THE

EVOLUTION

HAEMOGLOBINS

A

B.Sc.

THESIS

THE

THE

OF

Hon., THE

PROTON-TRIGGERED

REQUIREMENTS

SUBMITTED

UNIVERSITY

FACULTY

Queen’s

©

MATTHEW

MASTER IN

Matthew

BASAL

University,

October,

(Vancouver)

OF

(Zoology)

IN

OF

by OF

GRADUATE PARTIAL

P.

in

ACTINOPTERYGIAN

FOR

D.

BRITISH

Regan,

SCIENCE

2008

REGAN

Kingston,

OXYGEN

THE

FULFILLMENT

2008

DEGREE

COLUMBIA

STUDIES

Ontario,

PUMPS:

OF

2005

FISHES

ROOT

OF

EFFECT carrying

absence the

and basal hypoxia,

spat

alligator and

result

low

via lower

oxylabile

that analyses the proton-binding

values those There

potential

thus Hb

hula), buffering pirarucu

enough

the

actinopterygian

from

than

expected

It

was

titrations, of

of

capacity

Rb

02

gar

has and

of

Haldane

these

white

red

those

also

either

uptake

proton-binding ITh

levels

(Atractosteus for

recently

(Arapaima hypercarbia.

of

blood

02

to properties seven

these

a

the sturgeon

produced

at

which

RBC

correlation

be

in

saturation

effects,

at

low

intracellular

cell

the

produced

the been

lineage

species

species’

buffering

pH)

gigas)) simultaneously

general

(RBC)

gills,

(Acipenser

spatula),

of

I proposed

while

by

properties

analyzed

evolved

Root between

(from over

to

these

during

variably-sized

by

based

experience pH1-protecting

circulation

compartment

the

by

a a

effect ABSTRACT

basal

bowfin

generalized

pH

hypoxia-

Rb,

that

latter

in

transmontanus),

the

a

of the

on

the

generalized

spectrum

assessed

these or

the

species

haemoglobins

to

the

presence

was

basal

onset (Amia a

derived:

to

Root

11

hypothesis

reduction

Root

or

species’

activate of

investigated

I3NHE

acidoses.

exercise-induced

with

intrinsic actinopterygian

pH the

of

effect

calva),

effects

of

acidosis

pH

values RBC

American

countercurrent

3NHE.

spotted

activity.

their

Hbs

in

(Hbs)

5.5

(reduction

that

mooneye

Hb

blood

were

The

when

by

to

unlikely

of

Root

associated

RBC

buffer

gar

of

However, 8.5.

the spectrophotemetric

former

paddlefish Consequently,

considerably

02

compared

seven

lineage

(Lepisosteus

effects. generalized

Root pH1

The of

carrying

(Hiodon

capacities

play

choroid

blood

was

is

with

species

results

effect

the

not

of

a

This

investigated

(Polyodon

with

major

fishes

oxygen exercise,

capacity, onset

tergisus),

reduced

lower

retia

acidosis.

that

there

and

oculatus),

suggest

within

may

those

role

in

pH

and

are

than

is

the

to

the

Hb

in low Hb buffer values, large Bohr/Haldane effects, large Root effects, and high Root effect onset pH values, which collectively suggest a different process of Root effect evolution than is currently assumed. Taken together, it appears as though 02 uptake is not jeopardized in these early Root effect species, despite their relatively low Hb proton binding properties and lack of NHE.

111 Abstract Table List List Acknowledgements Statement Chapter Chapter Fishes Haemoglobin’s The The The Hypothesis Thesis References Chapter Introduction Material Results of of Variation The Animal Haemoglobin Haemolysate Data Haemoglobin Haemoglobin of basal basal Figures Abbreviations Root . Contents One: Two: objectives potential of analysis summary and effect actinopterygians: actinopterygians: Co-Authorship acquisition in General methods Haemoglobin-proton Hb-H function preparation cost titrations buffer titrations of Introduction binding

TABLE values the in blood Root A A transitional transitional among effect gas

OF Binding transport fishes

CONTENTS iv group group Properties in in haemoglobin haemoglobin of Basal evolution evolution Actinopterygian I II viii vii 16 14 iv ix xi 22 20 21 16 23 23 23 17 13 19 19 16 ii 4 3 6 1 8 1 7 Fishes Chapter

Discussion

References Chapter

Introduction

Materials

Results

Discussion

Chapter

Haldane Titratable

Influence Haemoglobin

Haemoglobin The Transitions Postulating

Conclusions

The

The

Animal Haemolysate Root Data

Statistical

Root Root Influence

Conclusions

3:

Haldane

Root basal

summary

analysis

effect

effect

effect

summary

Root

and

acquisition effects

of

groups

effect

actinopterygian

ofGTP

analysis

in methods pirarucu

quantification

without onset GTP

Effect

effect

buffer P50 preparation

Hb-H

buffer

pH

Haemoglobm

retia curves

binding

values

fishes:

in

light

A

Characteristics

transitional

of

V

the

Root

group

effect

in

Basal in

Root

Actinopterygian

effect

evolution . .

. .24

24

27 25 25 28 .46 29

30

44

34

41 40

32 44 44

47

47

49 48

49

50 50

52 53

54 57 60

65 Chapter

References

Future

Synthesis References Chapter

Four:

directions

summary

General

Discussion

vi

.66

70 71

76

83 82 LIST OF FIGURES

Figure 1.1. Movements of respiratory gases into and out of the red blood cell at the metabolically active tissue 10

Figure 1.2. Changes in Hb buffer value, Root effect magnitude, and f3NHEactivity in fishes situated on the Percomorpha evolutionary trajectory (modified from Berenbrink et al., 2005) 11

Figure 1.3. Phylogeny depicting the interrelationships of the living vertebrates, with particular reference to the “primitive” fishes (from Janvier, 2007) 12

Figure 2.1. Representative titration curves of oxygenated and deoxygenated haemoglobins of seven basal actinopterygian species in the presence and absence ofGTP 35

Figure 2.2. Representative buffer curves of oxygenated and deoxygenated haemoglobins of seven basal actinopterygian species in the presence and absence ofGTP 36

Figure 2.3. Haemoglobin P50 buffer values of seven basal actinopterygian species at three physiologically-relevant pH ranges (pH 6.8-7.0; 7.0-7.2; 7.2-7.4) in the presence of GTP 37

Figure 2.4. The Haldane effects as a function of pH of seven basal actinopterygian species in the presence and absence of GTP 38

Figure 2.5. The number of titratable groups in the haemoglobins of bowfin, mooneye, and pirarucu as determined from differentially-plotted haemoglobin titration curves 39

Figure 3.1. The Root effect as a function of pH in seven basal actinopterygian species in the presence and absence of GTP 62

Figure 3.2. The pH values at 5, 10, 15, and 20% Root effect onset in seven basal actinopterygian species in the presence and absence of GTP 63

Figure 3.3. The relationship between maximal Root effect and Root effect onset pH in seven basal actinopterygian species 64

Figure 4.1. Changes in Hb P50 buffer value, Haldane effect, maximal Root effect, and Root effect onset pH among basal actinopterygian fishes in relation to the appearance of the choroid rete 81

vii LIST OF ABBREVIATIONS

ATP Adenosine triphosphate AE Anion exchanger NHE Adrenergically-activated sodium/proton exchanger CA Carbonic anhydrase Cr Chloride 2CO Carbon dioxide GTP Guanosine triphosphate H Proton Hb Haemoglobin Hb P50 buffer value Mean of oxygenated and deoxygenated Hb buffer values 3HC0 Bicarbonate His Histidine residue NTP Nucleoside triphosphate 02 Oxygen 2Pco Partial pressure of carbon dioxide pHe Extracellular pH pH Intracellular pH pK Negative logarithm of dissociation constant 2Po Partial pressure of oxygen PWco2 Partial pressure of carbon dioxide in water 2Po Partial pressure of oxygen in water RBC Red blood cell R-state Relaxed (oxygenated) state of haemoglobin SEM Standard error of mean value T-state Tense (deoxygenated) state of haemoglobin

viii me However, likely Brauner. shall possible. me Gracias, particularly and research, enthusiasm. thankful on my have shout-out other in this a own Trish thank deeper played some never close It There In page. I CJ! have is experiences for Not but A if a Schulte. goes all probable capacity similar friends read we during more understanding in shepherd Of many are, only Having of made some to are particular my you however, important The a my speaking vein, I’ve Likely other many over that thesis deep way, here forgotten such to Dave committee made,

ACKNOWLEDGEMENTS I I my things the well would at will note shape, unbeknownst close anyway. of, a some lesson Toews, LTBC probabilistically, sheep friend course particularly and not of ones is as friends like or not-so-forgettable physiological Jeff, meeting, all well. love be — a experiencing form. in of for So, supervisor thank to the able who person! my for, over very thank to no more to those Milica MSc them, you! did proved sleep science ix the similar adequately most my enjoyable much knowledge, can in past older, similar their lost! Mandic thesis this insightful of and offer ones, reasons. three to those enthusiasm brief wiser, And its encourage committee thank things and their and has years, pursuit he if whom acknowledgement not rewarding. more And first certainly you student, all has at only all those similar in than do to and me helped I experienced of of may my the in with Drs. and foremost whom I ever earned who fact cannot project, dozen times Another forget instil foster Jeff regards I read have thought I a section. has Richards am or within imagine. will ones is few my this, helped big so Cohn to made own lines who my I Finally, I’d like to thank my family. My parents’ interest in what I do, a testament to their unconditional support more than it is to their love of science, has provided me with much encouragement and confidence. I feel both fortunate and grateful to have been the serendipitous product of their fused gametes! And to my sister, who, despite her two fewer years on earth, has proved quite the source of inspiration for following one’s chosen path.

x experimental Regan Dr. experiments, Cohn under Chapters J. Brauner. the the designs,

STATEMENT 2 supervision manuscripts and laboratory 3 of this of were thesis Dr. research, written Cohn are

OF co-authored. J. and solely Brauner.

CO-AUTHORSHIP xi data by analyses Matthew Following The research were Regan the conducted conclusion questions, in consultation by of Matthew the with

consumption,

blood

bloodstream

conformational

al., 02 transport

structure/function yielded

organism, cell the this,

information Furthermore,

reversibly meet

soluble

molecule

1998).

efficient

to

vertebrate

metabolic

cell

the

In

Oxygen

in

from

(Brauner

the

Hb

environment, water,

(RBC).

binds

Oxygen

not

transport

where

to

CHAPTER

the

blood

Hb

blood

has

Hb

species

changes demand

(02)

only

02

hydration

the

is

been

relationships

This and

increases

it

responsible

is

molecules

of

Po 2

oxygen

is

Haemoglobin

on

binds

then

of

fishes,

Randall,

have

a

thoroughly

occurs

through

decreases in the

at

necessary

prerequisite

transported

the

the

of

to

respiratory

evolved

the

carrying

Hb

CO 2

one

tetrameric

respiring

to

in

of

for

1996; 1:

02 both

each

plays

a

proteins

of

and

to

studied

quantities

the

function

cooperative

affinity

GENERAL

to

four

of

bicarbonate.

the

capacity

of

1998).

to

02

transport

physiology

a

use

tissue vertebrate

central

four

protein

the

direct

haem

is

for

in

haemoglobin

of

1

subsequently

tissues,

of

general.

iron-containing

At

in

well

in

the

of

binding

fashion,

groups

02

of

role

blood

multicellular

resulting

the

the

subsequent

life.

Due

waste

of

over

from

INTRODUCTION

where,

gill,

in plasma

vertebrates,

gas

on

the

to

However,

of

whereby

a

the

(Hb),

carbon

02

released

century

these

from

the

C0 2 ,

transport

tight

as

environment

diffuses

does

haem

three

Hb

organisms.

a

a

Hb

vital

as

coupling

result

dioxide

tetrameric

the

molecule

not

but

and

as

from

well haem

02

groups,

it

roles

binding

into

the allow

has

binding

of

is

as

Hb

(C0 2 )

not

groups

02

to

the

of

yielded

To

the

in

within

protein

for

allowing

supply

the

02

the

sufficiently

of

account

protons

(Jensen

passive

from

and

cells.

the

due

critical

the

to

that

first

CO 2

the

to

for

red for et

Fig.

transport).

it

methods

effect

such,

molecular

its

which

additional effect,

RBC

exposed

hydrated

fashions. in

presence

Approximately

capillaries

(bicarbonate;

CO 2 .

diffusion

with

exchange

oxygenated

1.1).

both

to

allows

A

is

the

an

be

on

metabolic

effectively

a

form

of

oxylabile

into

The

One

His

microenvironments

efficient

force function

where

taken

However,

the

the

for

for

HC0 3 )

the

binding

residues

to

method

(Tufts

surface

95%

catalyst

chloride

the

the

up

its

it

respiring

waste,

removal

of

method

is

reaction

account by

proficient

deoxygenated

of

and

due

through

quickly

the

of

is

of

and

fib

the

carbonic

via

through

protons

Perry,

to

number

the

CO 2

(Tufts

CO 2

cell.

the

of

of

its

band

for

in

molecule,

transferred

the

Hb

in

delivery

CO 2

diffuses

elevation

favour C0 2 +H 2 0

mechanistic

the

produced

1998;

This

which

anhydrase

to

the

and

proton-binding following

3

of

conformation

protons

from

anion

the

proton-binding

intrinsic

Perry,

process

of

Fig.

of

out

newly-exposed they

the

of

more

the

to

by

exchangers,

02

of

HCO 3 +H

1.1).

their

pK

produced

2

linkage

the

(CA),

the

reside

1998).

tissue

the

to

is

buffering the

HC0 3

values

RBC

reaction:

the

made

pK

body

(Brauner

Bicarbonate

that

respiring

it

(Brauner

(Jensen

tissue

with

This

is

values

sites

cytosol

in

results

production

more

of

is

while

converted

His

capacity

the

transported

the

these

can

(histidine

through

and

cell

when et

residues

hydration

efficient

Bohr

and the

(Fig.

from

occur

is

al.,

Randall,

residues,

and

excreted

protons

Randall,

of

to

Hb

(i.e.,

1989);

effect,

a

the

1.1).

in

(His)

the into

its

method

on

in

in

is

reaction,

either

exposure

hydrated

increased

the

Hb

fib

converted

the

1998).

the

and

Here,

remain

or,

the

from

residues)

1996;

blood

molecule,

(called

presence

tissue

the

that

of

the

Haldane

in

the

two

Both

and

Haldane

of

form

in

1998;

couples

CO 2

the

in

from

RBC

Bohr

the

as

this of

remnants

fishes.

spectrum As there

RBC,

are capacities possession

small

et

capacity

way,

02 RBC,

The

defined releasing

deoxygenated

groups)

a!.,

the

supply

two

more

exists

through

Haldane

2007),

each

and

elasmobranchs

Living

There

different

as

leads

and

of

(Janvier,

it

and

CO 2

thus,

can

a

resulting

of

The

a

for

taxa

reduction

species

the

paucity

primitive has the

to

large

“primitive”

elegantly

T(ense)-state

effects,

that

diffusion

basal

the

Haldane

that

the

but

oxylabile

been

2007),

more

is

Haldane

in

formation

equally such

dominated

of

actinopterygians:

released

and

while

in

the

shown

morphological

information

meet

into

Hb

02

effect

it

as

teleosts

species

efficient

binding

is

of

that

elasmobranchs

effective the

effects 02

the

02

possible

to

into

of

the

among

periods

affinity

more

be

demand

is

tissue

intramolecular

such

straddle

protein,

the

released

transport

of

an

(Jensen,

on

strategies

derived protons

that or inverse

blood,

the

the

(Jensen,

as

with

of

A

molecular at

the

transitional

the

the

decreasing

fishes.

nature

this

3

the

possess from

and

1989;

a

the

basal

teleost

to

basal

relationship

fossil

reduction

transition

metabolically

by

2004).

salt

particular

subsequent

more

Hb

of

Defined

which

Berenbrink

actinopterygians

characters

Hbs

actinopterygians

record,

this

bridges

for

species

its

group

protons

This

diffusion

of

transition

in

affinity

occurred

fishes

between

groups

as

blood

high

and

which

is

removal

possess

in

active

“plesiomorphic”

(Janvier,

et

the

that

Hb

bind

as

buffer

for

al.,

pH

into

on

within

Bohr

in

such,

stabilize

are

evolution

intrinsic

tissue.

are

02

2005).

Hbs

protons

on

Hb

(Bohr

the

of

the

produced

capacities

and

2007;

the

effect,

the CO 2 .

their

proton-binding.

Hb

this

of

tissue.

et

few

phylogenetic

thus

the

Hb

low

Thus,

molecule,

within

I

group

McKenzie

al.,

However,

due

and

buffer

extant

buffer

in

and

In

1904).

there

to

is

the

of

the

this the

of

of

2005).

their

facilitating

earth’s

species

derived

properties

actinopterygian

believe

analyzing

same physiological

question

depending

number

to

contemporaries

physiological

the

protons

this

Hbs,

protein.

way

known

At

idea.

Haemoglobin’s

of

class

that

of

may

a

extant

to

the

the

changed that

this

on

selective

phenomenon

they

particular

Extant

of

Hb

level

well

characters

the

living

functions

However,

morphological

explosive

fishes,

lineage,

(Janvier,

fishes

may

proton-binding

nature

have

of over

members

pressures

vertebrates

Hb,

contribute

groups

the

proton

(Nelson,

been

are

we

evolutionary

of

remain

the

called

species

2007;

the

teleosts.

these

believed

may

bonds

altered

Root

of

and

on

sensitivity

since

the

McKenzie

(Baillie

in

plesiomorphic

1994)

to

gain

radiation

characteristics

selective

the

molecular

a

formed

the

Root

effect

This

diverging

relatively

to

Hb

in

The

time.

an

patterning

and

mirror

the

et

understanding

(Root

effect,

group

has

Root

is

pressures,

(Moyle

et

al.,

upon

accounts

process.

4

much

characters

been

a!.,

from

2004). those

similar

effect

groups,

constitutes

lineages

is

of

proton-binding

2007).

of

believed

like

very

and

extant

their

phylogenetic

for

of

By

the

The

state,

the

Cech,

in

their

much

of

can

approximately

and

have

There

physiology

last

members

this

how

Bohr

increased

roughly

to

(Janvier,

to

fossil

large,

common

1996;

likely

have

amplified

case)

the

the

are,

effect

are

relationships

point

anatomical

however,

Hb

played

27

of

of

Berenbrink

stabilizes

been

so

of

proton

2007).

the

course,

in

000

proton-binding

ancestors,

strong

half

the

in

where

that

subjected

basal

a

the

of

species

of

sensitivity

major

these

So,

the

28

the

that

limitations

most

all

some

in

et

by

000

binding

and

T-state

the

al.,

the

role

in

to

any

of in

blood

a

exchangers the

the

with

gill

hypercarbia,

come pressures.

molecular

despite

particular:

that

arterial

1998; Together much

physiological

Root

from

atmospheres

complete

generalized

RBC

ability

during

allows

pH

under

the

a

effect

Peister

different

potential

their

supply,

sensitive

membrane

Bohr

with

to

saturation

02

a

But

are

the

for

stressful

is

time

regulate

are

avascularization,

blood

(Root,

and

directly

implications

therefore

an

activated

effect),

the

eye

as

than

teleosts

capable

trade-off.

when

Root

with

Decker, acid-producing

diffusion

and

acidosis.

(NHE;

those

with

1931;

conditions,

RBC

from

but

most

Hbs,

the

the

use

capable

through

of

oxygen

2004).

that

also

Root

pH

fish swimbladder

of

eliciting

the

Certain

the

this of

things

Nikinmaa,

As

this

and

result

bloodstream

02

through

its

Root

is

of

and

and

generalized

a

the

gas

most cannot

This

large

total

and

result,

allows

decreasing

physiological,

conditions,

a

Irving,

binding

are

effect

from

gland

generalized

protons

ensures

in

02

the

reduction

1992;

capable

(Peister

be

02

for

need

the

carrying

to

use

5

and

1943;

despite achieved,

loading

of

acidosis

the

deliver

Pelster

Bohr

not

from

including

the

of of

catecholamines

a

of

swimbladder

and

countercurrent

just

Na/H with

in

02.

acidosis

Scholander

retinal

keeping

capacity.

high

effect’s

the

Hb

at

Weber,

vast

and

could

even

the

To

the

the

venous

02

Po 2 s

hypoxia,

Randall,

compensate,

exchangers cells

quantities

blood’s

benefits

of

gill

carrying

jeopardize at

decreased

RBC

1991;

the

Furthermore,

Po 2 s

and

and

to

is

remain

that

supply

capillary

be

blood.

not

pH 1

large

02 Van

exercise,

1998).

of

of

Pelster

filled

are

of capacity

jeopardized.

affinity

high up

on

the

Hb

saturated

02

back

teleost

Dam,

02

released

hydrostatic

In

to

the

Root

with

02

uptake

network

to

These

in

and

conjunction

the 140

and

into

surface

two

(as

affinity.

are

the

1954).

fishes

pure

Randall,

effect

with

into

results

the

very

face

sites

at

(rete)

the

of

have

The

the

02

of in The basal actinopterygians: A transitional group in Hb evolution II

The evolutionary relationships of the Root effect and its related processes have recently been shown by Berenbrink et aL (2005) in a comparative study across a broad range of fish species. Their findings suggest that the Root effect first appeared in the last common ancestor of the polypteriformes and the acipenseriformes (Fig. 1.2). This was eventually followed by the appearance of the choroid rete in the last common ancestor of the amiiformes and the osteoglossiformes, which maximized and localized the acidosis necessary to drive 02 off the Root Hbs and into the retinal cells. And eventually, when the Root effect had already reached roughly 50% in the elopomorphs, !3NHEevolved to protect RBC pH, during generalized blood acidoses (Berenbrink et al., 2005; Fig. 1.2).

Although sensible on a number of levels, this evolutionary picture of Berenbrink et al.’s (2005) begs a question as to how species with a Root effect are capable of maintaining 02 uptake during a generalized acidosis in the absence of 3NHE. These species include the acipenseriformes (sturgeons and paddlefish), the ginglymods (gars), the amiiformes (bowfln), and the osteoglossoformes (the bony-tongues; the basal teleosts), all of whom, it seems, are at risk of losing up to 50% of their blood 02 carrying capacity during times of hypoxia, exercise, or hypercarbia (Fig. 1.2). That they are some of the very few “primitive” fishes to survive to the present day is likely a testament to their ability to compensate in such situations. How they manage this, however, is unresolved.

6

values

of vary

binding

in protection, both

actinopterygian

low measured

non-bicarbonate

than

species

means pH

of blood

evolution

activity,

rainbow

these

two

values

buffer

interspecifically,

those

intrinsic

acidosis.

lower

other

scenarios:

Regarding

Due

Despite

during

Root among

the

simultaneously

must

required

produced

value/large

trout

these

to

basal

than

than

effect

Hb

its

a

fishes

surely

the

species.

I

(Peister

generalized

buffer,

experiments

buffer

high hypothesize

3NHE;

those

the

actinopterygian

either

presence

to

species

by

from

second

from

activate

remain

Haldane

concentration

produced

exercise,

capacity Hb

these

and

or,

through

its

pH

would

lacking

acidosis.

the

Weber,

of

scenario,

could

capable

this

species

ancestral

these

6.15

effect,

a

Root

and

hypoxia,

significant

by

Hb-H

fishes

would

be

I3NHE

also

in

species’

exercise,

1990). in

the

the

effect

of

are

particularly

This

elephantnose

Root

Hypothesis

the

state

reveal

securing

occupying

oxylabile

likely

titrations. occur

capable

or

vary

R]3C

proton-binding

7

onset

Root

It

effect

Root

of

hypercarbia.

hypoxia,

is

the

only

candidate

high

in

therefore

and

02

effect

pH

of

such effects.

if

Haldane

onset

nature

the

As

fish if

done

buffer regulating

status at

values

RBC

transitional

or well

the

a

and

pH

(Berenbrink,

way

of for

on

hypercarbia.

possible

This

gills

ability

as

value/small

effect,

pH 1

values

the

a

as

of

protecting

the

the

that

lack

Hb’s

these

their

could

during

transition

remained

Hbs

blood’s

would

they

both

of

that phase

have

role

RBC

species

protective

of

result

2007)

a

all

the Haldane

of

This

RBC

basal

been

generalized

be

in

in

predominant

pH 1

in

higher

which

occur

onset

pH 1

a

Root

from

to

Hb

would

are

pH

function

shown

by

pH

I3NFIE

proton-

effect

at

lower

can

values

than

some

effect

either

in

7.35

pH

these

be

to

the

of to

properties phosphates),

over methodologies

excepted, two

(2005) mooneye paddlefish

basal

(Lepisosteus

species

role

generalized

a

that

saturation maximum

prevent

02

species’

uptake

representative

in at

detailed

to

pH

to

the

The

within

most

the

be

(Hiodon

Amia

of

Root

at

5.5

preservation

would

(Polyodon

value,

objective

Root

acidosis

particularly

I

species

oculatus),

many

pH

derived,

hope

(where

the

(Hb

calva

effect

spectra;

not effect

comparing

basal

tergisus),

species

points

titrations

to

within

will

be

being

of

spathula),

there

gain

is

the

alligator

of

from

actinopterygian

jeopardized.

this

activated.

demonstrative

all

make

along

species

02

from

a

is

this

the

experiments

and

thesis

fib

comprehensive

of

ever

uptake

maximal

clear

only

oxygenated

interesting

each

a

gar

white

02

pirarucu

pH

analyzed

being

is

Comparing

saturation

(Atractosteus

Thesis

extant

whether

at

to

of

spectrum

The

sturgeon

desaturation).

the

of

determine

the

lineage

activated

done

(Arapaima

transitions

Root

evolutionary. gill.

member).

in

objecti

four

and

8

understanding

this

this

at

in

this will

(Aczenser effect

deoxygenated

of

orders

pH

the

Hb

spatula),

under

thesis

the

fishes.

onset

s

elucidate

8.5

presence

gigas).

in

characteristic

fib However,

Combined is

outlined

Hb-H

(where

these

include commonly

pH

If

proton-binding

Expressed

transmontanus),

of

bowfin

I

value

This

can

the

the

conditions,

and

fibs;

sensitivity

there

by

measuring

(Fig.

exact

with

assume

Hb

selection

absence

to

Berenbrink

(Amia

is

just

Rb

is

proton-binding

in

those

indeed

1.3):

thorough

pH

no

02

measured

order

properties

the

and

(Amiiformes

calva),

value

Root

saturation

of

Hb

of

constitutes American

spotted

playing

Hb

organic

of

therefore,

a

et

02

effect)

most

at

al.

as

which

of

gar

its

a to

today

have

ways

(Janvier,

characteristics

allowed in

—

the

which

2007;

teleosts.

for

these

of

McKenzie

the

these

Hb

adaptive

extant

properties

et

a!.,

radiation

species

2007),

changed

to

of

these

be

the

9

representative

over

analyses

largest

evolutionary

group

may

of

of

allow

the

time,

vertebrates

ancestral

me

and

to

understand

how

on

states

the

they

of

planet

may

Hb the

diffusion haemoglobin

chloride which

(HC0 3 )

(C0 2 ) Figure

binds diffuses

1.1.

(Cl)

and

into

A

directly

protons

(Hb),

via

the

02

from

diagrammatic

the

tissue.

stabilizing

the

to

band

(Hj

Hb

tissue

The

in

3

is

anion

the

representation

not

reverse

its

into

Co 2

C0 2 +H 2 0(

presence

I

represented

deoxygenated

exchanger

the

+HbH(

scenario

red

of

blood

RBC

of

carbonic

10

in

(AE),

occurs

CA)HCOH+

gas

this

T-state

Ci

cell

exchange

while

diagram).

Hb0 2 ÷H

(RBC)

at

anhydrase

and

I

the

the

thereby

gills.

where

at

protons

Bicarbonate

the

(CA;

tissue.

it

releasing

HC0 3

is

are

the

hydrated

bound

small

Carbon

is

its Tissue

Plasma

exchanged

amount

by

bound

to

dioxide

bicarbonate

02

of

for

for

CO 2 I

0.

m

percentage physiological

uptake.

physiological the

deoxygenated progressive

refer

Figure

5.5

(squares),

presence

buffer

to

1.2.

deoxygenated,

Modified

relative and

decrease

order,

Changes

of

temperature

temperature

red

I3NHE

choroid

blood

from

to

from in

>

activity

in

Phyogenetlc

organic

pH

02

Berenbrink

and

basal

specific

cells

saturation

according and

8.0

swimbladder

in

to

(triangles)of

buffer.

RBC

phosphate-free

-‘

physiological

derived.

Hb Progression

choroid

/

et

pHi.

buffer

of

to

11

NHE

al.,

the Rb

R

retia,

Light

2005.

Root

extant

value

isoproterenol-activated in

activities

air-equilibrated

salines

haemolysates

_

effect

respectively.

grey

(circles), fishes

SwmdderRete

and

values

at

were

shown

arterial

dark

Root

measured

represent

in

Hb

haemolysates

grey

in

0.1

pH

effect

buffer

phylogenetically

rate

M

fields

and

I

KC1

on

the

magnitude

of

values

indicate

22 Na

maximal

at in

a pH

illustrations

to Figure

elasmobranchs; accepted

15,

10,

living

neopterygians;

lungfishes;

1.3.

taxa:

by

for

morphologists

Phylogeny

1,

16,

terminal

6,

craniates;

osteichthyans;

lepidosirenidae;

11,

halecostomes;

of

taxa

the

and

2,

from

vertebrates;

living

paleontologists.

7,

Janvier,

17,

actinopterygians;

vertebrates,

12,

Cladistians

rhipidistians.

teleosts;

12

Chiniäenforms

3,

1996.

gnathostomes;

(Polypteriformes)

Main

depicting

rphs

13,

Figure

elopocephalans;

8,

clades

actinopterans;

topology

4,

taken

and

chondrichthyans;

selected

from

that

14,

9,

Janvier,

is

acipenseriformes;

characters

sarcopterygians;

most

5,

widely

2007; applying

4.

2.

3.

1.

The

conditions blood

Two

Oxygen

fishes The

•

•

•

•

objectives

Root

hypotheses

brought

in

values.

comprehensive

H

To

of

To

Root

hypercarbia.

generalized

oxylabile

Significant

uptake

the

species

effect

titrations

characterize

simultaneously

are:

absence

effect

on

of

may

is

as

Haldane

within

this

by

believed

Hb

onset

acidosis

to

of

therefore

exercise,

of

thesis

how

proton-binding

their

fashion

the

RBC

-CHAPTER

the

pH

effect)

measure

Root

these

to

oxygenated

of

were:

basal

values

pH 1 -protecting

have

be

hypoxia,

the

so

effect

fishes

threatened

that

as

actinopterygian

blood

the

evolved

that

13

to

SUMMARY-

protect

properties

properties

Hb

determine

maintain

or

are

and

brought

hypercarbia.

buffer

lower

in

deoxygenated

during

I3NHE.

RBC

the

02

(intrinsic

on

of values

their

than

basal

lineage

pH 1 ,

generalized

uptake

these

by

Root

those

exercise,

and

actinopterygian

and

species’

of

Hb

Hbs.

during

thus,

effect

the

produced

fishes

buffering

acidoses

Haldane

hypoxia,

02

these

Hbs

onset

through

uptake.

by

in

and

lineage

pH

of

effects

a

a

or

the

the of References

Baillie, J. E. N., Bennun, L. A. and Brooks, T. M. (2004). A Global Species Assessment: 2004 IUCNRed List of Threatened Species. Gland: World Conservation Union.

Berenbrink, M. (2007). Historical reconstructions of evolving physiological complexity: 0-2 secretion in the eye and swimbladder of fishes. I Exp. Biol. 210, 1641-1652.

Berenbrink, M., Koldkjaer, P., Kepp, 0. and Cossins, A. R. (2005). Evolution of oxygen secretion in fishes and the emergence of a complex physiological system. Science 307, 1752-1757.

Bohr, C., Hasselbaich, K. A. and Krogh, A. (1904). Uber einen in biologischer beziehung wichtigen einfluss, den die kohiensaurespannung des blutes auf dessen sauerstoffbindung ubt. Skand. Arch. Physiol. 16, 402-412.

Brauner, C. J. and Randall, D. J. (1996). The interaction between oxygen and carbon dioxide movements in fishes. Comp. Biochem. Physiol. A 113, 83-90. Brauner, C. J. and Randall, D. J. (1998). The linkage between 02 and 2CO transport. In Fish Physiology: Fish Respiration, vol. 17 (eds S. F. Perry and B. L. Tufts), pp. 283- 321. New York: Elsevier.

Janvier, P. (2007). Living Primitive Fishes and Fishes from Deep Time. In Fish Physiology: Primitive Fishes, vol. 26 (eds D. J. McKenzie, A. P. Farrell and C. J. Brauner), pp. 2-53. New York: Academic Press.

Jensen, F. B. (1989). Hydrogen ion equilibria in fish hemoglobins. I Exp. Biol. 143, 225-234.

Jensen, F. B. (2004). Red blood cell pH, the Bohr effect, and other oxygenation-linked phenomena in blood 02 and 2CO transport. Acta Physiol. Scand. 182, 215-227.

Jensen, F. B., Fago, A. and Weber, R. E. (1998). Red blood cell physiology and biochemistry. In Fish Physiology: Fish Respiration, vol. 17 (eds S. F. Perry and B. L. Tufts), pp. 1-40. New York: Elsevier.

McKenzie, D. J., Farrell, A. P. and Brauner, C. J. (2007). Rish Physiology . Primitive Fishes, vol 26 (eds D. J. Mckenzie, A. P. Farrell and C. J. Brauner), pp xi-xiii. New York: Elsevier.

Moyle, P.B. and Cech, J.J. Jr. (1996). Fishes: An Introduction to Ichthyology. New Jersey: Prentice Hall.

14

Physiology:

York:

Tufts,

247-259.

the

Scholander, oxygen-combining Micron

onitis Root, Root,

Fish

61,

Pelster, Elsevier.

Peister, environ.

Peister,

nucleated

of

Peister,

Nikinmaa,

Nelson,

&

multiple

Sons,

swim

427-456.

Respiration,

(Linn).

Elsevier.

R.

R.

B.

35,

B.

B.

B.

J.

B.

Physiol.

Inc.

W.

W.

bladder

L.

erythrocytes.

S.

and

and

and

and

73-74.

fish

M.

Fish

and

and (1931).

P.

Biol.

(1994).

(1992).

F.

hemoglobins.

Decker,

Randall, Weber,

Weber,

Perry,

Respiration,

of

8,

Irving,

vol.

and

Bull.

power

51-77.

deep

The

Fishes

17

Van

Membrane

Physiol.

S.

R.

84,

R.

H.

respiratory

(eds

L.

sea

D.

of

F.

E.

E.

(2004).

Dam, 207-2

(1943).

whole

J.

of

fishes.

(1998).

I

vol.

S.

(1991).

(1990).

(1998).

the

Exp.

Rev.

F.

12.

L.

transport

17

World,

Perry

and

The

The

I.

function

Biol.

(1954). 72,

(eds

Carbon

The

Influence

Oxygen

Physiology

hemolyzed

Root

effect

301-321.

and

3rd

S.

physiology

149,

15

and

F.

Secretion

B.

of

dioxide

effect

Edition.

dissociation

of

Perry

425-437.

the

control

of

L.

carbon

of

blood

organic

Tufts), blood

—

the

and

transport

a

of

pp.

of

physiological

of

Root

dioxide

the

of

B.

gases

of

pp.

hemoglobin-oxygen

77-96.

phosphates

in

the

L.

Root

marine

blood.

effect.

141-184.

Tufts),

and

marine

against

and

New

effect.

excretion.

fishes.

Biol.

lactic

In

perspective.

pp.

on

York:

fish

high

New

Fish

Adv.

the

Bull.

229-282.

acid

Biol.

Tautogo

pressures

York:

Physiology:

Root

John

comp.

In

affinity

on

107,

Bull.

Fish

Wiley

effect

the

New

in in CHAPTER 2: HAEMOGLOBIN-PROTON BINDING PROPERTIES OF BASAL ACTINOPTERYGIAN FISHES

Introduction

Haemoglobin (Hb) is intimately involved in the transport and delivery of oxygen

(02) in all vertebrate species (Antarctic icefishes excepted), binding 02 at the respiratory surface and releasing it at the metabolically active tissue. Furthermore, it plays an important role in the transport of carbon dioxide )2(C0 in the reverse direction, binding the protons produced by the hydration of 2CO to bicarbonate and therefore increasing the blood’s carrying capacity of .2CO These dual roles of Hb in 02 and 2CO transport are coupled through the Bohr/Haldane effect (Brauner and Randall, 1996; 1998). The protons produced by the hydration of 2CO in the red blood cell (RBC) bind to specific groups (Bohr groups) on the oxygenated (relaxed) R-state Hb which stabilize its deoxygenated (tense) T-state, decreasing its affinity for 02 and subsequently releasing it for diffusion into the tissue (the Bohr effect; Riggs, 1988; Jensen et al., 1998; Jensen,

2004). This R to T transition exposes more histidine residues (His) on the surface of the

Hb molecule which bind additional protons, further stabilizing the T-state as well as aiding in proton-buffering and the transport of 2CO (the Haldane effect; Brauner and Randall, 1998).

Variation in Hb-I-t binding amongfishes

It has recently been proposed that the Bohr effect increased in relative proportion with the Root effect (decreased 02 carrying capacity with low PH; Root, 1931; Peister and Randall, 1998) among the basal actinopterygian lineage of fishes (Berenbrink et al.,

16 2005). The Root effect, something of an exaggerated Bohr effect used to deliver large amounts of 02 to the swimbladder and eye of fishes (Brittain, 1987), has been shown to be correlated with low intrinsic Hb buffer values (Jensen, 1989; Brauner and Weber,

1998; Jensen, 2001; Berenbrink et al., 2005; Berenbrink, 2006). This reduced intracellular proton-binding ability could left-shift the equilibrium of the 2CO hydration reaction, thereby hindering the formation of bicarbonate and the subsequent transport of .2CO However, there appears to be an inverse relationship between intrinsic Hb buffer capacity and oxylabile Hb proton-binding ability (Haldane effect) in the few species that have been studied. It has been shown that plesiomorphic fishes such as elasmobranchs possess Hbs of high buffer capacities and small Haldane effects, while the more derived teleosts possess Hbs of some of the lowest buffer capacities and largest Haldane effects seen among all vertebrate species (Jensen, 1989; Berenbrink et al., 2005). This suggests two different, but equally effective, mechanisms for intracellular proton-binding and subsequent 2CO transport in the blood. Assuming the Hb properties of extant plesiomorphic species to be representative of their respective ancestral states (Janvier,

2007; McKenzie et al., 2007), measuring the Hb buffer capacities and Haldane effects of those fishes intermediate to the elasmobranchs and teleosts on the vertebrate phylogeny may potentially shed light on the evolution of this Hb proton-binding transition.

Thepotential cost of the Root effect

The evolution of the Root effect, presumably among the basal actinopterygians

(Berenbrink et al., 2005), raises another interesting question in regards to Hb proton binding. Having evolved in the absence of RBC 1pH-protecting J3NHE(Berenbrink et al.,

17 2005), the presence of a Root effect in these fishes comes with the potential of losing a significant proportion of the blood’s 02 carrying capacity during a generalized acidosis, and thus may compromise 02 uptake at the gills. However, it is likely that these fishes experience generalized acidoses as a result of hypoxia and exercise in their natural environments (Nelson, 1994; Clack, 2007; Brauner and Berenbrink, 2007; lives and

Randall, 2007). Together with the fact these basal actinopterygian fishes have successfully survived over 250 million years to the present day (Janvier, 2007), it is likely they are capable of coping with such situations. Oxygen carrying capacity would be preserved in these intermediate species so long as their RBC 1pH remained higher than those pH values resulting from a generalized acidosis. This may be the result of either RBC 1pH-protection by some means other than I3NHE,or Root effect onset pH values in these species that are lower than those produced by a generalized acidosis in vivo. The former hypothesis is addressed in the present study, while the latter is addressed in the subsequent study (Chapter 3). Haemoglobin’s pivotal role in blood buffering and its overall abundance within the RBC cytosol suggest it as the potential protector of RBC

1pH during generalized blood acidoses in these intermediate species. Although it has been shown that those species with a Root effect generally have Hbs of low overall buffer value, this may be a trait exclusive to those Root effect species with protective 3NHE.

Berenbrink et al. (2005) reported buffer values for the deoxygenated Hbs of some basal actinopterygian species at a single physiological pH value, the results suggesting a gradual decrease in buffer value with phylogenetic progression among these species.

However, as Hb buffer value varies significantly as a function of both oxygenation status and pH (Jensen et al., 1998), accounting for these variables in the context of a species’

18 particular Root effect onset pH value may shed additional light on whether or not the buffering properties of Hb are in fact playing a key role in protecting RBC ,1pH and with it, 02 uptake, during a generalized acidosis.

It is pertinent, therefore, to measure the proton-binding properties of the oxygenated and deoxygenated Hbs of species within the basal actinopterygian lineage over a broad pH range. Both the intrinsic buffering capacity of Hb and the Haldane effect can be measured simultaneously through the acid-base titrations of oxygenated and deoxygenated Hbs. In the present study, these titrations were performed on isolated Hbs of the following species, expressed in order of most basal to most derived: American paddlefish (Polyodon spathula), white sturgeon (Acipenser transmontanus), spotted gar

(Lepisosteus oculatus), alligator gar (Atractosteus spatula), bowfin (Amia calva), mooneye (Hiodon tergisus), and pirarucu (Arapaima gigas), all species which occupy a transitional phase in the evolution of both the Bohr/Haldane effect and the Root effect

(Berenbrink et al., 2005). The objective of this study was to perform Hb titrations on the haemolysates of these fishes (with and without saturating levels of organic phosphates) to characterize the Hb proton binding properties of a particularly interesting group in the evolution of vertebrates, as well as to help understand how a large Root effect could have evolved in the absence of RBC f3NHE.

Materials and methods

Animal acquisition

White sturgeon (Acienser transmontanus; 1-2 kg) and bowfin (Amia calva; 0.5- 1.0 kg) were kept in large flow-through outdoor tanks (dechlorinated city water; 2>Po

19 130 torr; 2Pco <0.1 torr; T = 11-17°C; fish density < 25 kg fish per 3m water) at the University of British Columbia, Vancouver, Canada, prior to sampling. Pirarucu

(Arapaima gigas; 1-2 kg) were maintained in outdoor static tanks at the National Institute for Research of the Amazon (INPA). American paddlefish (Polyodon spathula), spotted gar (Lepisosteus oculatus), alligator gar (Atractosteus spatula) and mooneye (Hiodon tergisus) were all sampled immediately after being caught in their respective natural habitats. Samples for each species consisted of blood from at least three adult fish of either sex, with the exception of spotted gar which consisted of the blood of a single adult.

Haemolysate preparation

Whole blood was drawn from the caudal vein of anaesthetized animals (1 ml of

200 mmol l’ benzocaine per litre of water; Sigma E1501) into heparinized syringes, where red cells were then separated by centrifugation and washed three times in cold

Cortland’s physiological saline (Wolfe, 1963). The red cells were lysed by addition of two-times volume cold deionized water and subsequent freezing, and cell debris was removed by ten minutes of chilled (4°C) centrifugation at 14 000 r.p.m. (Thermo Electron

Corporation 21000R, Waltham, MA, USA). The haemolysates were purified by removing cell solutes and organic phosphates by repeated passage through mixed-bed ion exchange columns (Amberlite LRN-150mixed bed ion exchange resin), and were then divided into 0.5 ml aliquots and frozen at -80°C until use.

20 Haemoglobin titrations

Haemoglobin titrations were conducted on concentrated stripped haemolysates that were thawed and diluted to a final concentration of 40 jimol F’ Hb and 0.1 mol 1F

KC1according to the methods of Jensen (1989). Haemoglobin concentration was confirmed after conversion to cyanomethaemoglobin using a millimolar extinction coefficient of 11 at 540 nm, and methaemoglobin content was assessed on identical sub- samples using the spectrophotometric method of Benesch et al. (1973). Samples where methaemoglobin exceeded 10% of total Hb were discarded. A 2 ml volume of the haemolysate (in the absence or presence of GTP at a molar ratio of 3:1 relative to the tetrameric haemoglobin (GTP4)):Hb was then transferred to a chilled (12°C), magnetically stirred glass titration vessel where the it was equilibrated with humidified oxygen (100%) for 90 minutes. After reaching a stable pH reading, hydrogen ion titrations were performed with an automated Radiometer (Copenhagen, Denmark)

TitraLab 90 titration apparatus, where 0.01 mol F’ NaOH was added in 10 il increments to raise pH from isoionic to approximately pH 9.2. After five minutes of equilibration at a pH of 9.2, titration with 0.01 mol F’ HC1(10 iil increments) was initiated and continued until pH 5.2 was reached, allowing seven minutes of equilibration time between injections. Total amount of NaOH or HC1to reach these end points was recorded. The same procedure was performed on a separate 2 ml sample from the same stock haemolysate, this time equilibrated in humidified nitrogen (100%). Three to four separate H titrations were performed on the haemolysates of each species, yielding reproducible results.

21 Data analysis

The Hb buffer values were derived from the negative slope between adjacent points on the oxygenated and deoxygenated Rb titration curves for each species. Because there are large differences in Rb proton binding over small pH ranges (Jensen et al.,

1998), as well as large differences in proton binding between oxygenated and deoxygenated Rb at a constant pH (Jensen et al., 1998), the Rb P50 buffer value was used to derive a single value for each species that would be more representative of in vivo Hb buffer properties. These Rb P50 buffer values were calculated for each of three different, physiologically relevant pH ranges: pR 6.8-7.0, pR 7.0-7.2, and pR 7.2-7.4. The resulting Hb P50 buffer value for (example) pH 6.8-7.0 therefore accounted for variation in proton binding brought about by both oxygenation status and minor fluctuations in pH between 6.8 and 7.0.

The fixed acid Haldane effects of these species were determined by calculating the vertical distances between their oxygenated and deoxygenated titration curves (\ZH, mol H taken up per mol 4Hb upon deoxygenation at constant pR). And by plotting the inverse negative slope (-ApH/z\Zu) of the titration curves as a function of ZR,the inflection points on the titration curves become localized at two peaks. Quantification of the AZHbetween these peaks (ZKbeing a molar ratio of protons to )4Rb gives an accurate measurement of the number of groups being titrated in the physiological pR range (De

Bruin and van Os, 1968; Jensen, 1989) and was determined for bowfin, mooneye, and pirarucu. The peaks on the differentiated curves of the other four species could not be accurately resolved due to the linear nature of their titration curves.

22 Results

Haemoglobin titrations

Representative H titration curves of oxygenated and deoxygenated haemoglobin from American paddlefish, white sturgeon, spotted gar, alligator gar, bowfin, mooneye, and pirarucu are drawn to the same scale in Fig. 2.1, showing how net proton charge (ZR, mol H moi’ )4Hb of stripped Hbs in 0.1 M KC1changes as a function of pH in the absence and presence of GTP (3:1 molar ratio of 4).GTP:Hb Zero net proton charge refers to the ioselectric pH and is used as the reference point (Tanford, 1962).

Haemoglobin Buffer Values

The slope of each Hb titration curve indicates the Hb buffer value at that pH (mol H moF’ 4Hb pH unitj, which varies according to pH, oxygenation status, GTP concentration and species (Fig. 2.2). The Hb P50 buffer values in the presence of GTP varied among the seven species at physiological pH (pH 7.2-7.4), showing a decreasing trend with phylogenetic progression that becomes particularly pronounced starting with bowfin (Fig. 2.3). Pirarucu is a notable exception to this, its Hb exhibiting a relatively high buffer value at physiological pH compared to the other more derived species studied

(bowfin and mooneye). This trend of decreasing Hb buffer value with phylogenetic progression becomes less apparent under more acidic conditions (pH 7.0-7.2 and pH 6.8-

7.0). Whereas the Hb P50 buffer values of the four more plesiomorphic species tended to decrease at the more acidic pH ranges, those of the three more derived, rete-bearing species showed a marked increase (Fig. 2.3). The same general trends were apparent in

23

with paddlefish,

were performed

on

presence

plesiomorphic

(Fig.

absence

exhibiting curves

studied deoxygenation

and

the

lack

the

stripped

species

the of

determined

2.4b).

surfaces

indicates

Due

The

allosteric

species,

of

methods

of

considerably

for

GTP,

sturgeon,

vertical

GTP

(Fig.

to

Hbs,

the

a

species

at

of

with

lack

the

as

from

modification

(Fig.

2.4).

of

with

constant

Hbs

their

well

distance,

Jensen fixed

of

the

spotted

the

The

(paddlefish,

2.4a).

of

lower

Hbs

conspicuous

larger

as

three

bowfin,

acid

titration

pH),

magnitude

left

(1989).

were

overall

gar,

or

These

maximum

rete-bearing

through

Haldane

shifted which

AZH,

16, and

mooneye,

curves

sturgeon,

Titratable

maxima

Hb

inflection

Haldane

14,

between

alligator

of

also

on

GTP

effect

P 50

and

oxylabile

Haldane

the

of

varies

species

24

buffer

and

binding

deoxygenated,

were

spotted

17,

pH

(z\ZH,

effects

groups

the

gar,

points

pirarucu.

respectively

axis

according

effects

oxygenated

slightly

values

proton

(bowfin,

this

mol

(data

gar,

on

by

analysis

the H

approximately

alligator

than

for

binding

not

The less

stripped

to

mooneye,

taken

Hb

a

(Fig.

shown).

the

and

pH,

for

given

number

could

titration

four

gar),

up

varied

each

deoxygenated

2.5).

GTP

Hbs

pH

per

only

pirarucu)

more

of

particularly

species

0.2-0.4

concentration,

range

curves

These

in

mol

among

groups

be

accordance

Hb 4

objectively

due

numbers

in

pH

of

the

titration

titrated

the

upon

to

units

in

a the

reducing

(GTP)

bearing

none Hb

significant

the preserve

the

acidosis,

among

1975; of therefore Haldane

plesiomorphic

magnitudes

large

the

buffer

Hb

same

of

Haldane

Pelster

present

and

Nucleoside proton-binding

the

In

bowfin

the

02 effects,

Hb

despite

group

likely

value

fish

basal

role

adenosine

Hb

of uptake

02

and

study

effects

Hbs,

Hb

species

proton-binding

in

from

in

occurred

affinity

decreasing

actinopterygian

while

a

Weber,

the

which

buffer

lack

triphosphates

in

an

was

fairly

triphosphate

preservation

(Jensen

these

transition

(elasmobranchs)

inverse

derived

of

through

to

the

among capacities

1990;

RBC

steady

determine

and

intermediate

Root

et

relationship

properties

species

Haldane

NHE

Brauner

the

al.,

the

was

fishes,

(NTPs)

ancestral

effect

(ATP)

02

and

Influence

stabilization

1998).

intermediate

not

the

uptake

activity. (teleosts)

Discussion

tend

oxylabile

and

evolved

and

effect

Root

have

completely

nature

of

are

has

25

levels This

to

these

Weber, to

the

during

of

been

effect

see

been

increasing

have

The

of

GTP tend

transition

predominant

of

(Berenbrink

group,

Haldane

species

in

whether

this

the

shown

generalized

shown

high

the

findings

species

1998).

gradual,

to

protein’s

Hb

have

four

the

Hb

suggest markedly

to

in

effects,

proton-binding

these

to

basal

Guanosine

during

basal

buffer

Hb

influence

but

suggest

low

et

NTPs

exist

al.,

acidoses.

T-state

Hb

proton-binding

rather

Hb

they

actinopterygian

such

species.

between

a

2005).

starting

values

in

properties

generalized

buffer

that

play

the

Hb

triphosphate

punctuated,

that

(Weber

the

transition

RBCs

function

and

The

any

However,

with

values

the

nature

may

small

objective

et

strategy

the

of

blood

fishes,

al.,

and

by

with

help

of rete fishes (Va!, 2000), with GTP exerting a greater effect on Hb 02 affinity due to an additional T-state stabilizing bond in many teleosts (Peister and Weber, 1990; Val, 2000).

For this reason, all of the experiments in the present study were performed in both the presence and absence of saturating levels of GTP (3:1 4).GTP:Hb As well as reducing Hb 02 affinity, the allosteric effects of GTP expose more proton-binding groups (His residues) on the surface of the fib molecule, in addition to altering the molecular microenvironrnents of existing His residues in such a way that increases their pK values

(Peister and Weber, 1990; Jensen et al., 1998). This results in a respective increase in overall proton-binding ability and a right shift of the titration curve (and resulting buffer curve) along the pH scale for a GTP-bound fib when compared to a stripped Hb (Figs.

2.1, 2.2). The degree to which these effects are manifest varies among species. For instance, bowfin Hbs showed much greater GTP-dependent variation in buffer capacity than did sturgeon Hbs. This is likely the result of the conformational changes to Hb resulting from GTP binding exposing more titratable groups in bowfin Hb relative to sturgeon Hb.

The Haldane effect is also influenced by GTP in that it is manifest to a greater degree and over a narrower, higher pH range when GTP is bound to Hb. Again, this effect varies interspecifically, but among the species studied here it is particularly apparent in the three most derived species with the largest Haldane effects — bowfin, mooneye, and pirarucu (Figs. 2.1, 2.4).

The remainder of the discussion will focus primarily on those experiments done in the presence of GTP, as we believe these to be more representative of the in vivo conditions of these fishes.

26 Haemoglobin buffer curves

The capacity of Hb as a buffer is a function of the type and number of amino acid residues capable of binding protons over a given pH range, as well as the molecular microenvironments in which these residues are located. These titratable amino acid residues are divided into three pH-dependent classes (Tanford, 1962): the guariidyl group of arginine and the amino group of lysine are titrated at basic pH values (>pH 9.0); the carboxyl groups of glutamic acid and aspartic acid are titrated at acidic pH ranges (

6.0); and the imidazole group of histidine residues (His) and the terminal ct-amino groups are titrated in the “physiological” pH range (pH 6.0-9.0). It is the disparity in pK values of the titratable groups on the Hb molecule that leads to the variation in proton binding ability as a function of pH (Fig. 2.2). Furthermore, the T- and R-states of a Hb molecule come with changes in the protein conformation, leading to variation in the exposed titratable groups and their microenvironments. This results in intra-molecular differences in oxygenated and deoxygenated Hb buffer curves (Fig. 2.2), a phenomenon which becomes more apparent with phylogenetic progression due to increases in the Haldane effect and, subsequently, variation in the oxygenated and deoxygenated Hb titration curves for a given species (Fig. 2.1). Similarly, the allosteric effects of GTP influence the proton-binding abilities of Hb for the reasons outlined above (Figs. 2.1, 2.2).

The microenvironment of Rb within the RBC is in a constant state of flux, and, as outlined in the foregoing discussion, the ability of Hb to effectively bind protons is subject to this microenvironment. When alluding to the general ability of a particular Hb to bind protons, it is therefore important to account for such variation brought about by

27 microenvironmental fluctuations within the RBC. Hence, the Hb P50 buffer value.

Describing the ability of a Hb molecule to bind protons in this manner accounts for variation in both oxygenation status and pH fluctuation, two variables which it is likely to encounter in circulation.

Haemoglobin P50 buffer values There was a general trend of decreasing Hb P50 buffer values with phylogenetic progression among the seven studied species (Fig. 2.3). This trend was especially apparent in the physiological pH range (pH 7.2-7.4), and became less pronounced as conditions became more acidic (pH 7.0-7.2 and pH 6.8-7.0; Fig. 2.3). It has repeatedly been observed that the Hb buffer capacities of teleost fishes are significantly lower than those of almost all other vertebrate species (Jensen, 1989; Brauner and Weber, 1998;

Jensen et al, 1998; Jensen, 2001; Berenbrink et al, 2005; Berenbrink, 2006; Brauner and

Berenbrink, 2007), a trait ultimately attributable to a lower number of proton-binding groups (i.e., His residues) on the molecule itself (Riggs, 1970; Jensen, 1989; Berenbrink et al, 2005; Berenbrink, 2006). For instance, human HbA contains 38 His residues per tetramer (Braunitzer et a!., 1961), while that of the elasmobranch, spiny dogfish, contains

40 His residues (Aschauer et a!., 1985). In comparison, the tetramers of two teleost species, trout and carp, contain only 16 and 18 His residues, respectively (Hilse and

Braunitzer, 1968; Bossa et al., 1978). Appropriately, the Hb buffer capacities of these species reflect the number of His residues that make them up, with both human and dogfish displaying high Hb buffer capacites, and trout and carp displaying low Hb buffer capacities (Siggaard-Anderson, 1975; Jensen, 1989). However, the overall His content of

28 a Hb molecule does not fully account for its ability to buffer protons, as some of these residues may have pK values too low to be operational in vivo, or, more regularly, be buried within the protein moiety and unavailable for titration. Thus, the true reflection of a Hb’s ability to buffer protons is the number of titratable groups available on its surface.

For the aforementioned Hbs of human and dogfish, these numbers are 26 and 30, respectively, while trout and carp display 7 and 6 groups on the surface of their respective

Hbs that are available for titration (Berenbrink, 2006). Through the differential plotting of the Hb titration curves (i.e., -ApH/AZH)as a function of ZH, the inflection points on the regular titration curves become localized as two separate peaks (Fig. 2.5), the difference between which gives an accurate measure of the number of groups being titrated (De

Bruin and van Os, 1968; Jensen, 1989). Due to the linear shape of their Hb titration curves, well-resolved peaks were hard to discern in the differentially-plotted titration curves of the ancestral species (paddlefish, white sturgeon, spotted gar, and alligator gar).

However, the differential plots of the deoxygenated, stripped Hbs of bowfin, mooneye, and pirarucu indicated approximately 16, 14, and 17 titratable groups per tetramer, respectively (Fig. 2.5). These numbers agree well with the predictions of Berenbrink et al. (2005) based on a correlation of the buffer values of Hbs with known numbers of titratable groups on their surfaces.

The Haldane effect

It is believed that the ancestral state of tetrameric Hb among fishes is that of a high buffer value, while the Hbs of more derived teleost fishes display much lower buffer values (Jensen, 1989; Jensen et al., 1998). With a low Hb buffer value comes the

29 potential ofjeopardized 2CO transport in the blood. Teleost fishes compensate for this by way of a large Haldane effect, a different, but equally viable strategy for accounting for the protons produced by 2CO hydration and resulting in a tight interaction between 02 and 2CO exhange (Brauner and Randall, 1996; 1998). By and large, this trend was illustrated in the present study, where it was shown that those species with low Hb P50 buffer values were characterized by large Haldane effects, while those with high Hb P50 buffer values had small Haldane effects (Figs. 2.3, 2.4). The exception to this is pirarucu, a species that displays both a high Hb buffer capacity and a large Haldane effect (possible reasons for this are discussed below). In the absence of this anomalous species, there is a significant negative correlation between Hb P50 buffer value and the Haldane effect in the physiological pH range (data not shown; r2 = 0.805; P <0.05). Therefore, despite the lower intrinsic Hb buffer capacities of the more derived fishes in this study, their large Haldane effects likely ensure 2CO transport and removal is not jeopardized.

Transitions in Hb-I-f binding in light of the Root effect

The data for these transitional species do not display a gradual progression towards a low Hb P50 buffer value and a large Haldane effect so much as they do a dichotomy between the four basal species (paddlefish, sturgeon, spotted gar, alligator gar) and the three derived species (bowfin, mooneye, pirarucu; Figs. 2.1 — 2.4). This suggests a more punctuated evolution of these Hb traits as opposed to a more gradual evolution, and appears to be associated with the presence of a choroid rete in bowfin, mooneye, and pirarucu. The choroid rete is a countercurrent capillary network located at the eye that is used to both localize and maximize the acidosis necessary to activate the Root effect, an

30 arrangement these fishes use to deliver high levels of oxygen to the avascularized retinal tissue. This tissue therefore allows for the exploitation of the Root effect, itself a Hb property shown to be correlated with low Hb buffer capacity and large Haldane effect

(Jensen et al., 1998; Jensen, 2001; Berenbrink et al., 2005; Berenbrink, 2006). Together with the findings of the related study (Chapter 3) which show a similar dichotomy between rete-bearing and non-rete species in relation to their Root effect characteristics, it may be sensible to expect traits associated with the Root effect (i.e., low Hb buffer capacity; large Haldane effect) to vary correspondingly.

The Root effect is believed to have evolved among the basal ray-finned fishes in the absence of RBC 1pH-protecting I3NHEactivity (Berenbrink et al., 2005). Combined with the low Hb buffer capacities of these fishes, 02 transport could be jeopardized during a generalized blood acidosis resulting from exercise, hypoxia, or hypercarbia. The seven species of the present study all occupy this transitional group, and it was hypothesized that their Hb proton-binding properties may deviate from those of 13N}IE- equipped Root effect species in such a way that would protect RBC 1pH under such conditions, and with it, 02 uptake at the gill. This would be manifest as higher intrinsic buffer capacities and larger Haldane effects. However, the results suggest that this is not the case. The buffer capacities and Haldane effects of the four basal species’ Hbs

(paddlefish, sturgeon, spotted gar, alligator gar) do not deviate in any significant way from those of the more plesiomorphic elasmobranchs (Jensen, 1989) that lack the Root effect altogether (Berenbrink et al., 2005). Likewise, the Hb proton-binding properties of the rete-bearing bowfin, mooneye, and pirarucu do not differ from those of more derived

3NETE-equippedRoot effect species such as trout, carp, eel, and tuna (Jensen, 1989;

31 Brauner and Weber, 1998; Jensen, 2001; pirarucu excepted in the case of intrinsic Hb buffering for reasons stated below). This suggests that 02 uptake during generalized blood acidoses is not likely preserved in these species by way of RBC 1pH protection on the part of Hb. As the related study shows (Chapter 3), this is instead attributed to relatively low Root effect onset pH values.

Although the reasons for it are not fully resolved, a low Hb buffer value may help to optimize the utilization of the Root effect insomuch as fewer protons need transferring to the RBC to drive the cytosol down to Root effect onset pH. This would allow for optimal oxygen delivery via both Bohr and Root effects, but could potentially come at the detriment to 2CO transport. However, the correlation of large Haldane effects with low Hb buffer values and the presence of a Root effect ensures 2CO transport is not jeopardized. In this way, a species may take advantage of efficient proton-dependent 02 delivery without any cost to their ability to transport and remove .2CO

Postulating pirarucu

The Hb proton binding properties of pirarucu are somewhat anomalous in that they do not display the low intrinsic buffer capacity typical of other teleost Hbs. Rather, pirarucu displays Hb buffer capacities equal to, or even greater than, those of more plesiomorphic species (Figs. 2.2, 2.3). Moreover, unlike the presumably ancestral-state

Hbs of the more plesiomorphic species, the Hbs of pirarucu display a large Haldane effect (Figs. 2.1, 2.4), uncharacteristic of a Hb with a low intrinsic buffer value (Jensen,

1989; Brauner and Weber, 1998). Taken together, these two phenomena suggest pirarucu

32 as having an exceptional ability to bind protons in the blood, and subsequently, an exceptional ability to transport .2CO But why? Pirarucu are obligate air-breathers, securing roughly 80% of their 02 at their air- breathing organ (Randall et al., 1978). In fact, if denied access to air for more than 10 minutes, pirarucu will actually drown (Val and Almeida-Val, 1995). This breathing strategy results in high levels of blood ,2C0 with a 2Pco measured at approximately 27 mmHg in arterial blood (Randall et al, 1978) compared to the arterial 2Pco values of trout (2.9 mmHg; Perry et al, 1996), eel (2.4 nnnHg; Perry et al, 1996), turbot (2.5 mmHg;

Perry et a!, 1996), and dogfish (1.1 mmHg; Perry et al, 1996). As roughly 90% of 2CO is transported in its hydrated form in the blood of fishes (Tufts and Perry, 1998), the high levels of 2CO measured in the blood of pirarucu has the potential to produce a large quantity of acid once hydrated and dissociated into 3HC0 and H. In this situation, having a large Hb buffer capacity to account for these protons and, subsequently, facilitate 2CO transport and protect against a generalized blood acidosis would be beneficial. This is especially true for pirarucu, who, unlike many other air-breathing fishes, do not display haematocrits and whole blood Hb concentrations considerably higher than those of water-breathing fishes with regular 2Pco values (Graham, 1997). The occurrence of this high intrinsic Hb buffer capacity may then be an adaptation paralleling the increased aerial dependence of this species. As air-breathing is thought to have evolved as many as 67 times among the fishes (Graham, 1997), there is no shortage of species among which to compare Hb buffer values to test this hypothesis.

33 Conclusions

The findings of the present study suggest that the transition in Hb proton-binding strategy from large intrinsic buffering/small Haldane effect to low intrinsic buffering/large Haldane effect occurred not by gradual succession among the primitive ray-finned fishes, but rather a more punctuated method. A marked decrease in buffer value and increase in the Haldane effect appears to have occurred in the last common ancestor of bowfin and the teleosts, and is correlated with the first appearance of the choroid rete. As the choroid rete has also been shown to be correlated with a significant increase in the magnitude and onset pH of the Root effect (Chapter 3), it is possible that this transition in Hb proton-binding is in some way related to the optimization and/or utilization of the Root effect in these fishes — a low intrinsic Hb buffer value would reduce the number of necessary protons shifted to the RBC cytosol to activate the Root effect, while a large Haldane effect would ensure 2CO transport did not suffer despite the decreased Hb buffer capacity. Furthermore, being linked mechanistically to the Bohr effect, a large Haldane effect would allow for the tight coupling of 02 and 2CO transport in the blood of these fishes (Brauner and Randall, 1996; 1998). However, this transition in Hb proton-binding would not necessarily increase the overall proton-binding capacity of Hb in these basal ray-finned species to the extent that it would protect 02 uptake in the case of a generalized blood acidosis as hypothesized. In light of the findings of the related study, however, this is likely not a necessary characteristic of these species’ Hbs despite their lack of RBC I3NIIE, as the onset pH values of their respective Root effects all appear to be lower than those resulting from even a severe generalized acidosis of the blood (Chapter 3).

34 Figure 2.1. Representative Hb H titration curves, ZR(mol H mol )4Hb as a function of pH, for oxygenated (•) and deoxygenated (0) Hbs of American paddlefish (A), white sturgeon (B), spotted gar (C), alligator gar (D), bowfin (E), mooneye (F), and pirarucu (G) in the presence (black) and absence (grey) of organic phosphates (3:1 GTP:Hb Titrations were performed on haemolysates at a 4[Hb of 0.04 mmol l’ and a {KC1]4). of 0.1 mol F’. The presence of a choroid rete within a speciesj is indicated by ®.

35 25 A

5 6 7 8 95 6 7 8 95 6 7 8 95 6 7 8 9

220® E pH

(1) 2 15 >Cu i_ 10

0 5 6 7 8 95 6 7 8 95 6 7 8 9

pH

Figure 2.2. Representative buffer curves (-AZH/ApH)as a function of pH for oxygenated (•) and deoxygenated (0) Hbs of American paddlefish (A), white sturgeon (B), spotted gar (C), alligator gar (D), bowfin (E), mooneye (F), and pirarucu (0) in the presence (black) and absence (grey) of organic phosphates (3:1 GTP:Hb Titrations were performed on haemolysates at a ]4[Rb of 0.04 mmol F’ and4).a [KC1]of 0.1 mol F’. The presence of a choroid rete within a species is indicated by ®.

36 18 C ID 16 I- 14 12 E + 10 I — pH 6.8-7.0 0 pH 7.0-7.2 8 — pH 7.2-7.4

U) D 6 >a5 4

2

I 0 Paddefish Sturgeon Spotted Gar AlligatorGar Bowfin Mooneye Pirarucu ©

Figure 2.3. Haemoglobin P50 buffer values in the presence of GTP (3:1 GTP:Hb for seven basal actinopterygian species. Values were determined by averaging4 all points along the oxygenated and deoxygenated buffer curves (Fig. 2.2) within each) of the three physiologically-relevant pH ranges (pH 6.8-7.0; pH 7.0-7.2; pH 7.2-7.4) for each species. The presence of a choroid rete within a species is indicated by ®.

37 ______

5

A ——0-—— Paddlefish ——s-—- White sturgeon 4 — -G — Spotted gar — - — Alligatorgar • Bowfin ® 3 • Mooneye ® A I Pirarucu ® N 2

\ 0

5— B

4

3 I N 2

0

—1 5 6 7 8 9

pH

Figure 2.4. Fixed-acid Haldane effects (AZH; the number of protons taken up per 4Hb upon deoxygenation at constant pH) as a function of pH for seven basal actinopterygian species in the absence (A) and presence (B) of GTP (3:1 GTP4).:Hb AZHwas calculated from the vertical distance between the oxygenated and deoxygenated titration curves (Fig. 2.1) for a given species. The presence of a choroid rete within a species is indicated by ®.

38 0.35

030 zZH=l6

0.25

0 20

0.15

0.10

0.05

0.35

0.30 AZH=l4

0.05

0 35 7z\ZH=l

0.05 0 5 10 15 20 25 ZH(mol W 1mor )4Hb

Figure 2.5. Representative differential titration curves (-pWAZH; the inverse of buffer value) as a function of ZH, taken from the deoxygenated titration curves of bowfin (A), mooneye (B), and pirarucu (C) in the absence of GTP, according to the methods of Jensen (1989). The horizontal distance between the inflection points indicates the number of titratable groups in the neutral pH range (approx. pH 6.0-9.0).

39 -CHAPTER SUMMARY-

1. Assuming the Hb properties of these seven species to be representative of their

respective ancestral states, the transition in Hb proton-binding from a high Hb

buffer value/small Haldane effect to a low Hb buffer value/large Haldane effect

appears to have occurred in a somewhat punctuated manner in the last common

ancestor of the amiiformes and the teleosts. This is correlated with the first

appearance of the choroid rete, as well as a significant increase in the magnitude

and onset pH value of the Root effect (Chapter 3).

2. When compared to those of more plesiomorphic species lacking the Root effect,

as well as more derived species with both the Root effect and protective 3N}TE

activity, it is unlikely the Hb proton-binding properties of these seven species as playing a significant role in protecting RBC ,1pH and with it, 02 uptake, during generalized blood acidoses.

40 References

Aschauer, H., Weber, R. E. and Braunitzer, G. (1985). The primary structure of the hemoglobin of the dogfish shark (Squalus acanthias). Antagonistic effects of ATP and urea on oxygen affinity of an elasmobranch hemoglobin. Biol. Chem. Hoppe-Seyler 366, 589-599.

Benesch, R. E., Benesch, R. and Yung, S. (1973). Equations for spectrophotometric analysis of hemoglobin mixtures. Anal. Biochem. 55, 245-248.

Berenbrink, M. (2006). Evolution of vertebrate haemoglobins: Histidine side chains, specific buffer value and Bohr effect. Respir. Physiol. & Neurobiol. 154, 165-184.

Berenbrink, M., Koldkjaer, P., Kepp, 0. and Cossins, A. R. (2005). Evolution of oxygen secretion in fishes and the emergence of a complex physiological system. Science 307, 1752-1757.

Bossa, F., Barra, D., Petruzzelli, R., Martini, F. and Brunori, M. (1978). Primary structure of hemoglobin from trout (Salmo irideus). Amino acid sequence of a chain of trout I. Biochim. biophys. Ada 536, 298-305.

Brauner, C. J. and Randall, D. J. (1996). The interaction between oxygen and carbon dioxide movements in fishes. Comp. Biochem. Physiol. A 113, 83-90. Brauner, C. J. and Randall, D. J. (1998). The linkage between 02 and 2CO transport. In Fish Physiology: Fish Respiration, vol. 17 (eds S. F. Perry and B. L. Tufts), pp. 283- 321. New York: Elsevier.

Brauner, C. J. and Weber, R. E. (1998). Hydrogen ion titrations of the anodic and cathodic haemoglobin components of the European eel Anguilla anguilla. I Exp. Biol. 201, 2507-25 14.

Brauner, C.J. and Berenbrink, M. (2007). Gas Exchange. In Fish Physiology. Primitive Fishes, vol 26 (eds D. J. McKenzie, A. P. Farrell and C. J. Brauner), pp. 213- 282. New York: Elsevier.

Braunitzer, G., Gehring-Muller, R., Hilschmann, N., Hilse, K., Hobom, G., Rudloff, V. and Wittmann-Liebold, B. (1961). Die konstitution des normalen adulten Humanhamoglobins. Hoppe-Seyler Z.physiol. Chem. 325, 283-286.

Brittain, T. (1987). The Root effect. Comp. Biochem. Physiol. B: Biochem. Mol. Biol. 86, 473-48 1.

Clack, J. A. (2007). Devonian climate change, breathing, and the origin of the tetrapod stem group. Integ. Comp. Biol. 47, 510-523.

41 De Bruin, S. H. and van Os, G. A. J. (1968). Determination and interpretation of the differential titration curves of proteins: the differential curves of bovine methemoglobin, CO-hemoglobin and serum albumin. Rec. Tray. Chim. 87, 861-872.

Graham, J. B. (1997). Diversity and natural history. In Air Breathing Fishes. pp. 13-64. New York: Academic Press.

Hilse, K. and Braunitzer, G. (1968). Die Aminosauresequens der a-ketten der beiden Hauptkomponenten des Karphenhamoglobins. Hoppe-Seyler Z.physiol. Chem. 349, 433-450. lives, K. L. and Randall, D. J. (2007). Whyhave primitive fishes survived? In Fish Physiology: Primitive Fishes, vol. 26 (eds D. J. McKenzie, A. P. Farrell and C. J. Brauner), pp. 515-536. New York: Academic Press.

Janvier, P. (2007). Living Primitive Fishes and Fishes from Deep Time. In Fish Physiology: Primitive Fishes, vol. 26 (eds D. J. McKenzie, A. P. Farrell and C. J. Brauner), pp. 2-53. New York: Academic Press.

Jensen, F. B. (1989). Hydrogen ion equilibria in fish hemoglobins. I Exp. Biol. 143, 225-234.

Jensen, F. B. (2001). Hydrogen ion binding properties of tuna haemoglobins. Comp. Biochem. Physiol., A. Mol. Integr. Physiol. 129, 511-517.

Jensen, F. B. (2004). Red blood cell pH, the Bohr effect, and other oxygenation-linked phenomena in blood 02 and 2CO transport. Acta Physiol. Scand. 182, 215-227. Jensen, F. B., Fago, A. and Weber, R. E. (1998). Red blood cell physiology and biochemistry. In Fish Physiology: Fish Respiration, vol. 17 (eds S. F. Perry and B. L. Tufts), pp. 1-40. New York: Elsevier.

McKenzie, D. J., Farrell, A. P. and Brauner, C. J. (2007). Rish Physiology . Primitive Fishes, vol 26 (eds D. J. Mckenzie, A. P. Farrell and C. J. Brauner), pp xi-xiii. New

York : Elsevier.

Nelson, J. S. (1994). Fishes of the World,3rd Edition. pp. 77-96. New York: John Wiley & Sons, Inc.

Peister, B. and Weber, R. E. (1990). Influence of organic phosphates on the Root effect of multiple fish hemoglobins. I Exp. Biol. 149, 425-437.

Peister, B. and Randall, D. J. (1998). Physiology of the Root effect. In Fish Physiology: Fish Respiration, vol. 17 (eds S. F. Perry and B. L. Tufts), pp. 141-184. New York: Elsevier.

42 Perry S.F., Wood C.M., Walsh P.J. and Thomas S. (1996). Fish red blood cell carbon dioxide transport in vitro: a comparative study. Comp. Biochem. Physiol. A. 113, 121- 130.

Randall, 0. J., Farrell, A. P. and Haswell, M. S. (1978). Carbon dioxide excretion in the pirarucu (Arapaima gigas), an obligate air-breathing fish. Can. I Zool. 56, 977-982.

Riggs, A. F. (1970). Properties offish hemoglobins. In Fish Physiology, vol. 4 (eds W. S. Hoar and D. J. Randall), pp. 209-252. New York: Academic Press.

Riggs, A. F. (1988). The Bohr effect. Annu. Rev. Physiol. 50, 181-204.

Root, R. W. (1931). The respiratory function of the blood of marine fishes. Biol. Bull. 61, 427-456.

Siggaard-Andersen, 0. (1975). TheAcid—BaseStatus of the Blood, 4th ed. Munksgaard, Copenhagen.

Tanford, C. (1962). The interpretation of hydrogen ion titration curves of proteins. Adv. Protein Chem. 17, 69-165.

Tufts, B. L. and Perry, S. F. (1998). Carbon dioxide transport and excretion. In Fish Physiology: Fish Respiration, vol. 17 (eds S. F. Perry and B. L. Tufts), pp. 229-282. New York: Elsevier.

Val, A. L. (2000). Organic phosphates in the red blood cells of fish. Comp. Biochem. Physiol., A: Mol. Integr. Physiol. 125, 417-435.

Vat, A. L. and Almeida-Val, V. M. F. (1995). Fishes of the Amazon and their environment: physiological and biochemical aspects. Berlin: Springer-Verlag.

Weber, R. E., Lykkeboe, G. and Johansen, K. (1975). Biochemical aspects of adaptation of hemoglobin oxygen affinity of eels to hypoxia. Lfe Sci. 17, 1345-1349.

Wolfe, K. (1963). Physiological salines of fresh water teleosts. Frog. Fish Cult. 25, 135- 140.

43 CHAPTER 3: ROOT EFFECT HAEMOGLOBIN CHARACTERISTICS IN BASAL ACTINOPTERYGIAN FISHES

Introduction

Over its 450 million year history (Dickerson, 1971), haemoglobin (Hb) has been the subject of countless selective pressures in countless species due to its functional position at the interface of the organism and its environment (Powers, 1980; Berenbrink,

2006). This has resulted in a remarkable degree of heterogeneity in protein structure and function, with fishes in particular displaying extraordinary variation in Hb multiplicity and function (Peister and Weber, 1990; Weber, 1990; Jensen et al., 1998; Bonaventura et al., 2004). These molecular characteristics have been shaped by the selective pressures brought about by new niches and environments and, in some cases, are believed to have contributed to the relative success of whole classes of species. One such example is the

Root effect, a characteristic of certain fish Hbs allowing for the enhanced delivery of oxygen (02) to particular tissues. It is believed that this Root effect has played a major role in the explosive adaptive radiation of a group of fishes that accounts for half of all living vertebrate species in the world today — the teleosts (Weber, 2000; Bonaventura et a!., 2004; Berenbrink et al., 2005; Berenbrink 2007).

The Root effect

At the level of Hb, the Root effect is thought to be an exaggerated Bohr effect

(Brittain, 1987), with Root Hbs showing a large decrease in both 02 affinity and cooperativity at low pH (Pelster and Randall, 1998). Protons bound to particular groups

44 (Root groups) on the Root Hb stabilize the deoxygenated T(ense)-state to such an extent that complete saturation with 02 is unattainable (Root, 1931; Root and frying, 1943), even at s2Po of up to 140 atmospheres (Scholander and Van Dam, 1954). Physiologically, however, the Root effect has very different implications for gas transport than the Bohr effect. In conjunction with a gas gland and rete, tissues that respectively create and localize an acidosis within capillaries, fishes use these Root Hbs to offload large quantities of 02 to the swimbladder and eye (Brittain, 1987; Peister and Randall,

1998). This 02 multiplication system allows fishes to regulate their swimbladder volume, and thus their buoyancy in the water column, as well as to ensure adequate 02 delivery to the retinal cells despite the avascularization of the fish eye (Bridges et al., 1998; Peister,

2001). However, the Root effect comes with potentially unfavorable consequences.

Certain conditions, including hypoxia and exercise, are capable of producing generalized blood acidoses through the production of 2CO and metabolic acid which, in conjunction with pH-sensitive Root Hbs, may jeopardize 02 uptake at the gill. To compensate, teleost fishes have the ability to regulate RBC 1pH through the use of adrenergically-stimulated

Na7H exchangers on the RBC membrane (I3NHE). These exchangers are activated through the binding of catecholamines released to the blood under stressful conditions, and with the resulting proton extrusion, are capable of keeping red cell 1pH high

(Nikinmaa, 1983; Thomas and Perry, 1992; Nikinmaa and Salama, 1998). In this way, teleosts preserve 02 uptake during generalized blood acidoses despite possessing proton sensitive Root Hbs.

45 The basal actinopterygian fishes: A transitional group in Root effect evolution

The evolutionary relationships of the Root effect and its associated mechanisms have been recently investigated by Berenbrink and colleagues (2005) in a comparative study across a broad range of fishes. Their results indicate that the protective J3NHEdid not evolve until long after the appearance of the Root effect Hb traits (increased proton sensitivity; decreased Hb buffer capacity) and the choroid rete (Berenbrink et al., 2005).

From this, it can be assumed that those intermediate species possessing a Root effect but lacking 3NHE are at risk ofjeopardizing up to 50% of their 02 carrying capacity in the case of a generalized acidosis. These include species within the acipenseriformes

(sturgeons and paddlefish), ginglymods (gars), amiiformes (bowfin), and osteoglossomorphs (the bony-tongues; the basal teleosts). That these species are some of the very few ‘primitive’ fishes to have successfully survived to the present day (Janvier,

2007) is likely a testament to their ability to compensate in such situations. How they are accomplishing this, however, is unresolved, as relatively little is known about the blood properties of this most interesting group (Brauner and Berenbrink, 2007).

We hypothesize that despite a significant Root effect and lack of 3NHE, 02 binding at the gills of these transitional species during a generalized blood acidosis must be maintained. This would only occur if RBC 1pH did not get low enough to activate their Root effects in the general circulation, a likely result of either RBC pH regulation by some means other than f3NHE,or onset pH values of the Root effect in the Hbs of these species that are lower than those brought about by hypoxia or exercise. The previous chapter addressed the former, where it was shown that these species’ Hb buffer capacities are not uncharacteristically high so as to preserve 02 binding through the

46 protection of RBC 1pH (Chapter 2). The latter scenario is addressed in the present study.

Expressed in order from most basal to most derived, the species investigated in this study include: American paddlefish (Polyodon spathula), white sturgeon (Acienser transmontanus), spotted gar (Lepisosteus oculatus), alligator gar (Atractosteus spatula), bowfin (Amia calva), mooneye (Hiodon tergisus), and pirarucu (Arapaima gigas). Not only do these species all belong to the transitional group in the evolution of the Root effect, but they also straddle the first appearance of the choroid rete, a countercurrent network of capillaries that, in conjunction with the Root effect, plays a key role in the generation of high 02 tensions in the eye (Wittenberg and Wittenberg, 1974; Wittenberg and Haedrich, 1974; Bridges et al., 1998). The analysis of Root effect characteristics in species that both lack (paddlefish, white sturgeon, spotted gar, and alligator gar) and possess (bowfin, mooneye, and pirarucu) choroid retia may shed light on its influence on these Hb properties. The objective of this study was to measure the onset pH of the Root effect in the haemolysates of these species in the absence and presence of organic phosphates (GTP). Assuming the Hb properties of these seven species to be representative of their ancestral states (Janvier, 2007; McKenzie et al., 2007), this may allow us to gain insight into how a large Root effect could have evolved in this group of fishes, despite the absence of RBC I3NHE.

Materials and methods

Animal acquisition

White sturgeon (Acipenser transmontanus; 1-2 kg) and bowfin (Amia calva; 0.5- 1.0 kg) were kept in large flow-through outdoor tanks (dechlorinated city water; 2>Po

47 130 torr; 2Pco <0.1 torr; T = 11-17°C; fish density < 25 kg fish per 3m water) at the University of British Columbia, Vancouver, Canada, prior to sampling. Pirarucu

(Arapaima gigas; 1-2 kg) were maintained in outdoor static tanks at the National Institute for Research of the Amazon (INPA). American paddlefish (Polyodon spat hula), spotted gar (Lepisosteus oculatus), alligator gar (Atractosteus spatula) and mooneye (Hiodon tergisus) were all sampled immediately after being caught in their respective natural habitats. Samples for each species consisted of blood from at least three adult fish of either sex, with the exception of spotted gar which consisted of the blood of a single adult.

Haemolysate preparation

Whole blood was drawn from the caudal vein of anaesthetized animals (1 ml of

200 mM benzocaine per litre of water; Sigma El 501) into heparinized syringes, where red cells were then separated by centrifugation and washed three times in cold Cortland’s physiological saline (Wolfe, 1963). The red cells were lysed by addition of two-times volume cold deionized water and subsequent freezing, and cell debris was removed by ten minutes of chilled (4°C) centrifugation at 14 000 r.p.m. (Thermo Electron

Corporation 21000R, Waltham, MA, USA). The haemolysates were purified by removing cell solutes and organic phosphates by repeated passage through mixed-bed ion exchange columns (Amberlite RN-i 50 mixed bed ion exchange resin), and were then divided into 0.5 ml aliquots and frozen at -80°C until use. Upon thawing, Hb concentration was determined after conversion to cyanomethaemoglobin using a millimolar extinction coefficient of 11 at 540 rim, and methaemoglobin content was

48 assessed on identical sub-samples using the spectrophotometric method of Benesch et al.

(1973). Samples where methaemoglobin exceeded 10% of total Hb were discarded.

Root effect quantfication

The magnitude of the Root effect was determined by measuring oxygen saturation of Hb spectrophotometrically at atmospheric 2Po (157 mmHg) in Tris buffers (50 mmol r

‘;Trizma base, Sigma T6066) ranging in pH from 5.5 to 8.5. Air-equilibrated concentrated haemolysates, diluted to a final concentration of 160 p.molr’ 4Hb and 0.1 mol 1F KC1,were then mixed with the buffers in a 1 ml cuvette. This procedure was conducted for haemolysates in both the absence and presence of GTP (GTP4:Hb ratio of 3:1; guanosine 5’-triphosphate sodium salt hydrate, Sigma G8877). Absorption at wavelengths of 540, 560 and 576 nm were measured and recorded using a Shimadzu

(Columbia, MD, USA) UV- 160 spectrophotometer, and were used to calculate percent

Hb 02 saturation according to Benesch et al. (1973) as described below.

Data analysis

Percent Hb 02 saturation at each pH was calculated using the following equations according to Benesch et al. (1973),

[Oxy Hbj = (1.4747 A576 — 0.6820 A560 — 0.5329 )A540 (1) [Deoxy Hb] = (1.4749 A560 + 0.2141 A576 — 1.1042 )A540 (2)

49 where A is the optical density at the peak absorption wavelength for oxygenated Hb (540 and 576 nm) and for deoxygenated Hb (560 nm). Equations (1) and (2) were then added together to give total Hb in solution, by which the oxygenated Hb concentration was divided to yield the percent oxygenation status of the haemolysate at each pH.

The pH of Root effect onset was determined for four different degrees of Hb 02 desaturation: 5%, 10%, 15% and 20%. These values were subtracted from the average fully oxygenated percentage (between pH 7.5 and 8.5, where no Root effect is present), and the pH at the respective Hb 02 saturation percentages was determined using the sigmoidal line of best fit generated for each sample. A sample size of three to four was used for each species.

Statistical analysis

Statistical t-tests were performed when comparing the mean values of both maximal Root effect and Root effect onset pH values among the rete-bearing species and the non-rete species, as well as each species’ Root effect onset pH values in the presence and absence of GTP. Data were transformed in the few instances where equal variance tests were not passed.

Statistical tests were done using SigmaStat 3.0.

Results

Percent Hb 02 saturation as a function of pH for the stripped haemolysates in aerated Tris buffer of American paddlefish, white sturgeon, spotted gar, alligator gar, bowfin, mooneye and pirarucu is shown in Fig. 3.1. The Hb 02 saturation for each

50 species decreased with a decrease in pH, attributable to the Root effect. The magnitude of the Root effect varied among the species, with the Hb of the three rete-bearing species

(bowfin, mooneye, and pirarucu) showing a significantly greater mean level of Hb 02 desaturation than that of the four more plesiomorphic species (paddlefish, white sturgeon, spotted gar and alligator gar; Fig. 3.1; without GTP: t = -13.441; d.f. = 19; P <0.001; with GTP: t = -9.690; d.f. 18; P < 0.001).

Both Root effect magnitude and onset pH were elevated in the presence of allosterically-modifying GTP (Figs. 3.1, 3.2), with the onset pH values of the rete-bearing species in particular increasing significantly over those of their stripped haemolysates [at

10% desaturation: bowfin (t -3.561; d.f= 4; P = 0.024); mooneye (t = -7.211; d.f. = 4; P

0.002); pirarucu (t = -2.990; d.f. = 4; P = 0.04)]. Tnboth cases, there was a general dichotomy between the four more plesiomorphic species and the three rete-bearing species on the Root effect traces (Fig. 3.1).

The pH at which four different levels of Hb 02 desaturation (5, 10, 15 and 20% of total blood haemoglobin) were observed for each species’ stripped haemolysates in the presence and absence of GTP is shown in Fig. 3.2. Due to relatively small Root effects, the pH values at which 20% (and even 15% in the case of paddlefish) of total blood Hb was desaturated were indeterminable for three of the four more plesiomorphic species (a single spotted gar sample excepted). The relatively modest desaturation levels that were reached in these species were observed only at very low pH values, even for a 5% decrease in oxygen carrying capacity (Fig. 3.2). The three rete-bearing species, however, all exhibited large Root effects, and the mean pH values at which each of 5, 10, and 15% desaturation (20% not achieved in non-rete species) were observed at were significantly

51 higher than those of the four more plesiomorphic species (e.g. 5% desaturation without

GTP: t= -6.382; d.f. = 13; P <0.001; 5% desaturation with GTP: t= -6.240; d.f. = 21; P

<0.001). All seven species’ Root effect onset pH values were considerably lower than

the lowest average RBC pH that might be expected (based upon literature values) during

hypoxia and exercise. Furthermore, the variation in pH onset between the four different

levels of desaturation was less for each of the rete-bearing species than for the more

plesiomorphic species (Fig. 3.2), evident by the shallower slopes of their Root effect

curves (Fig. 3.1).

Root effect onset pH (5% Rb 02 desaturation) was positively correlated with the maximal Root effect of that species 2(r = 0.700; d.f. = 21; P < 0.001; Fig. 3.3). Again the

dichotomy of the four more plesiomorphic species and the three rete-bearing species is

evident, with the former segregated on the lower left of the graph signifying a low Root effect maximum and a low pH of onset, and the latter segregated on the upper right portion of the graph signifying a high Root effect maximum and a high pH of onset (Fig.

3.3).

Discussion

It has been proposed that the Root effect evolved in the basal actinopterygian lineage of fishes in the absence of I3NHEactivity (Berenbrink et a!., 2005), a situation that comes with the potential of losing a significant proportion of the blood’s 02 carrying capacity during times of generalized acidoses. In light of the recent finding that the proton-binding abilities of these species’ Hbs (intrinsic buffering and oxylabile Haldane effect) do not likely compensate for a lack of f3N}IEactivity (Chapter 2), the objective of

52 the present study was to gain insight into how a sizeable Root effect could evolve in the absence of I3NHE.The findings indicate that the onset pH of the Root effect is below even the lowest predicted pH values that would be elicited in the general circulation of these fishes following severe exercise or exposure to hypoxia.

Influence of GTP

Adenosine triphosphate (ATP) and guanosine triphosphate (GTP) are the predominant organic phosphates in fish RBCs and have been shown to act as major cofactors in Hb function by reducing Hb 02 affinity (Weber et al., 1975; Peister and

Weber, 1990; Brauner and Weber, 1998; Val, 2000). Of these, GTP exerts the stronger effect on Hb 02 saturation due to an additional stabilizing bond within the T-state of the

Hb molecule (Peister and Weber, 1990; Val, 2000). For this reason, all of the experiments in the present study were performed in both the presence and absence of saturating levels of GTP (3:1 4).GTP:Hb As well as reducing Hb 02 affinity, the allosteric effects of GTP alter the molecular microenvironments of the Root groups of Hb in such a way that increases their pK values (Pelster and Weber, 1990; Jensen et al.,

1998). These allosteric effects become manifest as increases in both Root effect magnitude and onset pH (Figs. 3.1, 3.2), although the degree to which GTP influences these Hb characteristics is species-specific. For instance, the Hbs of carp and eel each show a Root effect of 40% in the presence of saturating levels of organic phosphates, but no Root effect in their absence (Peister and Weber, 1990). Conversely, trout Hbs display large Root effects of 40% in the absence of organic phosphates, increasing to 55% when they are added to the haemolysate (Peister and Weber, 1990). All of the experiments in

53 the present study were performed in both the presence and absence of saturating levels of GTP (3:1 4),GTP:Hb where it was shown that the influence of GTP on these Hb characteristics varied interspecifically. Root effect magnitude was generally increased upon GTP binding, ranging from very slight (e.g., alligator gar; pirarucu; Fig. 3.1) to up to a 7% increase in overall Hb 02 desaturation (sturgeon; Fig. 3.1), while onset pH was increased in all species by up to 0.5 pH units (mooneye; Figs. 3.1, 3.2).

The remainder of the discussion will focus primarily on those experiments done in the presence of GTP, as we believe these to be more representative of the in vivo conditions of these fishes.

Root effect onset pH

Although believed to be a trait almost exclusive to the haemoglobins (Hbs) of teleosts (Pelster and Randall, 1998), this study supports the findings of Berebrink and colleagues’ (2005) that the Root effect is also a characteristic of the Hbs of more plesiomorphic fishes. Each of the seven species studied here possesses Hbs that display a

Root effect (Fig. 3.3), ranging from approximately 7% of total blood oxygen carrying capacity in paddlefish to 40% in mooneye. Variation also exists in the onset pH values of these species’ Root effects (Figs. 3.1, 3.2). To our knowledge, few, if any, studies have looked at the mechanism underlying the variation in Root effect onset pH. However, it is likely that this pH value is a function of the pK values of the particular amino acid residues responsible for binding the protons that bring about the R to T conformational change in the Hb. As it has been shown that the mechanistic basis of the Root effect varies interspecifically, that is, different amino acid residues on different Hbs have been

54 shown to be involved in the Root effect-associated conformational changes (Perutz and

Brunori, 1982; Nagai et al., 1985; Mylvaganam et aL, 1996; Jensen et a!., 1998;

Tsuneshige et al., 2002), one may also expect the pK values of these different groups to

vary. This would be manifest as interspecific differences in Root effect onset pH, as are

shown in Fig. 3.2. The onset pH value for each species was determined at four different

levels of Hb 02 desaturation (5, 10, 15, and 20% of total Hb), although not all seven

species had Root effects capable of achieving 20% desaturation. Of the four plesiomorphic species, only the haemolysates of spotted gar became 20% desaturated (a

single sample), while those of paddlefish, white sturgeon, and alligator gar showed at most 10 to 15% desaturation (Fig 2). Moreover, the pH values at which these levels of

desaturation were achieved were very low relative to in vivo conditions in the general

circulation, all of them occurring well below pH 6.3 (5% desaturation in white sturgeon

excepted; Fig. 3.2). Conversely, the haemolysates of the rete-bearing bowfin, mooneye,

and pirarucu all displayed Root effects larger than 20%, and the pH values at which the

four desaturation levels were achieved were in a more physiologically-relevant range.

Five percent desaturation occurred in these three species between pHs 6.8 and 7.0, while

20% was achieved between pHs 6.4 and 6.7 (Fig. 3.2).

Each of the seven species’ Root effect onset values were quite low compared to

those that would be expected during a generalized acidosis. Exercise (acute and

exhaustive) and hypoxia are ways of generating severe acidoses of the blood. However,

little information on the effect of these challenges on blood-gas transport exists for these

species (Brauner and Berenbrink, 2007). A survey of the responses to exhaustive exercise

in a broad range of fishes (Squalus acanthias; Lepisosteus osseus; Amia calva;

55

species 2008).

more

Animals

from time

lower uptake

gave relatively

which

compromised

studied

intracellular those

(Holeton

3.2).

translate

2003).

(Milligan

Oncorhynchus

1986;

when

rise

the

robust

pH

Similarly,

in

0.469

again

throughout

An

There

have

species

Assuming

02,

from

air;

to

to

and

values

large

and

the

example

RBC

the

at

and pH

in

been are

others

Randall,

are

in

this

Wood,

earth’s

dealing

white

mykiss)

Root

(Fig.

species

values hypoxia

considerably

that

these

much

pH 1

conditions

shown

era

a

evolutionary

abandoned

of

APHi/ApHe

would

effects

3.2).

values sturgeon,

1986;

atmosphere

dealt

species

with

the

more

1967;

between

studied

indicates

acidifies

to

latter

It

with

their

result

be

Burleson

and

which between

acidic

therefore

Butler

higher

following

among

their

Baker

here,

time,

value

is

7.15

no

these

harsh

that

in

and the

found

may

(Algeo

3NHE

and

aquatic

even

than

date

blood

and

7.1

however.

blood

et

et

of

the

water

challenges

appears

environments

have

Taylor,

al.,

al.,

exhaustive

0.47

in

and

back 7.3

the

most

a

activity.

et

56

the

pHe

5%

habitats in

1998;

to

systems

al.,

(Heming

Root

posed

7.3.

(0.471

press),

pH

as

over

acipenseriformes,

CO 2

Root

may

Stem

1975;

2001;

though

in

Gonzalez

7.4-7.7 These

effect

larger

exercise

300

different

tolerant

altogether;

in

be

were

effect

(Clack,

these

actinopterygian

Randall

et

Berner,

rainbow

reduced

million

are

02

onset

al.,

in

threats

significantly

in

extracellular

et

uptake

of

considerably

a

or

1986;

2007;

ways

the

comparable

aL,

et

pH

2006;

fishes

hypoxia,

while

trout,

years

to

al.,

whose

to

studied

2001;

values between

—

Baker

Brauner

is

these

some

1992),

Clack,

(Crocker

still not

species,

Heming

(Janvier,

higher

pH

extant

Richards

despite

species

in

likely

et

higher

species’

others

range

secured

translating

7.3

and

values

al.,

any

2007).

which

in

et

sturgeon

and

2007)

and

in

Baker,

of

of

became

than

C0 2 ,

al.,

(Fig.

et

press),

02

the

02

Cech,

fishes

7.7

al.,

to

to a 1998; Baker et al., in press). In conditions of hypercarbia that may be similar to those of its proposed ancestral habitats, white sturgeon has been shown to survive despite a blood acidosis resulting in a RBC 1pH of 6.8 (Baker et al., in press; Brauner and Baker, 2008). Taken as the extreme example of a potential generalized blood acidosis, 1pH 6.8 would not be capable of activating the Root effect in the four ancestral species (Fig. 3.2b).

However, it may be low enough to reduce blood oxygen carrying capacity by 10% in the three rete-bearing species (15% in the case of mooneye; Fig. 3.2b). Whether this degree of desaturation would pose a threat to these animals is not known. However, 3NHE first appeared in the elopomorphs, the order of fishes immediately following the osteoglossomorphs on the phylogenetic spectrum (Berenbrink et al., 2005; Berenbrink,

2007). Taken together, the potential for incomplete Hb 02 saturation under potentially hypercarbic ancestral conditions may have potentially provided the necessary selective pressure for RBC f3NHEevolution.

Root effect without retia

There is a dichotomizing trend in the data set between those species with a rete and those lacking one. These significant differences in both Root effect magnitude and onset pH suggest a potential role of the rete in manifesting a Root effect that is operational in vivo. All of the non-rete species analyzed show relatively small Root effects elicited at pH values that appear too low to be seen in vivo, while the rete-bearing species show significantly larger Root effects elicited at pH values that may occur at the eye in the presence of a gas gland and rete. This suggests that although the non-rete species do technically have a Root effect, it is not one that is likely operational at pH

57 values seen in vivo. Its presence within, and benefit to, the organism is therefore questionable.

There is a highly significant correlation between Root effect and Bohr effect magnitude among the ray-finned fishes starting around the acipenserformes (Berenbrink et al., 2005). It has been suggested that this correlation stems from the Root effect being the extreme byproduct of the mechanism by which these early ray-finned fishes activated their Bohr effects (Berenbrink et a!., 2005). The Root effect may therefore not have evolved specifically for enhanced 02 delivery itself, but rather as the extreme manifestation of a selection-favoured Bohr effect in the early ray-finned fishes. When the choroid rete eventually evolved at a particularly beneficial site (i.e., the avascular eye) in the last common ancestor of bowfin and the teleosts, the Root effect could then be used and selected for in and of itself, as the necessary activating acidoses were now capable of being generated by the organism. The findings of the present study strongly support this idea. The significant differences in both Root effect magnitude and onset pH between species with and without retia (Figs. 3.1, 3.2) suggest the rete as playing an important role in the optimization of the Root effect. Furthermore, the extremely low onset pH values of those species lacking retia (Figs. 3.1, 3.2) suggest the rete as a necessary mechanism in the utilization of the Root effect within the organism.

The degree to which a species is capable of offloading 02 to the tissue via the

Bohr effect is determined by the product of its Bohr coefficient and the areterio-venous pH change (Lapennas, 1983). As the Bohr effect is directly proportional to the Haldane effect (proton binding of Hb upon deoxygenation), the larger the Bohr coefficient, the smaller the arterio-venous pH change. In fact, it has been shown that the optimal Bohr

58 coefficient for 02 delivery to the aerobically respiring tissue is between 0.35 and 0.5, while coefficients greater than that are ideal for blood buffering/CO transport (Lapennas,

2 2) revealed the 1983). The Hb titrations performed in the previous study (Chapter Bohr coefficients of these seven basal actinopterygian species. Those species lacking retia had maximal Bohr coefficients between 0.35 and 0.5 (sturgeon excepted at 0.65), suggesting a Hb adaptation for 02 delivery via the Bohr effect, while those species with retia displayed maximal Bohr coefficients at or around unity, suggesting Hb adaptation for proton buffering and 2CO transport. This is sensible given the fact that the non-rete species had relatively high Hb buffer values, while those of the rete-bearing species were generally lower (Chapter 2). Furthermore, although a large Bohr coefficient may not be ideal for 02 delivery under regular aerobic conditions, it allows for a greater right shift of the 02 equilibrium curve if provided with a sufficient acidosis (Lapennas, 1983). In bowfin, mooneye, and pirarucu, such an acidosis is produced by the choroid rete at the eye, and thus their large Bohr coefficients allow for larger quantities of 02 to be delivered to the retinal cells.

There is a slight exception to the rete/non-rete dichotomy in sturgeon. Sturgeon

Hbs display buffer values markedly lower than those of the other non-rete species

(Chapter 2), a maximal Bohr coefficient of 0.65 (Chapter 2), and a 5% Root effect onset pH not significantly different than that of bowfin (pH 6.63; Fig. 3.2). These together suggest that sturgeon may in fact be capable of using their modest Root effect to some degree. Although there is no known gas glandlrete analogue at the eye or swimbladder of sturgeon, pH values approaching 6.6 may still be generated in capillaries where large quantities of 2CO could potentially have profound influence on the minute blood

59 volume’s pH, and thus, its capacity to deliver 02. Once in the efferent venous supply, this acidified blood would be quickly diluted so as to not negatively influence pH in the general circulation. And as roughly 98% of the 02 in the blood is Hb-bound, even a modest 5% decrease in Hb 02 saturation could result in large increases in blood .2Po Although no evidence of Root effect utilization is known to exist for sturgeon, it would be interesting to analyze the 02 tension of certain tissues (namely retinal tissue) to see if they are in fact using their modest Root effect to drive s2Po at these sites above those of arterial blood.

Conclusions

It was shown that the Root effect is indeed a feature of the Hbs of the basal actinopterygian fishes, in agreement with the recent findings of Berenbrink et al. (2005).

However, the in vivo relevance of these Root effects varies interspecifically in a manner that appears dependent on the presence or absence of the counter-current rete.

Paddlefish, white sturgeon, spotted gar, and alligator gar all lack retia, and although they possess Hbs that display Root effects, they are not likely capable of utilizing them under in vivo conditions by reason of low onset pH values and a lack of any known tissue capable of generating the appropriate acidosis. Bowfin, mooneye, and pirarucu all possess choroid retia, however, and their higher onset pH values indicate they are likely operational in vivo at the eye. Taken together, the fact that the onset pH values of all seven species’ Root effects are lower than even the lowest pH values likely to occur following exposure to hypoxia or exercise renders it unlikely that the presence of the

Root effect in the absence of I3NHEwould pose a threat to these species’ 02 uptake. This

60 would allow for the positive selection of the Root effect as a mechanism of 02 delivery to override any negative selection for it brought about by its potentially adverse effects on

02 uptake. In this way, the low onset pH values of the species within this transitional phase of Root effect evolution may explain how this important Hb characteristic was capable of becoming almost ubiquitous among fishes thereafter, despite a lack of protective RBC I3NHEactivity.

61 120

100

80

60

40

20 A ® E 0

120

100 0..__ 0 0 80

60 a— . 0 40

20 B ® F 0

120

100

80

60 oo

40

20 C ® G 0 5 6 7 8 9 120 pH

100 0 0

80

60

40

20 D 0 5 6 7 8 9 pH

Figure 3.1. The magnitude of the Root effect at different pHs in the presence (•) and absence (S) of 3:1 GTP4:Hb in American paddlefish (A), white sturgeon (B), spotted gar (C), alligator gar (D), bowfin (E), mooneye (F), and pirarucu (G). Values were determined spectrophotometrically on heamolysates at a ]4[Hb of 0.16 mmol 11 and a [KC1]of 0.1 mol r’. The presence of a choroid rete within the species is indicated by ®.

62 7.5 A

5% Desaturation 10% Desaturation 15% Desaturation 7.0a EZD 20% Desaturation -I-. I) Cl) C o 6.5 .4-. C)

ci) 6.0

5.5

7.5

6.0

5.5

Paddlefish Sturgeon Spotted gar Alligator gar Bowfin Mooneye Pirarucu ® ®

Figure 3.2. Root effect onset pH values in haemolysates of seven basal actinopterygian fishes in the absence (A) and presence (B) of saturating GTP (3:1 GTP:Hb Root effect onset was determined for 5, 10, 15, and 20% Rb 02 desaturation.4). Data are mean values ± SEM. Bars lacking SEM represent the lone sample to reach the respective desaturation percentage from that group. The presence of a choroid rete within the species is indicated by ®. Horizontal grey bars are indicative of predicted lowest RBC 1pH range that would occur during hypoxia or exercise, based upon literature values (see discussion for further details). Average onset value for the three rete-bearing species is significantly greater than that of the four non-rete species (P < 0.001; see Results section for further details).

63 50 0C . . . .

Cl) 0 0c’J 30

•Z5 . 20 . . . E 10 . >< (‘3 . . .

0• I I I 5.5 6.0 6.5 7.0 7.5 Root effect onset (pH)

Figure 3.3. Maximal Root effect as a function of Root effect onset pH (5% Root effect) for each haemolysate sample of American paddlefish, white sturgeon, spotted gar, alligator gar, bowfin, mooneye, and pirarucu. Haemolysates, in the presence of GTP (3:1 GTP:Hb were brought to a final ]4[Hb of 0.16 mmol r’ and [KC1Jof 0.14),mol f’. Individual points represent individual measurements. r2 = 0.700; d.f. = 21; P <0.001.

64 -CHAPTER SUMMARY-

1. The Root effect onset pH values of all seven species were shown to be lower than

even the lowest pH values likely to occur in the RBC following exposure to

hypoxia or exercise. The presence of a Root effect in the absence of I3NHE

therefore does not likely pose a threat to 02 uptake.

2. Assuming the Hb properties of these seven species to be representative of their

respective ancestral states, a significant increase in Root effect magnitude and

onset pH value appears to have occurred in the last common ancestor of the

amiiformes and the teleosts. This is correlated with the first appearance of the

choroid rete, as well as a punctuated increase and decrease in the Haldane effect

and Hb buffer value, respectively (Chapter 2).

3. The low onset pH values of the plesiomorphic non-rete species, the optimality of

their Bohr coefficients for 02 delivery, and the positive correlation of the Bohr

effect and Root effect together suggest that the ancestral state of the Root effect

may have been nothing more than the extreme manifestation of their Bohr effects,

only becoming of physiological relevance itself once a tissue capable of

producing the required acidosis to elicit it (i.e., the rete) evolved at a particularly

beneficial site (i.e., the avascularized retina).

65 References

Algeo, T. J., Scheckler, S. E., and Maynard, J. G. (2001). Effects of the middle to late Devonian spread of vascular land plants on weathering regimes, marine biotas, and global climate. In Plants Invade the Land. Evolutionary and Environmental Perspectives (eds P. G. Gensel and D. Edwards), pp. 213-236. New York: Columbia University Press.

Baker, D.W., Morgan, J.D., Wilson, J.M., Matey, V., Huynh, K.T. and Brauner, C.J. (2008). Complete intracellular pH protection during extracellular pH depression is associated with hypercarbia tolerance in white sturgeon, Aczpenser transmontanus. Amer. I of Physiol. (Submitted).

Benesch, R. E., Benesch, R. and Yung, S. (1973). Equations for spectrophotometric analysis of hemoglobin mixtures. Anal. Biochem. 55, 245-248.

Berenbrink, M. (2006). Evolution of vertebrate haemoglobins: Histidine side chains, specific buffer value and Bohr effect. Respir. Physiol. & Neurobiol. 154, 165-184.

Berenbrink, M. (2007). Historical reconstructions of evolving physiological complexity: 0-2 secretion in the eye and swimbladder of fishes. I Exp. Biol. 210, 1641-1652.

Berenbrink, M., Koldkjaer, P., Kepp, 0. and Cossins, A. R. (2005). Evolution of oxygen secretion in fishes and the emergence of a complex physiological system. Science 307, 1752-1757.

Berner, R. A. (2006). GEOCARBSULF: a combined model for Phanerozoic atmospheric 02 and .2CO Geochim. Cosmochim. Ac. 70, 5653-5664. Bonaventura, C., Crumbliss, A. L. and Weber, R. E. (2004). New insights into the proton-dependent oxygen affinity of Root effect haemoglobins. Acta Physiol. Scand. 182, 245-258.

Brauner, C. J. and Weber, R. E. (1998). Hydrogen ion titrations of the anodic and cathodic haemoglobin components of the European eel Anguilla anguilla. I Exp. Biol. 201, 2507-2514.

Brauner, C.J. and Berenbrink, M. (2007). Gas Exchange. In Fish Physiology: Primitive Fishes, vol 26 (eds D. J. McKenzie, A. P. Farrell and C. J. Brauner), pp. 213- 282. New York: Elsevier.

Brauner, C.J. and Baker, D. (2008). Patterns of acid-base regulation infish. In: Cardio-Respiratory Control in Vertebrates: Comparative and Evolutionary Aspects. (eds M.L. Glass and S.C. Wood), In Press. Berlin: Springer-Verlag.

Bridges, C. R., Berenbrink, M., Muller, R. and Waser, W. (1998). Physiology and biochemistry of the pseudobranch: An unanswered question? Comp. Biochem. Physiol.,

66 A: Mol. Integr. Physiol. 119, 67-77.

Brittain, T. (1987). The Root effect. Comp. Biochem. Physiol. B: Biochem. Mol. Biol. 86, 473-481.

Burleson M., Shipman, B. and Smatresk, N. (1998). Ventilation and acid-base recovery following exhausting activity in an air-breathing fish. I Exp. Rio!. 201, 1359- 1368.

Butler, P. J. and Taylor, E. W. (1975). The effect of progressive hypoxia on respiration in the dogfish (Scyliorhinus canicula) at different seasonal temperatures. I exp. Biol. 63, 117—130.

Clack, J. A. (2007). Devonian climate change, breathing, and the origin of the tetrapod stem group. Integ. Comp. Biol. 47, 510-523.

Crocker, C. E. and Cech, J. J. Jr (1998). Effects of hypercapnia on blood-gas and acid- base status in the white sturgeon, Aczenser transmontanus. I Comp. Physiol. B. 168, 50—60.

Dickerson, R. E. (1971). The structures of cytochrome c and the rates of molecular evolution. I Mol. Evol. 1, 26-45.

Gonzalez, R. J., Milligan, L., Pagnotta, A., and McDonald, D. G. (2001). Effect of air breathing on acid-base and ion regulation after exhaustive exercise and during low pH exposure in the bown, Amia calva. Phusiol. Biochem. Zoo!. 74, 502-509.

Heming, T. A., Randall, D. J., Boutilier, R. G., Iwama, G. K. and Primmett, D. (1986 Ionic equilibria in red blood cells of rainbow trout (salmo gairdneri) C1, 3HC0 and H . Respir. Physiol. 65, 223-234.

Holeton, G. F. and Randall, D. J. (1967). Effect of hypoxia upon partial pressure of gases in blood and water afferent and efferent to gills of rainbow trout. I Exp. Rio!. 46, 317-327.

Janvier, P. (2007). Living Primitive Fishes and Fishes from Deep Time. In Fish Physiology: Primitive Fishes, vol. 26 (eds D. J. McKenzie, A. P. Farrell and C. J. Brauner), pp. 2-53. New York: Elsevier.

Jensen, F. B., Fago, A. and Weber, R. E. (1998). Red blood cell physiology and biochemistry. In Fish Physiology: Fish Respiration, vol. 17 (eds S. F. Perry and B. L. Tufts), pp. 1-40. New York: Elsevier.

Lapennas, G. N. (1983). The magnitude of the Bohr coefficient: optimal for oxygen delivery. Respir. Phyiol. 54, 161-172.

67 McKenzie, D. J., Farrell, A. P. and Brauner, C. J. (2007). Fish Physiology : Primitive Fishes, vol 26 (eds D. J. Mckenzie, A. P. Farrell and C. J. Brauner), pp xi-xiii. New

York : Elsevier.

Milligan, C. L. and Wood, C. M. (1986). Intracellular and extracellular acid-base status and H exchange with the environment after exhaustive exercise in the rainbow trout. I Exp. Biol. 123, 93-121.

Mylvaganam, S. E., Bonaventura, C., Bonaventura, J. and Getzoff, E. D. (1996). Structural basis for the Root effect in haemoglobin. Nat. Struct. Biol. 3, 275-283.

Nagai, K., Perutz, M. F. and Poyart, C. (1985). Oxygen binding properties of human mutant hemoglobins synthesized in Escherichia coli. Proc. Nati. Acad. Sci. USA82, 7252-7255.

Nikinmaa, M. (1983). Adrenergic regulation of hemoglobin oxygen affinity in rainbow trout red cells. I Comp. Physiol. 152, 67-72.

Nikinmaa, M. and Salama, B. (1998). Oxygen transport infish. In Fish Physiology: Fish Respiration, vol. 17 (eds S. F. Perry and B. L. Tufts), pp. 141-184. New York: Academic Press.

Peister, B. (2001). The generation of hyperbaric oxygen tensions in fish. News Physiol. Sci. 16, 287-291.

Peister, B. and Weber, R. E. (1990). Influence of organic phosphates on the Root effect of multiple fish hemoglobins. I Exp. Biol. 149, 425-437.

Peister, B. and Randall, D. J. (1998). Physiology of the Root effect. In Fish Physiology: Fish Respiration, vol. 17 (eds S. F. Perry and B. L. Tufts), pp. 141-184. New York: Elsevier.

Perutz, M. F. and Brunori, M. (1982). Stereochemistry of cooperative effects in fish and amphibian hemoglobins. Nature 299, 421-426.

Powers, D. A. (1980). Molecular ecology of teleost fish hemoglobin: strategies for adapting to changing environments. Am. Zool. 20, 139-162.

Randall, D. J., McKenzie, D. J., Abrami, G., Bondiolotti, G. P., Natiello, F., Bronzi, P., Bolis, L. and Agradi, E. (1992). Effects of diet on response to hypoxia in sturgeon (Acipenser naccarii). .1 Exp. Biol. 170,113 -125.

Richards, J. G., Heigenhauser, G. J. F., and Wood, C. M. (2003). Exercise and recovery metabolism in the pacific spiny dogfish (Squalus acanthias) I Comp. Physiol. B. 173, 463-474.

68 Root, R. W. (1931). The respiratory function of the blood of marine fishes. Biol. Bull. 61, 427-456.

Root, R. W. and Irving, L. (1943). The effect of carbon dioxide and lactic acid on the oxygen-combining power of whole and hemolyzed blood of the marine fish Tautogo onitis (Linn). Biol. Bull. 84, 207-2 12.

Scholander, P. F. and Van Dam, L. (1954). Secretion of gases against high pressures in the swim bladder of deep sea fishes. I. Oxygen dissociation in blood. Biol. Bull. 107, 247-259.

Thomas, S. and Perry, S. F. (1992). Control and consequences of adrenergic activation of red blood cell Na/H exchange on blood oxygen and carbon dioxide transport in fish. I Exp. Zool. 263, 160-175.

Tsuneshige, A., Park, S. and Yonetani, T. (2002). Heterotropic effectors control the hemoglobin function by interacting with its T and R states - a new view on the principle of allostery. Biophys. Chem. 98, 49-63.

Val, A. L. (2000). Organic phosphates in the red blood cells of fish. Comp. Biochem. Physiol., A: Mol. Integr. Physiol. 125, 417-435.

Weber, R. E. (1990). Functional significance and structural basis of multiple hemoglobins with special reference to ectothermic vertebrates. In Animal Nutrition and Transport Processes. 2. Transport Respiration and Excretion: Comparative and Environmental Aspects (ed. J. P. Truchot and B. Lahlou), pp. 58-75. Karger, Basel.

Weber, R. E. (2000). Adaptations for oxygen transport: lessons from fish hemoglobins. In: Hemoglobin Function in Vertebrates, Molecular Adaptation in Extreme and Temperate Environments. (eds F. Di Prisco, B. Giardina and R. E. Weber), pp. 23-37. Milano: Springer-Verlag.

Weber, R. E., Lykkeboe, G. and Johansen, K. (1975). Biochemical aspects of adaptation of hemoglobin oxygen affinity of eels to hypoxia. Lfe Sci. 17, 1345-1349.

Wittenberg, J. B. and Haedrich, R. L. (1974). Choroid rete mirabile of fish eye .2. Distribution and relation to pseudobranch and to swimbladder rete mirabile. Biol. Bull. 146, 137-156.

Wittenberg, J. B. and Wittenberg, B.A. (1974). Choroid rete mirabile of fish eye .1. Oxygen secretion and structure - comparison with swimbladder rete mirabile. Biol. Bull. 146, 116-136.

Wolfe, K. (1963). Physiological salines of fresh water teleosts. Frog. Fish Cult. 25, 135- 140.

69 — CHAPTER 4: GENERAL DISCUSSION —

The Hb system of fish is a curious entity by relative standards. Within it, there exists a degree of functional variation that is unseen in the Rb systems of most, if not all, other vertebrate groups. This is made manifest in a number of ways — Rb 02 affinity is high in some fishes and low in others; Bohr/Raldane effects are large in some fishes and virtually non-existent in others; Hb multiplicity results in up to 15 isoforms per individual in some fishes, while others lack Rb altogether; and present in some fishes and not in others is a Rb-based 02 multiplication system (the Root effect) that is believed to have played a key role in the explosive radiation of the teleosts, a group of fishes which accounts for more than half of the world’s extant vertebrate species (Weber, 2000; Baillie et al., 2004; Bonaventura et al., 2004; Berenbrink et al., 2005; Berenbrink, 2007). This assortment of Rb traits has contributed to the successful survival of fishes in habitats of all sorts, from hypoxic to hyperoxic, acidic to basic, and everything in between. And what is more, thanks to a robust fish phylogeny and increasing knowledge of the Hbs of many species within it, we can start to gain an understanding of how some of these diverse Rb characteristics arose.

The objective of this thesis was to determine the Rb proton-binding properties of species within the basal actinopterygian lineage of fishes, including their Hb buffer capacities, Raldane effects, and Root effect characteristics. The pertinence of this question lies in the fact that many of the aforementioned variations in Rb function are pH-dependent, in that they are the result of conformational changes to Rb that come about with proton-binding. This is of specific relevance to the Bohr/Raldane effects, the

Root effect, and their respective influences on Hb 02 affinity. Furthermore, it is among

70 the basal actinopterygian fishes that the Root effect is believed to have evolved, eventually coming to play a significant role in a respiratory physiological strategy that is unique to the teleosts (Peister and Randall, 1998; Berenbrink et al., 2005). By analyzing the characteristics of these species’ Hb proton-binding properties, light can be shed on the nature of the evolution of these Hb traits.

Synthesis

The Root effect is one component of a multi-faceted physiological system used to deliver large quantities of 02 to the eye and swimbladder of fishes. This 02 delivery is enabled through the cohesive functioning of these proton-sensitive Root Hbs with an acid-producing gas gland, a countercurrent rete, a low fib buffer capacity, and protective

I3NHEactivity. As such, it is a fascinating model by which to help understand the evolution of physiological systems. The lone study that has addressed this question suggests an evolutionary sequence that is rational on a number of levels (Berenbrink et al., 2005). However, it begs a further question insomuch as those species in which the

Root effect and its related processes originally evolved appear to be at risk of a diminished 02 carrying capacity during generalized blood acidoses due to a lack of

3NHE activity. These species include those of the basal actinopterygian lineage, the fishes situated intermediate to the more plesiomorphic elasmobranchs and the more derived teleosts on the fish phylogeny. There are 36 non-teleost members of this group found living in the world today (Nelson, 1994), and while the vast majority of their contemporaries have long since gone extinct (Janvier, 2007), these 36 species have been able to successfully adapt to their ever-changing environments and squeak through the

71 sieve that is natural selection. Their survival is likely, in part, a testament to their adaptability, and certainly not excluded from this list of “dealt-with” selective pressures is their ability to acquire 02 amidst generalized acidoses of the blood, as the extant members are known to deal with instances of exercise, hypoxia, and hypercarbia in their natural environments (Nelson, 1994; Brauner and Berenbrink, 2007; Ilves and Randall,

2007). I hypothesized that this would result from RBC 1pH remaining higher than the pH required to trigger the Root effect in these species during generalized blood acidoses.

This could occur as a result of RBC 1pH protection by some means other than f3NHE

(likely Hb, for reasons discussed in Chapter 2), or Root effects in these species with onset pH values lower than those produced by the generalized blood acidoses resulting from exercise, hypoxia, or hypercarbia. The results yielded by the Hb titration experiments reported in Chapter 2 suggest the Hb proton-binding properties (intrinsic Hb buffering and oxylabile Haldane effects) of seven species within the basal actinopterygian lineage as not likely playing substantial roles in protecting the intracellular pH of the RBC.

However, the results of the Hb 02 saturation experiments reported in Chapter 3 suggest

Root effect onset pH values in these same species as being lower than those likely to be produced by generalized blood acidoses brought on by exercise, hypoxia, or hypercarbia

(Holeton and Randall, 1967; Butler and Taylor, 1975; Heming et al., 1986; Milligan and

Wood, 1986; Randall et al., 1992; Burleson et a!., 1998; Gonzalez et al., 2001; Richards et al., 2003; Baker et al., in press; Brauner and Baker, 2008). It would therefore appear as though 02 uptake is not jeopardized in these species during generalized blood acidoses on account of Root effect onset pH values that are lower than those produced by these conditions.

72 The methods employed in Chapter 2, combined with the particular species analyzed, allowed me to assess another pertinent question — the nature of the transition in

Hb proton-binding strategy among fishes. There are two methods by which the Hbs of fishes take up the protons yielded by 2CO hydration within the RBC: intrinsic buffering, and the oxylabile Haldane effect (see Chapter 2). An inverse relationship has been shown to exist in fishes between the magnitudes of their Hb buffer values and their

Haldane effects, such that plesiomorphic species such as elasmobranchs tend to have high

Hb buffer values and small Haldane effects, while derived species such as teleosts tend to have low Hb buffer values and large Haldane effects (Jensen, 1989; Jensen et al., 1998).

This transition in Hb proton-binding strategy therefore likely occurred among the basal actinopterygians, the group of fishes intermediate to the elasmobranchs and the teleosts on the phylogenetic spectrum (Janvier, 2007). As the Hb titrations employed in Chapter

2 simultaneously assess both the intrinsic buffer capacity and Haldane effect of the Hbs being titrated, so long as the Hb properties of these seven species are representative of their respective ancestral states (Janvier, 2007; McKenzie et al., 2007), titrating these Hbs could elucidate the nature of this transition in Hb proton-binding. The results suggest that the transition in Hb proton-binding strategy from large intrinsic buffering/small

Haldane effect to low intrinsic buffering/large Haldane effect occurred not by gradual succession among the primitive ray-finned fishes, but rather a more punctuated process.

A marked decrease in buffer value and increase in the Haldane effect occurred in the last common ancestor of bowfln and the teleosts, and is correlated with what is believed to be the first appearance of the choroid rete (Berenbrink et al., 2005). As the findings of

Chapter 3 show the presence of a choroid rete to be associated with a significant increase

73 in the magnitude and onset pH of the Root effect, it is possible that this transition in Hb proton-binding is in some way related to the optimization andlor utilization of the Root effect in these fishes — a low intrinsic Hb buffer value would reduce the number of necessary protons shifted to the RBC cytosol to activate the Root effect, while a large Haldane effect would ensure 2CO transport did not suffer despite the decreased Hb buffer capacity. Furthermore, being linked mechanistically to the Bohr effect, a large Haldane effect would allow for the elegant and tight coupling of 02 and 2CO transport in the blood of these fishes (Brauner and Randall, 1996; 1998).

Finally, the actinopterygian species analyzed in my thesis straddle the original appearance of the choroid rete in the last common ancestor of bowfin and the teleosts

(Berenbrink et a!., 2005). The rete is believed to be an integral part of the Root effect- related 02 delivery system (Pelster and Scheid, 1992; Pelster and Randall, 1998; Pelster, 2001; Berenbrink et al., 2005), allowing for the diffusion of 02, ,2C0 and protons from the venous capillaries back into the arterial supply. Analyzing the Hbs of these particular species allowed me to gain an understanding of how the presence of a rete may influence the manifestation of Root effect-related Rb characteristics. Indeed, the results of

Chapters 2 and 3 suggest the rete as being associated with these four related Rb characteristics, the relationships of which are made clear when the data are portrayed on the same graph (Fig. 4.1). Haemoglobin buffer capacity, Haldane/Bohr effect, and Root effect maximum and onset pH all change markedly in accordance with the appearance of the choroid rete. What is more, Fig. 4.1 suggests a somewhat different manner by which the Root effect and its related Rb properties may have evolved compared to that of our current understanding based on the findings of Berenbrink and colleagues (2005). Their

74 data show a gradual increase in maximal Root effect and a gradual decrease in Hb buffer value with phylogenetic progression among the basal actinopterygians, suggestive of

directional selection for these Hb traits (Endler, 1986). However, as my data show Root

effects in the basal orders of the actinopterygians (i.e., acipenseriformes; ginglymods) that are elicited only at very low pHs, and as these species possess no known mechanisms

of generating such extreme acidoses in the blood (i.e., gas gland; rete), a question arises

as to how a seemingly in vivo-irrelevant Hb trait can be selected for in a directional

fashion. My data, on the other hand, show these traits as changing markedly only once

an acid-producing mechanism appears at a particularly beneficial site (i.e., the

avascularized eye; Fig. 4.1). This is sensible, as it is presumably only with the help of a

rete that a species is capable of generating the appropriate acidosis to allow it to make use

its modest Root effect. Once this countercurrent system was in place in the last common

ancestor of bowfin and the teleosts, those individuals with larger Root effects elicited at

more in vivo-relevant pHs would have been capable of offloading larger amounts of 02 to

their avascularized retinas. This may have allowed for improved vision, assisting in such

things as foraging and mate selection, and ultimately leading to increased survival and/or

reproductive success of those individuals possessing these particular Hb traits. It is in

this fashion that favourable selection may have originally been placed on the Root effect

and its onset pH value, as it was now capable of being utilized to its own end due to the

species’ ability to generate the appropriate activating acidoses through the choroid rete.

Furthermore, the theoretical negative consequences of the Root effect (i.e., jeopardized

02 uptake during generalized blood acidoses in the absence of f3NFIE)were rendered

negligible in these species owing to a low Root effect onset pH value (Chapter 3).

75 Hence, we see a marked increase in these Root effect properties starting with bowfin, the most basal of extant species to possess a choroid rete (Fig. 4.1). But if the Root effect seemingly only became physiologically relevant to organisms with the appearance of the choroid rete, why would it exist, albeit modestly, in the more basal orders

(acipenseriformes; ginglymods) that lack retia? This is especially pertinent with regards to the potential negative consequences that come with a Root effect in the absence of

I3NHEactivity (i.e., jeopardized 02 uptake during generalized acidoses). The reason for this may lie in the positive correlation between Root effect and Bohr effect magnitudes, and the hypothesis of the Root effect being the extreme expression of the mechanism by which these early ray-finned fishes activated their Bohr effects (Berenbrink et al., 2005).

The Root effect may therefore not have evolved specifically for enhanced 02 delivery itself, but rather as the byproduct of a selection-favoured Bohr effect in the early ray- finned fishes (Berenbrink et a!., 2005). The Bohr coefficients determined for the four non-rete species in Chapter 2 suggest their Bohr effects as being selected-for as a mechanism of 02 delivery (Lapennas, 1983), as the magnitude of their Bohr coefficients are consistent with the optimal value determined for oxygen delivery under steady state conditions. This, therefore, may explain the presence of a seemingly in vivo-irrelevant

Root effect in these species.

Future directions

I do not assume I am anomalous among graduate students in terms of the lack of satiation I feel upon reaching the end of my MSc, for, in many ways, a completed thesis project raises as many questions as it answers. Once a difficult task, posing new and

76 interesting research questions has become much easier with the knowledge acquired over the course of a masters degree. I will use this section to briefly expand on some interesting ideas and questions that relate to my thesis project that I believe would be worthwhile pursuits for a future masters student.

As physiological traits do not tend to leave their mark in the fossil record, the study of the evolution of physiological systems hinges in part on the comparative phylogenetic approach (Bradely and Zamer, 1999; McKenzie et al., 2007; Garland et al.,

2005; Berenbrink, 2006). There are limitations to this approach in the form of assumptions that the physiology of an extant member of a “primitive” order is representative of the ancestral physiological state of that species’ (normally extinct) ancestors. This is a legitimate concern, one that is particularly relevant to the study of the evolution of a protein like Rb where, it has been shown, single amino acid substitutions on a relatively large 65 000 dalton protein can have marked effects on the protein’s function (Hochachka and Somero, 2002). For this, and other, reasons, it is therefore difficult to infer evolutionary relationships of species based on physiological function alone. An effective way of increasing the robustness of the comparative phylogenetic approach, however, is to include in the study as many species as possible so as to allow for a more highly resolved picture of the evolution of the physiological trait in question.

Although the seven species included in this study were fairly representative of the basal actinopterygian lineage, it would be interesting to see if the inclusion of as many of the species within this group as possible would alter our understanding of how these Hb traits came about. It could also illuminate the degree of intra-order variation in these traits. If there did in fact exist a large degree of variation within a particularly variable order (e.g.,

77

bowfin

rete)

physiological

(Berenbrink

Perciformes

(Anabas breathing

of

interesting

virtually

obligate

transport

2001),

the

Haldane

plesiomorphic

2005),

measured

Pirarucu’s

the

environments,

low

Haldane

osteoglossids),

evolved

buffer

One

and

suggesting

and

There

testudineus),

air-breathing

all

effect

large

fishes,

(Jensen,

to

the

Hb

is

of

other

et

order

effects

at

relevance

capacity.

see

much

was

the

al.,

teleosts.

buffer

quantities

species

that

and

a

namely

particularly

if

fishes’

conclusions whose

2005).

a

an

this

1989;

this

could

is

more

of

remarkable

value

anomaly

an

markedly

fish,

other

such

only

A phenomenon

would

This

those

obligatory

measured

species

of

Brauner

be

similar

likely

with

is

CO 2 . as

teleosts

when

tested

beneficial

is

significantly

of

with

possibly

elasmobranchs

found

a

ability

higher

arterial

candidate

have

Chapter

sensible in

and

This

an

using

(Randall

air-breather

a

magnitude

(Jensen,

was

Root

in

acid-producing

been

Weber,

to

than

is

be

Pco 2

site

pirarucu

phylogenetically

buffer

perhaps

paralleled

3

idea,

for

an

78

effect higher

involves

shown

(i.e.,

those

et

levels

1989;

indication

this

(Jensen,

1998;

al,

to

but

(Graham,

protons

with

the

which

those

species,

attributable

than

1978;

could

to

it

ten

Brauner

in

the

Jensen,

eye)

is

possess

mechanism

regards

the

those

times

1989).

Hb

based

should,

Perry

of

Root

in

be

in

1997)

one

Hbs

independent

the

adaptations

buffer

found

the

and

2001;

of

greater

a

on

effect

to

to

et

blood, that

However,

of

large

any

presumably,

last

Weber,

pirarucu

of

its

al,

the

(i.e.,

other

values

in

Berenbrink

is

the

high

other

common

1996).

coming

the

assumption

than

Root

not

and

the

highly

of

contrasts.

obligate

climbing

1998;

Hb

of

dissimilar

it

being

teleost

Hb

those

therefore,

gas

effect

displays

It

more

into

buffer

possess

to

ancestor

would

derived

gland

Jensen,

et

of

an

particular

air-

that

al.,

perch

value.

from

a

and

be

Hbs

all of fish eyes are avascularized, and thus, would benefit from an 02 delivery mechanism like the Root effect. However, I am unable to find any information in the literature

suggesting that any fish eyes other than those of teleosts are avascularized. If this is the

case, that only teleost fishes possess avascularized eyes, then perhaps the reason for this

is attributable to the fact that they have a Root effect and a choroid rete that together are

capable of saturating the retinal cells with 02, and thus, negate the need for a high degree

of retinal vascularization. Retinal cells are highly sensitive to low 02 levels (Peister and

Randall, 1998), and a way of ensuring sufficient 02 saturation is by increasing their

vascularization. However, this could come at the detriment to visual acuity (Chang et al.,

2001). So, if a different mechanism capable of saturating these cells with 02 evolved

(i.e., the Root effect and choroid rete), perhaps it would be selected-for over the

potentially vision-limiting high vascularization mechanism of retinal oxygenation. If this

were the case, one could hypothesize that avascularization of the fish eye only began with

the evolution of the choroid rete in the last common ancestor of bowfin and the teleosts,

while those more plesiomorphic species lacking choroid retia would retain eyes of high vascularization so as to ensure their sensitive retinal cells remained 02-saturated. Although speculative, this hypothesis is supported by the fact that the eyes of dogfish

(Squalus acanthias), an elasmobranch lacking choroid retia, appear to be highly

vascularized (Chris Wood, personal communication), as do those of white sturgeon

(personal observation). It would therefore be interesting to determine the vascularization

in a number of species lacking choroid retia, including those species plesiomorphic to

bowfin as well as those derived teleosts that have secondarily lost their choroid retia (e.g.,

Silurus glanis, Monopterus albus; Berenbrink et al., 2005). Because it is currently

79

compliment

some

levels vitreous

looking where

capillaries

that

and

arterial therefore

effect Together,

effect

species

sturgeon.

believed

selective

equipped

an

avascularized

Randall,

are

of

(Regan

in

onset

the

blood

Possibly

highly

(Chapter

at

the

humor

some

the

be pressures

fishes

diffusion

these

the

may

Sturgeon

ways

the

pH

of

selective

1998).

supply,

et

ocular

capacity,

metabolically

interest

potentially

immediately

ideas

not

suggest

were

al.,

related

2),

retina

we

driving

significantly

of

unpublished).

Hbs

a

“Candidate”

Po 2

view

mentioned

it

avascularized,

maximal

pressures

large

to

with

is

to

that

despite

display

values

measure

likely

the

this

drive

it

quantities

the

proximal

02,

active).

may

evolution

evolutionary

Bohr

its

of

white

that

acting

pH

buffer

if

different

above

tissues

be

white

the

If

it

lack

down

the

our

was

this

altered.

coefficient

If

to

sturgeon

Po 2

of

on

related

values

of

these

Root

of

understanding

their

may

sturgeon

CO 2

in

found

than

the

to

known

values

the

80

fact

idea

6.6.

Root

Po 2

include

retinas

effect

into

markedly

to

Root

that

may

that

is

of

of

retinal

at

retia

values

Indeed,

have

the

the

of

effect

0.65

the

certain

is

only

effect,

be

are

the

bowfin

minute

case,

involved

or

Root

of

shown

capable

vascularization.

(Chapter

lower

higher

were

those

were

eye

acid-producing

preliminary

Root

and

“candidate”

it

effect

and

(pH

blood

has

higher

that

than

in

eyes

could

effect

of

than

in

place

2),

the

the

6.63;

using

02

the

is

the

volume

of

are

and

than

the

potential

perhaps

exercising

delivery

evolution

experiments

02

Root

other

so

tissues

Chapter

its

their

case

a

tissue.

levels

those

as

5%

modest

of

effect/retia

non-rete

to

arterial

of

even

(Peister

to Root

the

(i.e.,

saturate

and

of

of

muscle,

3).

white

It

modify

supply

the

their

would

Root

those the 02 50

14 I 4.5 Rb 8ufe Vahe 40 I

10 G7 12 2.5 0

IL 5 uj

Max. x Rct Effect 2.5 I 8 RootffctOnsetpH 10

Rda,e Effect

I I 1.5 0 Gd 4 0% 0% 44 0% r Phylogermtlc Prvgresslon

Figure 4.1. Changes in mean values of haemoglobin (Hb) P50 buffer value (circles), Haldane effect (diamonds), maximal Root effect (squares), and Root effect onset pH (triangles) in seven species belonging to four orders of basal actinopterygian fishes. In phylogenetically progressive order, these include: acipenseriformes (American paddlefish; white sturgeon); ginglymods (spotted gar; alligator gar); amiiformes (bowfin); osteoglossiformes (mooneye; pirarucu). Grey field indicates the presence of a choroid rete, with species belonging to the amiiformes believed to be the first in which this system originally evolved. All values determined from experiments performed on haemolysates at 4[Hb of 0.04 mmol r1 (buffer value and Haldane effect) or 0.16 mmol r’ (Root effect maximum] and onset pH) and a [KC1] of 0.1 mol ii, in the presence of organic phosphates (3:1 4).GTP:Hb

81 -CHAPTER SUMMARY-

1. Although their Hb proton-binding properties (instrinsic buffering and Haldane

effect) do not appear to be playing any major role in protecting RBC 1pH during generalized acidoses of the blood, the basal actinopterygian fishes appear to

protect 02 uptake during these conditions by way of Root effect onset pH values

below those produced by the generalized acidoses themselves. This is despite these fishes displaying relatively large Root effects and lacking RBC -1pH protecting NHE.

2. Assuming the Hb properties of these seven species to be representative of their

respective ancestral states, numerous Hb properties changed in marked, and even

significant ways, in correlation with the appearance of the choroid rete in the last

common ancestor of the amiiformes and the teleosts. This Hb traits include:

buffer value (decreased); Haldane effect (increased); Bohr effect (increased);

Root effect magnitude (increased); Root effect onset pH value (increased).

3. The results yielded by this thesis have provided numerous further questions, and

with them, research avenues that would be interesting to pursue in the future.

82 References

Baillie, J. E. N., Bennun, L. A. and Brooks, T. M. (2004). A Global Species Assessment: 2004 IUCNRedList of Threatened Species. Gland: World Conservation Union.

Baker, D.W., Morgan, J.D., Wilson, J.M., Matey, V., Huynh, K.T. and Brauner, C.J. (2008). Complete intracellular pH protection during extracellular pH depression is associated with hypercarbia tolerance in white sturgeon, Acienser transmontanus. American Journal of Physiology (Submitted).

Berenbrink, M. (2006). Evolution of vertebrate haemoglobins: Histidine side chains, specific buffer value and Bohr effect. Respir. Physiol. & Neurobiol. 154, 165-184.

Berenbrink, M. (2007). Historical reconstructions of evolving physiological complexity: 0-2 secretion in the eye and swimbladder of fishes. I Exp. Biol. 210, 1641-1652.

Berenbrink, M., Koldkjaer, P., Kepp, 0. and Cossins, A. R. (2005). Evolution of oxygen secretion in fishes and the emergence of a complex physiological system. Science 307, 1752-1757.

Bonaventura, C., Crumbliss, A. L. and Weber, R. E. (2004). New insights into the proton-dependent oxygen affinity of Root effect haemoglobins. Acta Physiol. Scand. 182, 245-258.

Bradley, T. J. and Zamer, W. E. (1999). Introduction to the symposium: what is evolutionary physiology? Amer. Zool. 39, 321-322.

Brauner, C. J. and Randall, D. J. (1996). The interaction between oxygen and carbon dioxide movements in fishes. Comp. Biochem. Physiol. A 113, 83-90. Brauner, C. J. and Randall, D. J. (1998). The linkage between 02 and 2CO transport. In Fish Physiology: Fish Respiration, vol. 17 (eds S. F. Perry and B. L. Tufts), pp. 283- 321. New York: Elsevier.

Brauner, C. J. and Weber, R. E. (1998). Hydrogen ion titrations of the anodic and cathodic haemoglobin components of the European eel Anguilla anguilla. I Exp. Biol. 201, 2507-25 14.

Brauner, C.J. and Baker, D. (2008). Patterns of acid-base regulation infish. In Cardio Respiratory Control in Vertebrates: Comparative and Evolutionary Aspects. (eds M.L. Glass and S.C. Wood), in press. Berlin: Springer-Verlag.

Burleson M., Shipman, B. and Smatresk, N. (1998). Ventilation and acid-base recovery following exhausting activity in an air-breathing fish. I Exp. Biol. 201, 1359- 1368.

83

Biochem.

Jensen,

Jensen,

Physiology:

Brauner),

225-234.

Physiology:

Janvier,

Brauner),

lives, Biochemical

317-327. Holeton,

gases

Hochochka,

and Heming, New

breathing

(1986).

101-157. Graham,

exposure

Gonzalez,

comparative

Garland, Princeton,

Endler, neovascularization.

Chang,

in

Butler,

117—130.

the

H.

York:

in

K.

dogfish

Ionic

F.

F.

Respir.

blood

P.

J-H,

J.

P.

L.

New

Physiol.,

G.

pp.

T.

pp. in

on

J.

T.,

B.

B.

NJ:

R.

J.

A.

Academic

Primitive

and

Primitive

the

(2007).

A.,

F. B.

Adaptation:

physiology.

P.

acid-base

2-53.

5

equilibria

(2001).

(1989).

and

Jr,

Gabison,

J.,

(1986).

York: (Scyliorhinus

and

15-536.

Princeton

and

(1997).

Physiol.

bown,

W.

Randall,

Randall,

Milligan,

Bennett,

Taylor,

A:

water

New

and

Living

Randall,

Cur.

Oxford

Hydrogen

Hydrogen

Mol.

Fishes,

Fishes,

Modes

Press.

Amia

New

and

Diversity

in

E.

Somero,

York:

65,

afferent

University

I

Mechanisms

Opinion

D.

red

D.

E.

E.,

Integr.

A.

Primitive

ion

L.,

Exp.

223-234.

York:

calva. canicula)

University

J.,

J.

of

W.

vol. D.

vol.

blood

F.

Kato,

Elsevier.

Pagnotta,

regulation

selection.

Boutilier,

ion

ion

(2007).

Biol.

J.

and

and

and

(1975).

G.

Ophthalmol.,

Physiol. 26

26

Academic

Phusiol.

(1967).

binding

equilibria

Press.

cells

N. T.

Fishes

Rezende,

natural

efferent

(eds

(eds

208,3015

and

at

and

(2002).

Press.

Why

different

The

of

A.,

after

In

D.

R.

D.

Processes

84

129, Effect

Biochem.

properties

and

rainbow

Azar,

Natural

have

Press.

history.

and

J.

J.

G., effect

to

in

exhaustive

Influence

E.

McKenzie, 12,

McKenzie,

511-5

-3035.

Fishes

gills

fish

Iwama,

of

McDonald,

seasonal

primitive

L.

D.

242—249.

of

hypoxia

trout

hemoglobins.

Selection

of

in

(2005).

In

Zool.

T.

17.

progressive

of

from

Physiological

rainbow

Air

(2001).

tuna

of

G.

(salmo

exercise

temperatures.

A.

A.

74,

fishes

Breathing

oxygen

Deep

upon

K.

Phylogenetic

haemoglobins.

P.

D.

P.

in

502-509.

and

Comeal

trout.

Farrell

Farrell

the

gairdneri)

G.

survived?

partial

hypoxia

Time.

I

and

availability.

(2001).

Primmett,

Exp.

Wild.

Evolution,

I

Fishes.

during

and and

In

Exp.

I

pressure

Biol.

pp.

approaches

on

exp.

Fish

Effect

Cr,

C.

C.

In

Comp.

Biol.

pp.

respiration

low

15-26.

Fish

J. J.

143,

Biol.

In D.

HC0 3

pp.

13-64.

of

pH

of

46,

63,

air in

In:

Weber, B.

recovery

Richards,

(Acipenser

Randall,

P., the

Randall,

dioxide Fish

Perry

Elsevier.

Peister,

130.

the

Peister,

Sci.

Peister,

Exp.

& Nelson,

Milligan, Fishes,

and York

McKenzie,

delivery.

Lapennas,

Tufts),

Jensen,

biochemistry.

173,

Hemoglobin

Sons,

Bolis,

pirarucu

fish

H

16,

Respiration,

Biol.

:

S.F.,

463-474.

Elsevier.

pp.

exchange

R.

transport

swimbladder.

287-291.

vol

B.

B. J.

B.

metabolism

Inc.

F.

D.

D.

L.

Respir.

J.

C.

123,

E.

naccarii).

S.

and And

(2001).

B.,

1-40.

Wood

26

G.

(Arapaima

J., D.

J.,

and

G.,

L.

(1994).

(2000).

(eds

N. In

Fago,

J.,

McKenzie,

Function Farrell,

93-12

Randall,

and

Scheid,

Heigenhauser,

Agradi,

New

in

Phyiol.

vol. with

Fish

(1983).

Farrell,

C.M.,

The

D.

vitro:

Wood,

in

I

Fishes

1.

A.

Physiol.

Adaptations

17

York:

J.

the

Physiology:

the gigas),

Exp.

generation

A.

and

P.

Mckenzie,

(eds

54,

in

Walsh

D.

E.

a

The

environment

A.

pacific

P.

D.

comparative

(1992).

Vertebrates,

(1992).

Biol. J. of

C.

Elsevier.

161-172.

Weber,

and

S.

P.

J.,

magnitude

Zool.

an

(1998).

the

M.

F.

G.

P.J.

and

Abrami,

obligate

170,113

Haswell,

spiny

for of

World,

Perry

(1986).

Fish

Counter-current

J.

A.

Effects

65,

and

hyperbaric

Brauner,

R.

F.,

oxygen

Physiology

P.

after

Respiration,

1-16.

Molecular study.

dogfish

E.

and

and Farrell

of

Thomas

3rd

air-breathing

85

-125.

Intracellular

G.,

(1998).

of

M.

the

exhaustive

B.

Wood,

transport: Edition.

diet

Bondiolotti,

Comp.

C.

S.

Bohr

L.

(Squalus

oxygen

and

(1978).

of

J.

Tufts),

on

S.

Adaptation

Red

concentration

C.

the

vol.

(2007).

coefficient:

C.

(1996).

Biochem.

response

pp.

and

J.

exercise

blood

fish.

lessons

tensions

Root

M.

17

acanthias)

Brauner),

pp.

Carbon

77-96.

G. extracellular

(2003).

(eds

Rish

Can.

Fish

effect.

cell

14

P.,

in

to

Physiol.

from

in

1-184.

optimal

in

New

S.

Extreme

hypoxia

Physiology

and physiology

dioxide

I

Natiello,

red

the

F.

fish. pp

Exercise

Zool.

I

In

fish

blood

gas

Perry

rainbow

York:

xi-xiii.

Comp.

New

Fish

A.

News

acid-base

for

hemoglobins.

secretion

excretion

56,

and

in

113,

F.,

cell

and

York:

Physiology:

oxygen

sturgeon

John

and and

.

977-982.

Physiol.

New

Physiol.

Primitive

Bronzi,

trout.

121-

carbon

B.

Wiley

status

in

L.

in I Temperate Environments. (eds F. Di Prisco, B. Giardina and R. E. Weber), pp. 23-37. Milano: Springer-Verlag.

86