiño, R., 2017. Influence of genetic background, salinity, and inoculum size on growth of the ichthyotoxic golden alga (Prymnesium parvum). Harmful Algae 66, 97–104.

Rashel, R.H., Patiño, R., 2019. Growth response of the ichthyotoxic haptophyte, Prymnesium parvum Carter, to changes in sulfate and fluoride concentrations. PLoS ONE 14 (9), e0223266.

Roelke, D.L., Barkoh, A., Brooks, B.W., Grover, J.P., Hambright, K.W., LaClaire II, J.W., Moeller, P.D., Patiño, R., 2016. A chronicle of a killer alga in the west: ecology, assessment, and management of Prymnesium parvum blooms. Hydrobiologia 764, 29–50.

Solomon, C.M., Collier, J.L., Berg, G.M., Glibert, P.M., 2010. Role of urea in microbes in aquatic systems: a biochemical and molecular review. Aquat. Microb. Ecol. 59, 67–88.

Suttle, C.A., Chan, A.M., Fuhrman, J.A., 1991. Dissolved free amino acids in the Sargasso Sea: uptake and respiration rates, turnover times, and concentrations. Mar. Ecol. bog. Ser. 70, 189-199.

Vanlandeghem, M.M., Farooqi, M., Southard, G.M., Patiño, R., 2015. Spatiotemporal associations of reservoir nutrient characteristics and the Invasive, Harmful Alga Prymnesium parvum in West Texas. J. Am. Water Resour. Asso. 51, 487–501.

Wood, A.M., Everroad, R.C., Wingard, L.M., 2005. Measuring growth rates in microalgal cultures. In: Andersen, R.A (Ed.), Algal Culturing Techniques. Elsevier Academic Press, 598.

113

Texas Tech University, Rakib H. Rashel, May 2020

Tables

Table 4.1. Output of two-way ANOVA of data collected in this study. Growth indices examined in Prymnesium parvum cultures included exponential growth rate (r), maximum cell density, and early cell density. DFn, Response Effects DFd F-value p-value Effects of different ratios of inorganic r trial 1, 20 13.392 0.002 to organic nitrogen on growth of maximum 1, 20 golden alga (Organic nitrogen source: density 33.216 < 0.0001 Urea) early density 1, 20 3.566 0.074 r interaction 4, 20 0.38 0.82 maximum density 4, 20 1.883 0.153 early density 4, 20 1.16 0.358 r concentration 4, 20 0.849 0.511 maximum density 4, 20 208.334 < 0.0001 early density 4, 20 5.976 0.002 Effects of different ratios of inorganic r trial 1, 20 3.594 0.073 to organic nitrogen on growth of maximum 1, 20 golden alga (Organic nitrogen source: density 27.321 < 0.0001 Glycine) early density 1, 20 0.758 0.394 r interaction 4, 20 0.843 0.514 maximum density 4, 20 3.819 0.018 early density 4, 20 0.607 0.662 r concentration 4, 20 1.379 0.277 maximum density 4, 20 102.589 < 0.0001 early density 4, 20 4.97 0.006 DFn, degrees of freedom numerator; DFd, degrees of freedom denominator.

114

Texas Tech University, Rakib H. Rashel, May 2020

Table 4.2. Pearson partial correlation of maximum cell density with exponential growth rate (r) or early cell density in Prymnesium parvum cultures as a function of the relative concentration of inorganic and organic N. Partial Correlation with Maximum Cell Density Source of Correlated Control organic N Variable variable Pearson r p-value n Early r 0.646 <0.0001 30 density Urea Early r 0.652 <0.0001 30 density Early r 0.664 <0.0001 30 density Glycine Early r 0.788 <0.0001 30 density When r was the correlation variable, early cell density was used as a control variable, and vice versa. Data used for these analyses are those reported in figures 1 and 2. n, sample size.

115

Texas Tech University, Rakib H. Rashel, May 2020

Figures

Figure 4.1. Growth indices of Prymnesium parvum as a function of different molar ratios of inorganic to organic (urea) nitrogen in modified ASM. (A) Growth curves, (B) early cell density, (C) exponential growth rate (r), and (D) maximum cell density. Each time point or bar represents the mean (± SEM) of 6 replicates. Bars with the same letter codes do not differ significantly (Tukey’s HSD, p ˂ 0.05). Results of multiple comparisons shown in D were conducted separately for each trial, and statistical significances are separately shown for each in large and small case letters, respectively – for more detailed trial results see Supplemental Figure 1.

116

Texas Tech University, Rakib H. Rashel, May 2020

Figure 4.2. Growth indices of Prymnesium parvum as a function of different molar ratios of inorganic to organic (glycine) nitrogen in modified ASM. (A) Growth curves, (B) early cell density, (C) exponential growth rate (r), and (D) maximum cell density. Each time point or bar represents the mean (± SEM) of 6 replicates. Bars with the same letter codes do not differ significantly (Tukey’s HSD, p ˂ 0.05). Results of multiple comparisons shown in D were conducted separately for each trial, and statistical significances are separately shown for each in large and small case letters, respectively – for more detailed trial results see Supplemental Figure 3.

117

Texas Tech University, Rakib H. Rashel, May 2020

Supplemental Tables

Supplemental Figure 1. Maximum cell density of Prymnesium parvum as a function of different molar ratios of inorganic to organic (urea) nitrogen in modified ASM. (A) Maximum cell density for trial 1, (B) Maximum cell density for trial 2. Each bar represents the mean (± SEM) of 3 replicates. Bars with the same letter codes do not differ significantly (Tukey’s HSD, p ˂ 0.05).

118

Texas Tech University, Rakib H. Rashel, May 2020

Supplemental Figure 2. Transitional phase in log-transformed growth curve of Prymnesium parvum as a function of different molar ratios of inorganic to organic (urea) nitrogen in modified ASM.

119

Texas Tech University, Rakib H. Rashel, May 2020

Supplemental Figure 3. Maximum cell density of Prymnesium parvum as a function of different molar ratios of inorganic to organic (glycine) nitrogen in modified ASM. (A) Maximum cell density for trial 1, (B) Maximum cell density for trial 2. Each time point or bar represents the mean (± SEM) of 3 replicates. Bars with the same letter codes do not differ significantly (Tukey’s HSD, p ˂ 0.05).

120

Texas Tech University, Rakib H. Rashel, May 2020

CHAPTER Ⅴ

GROWTH OF THE HARMFUL HAPTOPHYTE, PRYMNESIUM PARVUM UNDER PAST, PRESENT, AND PROJECTED FUTURE AIR CONCENTRATIONS OF CARBON DIOXIDE

Abstract

Carbon dioxide is the primary source of carbon for photosynthetic fixation by

plants and algae. Because it is highly soluble in water, changes in air CO2 concentration

can lead to corresponding changes in dissolved CO2 concentration of surface waters. This

has led to concerns over the potential effects of the rising air CO2 concentration on

growth of harmful algae. Current knowledge suggests that these effects may be species-

specific and no information is available for Prymnesium parvum, a euryhaline haptophyte

that forms toxic blooms. This study determined and compared the effects on P. parvum

growth by air CO2 at concentrations of 280 (pre-industrial era), 400 (current), and 670 ppm (a projected scenario by 2100). Batch cultures were conducted in two different

media, Artificial Seawater Medium at salinity of 5 and Instant Ocean® at salinity of 30.