The Role of Mitochondrial Uncoupling in Temperature

THE ROLE OF MITOCHONDRIAL UNCOUPLING IN TEMPERATURE

RESPONSES IN ATLANTIC KILLIFISH, FUNDULUS HETEROCLITUS

Heather Jean Bryant

B.Sc., The University of British Columbia, 2015

A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF

THE REQUIREMENTS FOR THE DEGREE OF

MASTER OF SCIENCE

THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Zoology)

THE UNIVERSITY OF BRITISH COLUMBIA

(Vancouver)

February 2018

Abstract

Environmental temperature can greatly impact the functioning of ectothermic organisms through effects on mitochondria, which are crucial to aerobic metabolism.

Changes in temperature have the potential to influence mitochondrial ATP production and production of reactive oxygen species (ROS), both of which are influenced by the activity of the mitochondrial electron transport system, which generates the proton gradient necessary for mitochondrial ATP production. Thus, I hypothesized that ectothermic organisms have a mechanism for modulating the proton gradient in the face of changes in environmental temperature to maintain ATP production, and that this mechanism may act through uncoupling proteins (UCPs) which can cause a decrease in the proton gradient independent of the production of ATP.

Here, I investigate changes in UCPs and mitochondrial function following thermal acclimation in two populations of the eurythermal Atlantic killifish, Fundulus heteroclitus. I show that UCP mRNA expression is tissue-specific, changes with thermal acclimation, and differs between two populations of killifish. However, these changes vary depending on the isoform, tissue, and population (Chapter 2). I also demonstrate that changes in UCP function are not necessarily consistent with changes in mRNA expression in isolated liver and brain mitochondria, but that UCP function may differ in liver between the two populations

(Chapter 3). Cold-acclimated northern killifish increase liver mitochondrial capacity and coupling as indicated by increases in state III, respiratory control and ADP/O ratios (Chapter

3). Interestingly, I also observed increases in proton conductance in isolated liver mitochondria from cold-acclimated northern killifish as indicated by increased O2

ii consumption rate at a common membrane potential (Chapter 3). Mitochondrial properties in southern killifish did not differ with thermal acclimation.

Taken together, my data suggest that UCPs may play a role in thermal acclimation, although there is not a clear connection between UCP mRNA expression and function.

Furthermore, my data indicate that northern killifish may have a greater capacity to respond to low temperature acclimation than southern killifish, suggesting a potential role for adaptive variation in mitochondrial responses to temperature.

iii Lay Summary

Changes in environmental temperature can have large impacts on the physiology of ectothermic organisms, whose body temperatures closely match environmental temperature.

Many of the mechanisms involved in thermal responses occur at the level of the mitochondria, the organelle in the cell responsible for the production of the energy currency used to power biological processes. Uncoupling proteins (UCPs) are hypothesized to play a role in mitigating the negative side effects of thermal change in the mitochondria. I showed that UCPs may play a role in the response to thermal change in killifish. Furthermore, I demonstrated that populations of killifish that come from different local thermal conditions differ in their responses to temperature acclimation. Together, my data suggest that UCPs may be involved in responses to thermal change and that these responses may depend on geographic location, suggesting a putative role for mitochondria in thermal adaptation.

iv Preface

Chapter 2 is co-authored by Heather J. Bryant and Patricia M. Schulte. Experimental design was carried out by Heather J. Bryant and Patricia M. Schulte. Data collection, analysis, and write-up were performed by Heather J. Bryant. All work with animals was done in accordance with UBC ACC approved animal use protocol # A11-0732.

Chapter 3 is co-authored by Heather J. Bryant and Patricia M. Schulte. Experimental design was conducted by Heather J. Bryant and Patricia M. Schulte. Data collection, analysis, and write-up were performed by Heather J. Bryant. All work with animals was done in accordance with UBC ACC approved animal use protocol # A16-0028.

v Table of Contents

Abstract ...... ii

Lay Summary ...... iv

Preface ...... v

Table of Contents ...... vi

List of Tables ...... x

List of Figures ...... xi

List of Abbreviations ...... xii

Glossary ...... xiv

Acknowledgements ...... xv

Chapter 1: Introduction ...... 1

1.1 Effects of changes in temperature in ectothermic organisms ...... 1

1.1.1 Time-scales of temperature variation ...... 1

1.1.2 Processes influenced by temperature ...... 2

1.1.3 How does temperature constrain species ecology and distributions? ...... 3

1.1.4 Aerobic metabolism and the mitochondrion ...... 4

1.2 Temperature and mitochondria in ectotherms ...... 6

1.2.1 Effects of temperature on mitochondrial membranes ...... 6

1.2.2 Effects of thermal acclimation on mitochondrial respiratory capacity ...... 7

1.2.3 Effects of temperature on proton leak and mitochondrial coupling ...... 7

1.2.4 Effects of temperature on cellular oxidative damage ...... 9

1.3 Uncoupling proteins (UCPs) ...... 9

vi 1.3.1 Uncoupling protein 1 (UCP1 a.k.a. thermogenin) ...... 10

1.3.2 Uncoupling proteins 2 and 3 (UCP2 and UCP3) ...... 11

1.3.3 Other UCPs (4, 5, and 3-like) ...... 12

1.3.4 Mechanism of UCP proton transport ...... 13

1.3.5 Functional roles of UCPs ...... 13

1.4 Atlantic killifish as a model ...... 14

1.4.1 Atlantic killifish ...... 14

1.4.2 Thermal acclimation effects on killifish physiology ...... 15

1.4.3 Thermal acclimation effects on killifish mitochondrial function ...... 16

1.5 Thesis objectives ...... 17

Chapter 2: Thermal acclimation effects on uncoupling protein mRNA expression in two populations of Atlantic killifish, Fundulus heteroclitus ...... 18

2.1 Introduction ...... 18

2.2 Methods...... 20

2.2.1 Phylogenetic analysis ...... 20

2.2.2 Animals and temperature acclimations ...... 23

2.2.3 RNA isolations and cDNA synthesis ...... 24

2.2.4 Quantitative real-time PCR (qRT-PCR) ...... 24

2.2.5 Statistical analysis ...... 25

2.3 Results ...... 27

2.3.1 Phylogenetic analysis ...... 27

2.3.2 Tissue-specific mRNA expression ...... 27

vii 2.3.3 Thermal acclimation and population effects on Ucp isoform mRNA

expression ...... 31

2.3.3.1 Expression in brain tissue ...... 31

2.3.3.2 Expression in gill tissue ...... 33

2.3.3.3 Expression in liver tissue ...... 35

2.3.3.4 Expression in muscle tissue ...... 37

2.4 Discussion ...... 40

2.4.1 Phylogenetic relationships ...... 40

2.4.2 Tissue-specific mRNA expression ...... 41

2.4.3 Effects of thermal acclimation and population on Ucp mRNA expression .... 42

2.4.4 Conclusions ...... 50

Chapter 3: The effects of thermal acclimation on mitochondrial function and fatty-acid induced uncoupling in two populations of Atlantic killifish, Fundulus heteroclitus ...... 51

3.1 Introduction ...... 51

3.2 Methods...... 55

3.2.1 Animals ...... 55

3.2.2 Isolation of liver and brain mitochondria ...... 56

3.2.3 Mitochondrial respiration and membrane potential measurements ...... 57

3.2.4 Statistical analysis ...... 61

3.3 Results ...... 62

3.3.1 Mitochondrial respiration ...... 62

3.3.2 Thermal acclimation and population effects on mitochondrial membrane

potential…...... 68

viii 3.3.3 Effects of fatty acids on liver and brain mitochondria ...... 74

3.4 Discussion ...... 78

3.4.1 Respiratory capacity ...... 81

3.4.2 Liver proton leak kinetics ...... 84

3.4.3 Effects of palmitate on liver and brain mitochondria ...... 89

3.4.4 Conclusions ...... 91

Chapter 4: Discussion and conclusions ...... 93

4.1 Ucp mRNA expression and thermal acclimation ...... 93

4.2 Relationships between Ucp mRNA and function ...... 96

4.3 Mitochondrial responses to thermal acclimation between populations ...... 100

4.4 Future Directions ...... 101

4.5 Conclusions ...... 102

Bibliography ...... 104

Appendices ...... 118

Appendix A Supporting information ...... 118

Appendix B Whole animal data ...... 120

B.1 Chapter 2 whole animal parameters ...... 120

B.2 Chapter 3 whole animal parameters ...... 123

ix List of Tables

Table 2.1 Accession numbers for gene sequences used in phylogenetic analyses. Ensembl and NCBI gene IDs provided where available...... 22

Table 2.2 Primer pairs used for qRT-PCR...... 26

Table 2.3 p-values following Mann-Whitney U test in brain, gill and muscle and post-hoc Tukey HSD analysis for liver for tissue-specific Ucp expression in northern F. heteroclitus...... 30

Table 2.4 Ucp mRNA expression changes across tissues and acclimation temperatures for Ucp1, Ucp2, Ucp3, Ucp3L and Ucp5...... 43

Table 2.5 Ucp mRNA expression differences between populations for Ucp1, Ucp2, Ucp3, Ucp3L and Ucp5 across tissues...... 44

Table 3.1 Percent changes in state II with oligomycin (state IIol) in isolated rat brown adipose tissue mitochondria exposed to sodium palmitate, followed by GDP...... 59

Table 3.2 p-values for two-way ANOVA on whole animal measurements in thermally acclimated northern and southern F. heteroclitus...... 65

Table 3.3 p-values for post-hoc tests for respiration measurements in thermally acclimated northern and southern F. heteroclitus...... 65

Table 3.4 p-values for post-hoc tests for ADP/O in liver mitochondria from thermally acclimated northern and southern F. heteroclitus...... 67

Table 3.5 p-values for post-hoc tests for state II membrane potential and O2 consumption rate at the highest common membrane potential (HCmV, 142.4 mV) in liver mitochondria from thermally acclimated northern and southern F. heteroclitus...... 71

Table 3.6 p-values for post-hoc tests for O2 consumption rate at the highest common membrane potential (HCmV, 142.4 mV) in liver mitochondria from thermally acclimated northern and southern F. heteroclitus...... 71

Table 3.7 p-values for post-hoc tests for state II O2 consumption rate (A) and membrane potential (B) in brain mitochondria from thermally acclimated northern and southern F. heteroclitus...... 73

Table 3.8 p-values for post-hoc tests differences in O2 consumption rate at the highest common membrane potential (HCmV, 142.4 mV) or at the widest difference and for the membrane potential of liver mitochondria from thermally acclimated northern and southern F. heteroclitus...... 77

Table 3.9 FDR corrected p-values from one sample t-tests for the D O2 consumption rate at the highest common membrane potential control and palmitate treatments in isolated liver mitochondria from northern and southern F. heteroclitus across acclimation temperatures...... 77

Table 3.10 p-values for two-way ANOVA on O2 consumption rate and membrane potential measurements in brain mitochondria from thermally acclimated northern and southern F. heteroclitus...... 80

Table 3.11 p-values for post-hoc tests for O2 consumption rate and membrane potential in brain mitochondria from thermally acclimated northern and southern F. heteroclitus...... 80

x List of Figures

Figure 1.1 Schematic of the electron transport system (ETS) on the inner mitochondrial membrane (IMM)...... 5

Figure 2.1 Phylogenetic relationships among selected teleost and tetrapod Ucp gene sequences...... 28

-1 -1 Figure 3.1 O2 Consumption rate (nmol O2 min mg protein ) measurements for liver (A,C,E) and brain (B,D,F) mitochondria from northern (black circles, solid line) and southern (white squares, dashed line) Fundulus heteroclitus acclimated to 5, 15 and 25°C...... 64

Figure 3.2 ADP/O ratios for northern (black bars) and southern (white bars) F. heteroclitus acclimated to 5, 15 and 25°C...... 67

Figure 3.3 Thermal acclimation and population effects on liver mitochondrial proton leak kinetics in Fundulus heteroclitus...... 70

Figure 3.5 Proton leak kinetic curves for control (black circles, solid lines) and palmitate (35 µM, white squares, dashed lines) treatments measured in liver mitochondria from northern (A-C) and southern (D-F) Fundulus heteroclitus acclimated to 5 (A and D), 15 (B and E) and 25°C (C and F)...... 75

xi List of Abbreviations

12L:12D A light cycle with 12 hours of light and 12 hours of dark ADP Adenosine diphosphate ANOVA Analysis of variance ATP Adenosine triphosphate BLAST Basic local alignment search tool BSA Bovine serum albumin °C Degrees Celsius cDNA Complementary deoxyribonucleic acid

CTmax Critical thermal maximum

CTmin Critical thermal minimum CI Complex I of the electron transport chain, NADH dehydrogenase CII Complex II of the electron transport chain, succinate dehydrogenase

CIII Complex III of the electron transport chain, CoQH2-cytochrome c reductase CIV Complex IV of the electron transport chain, cytochrome c oxidase CV Complex V of the electron transport chain, ATP-synthase EGTA Ethylene glycol-bis(β-aminoethyl ether)-N,N,N’,N’-tetraacetic acid ETS Electron transport system FDR False discovery rate g Grams GDP Guanosine diphosphate GTP Guanosine triphosphate h hours HSI Hepatosomatic index HEPES 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid IMM Inner mitochondrial membrane l litres min minutes M Molar mg Milligrams µmol l-1 Micromolar µl Microliters mRNA Messenger ribonucleic acid mV millivolts NCBI National Centre for Biotechnology Information OCLTT Oxygen- and capacity-limited thermal tolerance hypothesis pmol Picomoles ppt Parts per thousand

xii Q10 Temperature coefficients qRT-PCR Quantitative real time polymerase chain reaction ROS Reactive Oxygen Species s Seconds S.E.M. Standard error of the mean

xiii Glossary

State II Conditions under which mitochondria are supplied with saturating substrates that donate electrons to the electron transport chain. No ADP is present and therefore not being phosphorylated. Any O2 consumption may be used as an estimate of leak respiration.

State III Conditions under which mitochondria are supplied with saturating substrates that donate electrons to the electron transport chain and a saturating amount of ADP is present. O2 consumption under these conditions is considered to be ADP-phosphorylating respiration.

State IV Conditions under which mitochondria are supplied with saturating substrates and saturating ADP, but ADP is not being phosphorylated to ATP either because the ATP-synthase has been inhibited (induced state IV) or all the ADP has been consumed (natural state IV). As with State II respiration, any O2 consumption may be used as an estimate of leak respiration.

xiv Acknowledgements

I want to express my gratitude towards my supervisor, Patricia Schulte. Her enthusiasm, guidance, and unwavering support have not only helped me progress as a scientist but have inspired me to always strive to continue learning and enjoy my scientific career. I also want to thank the members of the Schulte lab: Marina Giacomin, Taylor

Gibbons, Tara McBryan, Dave Metzger, Sara Northrup, and Xiang Lin for all their help and advice, and Dillon Chung and Tim Healy for their incredible mentorship during my time in the lab.

Thank you to my committee members, Jeff Richards and Colin Brauner for their perspectives, advice and encouragement. I also want to acknowledge the Department of

Zoology and my friends and peers within it, who have provided fantastic support and guidance throughout my degrees.

Thank you to my friends and family, without whom I would not be where I am today.

To my parents, your influence and support have not only helped me achieve my goals, but have allowed me the freedom to enjoy the journey. To Jackson, thank you for your continued support in everything I do, for bringing me dinner when I spent those late nights in the lab, and for continually reminding me to live life to the fullest.

xv Chapter 1: Introduction

Environmental temperature is a key abiotic factor that can greatly influence the physiological functioning of ectothermic organisms, whose body temperatures generally match that of the environment. In this thesis, I explore the role of changes in mitochondrial properties in response to changes in environmental temperature in ectotherms. I investigate these changes in response to both thermal acclimation and putative local adaptation. In the remainder of this general introduction I provide background and context on the effects of thermal change on ectothermic organisms in general and more specifically on mitochondrial function. I focus on the potential roles of uncoupling proteins (UCPs) in thermal responses in ectotherms. I conclude with the discussion of the biology of Atlantic killifish, Fundulus heteroclitus, which is the model organism I used to accomplish my thesis objectives.

1.1 Effects of changes in temperature in ectothermic organisms

1.1.1 Time-scales of temperature variation

Variation in temperature can occur over a range of time scales. For example, in many environments temperatures fluctuate diurnally and are typically warmer during the day than at night (Qu et al., 2014). Similarly, temperature can also vary seasonally, particularly at higher latitudes (Chan et al., 2016). There are also much longer-term changes in temperature that occur over geological time-scales due to the Milankovitch cycles that impact the advance and retreat of glaciers (Hays et al., 1976). When ectotherms experience variation in ambient temperature, they undergo physiological changes to maintain homeostasis. However, temperature fluctuations at different time-scales may require ectotherms to respond in different ways (Dillon et al., 2016). For example, seasonal changes in temperature may be associated with acclimation or acclimatization (in the lab or field, respectively), whereas

1 changes in temperature across longer time-scales may cause species to adapt. Characterizing the potential mechanisms by which ectotherms undergo these changes and maintain physiological functioning in the face of changing temperature has long been of interest to researchers and is currently of particular importance in order to develop predictions for both short- and long-term effects of global climate change on animals and their ecosystems

(Deutsch et al., 2015; Pörtner & Knust, 2007; Pörtner & Farrell, 2008).

1.1.2 Processes influenced by temperature

Temperature affects both the stability of biological structures and the rates of biochemical reactions (Hochachka & Somero, 2002). With respect to proteins, increased temperature is expected to reduce the stability of weak inter-molecular interactions such as hydrogen bonds, thus decreasing protein stability and potentially altering protein conformation and activity (Hochachka & Somero, 2002). With respect to biological membranes, increased temperature has been shown to increase membrane fluidity, which can affect the activity of integral membrane proteins (Hazel & Prosser, 1974). These processes at the molecular level are thought to be integrated across levels of biological organization to influence processes at the population, species, and ecosystem levels (Schulte, 2015). Because of these profound effects of temperature, many organisms have the capacity to alter their physiology to compensate for the acute effects of temperature through acclimation. Reviews examining the impacts of rising temperatures on coastal marine ecosystems have concluded that species’ distributions and abundances will greatly depend on their abilities to tolerate and adapt to changing temperatures, through either existing phenotypic plasticity or evolutionary responses (Harley et al., 2006). As a result, physiological responses to thermal acclimation have often been used as proxies to understand how organisms may respond to

2 changes in ambient environmental temperature (e.g., Stillman, 2003). Thermal acclimation has been previously shown to affect organismal performance, altering physiological processes such as mitochondrial function (e.g., Abele et al., 2002; Chung & Schulte, 2015;

Chung et al., 2017) and cardiac performance (e.g., Franklin et al., 2007; Stenseng et al.,

2005) as well as whole-organism phenotypes such as thermal limits, oxygen consumption and hypoxia tolerance (McBryan et al., 2016; Pörtner, 2001; 2010).

1.1.3 How does temperature constrain species ecology and distributions?

The physiological processes of many ectothermic organisms are expected to change in response to large-scale variation in environmental temperature. For instance, thermal tolerance breadth has been found to increase with increasing latitudes in ectothermic organisms (Sunday et al., 2011), indicating that differences in thermal tolerance may be found within and between species found across different latitudes. Species’ range limits are posited to be determined (at least in part) by the constraints on aerobic scope (the difference between maximum and routine oxygen consumption) at thermal extremes. This concept is known as the oxygen- and capacity- limited thermal tolerance (OCLTT) hypothesis (Pörtner,

2001; 2010; Pörtner & Farrell, 2008). Consistent with this hypothesis, aerobic scope has been shown to decrease upon exposure to temperatures representative of the edges of a species’ thermal range (e.g., Eliason et al., 2011; Healy & Schulte, 2012). Not only are species’ ranges postulated to be rooted in the effects of temperature on aerobic metabolism, but the metabolic theory of ecology postulates that differences in many ecological traits can also be attributed to the effects of temperature (combined with effects of body mass) on metabolic rate (Brown et al., 2004). Based on the OCLTT hypothesis and the metabolic theory of ecology, understanding the mechanisms through which temperature affects aerobic

3 metabolism in ectotherms will contribute to a greater ability to predict the effects of climate changes on ectothermic organisms.

1.1.4 Aerobic metabolism and the mitochondrion

One of the key cellular components involved in aerobic metabolism, and therefore potentially involved in determining whole animal thermal limits, is the mitochondrion

(Pörtner, 2001). A key functional component of the mitochondrion is the electron transport system (ETS), which is a series of membrane-spanning or membrane-associated proteins located on the inner mitochondrial membrane (IMM) (Figure 1.1). The ETS accepts electrons derived from the oxidation of nutrients. These electrons are then passed along the ETS to the final electron acceptor, O2, which combines with protons to form metabolic H2O. As a result of the electron transport process, protons (H+) are pumped from the mitochondrial matrix through to the intermembrane space establishing a proton gradient across the IMM (Figure

1.1). The proton motive force consists of a concentration gradient (pH gradient) and an associated electrical gradient (membrane potential), the latter of which is the primary driver of the proton motive force across the IMM (discussed in Nicholls & Ferguson, 2013). The

ATP synthase, a membrane-spanning enzyme in the IMM, harnesses the potential energy contained within the proton motive force in order to drive the synthesis of ATP from ADP and inorganic phosphate (Senior et al., 2002; Figure 1.1). Taken together, this process is known as oxidative phosphorylation because the oxidation of nutrients by the mitochondria requires O2 as the final electron acceptor along the ETS. In most animals, the majority of cellular ATP, the primary energy currency of the cell (Knowles, 1980), is produced through oxidative phosphorylation.

Intermembrane Space

H+ H+ H+ H+ H+

I III Cyt c IV V UCP

e- IMM Q - II e e- e-

FADH2 NAD+ NADH 2H2O FAD O2

- O2 O2 ROS ATP ADP + Pi

Mitochondrial Matrix

Figure 1.1 Schematic of the electron transport system (ETS) on the inner mitochondrial membrane (IMM). Substrates donate electrons to complexes I and II + and electrons are passed from complex to complex to the terminal electron acceptor (O2). Through this process, protons (H ) are translocated from the mitochondrial matrix through to the intermembrane space, establishing the proton gradient and membrane potential across the IMM; together these make up the proton motive force. This proton motive force is then used by the ATP synthase in order to produce ATP from ADP and inorganic phosphates (Pi). Uncoupling proteins (UCPs) allow for the movement of protons from the intermembrane space through to the matrix, independent of the ATP synthase (complex V) and the generation of ATP. Under situations when the proton motive force is high, some reactants are unable to be oxidized and are therefore able to reduce O2 to form reactive oxygen species (ROS).

5 1.2 Temperature and mitochondria in ectotherms

1.2.1 Effects of temperature on mitochondrial membranes

Cellular ATP production by mitochondria is highly dependent on the functioning of

ETS proteins and the maintenance of the proton motive force across the IMM. Given the semi-fluid nature of cellular membranes (Singer & Nicolson, 1972), temperature would be expected to influence the fluidity and structure of the IMM. Acute exposures to cold temperatures reduce membrane fluidity, while acute exposure to warm temperatures increases membrane fluidity. As a compensatory response, thermal acclimation typically induces membrane remodelling that results in increased fluidity during acclimation to colder conditions and decreased fluidity during acclimation to warmer conditions, a process known as homeoviscous adaptation (Sinensky, 1974). These changes in fluidity are often associated with changes in the concentrations of saturated and unsaturated fatty-acids in the membrane, with higher concentrations of saturated fatty acids following exposure to higher temperatures

(Sinensky, 1971). Consistent with the possibility of homeoviscous adaptation, several studies have demonstrated that mitochondrial phospholipid composition and fluidity change with temperature, with a focus on increases in fatty-acid unsaturation and fluidity following cold acclimation (e.g., Cossins & Prosser, 1982; Dahlhoff & Somero, 1993; Grim et al., 2010;

Hazel & Zerba, 1986; Itoi et al., 2003; Kraffe et al., 2007; Miranda & Hazel, 1996; Roy et al., 1992). Alterations to the structure and fluidity of membranes can greatly affect the functioning of embedded or associated enzymes (e.g., Itoi et al., 2003; Kraffe et al., 2007). In the case of the mitochondrion, ETS complexes may be affected by changes to the IMM.

Interestingly, changes in mitochondrial membrane composition with cold acclimation have been shown to affect the activity of ETS complex II in goldfish (Hazel, 1972). Combined

6 with the effects of temperature on the stability and reaction rates of enzymes, changes in the

IMM due to thermal acclimation may affect mitochondrial aerobic capacity.

1.2.2 Effects of thermal acclimation on mitochondrial respiratory capacity

Thermal acclimation can have large effects on the mitochondrial respiratory capacity of ectothermic organisms. The responses of mitochondrial respiratory capacity to thermal acclimation can vary depending on the thermal strategy (i.e., stenotherm or eurytherm) and geographic location (i.e., polar, temperate or tropical) of the species under investigation. In fish, cold acclimation has been associated with increases in mitochondrial capacity in conjunction with changes in membrane composition, potentially as compensation for thermal effects on biochemical reaction rates (Chung & Schulte, 2015; Fangue et al., 2009b; Grim et al., 2010; Lannig et al., 2005; St-Pierre et al., 1998). Cold acclimation has also been found to increase markers of mitochondrial volume density in fishes (Dhillon & Schulte, 2011;

Pörtner, 2001). However, increases in mitochondrial function in the cold are not seen consistently (e.g., Chung et al., 2017) and there is evidence for active metabolic suppression following prolonged cold exposure in some species of ectotherms (Campbell et al., 2008;

Richards, 2010). There is also evidence of decreases in mitochondrial function and respiratory capacity with increases in temperature in ectotherms (Chung & Schulte, 2015;

Chung et al., 2017; Kraffe et al., 2007; Lannig et al., 2005).

1.2.3 Effects of temperature on proton leak and mitochondrial coupling

When functioning optimally, the movement of protons to the mitochondrial matrix is mostly coupled to the production of ATP. This process is not perfectly coupled and protons leak through the IMM without resulting in the generation of ATP (Guderley, 2011), a process that can be highly temperature-sensitive (Hardewig et al., 1999; Portner et al., 1999; Pörtner,

7 2001). Basal proton leak contributes to 20-50% of cellular, tissue or whole-animal oxygen consumption rate (Brand 1990; Brand et al., 1994). A large portion of proton leak is due to mitochondrial carrier proteins such as the adenine nucleotide translocase (ANT) or uncoupling proteins (UCPs) (Azzu et al., 2010). Mitochondrial leak respiration is generally estimated under conditions where ATP synthesis has been halted, for example in the absence of ADP or in the presence of ATP synthase blockers, such as oligomycin (state II or state

IV). Under these conditions mitochondrial O2 consumption is indicative of the ETS activity necessary to maintain membrane potential despite proton leak. Therefore, the rate of O2 consumption without the production of ATP indicates the degree of uncoupling of the mitochondria. Leak respiration has been shown to increase following both warm (Hardewig et al., 1999; Portner et al., 1999) and cold (Chung & Schulte, 2015) acclimation in ectothermic organisms. ADP-phosphorylating respiration (i.e., state III) is generally used as an indicator of mitochondrial oxidative phosphorylation capacity. The respiratory control ratio (RCR) is calculated as state III/state IV and is an estimate of mitochondrial coupling, where higher RCRs indicate greater coupling. The relative changes of ADP-phosphorylating

(state III) respiration to leak respiration (state IV) can results in differences in mitochondrial coupling as a result of thermal acclimation (e.g., Chung & Schulte, 2015; Hardewig et al.,

1999; Portner et al., 1999). However, if only respiration rate is measured in order to determine proton leak, changes in mitochondrial O2 consumption rate due to changes in substrate oxidation may be misinterpreted as changes in proton conductance (Divakaruni &

Brand, 2011). To accurately characterize proton leak or proton conductance through the

IMM, it is arguably necessary to measure proton leak kinetics by simultaneously decreasing mitochondrial respiration and membrane potential (Divakaruni & Brand, 2011).

8 1.2.4 Effects of temperature on cellular oxidative damage

Mitochondria are a major site of reactive oxygen species (ROS) production, which can cause damage to cellular structures and impair cellular function. Ultimately, this damage may be associated with ageing (Brand, 2000). Mitochondrial ROS production can be highly dependent on the proton motive force (Skulachev, 1998). Under situations where the proton motive force is high, components of the ETS become completely reduced and unable to oxidize reactants, which then have time to reduce O2 without a catalyst thereby creating ROS

(Skulachev, 1998; Figure 1.1). Given the effects of temperature on the mitochondrial proton motive force it is predicted that ambient temperature will also influence ROS production. In fact, increases in ROS production with increases in temperature have been demonstrated in ectothermic species (e.g., Heise et al., 2003). At colder temperatures, where there is an increase in oxidative capacity there can be a corresponding increase in the production of ROS

(Guderley, 2004). In addition, others have postulated that the cell is more susceptible to lipid peroxidation, a consequence of ROS damage, at colder temperatures due to an increase in fatty acid unsaturation in the membrane (Grim et al., 2010). Provided that cellular oxidative damage and ATP production are both tightly linked to the maintenance of the proton motive force across the IMM, the ability to adjust the proton motive force to maintain homeostasis and mitigate ROS production is predicted to be highly beneficial in response to acute thermal effects (Azzu et al., 2010). This may be particularly important when organisms experience longer thermal exposures, which can have large effects on mitochondrial physiology.

1.3 Uncoupling proteins (UCPs)

Uncoupling proteins (UCPs) are a family of nuclear-encoded, membrane-spanning proteins that are responsible for the movement of protons (H+) from the intermembrane space

9 through to the matrix in the mitochondria of many eukaryotic animals (Figure 1.1) (Hughes

& Criscuolo, 2008; Klingenberg & Winkler, 1985; Woyda-Ploszczyca & Jarmuszkiewicz,

2017). UCPs were first identified by the characterization of UCP1 in hamster brown adipose tissue (Nicholls et al., 1978). Following the characterization of UCP1, four other uncoupling proteins have since been described (UCP2, UCP3, UCP4/SLC25A27 and

UCP5/SLC25A14/BMCP1) (Azzu et al., 2010; Bouillaud et al., 2001; Kwok et al., 2010).

Uncoupling proteins have been identified across the animal kingdom in various species including mammals, birds, fish, and invertebrates, and they have even been identified in plants (for a review see: Woyda-Ploszczyca & Jarmuszkiewicz, 2017). The relationship among the UCP gene sequences is well described, with UCPs 2 and 3 being the most closely related (Azzu et al., 2010; Tine et al., 2012). While the mitochondrial uncoupling function appears to be common among the isoforms, there is variation in the tissues they are expressed in, their abundance and potentially their functional role.

1.3.1 Uncoupling protein 1 (UCP1 a.k.a. thermogenin)

The best characterized uncoupling protein isoform is UCP1, or thermogenin, named for its role in heat production in mammalian brown adipose tissue (BAT, Nicholls, 2001).

This protein has been shown to play an important role in increasing body temperature in neonatal and hibernating mammals (Ballinger et al., 2016; Cannon & Nedergaard, 2004;

Nicholls & Locke, 1984) and has been implicated in the evolution of cold-tolerance.

However, UCP1 has not only been characterized in mammalian brown adipose tissue, but also in a variety of tissues in other taxa including non-mammalian endotherms such as birds, ectotherms such as fishes and invertebrates, and even non-animal taxa (Hughes & Criscuolo,

2008; Woyda-Ploszczyca & Jarmuszkiewicz, 2017). In ectothermic taxa such as fishes,

10 UCP1 has mostly been found in liver and brain and has also been shown to be responsive to temperature (Jastroch et al., 2005; Murakami et al., 2015; Wen et al., 2015), although the functional relevance of this temperature response is still being investigated (e.g., Jastroch et al., 2007). UCP1 is known to be activated by fatty-acids and the products of ROS such as reactive aldehydes and inhibited by purine nucleotides, such as GDP and GTP (Woyda-

Ploszczyca & Jarmuszkiewicz, 2017). These activators and inhibitors are known to influence other mitochondrial proteins, such as the ANT and therefore their specificity and use in functional assays has been debated (e.g., Woyda-Ploszczyca & Jarmuszkiewicz, 2014).

1.3.2 Uncoupling proteins 2 and 3 (UCP2 and UCP3)

UCP2 was identified in 1997 (Fleury et al., 1997) and UCP3 by multiple groups shortly thereafter (Gong et al., 1997; Liu et al., 1998; Matsuda et al., 1997; Vidal-Puig et al.,

1997). The protein sequences of UCPs 2 and 3 share > 70% identity to each other (Riquier &

Bouillaud, 2000). In humans, UCPs 2 and 3 are approximately 59% and 57% similar to

UCP1, respectively (Fleury et al., 1997; Vidal-Puig et al., 1997). UCP2 appears to be expressed ubiquitously across tissues, at least at low levels, whereas UCP3 is much more tissue-specific, appearing at relatively high levels in skeletal muscle in mammals and fish

(Riquier & Bouillaud, 2000; dos Santos et al., 2013; Wen et al., 2015). The avian UCP also shares approximately 70% similarity with UCPs 2 and 3 and shows a fairly ubiquitous tissue expression similar to mammalian UCP2 (Emre et al., 2007). UCPs 2 and 3 are thought to be involved in mitigating the production of ROS in the mitochondria, but their exact functions are still debated (Brand & Esteves, 2005), and some studies have suggested that they may have a thermogenic function under certain biochemical scenarios (Esteves & Brand, 2005;

Gong et al., 1997). Similar to UCP1, mammalian UCPs 2 and 3 have also been shown to be

11 activated by fatty acids, superoxides and reactive alkenals and inhibited by purine nucleotides (Esteves & Brand, 2005). In addition, genepin has also been found to be an inhibitor of UCP2 and also possibly UCP3 (Zhang et al., 2006). However, not all studies on

UCP3 across species have found that this protein is sensitive to purine nucleotide inhibitors.

For example, UCP3 in goldfish skeletal muscle was unable to be inhibited by purine nucleotides (dos Santos et al., 2013). Together with UCP1, UCPs 2 and 3 are said to make up the core UCP family (Cannon & Nedergaard, 2004).

1.3.3 Other UCPs (4, 5, and 3-like)

Other UCP isoforms do exist across eukaryotes. Lamprey UCP, thought to be the ancestral UCP isoform, has similar properties to other UCP isoforms but is insensitive to inhibition by purine nucleotides (Wang et al., 2010). In mammals, the isoforms UCP4 and

UCP5 were identified in 1999 (Mao et al., 1999) and 1998 (Sanchis et al., 1998). Despite being rather dissimilar to the core UCP protein family, studies in mammals suggest that

UCP4 and 5 act in similar manners as other UCPs, causing increased proton leak and uncoupling of oxidative phosphorylation (Kwok et al., 2010; Ramsden et al., 2012).

Furthermore, there is some evidence the proton conductance by UCPs 4 and 5 can also be activated by fatty-acids and inhibited by purine nucleotides (Hoang et al., 2012; D. Liu et al.,

2006), although fewer studies on these isoforms have been conducted. A fish-specific UCP

(UCP3-like), which is the result of a genome duplication event, has also been identified (Tine et al., 2012) although to my knowledge the tissue distribution, regulation and function of this protein have yet to be investigated.

12 1.3.4 Mechanism of UCP proton transport

UCPs cause the movement of protons from the intermembrane space through to the mitochondrial matrix (Figure 1.1). The exact mechanism of H+ movement as a result of

UCPs is debated, but theories include: a simple proton pore, fatty acids as cofactors for transport, transport of anionic fatty acids to the intermembrane space which then become protonated and move back through to the matrix, fatty acids that act as allosteric modifiers of

UCPs that overcome nucleotide inhibition, and coenzyme Q acting as a facilitator or mediator of proton transport by the UCPs (Bouillaud et al., 2001; Brand & Esteves, 2005;

Divakaruni & Brand, 2011; Garlid et al., 1996; Klingenberg, 2001; Riquier & Bouillaud,

2000; Shabalina et al., 2004; Woyda-Ploszczyca & Jarmuszkiewicz, 2017). Inhibition by nucleotides is based on a direct binding to a specific site on the UCP (Bouillaud et al., 1994;

Klingenberg, 2001; Riquier & Bouillaud, 2000).

1.3.5 Functional roles of UCPs

Aside from the thermogenic function of UCP1 in mammalian BAT, the functional role of other UCP isoforms and all UCPs in ectothermic species are unknown (e.g., Brand &

Esteves, 2005; Jastroch et al., 2005). One of the most common hypotheses is that the ancestral function of UCPs may be to mitigate the production of ROS and the accumulation of oxidative damage by increasing proton movement from the intermembrane space through to the mitochondrial matrix (Brand, 2000; Brand & Esteves, 2005; Brand et al., 2004; Echtay et al., 2002a; Echtay et al., 2002b; Esteves & Brand, 2005). Building from this hypothesis, I predicted that UCPs in ectothermic species should be regulated in response to thermal acclimation, which can have large effects on the production of and susceptibility to ROS, as mentioned above. The exact changes with thermal acclimation will be highly dependent on

13 whole-organism responses to temperature (i.e., metabolic compensation and up-regulation of mitochondrial function or metabolic suppression and down-regulation of function). In fact, studies in fish have found the mRNA expression and function of UCP isoforms change in response to thermal acclimation, which may be consistent with changes expected for the mitigation of reactive oxygen species (e.g., Jastroch et al., 2005; Jastroch et al., 2007; Mark et al., 2006; Murakami et al., 2015; Tseng et al., 2011; Wen et al., 2015). However, few studies have explored the potential for the regulation of UCPs during thermal acclimation and their putative role in thermal adaptation (e.g., Mark et al., 2006). Atlantic killifish,

Fundulus heteroclitus, present an excellent model species in which to explore the potential impacts of thermal acclimation and putative adaptation on physiological processes generally, and UCPs specifically.

1.4 Atlantic killifish as a model

1.4.1 Atlantic killifish

Atlantic killifish, Fundulus heteroclitus, are abundant topminnows (Taylor, 1999) that inhabit intertidal salt marshes on the east coast of North America from the Gulf of the St.

Lawrence, Canada south to Florida, USA. Across this range there are two subspecies: the northern subspecies, F. heteroclitus macrolepidotus, ranging from New Jersey, USA northward and the southern population, F. heteroclitus heteroclitus ranging from New Jersey,

USA southward (Morin & Able, 1983). There is substantial genetic divergence between the two subspecies, with many studies demonstrating a sharp transition between alleles from the two subspecies along the coast (Adams et al., 2006; Bernardi et al., 1993; Cashon et al.,

1981; Powers & Place, 1978). Life in a temperate intertidal zone is associated with large diurnal variations in ambient temperature due to the tidal cycle (Helmuth & Hofmann, 2001)

14 and large seasonal variations in temperature, where winter may be 15-20 °C lower than summer temperatures at any given latitude (Fangue et al., 2006). Between the northern and southern ends of the species range, there is approximately a 13 °C difference in mean monthly temperature at any given time of year (Fangue et al., 2006) and because killifish home ranges are relatively small (Fritz et al., 1975; Lotrich, 1975), populations at the ends of the species range experience quite different thermal regimes year-round. Therefore, F. heteroclitus provide a system in which we can address thermal responses on different time- scales in an ectothermic species, including the responses to thermal acclimation and the potential for adaptation to life in different thermal regimes.

1.4.2 Thermal acclimation effects on killifish physiology

Northern and southern killifish are extremely eurythermal and have been shown to exhibit intraspecific variation in their thermal preferences, thermal tolerances and responses to thermal acclimation (Chung & Schulte, 2015; Chung et al., 2017; Fangue et al., 2006;

Fangue et al., 2009a; Fangue et al., 2009b; Healy & Schulte, 2012; Healy et al., 2017;

McBryan et al., 2016). Despite living in colder habitats, northern killifish have been shown to prefer warmer temperatures than southern killifish demonstrating counter-gradient variation

(Fangue et al., 2009a). Consistent with latitudinal variation in habitat temperatures, northern killifish tend to have lower critical thermal minimum (CTmin) temperatures and southern killifish tend to have higher critical thermal maximum (CTmax) temperatures (Fangue et al.,

2006). Both populations of killifish have the ability to acclimate to a wide range of temperatures and intraspecific variation in CTmin and CTmax is generally maintained following acclimation (Fangue et al., 2006). Thermal acclimation has been shown to influence killifish whole-animal O2 consumption rates and aerobic scopes in ways that are

15 mostly consistent with the OCLTT hypothesis (Healy & Schulte, 2012). Furthermore, high temperature acclimation has been shown to improve hypoxia tolerance between northern and southern killifish at high temperatures (McBryan et al., 2016). In general, northern fish have shorter times to loss of equilibrium (LOE) in hypoxia than southern fish (McBryan et al.,

2016), which is consistent with the higher routine O2 consumption in northern fish than in southern fish assayed at a common temperature (Fangue et al., 2009b).

1.4.3 Thermal acclimation effects on killifish mitochondrial function

Consistent with differences in whole-animal O2 consumption rate, northern killifish have been shown to have greater liver mitochondrial maximal O2 consumption rate than southern killifish particularly following cold-acclimation (Fangue et al., 2009b). This increase in mitochondrial functional capacity in the northern population in the cold may be consistent with cold-compensation of mitochondrial function (Chung & Schulte, 2015;

Fangue et al., 2009b). However, in heart and brain permeabilized tissue preparations, there were very few differences in mitochondrial O2 consumption found between populations

(Chung et al., 2017). Furthermore, in contrast to the observation for the liver, there were decreases in maximal mitochondrial O2 consumption following cold acclimation (Chung et al., 2017). In addition to the observed increases in oxidative capacity of isolated mitochondria, there is also some evidence for increases in mitochondrial volume density following cold-acclimation in the northern population in muscle tissue (Dhillon & Schulte,

2011) and potentially in liver tissue as well (Fangue et al., 2009b). Furthermore, thermal acclimation effects on mitochondrial coupling as estimated by RCRs have been observed in

F. heteroclitus isolated liver mitochondria and brain and heart permeabilized tissue preparations (Chung & Schulte, 2015; Chung et al., 2017). In northern killifish liver

16 mitochondria, there were no differences detected in the production of ROS with thermal acclimation (Chung & Schulte, 2015); however, this may be due to the activation of antioxidant mechanisms in response to changes in ROS production that may occur earlier in the thermal exposure.

F. heteroclitus are eurythermal and can be acclimated to a wide range of temperatures, have physiological processes that are affected by thermal acclimation and differ between subspecies, and have mitochondrial properties that are responsive to thermal acclimation. These aspects make Atlantic killifish an ideal model in which to investigate the potential role of mitochondrial UCPs in response to thermal acclimation and putative thermal adaptation.

1.5 Thesis objectives

The objectives of my thesis were to 1) identify Ucp isoform gene sequences in F. heteroclitus, 2) determine in which tissues mRNA from each of the Ucp isoforms is expressed, 3) characterize changes in mRNA expression of the Ucp isoforms in response to acclimation to both relatively warm and cold temperatures and between a northern and a southern population of F. heteroclitus, 4) characterize changes in mitochondrial respiratory capacity in the liver and brain tissue with thermal acclimation (both warm and cold) and between a northern and a southern populations of F. heteroclitus, 5) determine the effects of thermal acclimation (both warm and cold and population on proton leak kinetics in liver tissue from F. heteroclitus, and 6) determine the effects of fatty-acids on uncoupling in mitochondria across acclimation temperatures and between populations in liver and brain tissue of F. heteroclitus.

17 Chapter 2: Thermal acclimation effects on uncoupling protein mRNA expression in two populations of Atlantic killifish, Fundulus heteroclitus

2.1 Introduction

Uncoupling proteins (UCPs) are membrane-spanning proteins found in the inner membrane of mitochondria across a wide variety of eukaryote taxa (Woyda-Ploszczyca &

Jarmuszkiewicz, 2017). Although the functions of these proteins are not fully understood, they are known to result in the movement of protons (H+) from the intermembrane space to the mitochondrial matrix, which decreases the proton motive force responsible for ATP generation in the mitochondria (Hughes & Criscuolo, 2008; Klingenberg & Winkler, 1985;

Woyda-Ploszczyca & Jarmuszkiewicz, 2017). Many different isoforms of UCPs exist, with the core and most well characterized family of UCP isoforms consisting of UCP1, UCP2 and

UCP3 (Azzu et al., 2010; Krauss et al., 2005). UCP1, or thermogenin, is the best-studied

UCP isoform. UCP1 is found in mammalian brown adipose tissue (BAT), where its primary function is thought to be thermogenesis, which has traditionally been implicated in the ability of mammals to cope in cold environments (Fedorenko et al., 2012; Gaudry et al., 2017;

Hughes et al., 2009). However, the other UCP family members are not thought to play a key role in thermogenesis, and their functions remain debated (Brand & Esteves, 2005; Esteves &

Brand, 2005).

UCPs, including UCP1, are also found in ectothermic animal taxa (e.g., Jastroch et al., 2005; Tine et al., 2012; Tseng et al., 2011; Wen et al., 2015), suggesting that the ancestral function of UCPs is unlikely to be thermogenesis. In general, UCPs are known to be activated by free fatty acids, aldehydes, retinoids and superoxides and inhibited by purine nucleotides (Brand et al., 2004; Considine et al., 2003; Echtay et al., 2003; Echtay et al.,

18 2002a; Echtay et al., 2002b; Woyda-Ploszczyca & Jarmuszkiewicz, 2014; 2017). The activation of UCPs by superoxides suggests that UCPs may play a role in the regulation of reactive oxygen species (ROS). Indeed, the production of superoxides likely provides a negative feedback signal resulting in a decrease of the proton motive force and in turn decreased production of free radicals (Brand et al., 2004; Echtay et al., 2002b). Thus, UCPs may play a role in regulating the mitochondrial proton motive force and protecting against the formation of ROS in response to changing cellular conditions (Echtay et al., 2002b).

In ectotherms, exposure to thermal extremes is known to result in an increase in oxidative damage (e.g., Abele et al., 2002; Crockett, 2008; Keller et al., 2004; Stier et al.,

2014). This is because temperature affects mitochondrial properties such as enzyme activity, membrane composition and proton leak, as well as the capacity for oxidative phosphorylation

(Guderley, 2011). As a consequence of these temperature effects on mitochondrial function, variation in temperature can also influence the production of ROS. Given that UCPs may play a role in the regulation of ROS production, this suggests that UCPs could be involved in thermal acclimation in ectotherms such as fish.

Consistent with this hypothesis, increases in Ucp mRNA expression in fish brains have been associated with and inferred to be a mechanism for attenuating the increased ROS production associated with exposure to colder temperatures (Jastroch et al., 2007; Kwok et al., 2010; Tseng et al., 2011). As ROS generation (Korshunov et al., 1997) and oxidative phosphorylation are both highly dependent on mitochondrial membrane potential, one might expect that any changes to mitochondrial membrane potential associated with temperature acclimation would cause corresponding changes in the regulation of UCPs in order to

19 mitigate these effects. However, there is currently little data available with which to assess the potential role of UCPs in thermal acclimation responses in ectotherms.

In this study, we characterized the effects of thermal acclimation on mRNA expression of five UCP isoforms in two populations of Atlantic killifish, Fundulus heteroclitus. These two populations experience different local thermal conditions throughout the year and have been previously shown to differ in their mitochondrial physiology, which may be indicative of adaptation in response to local thermal conditions (Chung & Schulte,

2015; Chung et al., 2017a; Fangue et al., 2009b; Healy et al., 2017). In addition, this species has been shown to undergo substantial remodeling of mitochondrial properties in response to temperature acclimation making it an ideal system in which to investigate the role of UCPs in thermal adaptation and acclimation. Therefore, the objectives of this study were to: 1)

Identify UCP isoform gene sequences in F. heteroclitus, 2) Characterize the mRNA expression patterns of each isoform across tissues, and 3) Determine how the mRNA expression patterns of the UCP isoforms differ across both warm and cold acclimation temperatures and between populations. We hypothesized that the regulation of UCPs is involved in thermal adaptation and acclimation and therefore predicted that Ucp mRNA expression would differ between populations and in response to thermal acclimation.

2.2 Methods

2.2.1 Phylogenetic analysis

Nucleotide BLAST was used to identify gene sequences in the Fundulus heteroclitus genome (Fundulus_heteroclitus-3.0.2, GCF_000826765.1, annotation release 101) that share significant sequence similarity with Ucp1, Ucp2, Ucp3, Ucp3L and Ucp5 as characterized in several other tetrapod and fish species. Sequences for the UCP isoform gene sequences from

20 an additional five fish species and eight tetrapod species were also gathered from the NCBI and Ensembl databases (see Table 2.1 for accession numbers). MUSCLE alignment (Edgar,

2004) and phylogenetic analysis was performed using Mega (v. 6.06) to identify which of the

F. heteroclitus Ucps were homologous with each of the Ucp family members identified in other species. A Maximum-Likelihood (ML) tree (substitution type: nucleotide, model/method: Tamura-Nei model, Rates among sites: Uniform rates, Gaps/Missing data treatment: complete deletion, all codon positions selected, ML heuristic method: Nearest-

Neighbor-Interchange) was constructed and the reliability of the tree was confirmed by bootstrapping with 1500 replicates.

As an additional method to confirm the identification of homology relationships, the genes adjacent to the various Ucp isoforms were identified in multiple species using

Genomicus (v. 89.02) with all Ensembl vertebrate genomes. The genes surrounding each Ucp were also identified in the F. heteroclitus genome, and these gene lists were compared to those identified in other species using Genomicus to confirm syntenic relationships.

The annotation of Ucp isoforms in the published fish genomes is poor. For example, some

Ucp isoforms in fish genomes (both in Ensembl and Genbank) were labeled similarly to each other or as unknown proteins. In these cases, the genes were given the names of their mammalian counterparts based on both phylogenetic and syntenic relationships identified here. In the case of Ucp3L, the F. heteroclitus gene was named based on relationships to the same isoform in other fish species.

21 Table 2.1 Accession numbers for gene sequences used in phylogenetic analyses. Ensembl and NCBI gene IDs provided where available.

Group Species Name Ucp1 Gene ID Ucp2 Gene ID Ucp3 Gene ID Ucp3L Gene ID Ucp5 Gene ID

Tetrapods Anolis carolinensis ENSACAG00000006491 ENSACAG00000009396 XM_008124024.2 XM_003229550.2 Gallus gallus ENSGALG00000017316 XM_015280964.1 Bos taurus ENSBTAG00000004647 ENSBTAG00000003692 ENSBTAG00000005259 ENSBTAG00000016263 NM_001166528.1 NM_001033611.2 NM_174210.1 NM_001046145.1 Loxodonta africana ENSLAFG00000007077 ENSLAFG00000015866 ENSLAFG00000005399 ENSLAFG00000002284 XM_010590606.1 XM_003420045.2 XM_003420044.2 XM_010594644.1 Homo sapiens ENSG00000109424 ENSG00000175567 ENSG00000175564 ENSG00000102078 NM_021833.4 NM_003355.2 NM_003356.3 NM_001282195.1 Mus musculus ENSMUSG00000031710 ENSMUSG00000033685 ENSMUSG00000032942 ENSMUSG00000031105 NM_009463.3 NM_011671.5 NM_009464.3 NM_011398.3 Rattus norvegicus ENSRNOG00000003580 ENSRNOG00000017854 ENSRNOG00000017716 ENSRNOG00000006871 NM_012682.2 NM_019354.3 NM_013167.2 NM_053501.2 Xenopus tropicalis ENSXETG00000014927 ENSXETG00000026447 NM_001113882.1 NM_203848.1

Fish Fundulus heteroclitus XM_012862869.1 XM_012867200.1 XM_012867201.1 XM_012851672.1 XM_012868877.1

Oryzias latipes ENSORLG00000018506 ENSORLG00000009082 ENSORLG00000009082 ENSORLG00000004325 ENSORLG00000015482 ENSORLT00000011391.1 ENSORLT00000011397.1 Gasterosteus ENSGACG00000017283 ENSGACG00000020354* ENSGACG00000020354* ENSGACG00000011266 ENSGACG00000020832 aculeatus Takifugu rubripes ENSTRUG00000013144 ENSTRUG00000014496 ENSTRUG00000006965 ENSTRUG00000014470 ENSTRUG00000009205

Tetraodon nigriviridis ENSTNIG00000006838 ENSTNIG00000011803 ENSTNIG00000016239 ENSTNIG00000011804 ENSTNIG00000007592

Danio rerio ENSDARG00000023151 ENSDARG00000043154 ENSDARG00000091209 ENSDARG00000026680

*Ucp2 and Ucp3 in the G. aculeatus genome are labelled as the same gene and transcript. Ucps2 and 3 are known to be adjacent to each other animal genomes (Tine et al., 2012). However, phylogenetic analysis indicates they are separate Ucp2 and Ucp3 genes. 22

22 2.2.2 Animals and temperature acclimations

Southern adult Fundulus heteroclitus were collected from Jekyll Island, GA, U.S.A.

(31º02'N; 81º25'W) by members of the Schulte lab and northern adult Fundulus heteroclitus were wild-caught by Aquatic Research Organisms (Hampton, NH) from Taylor River, NH,

U.S.A (42°55’N;70°51’W) in the summer of 2014 and brought to holding facilities at The

University of British Columbia in Vancouver, British Columbia, Canada. All F. heteroclitus were then held under common conditions of 15°C, 20ppt salinity, and a 12D:12L photoperiod in 190 l recirculating fiberglass tanks with biological filtration. All fish were fed daily to satiation throughout holding and acclimations with Nutrafin MaxÒ flake food. After at least four weeks under these conditions, 75 southern fish and 75 northern fish were transferred to 8 experimental glass tanks. Fish were held at densities of approximately 0.6 fish per gallon in three 25-gallon tanks, and one 50-gallon tank per population. Different populations remained in separate tanks. Fish were then held in the experimental tanks under the conditions described above for at least three weeks before temperature acclimations began. Temperature acclimations occurred at four different temperatures of 5, 15, 25 and 33

°C with one tank for each population acclimated to each temperature. Tanks were slowly brought to the acclimation temperature over a 24 h period and were then held at that temperature for three weeks. At the end of three weeks, between eight and nine fish from each tank were fasted for 24 h and then euthanized via cervical dislocation, and then sex, weight and length were assessed. Whole animal data may be found in Appendix B, B1

Whole brain, gill, liver and a lateral skeletal muscle sample from just ventral to the dorsal fin were dissected and then flash frozen in liquid nitrogen. The whole process from the fish being removed to the tissues being flash frozen took less than 5 min. All procedures and

23 animal care were done in accordance with the CCAC approved animal use protocol A11-

0732.

2.2.3 RNA isolations and cDNA synthesis

Tissues were homogenized using a Bullet Blender™ 24 (BBX24B, Next Advance,

Inc., Averill Park, NY, U.S.A.). RNA was then isolated using TRIzol® Reagent following the manufacturer’s instructions (Catalog #15596-018, Life Technologies, Grand Island, NY,

U.S.A.). Isolated RNA was then checked for genomic contamination by running qRT-PCRs with 16 randomly selected non-reverse transcribed samples per tissue using the primers for the most highly expressed isoform in each respective tissue. If genomic contamination was present at a ratio greater than 1:16,000, RNA was put through a Qiagen on-column RNA clean-up using a Qiagen RNeasy Mini Kit (Catalog #74104) and Qiagen RNase free DNase

(Catalog #79254). 2 µg of RNA for liver and gill tissue and 1 µg of RNA for brain and muscle tissue was then reverse transcribed into cDNA in a total volume of 20 µL using a

High Capacity cDNA Reverse Transcription Kit according to manufacturer’s instructions

(Catalog # 4368814, Life Technologies, Grand Island, NY, U.S.A.). cDNA for each tissue was diluted 5-fold and a 5-point standard curve was made for each tissue from a pooling all samples per tissue.

2.2.4 Quantitative real-time PCR (qRT-PCR)

qRT-PCR was performed on Bio-Rad CFX96 Touch™Real-Time PCR Detection

System (#185-5196, Bio-Rad Laboratories, Inc., Mississauga, Ontario, Canada). The qRT-

PCR reaction protocol was as follows: 10 min at 95 °C, and then 40 cycles of: 15 seconds at

95 °C, one min at 60 °C. A melt curve analysis from 65 to 95 °C at an increment of 0.5 °C was then conducted to ensure the presence of only one amplicon. Reactions of 20 µL total

24 volume contained 5 µL of diluted cDNA, 0.4 µL of each primer, 10 µL of SYBR green master mix and 4.2 µL water. Primer pairs for all genes run with qRT-PCR may be found in

Table 2.2. Primer pairs were designed using Primer Express (v. 3.0) on regions that were different between isoforms, but the same between any transcript variants. Specificity was confirmed via BLAST against the F. heteroclitus genome in Genbank. Thresholds across primer pairs within a tissue were set to the same value and absolute gene expression values

. (- ) presented are calculated as abs. exp. = (10 /0123 )-56 with the slope calculated from the standard curve and Cq representing the mean cycle number for each sample (run in duplicate, s<20%). No template controls (NTCs) showed no signal across all plates and other plate quality control parameters may be found in Table 2.2. All absolute gene expression values are normalized to 18S rRNA absolute expression, which has previously been used as a reference gene in F. heteroclitus across acclimation temperatures (Fangue et al., 2009b).

2.2.5 Statistical analysis

All statistical analysis was conducted in R (v3.3.2). All data are presented as a mean

± SEM and α = 0.05. Assumptions of normality and homogeneity of variances were assessed using a Shapiro-Wilk test and Bartlett test, respectively. Data were log transformed where necessary to meet assumptions. For tissue-specific expression one-way analyses of variance

(ANOVAs) were used where assumptions were met. For thermal acclimation and population effects on expression two-way ANOVAs were run. Where assumptions were met, post-hoc analyses were done using planned comparison t-tests followed by FDR correction. In cases where parametric assumptions were not met, post-hoc analyses were completed via pair-wise comparisons using a Mann-Whitney-U/Wilcoxon rank sum test with FDR corrected p-values.

25 Table 2.2 Primer pairs used for qRT-PCR. Unless otherwise noted, primers were specifically designed for this study. All primers are displayed in the 5’ to 3’ direction. Plate parameters (efficiency, slope, y-intercept and R2) are representative examples from the tissues where each primer pair had the highest expression. 18S and Ucp2 expression were similar across tissues, therefore liver was used as a representative example here.

Gene Primer Sequence Efficiency (%) Slope Y-intercept R2 Melt Temp (°C) Ucp1 F GCT TCG TCA CCA CCG TCA T 97.0 -3.397 16.323 0.997 80.0 R GGA GAG TTC ATG TAT CTG GTC TTC AC Ucp2 F CCT GCC TTG CCA CTT CGT 91.9 -3.535 28.644 0.995 81.5 R AAG CGA TCA CAG TCG TGC ATA

Ucp3 F GGC GAG GCT CAC AAC GTT 92.3 -3.521 22.272 0.998 80.0 R CGA ACA CGC CCC GAT ACT

Ucp3-like F CAG GTT CCA GGC TCA GAT ACG 91 -3.559 24.513 0.996 79.0 R TGC CGC TGT ACC TCT TCA CA

Ucp5 F GCC TCA TAC GGC ACA ATC AA 84.5 -3.760 23.262 0.999 76.5 R CGA CTG ACG AAC AGC CTC TTC

18S (designed F TTC CGA TTA ACG AAC GAG AC 98 -3.372 14.142 0.998 83.0 by Whitehead R GAC ATC TAA GGG CAT CAC AG et al., 2011) 26

26 All mRNA expression values in figures have been multiplied by a factor of 105 for ease of visual presentation.

2.3 Results

2.3.1 Phylogenetic analysis

Phylogenetic analysis allowed for the identification of five UCP homologues in F. heteroclitus (Figure 2.1). In any situation where isoforms were labelled similarly, they were given the name of their mammalian counterpart as identified by both phylogenetic and syntenic analysis. This phylogenetic analysis has also identified an apparent fish-specific isoform, Ucp3L. In general, the sequences for each isoform from fish are more similar to the same isoform gene sequence from tetrapods than they are to the other fish isoforms. Ucp1, 2,

3 and 3L appear to cluster as the core Ucp gene family, with Ucp5 gene sequences being the least similar and forming an outgroup.

2.3.2 Tissue-specific mRNA expression

The expression profiles from of the five Ucp isoforms investigated here differ between tissues. In brain, there was high expression of Ucp5 mRNA relative to the other isoforms (Figure 2.2A, see Table 2.3 for p-values). Ucp1, Ucp2, Ucp3 and Ucp3L mRNA were expressed at much lower levels in brain than Ucp5 and significantly differed from each other (Figure 2.2A, see Table 2.3 for p-values). In gill tissue, the fish specific isoform,

Ucp3L, was the most highly expressed with Ucp2, 3 and 5 showing the lowest expression and Ucp1 mRNA being expressed at an intermediate level (Figure 2.2B, see Table 2.3 for p- values). In liver tissue, Ucp1 was the most highly expressed isoform and Ucp2 and 3 were the two isoforms with the lowest expression (Figure 2.2C, see Table 2.3 for p-values).

Figure 2.1 Phylogenetic relationships among selected teleost and tetrapod Ucp gene sequences. This tree was constructed using maximum likelihood analysis of nucleotide sequences, rooted to the Ucp-5 gene sequences. Bootstrap values (from 1500 replicates) are shown at the nodes. Gene IDs for all sequences are listed in Table 1.

5 A. Brain 5 B. Gill 2.0 0.4 e d 1.5 0.3 rRNA) rRNA) 18S 18S c 1.0 0.2

ab b 0.5 0.1 b a (relative to a c d (relative to 0.0 0.0 Absolute mRNA expression x10 Absolute mRNA expression x10

Ucp1 Ucp2 Ucp3 Ucp5 Ucp1 Ucp2 Ucp3 Ucp5 Ucp3L Ucp3L Isoform Isoform

5 C. Liver 5 D. Muscle 200 1.5 a d 150 rRNA) rRNA) 1.0 18S 100 18S

0.5 50 c (relative to c d b b (relative to a ab b 0 0.0 Absolute mRNA expression x10 Absolute mRNA expression x10

Ucp1 Ucp2 Ucp3 Ucp5 Ucp1 Ucp2 Ucp3 Ucp5 Ucp3L Ucp3L Isoform Isoform

Figure 2.2 Absolute mRNA expression x105 (relative to 18S rRNA) of each Ucp isoform within brain (A), gill (B), liver (C), and muscle (D) tissue from northern F. heteroclitus acclimated to 15 °C. Means ± S.E.M. are shown with n=9 for each mean. Letters that differ indicate a significant difference.

29 Table 2.3 p-values following Mann-Whitney U test in brain, gill and muscle and post-hoc Tukey HSD analysis for liver for tissue-specific Ucp expression in northern F. heteroclitus. Significant values are indicated by an asterisk (*).

Comparison Brain Gill Liver Muscle Ucp1-Ucp2 0.040* <0.001* <0.001* 0.297 Ucp1-Ucp3 <0.001* 0.008* <0.001* <0.001* Ucp1-Ucp3L <0.001* 0.050* <0.001* 0.023 Ucp1-Ucp5 <0.001* 0.008* <0.001* <0.001* Ucp2-Ucp3 <0.001* 0.297 <0.001* <0.001* Ucp2-Ucp3L <0.001* <0.001* <0.001* 0.086 Ucp2-Ucp5 <0.001* 0.002 <0.001* <0.001*

Ucp3-Ucp3L 0.021* 0.001* <0.001* <0.001* Ucp3-Ucp5 <0.001* 0.069 <0.001* <0.001* Ucp3L-Ucp5 <0.001* <0.001* 0.882 <0.001*

30 In muscle, mRNA expression of the Ucp isoforms also significantly differed (Figure 2.2D), with Ucp3 being the most highly expressed isoform and Ucps 1, 2 and 3L having the lowest mRNA expression. Ucp5 showed an intermediate level of mRNA expression (Figure 2.2D see Table 2.3 for p-values).

2.3.3 Thermal acclimation and population effects on Ucp isoform mRNA expression

2.3.3.1 Expression in brain tissue

Acclimation temperature had a significant effect on mRNA expression of all Ucp isoforms in the brain of F. heteroclitus (Figure 2.3; Ucp1 p<0.001; Ucp2 p<0.001; Ucp3 p<0.001; Ucp3L p=0.020; Ucp5 p<0.001). For Ucp1 and Ucp3 the temperature effect in the brain was mostly driven by an increase in Ucp mRNA expression at warmer acclimation temperatures (Figure 2.3A & C, see Appendix A: Table A.2 for post-hoc p-values). For Ucp

2, this temperature effect was due to a decrease in expression at 5 °C for both populations and for Ucp3L in the brain, the temperature affect was most likely due to higher expression at

15 °C (Figure 2.3B & D, see Appendix A: Table A.2 for post-hoc p-values). For the most highly expressed Ucp isoform in the brain, Ucp5, expression appeared to be highest with acclimation to 5 °C and decreased at the warmer temperatures (Figure 2.3E, see Appendix A:

Table A.1 for post-hoc p-values).

There was a significant effect of population for four out of the five Ucp isoforms in the brain (Figure 2.3; Ucp1 p<0.001; Ucp2 p=0.046; Ucp3 p<0.001; Ucp5 p<0.001, see

Appendix A: Table A.1 for non-significant p-values). For Ucp1 and Ucp5, mRNA expression was generally higher in the northern population than the southern population (Figure 2.3A and E, see Appendix A: Table A.2 for post-hoc p-values) and for Ucp2 and Ucp3, the opposite pattern may be present, although these effects were not detected in post-hoc planned

5 5 A. Ucp1 B. Ucp2 0.08 * 0.3 b h h 0.06 rRNA) rRNA) 0.2 a a h 18S 18S * * 0.04 * a a a a i mRNA expression x10 mRNA expression x10 0.1 hi hi 0.02 h b i Ucp2 Ucp1 (relative to (relative to 0.00 0.0 5 5 15 25 33 15 25 33 Absolute Absolute Acclimation Temperature (oC) Acclimation Temperature (oC) 5

5 C. Ucp3 D. Ucp3L 0.025 0.006 h h 0.020 h rRNA) rRNA) 0.004 a 0.015 18S 18S h h h a a a 0.010 a mRNA expression x10 h h 0.002

0.005 a a

Ucp3 a (relative to (relative to 0.000 0.000 5 5 15 25 33 15 25 33

Absolute o o Acclimation Temperature ( C) Absolute Ucp3L mRNA expression x10 Acclimation Temperature ( C)

5 E. Ucp5 4

ab 3 rRNA)

18S * 2 a * b h * 1 b h h h (relative to 0 5 15 25 33

Absolute Ucp5 mRNA expression x10 Acclimation Temperature (oC)

Figure 2.3 Absolute mRNA expression x105 (relative to 18S rRNA) of Ucp1 (A), Ucp2 (B), Ucp3 (C), Ucp3L (D), and Ucp5 (E) in brain tissue from northern (grey bars) and southern (white bars) Fundulus heteroclitus across acclimation temperatures (°C). Means ± S.E.M. are shown with n=8-9 for each mean. For A-C, letters that differ indicate a significant difference within a population and an asterisk indicates a significant difference between populations at a given acclimation temperature as determined by FDR corrected planned comparisons. For D and E, letters that differ indicate a significant difference within a population and an asterisk indicates a significant difference between populations at a given acclimation temperature as determined by a Mann- Whitney-U/Wilcoxon rank sum test with FDR corrected p-values.

32 comparisons (Figure 2.3B and C, see Appendix A: Table A.2 for post-hoc p-values). Ucp5 was the only isoform in the brain to show an interaction between acclimation temperature and population (Figure 2.3E, p=0.015, see Appendix A: Table A.1 for non-significant p- values). There was little change in Ucp5 mRNA expression across acclimation temperatures in the brains of southern F. heteroclitus, but the northern population displayed a general decrease in Ucp5 mRNA expression with increasing temperatures (Figure 2.3E).

2.3.3.2 Expression in gill tissue

In the gills of F. heteroclitus, there was an effect of acclimation temperature on the mRNA expression of all Ucp isoforms (Figure 2.4; Ucp1 p=0.014; Ucp2 p=0.003;

Ucp3=0.001; Ucp3L p=0.001; Ucp5 p<0.001). For Ucp1 in the gill, mRNA expression increased with acclimation to extreme temperatures (5 and 33 °C) (Figure 2.4A, see

Appendix A: Table A.2 for post-hoc p-values). The temperature effect on Ucp2 and Ucp3 mRNA expression in the gill was primarily driven by an increase in mRNA expression following acclimation to warmer temperature (Figure 2.4B &C, see Appendix A: Table A.2 for post-hoc p-values). Ucp3L mRNA expression was generally highest following cold acclimation (Figure 2.4D), but these effects were not detectable in post hoc analysis (see

Appendix A: Table A.1). Ucp5 in the gill displayed a similar pattern as seen in the brain, with expression generally higher at 5 °C, although not significantly higher than expression at

15 °C, and lower at the two warmer acclimation temperatures (Figure 2.4E, see Appendix A:

Table A.2 for post-hoc p-values).

Four out of five Ucp isoforms in the gill showed a significant effect of population

(Figure 2.4; Ucp1 p<0.001; Ucp2 p=0.007; Ucp3 p=0.003; Ucp5 p<0.001, see Appendix A:

Table A1 for non-significant p-values). mRNA expression of Ucps 1, 2, and 5 in the gill was

5 5 A. Ucp1 B. Ucp2 0.8 0.25 * * b b 0.20 0.6 rRNA) rRNA) * ab 0.15 18S 18S 0.4 h ab 0.10 h mRNA expression x10 mRNA expression x10 * h a a h 0.2 a h 0.05 a Ucp2 Ucp1 h h h (relative to (relative to 0.0 0.00 5 5 15 25 33 15 25 33 Absolute Absolute Acclimation Temperature (oC) Acclimation Temperature (oC) 5

5 C. Ucp3 D. Ucp3L 0.4 0.8 h

i hi 0.3 0.6 rRNA) rRNA) b a 18S 18S 0.2 i 0.4 h a a h

mRNA expression x10 a mRNA expression x10 * h a 0.1 h a * 0.2 a Ucp3 Ucp3L (relative to (relative to 0.0 0.0 5 5 15 25 33 15 25 33

Absolute o o

Acclimation Temperature ( C) Absolute Acclimation Temperature ( C)

5 E. Ucp5 0.3 * ab

rRNA) 0.2 18S * a mRNA expression x10 0.1 h b *

Ucp5 h b

(relative to h h 0.0 5 15 25 33

Absolute Acclimation Temperature (oC)

Figure 2.4 Absolute mRNA expression x105 (relative to 18S rRNA) of Ucp1 (A), Ucp2 (B), Ucp3 (C), Ucp3L (D), and Ucp5 (E) in gill tissue from northern (grey bars) and southern (white bars) Fundulus heteroclitus across acclimation temperatures (°C). Means ± S.E.M. are shown with n=8-9 for each mean. For B-D, letters that differ indicate a significant difference within a population and an asterisk indicates a significant difference between populations at a given acclimation temperature as determined by FDR corrected planned comparisons. For A and E, letters that differ indicate a significant difference within a population and an asterisk indicates a significant difference between populations at a given acclimation temperature as determined by a Mann- Whitney-U/Wilcoxon rank sum test with FDR corrected p-values.

34 higher in the northern population than in the southern population (Figure 2.4A, B, and E, see

Appendix A: Table A.2 for post-hoc p-values). In contrast, expression of Ucp3 in the gill appeared higher in the southern population than the northern population at 15 and 25 °C

(Figure 2.4C, see Appendix A: Table A.2 for post-hoc p-values). There were significant interactions between thermal acclimation and population for Ucp 2, 3, and 3L in the gill

(Figure 2.4; Ucp2 0.015; Ucp3 0.003; Ucp3L 0.039, see Appendix A: Table A.1 for non- significant p-values). Ucp2 differed between populations at 33 °C and Ucp3 differed at 15 and 25 °C between populations (see Appendix A: Table A.2 for post-hoc p-values).

2.3.3.3 Expression in liver tissue

Four out of five of the Ucp isoforms displayed differences in mRNA expression with acclimation temperature in liver tissue (Figure 2.5; Ucp1 p<0.001; Ucp2 p<0.001; Ucp3L p<0.001; Ucp5 p<0.001, see Appendix A: Table A.1 for non-significant p-values). Ucp1, the most highly expressed isoform in F. heteroclitus liver tissue, displayed a pattern with acclimation temperature whereby expression was lowest at 25 °C in both populations (Figure

2.5A, see Appendix A: Table A.2 for post-hoc p-values). Ucp2 showed an increase in expression in the liver following acclimation to 33 °C (Figure 2.5B) and Ucp3L and Ucp5 showed a decrease in mRNA expression at warmer temperatures (Figure 2.5D & E, see

Appendix A: Table A.2 for post-hoc p-values).

The same four isoforms also showed a significant effect of population on mRNA expression in the liver (Figure 2.5; Ucp1 p<0.001; Ucp2 p=0.024; Ucp3L p<0.001; Ucp5 p<0.001, see Appendix A: table A.1 for noon-significant p-values). All four of these isoforms demonstrated slightly higher mRNA expression in the northern population than in the

5 5 A. Ucp1 B. Ucp2 200 0.20 a * a 150 * 0.15 rRNA) rRNA) a i 18S 18S 100 * 0.10 h a a h mRNA expression x10 mRNA expression x10 i 50 * 0.05 * b Ucp2

Ucp1 a a (relative to (relative to h j i i 0 0.00 5 5 15 25 33 15 25 33 Absolute Absolute Acclimation Temperature (oC) Acclimation Temperature (oC)

5 5 C. Ucp3 D. Ucp3L 0.04 0.0020 a h

h * 0.03 h 0.0015 ab rRNA) rRNA) a 18S 18S 0.02 a 0.0010 a * a mRNA expression x10 mRNA expression x10 b 0.01 0.0005 a h h i i Ucp3 h Ucp3L (relative to (relative to 0.00 0.0000 5 5 15 25 33 15 25 33

Absolute o o Acclimation Temperature ( C) Absolute Acclimation Temperature ( C) 5 E. Ucp5 1.5 * ab

rRNA) 1.0 18S

* mRNA expression x10 0.5 a h * Ucp5 b

(relative to hi h c i 0.0 5 15 25 33

Absolute Acclimation Temperature (oC)

Figure 2.5 Absolute mRNA expression x105 (relative to 18S rRNA) of Ucp1 (A), Ucp2 (B), Ucp3 (C), Ucp3L (D), and Ucp5 (E) in liver tissue from northern (grey bars) and southern (white bars) Fundulus heteroclitus across acclimation temperatures (°C). Means ± S.E.M. are shown with n=7-9 for each mean. For A, B and D, letters that differ indicate a significant difference within a population and an asterisk indicates a significant difference between populations at a given acclimation temperature as determined by FDR corrected planned comparisons. For C and E, letters that differ indicate a significant difference within a population and an asterisk indicates a significant difference between populations at a given acclimation temperature as determined by a Mann-Whitney-U/Wilcoxon rank sum test with FDR corrected p-values.

36 southern population (Figure 2.5A, B, D, & E, see Appendix A: Table A.2 for post-hoc p- values).

Three of five Ucp isoforms also showed an interaction between population and temperature in relation to mRNA expression (Figure 2.5; Ucp1 p<0.001; Ucp2 p=0.008;

Ucp3L p=0.050, see Appendix A: Table A.1 for non-significant p-values). Ucp1 decreased in mRNA expression at 5 °C in the southern population, but not the northern population (Figure

2.5A), Ucp2 decreased at warm acclimation temperatures in the southern population but not the northern population (Figure 2.5B), and Ucp3L decreased to a greater extent in the southern population at warmer acclimation temperatures than in the northern population

(Figure 2.5D, see Appendix A: Table A.2 for post-hoc p-values).

2.3.3.4 Expression in muscle tissue

Four out of five of the Ucp isoforms showed differences in expression with temperature acclimation in F. heteroclitus muscle tissue (Figure 2.6; Ucp1 p<0.001; Ucp3 p=0.005; Ucp3L p<0.001; Ucp5 p<0.001, see Appendix A: Table A.1 for non-significant p- values). Ucp1 mRNA expression in muscle displayed a similar pattern to that in the liver with slightly higher expression occurring at 15 and 33 °C and slightly lower expression at 5 and 25 °C (Figure 2.6A, see Appendix A: Table A.2 for post-hoc p-values). Ucp3, the most highly expressed isoform in muscle tissue, differed with acclimation temperature primarily in the northern population which showed an increase in Ucp3 mRNA expression at 33 °C

(Figure 2.6C, see Appendix A: Table A.2 for post-hoc p-values). Ucp2 expression decreased with acclimation to the cold (Figure 2.6B) and Ucp5 expression decreased with acclimation to the warm (Figure 2.6E, see Appendix A: Table A.2 for post-hoc p-values).

37 5 5 A. Ucp1 B. Ucp2 0.05 a 0.04 h 0.04 a 0.03 rRNA) rRNA) b h 0.03 h h 18S 18S ab a 0.02 ab 0.02 h h mRNA expression x10 h a mRNA expression x10 * 0.01 0.01

Ucp1 h a Ucp2 (relative to (relative to 0.00 0.00 5 5 15 25 33 15 25 33

Absolute o Acclimation Temperature ( C) Absolute Acclimation Temperature (oC)

5 5 C. Ucp3 D. Ucp3L 8 0.025 h h 0.020 h 6 rRNA) rRNA) 0.015 18S 18S h 4 h b

h mRNA expression x10 0.010 b

mRNA expression x10 * 2 * a * a a 0.005 c i

Ucp3 a Ucp3L d j (relative to (relative to 0 0.000 5 5 15 25 33 15 25 33

Absolute o o Acclimation Temperature ( C) Absolute Acclimation Temperature ( C)

5 E. Ucp5 0.20 * a

0.15

rRNA) ab

18S 0.10 h

mRNA expression x10 h 0.05 b h * Ucp5

(relative to c i 0.00 5 15 25 33

Absolute Acclimation Temperature (oC)

Figure 2.6 Absolute mRNA expression x105 (relative to 18S rRNA) of Ucp1 (A), Ucp2 (B), Ucp3 (C), Ucp3L (D), and Ucp5 (E) in muscle tissue from northern (grey bars) and southern (white bars) Fundulus heteroclitus across acclimation temperatures (°C). Means ± S.E.M. are shown with n=8-9 for each mean. For A, B and E, letters that differ indicate a significant difference within a population and an asterisk indicates a significant difference between populations at a given acclimation temperature as determined by FDR corrected planned comparisons. For C and D, letters that differ indicate a significant difference within a population and an asterisk indicates a significant difference between populations at a given acclimation temperature as determined by a Mann-Whitney-U/Wilcoxon rank sum test with FDR corrected p-values.

38 These same four isoforms showed a significant effect of population on mRNA expression (Figure 2.6; Ucp1 p<0.001; Ucp3 <0.001; Ucp3L 0.008; Ucp5 p<0.001, see

Appendix A: Table A.1 for non-significant p-values). Ucp1 and Ucp5 were more highly expressed in the northern population than in the southern population (Figure 2.6A & E), and

Ucp3 and Ucp3L showed slightly higher expression in the southern population than in the northern (Figure 2.6C & D, see Appendix A: Table A.2 for post-hoc p-values). Two out of five isoforms had an interaction between acclimation temperature and population (Figure 2.6;

Ucp3 p=0.013; Ucp3L p=0.001, see Appendix A: Table A.1 for non-significant p-values).

Only the northern population, not the southern population, demonstrated a change in muscle

Ucp3 expression with thermal acclimation (Figure 2.6C, see Appendix A: Table A.2 for post- hoc p-values), and the southern population showed a greater decrease in Ucp3L mRNA expression with increasing acclimation temperature than the northern population (Figure

2.6D, see Appendix A: Table A.2 for post-hoc p-values).

There were some consistent patterns in Ucp mRNA expression across tissues. Ucp5 displayed a similar pattern across tissues (Figures 2.3E, 2.4E, 2.5E, & 2.6E), whereby expression generally decreased with increasing acclimation temperature and was more highly expressed in the northern population across acclimation temperatures. Additionally, in liver and muscle Ucp1 mRNA may change in a similar pattern (Figures 2.5A & 2.6A). Expression appeared lowest at 5 and 15 °C for Ucp1 in both liver and muscle, but this pattern was only significant in liver tissue. However, in general, each Ucp isoform in each tissue appears to display a slightly different pattern with thermal acclimation and between populations.

39 2.4 Discussion

This study represents a comprehensive examination of the changes in mRNA expression of five different Ucp isoforms across four tissues and in two populations of a eurythermal teleost, Fundulus heteroclitus. The isoforms have tissue-specific distribution pattern such that each tissue examined in this study has a different Ucp isoform that is most highly expressed relative to the others at the level of mRNA, with Ucp5 most highly expressed in brain, Ucp3L in gill, Ucp1 in liver, and Ucp3 in muscle. Most tissues have a single dominant isoform that is expressed, except for gill which expressed both Ucp3L and

Ucp1 at high levels. The mRNA expression of most isoforms across tissues was affected by thermal acclimation and differed between the northern and southern populations of F. heteroclitus. The differences in mRNA expression of these Ucp isoforms with acclimation and between populations may indicate that the regulation of Ucp gene expression is an important response to thermal change. However, the pattern of response to thermal acclimation and between populations was not consistent across isoforms or tissues. This suggests that either different isoforms play different physiological roles or that different tissues have different responses to thermal acclimation at the level of the mitochondrion.

2.4.1 Phylogenetic relationships

The phylogenetic relationships between the Ucp isoform gene sequences from fish and tetrapod species supports the existence of a core Ucp family consisting of the closely related Ucp1, Ucp2, and Ucp3 with Ucp5 being less closely related (Jastroch et al., 2005;

Mark et al., 2006; Tine et al., 2012; Tseng et al., 2011; Wen et al., 2015). However, our ability to discriminate among these families was limited (as indicated by the low bootstrap values on the branches for Ucps 1, 2, and 3), which is most likely due to high similarity in the

40 sequences of the members of the core Ucp family (Hughes & Criscuolo, 2008). In addition, a fish-specific isoform (Ucp3L) was also identified in F. heteroclitus and shown to be most closely related to the fish Ucp3 isoform gene sequence. Ucp3L has previously been identified in other studies as a fish-specific Ucp isoform that is believed to be a product of the teleost genome duplication event (Tine et al. 2012; Wen et al., 2015). This isoform has been independently lost in several fish species, but is present in several others (Tine et al., 2002). I was unable to identify Ucp4 in the genome of F. heteroclitus, but it has previously been identified zebrafish and cave fish (Wen et al., 2015; Tseng et al., 2011), suggesting that it is present in at least some fish species. At present, the killifish genome is not entirely complete, and thus this isoform may be present in one of the unsequenced regions. Alternatively, this isoform may have been lost from the killifish genome.

2.4.2 Tissue-specific mRNA expression

The multiple UCP isoforms are each known to have expression patterns specific to certain tissues. Ucp1 is most well known in mammals to be specific to brown adipose tissue

(BAT) where it plays a thermogenic role (Nicholls & Locke, 1984). In fish species, Ucp1 has been found primarily in liver, where it has the highest expression relative to the other isoforms (Jastroch et al., 2005; Wen et al., 2015). We also found Ucp1 to be the most highly expressed isoform in the liver of F. heteroclitus, with no other isoform expressed at relatively appreciable levels (Fig. 2C), and it was also present at relatively high levels in gill tissue

(Figure 2B). Ucp3 is known to be a skeletal muscle-specific isoform across species (Boss et al., 1997; dos Santos et al., 2013), and is also found in F. heteroclitus at the highest levels in skeletal muscle (Fig. 2D). The fish-specific isoform, Ucp3L, has been identified in several other fish species, but its tissue distribution has not been fully characterized (Tine et al.,

41 2012; Wen et al., 2015). Here, we show that Ucp3L is primarily present in the gill of F. heteroclitus (Fig. 2B). In brain tissue, Ucp5 is the most highly expressed isoform and this has also been seen in zebrafish brains (Tseng et al., 2011).

2.4.3 Effects of thermal acclimation and population on Ucp mRNA expression

The response of Ucps to thermal acclimation varied greatly depending on the isoform and the tissue in which it was expressed. A summary of the significant changes in Ucp isoform mRNA expression across tissues can be found in Table 2.4. Furthermore, the expression patterns of each isoform across acclimation temperatures and tissues also differed between populations (summarized in Table 2.5). Although, as discussed above, each tissue has a specific isoform that is expressed at the highest level, there were detectable levels of all isoforms in most tissues. Results for all isoforms in all tissues are presented in Figures 2.3-

2.6, but to focus my discussion, here I discuss only those isoforms that are expressed above a minimum threshold of ~2.7% of the expression of 18S rRNA (unshaded cells in Tables 2.4 and 2.5). In practice, this equates the mRNAs whose crossing cycle was cycle 28 or earlier.

Cycle 28 is approximately 100 times the expression detection limit (~ cycle 35) of the technique and more than 1000 times lower than the expression of the most highly expressed isoforms (~ cycle 16) in each tissue. It is important to note that changes in mRNA levels do not necessarily represent changes in protein levels, and thus examination of changes in mRNA levels simply provides hypotheses that should be subsequently tested through studies at the protein levels or through studies of functional activity. However, isoforms expressed at extremely low levels are less likely to be physiologically significant in each tissue, and thus by focusing this discussion on the more highly expressed isoforms there is a greater chance that the observed changes have physiological significance.

42 Table 2.4 Ucp mRNA expression changes across tissues and acclimation temperatures for Ucp1, Ucp2, Ucp3, Ucp3L and Ucp5. Arrows indicate a directional change significantly different from expression levels seen at 15°C as determined by post-hoc analyses. Dashes represent no significant change. Grey shading indicates that mRNA expression was below the threshold level of ~2.7% of 18S rRNA expression.

Tissue Population Ucp1 Ucp2 Ucp3 Ucp3L Ucp5

5°C 25°C 33°C 5°C 25°C 33°C 5°C 25°C 33°C 5°C 25°C 33°C 5°C 25°C 33°C

Brain N ------

Brain S ------

Gill N ------

Gill S ------

Liver N ------

Liver S ------

Muscle N ------

Muscle S ------

43 Table 2.5 Ucp mRNA expression differences between populations for Ucp1, Ucp2, Ucp3, Ucp3L and Ucp5 across tissues. “N” indicates significantly higher expression in the northern population and “S” indicates significantly higher expression in the southern population as determined by ppopulation from two-way ANOVA results. Grey shading indicates isoforms where mRNA expression was below ~2.7% of 18S rRNA expression. For p-values see Table 4.

Tissue Ucp1 Ucp2 Ucp3 Ucp3L Ucp5

Brain N S S - N

Gill N N S - N

Liver N N - N N

Muscle N - S S N

44 In brain tissue, Ucp5 is the most highly expressed isoform, and Ucp1 and Ucp2 were also expressed at levels that exceed the threshold discussed above. The northern population demonstrated a decrease in Ucp5 mRNA expression with increasing acclimation temperature

(Figure 2.3E). This pattern is similar to that observed by Tseng et al. 2011, who showed that

Ucp5 mRNA expression in the brain of zebrafish increased after 1 h cold exposure with expression remaining slightly elevated after 24 h. Ucp5 is expressed in the brain in mammals, and in humans it has been shown to reduce oxidative stress in neuroblastoma cells when overexpressed (Kwok et al., 2010). Reasoning from this observation, I hypothesize that the increase in Ucp5 expression in fish brains in response to cold exposure may represent a defense against lipid peroxidation at colder temperatures (Jastroch et al., 2007; Tseng et al.,

2011). However, data from F. heteroclitus permeabilized brain tissue preparations indicate that there is no increase in mitochondrial capacity at colder temperatures, which suggests that there may not be a corresponding increase in lipid peroxidation in the cold in this species

(Chung et al., 2017). Permeabilized brain tissue preparations from F. heteroclitus demonstrate a decrease in mitochondrial oxidative phosphorylation capacity and leak at warmer acclimation temperatures (Chung et al., 2017). Thus, the decrease in Ucp5 expression at warm temperatures seen here (Figure 2.3E) may contribute to the previously observed decrease in leak respiration at warm temperatures (Chung et al., 2017).

In contrast to the pattern observed for Ucp5, Ucp1 expression increased with warm acclimation in northern killifish brains. This differs from the patterns observed in some other fish species. For example, Ucp1 mRNA has been shown to increase in the brains of common carp following 7-10 days of cold acclimation (Jastroch et al., 2007). However, there is no increase in Ucp1 mRNA expression upon cold exposure in brain tissue from F. heteroclitus

45 and this was also seen in zebrafish brain following 24 h cold exposure (Tseng et al., 2011). It is difficult to directly compare the responses among studies because the timing of cold exposures and the temperatures tested differ among these studies and therefore any differences in expression may be due to sampling at different points in time following cold exposure. Interestingly, Ucp2 mRNA was detected at higher levels overall than Ucp1 mRNA in killifish brains (Figure 2.2, and Figure 2.3A and B). Ucp2 decreased following cold acclimation in both the northern and southern population fish, which corresponds with previously measured decreases in mitochondrial capacity and leak respiration following cold acclimation seen in killifish brain preparations (Chung et al., 2017).

In addition to the effects of thermal acclimation on Ucp expression in the brain, there were also consistent differences between the two populations, with northern killifish expressing higher levels of Ucp1 and Ucp5 at most acclimation temperatures (see Table 2.5).

However, there were no differences between the populations in the expression of Ucp2, suggesting that the observed effect of population on the expression of Ucp1 and Ucp5 is not likely to be due simply to a difference in mitochondrial volume density or capacity. Previous studies have shown that killifish populations do not differ in mitochondrial capacity in brain tissue (Chung et al., 2017), and studies in muscle suggest that they do not differ in mitochondrial volume density, except following cold acclimation (Dhillon & Schulte, 2011).

If Ucp mRNA expression levels are indicative of differences in UCP protein levels and functions, it may be expected that northern and southern fish differ in proton leak. However, evidence from permeabilized killifish brain tissue suggests that the two populations do not differ in leak respiration with fuels provided to electron transport system complex I or both complexes I and II.

46 Ucp isoforms in the gill were expressed at more similar levels to each other and all above the threshold level discussed previously (Figure 2.2B). All Ucp isoforms mRNA levels except for Ucp3L changed with thermal acclimation in gill tissue from northern killifish, but all responded in different ways from each other (Table 2.4). Additionally, all isoforms except for Ucp3L were expressed at different levels between the two F. heteroclitus populations

(Table 2.5). The diversity of Ucp isoform gene expression relative to other tissues (Figure

2.2) may be reflective of the diversity of functions of the fish gill, and the complexity of gill cell types (Evans et al., 2005). It is possible that each isoform may be expressed in different gill cell types and may carry out a slightly different physiological function, and therefore respond in different ways to thermal acclimation.

In killifish liver tissue, Ucp1 responded to thermal acclimation and differed in the level of mRNA expression between the two killifish populations (Figure 2.5A). Ucp1 is known in mammals to be thermogenic in brown adipose tissue (BAT) (Cannon &

Nedergaard, 2004; Nicholls & Locke, 1984), and is present at levels much higher than any other Ucp1 isoform in any other tissue in mammals (Riquier & Bouillaud, 2000). Similarly,

Ucp1 mRNA in the liver of killifish is the most highly expressed of all isoforms from any tissue (Figure 2.2). This is not the first time Ucp1 mRNA has been measured in abundance in the fish livers (Jastroch et al., 2005; Wen et al., 2015; Murakami et al., 2015). Here, we have shown that Ucp1 mRNA expression responds in a similar manner in both northern and southern killifish, with lowest expression following warm acclimation to 25 °C (Figure 2.5A,

Table 2.4). The southern population also showed lower expression at 5 °C and the northern population displayed a trend toward lower Ucp1 mRNA expression in the liver at 5 °C as well (Figure 2.5A). This is similar to the Ucp1 mRNA expression pattern seen following cold

47 acclimation in common carp, which showed a decrease in Ucp1 mRNA expression following

2 days of cold acclimation and maintained following 4 weeks of cold acclimation (Jastroch et al., 2005; Murakami et al., 2015). This Ucp1 mRNA expression pattern in common carp liver was also demonstrated to be reflective of Ucp1 function in the liver, showing decreased response to free fatty acid and alkenal activators following cold acclimation (Jastroch et al.,

2007). However, this response does not appear to be consistent across all species, as Chinese perch shows an increase in liver Ucp1 mRNA expression following 4 weeks of cold acclimation and no change in Ucp1 mRNA expression following 4 weeks of warm acclimation (Wen et al., 2015). If compared at the warmest acclimation temperature (33 °C),

Ucp1 mRNA expression is no different from that seen in the 15 °C acclimated fish (Figure

2.5A). However, there is a decrease in Ucp1 mRNA expression at a more intermediate warm acclimation temperature of 25 °C, indicating that there may differences in responses to thermal acclimation at different temperatures despite them all being relatively warm.

Ucp2 and Ucp5 mRNAs are also expressed in killifish livers above the threshold level that I discussed above, although they are present at much lower levels than is Ucp1. These two isoforms also responded to thermal acclimation in F. heteroclitus livers. Ucp2 mRNA expression did not change much with temperature acclimation in the northern population, but there was a decrease in Ucp2 mRNA expression at 25 and 33 °C in the southern population

(Figure 2.5B). This means that the northern population had higher Ucp2 mRNA expression at those warmer temperatures than the southern population. Similar to both Ucp1 and Ucp2 in the liver, the northern population also had higher expression of Ucp5 in the liver than the southern population (Table 2.5). Ucp5 in the liver shows a similar pattern to Ucp5 expression in both the brain and the gill of F. heteroclitus, where mRNA expression in both populations

48 displays the trend of decreasing with increasing temperature (Figures 2.3E, 2.4E and 2.5E,

Table 2.4). The common expression pattern of this isoform across tissues may be indicative of a common function of this isoform across tissues. Indeed, liver mitochondrial capacity has been shown to increase with acclimation to cold temperatures and decrease with acclimation to warm temperatures in northern F. heteroclitus (Chung & Schulte, 2015). The expression patterns of some of the Ucp isoforms therefore follow the change in mitochondrial capacity, possibly to attenuate the production of reactive oxygen species (ROS) caused by the increase in function. Chung and Schulte (2015) observed no change in ROS production following thermal acclimation in F. heteroclitus livers, possibly because there are mechanisms such as

Ucps in place to attenuate ROS production in response to thermal acclimation.

In muscle, Ucp3 was the most highly expressed isoform and differed with thermal acclimation and between populations (Figure 2.6C, Tables 2.4 & 2.5). Ucp3 expression in the muscle tissue of southern F. heteroclitus did not change across acclimation temperatures, yet the northern population showed an increase in expression following acclimation to 33 °C. At

33 °C, the northern and southern populations have similar levels of Ucp3 mRNA expression whereas at 5 and 15 °C the southern population had higher Ucp3 expression. This could indicate that higher expression of Ucp3 in skeletal muscle is important for mitochondrial function at warmer temperatures. Similarly, (Mark et al., 2006) found that the Antarctic eelpout increased expression of Ucp2 in liver and muscle tissue in response to warm acclimation. Ucp2 was not detected at high levels in the muscle of F. heteroclitus in this study, nor was it detected at high levels in common carp skeletal muscle (Jastroch et al.,

2005). This indicates that there may be some interspecific differences in Ucp isoform expression across tissues. Ucp5 was also expressed higher than the threshold in muscle and

49 while there was a similar pattern for this isoform in muscle as there was in the other tissues

(Table 2.4), northern fish had higher expression of Ucp5 at 15 °C than at other temperatures or than the southern population (Figure 2.6E). Again, the similarities in Ucp5 expression across tissues may indicate a common function in response to thermal acclimation.

2.4.4 Conclusions

I have demonstrated that mRNA expression of Ucp isoforms across tissues responds to thermal acclimation and differs between populations of F. heteroclitus. However, the pattern of response varies depending on the isoform and on the tissue in which it is expressed. The difference in Ucp mRNA expression between populations also varied depending on the isoform and which tissue it was expressed in. It must be noted that the changes in mRNA expression seen here may not be indicative of protein levels, but there is some evidence in fish to suggest that patterns seen at the level of mRNA may also be reflected by protein expression and UCP function (Jastroch et al., 2007; Mark et al., 2006). If this is true, these different isoforms may be carrying out different functions in response to thermal acclimation within these tissues. While the changes in Ucp mRNA expression seen here suggest a functional significance of Ucps in response to thermal acclimation, further investigation into the function of Ucps across acclimation temperatures is required in order to elucidate their potential role in thermal acclimation. This question is pursued in Chapter 3 of this thesis.

50 Chapter 3: The effects of thermal acclimation on mitochondrial function and fatty-acid induced uncoupling in two populations of Atlantic killifish,

Fundulus heteroclitus

3.1 Introduction

Ambient temperature influences the geographic distribution of ectotherms due to the effects of temperature on biochemical reaction rates which, in turn, affects whole-animal performance (Hochachka & Somero, 2002; Chung & Schulte, 2015; Fangue et al., 2009b;

Iftikar & Hickey, 2013; Pörtner, 2001). Due to the importance of mitochondria in aerobic metabolism, thermal effects on mitochondrial biochemistry likely influence thermal performance in ectothermic organisms (Iftikar & Hickey, 2013; Pörtner, 2001). Thermal acclimation is known to affect many mitochondrial properties including membrane physical state, enzyme activity and oxidative phosphorylation capacity, the production of reactive oxygen species (ROS) and proton leak across the inner mitochondrial membrane (for review see: Guderley, 2011). The nature of thermal acclimation responses can vary greatly, and may involve (for example) absent, partial, complete or inverse compensation for the effects of temperature (Precht et al., 1958). Many fish mitochondrial properties change in ways that presumably compensate for Q10 thermal effects meaning that mitochondrial capacity and proton leak often increase in response to cold acclimation, representing partial or complete compensation for the effects of low temperature (Chung & Schulte, 2015; Guderley, 1990).

However, following acclimation to warm temperatures modification of mitochondrial function has been shown to be variable. Depending on the species or temperatures examined, mitochondrial function may increase due to Q10 effects, which would suggest absence of compensation, or may decrease as a compensatory response (e.g., Abele et al., 2002; Chung

51 et al., 2017). In instances where there are increases in mitochondrial capacity, increases in membrane potential may occur, resulting in increased ROS production and oxidative damage

(Abele et al., 2002; Abele & Puntarulo, 2004; Brand, 2000). It is possible that a mechanism for mitigating thermal effects on membrane potential and proton leak may exist within the mitochondria of ectothermic organisms, although this possibility has received only limited attention (e.g., Jastroch et al., 2007; Mark et al., 2006; Tseng et al., 2011; Wen et al., 2015).

Uncoupling proteins facilitate the movement of protons from the mitochondrial intermembrane space through to the mitochondrial matrix, thereby decreasing the proton motive force that drives the production of ATP (Nicholls & Locke, 1984). This uncoupling and decrease of the proton motive force has been postulated to be protective against oxidative damage (Brand, 2000). In mammalian brown adipose tissue (BAT), this decrease of the proton motive force is due to uncoupling protein 1 (UCP1), which causes futile energy cycling and results in thermogenesis (Nicholls, 2001). Functional similarities and phylogenetic analyses suggest that UCP1, UCP2 and UCP3 compose the core UCP protein family with other UCPs (4 and 5) being more distantly related (Esteves & Brand, 2005;

Rousset et al., 2004). UCPs have been identified in a variety of species and taxa, including plants and ectothermic organisms, but the functional roles of UCPs in non-mammalian taxa and of the paralogs are still unclear (Brand & Esteves, 2005). While UCP1 in mammalian

BAT has a thermogenic function, other proposed physiological roles of UCP proteins that have been proposed include attenuating ROS production, transporting fatty acids, and playing roles in ROS signalling and insulin secretion (see review Brand & Esteves, 2005).

UCP1, UCP2 and UCP3 are the most well characterized UCP isoforms and are known to be activated by free fatty acids, reactive alkenals and superoxide products and are

52 inhibited by purine nucleotides (ADP, ATP, GDP and GTP) (Klingenberg & Huang, 1999).

It has also been demonstrated that UCPs may be activated by fatty acids of a variety of chain lengths (Klingenberg and Huang, 1999; Davis et al. 2008; Shabalina et al. 2008).

Furthermore, the effectiveness of inhibition by different purine nucleotides is still debated

(e.g., Woyda-Ploszczyca & Jarmuszkiewicz, 2014) and in some scenarios, such as with goldfish UCP3 in skeletal muscle (dos Santos et al., 2013) and lamprey UCP (Wang et al.,

2010) inhibition by purine nucleotides has not yet been demonstrated. Despite this, the use of fatty acid activation and purine nucleotide inhibition has proven a useful tool to characterize the uncoupling capacity of the UCPs following a variety of treatments in a variety of organisms (e.g., Considine et al., 2003; Cunningham et al., 1986; Jastroch et al., 2007).

UCP gene sequences and mRNA expression profiles have been described in several fish species, with a particular focus on changes in expression in response to cold acclimation

(Jastroch et al., 2005; Wen et al., 2015; Chapter 2). UCP1 mRNA expression has been examined in the livers of fish following cold acclimation and does not appear to show a consistent response. UCP1 mRNA expression has been shown to decrease following cold acclimation in common carp (Jastroch et al., 2005; Murakami et al., 2015), but increase in

Chinese perch (Wen et al., 2015). Furthermore, Jastroch et al., (2007) showed that, corresponding to mRNA expression patterns, UCP1 activation in the liver decreased following cold acclimation in common carp. These changes in mRNA expression and mitochondrial uncoupling suggest that UCPs may be playing an important role in thermal acclimation in fish.

Here, I used Atlantic killifish, Fundulus heteroclitus, to look at the influence of thermal acclimation on mitochondrial membrane potential and proton leak. I further

53 investigate the influence of palmitate, a fatty acid and known activator of mammalian UCP, on the proton conductance of the IMM. I used a northern and a southern population of killifish because these subspecies come from different thermal habitats, allowing for the possibility for local adaptation to the different local thermal conditions they experience throughout the year. The mitochondrial properties of F. heteroclitus mitochondrial performance has been previously shown to respond to thermal acclimation and differs between the northern and southern subspecies (Chung et al., 2017; Fangue et al., 2009b).

Furthermore, differences in Ucp isoform mRNA expression have been demonstrated across tissues and across acclimation temperatures for both populations (Chapter 2). I chose to investigate the effects of thermal acclimation and population on liver and brain tissue. In F. heteroclitus, Ucp1 mRNA in the liver is the most highly expressed of any isoform in any tissue (Chapter 2). In F. heteroclitus brains, Ucp5 mRNA is most highly expressed (Chapter

2), and while not a member of the core UCP family, UCP5 has been shown to have similar functional properties to the other UCP isoforms (Kwok et al., 2010). Furthermore, Ucp5 mRNA in F. heteroclitus appears to show a consistent response to thermal acclimation across tissues, suggesting the potential for a common function.

The specific questions I sought to address in this study were 1) How is mitochondrial respiratory capacity in liver and brain influenced by population and thermal acclimation? 2)

How does thermal acclimation and population affect proton leak kinetics in killifish liver mitochondria? 3) How does the presence of a UCP activator (palmitate) influence IMM proton conductance? 4) If so, does this fatty acid effect differ between populations and with acclimation to different temperatures? More specifically, I predicted that northern, but not southern, F. heteroclitus would show an increase in mitochondrial capacity in the cold, as

54 shown in previous studies (Fangue et al., 2009; Dhillon & Schulte, 2011; Chung & Schulte,

2015). Furthermore, I predicted to see the largest increase in mitochondrial uncoupling

(decreased membrane potential and increased rate of O2 consumption) in liver mitochondria following acclimation to 15 °C in both populations. In brain mitochondria, I would predict less fatty-acid induced uncoupling following acclimation to 25 °C in northern, but not southern fish. In both tissues, I would expect northern killifish to show a greater extent of fatty-acid induced uncoupling than southern killifish. The predications regarding fatty-acid induced uncoupling are based on mRNA expression patterns of the most highly expressed isoform in each tissue (Chapter 2).

3.2 Methods

3.2.1 Animals

Adult northern Atlantic killifish of the northern subspecies (Fundulus heteroclitus macrolepidotus) were wild caught by Aquatic Research Organisms from Taylor River, NH,

U.S.A (42°55’N;70°51’W) in the fall of 2016. Adult Atlantic killifish of the southern subspecies (Fundulus heteroclitus heteroclitus) were wild caught from Jekyll Island, GA,

U.S.A. (31º02'N; 81º25'W) in the fall of 2016. Fish were then brought back to and housed at

The University of British Columbia in 114 l glass tanks at 15 °C, 20 ppt salinity, and a

12D:12L photoperiod for at least four months prior to acclimation. Fish were fed once daily to satiation with Nutrafin MaxÒ flake food. Following this holding period, fish were acclimated in the same tanks to 5, 15 or 25 °C with no other change in water parameters or photoperiod. Tanks were slowly brought to acclimation temperature over 24 h and then held at the acclimation temperature for 3 weeks prior to sampling. All procedures and animal care were performed in accordance with the UBC approved animal use protocol A16-0028.

55 3.2.2 Isolation of liver and brain mitochondria

The same isolation protocol was used for both liver and brain tissues and seven fish were pooled for each mitochondrial sample. Prior to euthanasia, fish were fasted for 24 h and their sex, weight and length were recorded. Fish were then euthanized via rapid cervical dislocation and liver and brain tissues were taken. Brains and livers were placed in separate aliquots of ice-cold homogenization buffer (250 mmol l−1 sucrose, 50 mmol l−1 KCl, 0.5

−1 −1 −1 mmol l EGTA, 25 mmol l KH2PO4, 10 mmol l HEPES, 1.5% w/v BSA, pH=7.4 at

20°C). Chemicals for buffers were obtained from Sigma-Aldrich (St. Louis, MO) with the exception of BSA which was obtained from Akron Biotech (Boca Raton, FL). Liver and brain tissues were then minced in the ice-cold homogenization buffer until pieces were approximately 1 mm3 in size. Each tissue was then homogenized using 7 passes of a loose- fitting Teflon pestle. This crude homogenate for each tissue was then filtered through 1-ply cheesecloth and then centrifuged at 600 g for 10 min at 4°C. The fat layer was then aspirated from the supernatant for each tissue and the supernatant subsequently filtered through 4-ply cheesecloth. This defatted supernatant was then centrifuged at 6000 g for 10 min at 4°C. The resulting pellet was then washed two times with fresh ice-cold homogenization buffer at

6000 g for 10 min at 4°C. Any remaining homogenization buffer was aspirated from the pellet and the pellet from each tissue was then suspended in 300 µl BSA-free ice-cold homogenization buffer. The liver and brain mitochondria were then kept on ice until respiration and membrane potential measurements were completed, which was within 6 h of mitochondrial isolation. Protein concentrations in each sample were determined prior to measurements using a Bradford (1976) assay with BSA standards.

56 3.2.3 Mitochondrial respiration and membrane potential measurements

All mitochondrial respiration and mitochondrial membrane potential measurements were determined via high resolution respirometry using an Oxygraph O2k system and tetraphenylphosphonium (TPP+) selective electrodes, respectively (Oroboros instruments,

Innsbruck Austria) at an assay temperature of 15°C. O2 electrodes were calibrated with air

−1 −1 saturated and O2 depleted assay buffer (MiRO5, 110 mmol l sucrose, 0.5 mmol l EGTA,

−1 −1 −1 −1 3 mmol l MgCl2, 60 mmol l K-lactobionate, 20 mmol l taurine, 10 mmol l KH2PO4, 20 mmol l−1 HEPES, 0.1% w/v BSA, pH=7.1 at 30°C; Gnaiger and Kuznetsov, 2002) and were calibrated for background O2 consumption at 15°C using published O2 solubilities (Gnaiger and Forstner, 1983). All respiration and membrane potential measurements were conducted at 15°C.

For liver samples, 0.1-0.2mg of mitochondrial protein was loaded into 2 ml MiRO5 followed by rotenone (0.5 µmol l-1 dissolved in ethanol) to inhibit complex I (CI) and succinate (10 mmol l-1, complex II substrate) to measure state II respiration, and then ADP

(125 µmol l-1, which is the substrate needed for ATP formation by complex V) was added to measure state III. The ADP was then allowed to be completely consumed to determine

ADP/O ratio. ADP/O ratios were calculated according to Gnaiger et al. (2000). Saturating

ADP (1.25 mmol l-1) was then added to reinstate state III and then oligomycin (2.5 µmol l-1 dissolved in ethanol, to inhibit complex IV) was added to induce state IV. Finally, FCCP (1

µmol l-1 dissolved in ethanol) was added to achieve maximum electron transport flux by completely uncoupling the mitochondria. For brain samples, 0.2-0.5mg of mitochondrial protein was loaded into 2 ml MiRO5 followed by rotenone (0.5 µmol l-1 dissolved in ethanol) and succinate (10 mmol l-1) to measure state II respiration, and then ADP (1.25 mmol l-1) to

57 measure state III. Oligomycin (2.5 µmol l-1 dissolved in ethanol) and FCCP (1 µmol l-1 dissolved in ethanol) to achieve state IV and maximum electron transport flux, respectively.

Respiratory control ratios were determined for each mitochondrial sample.

Palmitate is a known fatty acid activator of UCPs, causing increases in mitochondrial

O2 consumption rate and decreases in mitochondrial membrane potential (e.g., Jastroch et al.,

2007; dos Santos et al., 2013; Jastroch et al., 2012). Preliminary analyses (data not shown) indicated that while killifish liver mitochondria appeared to respond to palmitate in the expected manner, the response could not be inhibited by GDP or GTP, known UCP inhibitors. Several concentrations of GDP were tried in killifish mitochondria and both the palmitate and GDP treatments were checked for the expected responses in rat BAT mitochondria (Table 3.1). In killifish mitochondria, the increased respiration and decreased membrane potential that corresponded to the addition of palmitate was also not inhibited by the addition of 8 mmol l-1 glutamate (high extra-mitochondrial glutamate blocks the glutamate aspartate transporter, Samartsev et al., 1997), 6 µmol l-1 cyclosporine A (an inhibitor of the permeability transition pore), or 50 µmol l-1 Genepin (a known inhibitor of

UCP2) (data not shown). The addition of carboxyatractyloside (Cat, a known inhibitor of the adenine nucleotide translocator) resulted in the return of approximately 20% of the increase in respiration and decrease in membrane potential (data not shown) and thus 5 µmol l-1 Cat was included in every run, to exclude the effects of the ANT on proton leak.

Proton leak kinetic curves with and without palmitate were constructed from respiration and membrane potential measurements for liver mitochondrial samples. Previous work from our lab indicates inhibitory effects of nigericin on mitochondrial respiration in killifish (Chung & Schulte, 2015), which is normally used in the determination of proton leak

58 Table 3Table.1 Percent 1. Percent changes changes in state in isolated II with ratoligomycin brown adipose (state tissue IIol) mitochondriain isolated rat exposed brown palmitateadipose tissue mitochondriafollowed exposed by GDP. to sodium palmitate, followed by GDP.

Percent change in state IIol From state IIol to From state IIol +palmitate state IIol +palmitate to stateIIol +palmitate+GDP O2 consumption (nmol O2 +29.35% -41.57% min-1 mg protein-1)

Membrane potential (mV) -5.00% +9.52%

59 kinetic curves to eliminate the proton gradient across the inner mitochondrial membrane, while maintaining the electrical gradient. Because of its inhibitory effects in killifish, nigericin was not added, and thus our measurement of proton motive force through the measurement of membrane potential may represent an underestimate of the true driving force for ATP synthesis by complex V. At the beginning of each run, rotenone (0.5 µmol l-1 dissolved in ethanol), oligomycin (2.5 µmol l-1 dissolved in ethanol) and Cat (5 µmol l-1) were added to the chamber. For palmitate treatments only, sodium palmitate (35 µmol l-1 dissolved in 50% ethanol) was added at the beginning of each run. This resulted in approximately 11.8 nmol l-1 free palmitate in the respiration medium, calculated using

Richieri et al.’s (1993) equation (Free palmitate = 4.4n – 0.03 + 0.23exp(1.16n), where n is the molar ratio of palmitate to albumin). Preliminary data indicated the addition of the same volume of 50% ethanol did not influence membrane potential or respiration rates. TPP+- selective electrodes were introduced to 2 ml MiRO5 and calibrated through five additions of

TPP+ (one 1 µmol l-1 addition followed by four 0.5 µmol l-1 additions for a total of 3 µmol l-1

TPP+). Liver mitochondrial protein (0.7-0.8mg) was then added, followed by succinate (10 mmol l-1) to achieve state II respiration and membrane potential. State II respiration was then titrated using nine additions of malonate (0.5 mmol l-1 each, for a total of 4.5 mmol l-1) and then mitochondria were uncoupled through the addition of FCCP (1 µmol l-1 dissolved in ethanol). Each proton leak kinetic curve was fit to a combined model equation, f(x)=ax+b*exp(cx), where x is membrane potential (mV), using GraphPad Prism 7 software

(La Jolla, CA). This combined model accounts for both the ohmic (linear) and non-ohmic

(exponential) behavior that has been previously described for mitochondrial proton leak kinetics (Jastroch et al., 2012; Murphy, 1989; Nicholls, 1977).

60 Due to limited sample, proton leak kinetic curves were not possible for brain mitochondrial isolations, but respiration and membrane potential measurements were still conducted. TPP+ electrodes were calibrated in 2 ml MiRO5 assay buffer as stated above and

0.8mg of brain mitochondrial protein was added. Succinate (10 mmol l-1) was the added to achieve state II followed by five additions of sodium palmitate (one addition of 35 µmol l-1, followed by four additions of 17.5 µmol l-1 all dissolved in 50% ethanol). Preliminary data indicated the addition of the same volume of 50% ethanol did not influence membrane potential or respiration rates. The last titration step, when a total of 105 µM sodium palmitate had been added to the chamber, was when a palmitate response was observed and the respiration rate and membrane potential was then compared under these conditions between treatments. Finally, FCCP (1 µmol l-1 dissolved in ethanol) was added to uncouple the mitochondria. All values measured by TPP+-selective electrodes were corrected for chemical background and dilution effects. Membrane potential was calculated as described in Chung et al. (2015) for both liver and brain mitochondrial samples. Chemicals were obtained from

Sigma-Aldrich (St. Louis, MO) except for carboxyatractyloside which was obtained from

EMD Millipore (Etobicoke, ON).

3.2.4 Statistical analysis

All statistical analyses were conducted in R (v3.3.2). All data are presented as a mean

± S.E.M. and analyzed with α = 0.05. The assumption of normality was assessed using

Shapiro-Wilk tests and homogeneity of variances was assessed using Bartlett tests. All data were analyzed using two-way analyses of variance (ANOVAs). Where assumptions were met, post-hoc analyses were done using planned comparison t-tests with FDR corrections.

Where assumptions were not met, a log transformation followed by planned comparison t-

61 tests with FDR corrections was used. If assumptions still were not met following transformation, pair-wise comparisons were conducted using a Mann-Whitney-U/Wilcoxon rank sum test with FDR corrections. The DO2 consumption rate and the D membrane potential at the points where they were each the widest between the average palmitate and control curves for each acclimation temperature in each population were determined and analyzed using a two-way ANOVA. One-sample t-tests were used to compare the DO2 consumption rate of each population at each acclimation temperature relative to zero. The p- values from the one-sample t-tests were then FDR corrected. Condition factor was calculated as condition factor = M/L3, where M is mass in g*100 and L is length in cm (Nash et al.,

2006). Whole animal data may be found on Appendix B, B2. Hepatosomatic index (HSI) was calculated as !"#$% '()) and the craniosomatic index was calculated as -%("0 '()) . *+,!$ -,./ '()) *+,!$ -,./ '())

Respiratory control ratios (RCRs) were calculated as state III/state IV. Comparisons of proton leak kinetic curves across both populations were achieved by comparing leak respiration at the highest common membrane potential (142.4 mV) among all individual curves. Comparisons of proton leak within a population were conducted by comparing oxygen consumption rate at a representative membrane potential within the ohmic (linear) portion of the curves; the representative membrane potentials were 152 mV and 142 mV for the northern and southern population, respectively. The oxygen consumption rate at a representative membrane potential within each population was analyzed using a one-way

ANOVA with Tukey honest significant differences post-hoc test.

3.3 Results

3.3.1 Mitochondrial respiration

Liver state III respiration differed significantly among acclimation temperatures in

62 northern killifish, but not in southern killifish (Figure 3.1A, see Tables 3.2 and 3.3 for p- values). Northern killifish had higher state III respiration following acclimation to 5°C, making state III at 5°C higher in northern killifish liver mitochondria than in southern killifish liver mitochondria (Figure 3.1A, see Tables 3.2 and 3.3 for p-values). State IV respiration in the liver did not significantly differ between northern and southern killifish, or with temperature (Figure 3.1C, see Tables 3.2 and 3.3 for p-values). However, there was a trend for state IV respiration to be higher in the northern population than in the southern population at 5°C (Figure 3.1C, see Tables 3.2 and 3.3 for p-values). RCRs in liver mitochondria, an estimate of mitochondrial coupling, were above 6 for both populations across acclimation temperatures and were highest in the northern population following acclimation to 5°C (Figure 3.1E, see Tables 3.2 and 3.3 for p-values). Brain state III respiration and RCR showed a significant effect of temperature, with a slight decrease in state III and the RCR at 25°C primarily in the southern population (Figure 3.1B and F, see

Table 3.2 for p-values), although these differences were not evident following post-hoc analyses. State IV respiration in the brain was not significantly affected by temperature and state III, state IV and RCRs in brain mitochondria did not significantly differ between the two populations of killifish at any acclimation temperature (Figure 3.1B, D, and F, see Tables

3.2 and 3.3 for p-values). The ADP/O ratio is a metric of how many moles of ADP are consumed (i.e., converted to ATP) per mole of oxygen consumed. Here, ADP/O ratios differed between populations and with thermal acclimation (Figure 3.2, see Tables 3.2 and

3.4 for p-values). There was an increase in ADP/O at 5°C, primarily in the northern population and at 25°C the southern population had greater ADP/Os than the northern population (Figure 3.2, see Tables 3.2 and 3.4 for p-values).

63 A. B. ) ) -1 -1 40 * 40 a

30 b 30 h mg protein mg protein h -1 20 h -1 20 b min min Consumption Consumption 2 2 h a 2 10 2 10 a O O a h h (nmol O 0 (nmol O 0 5 5 15 25 15 25 Acclimation Temperature (°C) Acclimation Temperature (°C)

C. D. ) ) -1 4 -1 4 a a h 3 3 h a mg protein h mg protein -1 2 -1 2 h h a min min Consumption Consumption 2 2 2 1 2 1 a h a O O (nmol O 0 (nmol O 0 5 5 15 25 15 25 Acclimation Temperature (°C) Acclimation Temperature (°C)

E. F. * 12 a 12 10 b 10 ab 8 h 8 a h a 6 h 6 a h h 4 4 h 2 2 RCR (State III/State IV) RCR (State III/State IV) 0 0 5 5 15 25 15 25 Acclimation Temperature (°C) Acclimation Temperature (°C) -1 -1 Figure 3.1 O2 Consumption rate (nmol O2 min mg protein ) measurements for liver (A,C,E) and brain (B,D,F) mitochondria from northern (black circles, solid line) and southern (white squares, dashed line) Fundulus heteroclitus acclimated to 5, 15 and 25°C. Letters that differ indicate a significant difference within a population as determined by planned comparison t-tests followed by FDR correction for liver state III (A) and liver state IV (C). Letters that differ indicate a significant difference within a population as determined by pairwise comparisons using a Mann Whitney U rank sum test followed by FDR correction for liver RCRs (E), brain state III (B), brain state IV (D), and brain RCRs (F). An asterisk indicates a significant difference between populations at a given acclimation temperature as determined by the previously stated post-hoc analyses.

64 Table 3.2 p-values for two-way ANOVA on whole animal measurements in thermally acclimated northern and southern F. heteroclitus. An asterisk indicates p<0.05 following FDR correction

Measurement Population Temperature Population*Temperature Liver State III (Fig. 2A) 0.055 0.025* 0.012* Brain State III (Fig. 2B) 0.357 0.030* 0.117 Liver State IV (Fig. 2C) 0.147 0.067 0.032* Brain State IV (Fig. 2D) 0.529 0.852 0.228 Liver RCR (Fig. 2E) 0.327 0.011* 0.258 Brain RCR (Fig. 2F) 0.216 0.014* 0.302 ADP/O (Fig. 3) 0.015* <0.001* 0.022* Liver State II membrane <0.001* 0.016* 0.001* potential (Fig. 4E) O2 consumption rate at highest 0.603 <0.001* 0.222 common membrane potential (Fig. 4F) Brain State II O2 consumption 0.451 0.072 0.913 rate (Fig. 5A) Brain State II membrane 0.009* 0.065 0.730 potential (Fig. 5B) D O2 consumption rate at 0.313 0.494 0.815 highest common membrane potential (Fig. 7A) D O2 consumption rate at widest 0.044* 0.224 0.259 point (Fig. 7B) D membrane potential at widest 0.115 0.933 0.614 point (Fig. 7C)

65 Table 3.3 p-values for post-hoc tests for respiration measurements in thermally acclimated northern and southern F. heteroclitus. An asterisk indicates p<0.05 following FDR correction.

Liver Brain Comparison State III State IV RCR State III State IV RCR N5 vs. S5 0.003* 0.112 0.010* 0.945 0.594 0.687 N15 vs. S15 0.794 0.484 0.643 0.945 0.709 0.687 N25 vs. S25 0.723 0.322 1.000 0.873 0.786 0.699 N5 vs. N15 0.036* 0.484 0.010* 0.945 0.728 0.699 N5 vs. N25 0.021* 0.238 0.078 0.873 0.786 0.687 N15 vs. N25 0.204 0.112 1.000 0.304 0.709 0.329 S5 vs. S15 0.677 0.213 1.000 0.945 0.459 0.687 S5 vs. S25 0.826 0.112 0.643 0.078 0.709 0.687 S15 vs. S25 0.724 0.812 0.643 0.304 0.945 0.329

1.8

1.6 b h h h

1.4 a * a 1.2 ADP/O 1.0

0.2

0.0

5 15 25 Acclimation Temperature (oC)

Figure 3.2 ADP/O ratios for northern (black bars) and southern (white bars) F. heteroclitus acclimated to 5, 15 and 25°C. Letters that differ indicate a significant difference within a population as determined by planned comparison t-tests followed by FDR correction. An asterisk indicates a significant difference between populations at a given acclimation temperature as determined by the previously stated post-hoc analysis.

67 Table 3.4 p-values for post-hoc tests for ADP/O in liver mitochondria from thermally acclimated northern and southern F. heteroclitus. An asterisk indicates p<0.05 following FDR correction.

Comparison p-value N5 vs. S5 0.324 N15 vs. S15 0.133 N25 vs. S25 0.027* N5 vs. N15 0.020* N5 vs. N25 0.002* N15 vs. N25 0.650 S5 vs. S15 0.177 S5 vs. S25 0.078 S15 vs. S25 0.726

68 3.3.2 Thermal acclimation and population effects on mitochondrial membrane potential

Proton leak kinetic curves were strikingly different between northern and southern killifish (Figure 3.3A & C). In northern killifish, the highest starting membrane potential and highest respiration rates of the proton leak kinetic curves were seen following 5°C acclimation (Figure 3.3A). Although state II membrane potential was not significantly different between 5°C and 15°C acclimations, it was lower following acclimation to 25°C in northern fish (Figure 3.3E, see Tables 3.2 and 3.5 for p-values). O2 consumption rate at the highest common membrane potential (142.4 mV) in northern fish was higher following acclimation to 5°C (Figure 3.3F, see Tables 3.2 and 3.5 for p-values). This pattern is also present at a representative ohmic membrane potential (152 mV) in the northern population

(Figure 3.3B, see Tables 3.2 and 3.6 for p-values). Proton leak kinetic curves were similar between acclimation temperatures in the southern population (Figure 3.3B). Accordingly, state II membrane potential (Figure 3.3E), respiration at the highest common membrane potential (142.4 mV, Figure 3.3F), and respiration at a representative ohmic membrane potential (142 mV, Figure 3.3D) did not differ with thermal acclimation in the southern population (see Tables 3.2, 3.5, and 3.6 for p-values). At the highest common membrane potential (142.4 mV), there was no significant difference in O2 consumption rate between populations (Figure 3.3F, see Tables 3.2 and 3.5 for p-values). However, state II membrane potential did significantly differ between populations at all acclimation temperatures, with the northern population demonstrating consistently higher membrane potentials than the southern population (Figure 3.3E, see Tables 3.2 and 3.5 for p-values).

69 A. B. ) ) -1 3.5 -1 3.5

3.0 3.0 2.5 mg protein 2.5 mg protein

-1 a -1 2.0 2.0 min Consumption 2 min Consumption 2

2 b

2 b O 1.5 O 1.5

(nmol O 1.0

(nmol O 1.0 130 140 150 160 170 5 Membrane Potential (mV) 15 25 Acclimation Temperature (oC)

C. D. ) ) -1 3.5 -1 3.5

3.0 3.0

2.5 2.5 mg protein mg protein -1 -1 2.0 2.0 h min min Consumption Consumption 2 2 2 2 h h O 1.5 O 1.5 (nmol O 1.0 (nmol O 1.0 130 140 150 160 170 5 15 25 Membrane Potential (mV) Acclimation Temperature (oC)

) F.

* -1 3.5 180 * a * 170 a b 3.0 h h 160 h 2.5 150 mg protein

-1 a 140 2.0 min 20 Consumption 2 2 h h 10 O 1.5 h b b

Membrane Potential (mV)

0 (nmol O 1.0 5 5 15 25 15 25 Acclimation Temperature (oC) Acclimation Temperature (oC)

Figure 3.3 Thermal acclimation and population effects on liver mitochondrial proton leak kinetics in Fundulus -1 -1 heteroclitus. (A, C) Proton leak kinetic curves, O2 Consumption rate (nmol O2 min mg protein ) (B) at a representative northern membrane potential (152 mV), (D) at a representative southern membrane potential (142 mV) and (F) at the highest common membrane potential (142.4 mV), and (E) state II membrane potential measured in liver mitochondria from northern (panels A & B, black bars in panel E, black circles and solid line in panel F) and southern (panels C & D, white bars in panel E, white squares and dashed line in panel F) acclimated to 5°C (black circles and solid line in panels A and C), 15°C (white squares and dashed line in panels A and C) and 25°C (white triangles and dotted line in panels A and C). Letters that differ in panels B and D indicate a significant difference as determined by one-way ANOVA. Letters that differ in panels E and F indicate a significant difference within a population and an asterisk indicates significant differences between populations at a given acclimation temperature as determined by planned comparison t-tests followed by FDR corrections.

70 Table 3.5 p-values for post-hoc tests for state II membrane potential and O2 consumption rate at the highest common membrane potential (HCmV, 142.4 mV) in liver mitochondria from thermally acclimated northern and southern F. heteroclitus. An asterisk indicates p<0.05 following FDR correction.

Comparison State II (Fig. 4E) O2 consumption rate at HCmV (Fig. 4F) 5 vs. 15 <0.001* 0.092 5 vs. 25 0.001* 0.025* 15 vs. 25 0.029* 0.152 N5 vs. N15 0.100 0.010* N5 vs. N25 0.003* 0.010* N15 vs. N25 0.011* 0.937 S5 vs. S15 0.163 0.177 S5 vs. S25 0.594 0.786 S15 vs. S25 0.085 0.152

Comparison Northern O2 consumption rate Southern O2 consumption rate (Fig. 4B) (Fig. 4D) ANOVA <0.001* 0.1333 5 vs. 15 <0.001* 0.2046777 5 vs. 25 <0.001* 0.1616897 15 vs. 25 0.5136797 0.9744742

A. )

-1 1.4 a

1.2 h a h

mg protein a h

-1 1.0

min Consumption 2

2 0.8

O 0.2

(nmol O 0.0 5 15 25 Acclimation Temperature (oC)

B. 140 a a 130 h a h h 120

110

100 20

Membrane Potential (mV) 0 5 15 25

o Acclimation Temperature ( C)

-1 -1 Figure 3.4 State II (A) O2 consumption rate (nmol O2 min mg protein ) and (B) membrane potential (mV) for brain mitochondria from northern (black bars) and southern (white bars) F. heteroclitus acclimated to 5, 15 and 25°C. Letters that differ indicate a significant difference within a population as determined planned comparison t-tests followed by FDR correction for (A) and by pairwise comparisons using a Mann Whitney U rank sum test followed by FDR correction for (B). An asterisk indicates a significant difference between populations at a given acclimation temperature as determined by the previously stated post-hoc analyses.

72 Table 3.7 p-values for post-hoc tests for state II O2 consumption rate (A) and membrane potential (B) in brain mitochondria from thermally acclimated northern and southern F. heteroclitus. An asterisk indicates p<0.05 following FDR correction.

Comparison State II O2 consumption rate State II membrane potential (Fig. 5A) (Fig. 5B) 5 vs. 15 0.745 0.324 5 vs. 25 0.917 0.324 15 vs. 25 0.745 0.324 N5 vs. N15 0.639 0.920 N5 vs. N25 0.745 0.324 N15 vs. N25 0.745 0.624 S5 vs. S15 0.343 0.443 S5 vs. S25 0.745 0.324 S15 vs. S25 0.745 0.945

73 In contrast to liver mitochondria, state II O2 consumption rate in brain mitochondria did not differ with thermal acclimation or between populations (Figure 3.4A, see Tables 3.2 and 3.7 for p-values). However, similar to liver mitochondria, there was a difference in state

II membrane potential between populations in brain mitochondria (Figure 3.4B, see Table 3.2 for p-value) with southern fish having marginally lower membrane potential than northern fish, but this difference was not detectable with post-hoc analysis (see Table 3.7 for p- values).

3.3.3 Effects of fatty acids on liver and brain mitochondria

The presence of palmitate caused a left shift in the proton leak kinetic curves for liver mitochondria in northern F. heteroclitus across acclimation temperatures (Figure 3.5A-C), but had little effect in the southern population (Figure 3.5D-F). However, when compared at the highest common membrane potential (142.4 mV) the D O2 consumption rate between the control and palmitate curves did not significantly differ between populations or with thermal acclimation (Figure 3.6A, see Tables 3.2 and 3.8 for p-values). When examined to see whether there is a difference in O2 consumption rate between the palmitate and control curves (i.e., if D O2 consumption rate is different from zero), only the northern population at

15°C came close to being significantly different (Figure 3.6B, p=0.050, see Table 3.9 for p- values). When the D O2 consumption rate between the palmitate and control curves was compared at the widest point for each population at each temperature, there was a significant difference between populations (Figure 3.6B, see Table 3.2 for p-values), with northern fish having a greater palmitate response than southern fish. However, this difference was not detected in post-hoc analyses (see Table 3.8 for p-values). When the change in membrane potential between the palmitate and control curves was compared at its widest for each

A. B. C. ) ) ) -1 -1 -1 3.5 3.5 3.5

3.0 3.0 3.0

2.5 2.5 2.5 mg protein mg protein mg protein -1 -1 -1 2.0 2.0 2.0 min min min Consumption Consumption Consumption 2 2 2 2 2 2 O O 1.5 O 1.5 1.5 (nmol O (nmol O 1.0 (nmol O 1.0 1.0 120 130 140 150 160 170 120 130 140 150 160 170 120 130 140 150 160 170 Membrane Potential (mV) Membrane Potential (mV) Membrane Potential (mV)

D. E. F. ) ) ) -1 -1 3.5 3.5 -1 3.5

3.0 3.0 3.0

2.5 2.5 2.5 mg protein mg protein mg protein -1 -1 -1 2.0 2.0 2.0 min min min Consumption Consumption Consumption 2 2 2 2 2 2 O O 1.5 1.5 O 1.5 (nmol O (nmol O 1.0 1.0 (nmol O 1.0 120 130 140 150 160 170 120 130 140 150 160 170 120 130 140 150 160 170 Membrane Potential (mV) Membrane Potential (mV) Membrane Potential (mV)

75 ) A. -1 0.8 0.6 0.4 a a a h h 0.2 mg protein

-1 0.0

Consumption -0.2 min 2 2 h O

-0.4 Δ -0.6

(nmol O -0.8 5 15 25 Acclimation Temperature (oC)

B. )

-1 1.0

a 0.5 a a h h mg protein

-1 0.0

Consumption min 2 2 O

-0.5 h Δ

(nmol O -1.0 5 15 25 Acclimation Temperature (oC)

C. 20 a a

15 a h h

10 h 5 Membrane Potential (mV) Δ 0 5 15 25 Acclimation Temperature (oC)

-1 -1 Figure 3.6 Change in O2 consumption rate (nmol O2 min mg protein ) at (A) the highest common membrane potential (142.4 mV) and (B) at the widest D O2 consumption rate and (C) the change in membrane potential at the widest D membrane potential between control and palmitate (35 µM) treatments measured in liver mitochondria from northern (black bars) and southern (white bars) F. heteroclitus across acclimation temperatures. Letters that differ indicate a significant difference within a population as determined by Mann Whitney U rank sum test followed by FDR correction (A-B) and planned comparison t-tests followed by FDR correction (C). An asterisk indicates a significant difference between populations at a given acclimation temperature as determined by previously mentioned post-hoc analysis.

76 Table 3.8 p-values for post-hoc tests differences in O2 consumption rate at the highest common membrane potential (HCmV, 142.4 mV) or at the widest difference and for the membrane potential of liver mitochondria from thermally acclimated northern and southern F. heteroclitus. An asterisk indicates p<0.05 following FDR correction.

Comparison D O2 at highest common D O2 at widest difference D membrane potential at widest membrane potential (Fig. 7A) between curves (Fig. 7B) difference between curves (Fig. 7B) 5 vs. 15 0.950 0.540 0.887 5 vs. 25 0.950 0.126 0.282 15 vs. 25 0.950 0.786 0.887

N5 vs. N15 0.950 0.786 0.887 N5 vs. N25 0.950 0.786 0.890 N15 vs. N25 0.950 0.540 0.887 S5 vs. S15 0.950 0.786 0.887 S5 vs. S25 0.950 0.786 0.887

S15 vs. S25 0.950 0.836 0.887

Population Temperature p-value Northern 5 0.119 15 0.050 25 0.222 Southern 5 0.878 15 0.149 25 0.153

77 population at each acclimation temperature, there were no significant differences based on population or thermal acclimation (Figure 3.6C, see Tables 3.2 and 3.8 for p-values).

However, on average the northern population had a decrease of approximately 10 mV with the addition of palmitate, while the southern population had an approximate decrease of 5 mV (Figure 3.6C). In brain mitochondria, there was a significant effect of palmitate on O2 consumption rate, such that in the presence of the fatty acid respiration increased and membrane potential decreased in both the northern and southern population (Figure 3.7, see

Table 3.10 for p-values). The increase in mitochondrial O2 consumption rate in the presence of palmitate was most pronounced at 15°C in the northern population and at 5°C in the southern population (Figure 3.7A and B, see Table 3.11 for p-values). There was a significant effect of temperature on mitochondrial membrane potential in the southern population where membrane potential was slightly higher at 5°C under both control and palmitate conditions

(Figure 3.7D, see Tables 3.10 and 3.11 for p-values).

3.4 Discussion

The data presented here clearly indicate that there are differences in the thermal acclimation response of liver mitochondrial respiratory capacity under ADP phosphorylating conditions (i.e., state III respiration) between killifish subspecies. In liver mitochondria from northern killifish, cold acclimation caused an increase in respiratory capacity and this effect was absent in southern killifish. Similarly, cold acclimation caused an increase in proton leak in liver mitochondria from northern killifish, but not in southern killifish. Palmitate resulted in modest uncoupling of mitochondria from northern killifish, with very little or no effect in southern killifish, suggesting there may be differences in uncoupling protein (UCP) activity between the two populations. Taken together, these data suggest that northern killifish have

A. B. ) ) * -1 -1 1.6 h 1.6 * h h h h h 1.4 1.4 mg protein mg protein -1 -1 1.2 1.2 min min Consumption Consumption 2 2 2 2 1.0 a 1.0 a O O a a a a (nmol O (nmol O 0.8 0.8 5 5 15 25 15 25 Acclimation Temperature (oC) Acclimation Temperature (oC)

C. D.

130 a a 130

a a 125 125 a a h 120 h 120 h

115 h 115 i hi Membrane Potential (mV) 110 Membrane Potential (mV) 110 5 5 15 25 15 25 Acclimation Temperature (oC) Acclimation Temperature (oC)

-1 -1 Figure 3.7 Sate II O2 Consumption rate (nmol O2 min mg protein ) and membrane potential (mV) measured for control (black circles) and palmitate (105 µM, white squares) conditions for isolated brain mitochondria from northern (A,C) and southern (B,D) Fundulus heteroclitus acclimated to 5, 15 and 25°C. Letters that differ indicate a significant difference within a given treatment and an asterisk indicates a significant difference between populations at a given acclamation temperature as determined by planned comparison t-tests followed by FDR correction for O2 Consumption rate measurements (A,B) and using a Mann Whitney U rank sum test followed by FDR correction for membrane potential measurements (C,D).

79 Table 3.10 p-values for two-way ANOVA on O2 consumption rate and membrane potential measurements in brain mitochondria from thermally acclimated northern and southern F. heteroclitus. An asterisk indicates p<0.05 following FDR correction.

Population Measurement Treatment Temperature Treatment*Temperature North O2 consumption rate <0.001* 0.482 0.530 Membrane Potential 0.013* 0.210 0.213 South O2 consumption rate <0.001* 0.232 0.890 Membrane Potential 0.008* 0.010* 0.183

Table 3.11 p-values for post-hoc tests for O2 consumption rate and membrane potential in brain mitochondria from thermally acclimated northern and southern F. heteroclitus. An asterisk indicates p<0.05 following FDR correction.

North South Comparison O2 consumption Membrane Potential O2 consumption Membrane Potential rate rate C5 vs. P5 0.319 0.185 0.024* 0.323 C15 vs. P15 0.010* 0.185 0.057 0.079 C25 vs. P25 0.226 0.545 0.147 0.662 C5 vs. C15 0.319 0.818 0.114 0.379 C5 vs. C25 0.722 0.360 0.515 0.297 C15 vs. C25 0.589 0.545 0.669 0.945 P5 vs. P15 0.951 0.279 0.515 0.011* P5 vs. P25 0.917 0.360 0.515 0.360 P15 vs. P25 0.917 0.360 0.902 0.105

80 enhanced capacity to respond to cold acclimation relative to southern killifish, possibly associated with differences in the regulation or activity of UCPs.

3.4.1 Respiratory capacity

Consistent with my predictions, I detected a strong effect of acclimation temperature on state III (ADP phosphorylating) respiratory capacity in liver mitochondria from northern killifish (Figure 3.2A). Such cold compensation has been described in other fish species (e.g.,

St-Pierre et al., 1998) and has indeed been demonstrated before in northern F. heteroclitus from Nova Scotia (Chung & Schulte, 2015). Given that this observation was made in a different population of northern F. heteroclitus (Chung & Schulte, 2015; Chung et al., submitted), these data suggest that cold compensation of respiratory capacity is a general effect within the northern subspecies of killifish. In addition, these two studies used different fuels to supply electrons to the electron transport system (ETS). Chung et al. (submitted) used pyruvate, which supplies electrons via complex I (CI), or pyruvate + succinate, supplying electrons to both CI and CII. Chung et al., (2015), used either pyruvate or succinate, supplying electrons to either CI or CII, respectively. In all scenarios, similar increases in ADP-phosphorylating (state III) respiratory capacity were seen. The consistency between these studies suggests either that cold-compensation occurs as a result of changes at multiple steps within the ETS, or at a limiting step downstream of both CI and CII. I also detected and increase in ADP/O following cold acclimation in northern killifish (Figure 3.3).

This indicates that liver mitochondria from northern killifish not only have increased capacity under state III conditions, but may also be more coupled (i.e., consume fewer moles of oxygen per mole ADP converted to ATP). Furthermore, Fangue et al. (2009) provided some evidence that there may be an increase in mitochondrial amount or volume density following

81 cold exposure in liver tissue from northern killifish. Similarly, Dhillon & Schulte (2011) detected increases in mitochondrial volume density in the muscle of northern killifish following cold acclimation. Changes in volume density, membrane composition and the up- regulation of mitochondrial function in the cold together may explain the effects of cold acclimation seen in F. heteroclitus whole-animal thermal tolerance (Chung & Schulte, 2015;

Fangue et al., 2006).

Unlike the pattern observed in northern killifish, I did not detect an effect of thermal acclimation on state III respiratory capacity in liver mitochondria from southern killifish

(Fig. 2A). A similar phenomenon has also been observed by Chung et al. (submitted). In that study, cold acclimation had limited effects on state III respiratory capacity in southern killifish, which were only detectable at higher assay temperatures than those used here.

Taken together, these data strongly suggest that northern killifish exhibit a greater degree of cold-compensation in state III mitochondrial respiratory capacity than do southern killifish.

Furthermore, there were only modest differences in ADP/O detected across acclimation temperatures in southern killifish as opposed to northern killifish (Figure 3.3). Similarly,

Dhillon and Schulte (2011) did not detect an increase in mitochondrial volume density in the muscle of southern killifish in response to cold acclimation, unlike the clear pattern of cold- compensation detected in northern killifish. Fangue et al. (2009) also found little evidence for an increase in mitochondrial volume density in southern killifish as opposed to northern killifish. Taken together these data strongly suggest that northern and southern killifish differ in their ability to alter mitochondrial capacity in response to the cold.

The differences in the extent of cold-compensation between killifish populations resulted in a significant difference in mitochondrial respiratory capacity under state III (ADP

82 phosphorylating conditions) in liver mitochondria from cold-acclimated fish (Figure 3.2A).

This result is consistent with those of previous studies (Fangue et al., 2009b; Chung et al., in prep). In contrast, differences between populations in liver mitochondrial respiratory capacity were not evident at higher acclimation temperatures (Figure 3.2A). There is some variability among studies in the detection of inter-population differences in killifish liver mitochondria at higher acclimation temperatures. Fangue et al. (2009b) detected differences between populations at all acclimation temperatures, but the direction and extent of these differences varied depending on the acclimation and assay temperature. At an assay temperature of 15°C

(which was used in the current study), there were no differences detected between populations at any acclimation temperature (Fangue et al., 2009b). Chung et al. (submitted) detected higher respiratory capacity in northern compared to southern killifish at all acclimation temperatures, although the differences between populations were generally largest in the cold-acclimated fish. The reason for the differences between studies is not clear. One possible explanation is that each study used different substrates (succinate, this study; pyruvate, Fangue et al., 2009; pyruvate or pyruvate+succinate, Chung et al., submitted). Alternatively, the studies used different combinations of northern and southern populations, and inter-population differences within the northern and southern subspecies could potentially exist.

In the current study, following acclimation to 25°C, one difference seen between populations in liver mitochondria was that southern killifish had greater ADP/Os than northern killifish, indicating southern killifish mitochondria are more coupled than northern killifish mitochondria following warm acclimation (Figure 3.3). However, this is inconsistent with state IV measurements from this study, which indicate no differences in leak respiration

83 (an indicator of mitochondrial uncoupling). This could indicate differences in thermal acclimation effects on different components of the ETS on coupling under non- phosphorylating (state IV) or phosphorylating (state III) conditions.

Unlike the clear patterns for state III respiration in liver mitochondria, the response of state III respiration in brain mitochondria to thermal acclimation was relatively modest

(Figure 3.2B), which is again consistent with the results of previous studies (Chung et al.,

2017), which did not detect evidence of cold-compensation of respiratory capacity in brain.

Indeed, Chung et al. (2017) detected a decrease in respiratory capacity with cold acclimation.

It must be noted, however, that the state III respiration measurements made by Chung et al.

(2017) were made using fuels to CI or CI and CII, with no measurements made with fuels to only CII. Furthermore, differences between the two studies may be due to the difference in measurements being made on permeabilized tissue (Chung et al., 2017) or on isolated mitochondria (current study).

Brain mitochondrial respiratory capacity under state III conditions did not differ between northern and southern of F. heteroclitus when provided with fuels through CII

(Figure 3.2B). This is not the first time few differences between populations brain mitochondrial function have been seen; (Chung et al., 2017) did not detect differences in mitochondrial respiration during coupled oxidative phosphorylation between northern and southern killifish in permeabilized brain tissue preparations.

3.4.2 Liver proton leak kinetics

Proton leak across the mitochondrial membrane is often approximated by examining state II or IV respiration with oligomycin added to inhibit the ATP-synthase, as oxygen consumption under non-phosphorylating conditions is assumed to be largely due to proton

84 conductance across the membrane (Brand, 2011). I detected few differences between populations and little effect of acclimation temperature on oligomycin-inhibited state IV respiration in killifish liver mitochondria (Figure 3.2C). However, accurate quantification of proton conductance across the mitochondrial inner membrane requires measuring state II or

IV respiration rate at a defined membrane potential, as membrane potential represents the driving force for proton conductance and may vary between treatment conditions

(Divakaruni & Brand, 2011). This quantification can be accomplished by constructing a proton leak kinetic curve. The measurement of proton leak kinetic curves also allow one to tease apart differences in proton conductance as opposed to differences in substrate oxidation, which can both result in increased respiration rates (Divakaruni & Brand, 2011).

Acclimation to 5°C resulted in a pronounced change in the proton leak kinetic curve in northern killifish (Figure 3.4A). This shift in the position of the curve resulted in an increase in the measured oxygen consumption rate at the highest common membrane potential (142.4 mV), which indicates an increase in proton conductance in 5°C acclimated fish (Figure 3.4F). This increase in proton conductance is also evident when examined at a representative membrane potential that fell within the ohmic (linear) portions of the northern curves (Figure 3.4B). In contrast, the changes in the proton leak kinetic curve with acclimation temperature were much smaller in southern killifish, and the curve for fish acclimated to 15°C was slightly left-shifted compared to the curves at other acclimation temperatures (Fig. 4C). However, this effect was not statistically significant when considering state II membrane potential across acclimation temperatures (Figure 3.4E), O2 consumption rate at the highest common membrane potential (Figure 3.4F), or when O2 consumption rate is compared at a representative membrane potential within the ohmic

85 portion of southern curves (Figure 3.4D). The initial state II membrane potential is significantly lower following warm acclimation in northern killifish liver mitochondria

(Figure 3.4E), suggesting a decrease in mitochondrial capacity following warm acclimation.

However, a decrease in capacity in northern killifish liver mitochondria is not significant in other measures, such as state III respiration following warm acclimation, although the trend may be present.

The net result of these changes in the two populations is that, at the highest common membrane potential (142.4 mV), the populations do not significantly differ in proton conductance at any acclimation temperature (Figure 3.4F). However, there is a trend that northern fish liver mitochondria have greater proton conductance 5°C than the southern population, although this is not statistically significant. The proton leak kinetic curve for northern fish at 5°C also has higher initial membrane potential under state II conditions, suggesting that there may be an increase in substrate oxidation or ETS capacity following cold acclimation (Figure 3.4A). Yet, the state II membrane potential in northern killifish liver mitochondria acclimated to 5°C does not significantly differ from that at 15°C (Figure 3.4E).

The trend of a slightly increased state II membrane potential in northern killifish livers following cold acclimation is supported by the increase in state III respiration (Figure 3.2A), the coupling ratio (Figure 3.2E) and ADP/O (Figure 3.3) fueled by succinate to CII seen in the northern population following cold acclimation. In contrast, southern killifish do not demonstrate this effect (Figure 3.2A, 3.4C & D).

The data presented here provide a clear example of the difficulty of interpreting only measures of state IV respiration in terms of proton leak and RCR in terms of mitochondrial coupling. The increase in proton conductance seen following 5°C acclimation in northern

86 killifish livers (Figure 3.4A, B, and D) suggests increased proton leak or increased uncoupling following cold acclimation. However, this is in direct contrast to the ADP/O ratios (Figure 3.3), state IV respiration rates (Figure 3.2C), and RCRs (Figure 3.2E) observed in the same liver mitochondrial samples, which indicate no change in leak respiration and an increase in mitochondrial coupling following cold acclimation. This suggests that liver mitochondria from cold-acclimated northern fish experience both an increase in proton conductance and an increase in mitochondrial capacity.

When considering the proton leak kinetic curve for northern killifish at 5°C alone, there is a trend toward increased substrate oxidation at higher membrane potentials (i.e., a higher starting membrane potential), suggesting an increase in capacity (Figure 3.4A). Yet, at lower membrane potentials we see an increase in proton conductance (Figure 3.4A). These two differences coincide with different portions of the proton leak kinetic curve, with the increase proton conductance appearing to occur in the ohmic (linear) portion and increased capacity appearing within the non-ohmic (exponential) portion. The ohmic and non-ohmic behavior of proton leak rates have been described as being due to the biophysical properties of bilayer membranes (Jastroch et al., 2012; Nicholls, 1977). It is suggested that this is due to constant conductance across the membrane at low proton motive forces (ohmic) and that at high values of the proton motive force, conductance increases substantially with an increase in proton motive force (non-ohmic), causing increased proton slip (i.e., fewer protons are translocated across the membrane per O2 consumed) (Murphy, 1989). This suggests that differences in capacity or leak observed at different membrane potentials may be due to in part to differences in the biophysical properties of the whole system at different membrane potentials. In part, this may explain why we observe signs of increased capacity at higher

87 membrane potentials and increased leak at lower membrane potentials. However, it must be noted that we do not have simultaneous membrane potential measurements corresponding to state III, state IV, RCR, or ADP/O respiration rates.

Taken together, these results suggest that northern, but not southern, killifish recruit substantial changes in mitochondrial capacity and proton conductance in response to cold acclimation. The effects of cold acclimation on mitochondrial capacity and proton conductance in northern killifish are in contrast to those observed in few other studies that have used proton leak kinetic curves to investigate the effects of cold acclimation on proton leak (Jastroch et al., 2007; Trzcionka et al., 2008). Trzcionka et al. (2008) and Jastroch et al.

(2007) observed a marked decrease in proton conductance in response to cold acclimation in liver mitochondria from cane toads (Bufo marinus) and liver mitochondria from common carp (Cyprinus carpio), respectively. Both studies interpreted the decrease in proton leak as a mechanism to increase mitochondrial efficiency as an energy saving mechanism related to cold-induced metabolic depression. In contrast, in northern killifish we observe an increase in mitochondrial capacity and proton leak in the cold, suggesting a pattern of cold compensation. Similarly, in goldfish, dos Santos et al. (2012) detected increases in state IV respiration rate in skeletal muscle mitochondria with cold acclimation, suggesting increases in proton conductance in the cold. However, they did not develop proton leak kinetic curves to accurately estimate proton leak at a common membrane potential. Interestingly, in goldfish, state III and state IV respiration rates increased in parallel in response to cold acclimation, resulting in no change in RCR (dos Santos et al., 2012). Whereas in northern killifish we observed a greater change in state III respiration compared to state IV respiration, resulting in an increase in RCR following cold acclimation in northern killifish.

88 There are two strategies in response to cold acclimation: cold compensation in which function is up-regulated to counteract thermal effects or cold suppression whereby function is down-regulated in order to conserve energy. Trzcionka et al. (2008) and Jastroch et al. (2007) provide evidence for metabolic suppression in response to cold acclimation in two different ectothermic species. In killifish, there is evidence that routine metabolic rate (MO2routine) decreases following cold acclimation in both the northern and southern populations (Healy &

Schulte, 2012). However, maximum metabolic rate (MO2max) does not change with thermal acclimation and combined with the decrease in MO2routine actually results in an increase in aerobic scope (Healy & Schulte, 2012). This indicates a greater ability to increase aerobic metabolism in order to carry out functions following cold acclimation. Furthermore, Fangue et al. (2008) observed that acute exposure to 5°C resulted in a decrease in critical swimming speed Ucrit to roughly 25% of that seen at 18°C, whereas acclimation to 5°C resulted in Ucrit closer to 40% of what was seen at 15°C and 20°C when swum at 5°C. This pattern was not present for the southern population in that study (Fangue et al., 2008) and while not directly comparable, this may indicate that cold acclimation improves swimming performance in northern killifish in the cold. Taken together, my study as well as other evidence for increased performance of the northern killifish in the cold suggests that a general strategy for northern killifish may be to compensate in the cold and that this cold-compensation may not be present in southern killifish.

3.4.3 Effects of palmitate on liver and brain mitochondria

In liver mitochondria, addition of palmitate tended to left-shift the proton-leak kinetic curves, primarily in the northern population (Figure 3.6). Consistent with my predictions, this effect occurred at all acclimation temperatures in northern killifish, whereas there was

89 limited effect of palmitate in liver mitochondria from southern killifish (Figure 3.6). A left shift of the proton leak kinetic curves represents an increase in proton conductance. Palmitate is thought to act as an uncoupler by activating uncoupling proteins (UCP family members).

In the liver, UCP1 is the dominant protein isoform. In Chapter 2, I demonstrated that Ucp1 mRNA levels in the liver are greater in northern fish than in southern fish. If this pattern is also present at the protein level, this could be consistent with the differences in palmitate stimulated proton conductance observed here. I also showed that Ucp1 mRNA levels changed with thermal acclimation (Chapter 2) and predicted that the extent of mitochondrial uncoupling would change accordingly. However, I observed no clear effects of thermal acclimation on palmitate-induced uncoupling in liver mitochondria from both northern and southern F. heteroclitus.

Similarly, in brain mitochondria, palmitate addition significantly increased brain mitochondrial respiration and decreased membrane potential, consistent with a modest uncoupling effect (Figure 3.8). Ucp5 is the most highly expressed uncoupling protein isoform in the brain (Chapter 2), and it also increases in response to cold acclimation in northern, but not southern, killifish. Based on this mRNA expression data for Ucp5, I predicted greater uncoupling in the northern population than in the southern population and a decrease in the uncoupling effect with increases in acclimation temperature in the northern population, but not in the southern population (Chapter 2, Figure 2.3E). However, no such pattern was evident in the data presented here (Figure 3.8). Thus, the mRNA and functional data are not consistent in this tissue. There are several possible reasons why this might occur. First, it is possible that differences observed at the mRNA level are not translated into differences at the protein level, and thus are not correlated with functional uncoupling. This is because the

90 amount of mRNA present does not necessarily reflect the amount of protein present (e.g.,

Richards et al., 2008) and depending on post-translational modifications or protein turnover the amount of protein present may not reflect activity or function (Maier et al., 2009).

Alternatively, it is possible that the uncoupling effect of palmitate that we observe is not due to an action on uncoupling proteins.

While the presence of fatty acids appears to cause at least modest uncoupling in liver and brain mitochondria of F. heteroclitus, we cannot confirm that this uncoupling is due to the presence of UCPs, because we could not inhibit this palmitate-uncoupling effect using known UCP inhibitors. The exact mechanism through which fatty acids act on UCPs is still debated (Klingenberg & Winkler, 1985; Garlid et al., 1996; Rial et al., 2004; Shabalina et al.

2004; Jastroch et al., 2007) and there could be additional effects of fatty acids on the mitochondrion that could be causing the modest uncoupling observed here. There is some evidence that fatty acids can influence membrane stability and ion pumps (Ibarguren et al.,

2014; Zavodnik et al., 1996), suggesting there are other mechanisms through which palmitate may be acting, independent of uncoupling proteins, in order to cause the resulting uncoupling.

3.4.4 Conclusions

In this study, we have characterized changes in mitochondrial O2 consumption rate and membrane potential in response to thermal acclimation and in the presence of a fatty acid in liver and brain mitochondria of two populations of Fundulus heteroclitus. Palmitate caused modest uncoupling in liver and brain mitochondria and this uncoupling effect did not differ with thermal acclimation, which is inconsistent with previously observed changes in

Ucp mRNA expression following thermal acclimation (Chapter 2). Additional exploration

91 into thermal acclimation effects on changes in mitochondrial volume density and changes in the amount UCP protein amount would clarify and add to the findings found here. Cold acclimation resulted in increases in state III respiration, RCRs and ADP/O ratios as well as an increase in proton conductance and membrane potential in the liver mitochondria of putatively cold-adapted northern killifish. Thermal acclimation had little effect on mitochondrial respiration measurements and membrane potential in southern killifish.

Together these data suggest that northern killifish have the capacity to increase function in the cold (cold compensation) while southern killifish do not show any evidence of being able to do so, suggesting putative thermal adaptation between populations from different thermal regimes.

92 Chapter 4: Discussion and conclusions

In my thesis, I have investigated mitochondrial uncoupling in response to thermal acclimation in two populations of Atlantic killifish, Fundulus heteroclitus. With respect to my thesis objectives, I have been able to identify the gene sequences of 5 Ucp isoforms in the

F. heteroclitus genome and confirm these sequences with a phylogenetic and synteny analysis (Chapter 2). I have characterized the tissues in which these isoforms are expressed and described the changes in Ucp mRNA expression following thermal acclimation in both populations of killifish (Chapter 2). I have also conducted an investigation into the potential function of the Ucps within the mitochondria in response to thermal acclimation by characterizing fatty-acid stimulated uncoupling (Chapter 3). Furthermore, I have characterized changes in basic mitochondrial function in the livers and brains from both populations of F. heteroclitus in response to thermal acclimation (Chapter 3). Overall, my data demonstrate that mitochondrial responses to temperature differ between killifish populations, and suggest that Ucps may be playing an important role in thermal acclimation and potentially thermal adaptation in this species.

4.1 Ucp mRNA expression and thermal acclimation

The five Ucp isoforms that I identified in F. heteroclitus were expressed across brain, gill, liver and skeletal muscle tissues and Ucp mRNA displayed tissue-specific expression patterns. Each tissue had one most highly expressed isoform: Ucp5 in brain, Ucp3L in gill,

Ucp1 in liver and Ucp3 in muscle, consistent with previous studies in fish (Jastroch et al.,

2005; dos Santos et al., 2013; Tseng et al., 2011; Wen et al., 2015). Gill tissue displayed the most diversity in Ucp isoform expression, potentially indicating that the different UCP isoforms may play separate functional roles or may differ by cell types even within a tissue.

93 Furthermore, the most highly expressed Ucp isoform mRNA across any tissue was that of

Ucp1 in killifish liver tissue (Chapter 2). This is consistent with the high expression of Ucp1 across other taxa, and in mammalian brown adipose tissue in particular, where the presence of high amounts of Ucp1 are crucial to its thermogenic role (Nicholls, 2001). The skeletal- muscle specificity of Ucp3 expression in killifish is also consistent across animal taxa (Vidal-

Puig et al., 1997), further indicating that the tissue-specific expression patterns of the Ucp isoforms are conserved and potentially of functional importance.

The mRNA expression pattern of all Ucp isoforms in response to thermal acclimation varied greatly depending on isoform, tissue, and killifish population. This may be indicative of each isoform having a different functional role in response to thermal acclimation and this role varying depending on the tissue and population being investigated. One possible mechanism that could account for differences in gene expression between acclimation temperatures could be differences in mitochondrial volume density with thermal acclimation.

However, if changes in mitochondrial volume density within a tissue were causing the changes I observed in Ucp mRNA expression, without any changes in the chemical and physical properties of the mitochondria themselves, it would be expected that all isoforms within a given tissue would be changing in a similar manner to each other. In my thesis, I have found that within a single tissue the Ucp isoforms are responding in different ways to acclimation temperature suggesting that it is unlikely to be an overall change to the tissue

(such as mitochondrial volume density) driving the observed Ucp mRNA expression patterns.

Although many of the isoforms changed in different ways with thermal acclimation across tissues, there were some consistent patterns seen in response to thermal acclimation

94 and between populations. For example, in F. heteroclitus, Ucp5 mRNA expression in northern killifish displayed the same pattern of decreasing mRNA expression with increasing acclimation temperature across all four tissues (Chapter 2). Ucp5 is an isoform not generally considered to be part of the core Ucp family, but there is some evidence that it shares similar properties to other UCP isoforms (Hoang et al., 2012). This may indicate that Ucp5 has a common function in response to thermal acclimation across tissues in F. heteroclitus. It is possible that the general decrease in Ucp5 expression could be related to an overall suppression of mitochondrial activity at higher temperatures due to thermally-driven increases in the biochemical reaction rates leading to increases in ROS production (Abele et al., 2002; Hochachka & Somero, 2002). This is consistent with data from F. heteroclitus permeabilized brain tissue (Chung et al., 2017), the tissue where Ucp5 is most highly expressed, and isolated liver mitochondria (Chung & Schulte, 2015) which both show decreased mitochondrial function in response to 33 °C acclimation. However, in Chapter 3, I observed no decreases in mitochondrial function in response to 25 °C acclimation, although

Ucp5 mRNA decreases at this temperature in liver (Chapter 2). Furthermore, if there is an overall down-regulation of mitochondrial function and mRNA expression, we would expect to see mRNA expression of all Ucp isoforms decrease with increasing acclimation temperature, which is not the case (Chapter 2).

Another clear pattern in my data is that for many of the isoforms across tissues, Ucp expression in northern killifish appears to change with thermal acclimation, whereas expression in the southern killifish does not. This may indicate that the local thermal conditions from which these two populations come influence their mitochondrial responses to thermal acclimation. Indeed, this is also suggested by other studies in these populations

95 and similar populations of killifish (e.g., Fangue et al., 2009b). This is interesting because while both populations experience similar variation in temperature throughout the seasons, the range of temperatures at the northern latitudes are colder than those at the southern latitudes (Fangue et al., 2006). This may indicate that animals or populations that experience colder temperatures throughout the year may have a greater plasticity of thermal responses than those that experience overall warmer temperatures. Further investigation would be required to see if this is a general pattern where plasticity is influenced by the range of temperatures associated with a given latitude. Overall, these patterns are consistent with the hypothesis that northern killifish have a different level of plasticity in mitochondrial function compared to southern killifish.

In general, it is difficult to interpret changes in mRNA expression as having functional consequences. Firstly, mRNA expression levels are not always correlated with protein levels (de Sousa Abreu et al., 2009; Greenbaum et al., 2003; Koussounadis et al.,

2015; Lundberg et al., 2010; Schwanhäusser et al., 2011; Vogel et al., 2010), although there are scenarios in killifish where the level of mRNA has been indicative of protein level and activity (e.g., Crawford & Powers, 1989). By measuring mRNA expression alone at one time point, we also miss key information such as post-transcriptional and -translational modifications, and protein turnover that ultimately determine protein function.

4.2 Relationships between Ucp mRNA and function

Many of these limitations in interpreting mRNA data may be factors in the disassociation between Ucp mRNA expression and Ucp function observed in my thesis. The two tissues where both expression and function were investigated were killifish liver and brain. In Chapter 2, I showed that Ucp1 was the most highly expressed isoform of any

96 isoform across any tissue in F. heteroclitus. Ucp1 in liver was more highly expressed in the northern population than in the southern population (Chapter 2). If function is related to mRNA expression, it would have been expected that fatty-acid induced uncoupling would be greatest in the northern population. Palmitate did left-shift the proton-leak kinetic curves in liver mitochondria from northern killifish than from southern killifish, indicating a greater increase in proton conductance in the presence of palmitate (Chapter 3). Therefore, the trend of greater fatty-acid induced uncoupling in northern killifish mitochondria did appear to be present (Chapter 3), but this difference between populations was not significant.

Furthermore, Ucp1 mRNA expression in liver showed the general pattern of decreased expression following cold acclimation to 5 °C and warm acclimation to 25 °C in both populations (Chapter 2). Inconsistent with the Ucp1 mRNA expression pattern, fatty-acid induced uncoupling in liver mitochondria also did not appear to be dependent on acclimation temperature (Chapter 3).

In brain tissue, Ucp5 was the most highly expressed isoform (Chapter 2). Ucp5 expression in the brain decreased in response to acclimation to warmer temperatures, and was generally higher in northern fish than in southern (Chapter 2). Therefore, in isolated mitochondria from brain tissue, I expected to see a decrease in fatty-acid induced uncoupling with acclimation to warm temperatures and uncoupling to be greatest at 5 and 15 °C. Fatty- acid induced uncoupling by palmitate is associated with a decrease in mitochondrial membrane potential and an increase in mitochondrial O2 consumption rates. However, there does not appear to be a significant drop in mitochondrial membrane potential or increase in

O2 consumption rate following the addition of fatty acids in killifish brain mitochondria. The only conditions under which there was a significant increase in O2 consumption rate was at

97 15 °C in the northern population and at 5 °C in the southern population (Chapter 3). While this is somewhat consistent with the expression of Ucp5 mRNA in the brain, the fact that there is no corresponding significant decrease in membrane potential is inconsistent with uncoupling protein function. I also expected fatty-acid induced uncoupling in isolated brain mitochondria to be greater in the northern population than in the southern population.

However, the data show that fatty-acid induced uncoupling does not appear to differ between the populations (Chapter 3).

There are some further limitations in interpreting the fatty-acid induced uncoupling that I observed as being due to UCPs. I was not able to inhibit fatty-acid induced uncoupling using classic UCP inhibitors such as GDP or GTP. However, the contributions of other proton-moving proteins were also excluded as a cause of observed uncoupling (Chapter 3). A couple of possibilities exist as to why purine nucleotides did not work to inhibit fatty-acid induced uncoupling. Firstly, the amino acid sequence hypothesized to be associated with purine nucleotide binding (amino acids 261-269, Bouillaud et al. 1994) differs by two amino acids between mammalian and killifish genomes. Phe267 is Tyr267 in the killifish UCP1 and

Ala264 is Thr264 in the killifish UCP1 (based on NCBI killifish genome, v. 101). While the

AlaàThr substitution has not been investigated, there is some evidence the PheàTyr substitution did not appear to qualitatively alter UCP function (Bouillaud et al. 1994). Other amino acids that may be involved in UCP purine nucleotide binding include three Arg residues (Modriansky et al., 1997) of which killifish have all three. Another possibility is that any observed fatty-acid induced uncoupling is not due to UCPs at all and is a result of another effect palmitate is having on the mitochondrial membrane (Ibarguren et al., 2014;

Zavodnik et al., 1996). If any fatty-acid induced uncoupling is due to UCPs, one further

98 limitation is that there is no way of determining which isoform(s) is contributing based on this assay. The contributions of each isoform may be further elucidated by conducting western blots to look at the actual protein amount present for each of the UCP isoforms in liver and brain tissue. Isolated mitochondria were saved from the current study to conduct assays to determine protein amount in the future.

If the fatty-acid induced uncoupling seen here does have to do with UCPs, the fact that the uncoupling does not significantly differ with thermal acclimation may suggest that

UCPs are not playing an important role in thermal acclimation. This further emphasises that it is difficult to make functional conclusions based solely on mRNA expression. However, the fact that we were unable to inhibit fatty-acid induced uncoupling with known inhibitors of UCPs may suggest that the pathways being assessed in these function assays are UCP- independent.

Furthermore, only one time-point was investigated in this study following a three- week acclimation. The lack of differences in fatty-acid induced uncoupling may be a result of responses that happened earlier in the time-course of the acclimation. For example, if exposure to a thermal extreme does result in increased ROS production or oxidative damage, the mechanisms required to mitigate this may be up-regulated initially and possibly return to control levels following transition to a new steady state. Moreover, there are many mechanisms and enzymes involved in the mitigation of ROS production and oxidative damage, such as superoxide dismutases and reductases (Sheng et al., 2014). It could be that different mechanisms may be up-regulated at different times during the mitochondrial thermal response and that UCPs may be involved earlier on in the acclimation process.

However, changes in mRNA expression and function of UCPs have been seen to change in

99 as little as a few hours (e.g., Tseng et al., 2011) and have been shown after as long as two months (e.g., Mark et al., 2006). In order to elucidate the potential time-course of UCP any

UCP involvement in antioxidant mechanisms with thermal acclimation, several time-points over the course of acclimation to both cold and warm temperatures would need to be investigated.

4.3 Mitochondrial responses to thermal acclimation between populations

Despite fatty-acid induced uncoupling appearing to not be responsive to thermal acclimation, other liver mitochondrial properties did appear to be affected by acclimation temperature, particularly in the northern F. heteroclitus. Unlike liver mitochondria, many properties of brain mitochondria did not appear to differ with thermal acclimation or between populations (Chapter 3), suggesting tissue-specific responses to thermal change. Liver mitochondria from northern killifish demonstrated increases in ADP-phosphorylating respirations (state III), mitochondrial coupling (RCRs & ADP/O), and proton conductance following cold-acclimation (Chapter 3). These changes in northern killifish liver mitochondria are consistent with an up-regulation of mitochondrial capacity as compensation for decreased enzyme activity and membrane fluidity associated with cold exposure

(Guderley, 2004; dos Santos et al., 2012; Chung & Schulte, 2015). Unlike northern killifish liver mitochondria, cold-compensation was not evident in southern killifish liver mitochondria. This suggests differences between the populations in the ability to change mitochondrial physiology in response to thermal acclimation.

Similarly, UCP expression was higher in northern killifish than southern killifish for many isoforms across tissues (Chapter 2). However, this was only reflected by modestly greater uncoupling in the liver mitochondria of northern killifish (Chapter 3). If northern

100 killifish are increasing mitochondrial function in the cold, this may be evidence of adaptation to more northern latitudes and colder temperatures. The increase in liver mitochondrial function plus the increased susceptibility of the membranes to lipid peroxidation in the cold

(Grim et al., 2010) may be a reason that UCP expression (and possibly uncoupling capacity) is often greater in the northern population than in the southern population.

The greater responses of the northern population to thermal change (as indicated by both Ucp mRNA expression and mitochondrial function) may have to do with the evolutionary history of F. heteroclitus. The southern subspecies of Atlantic killifish, F. heteroclitus heteroclitus is the ancestral subspecies first identified in South Carolina (Morin

& Able, 1983), and any changes in thermal responses in the northern subspecies, F. heteroclitus macrolepidotus may represent adaptations to life in colder environments. It should be noted that in this thesis only one population has been investigated from each of the subspecies and therefore conclusions cannot necessarily be extended to an entire subspecies.

However, there is some evidence that other northern and southern populations do have similar liver mitochondrial responses as seen in this study (Fangue, Richards, & Schulte,

2009b). Therefore, we can hypothesize that differences in mitochondrial properties seen between northern and southern populations may be reflective of subspecies differences and may be reflective of adaptations to different thermal regimes.

4.4 Future Directions

To further elucidate the potential function of UCPs in thermal acclimation and adaptation in ectothermic organisms, the information from this thesis could inform future studies. Firstly, the amount of UCPs present as determined by western blotting would allow for further connection between the level of gene expression and mitochondrial function.

101 Furthermore, a time-course acclimation study with sampling to determine mRNA expression, protein expression, and mitochondrial function at many different time points throughout thermal acclimation could provide information on if and when UCPs are important in responding to thermal change. It is highly likely that other antioxidant mechanisms, aside from UCPs, are also important to responses to thermal change and therefore assays to determine the contributions of these other mechanisms at time points through acclimation could potentially inform the relative influence of UCPs in the context of antioxidant responses. With respect to the increase in mitochondrial function in the northern population but not the southern, more accurate measures of mitochondrial volume density with thermal acclimation through microscopy would give us an indication of how mitochondrial function is changing in the context of the whole tissue.

4.5 Conclusions

In my thesis, I have shown that Ucp mRNA expression responds to thermal acclimation and differs between populations of northern and southern F. heteroclitus. I have shown that modest differences in liver mitochondrial uncoupling between populations may correspond to differences between populations in Ucp1 gene expression. However, the observed fatty acid induced uncoupling does not appear to be dependent on acclimation temperature, despite observed changes in Ucp1 mRNA expression with thermal acclimation.

It is difficult to therefore tease apart whether UCPs may be contributing to thermal acclimation in these two populations of killifish. Furthermore, there are marked differences in the overall functional responses of northern and southern killifish liver mitochondria to thermal acclimation, which may be suggestive of adaptation to local thermal regimes. Brain and liver mitochondria respond differently to thermal acclimation and further suggest that

102 tissue specific processes may be involved in the thermal responses of killifish. These findings contribute to a greater understanding of the mechanisms behind temperature responses in ectothermic species and the potential role for uncoupling proteins in these responses.

103 Bibliography

Abele, D., & Puntarulo, S. (2004). Formation of reactive species and induction of antioxidant defence systems in polar and temperate marine invertebrates and fish. Comparative Biochemistry and Physiology Part A: Molecular & Integrative Physiology, 138(4), 405– 415.

Abele, D., Heise, K., Pörtner, H. O., & Puntarulo, S. (2002). Temperature-dependence of mitochondrial function and production of reactive oxygen species in the intertidal mud clam Mya arenaria. Journal of Experimental Biology, 205, 1831–1841.

Adams, S. M., Lindmeier, J. B., & Duvernell, D. D. (2006). Microsatellite analysis of the phylogeography, Pleistocene history and secondary contact hypotheses for the killifish, Fundulus heteroclitus. Molecular Ecology, 15(4), 1109–1123.

Azzu, V., Jastroch, M., Divakaruni, A. S., & Brand, M. D. (2010). The regulation and turnover of mitochondrial uncoupling proteins. Biochimica et Biophysica Acta, 1797, 785–791.

Ballinger, M. A., Hess, C., Napolitano, M. W., Bjork, J. A., & Andrews, M. T. (2016). Seasonal changes in brown adipose tissue mitochondria in a mammalian hibernator: from gene expression to function. American Journal of Physiology - Regulatory, Integrative and Comparative Physiology, 311(2), R325-R336.

Bernardi, G., Sordino, P., & Powers, D. A. (1993). Concordant mitochondrial and nuclear DNA phylogenies for populations of the teleost fish Fundulus heteroclitus. Proceedings of the National Academy of Sciences, 90, 9271-9274.

Boss, O., Samec, S., Paoloni-Giacobino, A., Rossier, C., Dulloo, A., Seydoux, J., Muzzin, P., & Giacobino, J.-P. (1997). Uncoupling protein-3: A new member of the mitochondrial carrier family with tissue-specific expression. FEBS Letters, 408, 39–42.

Bouillaud, F., Arechaga, I., Petit, P. X., Raimbault, S., LeviMeyrueis, C., Casteilla, L., Laurent, M., Rial, E., & Ricquier, D. (1994). A sequence related to a DNA recognition element is essential for the inhibition by nucleotides of proton transport through the mitochondrial uncoupling protein. The EMBO Journal, 13(8), 1990–1997.

Bouillaud, F., Couplan, E., Pecqueur, C., & Ricquier, D. (2001). Homologues of the uncoupling protein from brown adipose tissue (UCP1): UCP2, UCP3, BMCP1 and UCP4. Biochimica et Biophysica Acta - Bioenergetics, 1504(1), 107–119.

Bradford, M. M. (1976). A rapid and sensitive method for quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Analytical Biochemistry, 72, 248-254.

Brand, M. D. (1990). The contribution of the leak of protons across the mitochondrial inner membrane to standard metabolic rate. Journal of Theoretical Biology, 145, 267-286.

104 Brand, M. D. (2000). Uncoupling to survive? The role of mitochondrial inefficiency in ageing. Experimental Gerontology, 35, 811–820.

Brand, M. D., & Esteves, T. C. (2005). Physiological functions of the mitochondrial uncoupling proteins UCP2 and UCP3. Cell Metabolism, 2, 85–93.

Brand, M. D., Affourtit, C., Esteves, T. C., Green, K., Lambert, A. J., Miwa, S., Pakay, J. L., & Parker, M. (2004). Mitochondrial superoxide: production, biological effects, and activation of uncoupling proteins. Free Radical Biology and Medicine, 37(6), 755–767.

Brand, M. D., Chien, L.-F., Ainscow, E. K., Rolfe, D. F. S., & Porter, R. K. (1994). The causes and functions of mitochondrial proton leak. Biochemica et Biophysica Acta, 1187, 132- 139.

Brand, M. D., Chien, L. F., & Diolez, P. (1994). Experimental discrimination between proton leak and redox slip during mitochondrial electron transport. Biochemical Journal, 297, 27–29.

Brown, J. H., Gillooly, J. F., Allen, A. P., Savage, V. M., & West, G. B. (2004). Toward a metabolic theory of ecology. Ecology, 85(7), 1771–1789.

Campbell, H. A., Fraser, K. P. P., Bishop, C. M., Peck, L. S., & Egginton, S. (2008). Hibernation in an Antarctic fish: on ice for winter. PLoS ONE, 3, e1743.

Cannon, B., & Nedergaard, J. (2004). Brown adipose tissue: function and physiological significance, 84(1), 277–359.

Cashon, R. E., Van Beneden, R. J., & Powers, D. A. (1981). Biochemical genetics of Fundulus heteroclitus (L.). IV. Spatial variation in gene frequencies of Idh-A, Idh-B, 6-Pgdh-A, and Est-S. Biochemical Genetics, 19(7/8), 715–728.

Chan, W.-P., Chen, I.-C., Colwell, R. K., Liu, W.-C., Huang, C.-Y., & Shen, S.-F. (2016). Seasonal and daily climate variation have opposite effects on species elevational range size. Science, 351(6280), 1437–1439.

Chung, D. J., & Schulte, P. M. (2015). Mechanisms and costs of mitochondrial thermal acclimation in a eurythermal killifish (Fundulus heteroclitus). The Journal of Experimental Biology, 218, 1621–1631.

Chung, D. J., Bryant, H. J., & Schulte, P. M. (2017). Thermal acclimation and subspecies- specific effects on heart and brain mitochondrial performance in a eurythermal teleost (Fundulus heteroclitus). The Journal of Experimental Biology, 220, 1459-1471.

105 Considine, M. J., Goodman, M., Echtay, K. S., Laloi, M., Whelan, J., Brand, M. D., & Sweetlove, L. J. (2003). Superoxide stimulates a proton leak in potato mitochondria that is related to the activity of uncoupling protein. Journal of Biological Chemistry, 278(25), 22298–22302.

Cossins, A. R., & Prosser, C. L. (1982). Variable homeoviscous responses of different brain membranes of thermally-acclimated goldfish. Biochimica et Biophysica Acta, 687, 303– 309.

Crawford, D. L. & Powers, D. A. (1989). Molecular basis of evolutionary adaptation at the lactate dehydrogenase-B locus in the fish Fundulus heteroclitus. Proceedings of the National Academy of Sciences, 86, 9365-9369.

Crockett, E. L. (2008). The cold but not hard fats in ectotherms: consequences of lipid restructuring on susceptibility of biological membranes to peroxidation, a review. Journal of Comparative Physiology B: Biochemical, Systemic, and Environmental Physiology, 178, 795–809.

Cunningham, S. A., Wiesinger, H., & Nicholls, D. G. (1986). Quantification of fatty acid activation of the uncoupling protein in brown adipocytes and mitochondria from the guinea pig, European Journal of Biochemistry, 157, 415-420.

Dahlhoff, E., & Somero, G. N. (1993). Effects of temperature on mitochondria from abalone (genus haliotus): adaptive plasticity and its limits. The Journal of Experimental Biology, 185, 151–168. de Sousa Abreu, R., Penalva, L. O., Marcotte, E. M., & Vogel, C. (2009). Global signatures of protein and mRNA expression levels. Molecular BioSystems, 5, 1512–1526.

Deutsch, C., Ferrel, A., Seibel, B., Pörtner, H.-O., & Huey, R. B. (2015). Climate change tightens a metabolic constraint on marine habitats. Science, 348(6239), 1132–1135.

Dhillon, R. S., & Schulte, P. M. (2011). Intraspecific variation in the thermal plasticity of mitochondria in killifish. The Journal of Experimental Biology, 214, 3639–3648.

Dillon, M. E., Woods, H. A., Wang, G., Fey, S. B., Vasseur, D. A., Telemeco, R. S., Marshall, K, & Pincebourde, S. (2016). Life in the frequency domain: the biological impacts of changes in climate variability at multiple time scales. Integrative and Comparative Biology, 56(1), 14–30.

Divakaruni, A. S., & Brand, M. D. (2011). The regulation and physiology of mitochondrial proton leak. Physiology, 26, 192–205. dos Santos, R. S., Galina, A., & Da-Silva, W. S. (2012). Cold acclimation increases mitochondrial oxidative capacity without inducing mitochondrial uncoupling in goldfish white skeletal muscle. Biology Open, 2(1), 82-87.

106 dos Santos, R. S., Peçanha, F. L. M., & Da-Silva, W. S. (2013). Functional characterization of an uncoupling protein in goldfish white skeletal muscle. Journal of Bioenergetics and Biomembranes, 45, 243–251.

Echtay, K. S., Esteves, T. C., Pakay, J. L., Jekabsons, M. B., Lambert, A. J., Portero-Otin, M., Pamplona, R., Vidal-Puig, A. J., Wang, S., Roebuck, S. J., & Brand, M. D. (2003). A signalling role for 4-hydroxy-2-nenonal in regulation of mitochondrial uncoupling. The EMBO Journal, 22(16), 4103–4110.

Echtay, K. S., Murphy, M. P., Smith, R. A. J., Talbot, D. A., & Brand, M. D. (2002a). Superoxide activates mitochondrial uncoupling protein 2 from the matrix side. Journal of Biological Chemistry, 277(49), 47129–47135.

Echtay, K. S., Roussel, D., St-Pierre, J., Jekabsons, M. B., Cadenas, S., Stuart, J. A., Harper, J. A., Roebuck, S. J., Morrison, A., Pickering, S., Clapham, J. C., & Brand, M. D. (2002b). Superoxide activates mitochondrial uncoupling proteins. Nature, 415, 96–99.

Edgar, R. C. (2004). MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Research, 32, 1792-1797.

Eliason, E. J., Clark, T. D., Hague, M. J., Hanson, L. M., Gallagher, Z. S., Jeffries, K. M., Gale, M. K., Patterson, D. A., Hinch, S. G., & Farrell, A. P. (2011). Differences in thermal tolerance among sockeye salmon populations. Science, 332, 109–112.

Emre, Y., Hurtaud, C., Ricquier, D., Bouillaud, F., Hughes, J., & Criscuolo, F. (2007). Avian UCP: The killjoy in the evolution of the mitochondrial uncoupling proteins. Journal of Molecular Evolution, 65, 392–402.

Esteves, T. C., & Brand, M. D. (2005). The reactions catalysed by the mitochondrial uncoupling proteins UCP2 and UCP3. Biochimica et Biophysica Acta, 1709, 35–44.

Evans, D. H., Piermarini, P. M., & Choe, K. P. (2005). The multifunctional fish gill: dominant site of gas exchange, osmoregulation, acid-base regulation, and excretion of nitrogenous waste. Physiological Reviews, 85, 97–177.

Fangue, N. A., Hofmeister, M., & Schulte, P. M. (2006). Intraspecific variation in thermal tolerance and heat shock protein gene expression in common killifish, Fundulus heteroclitus. Journal of Experimental Biology, 209, 2859–2872.

Fangue, N. A., Podrabsky, J. E., Crawshaw, L. I., & Schulte, P. M. (2009a). Countergradient variation in temperature preference in populations of killifish Fundulus heteroclitus. Physiological and Biochemical Zoology, 82(6), 776–786.

107 Fangue, N. A., Richards, J. G., & Schulte, P. M. (2009b). Do mitochondrial properties explain intraspecific variation in thermal tolerance? Journal of Experimental Biology, 212, 514– 522.

Fedorenko, A., Lishko, P. V., & Kirichok, Y. (2012). Mechanism of fatty-acid-dependent UCP1 uncoupling in brown fat mitochondria. Cell, 151, 400–413.

Fleury, C., Neverova, M., Collins, S., Raimbault, S., Champigny, O., Levi-Meyrueis, C., Bouillaud, F., Seldin, M. F., Surwit, R. S., Riquier, D, & Warden, C. H. (1997). Uncoupling protein-2: A novel gene linked to obesity and hyperinsulinemia. Nature Genetics, 15, 269–272.

Franklin, C. E., Davison, W., & Seebacher, F. (2007). Antarctic fish can compensate for rising temperatures: thermal acclimation of cardiac performance in Pagothenia borchgrevinki. Journal of Experimental Biology, 210, 3068–3074.

Fritz, E. S., Meredith, W. H., & Lotrich, V. A. (1975). Fall and winter movements and activity level of the mummichog, Fundulus heteroclitus, in a tidal creek. Chesapeake Science, 16(3), 211–215.

Garlid, K. D., Orosz, D. E., Modrianský, M., Vassanelli, S., & Jezek, P. (1996). On the mechanism of fatty acid-induced proton transport by mitochondrial uncoupling protein. Journal of Biological Chemistry, 271(5), 2615–2620.

Gaudry, M. J., Jastroch, M., Treberg, J. R., Hofreiter, M., Paijmans, J. L. A., Starrett, J., Wales, N., Signore, A. V., Springer, M. S., & Campbell, K. L. (2017). Inactivation of thermogenic UCP1 as a historical contingency in multiple placental mammal clades. Science Advances, 3, e1602878.

Gnaiger, E. & Forstner, H. (1983). Calculation of equilibrium oxygen concentration. In Polarographic Oxygen Sensors, pp. 321-333. Berlin; Heidelberg: Springer Berlin Heidelberg.

Gnaiger, E., Méndez, G., & Hand, S. C. (2000). High phosphorylation efficiency and depression of uncoupled respiration in mitochondria under hypoxia. Proceedings of the National Academy of Sciences, 97(20), 11080-11085.

Gnaiger, E. & Kuznetsov, A. V. (2002). Mitochondrial respiration at low levels of oxygen and cytochrome c. Biochemical Society Transactions, 30, 252-258.

Gong, D. W., He, Y. F., Karas, M., & Reitman, M. (1997). Uncoupling protein-3 is a mediator of thermogenesis regulated by thyroid hormone, beta 3-adrenergic agonists, and leptin. Journal of Biological Chemistry, 272(39), 24129–24132.

Greenbaum, D., Colangelo, C., Williams, K., & Gerstein, M. (2003). Comparing protein abundance and mRNA expression levels on a genomic scale. Genome Biology, 4(9), 117.

108

Grim, J. M., Miles, D. R. B., & Crockett, E. L. (2010). Temperature acclimation alters oxidative capacities and composition of membrane lipids without influencing activities of enzymatic antioxidants or susceptibility to lipid peroxidation in fish muscle. The Journal of Experimental Biology, 213, 445–452.

Guderley, H. (1990). Functional significance of metabolic responses to thermal acclimation in fish muscle. American Journal of Physiology-Regulatory, Integrative and Comparative Physiology, 259(2), R245–R252.

Guderley, H. (2004). Metabolic responses to low temperature in fish muscle. Biological Reviews of the Cambridge Philosophical Society, 79, 409–427.

Guderley, H. (2011). Mitochondria and Temperature. In Encyclopedia of Fish Physiology, pp. 1709–1716. Elsevier.

Hardewig, I., Peck, L. S., & Pörtner, H. O. (1999). Thermal sensitivity of mitochondrial function in the Antarctic Notothenioid Lepidonotothen nudifrons. Journal of Comparative Physiology B: Biochemical, Systemic, and Environmental Physiology, 169, 597–604.

Harley, C. D. G., Randall Hughes, A., Hultgren, K. M., Miner, B. G., Sorte, C. J. B., Thornber, C. S., Rodriguez, L. F., Tomanek, L., & Williams, L. (2006). The impacts of climate change in coastal marine systems. Ecology Letters, 9, 228–241.

Hays, J. D., Imbrie, J., & Shackleton, N. J. (1976). Variations in the earth's orbit: pacemaker of the ice ages. Science, 194(4270), 1121–1132.

Hazel, J. R. (1972). The effect of temperature acclimation upon succinic dehydrogenase activity from the epaxial muscle of the common goldfish (Carassius auratus L.)--II. Lipid reactivation of the soluble enzyme. Comparative Biochemistry and Physiology. B, Comparative Biochemistry, 43, 863–882.

Hazel, J. R. & Prosser, C. L. (1974). Molecular mechanisms of temperature compensation in poikilotherms. Physiological Reviews, 54, 620-677.

Hazel, J. R., & Zerba, E. (1986). Adaptation of biological membranes to temperature: molecular species compositions of phosphatidylcholine and phosphatidylethanolamine in mitochondrial and microsomal membranes of liver from thermally-acclimated rainbow trout. Journal of Comparative Physiology B: Biochemical, Systemic, and Environmental Physiology, 156, 665–674.

Healy, T. M., & Schulte, P. M. (2012). Thermal acclimation is not necessary to maintain a wide thermal breadth of aerobic scope in the common killifish (Fundulus heteroclitus). Physiological and Biochemical Zoology, 85(2), 107–119.

109 Healy, T. M., Bryant, H. J., & Schulte, P. M. (2017). Mitochondrial genotype and phenotypic plasticity of gene expression in response to cold acclimation in killifish. Molecular Ecology, 26, 814–830.

Heise, K., Puntarulo, S., Pörtner, H. O., & Abele, D. (2003). Production of reactive oxygen species by isolated mitochondria of the Antarctic bivalve Laternula elliptica (King and Broderip) under heat stress. Comparative Biochemistry and Physiology. Toxicology & Pharmacology, 134, 79–90.

Helmuth, B. S., & Hofmann, G. E. (2001). Microhabitats, thermal heterogeneity, and patterns of physiological stress in the rocky intertidal zone. The Biological Bulletin, 201(3), 374– 384.

Hoang, T., Smith, M. D., & Jelokhani-Niaraki, M. (2012). Toward understanding the mechanism of ion transport activity of neuronal uncoupling proteins UCP2, UCP4, and UCP5. Biochemistry, 51, 4004–4014.

Hochachka, P. W., & Somero, G. N. (2002). Biochemical Adaptation: Mechanism and Process in Physiological Evolution. New York, NY: Oxford University Press.

Hughes, D. A., Jastroch, M., Stoneking, M., & Klingenspor, M. (2009). Molecular evolution of UCP1 and the evolutionary history of mammalian non-shivering thermogenesis. BMC Evolutionary Biology, 9, 4.

Hughes, J., & Criscuolo, F. (2008). Evolutionary history of the UCP gene family: gene duplication and selection. BMC Evolutionary Biology, 8, 306.

Ibarguren, M., López, D. J., & Escribá, P. V. (2014). The effect of natural and synthetic fatty acids on membrane structure, microdomain organization, cellular functions and human health. Biochimica et Biophysica Acta, 1838, 1518–1528.

Iftikar, F. I., & Hickey, A. J. R. (2013). Do mitochondria limit hot fish hearts? Understanding the role of mitochondrial function with heat stress in Notolabrus celidotus. PLoS ONE, 8(5), e64120.

Itoi, S., Kawahara, S., Kondo, H., Sakai, T., & Watabe, S. (2003). Changes in mitochondrial fatty acid composition following temperature acclimation of carp and their possible effects on FoF1-ATPase activity. Fish Physiology and Biochemistry, 29, 237–244.

Jastroch, M., Buckingham, J. A., Helwig, M., Klingenspor, M., & Brand, M. D. (2007). Functional characterisation of UCP1 in the common carp: uncoupling activity in liver mitochondria and cold-induced expression in the brain. Journal of Comparative Physiology B: Biochemical, Systemic, and Environmental Physiology, 177, 743–752.

110 Jastroch, M., Hirschberg, V., & Klingenspor, M. (2012). Functional characterization of UCP1 in mammalian HEK293 cells excludes mitochondrial uncoupling artefacts and reveals no contribution to basal proton leak. Biochimica et Biophysica Acta - Bioenergetics, 1817, 1660–1670.

Jastroch, M., Wuertz, S., Kloas, W., & Klingenspor, M. (2005). Uncoupling protein 1 in fish uncovers an ancient evolutionary history of mammalian nonshivering thermogenesis. Physiological Genomics, 22, 150–156.

Keller, M., Sommer, A. M., Pörtner, H. O., & Abele, D. (2004). Seasonality of energetic functioning and production of reactive oxygen species by lugworm (Arenicola marina) mitochondria exposed to acute temperature changes. Journal of Experimental Biology, 207, 2529–2538.

Klingenberg, M. (2001). Uncoupling Proteins--How Do They Work and How Are They Regulated. IUBMB Life, 52(3-5), 175–179.

Klingenberg, M., & Winkler, E. (1985). The reconstituted isolated uncoupling protein is a membrane potential driven H+ translocator. The EMBO Journal, 4(12), 3087–3092.

Knowles, J. R. (1980). Enzyme-catalyzed phosphoryl transfer reactions. Annual Review of Biochemistry, 49, 877-919.

Korshunov, S. S., Skulachev, V. P., & Starkov, A. A. (1997). High protonic potential actuates a mechanism of production of reactive oxygen species in mitochondria. FEBS Letters, 416, 15–18.

Koussounadis, A., Langdon, S. P., Um, I. H., Harrison, D. J., & Smith, V. A. (2015). Relationship between differentially expressed mRNA and mRNA-protein correlations in a xenograft model system. Scientific Reports, 5, 10775.

Kraffe, E., Marty, Y., & Guderley, H. (2007). Changes in mitochondrial oxidative capacities during thermal acclimation of rainbow trout Oncorhynchus mykiss: roles of membrane proteins, phospholipids and their fatty acid compositions. Journal of Experimental Biology, 210, 149–165.

Krauss, S., Zhang, C. Y., & Lowell, B. B. (2005). The mitochondrial uncoupling-protein homologues. Nature Reviews in Molecular Cell Biology, 6(3), 248–261.

Kwok, K. H.-H., Ho, P. W.-L., Chu, A. C.-Y., Ho, J. W.-M., Liu, H.-F., Yiu, D. C.-W., Chan, K.-H., Kung, M. H.-W., Ramsden, D. B., & How, S.-L. (2010). Mitochondrial UCP5 is neuroprotective by preserving mitochondrial membrane potential, ATP levels, and reducing oxidative stress in MPP+ and dopamine toxicity. Free Radical Biology & Medicine, 49, 1023–1035.

111 Lannig, G., Storch, D., & Pörtner, H. O. (2005). Aerobic mitochondrial capacities in Antarctic and temperate eelpout (Zoarcidae) subjected to warm versus cold acclimation. Polar Biology, 28, 575–584.

Liu, D., Chan, S. L., de Souza-Pinto, N. C., Slevin, J. R., Wersto, R. P., Zhan, M., Mustafa, K., de Cabo, R., & Mattson, M. P. (2006). Mitochondrial UCP4 mediates an adaptive shift in energy metabolism and increases the resistance of neurons to metabolic and oxidative stress. Neuromolecular Medicine, 8, 389–414.

Liu, Q. Y., Bai, C., Chen, F., Wang, R. P., MacDonald, T., Gu, M., Zhang, Q., Morsy, M.A., & Caskey, C. T. (1998). Uncoupling protein-3: a muscle-specific gene upregulated by leptin in ob/ob mice. Gene, 207, 1–7.

Lotrich, V. A. (1975). Summer home range and movements of Fundulus heteroclitus (Pisces: Cyprinodontidae) in a tidal creek. Ecology, 56(1), 191–198.

Lundberg, E., Fagerberg, L., Klevebring, D., Matic, I., Geiger, T., Cox, J., Älgeniäs, C., Lundeberg, J., Mann, M., & Uhlen, M. (2010). Defining the transcriptome and proteome in three functionally different human cell lines. Molecular Systems Biology, 6, 450.

Maier, T., Güell, M., & Serrano, L. (2009). Correlation of mRNA and protein in complex biological samples. FEBS Letters, 583, 3966-3973.

Mao, W., Yu, X. X., Zhong, A., Li, W., Brush, J., Sherwood, S. W., Adams, S. H., & Pan, G. (1999). UCP4, a novel brain-specific mitochondrial protein that reduces membrane potential in mammalian cells. FEBS Letters, 443, 326–330.

Mark, F. C., Lucassen, M., & Pörtner, H. O. (2006). Thermal sensitivity of uncoupling protein expression in polar and temperate fish. Comparative Biochemistry and Physiology Part D: Genomics and Proteomics, 1, 365–374.

Matsuda, J., Hosoda, K., Itoh, H., Son, C., Doi, K., Tanaka, T., Fukunaga, Y., Inoue, G., Nishimura, H., Yoshimasa, Y., Yamori, Y., & Nakao, K. (1997). Cloning of rat uncoupling protein-3 and uncoupling protein-2 cDNAs: their gene expression in rats fed high-fat diet. FEBS Letters, 418, 200–204.

McBryan, T. L., Healy, T. M., Haakons, K. L., & Schulte, P. M. (2016). Warm acclimation improves hypoxia tolerance in Fundulus heteroclitus. The Journal of Experimental Biology, 219, 474–484.

Miranda, E. J., & Hazel, J. R. (1996). Temperature-induced changes in the transbilayer distribution of phosphatidylethanolamine in mitoplasts of rainbow trout (Oncorhynchus mykiss) liver. The Journal of Experimental Zoology, 274, 23–32.

112 Modriansky, M., Murdza-Inglis, D. L., Patel, H. V., Freeman, K. B., & Garlid, K. D. (1997). Identification by site-directed mutagenesis of three arginines in uncoupling protein that are essential for nucleotide binding and inhibition. The Journal of Biological Chemistry, 272(40), 24759-24762.

Morin, R. P., & Able, K. W. (1983). Patterns of geographic variation in the egg morphology of the Fundulid fish, Fundulus heteroclitus. Copeia, 1983(3), 726-740.

Murakami, M., Ohi, M., Ishikawa, S., Shirai, M., Horiguchi, H., Nishino, Y., & Funaba, M. (2015). Adaptive expression of uncoupling protein 1 in the carp liver and kidney in response to changes in ambient temperature. Comparative Biochemistry and Physiology Part A: Molecular & Integrative Physiology, 185, 142–149.

Murphy, M. P. (1989). Slip and leak in mitochondrial oxidative phosphorylation. Biochimica et Biophysica Acta, 977(2), 123–141.

Nash, R. D., Valencia, A. H., & Geffen, A. J. (2006). The origin of Fulton's condition factor- setting the record straight. Fisheries, 31, 236-238.

Nicholls, D. G. (1977). The Effective Proton Conductance of the Inner Membrane of Mitochondria from Brown Adipose Tissue. Dependency on Proton Electrochemical Potential Gradient. European Journal of Biochemistry, 77, 349–356.

Nicholls, D. G. (2001). A history of UCP1. Biochemical Society Transactions, 29, 751–755.

Nicholls, D. G., Bernson, V. S. M., & Heaton, G. M. (1978). The identification of the component in the inner membrane of brown adipose tissue mitochondria responsible for regulating energy dissipation. In Effectors of Thermogenesis. Experientia Supplementum, pp. 89-93. Basel, Switzerland: Birkhäuser.

Nicholls, D. G., & Locke, R. M. (1984). Thermogenic mechanisms in brown fat. Physiological Reviews, 61(1), 1-64.

Nicholls, D. G., & Ferguson, S. J. (2013). Bioenergetics: 4th Edition. London, UK: Academic Press.

Powers, D. A., & Place, A. R. (1978). Biochemical genetics of Fundulus heterolitus (L.). I. Temporal and spatial variation in gene frequencies of Ldh-B, Mdh-A, Gpi-B, and Pgm- A. Biochemical Genetics, 16(5-6), 593–607.

Pörtner, H. O., Hardewig, I., & Peck, L. S. (1999). Mitochondrial function and critical temperature in the Antarctic bivalve, Laternula elliptica. Comparative Biochemistry and Physiology Part a: Molecular & Integrative Physiology, 124, 179–189.

Pörtner, H. O. (2001). Climate change and temperature-dependent biogeography: oxygen limitation of thermal tolerance in animals. Naturwissenschaften, 88, 137–146.

113 Pörtner, H. O. (2010). Oxygen- and capacity-limitation of thermal tolerance: a matrix for integrating climate-related stressor effects in marine ecosystems. Journal of Experimental Biology, 213, 881–893.

Pörtner, H. O., & Farrell, A. P. (2008). Physiology and Climate Change. Science, 322(5902), 690–692.

Pörtner, H. O., & Knust, R. (2007). Climate change affects marine fishes through the oxygen limitation of thermal tolerance. Science, 315(5808), 95–97.

Precht, H. (1958). Concepts of temperature adaptation on unchanging reaction systems of cold- blooded animals. In Physiological Adaptations, Symposium General Physiology, pp.50- 78. Washington, DC: American Physiology Society.

Qu, M., Wan, J., & Hao, X. (2014). Analysis of diurnal air temperature range change in the continental United States. Weather and Climate Extremes, 4, 86–95.

Ramsden, D. B., Ho, P. W. L., Ho, J. W. M., Liu, H. F., So, D. H. F., Tse, H. M., et al. (2012). Human neuronal uncoupling proteins 4 and 5 (UCP4 and UCP5): structural properties, regulation, and physiological role in protection against oxidative stress and mitochondrial dysfunction. Brain and Behavior, 2(4), 468–478.

Rial, E., Aquirregoitia, E., Jiménez-Jiménez, J., & Ledesma, A. (2004). Alkylsulfonates activate the uncoupling protein UCP1: implications for the transport mechanism. Biochemica et Biophysica Acta, 1608, 122-130.

Richards, J. G. (2010). Metabolic rate suppression as a mechanism for surviving environmental challenge in fish. In Aestivation, pp. 113-139. Springer Berlin Heidelberg.

Richards, J. G., Sardella, B. A., and Schulte, P. M. (2008). Regulation of pyruvate dehydrogenase in the common killifish, Fundulus heteroclitus during hypoxia exposure. American Journal of Physiology, 295, R979-R990.

Richieri, G. V., Anel, A., & Kleinfeld, A. M. (1993). Interactions of long chain fatty acids and albumin: determination of free fatty acid levels using the fluorescent probe ADIFAB. Biochemistry, 32, 7574-7580.

Riquier, D., & Bouillaud, F. (2000). The uncoupling protein homologues: UCP1, UCP2, UCP3, StUCP and AtUCP. Biochemical Journal, 345(2), 161–179.

Roy, R., Ghosh, D., & Das, A. B. (1992). Homeoviscous adaptation of different membranes in the brain of an air-breathing indian teleost, Channa punctatus, during seasonal variation of environmental temperature. Journal of Thermal Biology, 17(4-5), 209–215.

114 Rousset, S., Alves-Guerra, M.-C., Mozo, J., Miroux, B., Cassard-Doulcier, A.-M., Bouillaud, F., & Riquier, D. (2004). The biology of mitochondrial uncoupling proteins. Diabetes, 53, S130-S136.

Samartsev, V.N., Smirnov, A.V., Zeldi, I.P., Markova, O.V., Mokhova, E.N., Skulachev, V.P., (1997). Involvement of apartate/glutamate antiporter in fatty acid-induced uncoupling of liver mitochondria. Biochemica Biophysica Acta, 1319, 251–257

Sanchis, D., Fleury, C., Chomiki, N., Goubern, M., Huang, Q., Neverova, M., Gregoire, F., Easlick, J., Raimbault, S., Levi-Meyrueis, C., Miroux, B., Collins, S., Seldin, M., Richard, D., Warden, C., Bouillaud, F., & Riquier, D. (1998). BMCP1, a novel mitochondrial carrier with high expression in the central nervous system of humans and rodents, and respiration uncoupling activity in recombinant yeast. Journal of Biological Chemistry, 273(51), 34611–34615.

Schulte, P. M. (2015). The effects of temperature on aerobic metabolism: towards a mechanistic understanding of the responses of ectotherms to a changing environment. The journal of Experimental Biology, 218, 1856-1866.

Schwanhäusser, B., Busse, D., Li, N., Dittmar, G., Schuchhardt, J., Wolf, J., Chen, W., & Selbach, M. (2011). Global quantification of mammalian gene expression control. Nature, 473(7347), 337–342.

Senior, A. E., Nadanaciva, S., & Weber, J. (2002). The molecular mechanism of ATP synthesis by F1F0-ATP synthase. Biochimica et Biophysica Acta - Bioenergetics, 1553, 188–211.

Shabalina, I. G., Jacobsson, A., Cannon, B., & Nedergaard, J. (2004). Native UCP1 displays simple competitive kinetics between the regulators purine nucleotides and fatty acids. Journal of Biological Chemistry, 279(37), 38236–38248.

Sheng, Y., Abreu, I. A., Cabelli, D. E., Maroney, M. J., Miller, A.-F., Teixeira, M., & Selverstone Valentine, J. (2014). Superoxide dismutases and superoxide reductases. Chemical Reviews, 114, 3854-3918.

Sinensky, M. (1971). Temperature control of phospholipid biosynthesis in Escherichia coli. Journal of Bacteriology, 106, 449-455.

Sinensky, M. (1974). Homeoviscous adaptation--a homeostatic process that regulates the viscosity of membrane lipids in Escherichia coli. Proceedings of the National Academy of Sciences, 71(2), 522-525.

Singer, S. J. & Nicolson, G. L. (1972). The fluid mosaic model of the structure of cell membranes. Science, 175, 720-731.

Skulachev, V. P. (1998). Uncoupling: new approaches to an old problem of bioenergetics. Biochimica et Biophysica Acta - Bioenergetics, 1363, 100–124.

115

St-Pierre, J., Charest, P.-M., & Guderley, H. (1998). Relative contribution of quantitative and qualitative changes in mitochondria to metabolic compensation during seasonal acclimatisation of rainbow trout Oncorhynchus mykiss. The Journal of Experimental Biology, 201, 2961–2970.

Stenseng, E., Braby, C. E., & Somero, G. N. (2005). Evolutionary and acclimation-induced variation in the thermal limits of heart function in congeneric marine snails (genus Tegula): implications for vertical zonation. The Biological Bulletin, 208, 138–144.

Stier, A., Bize, P., Habold, C., Bouillaud, F., Massemin, S., & Criscuolo, F. (2014). Mitochondrial uncoupling prevents cold-induced oxidative stress: a case study using UCP1 knockout mice. Journal of Experimental Biology, 217, 624–630.

Stillman, J. H. (2003). Acclimation capacity underlies susceptibility to climate change. Science, 301(5629), 65–65.

Sunday, J. M., Bates, A. E., & Dulvy, N. K. (2011). Global analysis of thermal tolerance and latitude in ectotherms. Proceedings of the Royal Society B: Biological Sciences, 278(1713), 1823–1830.

Taylor, M. H. (1999). A Suite of Adaptations for Intertidal Spawning. American Zoology, 39, 313-320.

Tine, M., Kuhl, H., Jastroch, M., & Reinhardt, R. (2012). Genomic characterization of the European sea bass Dicentrarchus labrax reveals the presence of a novel uncoupling protein (UCP) gene family member in the teleost fish lineage. BMC Evolutionary Biology, 12, 62.

Trzcionka, M., Withers, K. W., Klingenspor, M., & Jastroch, M. (2008). The effects of fasting and cold exposure on metabolic rate and mitochondrial proton leak in liver and skeletal muscle of an amphibian, the cane toad Bufo marinus. Journal of Experimental Biology, 211, 1911–1918.

Tseng, Y. C., Chen, R. D., Lucassen, M., Schmidt, M. M., Dringen, R., Abele, D., & Hwang, P.- P. (2011). Exploring uncoupling proteins and antioxidant mechanisms under acute cold exposure in brains of fish. PLoS ONE, 6(3), e18180.

Vidal-Puig, A., Solanes, G., Grujic, D., Flier, J. S., & Lowell, B. B. (1997). UCP3: an uncoupling protein homologue expressed preferentially and abundantly in skeletal muscle and brown adipose tissue. Biochemical and Biophysical Research Communications, 235, 79–82.

116 Vogel, C., Abreu, R. de S., Ko, D., Le, S.-Y., Shapiro, B. A., Burns, S. C., Sandhu, D., Boutz, D. R., Marcotte, E. M., & Penalva, L. O. (2010). Sequence signatures and mRNA concentration can explain two-thirds of protein abundance variation in a human cell line. Molecular Systems Biology, 6(1), 400.

Wang, C., Sun, G., Chen, K., Lv, Z., Peng, S., Jiang, X., Xiang, Y., & Zhang, C. (2010). Molecular cloning of lamprey uncoupling protein and assessment of its uncoupling activity using a yeast heterologous expression system. Mitochondrion, 10, 54–61.

Wen, Z.-Y., Liang, X.-F., He, S., Li, L., Shen, D., & Tao, Y.-X. (2015). Molecular cloning and tissue expression of uncoupling protein 1, 2 and 3 genes in Chinese perch (Siniperca chuatsi). Comparative Biochemistry and Physiology Part B: Biochemistry and Molecular Biology, 185, 24–33.

Whitehead, A., Roach, J. L., Zhang, S., & Galvez, F. (2011). Genomic mechanisms of evolved physiological plasticity in killifish distributed along an environmental salinity gradient. Proceedings of the National Academy of Sciences, 108, 6193-6198.

Woyda-Ploszczyca, A. M., & Jarmuszkiewicz, W. (2014). Different effects of guanine nucleotides (GDP and GTP) on protein-mediated mitochondrial proton leak. PLoS ONE, 9(6), e98969.

Woyda-Ploszczyca, A. M., & Jarmuszkiewicz, W. (2017). The conserved regulation of mitochondrial uncoupling proteins: From unicellular eukaryotes to mammals. Biochimica et Biophysica Acta, 1858(1), 21–33.

Zavodnik, I. B., Lapshina, E. A., Palecz, D., & Bryszewska, M. (1996). The effects of palmitate on human erythrocyte membrane potential and osmotic stability. Scandinavian Journal of Clinical and Laboratory Investigation, 56(5), 401–407.

Zhang, C. Y., Parton, L. E., Ye, C. P., Krauss, S., Shen, R., Lin, C.-T., Porco Jr., J. A., and Lowell, B. B. (2006). Genipin inhibits UCP2-mediated proton leak and acutely reverses obesity- and high glucose-induced b cell dysfunction in isolated pancreatic islets. Cell Metabolism, 3, 417–427.

117 Appendices

Appendix A Supporting information

Table A.1 p-values from two-way ANOVA for Ucp mRNA expression in a northern and southern population of F. heteroclitus across acclimation temperature. Significant values are indicated by an asterisk (*).

Tissue Gene Temperature Population Temperature*Population Brain Ucp1 <0.001* <0.001* 0.071 Ucp2 <0.001* 0.046* 0.821 Ucp3 <0.001* <0.001* 0.640 Ucp3L 0.020* 0.746 0.779 Ucp5 <0.001* <0.001* 0.015* Gill Ucp1 0.014* <0.001* 0.704 Ucp2 0.003* 0.007* 0.015* Ucp3 0.001* 0.003* 0.003* Ucp3L 0.001* 0.494 0.039* Ucp5 <0.001* <0.001* 0.265 Liver Ucp1 <0.001* <0.001* <0.001* Ucp2 <0.001* 0.024* 0.008* Ucp3 0.666 0.630 0.533 Ucp3L <0.001* <0.001* 0.050* Ucp5 <0.001* <0.001* 0.499 Muscle Ucp1 <0.001* <0.001* 0.270 Ucp2 0.054 0.055 0.071 Ucp3 0.005* <0.001* 0.013* Ucp3L <0.001* 0.008* 0.001* Ucp5 <0.001* <0.001* 0.206

118

Table A.2 p-values from post-hoc comparisons of Ucp isoform mRNA expression across acclimation temperatures and between populations of F. heteroclitus. All values are FDR corrected and significant values are indicated by an asterisk (*).

119 Gene Comparison p-value in Brain p-value in Gill p-value in Liver p-value in Muscle Ucp1 N5 vs. S5 0.011* <0.001* <0.001* 0.095 N15 vs. S15 0.007* <0.001* 0.169 0.492 N25 vs. S25 0.001* 0.243 <0.001* 0.097 N33 vs. S33 0.001* 0.018* 0.024* 0.593 N5 vs. N15 0.681 0.071 0.105 0.728 N5 vs. N25 0.427 0.063 0.004* 0.107 N5 vs. N33 <0.001* 0.071 0.169 0.963 N15 vs. N25 0.636 0.711 <0.001* 0.227 N15 vs. N33 <0.001* <0.001* 0.620 0.914 N25 vs. N33 <0.001* 0.007* <0.001* 0.154 S5 vs. S15 0.861 0.254 <0.001* 0.931 S5 vs. S25 0.107 0.619 0.034* 0.095 S5 vs. S33 0.069 0.095 <0.001* 0.635 S15 vs. S25 0.104 0.963 <0.001* 0.095 S15 vs. S33 0.100 0.063 0.468 0.948 S25 vs. S33 0.007* 0.470 <0.001* 0.095 Ucp2 N5 vs. S5 0.954 0.534 0.618 0.024* N15 vs. S15 0.495 0.717 0.624 0.884 N25 vs. S25 0.246 0.561 0.036* 0.884 N33 vs. S33 0.411 0.028* 0.036* 0.940 N5 vs. N15 0.034* 0.909 0.304 0.021* N5 vs. N25 0.024* 0.213 0.167 0.105 N5 vs. N33 0.023* 0.005* 0.767 0.150 N15 vs. N25 0.954 0.140 0.616 0.643 N15 vs. N33 0.979 0.001* 0.484 0.643 N25 vs. N33 0.954 0.080 0.324 0.949 S5 vs. S15 0.010* 0.392 0.324 0.884 S5 vs. S25 0.007* 0.175 <0.001* 0.884 S5 vs. S33 0.010* 0.690 <0.001* 0.643 S15 vs. S25 0.495 0.690 0.023* 0.795 S15 vs. S33 0.954 0.717 0.047* 0.553 S25 vs. S33 0.731 0.561 0.356 0.879 Ucp3 N5 vs. S5 0.182 0.822 1.000 0.007* N15 vs. S15 0.095 0.030* 1.000 0.018* N25 vs. S25 0.477 0.011* 1.000 0.215 N33 vs. S33 0.072 0.822 1.000 0.948 N5 vs. N15 0.927 0.551 1.000 0.215 N5 vs. N25 0.690 0.370 1.000 0.948 N5 vs. N33 0.095 0.013* 1.000 0.008* N15 vs. N25 0.570 0.894 1.000 0.948 N15 vs. N33 0.072 0.011* 1.000 0.010* N25 vs. N33 0.158 0.003* 1.000 0.008* S5 vs. S15 0.767 0.030* 1.000 0.948 S5 vs. S25 0.857 0.026* 1.000 0.948 S5 vs. S33 0.054 0.062 1.000 1 S15 vs. S25 0.687 0.636 1.000 0.948 S15 vs. S33 0.095 0.822 1.000 0.948 S25 vs. S33 0.054 0.822 1.000 0.948 Ucp3L N5 vs. S5 0.870 0.242 0.966 <0.001* N15 vs. S15 0.318 0.849 0.234 0.395 N25 vs. S25 0.899 0.369 0.011* 0.888 N33 vs. S33 0.931 0.176 0.031* 0.484 N5 vs. N15 0.100 0.176 0.127 0.015* N5 vs. N25 0.899 0.085 0.982 0.010* N5 vs. N33 0.847 0.227 0.028* <0.001* N15 vs. N25 0.090 0.762 0.233 0.001* N15 vs. N33 0.100 0.476 0.003* <0.001* N25 vs. N33 0.910 0.160 0.063 0.030* S5 vs. S15 0.899 0.161 0.966 0.869 S5 vs. S25 0.931 0.110 0.019* <0.001* S5 vs. S33 0.847 0.085 0.007* <0.001* S15 vs. S25 0.847 0.243 0.011* 0.004* S15 vs. S33 0.363 0.243 0.003* <0.001* S25 vs. S33 0.899 0.879 0.465 0.037* Ucp5 N5 vs. S5 0.072 0.045* 0.025* 0.278 N15 vs. S15 <0.001* <0.001* <0.001* 0.003* N25 vs. S25 0.010* 0.064 0.073 0.602 N33 vs. S33 0.007* 0.015* <0.001* 0.027* N5 vs. N15 0.198 0.067 0.151 0.174 N5 vs. N25 0.101 0.067 0.050 0.073 N5 vs. N33 0.072 0.064 0.011* 0.002* N15 vs. N25 0.045* 0.010* 0.009* <0.001* N15 vs. N33 0.006* 0.004* <0.001* <0.001* N25 vs. N33 0.206 0.095 0.018* 0.023* S5 vs. S15 0.150 0.064 0.073 0.785 S5 vs. S25 0.646 0.481 0.151 0.569 S5 vs. S33 0.150 0.064 0.003* <0.001* S15 vs. S25 0.619 0.395 0.837 0.614 S15 vs. S33 0.863 0.367 0.025* <0.001* S25 vs. S33 0.206 0.064 0.089 0.001*

120 Appendix B Whole animal data

B.1 Chapter 2 whole animal parameters

Whole animal mass did not significantly differ with thermal acclimation or population in F. heteroclitus (Figure B.1A, ppopulation=0.156, ptemperature=0.051, ptemperature*population=0.621, see Table B.1 for post-hoc p-values). Fish length also did not differ with thermal acclimation or population (Figure B.1B, ppopulation=0.157, ptemperature=0.068, ptemperature*population=0.928, see Table B.1 for post-hoc p-values). Condition factor did differ between populations (ppopulation<0.001) as the southern population had a greater condition factors than the northern population of F. heteroclitus. There was no effect of population or interaction on condition factor (ptemperature=0.390, ptemperature*population=0.167, see Table B.1 for post-hoc p-values).

Table B.1 Post-hoc p-values for whole animal parameters in thermally acclimated northern and southern F. heteroclitus as determined by planned comparison t-tests followed by FDR correction for mass and conditions factor and Mann-Whitney U Rank Sum test followed by FDR correction for length. Significant values are indicated by an asterisk (*).

Comparison Mass Length Condition Factor N5vsS5 0.8473600 0.5528889 0.00388840 N15vsS15 0.5740000 0.7933333 0.00024192 N25vsS25 0.8482286 0.5528889 0.00388840 N33vsS33 0.8473600 0.5528889 0.00024192 N5vsN15 0.8482286 0.9650000 0.38368000 N5vsN25 0.1037333 0.2400000 0.60073846 N5vsN33 0.8929067 0.5984000 0.38368000 N15vsN25 0.1037333 0.2400000 0.12765333 N15vsN33 0.8473600 0.5528889 0.94560000 N25vsN33 0.1037333 0.5528889 0.04553600 S5vsS15 0.8482286 0.9650000 0.94869333 S5vsS25 0.8473600 0.5528889 0.50933333 S5vsS33 0.8482286 0.9650000 0.50933333 S15vsS25 0.8473600 0.5528889 0.38368000 S15vsS33 0.9975000 0.9650000 0.38368000 S25vsS33 0.8473600 0.5992727 0.99630000

121

A. 8 a h 6 h h

h 4 a a

Mass (g) a 2

0 5 15 25 33 o Acclimation Temperature ( C)

B. 100 a a a 80 a h 60 h h h

40 Length (mm) 20

0 5 15 25 33 Acclimation Temperature (oC)

C. 1.5 * * h h * * h h

1.0 a ab ab b

0.5

Condition Factor

0.0 5 15 25 33 Acclimation Temperature (oC)

Figure B.1 Whole animal measurements for northern (black circles, solid line) and southern (white squares, dashed line) Fundulus heteroclitus acclimated to 5, 15, 25, and 33 °C. Shared letters indicate no significant difference as determined by planned comparison t-tests followed by FDR correction for mass (A) and condition factor (C). Shared letters indicate no significant difference as determined by Mann-Whitney U rank sum test followed by FDR correction for length (B). An asterisk indicates a significant difference between populations at a given acclimation temperature as determined by the previously stated post-hoc analyses.

122 B.2 Chapter 3 whole animal parameters

Whole animal mass differed significantly between populations with the northern population having larger overall body mass than the southern population at 15 (Figure B.2A, see Tables B.2 and B.3 for p-values). Whole animal mass was not significantly different between populations at 5°C and 25°C (Figure B.2A, see Table B.3 for p-values).

Temperature affected whole-animal body mass with a trend of increasing mass with increasing acclimation temperatures (Figure B.2A, see Table B.2 for p-values). Length differed in a similar manner as mass between populations, with northern fish being longer than southern fish (Figure B.2B, see Tables B.2 and B.3 for p-values). There was also a significant effect of temperature on length, with length in both populations being slightly shorter at 5°C (Figure B.2B, see Tables B.2 and B.3 for p-values). While northern killifish both weigh more and are longer, their condition factor was lower than that of the southern population at all acclimation temperatures (Figure B.2C, see Tables B.2 and B.3 for p- values), suggesting a possible difference in body shape. There was a significant effect of temperature on condition factor, but only in the southern population where condition factor was lowest at 15°C (Figure B.2C, see Tables B.2 and B.3 for p-values). Northern killifish had greater liver masses than southern killifish at both 5°C and 15°C (Figure B.2C, see

Tables B.2 and B.3 for p-values). There was also an effect of temperature on liver mass in both populations, with livers being largest at 5°C in both populations (Figure B.2D, see

Tables B.2 and B.3 for p-values). Southern killifish had the smallest livers at 15°C and intermediate sized livers at 25°C, while northern killifish had livers that were similar sizes at both 15°C and 25°C (Figure B.2D, see Tables B.2 and B.3 for p-values). Brain mass significantly differed between the two populations at all acclimation temperatures, with the

123 northern population having slightly larger brains (Figure B.2E, see Tables B.2 and B.3 for p- values). Unlike liver mass, brain mass appears slightly larger at 15°C in both populations

(Figure B.2E, see Tables B.2 and B.3 for p-values). The hepatosomatic Index (HSI) differed between populations at all acclimation temperatures, with northern killifish having a larger

HSI at 5°C and 15°C and southern killifish having a larger HSI at 25°C (Figure B.2F, see

Tables B.2 and B.3 for p-values). Both populations have larger HSIs at 5°C compared to other acclimation temperatures, driven by larger liver masses and smaller whole body masses at 5°C (Figure B.2A, D, and F, see Tables B.2 and B.3 at p-values). The cranisomatic index did not differ between populations but was significantly affected by thermal acclimation

(Figure B.2G, see Tables B.2 and B.3 for p-values). The cranisomatic index was greatest at

5°C and lower at 15°C and 25°C in northern killifish and greatest at 5°C and 15°C and lower at 25°C in southern killifish (Figure B.2G, see Tables B.2 and B.3 for p-values).

124 A. B. 6 * 90 b * b * 5 ab 80 * b a a 4 h 70 h h i hi h

3 Length (mm) 60 1 10

Whole Animal Mass (g) 0 0 5 5 15 25 15 25 Acclimation Temperature (°C) Acclimation Temperature (°C) C. * D. 1.3 h * 0.20 h * 1.2 * a i 0.15 1.1 a a a * b b 1.0 0.10 h j 0.9 Liver Mass (g)

Condition Factor 0.05 0.1 i 0.0 0.00 5 5 15 25 15 25 Acclimation Temperature (°C) Acclimation Temperature (°C)

E. F. 0.08 6 * a * * b 0.06 a * a 4 * b 0.04 * b hi h 2 i h Brain Mass (g) 0.02 j i 0.00 Hepatosomatic Index (%) 0 5 5 15 25 15 25 Acclimation Temperature (°C) Acclimation Temperature (°C)

G. 2.0

b 1.5 a b 1.0 h h i 0.5

Cranisomatic Index (%) 0.0 5 15 25 Accliamtion Temperature (°C)

Figure B.2 Whole animal measurements for northern (black circles, solid line) and southern (white squares, dashed line) Fundulus heteroclitus acclimated to 5, 15 and 25°C. Shared letters indicate no significant difference as determined by planned comparison t-tests followed by FDR correction for condition factor (C), liver mass (D), brain mass (E), and hepatosomatic index (F). Shared letters indicate no significant difference as determined by pairwise comparisons using a Mann Whitney U rank sum test followed by FDR correction for whole animal mass (A), length (B), and cranisomatic index (G). An asterisk indicates a significant difference between populations at a given acclimation temperature as determined by the previously stated post-hoc analyses.

125 Table B 2 p-values for two-way ANOVAs for whole animal parameters from thermally acclimated northern and southern F. heteroclitus. An asterisk indicates p<0.05.

Parameter Population Temperature Population*Temperature Whole Body Mass (Fig. 1A) 0.003* 0.042* 0.171 Length (Fig. 1B) <0.001* <0.001* 0.725 Condition Factor (Fig. 1C) <0.001* <0.001* 0.002* Liver Mass (Fig. 1D) <0.001* <0.001* <0.001* Brain Mass (Fig. 1E) <0.001* <0.001* 0.818 Hepatosomatic Index (Fig. <0.001* <0.001* <0.001* 1F) Cranisomatic Index (Fig. 0.223 <0.001* 0.023* 1G)

Table B.3 P-values for post-hoc tests for whole animal measurements in thermally acclimated northern and southern F. heteroclitus. An asterisk indicates p<0.05 following FDR correction.

Comparison Whole Length Condition Liver Brain Mass HIS Brain Body Factor Mass Mass/Body Mass Mass N5 vs. S5 0.792 <0.001* <0.001* <0.001* 0.002* <0.001 0.097 N15 vs. S15 0.015* <0.001* <0.001* <0.001* 0.002* <0.001 0.176 N25 vs. S25 0.119 <0.001* <0.001* 0.334 0.001* 0.050* 0.097 N5 vs. N15 0.015* <0.001* 0.211 <0.001* 0.022* <0.001* 0.026* N5 vs. N25 0.119 0.008* 0.211 <0.001* 0.254 <0.001* 0.002* N15 vs. N25 0.357 0.191 0.853 0.943 0.001* 0.197 0.295 S5 vs. S15 0.984 0.013* <0.001* <0.001* 0.055 <0.001* 0.421 S5 vs. S25 0.914 0.191 0.211 0.013* 0.062 <0.001* 0.003* S15 vs. S25 0.858 0.191 <0.001* <0.001* 0.001 <0.001* <0.001*

126

126v