SODQUM UPTAKE BY ISOLATED,
PERFUSED GiLLS 0F RAINBOW TROUT
(SALMO GAIRDNERI), WITH SPECIAL
REFERENCE TO PATTERNS OF
GILL BLDDD FLOW

Thesis for the Degree of Ph. D.
MTCHIGAN STATE UNIVERSITY
BRENT D. RICHARDS
1968

THESIS

0-169

EUN NTVERSIT

TTTTTTTT TTTTT TT TTT

29310569 93

O

MIC HT

TTTTT

“3

Ag:

 

This is to certify that the

thesis entitled

SODIUM UPTAKE BY ISOLATED, PERFUSED GILLS
OF RAINBOW TROUT (SALMO GAIRDNERI),
WITH SPECIAL REFERENCE TO PATTERNS

OF GILL BLOOD FLOW

presented by

Brent D. Richards

has been accepted towards fulfillment
of the requirements for

_P_h_:_D_o___ degree in My

WM 0 V/w‘

Major professor

 

Date_AL1g2§_t_8.._l9_68.__

    
      

' mam .v
HOME & SITNS'
800K BINDEIIY INC.

umnv mans-ac

  

 

ABSTRACT

SODIUM UPTAKE BY ISOLATED, PERFUSED GILLS OF RAINBOW
TROUT (§ALMO GAIRDNERI), WITH SPECIAL REFERENCE
TO PATTERNS OF GILL BLOOD FLOW

by Brent D. Richards

The pattern of fluid flow through the isolated,
perfused rainbow trout gill can be altered from sinus flow
to lamellar flow. Under control conditions and with
acetylcholine, a majority of the fluid flows through the
filamental sinus. 'When epinephrine is added to the
perfusion fluid, the majority of the fluid flow is via
the lamellae. These blood flow patterns were determined by
perfusing the gills with india ink containing the various
vasoactive substances, and then examining the gills
histologically.

The filamental sinus is apparently a portion of the
lymphatic system and connects to the filamental lymph
vessels at frequent intervals. There are also direct
connections from the afferent filamental artery to the
lymph vessels and sinus and from the sinus to the efferent
filamental artery. The lamellar-afferent and lamellar-
efferent vessels have muscular walls for at least part of

their lengths and probably function in the regulation of

Brent D. Richards
2
the pattern of fluid flow through the gill.

Isolated, perfused rainbow trout gills are capable
of actively taking up sodium against a concentration
gradient of at least 60:1. This sodium uptake was a
apparently stimulated directly by the low sodium content
of the choline and sucrose Ringer solutions used. It
could also be produced by the addition of epinephrine to
perfusion fluid which was not low in sodium. The amount
of sodium in the various solutions studied was determined
by flame photometry.

Studies using metabolic inhibitors suggest that the
ATP necessary for the uptake of sodium is derived almost
exclusively from oxidative metabolism.

Although sodium and chloride can be taken up
independently, evidence is presented that there is an
interaction between sodium and chloride uptake. The rate
of sodium uptake is not primarily dependent upon either
the rate or the pattern of blood flow through the gill.

SODIUM UPTAKE BY ISOLATED, PERFUSED GILLS OF RAINBOW
TROUT (fiALMO GAIRDNERI), WITH SPECIAL REFERENCE
TO PATTERNS OF GILL BLOOD FLOW

By

Brent D: Richards

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Department of Physiology

1968

ACKNOWLEDGMENTS

The author wishes to thank Dr. P. 0. Fromm for his
support and guidance during this study. Thanks are also
extended to Dr. J. R. Hoffert for his many helpful
suggestions.

He also wishes to thank the other members of his
committee, Dr.‘W. D. Collings, Dr. J. M. Dabney and Dr.

'W. L. Frantz for their interest and help in this project.

In addition, the writer is indebted to the Department
of the Interior, Federal Water Pollution Control
Administration grant SIWP-00807 for support of this project.

11

TABLE OF CONTENTS

Page
INTRODUCTION 0 O O O C O O C O O O O O O O O O O 1
REVIEN OF THE LITERATURE . . . . . . . . . . . . 3

Circulatory Anatomy . . . . . . . . . . . . 3
Sodium Movement . . . . . . . . . . . . . . 5

MATERIALS AND METHODS . . . . . . . . . . . . . 11

Animals . . . . . . . . .
Surgery . . . . . . . . .
Perfusion Apparatus . . .
Cannulation and Perfusion
Circulatory Anatomy . . .
Sodium Movement . . . . .
Sodium Analysis . .
Uptake of Sodium and Inhibitor Studi
Statistics . . . . . . . . . . . .

O O O O O O O O O
N
O

08

O
N
(7\

RESULTS 0 O O O O O O O O O O O O O O O O O

Circulatory Anatomy . . . . . . . . . . . . 26
SOdium Movement 0 o o e o e o o o e o o o o 55

DISCUSSION 0 O O O O O O O O O O O O O O O O O O 62

Circulatory Anatomy . . . . . . . . . . . . 62
Sodium Movement . . . . . . . . . . . . . . 68

CONCIUSIONS o o o o e o e e e o o e o o e o o o e 72
APPENDIX 0 e e e o o o o e o o o o e o o e e o o 7Ll
LITERATURE CITED 0 e e o o o e o o o e e e o e o 77

iii

Table

1.

LIST OF TABLES

Page

Sodium concentrations of the perfusion

fluid and the fluid collected after

passing through the gill, and the rate

of sodium uptake for the various

experimental perfusions . . . . . . . . . . 57

The rates of fluid flow through the

gill during control and experimental

perfusions and the percent change in

rate from the control to the experimental
perfusion . . . . . . . . . . . . . . . . . 60

iv

LIST OF FIGURES

Figure

1.

IO.

ll.

12.

The locations of the incisions made in
removing the opercula and anterior
portion of the head . . . . . . . . . . .

The locations of the incisions made in
separating the first pair of gill

arches from the remainder of the
branchial apparatus . . . . . . . . . . .

The perfusion apparatus . . . . . . . . .

The major circulatory paths from
the ventral to the dorsal aorta . . . . .

Longitudinal section of a gill filament .

A diagrammatic cross-section of a gill
filament from a rainbow trout . . . . . .

Photomicrograph of the afferent end of

a cross-section of a filament from a
rainbow trout gill which had been
perfused with india ink . . . . . . . . .

Photomicrograph of the efferent end of

a cross-section of a filament from a
rainbow trout gill which had been
perfused with india ink . . . . . . . . .

Photomicrograph of a longitudinal
section of a gill lamella . . . . . . . .

Photomicrograph of a longitudinal section
of a gill filament, showing the
lamellareafferent arterioles . . . . . .

Photomicrograph of a longitudinal
section of a gill filament, showing

the lamellar-efferent arterioles . . . ._

Photomicrograph of a longitudinal
section of a gill filament, showing
the sinus-efferent vessels . . . . . . .

V

Page

1%

l6
19

28
3O

36

38

38

Figure

13.

1%.

15.

16.

170

l8.

19.

20.

21.

22.

23.

2%.

Photomicrograph of a cross—section of a
gill filament in which the lymphatics
had been perfused with india ink . . . . .

Photomicrograph of a cross-section of a
gill filament, showing the beginnings of
the connection from the afferent filamental
artery to the afferent collateral and

the filamental sinus . . . . . . . . . . .

Photomicrograph of the section serially
succeeding that in Figure 1% . . . . . . .

Photomicrograph of the section serially
succeeding that in Figure 15 . . . . . . .

Photomicrograph of the section serially
succeeding that in Figure 16 . . . . . . .

Photomicrograph of the section serially
succeeding that in Figure 17 . . . . . . .

Photomicrograph of a cross-section of a
gill filament which had been perfused
with india ink containing acetylcholine .

Photomicrograph 03 a cross-section of a
gill filament which had been perfused
with india ink containing epinephrine . .

Photomicrograph of a cross-section of a
gill filament which had been perfused
with india ink in chiline Ringer solution.

Photomicrograph of a cross-section of a
gill filament which had been perfused
with india ink in sucrose Ringer solution.

Photomicrograph of a cross-section of a
gill filament which had been perfused with
india ink in choline Ringer solution
containing cyanide . . . . . . . . . . . .

Photomicrograph of a cross-section of a
gill filament which had been perfused
with india ink in choline Ringer solution
containing iodoacetate . . . . . . . . . .

vi

Page

1+0

#2

NZ

Ah

m.

#6

M6

M8

#8

50

SO

Figure

25.

26.

Page

Photomicrograph of a cross-section of a

gill filament which had been perfused with
india ink in choline Ringer solution

containing ouabain . . . . . . . . . . . . S2

Photomicrograph of a cross-section of a

gill filament which had been perfused with
india ink in choline Ringer solution

containing atropine . . . . . . . . . . . 52

vii

INTRODUCTION

Since Smith (1930) postulated an osmoregulatory role
for the teleost gill, there have been numerous studies
which support the idea that ions, particularly sodium and
chloride, are actively transported across the surface of
the gill. Marine teleosts transport sodium from the blood,
which has a sodium concentration of about 200 mEq/l, into
sea water, with a sodium concentration of about #50 mEq/l.
Sodium transport by freshwater fish is from the water,
with a sodium concentration of 0.1 to 20 mEq/l, into the
blood which has a sodium concentration of about 150 mEq/l.
Thus, the concentration ratios against which sodium
transport takes place range from about 2:1 (water to blood)
in marine teleosts to about 300:1 (blood to water) for
freshwater fish.

There are four major factors which must be taken into
account in analysing the net uptake or excretion of ions
by the gill. They are: (l) the mechanism by which ions are
moved (active transport, exchange diffusion, etc.), (2) the
effective surface area available for exchange, (3) the
duration of exposure between the blood and the water, and
(h) the concentration gradient against which the ions must
move.

The purpose of this study was to examine the net

1

2
sodium flux across isolated, perfused rainbow trout gills,
with respect to each of the above parameters. Some
information on the mechanisms involved was obtained by
examining the effects of metabolic inhibitors on the net
flux of sodium across the gill membranes. The effective
surface area available for exchange was studied by
histological examination of gills which had been perfused
with india ink containing various vasoactive agents. The
rate of flow of fluid through the branchial vasculature was
used as a measure of the duration of exposure between the
blood and the water. For all perfusions, flow rates were
used as an index of the vasoactive nature of the various
experimental solutions. The concentration gradients were
kept relatively constant in all experiments.

Since two separate and distince approaches were used
in this study, the major sections of the thesis have been
divided into at least two parts. The first part of each
section deals with the circulatory anatomy and the second

part with sodium movement.

REVIEH OF THE LITERATURE

Circulatory Anatomy

Although there have been numerous studies dealing
with the osmoregulatory and respiratory functions of the
teleost gill, only a few detailed studies of the functional
circulatory anatomy have been made.

The general path of blood flow through the gill has
been well established by Allis (1912, from Goodrich, 1958)
and by Mbtt (1950 a, 1951) and is as follows: ventral
aorta, afferent branchial artery afferent filamental artery,
lamellar lacunae, efferent filamental artery, efferent
branchial artery, and then to the dorsal aorta. The
counter current nature of the lamellar blood flow and the
flow of water across the gill surface was demonstrated by
van Dam (1938, from Steen and Kruysse, 196%).

Steen and Kruysse (196%) examined the pattern of
blood flow through isolated teleost gills and found that
there were three routes by which blood could flow from the
afferent to the efferent filamental arteries. In addition
to the previously established lamellar pathway, they found
that blood could go directly from afferent to efferent
filamental artery at the apical end of the filament and that
it could also flow through the large sinus which occupied

the central portion of the filament. Hughes and Grimstone
3

T,
(1965) suggested further that lamellar blood flow might
occur along two separate paths. The first, and perhaps
"preferred" path is along the margin of the lamella by a
direct channel formed by the pillar cells and the lamellar
epithelium. The second path is through the body of the
lamella among the pillar cells. These authors also
suggested that the pattern of blood flow through the gill
is regulated by the state of contraction or elongation of
the pillar cells, a view which is shared by Newstead (1967).
The effects of various pharmacological agents on the
branchial vascular resistance have been studied using both
isolated gills (Keys and Bateman, 1932; Dstlund and Fange,
1962) and in intact fish (Mott, 1951b; Randall and Stevens,
1967; Stevens and Randall, 1967). It is difficult to
interpret the results of the lg 1129 experiments since
other portions of the circulatory system, in addition to
the branchial vasculature, were undoubtedly affected by the
injected substances. No quantitative comparisons could be
made among the results presented in the above studies since
only one paper (Keys and Bateman, 1932) gave the actual
concentrations of the vasoactive substances tested. In the
other papers, the doses administered were reported but
since the volumes of blood or perfusion fluid in which the
agents were dissolved were not specified, the actual
concentrations could not be determined. Qualitatively,
however, these workers found that epinephrine decreased the

resistance and acetylcholine increased the resistance to

5

branchial blood flow in all of the species studied.
Histamine decreased the resistance to branchial blood flow
in one species of fish and apparently increased it in
another. It was also found that the response to acetyl-
choline could be blocked by atropine (Dstlund and thge,
1962) and the response to epinephrine could be blocked by
phenoxybenzamine (Randall and Stevens, 1967). The only
fish in which the branchial vascular resistance has been
shown to be under some direct nervous control is the eel,
which was shown by Mott (1951b) to have a vagal depressor
reflex.

Steen and Kruysse (196%) presented some evidence that
epinephrrne and acetylcholine not only change the
resistance of the branchial vasculature, but also change

the pattern of blood flow through the gill.
Sodigm Movement

Both marine and freshwater teleosts have the problem
of maintaining the osmotic concentration of the blood
against large concentration gradients. To prevent
excessive dehydration, marine teleosts must continually
drink sea water and excrete salts both renally and extra-
renally (Smith, 1932). The freshwater teleost, on the
other hand, must constantly take up salt and excrete a
copius volume of dilute urine (Pitts, 193%; Keys, 1937) in
order to maintain a relatively constant osmotic concen-

tration of its body fluids.

6

With very few exceptions, (notably the work of Keys
and Bateman, 1932 and of Bateman and Keys, 1932), the
information available relative to sodium and chloride
uptake and excretion by teleost gills is based on studies
of osmoregulation by intact fish maintained under a variety
of conditions.

By comparing the amount of water ingested and absorbed
to the volume of urine excreted per unit time, Smith (1930)
demonstrated that there was a significant extra-renal water
loss from the sculpin and eels adapted to sea water. 'With
another group of fish, he also determined that the amount
of chloride which was absorbed along with the water exceeded
renal chloride excretion and thus obtained indirect evidence
for extra-renal chloride excretion by marine fish. In
order to actually demonstrate the extra-renal excretion of
salts, eels which had been adapted to fresh water were used.
They were injected with salt solutions, fecal salt loss was
prevented by ligating the anus and the urine was collected
from free swimming fish in fresh water with a retention
catheter and a balloon. ‘With this preparation, any increase
in the salt concentration of the bathing solution was an
indication of extra-renal salt loss and Smith (1930) refers
to this loss as extra-renal excretion. The possibility
still remained, however, that this extra-renal loss
represented merely passive diffusion or leakage for which
the fish could not compensate by active uptake of salt from

the surrounding medium.

7

Keys (1931), Keys and Bateman (1932) and Bateman and
Keys (1932) used a perfused heart-gill preparation (in which
the heart pumped the fluid through the gills) and a perfused
gill preparation to study the secretory activity and some of
the branchial vascular responses in the salt water adapted
eel. They found that adrenaline increased the rate of flow
of fluid from the ventral to the dorsal aorta and that the
rate of chloride excretion decreased with an increase in the
rate of fluid flow through the branchial vasculature.

Changes in weight as well as changes in the freezing
point depression and chloride and protein concentrations of
the blood of Anguilla vulgarig during adaptation from fresh
water to salt water and back to fresh water were examined
by Keys (1933). He found that there was an initial stage
during which the fish responded passively and gained or
lost salts and/or water in response to the concentration
gradient between the blood and the surrounding medium. The
duration of this stage depended on whether the fish was
transferred from salt to fresh or from fresh to salt water.
This passive stage ended when the fish began to actively
control the composition and osmotic concentration of the
blood, and the weight was returning toward its pre-
treatment value.

Krogh (1937) measured the uptake of chloride by seven
species of freshwater fish which had been previously "washed
out" in distilled water. He found large variations among

the species studied but, in general, the fish were able to

8
absorb chloride from millimolar salt solutions.

Using a divided chamber which kept the salts lost in
the urine and feces from mixing with the water which
surrounded the head of the fish, Krogh (1938) found that
fish can absorb sodium and chloride separately. He also
hypothesized that the sodium uptake was at least partially
in exchange for ammonium ions and that the chloride must
exchange for bicarbonate. Romeu and Maetz (196%), using
radioactive tracers, provided additional evidence for the
independence of sodium and chloride transport by the in
yiyg gills of goldfish (Cara§sig§ aurgtug).

Maetz and Romeu (196%) re-examined the theory that
sodium exchanges for ammonium and that chloride exchanges
for bicarbonate. They found that increasing the internal
ammonium ion concentration increased the uptake of Na2l+ by
goldfish and that increasing the external ammonium ion
concentration inhibited the uptake of N32%. These
treatments, however, did not affect the chloride balance.

If the internal bicarbonate concentration was increased,
however, Cl36 uptake increased and if the bicarbonate
concentration in the external medium was increased, 0136
uptake was inhibited. These changes in C136 uptake occurred
with no significant change in the sodium balance. They also
found that injection of Diamox (a carbonic anhydrase
inhibitor) caused an inhibition of both Na2“ and 0136

uptake. A significant increase in blood chloride occurred

after inhibition of carbonic anhydrase in the marine teleost

9
§erranus gp. (Maetz, 1953) and a decrease in blood chloride
was observed in the freshwater teleost‘ggggg‘gp, (Maetz,
1956).

Gordon (1963) compared the losses of Cl36 from
rainbow trout adapted to fresh water, 1/7, 1/3, 2/3, and
3/3 sea water. In one group of fish the renal papillae
were ligated, the other group was unoperated. By comparing
the rates of appearance of the injected 0136 in the bathing
medium for both ligated and unoperated fish, he determined
that the total 0136 exchange in fresh water and in 1/3
sea water took place via the gills.

Kamiya (1967) found that cyanide and dinitro-phenol
inhibited the excretion of sodium by isolated gills of salt
water adapted eels, however, fluoride and iodoacetate had
little or no effect. At least a portion of the excretion of
sodium by the gills of salt water adapted eels is apparently
dependent upon oxidative metabolism. In addition, it has
been found that the level of Na-K activated ATPase is about
six times higher in the gills of sea water adapted fish
than in gills from fresh water adapted fish of the same
species (Epstein, Katz, and Pickford, 1967).

After making histological examinations of the gills
from 10 species of fish, Keys and Willmer (1932) put forth
the theory that there was a specific type of cell (the
"chloride cell," acidophil cell, Kestdillmer cell, or
Granel cell) which was responsible for chloride transport

across the gill epithelium. Bevelander (19%5) examined

10
the gills from 36 species of fish and could find no
histological evidence for specialized chloride secretory
cells. Virabhadrachari (1961) found that the number of
"chloride cells" increased during adaptation from 0-50%
sea water, did not change during adaptation from 50-75%
sea water and decreased during adaptation from 75-100% sea
water. Philpott (1965) used electron microscopy and
electron diffraction of silver acetate treated gills and
found that there was a greater density of silver salts in
the area of the "chloride cells" than in other areas.

From these studies it can be seen that, although fish
gills are certainly involved in osmoregulation and active
transport of sodium and chloride, it is not at all certain
whether this ability to transport ions is a general property
of gill epithelial cells or is reserved for certain

"transport cells" which may or may not have been identified.

MATERIALS AND METHODS

Animal;

The fish used in this study were rainbow trout (figlmg
gairdneri) weighing between 130g and 250g. They were
obtained from the Michigan Conservation Department Hatchery
in Grayling, Michigan, and were transported from the
hatchery to Michigan state University in a large, well
insulated metal tank, the inside of which had been painted
with non-toxic paint. The water in the tank was
mechanically agitated during the entire trip in order to
provide adequate aeration.

The holding facilities at Michigan State University
consisted of two rooms maintained at a temperature of
12 to 13°C with 1% hours of light and 10 hours of darkness
per day. The main cold room contained six 300 liter wooden
tanks which were lined with fiberglass. The smaller room
contained a variable number of smaller (100 liter) tanks
into which groups of fish were transferred shortly before
use. The fish were kept in aerated, flowing water from
which the chlorine and excess iron had been removed. Fish
in the main holding tanks were fed twice a week, those in

the smaller tanks were fasted.

ll

12

Man

Isolated gills were surgically prepared in the
following manner. An unanesthetized fish was removed from
a tank, wrapped in a cloth towel and weighed. Heparin was
then injected into the caudal portion of the dorsal aorta
and the fish was returned to the water for at least one
minute to allow the heparin to mix with the blood. The
fish was then removed from the water and the spine was cut
at the level of the pectoral fins. A mid-ventral incision
was made from the anus to a point just anterior to the
pectoral fins. The head was removed and the caudal portion
of the fish was discarded.

The opercula and anterior portion of the head were
removed by cutting as indicated by the dotted lines in
Figure 1 (A and B). The branchial apparatus was then
dissected free from the remaining section of the spinal
column and the dorsal musculature. Care was taken to
prevent damage to the efferent branchial arteries and the
dorsal aorta. The right and left halves of the branchial
apparatus were separated by a mid-dorsal incision as shown
in Figure 2 (incision A). The first pair of arches was
then removed by cutting as indicated by lines B, C, and
D (Fig. 2). The remaining gills were placed in a beaker
of dechlorinated tap water. The right and left first arches
were separated ventrally by a midline incision and, if one

of the first arches failed to perfuse, it was replaced by

13

Figure l
The locations of the incisions made in removing the opercula

and anterior portion of the head.

1%

 

 

A" venTreT View

 

B“ Te’reraT VTET.)

Figure l

15

Figure 2
The locations of the incisions made in separating the first
pair of gill arches from the remainder of the branchial

apparatus.

muster//:\\

FFFFFFF

17
the corresponding second arch. During surgery, the gills
were kept moist by periodic immersion in dechlorinated tap

water.

We“ 122m

The gills were perfused using a constant pressure
perfusion apparatus shown diagrammatically in Figure 3. It
consisted of a pair of plastic 10cc syringe barrels
connected by a piece of plastic tubing and a 3-way valve.
The valve was attached to a blunt-tipped 25 gage needle.
Leading from the needle was a piece of PE 20 tubing 50cm
long, into which had been inserted a blunt-tipped, cut-off
barrel of a 25 gage needle. This needle served as the
afferent cannula and was inserted into the afferent
branchial artery at the ventral end of the gill. The
efferent portion of the perfusion apparatus consisted of a
blunt-tipped, cut-off 25 gage needle barrel inserted into a
25cm piece of PE 20 tubing. This cannula was inserted into
the afferent branchial artery at the dorsal and of the gill.

The inflow and collection pressures could be altered
by changing the height of the syringe barrel and the free
end of the efferent cannula with respect to the gill. The
pressures used were 5% and 20cm of Ringer solution (%0 and
15mm Hg) for the inflow and collection pressures
respectively. These approximate the pressures found by
Stevens and Randall (1967) for the ventral and dorsal aortic

blood pressures in intact, swimming rainbow trout.

18

Figure 3
The perfusion apparatus which consists of two 10cc syringe

barrels connected by a piece of plastic tubing and a 3-way

valve. Leading from the valve is the afferent cannula which
consists of 50cm of PE 20 tubing, into which is inserted a

cut-off, blunt—tipped 25 gage needle.

FFFFFFF

20

Caggulation agd Perfgaion

The dorsal and ventral ends of the gill were trimmed
of excess bone and tissue, exposing the afferent and
efferent branchial arteries respectively. Prior to
cannulation of a gill, the valve on the perfusion apparatus
was Opened and a steady, rapid flow of fluid was established.
The afferent cannula was inserted into the afferent
branchial artery and tied in place with No. 30 cotton thread.
The ligature was tied around the arch rather than directly
around the artery, thus preventing loss of fluid due to
leakage at the cut end of the arch. The efferent artery
was cannulated in the same manner and the gill was

immersed in 50ml of 1% Ringer solution.

Cirgulatggy Agatomy

The gill was first perfused with Ringer solution
(Appendix) containing the test substance (control,
epinephrine, acetylcholine) in the required concentration,
or with one of the experimental solutions used in the sodium
uptake studies. This perfusion was carried out with the
free end of the efferent cannula at approximately the same
height as the gill, and was continued until no more blood
could be seen entering the collecting tube. The volume
collected, however, was never less than lOQpl. At the end
of this initial perfusion, the free end of the efferent
cannula was raised to a height of 20cm above the gill and

21
was inserted into another 10Qpl pipette. The valve on the
perfusion apparatus was turned and the india ink (Appendix),
which contained the test substance or was diluted with the
experimental solution, was allowed to flow through the gill
until at least lOQpl of ink solution had been collected
from the efferent cannula.

Additional ligatures were placed around the ends of
the gill and were tightened as soon as the cannulas were
removed, reducing the loss of ink from the branchial
‘vasculature to a minimum. Fixation was carried out over
night in Dietrich's fixative (Appendix), and the gill was
dehydrated, infilatated with and embedded in Paraplast
(Arthur H. Thomas 00., Philadelphia, Pa.) using standard
procedures. The gills were sectioned at 8p, the sections
were mounted on glass microscope slides and stained using
standard procedures for either hematoxylin and eosin or
Masson's trichrome stains. Only gills from which there was
no visible leakage of india ink were prepared for

histological examination.
Sodium Movement

After cannulation, the gills were placed in 50ml of
1% Ringer solution and perfused with 100% Ringer solution
until no more blood was observed entering the collecting
tube. As in the anatomical studies, the free end of the
outflow tube was at approximately the same level as the gill

and the perfusion was continued until at least 10Qpl of

22
fluid had been collected from the outflow tube.

At the end of this initial perfusion, the free end of
the outflow tube was raised to about 20cm above the gill,
which was placed in 50ml of fresh 1% Ringer solution. A
clean lOQpl pipette was placed over the end of the outflow
tube and the stopwatch started. The time required to
collect lOQpl of fluid was recorded. A 5Qpl aliquot of the
fluid collected was diluted 1:200 for sodium analysis, and
a 5ml sample of the bathing medium was also placed in a
vial for sodium analysis. The valve on the perfusion
apparatus was turned, and a lOQpl sample of the experimental
fluid was collected from the efferent cannula and discarded.
The gill was then placed in 50ml of fresh 1% Ringer solution,
a clean lOQpl pipette was placed over the free end of the
outflow tube, and the time required for the collection of
100ul of fluid was determined and recorded. Samples of the
bathing solution and fluid collected from the outflow tube
were again taken for sodium analysis.

In addition to the four samples already mentioned,

a 5ml sample of the bathing solution stock solution and
5Qpl aliquots of the control and experimental perfusion
fluids were also taken for sodium analysis. Both the
perfusion fluid (fluid which had not yet passed through the
gill, P) and the fluid collected after passing through the
gill (F) were diluted 1:200 for sodium analysis. Since the
bath fluid had a very low sodium concentration, samples

were only diluted 1:5 for sodium analysis.

23
Sodium Analyaia

Sodium analysis was performed using a Coleman Model
21 Flame Photometer and a Coleman Model 22 Galv-o-meter
galvanometer (Coleman Inst. 00., Maywood, Ill.). A11
dilutions were made using 0.02% Sterox SE (Scientific
Products, Evanston, Ill.) in distilled water. Sodium
concentration was read as percent transmission and converted
to mEq/l using a standard curve. As a constant check on the
accuracy of the readings, the solutions were read in the
following order: blank, l50mEq/1 standard, blank, sample,
blank, etc. The blank also served as a cleaning solution,
washing any remaining sodium out of the aspirator and
burner assembly of the flame photometer.

The sodium concentrations of the perfusion fluid (P)
and the fluid collected after passing through the gill (F)
were determined directly from the standard curve. Since
bath solutions were diluted 1:5 rather than 1:200, the
sodium concentrations determined form the standard curve
had to be divided by %0 to determine the actual sodium
concentration in the bath. The accuracy with which the
sodium concentration of a given sample could be determined

was,:l.0mEq/l.

Uptake 9f Sodium agd Ingipitor.§tggia§

To stimulate sodium uptake, gills were perfused with
Ringer solution in.which 50% of the sodium had been replaced

2%
by choline (choline Ringer solution, Appendix) and with
Ringer solution in which 50% of the sodium (as NaCl) had
been replaced by sucrose (sucrose Ringer solution,
Appendix).
The inhibitors studied (ouabain, cyanide, and

LIM in

iodoacetate) were all used at concentrations of 10'
choline Ringer solution. Atropine (10'3M and lO'hM)'was
also used in some experiments in an attempt to block any
acetylcholine-like effects of the choline in the Ringer
solution. Perfusions were also performed using sucrose
Ringer solution containing 10'3M atropine and served as

controls on any effects of atropine in addition to its

anti-cholinergic effects.

Statistica

All statistical analyses in this study were performed
using nonparametric statistical tests. These tests have a
number of advantages of parametric tests, such as Student's
t test. The assumptions associated with nonparametric tests
are fewer and less critical than those associated with
parametric tests. Nonparametrics can be used with very
small sample sizes (N56) and the tests are generally much
more rapidly and easily performed.

If all of the underlying assumptions for parametric
statistical analysis are met £11219 independent observations,
normal distribution, equal variances, additivity of

effects), than the parametric test is more powerful. That

25
is, it takes fewer observations to achieve a given
significance level with the parametric test than with its
nonparametric counterpart. If, as in the present study,
the differences observed are large enough to show
significance using nonparametric tests, the greater power of
the parametric test is of no particular advantage.

The three tests used in this study were: the Sign
test, which was used to compare related samples when N was
large, theTflalsh test, which was used to compare related
samples when N was small, and the MannAdhitney U test,
which was used to compare independent samples. The first
two tests correspond to the parametric paired t test and

the third corresponds to the unpaired 3 test.

RESULTS

Circulatory tom

The results of the studies of the circulatory anatomy
of the rainbow trout gill are presented in figures %
through 26. The orientation of the gills in all figures is
approximately the same, with the afferent artery at the
top or right-hand side of the figure.

Figure % is a diagrammatic representation of the
major circulatory paths from the ventral to the dorsal
aorta. Figure 5 is a diagrammatic representation of a
longitudinal section of a gill filament (a cross-section
of a gill arch) showing the various paths by which blood
can flow from the afferent to the efferent filamental artery.
Figure 6 shows a diagrammatic cross-section of a filament
from a rainbow trout gill. The major vascular connections
are also shown in figures 7 through 9 which are photo-
micrographs of control gills perfused with india ink.
Figure 7 shows the ink in the afferent filamental artery,
the lamellar-afferent arteriole, part of the "preferred
path" along the lamellar margin, the filamental sinus, and
the right afferent collateral. In Figure 8 the ink is again
present in the filamental sinus and the "preferred path"

but it can also be seen in the lamellar lacunae, the

26

27

Figure %
The major circulatory paths from the ventral to the dorsal

aorta.

28

: ousmfih

 

 

 

 

 

 

 

 

 

mallow ’mbesm) #506
T \ m .zrvfm.

’0 7)U¢0.vfi
15.3%

\‘m.-.
\‘N
\...

    

 

 

 

 

 

 

1 ATI rsw’Lm

’07.? mep
V, +cwyouum

m4som ,mnsow

 

 

 

 

 

T %

 

 

29

Figure 5
Longitudinal section of a gill filament.

9“"9‘”
branchTaT l

ar'AErW q’

 

2(‘evem+ “‘\
EulamenTQT \ tI ‘,
Grier\\ , ".l'
O,/a§§cren1’
0 branchiaj
It,"
In'
I!

a")! V\'

23$?cren‘t xdameniaT
arterT

lamella

 

Figure 5

Figure 6
A diagrammatic cross-section of a gill filament from a

rainbow trout.

32

 

Figure 6

33

smaller-efferent arteriole, the efferent filamental artery
and in the left efferent collateral. Although there is
little ink in the section shown in Figure 9, two important
features of lamellar circulation are apparent. The first
is the "preferred path" which can be seen along the upper
edge of the lamella and the second is the close contact
between the pillar cells and the red blood cells in the
lamellar lacunae.

Figures 10, 11, and 12 are photomicrographs showing
the lamellar-afferent, lamellar-efferent, and sinus—efferent
blood vessels respectively. As can be seen from Figures
10 and 11, the lamellar-afferent and lamellar -efferent
arterioles definitely contain muscular elements within
their walls. It has not been possible to clearly
demonstrate the presence of contractile elements in the
walls of the sinus-efferent vessels (Fig. 12), but neither
the dimensions of the vessels themselves nor the thickness'
of the vessel walls precludes such a possibility.

The photomicrograph in Figure 13 is the result of
accidentally inserting the afferent cannula into what was
apparently the branchial lymph duct rather than the afferent
branchial artery. It is a representative section from this
tissue showing the ink in the filamental sinus and afferent
and efferent collaterals, but with no ink in the afferent
filamental artery or the lamella and little or no ink in the
efferent filamental artery. Ink was also absent from the

afferent branchial artery, there was, however, some ink in

3%
the efferent branchial artery.

Figures 1% through 18 are photomicrographs of serial
sections and show the direct connection from the afferent
filamental artery to the afferent collateral and then to the
filamental sinus. Such connections are only evident near
the base of the filament.

The sections shown in Figures 19 and 20 are from gills
perfused with india ink containing acetylcholine and
epinephrine respectively. They represent the extremes of
fluid flow patterns through the gill. The ink in the
acetylcholine treated gill is concentrated in the filamental
sinus and the collateral vessels with very little ink
appearing in the "preferred"channel or the lamellar lacunae.
In the epinephrine treated gill, on the other hand, the ink
is concentrated in the lamellae with little or no ink in the
filamental sinus or collateral vessels.

Figures 21 through 25 are photomicrographs of cross—
sections of gills perfused with india ink in choline Ringer
solution (Fig. 21), sucrose Ringer solution (Fig. 22),
choline Ringer solution containing cyanide (Fig. 23),
choline Ringer solution containig iodoacetate (Fig. 2%),
and choline Ringer solution containing ouabain (Fig. 25).

By comparing these figures with Figures 19 and 20, it can be
seen that the pattern of ink distribution in these
experimental gills resembles that seen in acetylcholine
treated gills much more closely than it does that seen in

epinephrine treated gills.

35

Figure 7
Photomicrograph of the afferent end of a cross-section of a
filament from a rainbow trout gill which had been perfused
with india ink.

(Masson's trichrome x 53%)

Figure 8
Photomicrograph of the efferent end of a cross section of a
filament from a rainbow trout gill which had been perfused
with india ink.

(Masson's trichrome x 53%)

36

 

Figure 8

37

Figure 9
Photomicrograph of a longitudinal section of a gill lamella.

(Masson's trichrome x 1066)

Figure 10
Photomicrograph of a longitudinal section of a Bill filament,
showing the lamellar—afferent arterioles.

(Masson's trichrome x 1066)

38

 

Figure 10

39

Figure 11
Photomicrograph of a longitudinal section of a gill filament,
showing the lamellar-efferent arterioles.

(Masson's trichrome x 1066)

Figure 12
Photomicrograph of a longitudinal section of a gill filament,
showing the sinus-efferent vessels.

(Masson's trichrome x 1066)

%0

 

Figure 12

%1

Figure 13
Photomicrograph of a cross-section of a gill filament in
which the lymphatics had been perfused with india ink.

(Masson's trichrome x 53%)

Figure 1%
Photomicrograph of a cross-section of a gill filament,
showing the beginnings of the connection from the afferent
filamental artery to the afferent collateral and the
filamental sinus.

(Masson's trichrome x 266)

%2

 

Figure 13

 

Figure 1%

H3

Figure 15
Photomicrograph of the section serially succeeding that in
Figure 1%.

(Masson's trichrome x 266)

Figure 16
Photomicrograph of the section serially succeeding that in
Figure 15.
(Masson's trichrome x 266)

%%

'. "\‘leg.’ I
-‘{jr=;‘ . ..
T g. . . N.
W'i'ti‘l‘fialg

Figure 15

 

      

“A ‘:-‘ v . ‘ ‘ g '7
v ‘ .’.;r,‘ ‘
"It.” . (7.53.: a,

Figure 16

%5

Figure 17
Photomicrograph of the section serially succeeding that in
Figure 16.

(Masson's trichrome x 266)

Figure 18
Photomicrograph of the section serially succeeding that in
Figure 17.

(Masson's trichrome x 266)

%6

 

Figure 18

“7

Figure 19
Photomicrograph of a cross-section of a gill filament which
had been perfused with india ink containing acetylcholine.

(Masson's trichrome x 53%)

Figure 20
Photomicrograph of a cross-section of a gill filament which
had been perfused with india ink containing epinephrine.

(Masson's trichrome x 266)

%8

 

Figure 19

 

Figure 20

%9

Figure 21
Photomicrograph of a cross-section of a gill filament which
had been perfused with india ink in choline Ringer solution.

(Masson's trichrome x 266)

Figure 22
Photomicrograph of a cross-section of a gill filament which
had been perfused with india ink in sucrose Ringer solution.

(Masson's trichrome x 266)

50

 

Figure 21

 

Figure 22

51

Figure 23
Photomicrograph of a cross-section of a gill filament which
had been perfused with india ink in choline Ringer solution
containing cyanide.

(Masson's trichrome x 266)

Figure 2%
Photomicrograph of a cross—section of a gill filament which
had been perfused with india ink in choline Ringer solution
containing iodoacetate.

(Masson's trichrome x 266)

 

52

 

Figure 23

 

Figure 2%

53

Figure 25
Photomicrograph of a cross-section of a gill filament which
had been perfused with india ink in choline Ringer solution
containing ouabain.

(Masson's trichrome x 266)

Figure 26
Photomicrograph of a cross-section of a gill filament which
had been perfused with india ink in choline Ringer solution
containing atropine.

(Masson's trichrome x 266)

 

Figure 26

55 A
Figure 26 shows the distribution of india ink in
gills perfused with choline Ringer solution containing
10'3M atropine. The high concentration of ink in the
lamellae and its virtual absence from the filamental sinus
and collaterals resembles the pattern seen with epinephrine
perfusion rather than that seen.when gills were perfused

with india ink containing acetylcholine.

Sodigm Movemegt

The results of the experiments on sodium uptake are
presented in Table l which gives the concentrations (mEq/l)
of the perfusion fluid (P) and of the fluid collected after
passing through the gills (F), as well as the rates of
sodium uptake (qu/min). The most rapid uptake of sodium
occurred when gills were perfused with sucrose Ringer
solution. Perfusion with choline Ringer solution resulted
in a significantly (p=.025) slower uptake of sodium compared
to the rate of uptake during perfusion with sucrose Ringer
solution. No significant net uptake of sodium was observed
when the gills were perfused with Ringer solution containing
a normal concentration of sodium (158mEq NaI/l). With all
of the other perfusion solutions tested, both the amount and
the rate of sodium uptake were significantly (p(}10)
greater than zero.

The rates of sodium uptake by gills perfused with
choline Ringer solution containing ouabain or cyanide were

86% and 65% lower respectively than the rates of uptake by

56

Table 1
Sodium concentrations of the perfusion fluid (P) and the
fluid collected after passing through the gill (F), and
the rate of sodium uptake for the various experimental

perfusions.

57

uma.o«wmm.o
Hwo.OHwHN.o
ouo.oummm.o
:ao.onaom.o
mmo.onomo.o
uao.on:mo.o
o:a.onm:m.o
amo.onwmm.o
wmo.ouum:.o
-T---.ooo.o

seemmfl.oxmuab
no uhom

n.mfl:.uma

o.:um.uo
s.muc.mm
e.auw.mo
H.mflm.ms

m.dflm.mm

m.muw.uoa

n.ano.os

u.muw.moa
m.ou:.mma

*smmflw

w.afl:.oma

m.ou«.sa
o.ou:.aw
m.anm.mw
a.aflm.:m
m.aflm.aw
o.onw.mm
m.oum.am
o.ono.mm

m.ouw.mma

*emmflm

H taste

2

CHE\+NZ g‘ ***
H\ mz uma a*
henchmen mHHau mm

genes: *

mnfinzaocaao 2mm.o + ammsfim gooa

mafiaonpw EJTOH +
enamonpm Emuoa +
mumpoomocofi +
ocacmmo +

cfimnmso +

ocHQoupm EMTOH +

nmmnfim
newcam
nomcfim
nowawm
nmmcam
ammnam
newcam

nommam

osaaoso
ocaaono
ocaaono
osaaoho
ouaaoeo
ouaaoso
omouosm

omonosm

nomcfim Rooa

coausaom

58
gills perfused with choline Ringer solution which contained
no inhibitor. These differences were significant at the
=.025 and p .10 levels respectively. The rate of sodium
uptake by ouabain perfused gills was significantly less
(p<.l5) than the rate of uptake by cyanide perfused gills.
The rates of sodium uptake by gills perfused with

choline Ringer solution containing iodoacetate or 10'3M

or lO’hM atropine were not statistically different from the
rates of uptake by gills perfused with choline Ringer
solution alone. Addition of atropine (lo-BM) to sucrose
Ringer solution did not result in a statistically
significant change in the rate of sodium uptake.

The rates of fluid flow through the gills are given in

Table 2 which compares the flow rates during perfusion with
100% Ringer solution (control perfusion rate, R1) to the
flow rates during perfusion with the various experimental
solutions (experimental perfusion rate, R2). ‘When the
experimental solution was the same as the control solution
(100% Ringer solution), there was a significant (p(.001)
increase in the flow rate. There was also a significant
(p=.03l) increase in the flow rate when the experimental
solution was 100% Ringer solution containing epinephrine
(0.27M). This increse was significantly greater (p(.01)
than when the experimental solution was 100% Ringer
solution without epinephrine. There was no significant
change in flow rate with the sucrose Ringer solution alone,

with sucrose Ringer solution containing 10'3M atropine, or

 

59

Table 2
The rates of fluid flow through the gill during control (R1)
and experimental (R2) perfusions and the percent change

in rate from the control to the experimental perfusion.

60

.mosam> omwno>m one so no: .mpwc HwGHMHuo co comma has

GHE\H3 **
comsmnom madam mo nonasc *

m.osno.uea+ o.mnxamm u.m«x.mm m ocansdosado 25m.o + somcam mooa
u.m HH.AH n o.mnd.mm H.mMN.wm m maggoaum 2:Toa + nomcfim mafiaonu
0.:Hum.a - H.muo.mm o.mum.mm m odanouam ZmTOH + somsam osaaoho
a.maflo.am . a.:u4.wa w.mfla.am a onshoomoooa + tomcam ohaaono
m.: “4.:5 T m.HHN.m m.NHw.mH m ocHCmmo + ummcfim ocfiHoso
:.Nafli.ww I u.aflm.: :.:flv.ma m Cfiwnmso + nmmcfim mafiaoso
m.u flw.mm T m.aflm.u N.mflm.mm wa nomcfim mafiaoso
w.m HM.H u N.mflf.:m u.uflN.mm m mcfiaoupm EMTOH + nomcfim omonosm
m.wafl4.m u w.mflw.mm :.:HH.Nm w nomcwm mmonozm
:.: HM.AH + m.mum.wm w.mflu.:m mm humane mooa
.ssmmnumcsnos *smmumm simmn.m *z coaosaom

N magma

,
A.
r-
O
~nv

61
with choline Ringer solution containing 10'3M or lO'hM
atropine. The flow rates of all of the other experimental
perfusions were significantly less (p(.10) than the control
rates.

The rate of sodium uptake by gills perfused with
choline Ringer solution was plotted as a function of gill
weight, and flow rate. There appeared to be a slight
decrease in the rate of sodium uptake with increasing gill
weight and no change as a function of flow rate. There was

also no relationship between the rate of sodium uptake and

the time, during the study, when the experiment was

performed.

 

DISCUSSION
Circ tor Anatomy

According to Steen and Kruysse (196%), the two main
paths by which blood can pass through the filament of a
teleost gill (the lamellar path and the filamental sinus
path) were first described by Riess (1881). It was also
Riess (1881) who suggested that the anatomical and cyto-
logical nature of the vessels within the filamental sinus
and of the sinus itself, would classify them as lymph
vessels rather than as blood vessels. The investigations
by Steen and Kruysse (196%), which involved primalily the
common eel (Anguilla yglgagia), supported and expanded
these earlier studies. They found the direct connections
between the filamental sinus and the afferent and efferent
filamental arteries. They also pointed out that there is a
direct connection from the sinus to the major lymph vessel
at the base of the filament, which provided further evidence
for the lymphatic nature of the filamental sinus.

Evidence is presented herein that, in the rainbow
trout, as opposed to the eel, the lymphatics (or afferent
collaterals) are directly connected to the filamental sinus
at frequent intervals along the filament. These lymph

vessels are also connected directly to the afferent

62

 

63
filamental artery near the base of the filament. The
presence of numerous red blood cells in the filamental
sinus and the afferent collaterals (or lymphatics) and
their virtual absence from the major lymph vessels at the
base of the filament, indicates that there must be some
sort of filtering mechanism whereby the red cells are
removed from the lymph and returned to the blood. The
precise nature and location of this filtering system are
unknown.

The presence of such a well developed lymphatic-
blood vascular system would seem to be advantageous in the
maintenance of the ionic and osmotic composition of the
blood in both marine and freshwater fish. In both
environments, the large filamental sinus would serve to
decrease the amount of exposure between the blood and the
surrounding environment. This would decrease the amount of
salt lost and water gained by freshwater fish and the
amount of salt gained and water lost by marine fish. It is
also possible that the branchial lymphatic system acts as a
blood volume buffer during acute osmotic stress in both
freshwater and marine teleosts. In the freshwater fish,
the branchial lymphatics undoubtedly serve to drain off
some of the excess fluid formed by the osmotic uptake of
water across the gill epithelium. In the marine teleost,
the flow of fluid might concievably be from the lymphatics
into the blood, tending to restore any fluid which had

been lost across the gill epithelium.

6%

In their studies of the pattern of blood flow through
the gill, Steen and Kruysse (196%) placed freshly excised
gill filaments into physiological salt solutions on a
microscope slide. They then placed a cover slip over the
filament and observed the patterns of blood flow when
pressure was applied to the cover slip. Using this method,
they found that in untreated gills blood flowed, often
simultaneously, between the afferent and the afferent
filamental arteries by way of the lamellae, by way of the
filamental sinus and directly by way of the connection
between the afferent and efferent filamental arteries at
the tip of the filament. ‘When acetylcholine was added to
the salt solution bathing the filament, they found that the
blood flowed through the filamental sinus and around the tip
of the filament. Addition of adrenalin, on the other hand,
caused all of the blood to flow through the lamellae. The
absence of a normal, unidirectional, pressure gradient from
afferent to efferent, however, makes it difficult to
determine the true physiological significance of the flow
patterns obtained in these experiments.

In the present study, using approximately normal
pressure gradients from afferent to efferent, and with the
acetylcholine or epinephrine in the perfusing fluid rather
than applied to the outside of the gill, the results
obtained were similar to those of Steen and Kruysse (196%).
The effects of epinephrine and acetylcholine, however, were

not as absolute as those reported by these investigators.

65
Perfusion with india ink containing acetylcholine, for
example, did not completely eliminate the flow of fluid
through the lamellae, and perfusion with epinephrine did
not result in 100% lamellar flow.

Since there was very little difference between the
india ink distribution in the control gills and that in
acetylcholine treated gills, it would appear that the
pattern of blood flow through the gills is primarily under
adrenergic control. The effect of atropine on the flow
pattern through gills perfused with choline Ringer solution,
however, indicates that there is probably also some tonic
cholinergic control of gill blood flow. Thus, the pattern
of blood flow through the rainbow trout gill seems to vary
from "purely cholinergic" flow (exclusively through the
filamental sinus) to "purely adrenergic" flow (exclusively
through the lamellae.

The similarity between the india ink distribution in
the gills perfused with sucrose Ringer solution and that in
the control gills, indicates that a low internal sodium
concentration does not tend to favor lamellar perfusion.

No conclusions can be drawn with respect to the effects of
cyanide, iodoacetate or ouabain on the pattern of fluid
flow through the gills. This is true since any effects
which they might have were probably blocked or overshadowed
by the cholinergic effect of the choline Ringer solution.

In his studies on the juvenile coho salmon

(Ogaorgynchus kisutch) and seven other species, Newstead

66
(1967) was unable to find any structural specialization in
the walls of the afferent or efferent arteries which could
account for the variations seen in the pattern of blood
flow through the gills. Based on this apparent absence of
arterial control sites, and on electron microscopic studies
of the pillar cells, he suggested that the pattern of blood
flow through the teleost gill is regulated by active
alteration of the cross-sectional area and thus the
resistance of the lamillar lacunae (an hypothesis previously
advanced by Hughes and Grimstone, 1965). There are two
major objections to accepting this as a general theory.
They are: (l) the fact that the arterioles which branch
off of the afferent filamental artery and those which enter
the effernet filamental artery do possess muscular elements
within their walls (at least in‘fialgg‘gaigggazi), (2) the
very large changes in lamellar resistance which must exist
in order to produce the observed changes in the pattern of
fluid flow through the gill.

The following diagram is a simplified schematic
representation of the circulatory paths through the gill,
showing the apical path (A), the lamellar path (L), and the
sinus path (S). As can be seen from this diagram, the paths

 

67
are arranged in parallel and therefore, the amount of flow
through a given path will depend not only on the resistance
of that path, but also on the resistance of the other paths.
If, as is suggested by Newstead (1967), the lamellar
resistance (L) is the only variable resistance in the
circuit, then it must by able to vary from near zero to
infinity, with respect to the sinus (S) and apical (A)
resistances in order to change the circulatory pattern from
completely lamellar to completely sinus flow. If, on the

other hand, there are two variable resistances in the

w.— 0” u-"~ “~. 5 on m.--w-vdv ‘-'~ "" -" - '1‘ Sc-“

 

 

 

%~ cm

circuit, than much smaller changes in the sinus and the
lamellar resistances could result in the observed changes
in flow pattern.

Thus, the anatomical sites for the control of the
blood flow pattern are probably some combination of the
lamellar-afferent and lamellar-efferent arterioles and the
sinus-afferent and sinus-efferent vessels. Acetylcholine,
since it not only causes blood to flow through the
filamental sinus, but also decreases the flow rate (Estlund
and FSnge, 1962), probably acts solely to cause vaso-
constriction of the lamellar-afferent and lamellar-efferent
arterioles. Epinephrine, which stimulates blood flow
through the high resistance lamellar circulation, also

increases the flow rate. Thus, epinephrine must cause both

68
vasoconstriction of the sinus-afferent and sinus-efferent
vessels and vasodilation of the lamellar-afferent and the

lamellar-efferent arterioles.

__.12_30d I! mm

The results of these experiments clearly indicate
that the isolated, perfused gills of rainbow trout are
capable of taking up sodium against a concentration gradient
of at least 60:1. The differences between the rates of
sodium uptake by gills perfused with sucrose Ringer
solution and by gills perfused with choline Ringer solution
may have been brought about by: a stimulatory effect of
sucrose of sodium uptake, an inhibitory effect of choline on
sodium uptake or, an interaction between the uptake of
chloride and that of sodium. The first of these
possibilities seems unlikely since, to the best of my
knowledge, there is no evidence for any direct effect of
sucrose on the transport of sodium by membranes in general
or by fish gills in particular. The studies in which gills
were perfused with choline Ringer solution containing
atropine, suggest that any inhibitory effects that choline
might have on sodium uptake by the gill are either very
slight or are particularly insensitive to inhibition by
atropine. This is particualrly apparent when the effect
of atropine on flow rate is compared to its effect on sodium
uptake. Atropine, at a concentration of lO‘hM blocked the
effects of choline Ringer solution on the rate of fluid

69
flow through the gill, but the rate of sodium uptake was not
increased even when the concentration of atrOpine was
raised to 10'3M. Thus, any inhibitory effects of choline on
the rate of sodium uptake are probably insignificant. ‘With
respect to the third possibility, an interaction between the
uptake of chloride and the uptake of sodium, Krogh (1938)
found that either sodium or cholride could be taken up
independently by goldfish (Caraaaiga aggatga). ‘When this
occurred, however, the chloride had to exchange for
bicarbonate and the sodium had to either exchange for
ammonium or be accompanied by bicarbonate. This theory has
since been supported by Maetz and Romeu (196%) and by
Romeu and Maetz (196%). Since no source of ammonium ions
was supplied in the perfusion fluid and no bicarbonate was
added to the bathing solution, most of the sodium taken up
by gills being perfused with choline Ringer solution was
probably taken up as sodium chloride. Lesser amounts may
have exchanged for endogenously available ammonium ions or
may have been tsken up in combination with bicarbonate ions
derived from the carbon dioxide present in the bathing
solution. However, any uptake of sodium chloride would
tend to increase the chloride level of the perfusing fluid
above the original value. An increased chloride
concentration.would tend to increase the passive loss of
chloride ions, but since there is little bicarbonate
available in the bath for exchange, maintenance of

electrical neutrality requires that the chloride loss be

70
accompanied by an equivalent loss of cations, probably
sodium ions. This sodium-chloride interaction would tend
to decrease the rate of sodium uptake by gills being
perfused with choline Ringer solution.

In sucrose Ringer solution, however, both the sodium
and the cholride concentrations were below normal blood
levels. This would not only decrease the passive losses
of sodium as described above, but might actually result in
an increase in the rate of sodium influx, due to sodium
ions "passively" following the actively transported chloride
ions. In either case, the overall effect would be an
increase in the net uptake of sodium ions by the gill.

The inhibition of sodium uptake by ouabain and
cyanide, indicates that sodium uptake by isolated, perfused
rainbow trout gills is an ATP-dependent process and that
much of the ATP used is derived from oxidative metabolism.
The complete lack of inhibition of sodium uptake by
iodoacetate suggests that glycolysis is not required for
sodium uptake by the gill. These results, along with those
of Kamiya (1967) using salt water adapted eels, suggest that
glycolysis may generally be unimportant as a metabolic path
in the gills of fish. As yet, no attempt has been made to
discover what other paths of energy production may be of
significance in this tissue.

Comparisons of the rates of sodium uptake by gills
perfused with choline Ringer solution, sucrose Ringer

solution, and 100% Ringer solution containing epinephrine

71
indicate that the rate of sodium uptake is not primarily
dependent upon either flow rate or flow pattern throuth

the gill.

CONCLUSIONS

1. The filamental lymphatic system connects directly to
the blood circulatory system at frequent intervals
along the filament.

2. The patterns of filamental blood flow found in gills
perfused using a normal pressure gradient vary from
mostly sinus flow ("purely cholinergic") to mostly
lamellar flow ("purely adrenergic").

3. The pattern of blood flow through the gill is probably
regulated by vasoconstriction and vasodilation of the
lamellar-afferent and lamellar-efferent arterioles and
the sinus-afferent and sinus-efferent blood vessels.

%. Isolated, perfused rainbow trout gills are capable of
taking up sodium against a concentration gradient of
at least 60:1.

5. Sodium uptake by gills is at least partially dependent
upon the availability of other cations in the perfusion
fluid for which the sodium can exchange or the avail-
ability of anions in the bathing medium which can
accompany the transported sodium.

6. Sodium uptake by isolated, perfused rainbow trout gills
is ATP dependent and the ATP used is derived mainly

from oxidative metabolism.

72

73
The rate of sodium uptake is not primarily dependent
upon the pattern or rate of fluid flow through the
gill and is probably a function of the internal sodium

concentration.

APPENDIX

75
SOLUTIONS USED

Camposition.gf 109% Rigger Sglutiog

NaCl 7.37 gm/liter
KCl 0.31 gm/liter
CaCl2 0.10 gm/liter
Mgsog 0.1% gm/liter
KHQPOg 0.%6 gm/liter
Na2HPOh 2.02 gm/liter
Glucose 0.90 gm/liter

Composition 9; CQoline Riggar Solution

Choline Cl 6.91 gm/liter
NaCl 3.85 gm/liter
KCl 0.31 gm/liter
CaC12 0.10 gm/liter
MgSOu 0.1% gm/liter
KH’ZPO)+ 0.%6 gm/liter
NaZHPOh 2.02 gm/liter
Glucose 0.90 gm/liter
Compositian12f Spagoaa.§inga;.§glptign
Sucrose 23.96 gm/liter
NaCl 3.85 gm/liter
KCl 0.31 gm/liter
CaC12 0.10 gm/liter
MgSOh 0.1% gm/liter
KH2P0h 0.%6 gm/liter
NaZHPOg 2.02 gm/liter

Glucose 0.90 gm/liter

76
All of the Ringer solutions used were adjusted to
pH 7.% and 300m03M L: SmOSM).

Dietrich's Fixative

Tap‘Water _ 150 ml

95% Ethanol 75 ml

h0% Formalin 25 m1

Glacial Acetic Acid 5 ml
India Ink

The india ink used was Pelikan Biological India
Ink and was obtained from the John Henschel Co., New York.
According to Peterson, Ringer, Terzaloff and Lucas (1965),
this ink is 10% carbon with a particle size of 0.02 to 0.03».
It contains #.3% fish glue, 1.0% phenol and none of the
shellac or ammonia normally found in other india ink
preparations. The sodium content of the ink was adjusted
to equal that of 100% Ringer solution and the resulting

osmotic pressure was 30% mOSM.

LITERATURE CITED

Bateman, J. B. and A. Keys. 1932. Chloride and vapour-
pressure relations in the secretory activity of the
gills of the eel. J. Physiol., 75: 226-2k0.

Bevelander, G. J. 19H5. A comparative study of the branchial
epithelium in fishes, with special reference to
extrarenal excretion. J. Morphol. 57: 335-397.

Epstein, F. H., A. I. Katz, and G. E. Pickford. 1967.
Sodium- and potassium-activated adenosine
triphosphatase of gills: Role in adaptation of
teleosts to sea water. Science, 156: 3779.

Goodrich, E. S. 1958. Studies on the structure and
development of vertebrates. Vol. II. Dover
Publications Inc., New York.

Gordon, M. S. 1963. Chloride exchange in rainbow trout
(Salmo aird1 21) adapted to different salinities.
Biol. Bull. :MS-W

Hughes, G. M., and A. V. Grimstone. 1965. The fine
structure of the secondary lamellae of the gills of

G d Bollgchiug. Quart. J. Microscop. Sci.
10 = 3 3-3 3.

Kamiya, M. 1967. Changes in ion and water transport in
isolated gills of the cultured eel during the
course of salt adaptation. Annot. Zool. Jap.
#0: 123- 129.

Keys, A. 1931. The heart- gill preparation of the eel and
its perfusion for the study of a natural membrane
in sit . Z. VergL Physiol., 15: 352- 363.

Keys, A. 1933. The mechanism of adaptation to varying
salinities in the common eel and the general problem
of osmotic regulation if fishes. Proc Roy. Soc.
London, Ser. B., 112: 18h-199.

Keys, A. 1937. Properties of the gill membranes of fishes.
Trans. Faraday Soc., 33: 972-98h.

\l
“J

78

Keys, A. and J. B. Bateman. 1932. Branchial responses to
adrenalin and to pitressin in the eel. Biol. Bull.,
63: 327-336.

Keys, A. and E. N,‘Willmer. 1932. "Chloride secreting cells"
in the gills of fishes, with special reference to
the common eel. J. Physiol. 76: 368-378.

Krogh, A. 1937. Osmotic regulation in fresh water fishes
by active absorption of chloride ions. Z. Vergl.
Physiol., 2%: 656-666.

Krogh, A. 1938. The active absorption of ions in some
freshwater animals. Z. Vergl. Physiol., 25: 335-350.

Maetz, J. 1953. Action des sulfamides sur la chlorémie
du Serran. Compt. Rend. Soc. Biol., 197: 291-295.

Maetz, J. 1956. Le role biologique d9 l'anhydrase
carbonique chez quelques teleosteens. Bull. Biol.
de France et de Belgique. Suppl. #0: 1-169.

Maetz, J. and F. G. Romeu. 196%. The mechanism of sodium

uptake by the gills of a freshawater fish Qa§ggsig§
auratug, II. Evidence for NHh /Na and HC03‘ Cl'
exchanges. J. Gen. Physiol., H7: 1209-1227.

Mott, J. C. 1950. Radiological observations on the

cardiovascular system in Anguilla anguilla. J. Exp.
Biol., 27: 32h-332.

Mott, J. C. 1951a. The gross anatomy of the blood vascular
system of the eel ui 8.22321113- Proc. 2001.
Soc., London. 120: 3- .

Mott, J. C. 1951b. Some factors affecting the blood

circulation in the common eel (Aggu1118.aggu;;;§).
J. Physiol. 11“: 387-398.

Newstead, J. D. 1967. Fine structure of the respiratory
lamellae of teleostean gills. Z. Zellforschung.

fistlund, E. and R. Fange. 1962. Vasodilation by adrenaline
and noradrenaline, and the effects of some other
substances on perfused fish gills. Comp. Biochem.

Physiol., 5: 307-309.

Peterson, R. A., R. K. Ringer, M. J. Tetzloff, and A. M.
Lucas. 1965. Ink perfusion for displaying
capillaries in the chicken. Stain Technol. #0: 351-

356-

79

Philpott, C. w. 1965. Halide localization in the teleost
chloride cell and its identification by selected
area electron diffraction. Protoplasma, 60: 7-23.

Pitts, R. F. 193%. Urinary composition in marine fish.
J. Cell. Comp. Physiol., %: 389-395.

Randall, D. J. and E. D. Stevens. 196”. The role of
adrenergic receptors in cardiovascular changes
associated with exercise in salmon. Comp. Biochem.
Physiol. 21: %l5-%2%.

Romeu, F. G. and J. Maetz. 196%. The mechanism of sodium
and chloride uptake by the gills of fresh-water

fish, ngassius snratus. I. Evigsnce fsr independent
unsake.%£ sodium and onloride ions, J. Gen. Physiol.
7: 119 -1207.

Smith, HL‘W. 1930. The absorption and excretion of water
and salts by marine teleosts. Amer. J. Physiol.
93: %80—505.

Smith, H.\N. 1932. Water regulation and its evolution in
fishes. Quart. Rev. Biol., 7: 1—26.

Steen, J. B. and A. Kruysse. 196%. The respiratory function
of teleostean gills. Comp. Biochem. Physiol.,

Stevens, E. D. and D. J. Randall. 1967. Changes in blood
pressure, heart rate and breathing rate during
moderate swimming activity in rainbow trout. J.
Exp. Biol. %6: 307-315.

Virabhadrachari, V. 1961. Structural changes in the gills,
intestine and kidney of Etuontus.ns§nls§ns
(Teleoster) adapted to different sa inities. Quart.
J. Microscop. Sci., 102: 361-369.

 

MICHIGQN STQTE UNIV. LIBRQRIES
IIHI111H1VIIWIWIIWIIH1111111111111111
31293105699361