UPTAKE PATHWAYS OF A
METHYL MERCURIC CHLORIDE
AND MERCURIC CHLORIDE BY '

RAINBOW TROUT (SALMO GAIRDNERI) 7 ' I
WITH SPECIAL REFERENCE TO ‘
ULTRASTRUCTURAL CHANGES

OF THE. GILL

Thesis for the Degree of Ph. 0‘.
MICHIGAN STATE UNIVERSITY
KENNETH R. OLSON
1972 ~

Date

0-7639

”III III III III IIIIIII

93 01068 9143

 

 

 

This is to certify that the

thesis entitled

Uptake Pathways of Methyl Mercuric Chloride and Mercuric
Chloride by Rainbow Trout (Salmo gairdneri) with Spe01al
Reference to Ultrastructural Changes of the G111

presented by

Kenneth Robert Olson

has been accepted towards fulfillment
of the requirements for

Ph.D. degree in Physiology

Major professor

11/6/72

 

 

      
  

Er amomc; av
HBAG & SUNS’
800K BINDERY INC II
LIBRARY Bmoms III
snmmm mcmm I
L=/'\_,."\ "m '_:../J

  
 
 
  

:26st ~

ABSTRACT

UPTAKE PATHWAYS OF METHYL MERCURIC CHLORIDE
AND MERCURIC CHLORIDE BY RAINBOW TROUT
(SALMO GAIRDNERI) WITH SPECIAL REFERENCE
TO ULTRASTRUCTURAL CHANGES OF THE GILL

 

BY

Kenneth R. Olson

The environmental mercury problem has received
considerable attention with respect to its effect as a
human health hazard, however, little is known of the
effects of mercury on aquatic animals. The purpose of
this study was to determine the pathway of mercury uptake
into rainbow trout and the effects of two mercurials,
HgCl2 and CH3HgCl, on the tissues most closely associated

with the uptake process.

Uptake of either 203HgCl2 or CH3203HgCl by trout was
not affected by esophageal ligation thus uptake of these
two mercurials appears to be by way of either the gills or
skin. Low transcutaneous molecular fluxes reported by
numerous investigators support the hypothesis of the gill
as the primary pathway. In 24 hours both mercurials are

found in tissues at concentrations exceeding those of the

environment with gill, liver and kidney showing the

Kenneth R. Olson

greatest accumulation. Uptake of methyl mercury is much

greater than inorganic mercury uptake after 24 hours and
relative tissue distributions show preference for either
mercurial depending on the individual tissue.

Electron microprobe tissue analysis and radioauto-
graphic studies of gill tissue failed to show specific
concentration of HgCl2 at either the respiratory or ion
exchange sites, however, large accumulations of mercury
were found trapped in the mucus coating the gill surface
and.in.the cartilaginous filamental support.

Transmission and scanning electron micrographs

revealed ultrastructural changes following either methyl

or inorganic mercury exposure. Loss of lamellar epithelial

cell ridges, appearance of highly vacuolated epithelial
cells and chloride cell degeneration was common after

exposure to either mercurial. In addition, methyl

mercury specifically produced red cell vacuolation and
exposure to inorganic mercury resulted in appearance of

a new lamellar epithelial cell type.
£2_vitro experiments using radioactive mercurials
demonstrated high affinity of methyl mercury for red cells

(up to 90% of the mercurial was bound to red cells in

40 minutes). Only 9% of inorganic mercury was taken up by

red cells, however, this percentage was increased up to 65%

if the cells were washed and suspended in Ringer solution

Kenneth R. Olson

prior to incubation with mercury. In spite of the affinity
of washed cells for mercury, concentrations as high as
100 ppm of either mercurial had no effect on the oxygen

carrying capacity .

UPTAKE PATHWAYS OF METHYL MERCURIC CHLORIDE
AND MERCURIC CHLORIDE BY RAINBOW TROUT

(SALMO GAIRDNERI) WITH SPECIAL REFERENCE

 

TO ULTRASTRUCTURAL CHANGES OF THE GILL

BY

4/
--I

Kenneth REIOlson

A THESIS

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

DOCTOR OF PHILOSOPHY
Department of Physiology

1972

C: $030,),

Dedicated to my wife, Marilyn,
and daughter, Nicole,
for their understanding, tolerance and sacrifice
during the completion of this dissertation
and for the many words of encouragement and support

along the way.

Dedicated also to my parents,
Mildred and Robert Olson,
for their continual encouragement and concern

which made my entire education possible.

ii

ACKNOWLEDGMENTS

The author wishes to express his sincere appreciation
to Dr. P. O. Fromm for his guidance, suggestions and sup-
port in the preparation of this dissertation.

I am also grateful to Dr. J. R. Hoffert for his
valuable suggestions on many laboratory procedures as well
as help with the photographic preparation of the micro-
graphs in this thesis.

Thanks is also expressed to Harold Bergman, Dan
O'Connor, Dick Walker and other graduate students on this
campus for their assistance, and to my wife, Marilyn, and
Mrs. Esther Brenke for laboratory assistance and aid in
preparation and typing of the rough draft of this thesis.
Special recognition is also given to Mrs. C. Smith for her
interest in environmental problems.

The author is also indebted to Drs. W. Willford and
R. E. Reinert, Great Lakes Fishery Laboratory, Bureau of
Sport Fisheries and Wildlife, for their discussion and
generous supply of mercury exposed fish.

This researchtwas supported by Environmental ProteCe
tion Agency, Grant No. R-801034, and Public Health Service

Training Grant No. HL-05873.

iii

TABLE OF CONTENTS

LIST OF TABLES O O O O O O O 0
LIST OF FIGURES . . . . . . . .
INTRODUCTION 0 o o o o O o o 0
LITERATURE REVIEW . . . . . . .
Mercury Uptake . . . . . .
Primary Uptake Pathway .

(1) Gills . . .
(2) Skin Absorption of
Tract . . . .

Secondary Uptake Pathway
Inorganic Mercury: Hg

Organic Mercury Compounds
Effects of Mercurials on Tissues

Biochemical Effects . . .

Physiological and Morphologica

Gills O O O O O O O 0
Blood . . . . . . . .
smary O O I O O O O O O 0

MATERIALS AND METHODS . . . . .

Experimental Animals . .
Exposure Protocol . .
Radioisotope Methodology
Blood Studies ... . . . .
Hematocrit and Hemoglobin

Mercury
(3) Uptake of Mercury by the

Determinations

1

Alimen

000M000.-

ooomoooooofloooo

oooooooooofloooo

Mercury Uptake by Red Blood Cells (Uptake

Rate 0 O O O O O I O 0
Binding Characteristics .

Oxygen-Carrying Capacity--Effect of CH

and HgCl3 . . . . . .

Apparatus for CO Saturation of Ringer

Solution . . . . .
Blood Incubation . . .

iv

0 “

3

HgCl

Page
vii

viii

us}.
F.

fly.

“*0!

‘5.

urtt‘h

 

‘\
‘-

u -..~

 

APPENDIX II:

Liberation of Oxygen from Red Cells .
Oxygen Tension Determinations and
Calculations . . . . . . .
Gill Histology . . . . . . . . . .
Light Microscopy . . . . . . . .
. Transmission Electron Microscopy
Scanning Electron Microscopy .
Electron-Microprobe . . .
Sample Preparation
Microprobe Analysis
Pathway of Mercury Uptake
Esophageal Ligation .
Ligation Procedure
Radioautography . . .
Species Study . . . . . .

RESULTS AND DISCUSSION 0 O I O O O O O O O O O

Pathway of Mercury Uptake . . . . . . . .
Pathway of Mercury through the Gill . . . .
Mercurial Induced Structural Changes
of the Gill--Light Microscopy . . . . .
Effects of Mercury on Gill Ultrastructure
Ultrastructure of the Normal Trout Gill
Chloride Cells . . . . . . . . . . .
Effects of Methyl Mercury on Gill
Ultrastructure . . . . . . . . . . .
Effects of Inorganic Mercury on Gill
Ultrastructure . . . . . . . . . . .

Uptake and Distribution of Mercury in the Blood

Uptake Rates and Percent Incorporation
of Mercury into Red Cells . . . . . .
Effect of Mercurials on Oxygen Carrying
Capacity of Red Cells . . . . . . . . .

CONCLUSIONS 0 O O O O O O O O O O O O O O O O
MFERENCES O O C I I O O O O O O O O O O O O 0

APPENDIX I : SOLUTIONS O O O C C O O O O O O O

EQUILIBRIUM CURVE . . . . . . .

COMPONENT ANALYSIS OF MERCURY UPTAKE

Page
38

39
40
4O
41
42
44
44
45
46
46
47
50
52

54

54
64

80
81
81
100
120

144
167

168
179
181
185

195

196

Page

APPENDIX III: TRANSMISSION ELECTRON MICROSCOPY
TISSUE EMBEDDING PROCEDURE . . . . . . 201

APPENDIX IV: SCANNING ELECTRON MICROSCOPY
TISSUE PREPARATION . . . . . . . . . . . 203

APPENDIX V: PILLAR CELL STRUCTURE . . . . . . . . . . 205
APPENDIX VI: SURFACE FEATURES OF LAMELLAE FROM

THE BLACK BULLHEAD AND NORTHERN
PIKE C C C O O C O I O C C O O O O O O O 208

vi

Table
l.
2.

3.

LIST OF TABLES

Effect of mercurials on gill structure
Ambient mercury concentrations . . . .

Effect of esophageal ligation on twenty-
four hour uptake of 27S ppt mercury as
either HgCl2 or CHBHgCl . . . . . . . .
Effects of surgical incision on tissue
hYdration O O O O O O O I O O O O O O 0

Effects of sulfhydryl agents on mercury
binding to red cells . . . . . . . . .

Effects of CH3HgCl and HgClz on oxygen
carrying capacity of washed red cells .

vii

Page

23

35

55

57

177

179

9B.

10.

11A.
11B.

12.

LIST OF FIGURES

Position of incision and ligature for
esophageal ligation experiments . . . .

Mercury concentrations in tissues from
trout exposed for 24 hours to eith r
inorganic (Hg++) or organic (CH3Hg )
mercury . . . . . . . . . . . . . . . .

Cross section through a gill arch showing

the relationship between the filaments,

lamellae, and the pathway of water flow

and blood flow . . . . . . . . . . . .
Electron microprobe: Tissue scan . . .
Line scan of lamellar cross section . .
Line scan of filamental cross section .
Tissue scan of lamellar cross section .

Scanning electron micrograph of a pair
Of filaments o o o o o . '0 o o o o o o 0

Scanning electron micrograph of a gill
filament; 50% glutaraldehyde fixation .

Scanning electron micrograph of filament

fixed in 4% glutaraldehyde . . . . . .

Scanning electron micrograph of the
lamellar surface . . . . . . . . . . .

Scanning electron micrograph of lamellar

surface showing short (A) and long (B)
ridges O O O O O O O O O O O O O O O 0
Transmission electron micrograph of a
cross section through the lamella . . .

viii

Page

49

60

66
70
72
77

79

83

87

87

90

92

95

Figure Page

13A. Scanning electron micrographs of the
13B. lamellar surface showing the microridges
of the epithelial cells . . . . . . . . . . . 99

14A. Transmission electron micrograph of a
cross section through a lamella . . . . . . . 102

143. Transmission electron micrograph of
cross section through epithelial cell
ridges O O O O O O O O O r. O O I O O O O O C 102

15A. Transmission electron micrograph of
15B. lamellar epithelial cell junctions . . . . . 104

16. Transmission electron micrograph of a
normal chloride cell on a lamella . . . . . . 107

17A. Transmission electron micrograph of a
mucus cell on the lamellar surface . . . . . 110

173. Transmission electron micrograph of the
apical portion of a chloride cell . . . . . . 110

18. Scanning electron micrograph of the
filamental surface . . . . . . . . . . . . . 113

19. Scanning electron micrograph of several
chloride cells on the filament . . . . . . . 115

20A. Scanning electron micrograph of chloride
cells on filament . . . . . . . . . . . . . . 117

20B. Scanning electron micrograph of chloride
cells on a lamella . . . . . . . . . . . . . 117

21A. Transmission electron micrographs of
213. degenerating chloride cells . . . . . . . . . 119

22A. Scanning electron micrograph of mucus
22B. cells on the filament in922A and on the
lamella in 223 O I O O O O O O O O O O O O I 122

23. Scanning electron micrograph of filament
between lamellae . . . . . . . . . . . . . . 124

ix

Figure

24A.
24B.

25.

26A.
263.

27.

28.

29A.

29B.

30.

31A.

31B.

32A.

Scanning electron micrograph of trout
exposed to 0.05 ppm Hg as (CH 3HgCl) for
2 4 hours 0 O O O C O O O O O O O O O O O 0

Scanning electron micrograph of the
filament between the lamellae . . . . . . .

Scanning electron micrograph of trout
exposed to 0.05 ppm Hg as (CH3 HgCl) for
5 days 0 O O O I O O O O O O O O I O O O 0

Transmission electron micrograph of a
cross section through the lamella from

a trout exposed to 0.05 ppm Hg as (CH3H9C1)
for 5 days . . . . . . . . . . . . . . . .

Transmission electron micrograph of a
cross section through a lamella from a
trout exposed to 300 ppt Hg as (CH 3HgCl)
for 4 weeks . . . . . . . . . . . . . . . .

Transmission electron micrograph of
vacuolated red cells found in the
afferent filamental artery . . . . . . . .

Transmission electron micrograph of a
lamellar cross section from trout

exposed to 300 ppt Hg as (CH 3HgCl) for

4 weeks . . . . . . . . . . . . . . . . . .

Transmission electron micrograph of cross
ections through several lamellae of

trout exposed to 300 ppt Hg as (CH 3HgCl)
for 8 weeks at 5° C . . . . . . . . . . .

Transmission electron micrograph of
epithelial cells of low electron density
on lamellae of trout exposed to 300 ppt

Hg as (CH3HgC1) for 8 weeks at 5° C . . . .

Transmission electron micrograph of a
degenerating chloride cell on the
lamella O O O O O O O O O O O O O O O O O O

Page

126

129

131

133

136

138

138

141

143

146

Figure

33.

34A.

34B.

35.

36A.
36B.

37.

38A.
38B.

39.

40.

41A.

413.

42.

43.

Scanning electron micrograph of
degenerating chloride cells on the
filament O I O O O O O O O O O O O O 0

Transmission electron micrograph of
an epithelial cell devoid of ridges . .

Transmission electron micrograph of
a degenerating chloride cell and
amorphous mitochondria . . . . . . . .

Scanning electron micrograph of the
lamella surface from a trout exposed
to 0.25 ppm Hg as (HgClz) for 5 days .

Scanning electron micrographs of
lamellae with smooth epithelial cells .

Transmission electron micrOgraph of a
lamellar cross section showing smooth
epithelial cell . . . . . . . . . . . .

Transmission electron micrograph of the
cell junctions between a smooth
epithelial cell and a "normal"
epithelial cell . . . . . . . . . . . .

Transmission electron micrograph of
degenerating chloride cells on the
base of a lamella . . . . . . . . . . .

Scanning electron micrograph of the
lamellar surface from a trout exposed
to 0.05 ppm Hg as (HgClz) for 4 weeks .

Scanning electron micrograph of the
lamellar surface from trout exposed
to 0.05 ppm Hg as (HgClj) for 4 weeks .

Percent uptake of methyl mercury or
divalent mercury by red blood cells as
a function of time . . . . . . . . . .

Concentration of mercury in the red

blood cell fraction as a function of
ambient mercury levels . . . . . . . .

xi

Page

148

150

150

152

155

157

159

161

164

166

170

175

Figure

44.

45.

46A.
46B.

Page

Component analysis of methyl mercuric
uptake by red blood cells . . . . . . . . . . 198

Transmission electron micrograph of the
pillar cell bOdY O O O O O O O O O O O O O O 207

Scanning electron micrographs of the
lamellar surface from two species of

teleosts O O O O O O O O O O O O O O O O O O 210

xii

INTRODUCTION

Although known to be toxic to man since the first
century AD, the status of mercury as an environmental
contaminant was largely overlooked until the disastrous
epidemics of Minamata and Niigata in the 1950's and early
1960's which left numerous people dead and many more
crippled. Subsequent poisonings due in part to the use
of mercurials as seed dressing and for medicinal purposes
has promoted a revision in the heretofore promiscuous
use of mercury compounds and their Subsequent release
into the environment. As somewhat of an afterthought came
the realization that mercury contamination of the environ-
ment might also affect non-human organisms, and since much
of the mercury is eventually carried in its water soluble
form, the aquatic environment became one of the first
subjects of investigation.

With respect to teleosts, the two major research
foci became: 1) how much mercury is present in the tissues
of fish from different geographical areas (with reference
to polluted and unpolluted.waters), and 2) what concen-
trations of the various forms of mercury are lethal to

fish under controlled conditions for short time periods

(toxilogical bioassay procedures)? Relatively little
emphasis has been placed on studies of the pathway of
mercury uptake by fish and on the effects of sublethal
mercury concentrations on the vital homeostatic mechan-
isms which operate in teleosts. The purpose of the
present study was to compare and contrast two forms of
mercury most commonly involved in water pollution (mer-
curic chloride and methyl mercuric chloride) with respect
to: 1) pattern of uptake by teleosts, 2) interaction with
tissues primarily involved in uptake, and 3) morphological
and physiological changes which occur in tissues most
closely involved with the uptake process.

Initially, study was directed toward mercury-induced
physiological changes which could occur at one of the
possible uptake sites, the gill. An isolated-perfused gill
preparation was deve10ped which, for up to 10 hours,
demonstrated capabilities for active sodium transport (as
evidenced by saturation kinetics and metabolic inhibition)
and exhibited a characteristic membrane permeability shown
by urea diffusion. It became evident that while mercury
might have an effect on sodium transport and permeability,
it could also affect any one of a number of other physio-
logical processes carried out by the gill. Even if
measurements of physiological parameters were limited to

only the organs involved in mercury uptake, the magnitude

of the studies involved therein would be prohibitive.
Thus, the most logical approach was to determine the
primary pathway for mercury uptake and then investigate
the action of mercury on the physiology and morphology of
tissues involved in the uptake process. Therefore, this
dissertation was limited mainly to a study of morpho-
logical changes in rainbow trout exposed to two mercurials.
The physiological alterations could then be inferred from
the morphological information and a "hit-or-miss" study of
pathophysiology thereby avoided.

Very few anatomical studies of gill ultrastructure
have been reported on rainbow trout which involved the
use of the transmission electron microscope and, to the
author's knowledge, no information on gills from any teleost
has been obtained with either the scanning electron micro-
scope or the electron microprobe. In depth studies of this
nature on control fish are, of necessity, included in this

thesis.

LITERATURE REVIEW

Mercury Uptake

The extensive literature dealing with the importance
of mercury as a health hazard and its environmental dis-
position has been recently reviewed Lofroth (1969), Selikoff
(1971), Friberg and Vostal (1972), and D'Itri (1972).
Other information has appeared as a published discussion
from a symposium on ”Mercury in Man's Environment" and a
geological survey, "Mercury in the Environment." These
extensive reviews obviate the need to recapitulate the
ecological notoriety of mercury compounds, hence, the
present review will be limited to only those articles per-
tinent to a discussion of the action of mercury on fish‘
with respect to pathways for uptake, and possible effects

of mercury on the organ systems involved.
Primary Uptake Pathway

The movement of mercury into fish could be by one or
more routes: the gills, skin and gastrointestinal (GI)

tract. For reasons discussed below, the gills appear to

have the greatest potential for mercury transport

irrespective of whether the uptake is passive or active.

(1) Gills.--In addition to their well known respira-
tory function, the gills also function as the major site
for nitrogen excretion (Smith, 1929; Goldstein & Forster,
1961; Goldstein gt_al., 1964), ion regulation (Smith,

1930; Keys, 1931; Krogh, 1937) and probably blood pH regu-
lation (see reviews by Maetz, 1971, and Motais and Garcia-
Romeu, 1972). To attain optimal functional activity,
adequate ventilation of both the internal and external
surfaces of the gill epithelium is required. On the
external surface relatively large volumes of water (31 to
571 ml/min for 210-770 g rainbow trout) continually perfuse
the gill epithelium (Holeton & Randall, 1967; Stevens &
Randall, 1967; Cameron . Davis, 1970; Davis & Cameron, 1970)
and insures that considerable quantities of any water borne
pollutant (e.g., mercurials) are exposed to this potential
uptake pathway. In addition to a large water flow, the
surface area (SA) of the gills is large in comparison to
body surface area. Hughes (1966) reports that the total

lamellar area for a 175 g brown trout (Salmo trutta) is

 

around 600 cm2 and Muir (1969) found similar surface area
(SA)/unit weight (wt) values for other fish. The gill

SA/wt ratio is even higher in some species of marine fish

such as the skipjack tuna (Muir & Hughes, 1969) in which
the total surface area for a 100 9 fish has been calculated
to be 2,620 cmz. Hughes (1966) found that "more active
fish" have larger gill surface areas and conditions for gas
exchange are better than in "less active forms." Rainbow
trout are considered ”active“ thus the postulated increase
in surface area would in this species further enhance the
probability that the gills are sites for mercury uptake.
Exposure of fish to mercury compounds routinely
results in very high (up to several thousand times that
of the bathing media) mercury concentrations being found
associated with gill tissues (Hannerz, 1968; Rucker & Amend,
1969; Backstr6m, 1969). These very high concentrations,
however, may not always indicate that mercury is present in
gill tissue, i.e., the mercury compound might be adsorbed
’on the epithelial surface or merely trapped in the mucous
coat. Mercury has been shown to enter the gill at a rela-
tively high rate (0.3-3.2 x 10-5 mg Hg(NO3)2/hr in perfused
eel gills with ambient Hg at 0.74 ppm) (Hibiya & Oguri,
1961). In these experiments the gills were perfused with
Key's physiological saline (lack of plasma proteins and
red blood cells may affect uptake rates) and whole body
uptake was not determined under comparable conditions, so

the data presented do not provide definitive proof that

the gill is the primary site for uptake of mercury in
teleosts.

Three characteristics (e.g., large transepithelial
molecular fluxes, surface area greater than remainder of
body, and continual ventilation with environmental water)
coupled with demonstrated uptake of divalent mercury pro-
vide the gill with many criteria which would enable it to
be the major pathway for mercury uptake. There are, how-
ever, no conclusive studies reported in the literature to

verify this postulate.

(2) Skin Absorption of Mercury.--Friberg & Vostal
(1972) have reviewed the literature on skin absorption of
mercurials in mammals and their conclusions can be sum-
marized as follows: The skin can serve as a route of entry
for both inorganic mercury (such as HgO, HgClz, and HgNHZCl),
and organic mercury (methyl mercury, dicyandiamide or methyl
mercury thiacetamide). Absorption rates for both forms of
mercury are slow (maximum of 6% of the total mercury dose
absorbed via the cutaneous route by guinea pigs after five
hours exposure to either mercuric chloride or mercuric
dicyandiamide). Other pathways for uptake during many of
the above experiments cannot be excluded, that is, mercury
applied topically could enter the body as a mercury vapor

or through the digestive tract due to the experimental

animals licking the skin. Silberberg (1968 & 1972)
demonstrated that within five minutes after cutaneous
applications of mercuric chloride to humans, a gold-mercury
reaction product could be detected intracellularly and
extracellularly within the stratum corneum by means of
electron microscopy. This reaction product was observed
to spread deeper and become more concentrated with time
(up to 48 hours). Characteristics of the histochemical
procedures, however, prevented quantification of the
mercury movement and the extent of skin participation.

High molecular weight organic mercurials have also
been shown to penetrate the cornea (Ancill et_al., 1968).
Although descriptions of their experimental methods and
results are very abbreviated, they found an increase in
blood mercury after topical application of the mercurials
on the corneas of rats. Relatively little is known of the
movement of molecules across the teleostean skin and to
this author's knowledge there are no papers dealing with
the subject of mercury absorption by this organ. Even
water, which permeates tissues more readily than ions,
traverses the skin at a slow rate. Bentley (1962), using
his own data and values from the literature, calculated
that the time required to move 1 ml of water across 1 cm2

of membrane (skin) at AP of 1 atmosphere was 91 days for

a lamprey, 30 days for a frog, and 5 years for the teleost
Anguilla. The teleost skin is approximately 0.016 times
as permeable to water as the frog skin. Evans (1969) and
Motais gt_§l. (1969), using 3HOH, conclude that compared
to the gill, the skin plays a minor role in diffusional
water flow. With respect to ions, Knoll (1962) has
demonstrated the inability of hexavalent chromium to pene-
trate the skin of rainbow trout.

In conclusion, while there is no direct evidence on
the absorption of mercury compounds by the skin of teleosts,
it is probable that although some mercury can be absorbed
cutaneously, the efficiency of such absorptive processes
must be low and thus relegate this pathway as one of minor

importance.

(3) Uptake of Mercury by the Alimentary Tract.--

 

Uptake of mercury via the digestive system requires:
1) ingestion of mercury into the gut, and 2) absorption of
merCury from the lumen of the gut into the blood stream.
The latter criteria has been investigated in animals with
respect to both inorganic and organic mercury compounds.
Mercury was administered orally as mercuric acetate
to rats by Ellis and Fang (1967), who found that with
increased doses (from 3 to 9 pg mercuric acetate per rat)

the percent of the initial dose of mercury excreted in the

10

feces after 48 hours decreased from 87.3% to 77.4%. After
168 hours, 97.8% of the 3 ug/rat dose and 88.6% of the

9 ug/rat dose had been recovered in the feces. These
results demonstrated that around 13 to 23% of Hg++ is
either absorbed by the GI tract in 48 hours or retained

in the lumen or lumenal boundaries and excreted over a
longer time period. Appearance of mercury in the urine at
48 hours suggests the former is in part responsible.
Intestinal uptake of Hg++ folloWing oral doses of mercuric
chloride or mercuric nitrate occurs in rats (Swensson
gt_gl., 1959) but low adsorption rates were reported.
Alimentary tract absorption of HgC12 has also been inferred
in the dairy cow (Potter gt;21,, 1972) from the appearance
of mercury in milk, urine and body organs. This absorption
is also probably low in that 95% of the ingested dose
appears in the feces; however, absorption and subsequent
bilary excretion cannot be ruled out.

In the quail, intestinal absorption of Hg++ has also
been observed using whole-body autoradiography following
1.0 mg/kg peroral administration (Backstrdm, 1969). Only
traces of mercury were found in the body tissues indicating
low resorption and no quantitative data were given.

Studies on the quantitative absorption of mercuric

salts in the GI tract of teleosts have been limited.

11

Backstrom (1969), again using autoradiography and presenting
no quantitative data, reported that 10 days following per-
oral administration of mercuric nitrate to salmon "large
amounts" remained unabsorbed.

Inorganic mercury appears to be poorly absorbed from
the digestive tract in all vertebrates studies; however,
intestinal absorption cannot be ruled out as an important
pathway for mercury uptake in trout.

Organic mercurials, especially methyl mercury, are
rapidly absorbed from the digestive system. Friberg and
Vostal (1972) have reviewed the recent findings on rats,
cats, and monkeys and conclude that more than 90% of the
ingested mercury is absorbed via this pathway. The
absorption is the same irrespective of the form of methyl
mercury, i.e., methyl mercuric chloride or protein bound
methyl mercury. They also report that in humans, absorp—
tion of either form of methyl mercury is nearly complete.

Study of GI absorption of methyl mercury (as well as
inorganic mercury) by fish is hindered by three problems:
1) mercury must be rapidly and completely ingested to pre-
vent increased mercury concentrations in the ambient water
and thus expose the mercury to other routes of entry,

2) routinely, when mercury is force-fed, all or part of the

bolus is regurgitated (Miettinen, J. K., et al., 1969;

12

Miettinen, J. K., gt_gl., 1970; Miettinen, V. e£_al,, 1970;
and Jarvenpaa gt_al., 1970), and 3) if mercury is fed to a
"food chain" animal which is in turn fed to a fish, the
mercury might not retain its original chemical character-
istics and absorption rates would not typify the compound
in question. In spite of these problems, the general con-
sensus is that organic mercury, especially methyl mercury,
is rapidly absorbed by the gut of teleosts. Low fecal
excretion rates, down to 8% of the administered dose,
accompanied by high tissue levels following forced feeding
of methyl mercury or other Hg-organics, support this con-
tention.

If the hypothesis can be accepted that methyl mercury
(and to a lesser extent mercuric chloride) is absorbed by
the GI tract, then an indication of the contribution of
this pathway to total uptake can be ascertained from measure-
ments of the amount of mercury "carrier" viz., environmental
water, which reaches these tissues. Drinking rates of tele-
osts would thus reflect exposure of the GI mucosa to
mercury solutions and thereby provide an index of contribu-
tion of this pathway to the total mercury uptake rate.
Drinking rates in salt water fish have been reported to be
from 40 to 200 ml salt water/kg/day for an eel or sculpin

(Prosser, 1966); whereas, freshwater species such as rainbow

13

trout drink little if any water (Shendadeh and Gordon, 1969).
Other evidence indicates, however, that freshwater adapted
eels do in fact drink water with regularity. Maetz and

Skadhauge (1968) using 131

I phenol red as a tracer calcula-
ted the drinking rate of an eel in fresh water to be 13.5

+ 27.7 ul/hr/lOOg for 17 fish. Even at this rate only

3.24 ml of water would enter the stomach in a 24 hour period
and this would hardly account for the high tissue to water
mercury ratios (of up to 2,000:1 after 10 days exposure)

as reported by Hannerz (1968).

The high absorptive capacity of the GI tract for
mercury suggests this tissue could be of prime importance
as a pathway for mercury uptake; however, the relatively
low drinking rates found in freshwater teleosts would
severly limit the quantity of mercury delivered to the
absorptive site. Under stressful conditions such as

mercury exposure, drinking rates might change and increase

the importance of this uptake route.

SecondaryyUptakePathway

 

Subsequent to its entry into a fish, mercury is
distributed to the various organs of the body as demonstrated
by tissue accumulation of mercury after im, iv, or ip injec-

tion or oral administration. The most probable carrier

14

during this distribution process is the blood. Mercury
can be transported in either the plasma or cells and its
relative disposition appears to be more dependent on the
chemical form of mercury (organic or inorganic) than on

the species investigated.

Inorganic Mercury: Hg++.--Distribution of the

 

inorganic mercuric ion generally favors the plasma fraction
although much variation has been reported. In humans,
Swensson §£_gl, (1959, p. 432) found the plasma mercury/
whole blood mercury ratio for Hg(NO3)2 to be 0.80, whereas
Lundgren gt_al. (1967) reported an equal distribution
between blood and plasma fractions. Rahola gt_al. (1971)
and Miettinen (in press, quoted by Friberg & Vostal, 1972)
gave RBC/plasma ratios ranging from 0.4 to l in humans

203Hg labeled proteinate.

following oral ingestion of
Swensson gt_gl. (1959, p. 467) report that in rats "nearly
all of the Hg++ appears in plasma," however, no data were
presented. Ulfarvson (1962, quoted in Norseth & Clarkson,
1970) reported equal distribution of mercury in the same
species. Somewhat different distribution has been reported
for the cow by Potter gt_gl. (1972). Calculations using
their data indicate that 86% of the Hg++ (given orally as

HgClZ) is carried by the red blood cells.

15

Mercuric ions penetrate the red cell slowly
(equilibrium achieved after two hours in the rabbit,

Berlin & Gibson, 1963) if whole blood is used as the
incubation medium. Penetration is extremely rapid
(equilibrium under 5 min., Weed §E_§l., 1962) if red cells
are washed to remove plasma proteins and then suspended

in Ringer solution. Thus plasma proteins appear to be
involved in binding the mercury and establishing an
equilibrium (mercury-plasma protein :FtIred cell (protein—
mercury) with a characteristically slower uptake of Hg by
red cells than would be seen if mercury was exposed to

red cells in a nonprotein environment.

Relatively little information is available pertaining
to inorganic mercury uptake by non-mammalian blood. Back-
strdm (1969) reported that in the quail the ratio of mercury
between the cell and plasma fractions after injection of
Hg(NO3)2 is about 0.8. He found this ratio to be fairly
constant from 10 min to 10 days after injection which would
imply a rapid attainment of equilibrium. He also reported

that less than 1% of plasma 203

Hg activity remained in
solution after protein precipitation with alcohol. To the
author's knowledge there are no reports in the literature
on the binding of inorganic mercury compounds by teleost

blood.

16

Organic Mercury Compounds.--Transport of organic
mercurials in blood is generally agreed to be effected by
the cellular components irrespective of the species from
which the blood was sampled or the organic mercurial in
question. In rats, mercury administered as phenyl mercuric
acetate, methyl mercuric hydroxide or cyano (methyl mercuri
guanidine) is distributed almost exclusively in the red
cell (Swensson §E_al., 1959, 1967). Also in rats, Norseth
and Clarkson (1970) found a whole blood to plasma ratio of
mercury of about 271 to 1 after iv injection of methyl
mercuric chloride. This ratio remained fairly constant
even if the mercury was injected ip and it remained steady
over the experimental concentrations studied (1 to 10 mg
Hg/kg). Methyl mercury is also tightly bound to the red
cells and even after 10 days there was still 23% of the
injected dose of mercury in the blood compared to 6 and
7% in the liver and kidney respectively. In the dog, rabbit
and human, ‘75, 80, and 90% respectively of the methyl mercuric
hydroxide is bound to the red cell (Swensson et_al,, 1959,
p. 432). These authors report that in the rabbit, erythro-
cyte binding of mercury is weak, as demonstrated by a rapid
loss of 80% of the mercury from the blood to the tissues
without any accompanying change in percent mercury bound to

the red cell.

17

Backstram (1969) has described a similar red cell
affinity for organic mercurials in quail blood. He found
that the ratio of mercury between blood cells and plasma,

10 min., 1 hour, 1 day and 10 days post iv injection remained
relatively constant for methyl mercury (12.7, 32.5, 28.2,

and 28.7) and decreased for phenylmercury (16.5, 26.2, 1.6,
and 0.5) and methoxyethyl mercury (5.3, 4.4, 1.1, and 0.3).
All (99% or more) mercury in the plasma fraction was protein
bound.

Rucker and Amend (1969) reported the distribution of
ethylmercuric phosphate in rainbow trout to be mainly in
the cellular fraction with only traces in the serum; how-
ever, no quantitative data were presented.

Again, lack of literature pertinent to binding studies
of organic mercurials, especially methyl mercury, precludes
‘ definitive agreement on its disposition. From data
obtained using higher vertebrates it may be inferred that

when present, organic mercury should be mostly in the

+)

cellular fraction of fish blood and inorganic mercury (Hg+
is either evenly distributed between cells and plasma or
primarily in the plasma fractions. These extrapolations

and assumptions may or may not be warranted.

18

Effects of Mercurials on Tissues

 

Biochemical Effects

 

The chemistry and biochemistry of the various inorganic

and organic compounds has been reviewed extensively by

Webb (1966) and is beyond the scope of this thesis. For

general perspective, a brief summary of some biochemical

properties of inorganic mercury (Hg++) and monomethyl

mercury (CH3Hg+) are listed below:

1.

Specificity of both mercurials for sulfhydryl (SH)
enzymes is great, and, more often than not, SH groups
are in some way related to enzyme function which

could be altered by mercurial interactions.

Hg++ mercurials co-react with two SH groups: CH3Hg+

reacts with only one (see No. 8).

Simple amines, amino acids (especially SH amino
groups), nucleotides and nucleic acids form complexes

with Hg++.

Hg halides are more soluble in water than organic
mercurials and the reverse is true in organic sol-
vents. Thus the organic mercurials can penetrate

cell membranes more readily.

l9

Inorganic mercurials are smaller than the organic
molecules and as such are more likely to bind with
SH groups within an enzyme and are less affected by

steric hindrance.

The affinity of organic Hg for ligands is slightly
greater than that of inorganic compounds; however,

this difference is probably of minor importance.

The C-Hg bond is relatively weak, 15-19 kcal/mole,
and it is quite possible that many effects attributed
to organic mercurial compounds may be actually due to

Hg++.

Several inorganic Hg and organic Hg combinations with
monothiols and dithiols are shown below. This variety
of reactions enhances the probability of protein SH
interaction and subsequent enzymatic or structural

dysfunction.

Monothiols
R-SH

20

Organic mercurial
combinations

R'-Hg+
R-SHg-R'

Inorganic mercurial
combinations

Hg++

R-s-Hg+
R-S-Hg-S-R

/SH-Hg-R '

\\SH-Hg-R'

S-Hg-S

/

R ::R
\\S-Hg-S

(-R-S-Hg-—S-)n

9. Affinity of mercurials for SH groups is greater than

for any other single ligand thus in a mixture with

thiols and other complexing ligands the mercurial

will be primarily associated with the thiol.

10. Reactions of either of the mercurials with -S—S-

bonds of most protein molecules at physiological

conditions generally do not occur.

11. Mercurial-protein combinations are stoichiometric

and do not necessarily result in protein precipita-

tion or denaturation.

12.

13.

14.

15.

16.

21

Mercurial-SH protein bonds are usually not as strong
as mercurial-thiol bonds and thus mercury can be

freed from a protein by addition of a thiol.

Mercurials are capable of binding to non-SH sites on

proteins possibly COO- or CONH groups.

Enzyme inhibition by mercurials is generally through
formation of enzyme-mercurial complexes and not
through nonspecific denaturation and precipitation.
Inhibition can be due to mercurial bond to a SH group
on or near the active or catalytic site of the enzyme
or due to conformational changes in secondary, ter-

tiary or quaternary structure.

Low concentrations of mercurials may act to increase

enzyme activity.

Mercurials at the concentrations given have been

shown to inhibit one or more enzymes in the following
biochemical pathways: Glycolysis (1 mM), TCA cycle
(0.01 mM), ATPase, electron transport and oxidative
phosphorylation (0.001 mM), lipid synthesis (0.1 mM),
protein synthesis (5 mM), porphyrin synthesis (0.04 mM),
bioluminescence, photosynthesis (0.005 mM), active

glucose and ion transmembrane transport and diffusion,

22

erythrocyte hemolysis, generalized inhibitory effects

on mitosis, growth and differentiation.

Physiological and
Morphological Effects

 

 

Gills.--Histopathological changes in gill epithelium
after exposure to mercurials has been noted by several
investigators. Characteristically, the changes followed
exposure to high doses of the mercurial and were observed
using standard light microscopic techniques.

In general, both inorganic and organic mercurials
produce marked hyperplasia and hypertrophy of gill epithelial
cells with occasional separation of the epithelial layer
from the supporting cells. Table 1 summarizes these obser-
vations.

Impairment by mercurials of physiological functions
performed by the gill has received little attention. In
the only reported study of this nature, Meyer (1952)
described effects of mercuric ion on ion fluxes across

5

goldfish gills. He reported that 10- M Hg++ (approxi-

mately 2 ppm) administered as HgCl completely inhibited

2
active sodium uptake by the gills and caused a lO-fold

increase in branchial passive sodium loss. A lO-fold

6

- . - ++
reduct1on of the mercury concentration (10 M Hg )

depressed active sodium uptake by 25% from the control and

23

 

hmmH

 

 

 

 

 

 

.Homflwssom + mumo cw>wm uoz ++mm
mwusmomxm
Hauv Hmuos
mmma .csmE¢ usouu amp Mom mumsmmonm
+ + annouofi
paw Horosm soncamm wHDmOQXU
durum
an H
see mNH
Acowumwmcw
osma usonu annoy unsouoa
m m
.smsfiuumfiz o wasp AHA oz 3onsfimm wx\m= ahaumz
ms 0H 8
Acofiumomnfl
Hmuov
.swsaumwmm manomwmw Hmsmsoo oxen mx\mm ms mmmmwwm
. . . e.e~no.ma
mo ommum>¢
wow you
whomomxw mumsmmonm
mama usouu
..Hm no vsosd + + 3oncamm H: H ausonmfi
. Harem
6mm mmH.o
cowumummom Show
mousom Hmfiawnuflmm msmouuuwmmm mamMHmHmmhm pmflosum :mwm coaumuusousoo Hmowaono

 

 

 

 

 

 

 

.mnsuosuum HHflm co manausouwfi mo uoowmmII.H UHQMB

24

increased sodium loss by 43% over the control. No other
studies on mercury induced pathophysiology of the gill

have been reported.

Bleed.--Physiological effects of mercurials on
erythrocytes have been studied in detail; however, the
primary objective of the research in this area has been
the localization and identification of various sulfhydryl
groups either associated with the red cell membrane or
within the cell. Frequently differentiation of SH groups
has been accomplished by the use of various high molecular
weight organic mercurials which penetrate the cell membrane
to different degrees depending on molecular weight and side
chain characteristics. Inorganic mercury compounds have
also been used, but there is no data on methyl mercury salts.

Inorganic mercury (around 10-5

M HgClz) produces
increased potassium loss from rabbit erythrocytes (Joyce
23‘313, 1954; Vincent and Blackburn, 1958; Weed et_§l,,
1962). Concomitant with the K+ loss Weed et_al. (1962)
has described inhibition of active glucose uptake by these
cells. A

Organic mercurials such as chlormerodrin (CM), para-
chloromercuribenzoate (PCMB), parachloromercuribenzenesul-

fonate (PCMBS) and parachloromercuriphenylsulfonic acid

(PCMPS) affect ion fluxes and glucose absorption in

25

erythrocytes. In the human, CM, PCMB, and PCMBS inhibit
active or facilitated transport of many sugars and amino
acids into red cells (VanSteveninck gt_al., 1965; Martin,
1971) and produce potassium loss, sodium accumulation and
hemolysis of the cells (Sutherland et_al,, 1967; Shapiro
gt_21,, 1970; Knauf, I and II, 1971). These authors indi-
cate that sulfhydryl groups on the interior of the red
cell membrane are active in the regulation of cation flux
and that anion flux is governed by membrane protein amino
groups near the exterior of the membrane. This would
account for the lack of effect of mercurials on anion
fluxes. PCMB has also been shown to readily dissociate
the a and B chains of hemoglobin (Bucci, 1965) and this
technique is now used routinely for such purposes. In spite
of the actions of mercurials on red cell sulfhydryl groups,
none of the organic mercurials studied decrease the oxygen
carrying capacity of blood. In fact, PCMB, 2 chloromercuri-
4-nitrophenol and 4 chloromercuri-Z-nitro phenol have been
shown to slighly increase the amount of oxygen carried by
hemoglobin (Webb, 1966; Giardina e531” 1971).
Histological effects (other than hemolytic studies)
of mercurials on erythrocytes has received surprisingly
little attention. Kostenko and Iakouleva (1971) described

a change in mouse erythrocyte shape after exposure to

26

mercuric ions. These authors reported increasing
concentrations of Hg+ up to 10-5 M initially caused the
cells to become crenated in shape, but without loss of
volume. As the Hg++ concentration approached 10-5 M the
cells became spherical and somewhat swollen. Very rapid

4 M Hg++.

hemolysis was observed at 10-
Mercurials have variable effects on the red cell
in 2122. Yamaguchi (1969) has shown that both hemoglobin
concentration and red cell count decreased after admin-
istration of either one-third or one-fifty LD50 of ethyl
mercuric phosphate. The author, however, failed to men-
tion what the LD50 was. Erythrocytes from rainbow trout
exposed to 0.125 ppm ethyl mercury phosphate (EMP) one
hour daily for eleven days appeared normal even though
the blood mercury concentration was near 23 ppm and gill
tissues from the same fish exhibited extensive hypertrophy
and hyperplasia (Rucker and Amend, 1969). The authors
also reported that twelve weekly, one hour exposures to
0.125 ppm EMP produced gill hyperplasia with no concurrent
red cell abnormalities. Miettinen, V. et_al, (1970)
observed that after ingestion of around 10 mg methyl
mercury per Kg body weight, the spleen of rainbow trout

became swollen and darker than normal. These authors

postulated erythrocyte damage.

27

Summary

The environmental mercury problem is a well
appreciated fact; however, the effects of sublethal con-
centrations of this all too common pollutant on aquatic
animals, and especially teleosts, are virtually unknown.
Much of the mercurial induced pathophysiology and patho-
morphology can only be deduced via extrapolations from
mammalian experiments which involved chemical forms of
mercury not routinely encountered in polluted waters.
Considering the obvious differences between mammals and
fish and possible chemical dissimilarity between the small
molecular weight mercurials commonly associated with pol-
lution and larger organic mercurials utilized in the
mammalian experiments, such inferences can only be tenuous
at best. Universality of sulfhydryl groups associated
with enzymes and preponderant interactions of mercury
with sulfhydryl groups promulgate the need for adequate
assessment of the physiological consequences of mercury

on aquatic organisms.

MATERIALS AND METHODS

Experimental Animals

 

Rainbow trout (Salmo gairdneri) were obtained from

 

the Department of Natural Resources, Fish Research Labora-
tory at Grayling, Michigan, and from the Great Lakes
Fishery Laboratory, Bureau of Sport Fisheries and Wildlife
at Ann Arbor, Michigan. They were transported to East
Lansing, Michigan, in an insulated metal tank, the lining
of which had previously been coated with nontoxic paint.
Viability of the fish was insured en route by oxygenation
of the water with mechanical aerator and the addition of
ice to keep water temperature near the optimal 13° C. At
Michigan State University the trout were maintained at

13 1 1° C in 120 gallon fiberglass tanks under a 14 hour
daily photoperiod. Water in the holding tanks was aerated
and continually flushed with tap water from which iron

and chlorine had been removed. Commercial trout pellets

3/16 inch) were fed to the fish biweekly.

Exposure Protocol

 

Trout (60-200 9) were exposed to various concentra-

tions of either CH3HgCl or HgCl2 in 100 liter tanks. To

28

29

prevent build-up of toxic ammonia and to maintain the
mercury concentrations, 80% of the water was removed daily
and replaced with aged tap water containing the appropriate
mercury concentrations. Exposure schedules applicable to

the respective experiments were as follows:

 

 

 

 

Duration
Dose (days) Experiment
HgCl2
l . . . 2
0.25 ppm 24 hrs Microprobe, SEM, TEM, and nght Microscopy
0.25 ppm 72 hrs 1 SEM, TEM, and Light Microscopy
0.25 ppm 5 days Microprobe, SEM, TEM, and Light Microscopy
0.05 ppm 40 days SEM, and Light Microscopy
275.0 ppt 24 hrs Isotope Tissue Distribution Experiment and
Radioautography
CH3HgCl

300.0 ppt 28 days SEM, TEM, and Light Microscopy

300.0 ppt 56 daysii"5 SEM, TEM, and Light Microscopy
0.05 ppm 24 hrs 4 Microprobe, SEM, TEM, and Light Microscopy
0.05 ppm 5 days Microprobe, SEM, TEM, and Light Microscopy
275.0 ppt 24 hrs Isotope Tissue Distribution Experiment and
Radioautography

lDose equivalent of Hg salt added 1 hr prior to tissue sampling
for microprobe studies only.

2 . . . .
SEM = Scannlng electron microscopy; TEM = Transm1551on electron

microscopy.

3Fish exposed at Ann Arbor to methyl mercury through the courtesy
of Drs. Robert E. Reinert and Wayne Willford.

4Dose equivalent of Hg salt added to tank 1 hr prior to tissue
sampling.

5Exposed to mercury at 5° and 15°.

6ppm = parts per million (mg/Kg); ppt = parts per trillion (ng/Kg).

30

Radioisotope Methodology

Radioactive mercury, as either 203HgCl2 or CH3203HgCl,
was obtained from ICN Chemical and Radioisotope Division,
Irvine, California. The specific activity for each compound
was indicated with each shipment and assumed to be correct
as direct mercury analysis in the laboratory was not avail-
able. Detection of the 0.279 Mev gamma radiation was
achieved either with a two-inch thallium activated sodium
iodide NaI(T1) crystal Model DS 202(v) well detector and
activity displayed on a Model 8725 Analyzer/Sealer, both
from Nuclear-Chicago, Des Plaines, Illinois, or with a
multichannel analyzer and automatic 100 tube sample changer
also from Nuclear-Chicago. Use of the multichannel analyses
and tube changer system was through the courtesy of Dr. Niles
Kevern and Mr. James Seelye of the Department of Fisheries
and Wildlife, Michigan State University. Constant count-
ing geometry was not always possible during the various
experiments. Correction for this was made by counting known
amounts of the isotope under the various geometries encoun-
tered and determination of the respective counting effi-
ciencies. During routine counting the window width was
centered around the 0.279 Mev gamma peak. The region of

interest was determined by multichannel analysis to give a

high peak to background ratio.

31

The amount of mercury in a given experiment was

determined by the equation:

Cpm

 

 

 

o
% efficiency mgHg
mgHg = 9
2.22 x 10 dpm/mCi mCi
Where: cpmo = counts per minute at tO (time at which

specific activity determinations were
made at ICN)
% efficiency = calculated efficiency at constant
geometry

mgHb/mCi = specific activity of the Hg at tO

cpmo is in turn calculated from the formula:

_ t
cpmo — ;:TE
Where: cpmt = the counts at time of experiments
A = 0.693/46.6

t = time between to and experiment (in days)

_ 203
46.6 — t1/2 for Hg

Blood Studies

 

Blood samples were collected from the caudal vein

by insertion of 1-1/2 inch 21 gauge needle into the ventral

32

side of the haemal arch approximately 0.5 cm distal to the
anal fin. Two or three m1 of blood were then slowly drawn
into a heparinized syringe. Blood experimentsawere
initiated within two hours after the blood sample was taken.

203

All glassware used in conjunCtion with Hg-blood studies

was coated with silicone (Siliclad® , Clay Adams, Inc.,
New York, N.Y.) to prevent adsorption of 203Hg onto the
glass.

Hematocrit and Hemoglobin
Determinations

 

 

Heparinized microhematocrit capillary tubes (Division
American Hospital Supply Corp., Miami, Fla.) containing
aliquots of blood used during each experiment were centri-
fuged for 3 min at 11,500 rpm on an IEC Model MB Micro-
hematocrit centrifuge. Percent red cell volume was deter-
mined with an IEC Circular Microcapillary Tube Reader.

Hemoglobin concentrations were determined by the
cyanmethemoglobin method (Oser, 1965). Twenty ul of blood
were added to 5 ml of cyanmethemoglobin reagent (Hycel,
Houston, Tex.) and compared spectrophotometrically (B & L
Spectronic 20 Colorimeter, Rochester, N.Y.) to known hemo-
globin concentration standards (cyanmethemoglobin certified
standards, Hycel, Houston, Tex.). The unknown hemoglobin

concentrations were then calculated from the formula:

33

Hb concentration O.D. unknown (Hb conc. of standard)

-
J

unknown '— O.D. standard

Mercury Uptake_by Red Blood
Cells (Uptake Rate)

 

The rate of mercury uptake by red blood cells was
determined by incubating either whole blood or washed cells
with 203HgCl2 or CH3203HgC1 at 14° C for various time
periods. Heparinized whole blood (2.2 ml) or 2.4 ml washed
cells (0.40 ml cells washed 3 times in 3 volumes Ringer
and resuspended in 2 ml Ringer or phosphate Ringer--see
Appendix I for composition of Ringer solutions-~was placed
in a 25 m1 stoppered Erlenmeyer flask. The mercury solu-
tion was added to the blood at time zero and the flask
stirred at 16 rpm with a multipurpose rotator, model 150V
(Scientific Industries, Inc., Springfield, Mass.). At
timed intervals 0.20 ml of blood was removed and immediately
placed in a 0.4 ml polyethylene microcentrifuge tube
(A. H. Thomas, Phila., Pa.) and centrifuged in a Beckman
Microfuge Model 152 (Fullerton, Calif.). The supernatant
was drawn off and placed in a second microcentrifuge tube.
The packed red cells were then resuspended in Ringer solu-
tion, centrifuged and this supernatant "wash" was placed
in a third microcentrifuge tube. The concentration of

mercury in the packed red cells, plasma (or Ringer) and

34

wash solution in the microcentrifuge tubes was determined
by scintilation counting. The percent uptake of mercury
into the cellular fraction was corrected for equal volumes
of red cells and plasma and calculated according to the

following formula:

l-Hct

 

% incorp. z —fiEE—-(act1v1ty 1n red cell fractlon) x 100
into REC lfiggg (activity in red cell fraction)

+ (activity in plasma fraction)

+ (activity in wash)

The percent uptake vs. time was plotted. Component analysis
of the equilibrium curve and flux rate determinations are
given in Appendix 2. .

If mercury is bound to the red cells it must exhibit
certain binding characteristics (i.e., it must have an
affinity for the red cell), there must be a quantity of
binding sites which can be saturated and there must be
certain chemicals which inhibit mercury binding to the red
cell either through competition for the binding site on
the hemoglobin or through binding of the mercury to the

inhibitor, thus sequestering it (Hg) from the solvent.

Binding Characteristics

The binding characteristics of mercury as CH3HgC1

and HgCl2 was determined on washed red cells suspended in

35

Ringer solution by incubating the cells with various
concentrations of mercury for 90 minutes (equilibrium
conditions were reached in this time period). Following
incubation the suspended cells were centrifuged, washed,
and the percent activity in the cellular fraction deter—
mined as described above. The radioactive mercury was
diluted with non-radioactive mercury of the same molecular
form to maintain the total amount of 203Hg constant. The
mercury concentrations used in the medium (Table 2) are

calculated both in terms of ppm in the total cell-Ringer

mixture and as ppm in the red cell volume only.

Table 2.--Ambient mercury concentrations.

 

 

Methyl mercury (CH3HgCl)

 

Egfial 0.0133 0.133 1.133 10.133 100.1 1000
ppm 0.0947 0.947 8.09 72.38 715.0 7144

(m9 Hg/l RBC)

 

 

Inorganic mercury HgCl

 

2
ppm '

total 0.054 0.540 1.454 14.05 140.05
ppm 0.387 3.87 10.39 100.4 1000.0

(mg Hg/l RBC)

 

 

Note: Mercury concentrations lower than thgsg used
were not possible due to lack of sufficient total Hg
necessary for reliable counts, and higher concentrations
produced RBC hemolysis. ‘

36

This facilitates comparison of environmental blood mercury
analysis (ppm whole blood) and gives an idea of the mercury
concentration "seen" by the red cell (this is usually most
important with respect to mercury uptake by cells, Vincent
and Blackburn, 1958).

The binding affinity of the two mercury compounds at
various concentrations was tested by adding sulfhydryl
agents to red cells previously equilibrated for 90 mins

with mercury. Sulfhydryl agents selected were 1.37 x lO-SM

bovine albumin (Mann Res. Labs, N.Y., N.Y.), 1.37 x 10-4M
l-cysteine (Sigma Chem. Co., St. Louis, Mo.) and 1.37 x
10-4M reduced glutathione (Sigma Chem. Co.). Red cells and
sulfhydryl agents were incubated an additional 90 minutes
after which the percent mercury in the cellular fraction
was determined.

Ox en-Carr in Capacity--
Effect of CH HgCl and HgCl

3

 

2

The oxygen carrying capacity of red blood cells
exposed to various concentrations of organic and inorganic
mercury was determined using the method of Klingenmaier
§£_31. (1969) as modified by Hoffert (in press). This tech-
nique allows measurement of red cell oxygen content by

determination of the increase in oxygen tension following

37

oxygen release by exposure to carbon monoxide. Change in

oxygen tension is monitored with an oxygen electrode.

Apparatus for CO Saturation of Ringer Solution.—-
Approximately 400 ml phosphate buffered Ringer solution
was placed in a pyrex gas washing bottle (tall form with
frittered disc, Corning Glass Works, Corning, N.Y.) and
deoxygenated by bubbling nitrogen through it for 8-10 hours.
After deoxygenation 200 ml of this solution was flushed
into a round bottom flask containing 100% CO (Matheson Gas
Co., Joliet, 111.). Saturation of the Ringer solution
with CO was accomplished in 2 to 3 hours by flushing the
system several times with fresh CO and mixing with a
magnetic stirrer. Prior to each day's experiment 30-40 m1
of the CO-saturated Ringer was drawn off into a 50 ml glass
syringe and placed in the cold room to equilibrate to
15° C. The luer end of the syringe was fitted with a two—
way stopcock and a 2.5 ml glass syringe minus the barrel
which permitted rapid sampling of the CO-Ringer solution
by introduction of a small volume into the 2.5 ml syringe
without contamination of.the entire Ringer solution with

room air.

Blood Incubation.--Washed red cells (0.15 ml) were

.added to 1.0 m1 of phosphate buffered Ringer containing

38

0.1 or 100 ppm Hg as HgCl2 or CH3HgCl and incubated either
with room air or low PO2 gas for 3 hours. Regulation of
the low PO2 environment was achieved by a constant flow of
6% oxygen, 94% nitrogen gas saturated with water over the
cells. Mixing of the cells was accomplished by use of a

shaker (L. Olson, Ph.D. Thesis, 1969).

Liberation of Oxygen from Red Cells.--At the end of

 

the mercury incubation period a sample of known volume was
taken, mixed with CO-Ringer for l min and the oxygen tension
determined. Collection of the blood and CO-Ringer was by
means of a 1 cc syringe fitted with a blunt 18 gauge needle
which in turn was inserted into approximately 4.5 cm of
intramedic polyethylene tubing, PE 190. After all air and
liquid was eliminated from the syringe by a mercury
(metallic) plug, blood was drawn into the PE tubing until
the Hg-blood interface reached the needle. The PE tubing
was then inserted into the 2.5 cc syringe and CO-Ringer
drawn into the 1 cc syringe until the plunger volume was

at 0.6 ml. The solution was then mixed. The exact volume
of blood and CO-Ringer taken up was determined by filling
the syringe to the appropriate mark with mercury and then
weighing the syringe and calculating the volume of mercury

contained. Correction was also made for the mercury plug.

39

Oxygen Tension Determinations and Calculations.--

 

Oxygen tension in the solution after release by the red
cells was determined on a blood gas analyzer Radiometer
BMS 3 Blood microsystem with PHM 71 Acid Base Analyzer

equipped with P0 and PCO modules (Radiometer, Copenhagen,
2 2

Denmark). Blood incubations as well as PO determinations
2
were carried out in the cold room (15° C) during the HgCl2

study. The blood gas analyzer was calibrated prior to each
determination with air saturated water (20.95% 02) and 0%
Oz-zero solution type 10-S4150, London
Co.). The procedures used in calibration and sample

oxygen solution (P

determination were as described in the London Technical
Manual with the exception that due to the low working
temperature, response time was extremely long and therefore
it became necessary to read the sample on the analog read-

out at 4 minutes after introduction into the P electrode.

02

Although this procedure was not ideal, reproducibility was
good enough (i 2 mm Hg) to warrant its use. Blood incuba-
tion with CH3HgCl was also at 15° C, but the radiometer

was calibrated and maintained at 37° during the experiment.

The oxygen carrying capacity was calculated using

the following formula (Klingenmaier et al., 1969):

_ 100 - 1 . _ _
c02 — U . PB 92v pl(v U)

 

40

Where: C = oxygen content (ml/100 ml)
= barometric pressure corrected for water vapor
P1 = partial pressure of oxygen in the CO—Ringer

P2 = partial pressure of oxygen in 1 cc diluting
syringe

U = volume of PE tubing

V = total volume of diluting syringe at 0.6 cc
mark minus Hg plug

A = solubility of oxygen in Ringer at 15° C =
0.034
For comparisons within each experiment, the volume of oxygen
is expressed in terms of ml 02/9 Hb. Since only the effects
of mercury on 02 carrying capacity were of interest, no
attempt was made to correct for the higher Radiometer

temperatures used in the methyl mercury experiment.

Gill Histology

 

Light Microscopy

 

The first and third pair of gill arches were excised
from control fish and fish exposed to various concentrations
of both CH3HgC1 and HgCl2 (see exposure schedule) and fixed
for 48 hours in Detrich's fixative (see Appendix I). Tis-
sues were dehydrated in tetrahydrofuran (Fisher Sci. Co.,

Fairlawn, N.J.), vacuum embedded in Paraplast (A. H. Thomas,

41

Phila., Pa.) and sectioned at Bu. One-half of the sections
were stained with hematoxylin and eosin and the remainder
with Alcin blue (ph 1.0 method) for identification of
sulfated mucosubstances. Both procedures are described in
the Manual of Histologic Staining Methods of the Armed
Forces Institute of Pathology (Luna, 1968).

Transmission Electron
Microscopy

 

Filaments from the second arch on the right side
of the fish were excised and fixed at 4° C in glutaraldehyde
(Fisher Sci. Co.) diluted to 4% or 6% with S¢rensen buffer
(see Appendix III). The second arch was chosen for all
TEM and SEM work because this arch is normally exposed to
the greatest volume of water during normal respiration
(Paling, 1968). After fixation for 3-6 hours the tissues
were rinsed twice (30 min each) in Sorensen buffer solution
and post fixed in 1% osmium tetraoxide (0304) for 3 hours.
(As tissue stains for electron microscopy are of necessity
electron dense, an attempt was made to use mercury as a
supravital stain, thus some tissues were neither post fixed
with 0504 nor stained after sectioning.) Following post
fixation the tissues were dehydrated in graded concentra—
tions of either ethanol or acetone and embedded in Spurr's

low viscosity embedding media (Polyscience Inc., Warrington,

\ ' 42

Pa.). Sections approximately 600A thick were cut with a
Sorvall MT2-B "Porter-Blum" ultra-microtome (Ivan Sorvall,
Inc., Norwalk, Conn.) mounted on 300 mesh c0pper grids and
stained with uranyl acetate (UA) (Watson, 1958) followed
by lead citrate (LC) (Reynolds, 1963). Thick sections
(2-3u) of the same area were cut and observed under phase
contrast microscopy for orientation purposes.

Transmission Electron Microscopy was done on a
. Philips EM 300 Electron Microscope (Philips Electronic
Instruments, Mt. Vernon, N.Y.). Accelerating voltage of
40-60 KV was used for all stained specimens and 20 or 40 KV
accelerating voltage for unstained tissues. Photomicro-
graphs were recorded on Kodak Electron Microscope film,
ESTAR, thick base, 3-1/4 x 4 inches (Eastman Kodak Co.,

Rochester, N.Y.).
Scanning Electron Microscopy

The second gill arch on the left side was routinely
used for all SEM studies. After removal from the fish the
arch was fixed for 6 hours at 4° C in either 4% glutaralde-
hyde-S¢rensen or 50% glutaraldehyde stock solution.
Preliminary studies showed that when 4% glutaraldehyde was
used most of the lamellar surface was coated with mucus

(see Figure 9B). Fifty percent glutaraldehyde prevented

43

the mucus coat and produced no significant change in the
appearance of the ultrastructure of the lamella and it was
used exclusively in latter experiments. After fixation
the gills were rinsed 6 times in S¢rensen buffer (30 min
each) and dehydrated by freeze-drying (Atmo-Vac Freeze
Dryer, Refrigeration for Science, Inc., Island Park, N.Y.)
or by the critical point drying method (DCPl Critical Point
Dryer, Denton Vacuum Inc.). See Appendix IV for both
methods. As critical point dried tissues are generally
superior to the freeze-dried specimens (Anderson, 1951,
and Horridge & Tomm, 1969) in terms of ultrastructure
preservation, this method was used whenever possible.
Filaments from the dry gill arches were dissected free
from each other and glued with Duco cement (E. I. du Pont
de Nemours and Co., Wilmington, Del.) to a 12 mm glass
cover slide, lamellar side up. The tissue and slide were
then glued to aluminum stubs with an electron conducting
cement and coated withcarbon (approx. 100-200A thick)

and gold palladium (100A) in a Ladd evaporator (Ladd
Research Industries, Inc., Burlington, Vt.). They were
then placed in an AMR Model 900 Scanning Electron Micro-
scope (Advanced Metals Research, Burlington, Mass.) for
observation. Photomicrographs were recorded on Polaroid

Type N/P positive-negative film.

44

Electron-Microprobe

SamplefPreparation.--The second pair of gill arches
from fish exposed to pharmacological concentrations of
either methyl mercuric chloride or mercuric chloride were
removed within 15 sec after the fish was killed and quick
frozen in isopentane (2 methyl butane, Eastman Kodak Co.)
cooled to near -160° C with liquid nitrogen. Frozen tissues
were then mounted on a brass stub and embedded with Ames
O.C.T. compound at -30° C (Ames Company Division, Miles
Labs., Inc., Elkhart, Ind.). Originally tissues were
embedded in ice; however, sectioning was difficult due to
crystal rearrangement and subsequent shattering. This was
overcome by using the O.C.T. compound which sections easily
at -30° C.‘ Four or ten micron frozen sections were cut on
a Model CTD-International Harris Cryostat (International
Equip. Co., Needham Heights, Mass.) and fastened with East-
man 910 adhesive (Eastman Chemical Products, Kingsport,
Tenn.) to either quartz slides or carbon discs. Standard
glass slides could not be used as the high amounts of
impurities they contain would produce excess background
radiation in the microprobe. After the 910 adhesive had
polymerized (1-3 hours) the tissues were freeze dried for

36 hours. Although all tissues were stored in a dessicator

45

until just prior to examination, care was taken to schedule
the experiments such that no more than 12 hours elapsed
between completion of freeze drying and initiation of the
microprobe studies to prevent small molecular weight ion
shifts due to tissue rehydration. Small crystals of HgCl2
were glued to a corner of each slide in such a position
that they were at least 5 mm from the tissues. These

crystals were subsequently used to localize the L and M

peaks in the microprobe.

Microprobe Analysis.--Dried tissues were carbon

 

coated (about 100A thickness) with a varian model VElO
evaporator to reduce mercury evaporation and placed in

an ARL Microprobe, Model EMX-SM (Applied Research Labo-
ratories, Sunland, Calif.). Accelerating electron voltage
ranged up to 30 KVwith the elements in question and was
generally in excess of 23 KV with the mercury analyses.
Elemental analyses were accomplished either by a tissue
scan (for characteristic x-ray emission) which was then
compared to an image scan (secondary electron detection)
of the tissues or by a single line scan with x-ray activity
plotted on a x-y recorder as a function of distance across
the tissue. The tissue scan and image scan micrographs

were recorded on Polaroid 200 black and white film and the

46

line scan recorded on a Hewlett Packard x-y recorder. Low
peak to background ratios and inherent difficulties with
the L and M line spectra of mercury made the tissue scan
procedure questionable for this element; however, line scan
procedures proved adequate. Line scan and tissue scan
procedures were also used for detection of other elements
as both an aid in tissue orientation and a check on dif-
fusional ion movements which might have occurred during
sample preparation. The small (0.5u) diameter of the pri-
mary electron beam permits fairly high resolution of ions
with a given area especially with the line scan procedure.
All ion analyses were qualitative in nature and only semi-
quantitative due primarily to the existing state of the

art (Rasmussen §E_al., 1968) and partially to the tremendous

time requirements for quantitative analysis.

Pathway of Mercury Uptake

 

Esophageal Ligation

 

The pathway for mercury uptake was determined by

203

comparing the relative amounts of Hg in various tissues

of normal rainbow trout to concentrations of the isotope
in trout whose esophagus had been ligated or sham ligated.

Uptake of both 203HgCl2 and CH3203HgCl was determined.

47

Ligation Procedure.--Prior to surgery the fish were

 

anesthetized with tricainemethane sulfonate (MS 222) and
placed ventral side up in a trough. Gill ventilation and
anesthesia were maintained by a pulsatile flow of aerated-
MS 222 treated water through a rubber tube which was placed
in the oral cavity. An incision was made on the ventral
side about 3-4 mm lateral and parallel to the midline. The
anterior portion of the 30-40 mm incision approached the
pericardial sac, but did not penetrate it. The anterior
portion of the stomach was located and ligated with heavy
thread after which the incision was closed with 9 mm wound
clips (Auto Clips No. B—2355-100, Clay Adams Inc., N.Y.,
N.Y.). See Figure 1. Sham surgery was performed in a
similar manner except that the anterior stomach was located
but not ligated. All animals were allowed to recover for
6 days before exposure to the mercury solution. Autopsies
of the ligated fish revealed that the ligatures were at
the anterior end of the stomachs and had remained tight
during the course of the experiment.

Fish from each group (control, sham, and ligated)
were placed into 100 liters of aged tap water which con-
tained either 203HgCl2 or 203HgCh3Cl. After 24 hours expo-

sure the fish were randomly removed and blood samples taken

as described earlier. They were then killed by cervical

48

.musoeflummxm
coflummfia Hommmnmomm How Aqv musummfla pom AHV coflmwocfl mo cofluwmomll.a madman

 

 

50

dislocation and tissue samples from the gills, liver, kidney,
heart, skeletal muscle, stomach and intestine were removed
and placed into pre-weighted 12 x 75 mm disposable culture
tubes. The tubes were then covered and weighed on a Mettler
balance to the nearest pg to obtain tissue wet weights.
Radioactivity in the samples was then counted as described
previously using a multichannel analyser-automatic sample
changer system. The tissues and tube were then dried at
110° C for 48—72 hours and reweighed to obtain tissue dry
weights. Analysis of these data was done on a Control Data
Corporation 6500 computer.

In conjunction with the above experiment, several
samples of gill and stomach tissue were prepared for radio-

autography.

Radioautography

 

Radioautographs were used to aid in the localization
of the pathway of mercury uptake in trout. The 0.279 Mev

203

gamma ray emitted by Hg is of suitable energy to permit

such studies. Tissues were removed from a trout which had

203Hg as noted above and quick frozen in

been exposed to
ethanol cooled with dry ice. Frozen tissues were embedded
on a brass stub with distilled water and 60 sections cut

with the Cryostat-Microtome at -30° C. The frozen sections

were attached to albumin coated glass slides (Albumin

51

Fixative, Mayer Scientific Products, Allen Park, Mich.)

by finger thawing and refreezing. The slides were then
freeze dried for 24 hours. After drying they were dipped
in 0.05% celloidin (Parlodion, Mallinckrodt, N.Y., N.Y.)
dried and dipped in NBT-Z Nuclear Tracking emulsion
(Eastman Kodak, Rochester, N.Y.). The slides were dried
in a vertical position for two hours (horizontal drying
often results in a very thick emulsion coat) and stored in
a Con-Rad/Joftes exposure chamber (Controls for Radiation,
Inc., Cambridge, Mass.) until sufficient time had elapsed
and the emulsion could be developed. Developing was
accomplished with D-l9 developer as recommended by Eastman
Kodak.

Numerous problems were encountered with this pro—
cedure. It became apparent that some radioactivity was
lost in the radioautography procedure. This was indicated
by the lack of correspondence between tissue activity
(cpm/g tissue) and grain count especially with respect to
gill tissues. The loss of activity could have resulted
from any one of the numerous steps involved; however, no
attempt was made to investigate which one(s) were critical.
Instead an entirely different approach was tried.

Tissues were removed and quick frozen in isopentane

cooled with liquid nitrogen after which they were embedded

52

on brass stubs with O.C.T. embedding media. Again 6D
sections were cut with the Cryostat-Microtome and were
glued to glass slides with Eastman 910 adhesive followed
by lyophilization for 24 hours. This procedure results in
slight ion movement as shown by the electron microprobe
studies. Dried sections were processed as described pre-
viously; thus, failure to localize mercury could only be
due to (1) failure of mercury to bind at sites specific
enough to be resolved by emulsion thickness and/or light
microscope methods, (2) movement of mercury during coating
with celloidin or the tracking emulsion, (3) gross emulsion
film shifts during drying or staining or (4) sublimation

of 203Hg.

Species Study

 

To compare the normal ultrastructure of gills from
various species of freshwater teleosts samples were taken
from ten species ranging on the evolutionary scale from
very "old," i.e., fresh-water dogfish (bowfin) and gar, to
more "modern" species such as trout and perch. With the
exception of the trout, all Species studied were obtained
through the courtesy of Dr. J. B. Hunn and the Fish Control
Laboratory, Bureau of Sport Fisheries and Wildlife at

La Crosse, Wisconsin. To minimize gill damage as a result

53

of transportation of live fish, representatives of each
species were killed and gill tissues fixed for SEM and

TEM studies prior to the return to East Lansing.

RESULTS AND DISCUSSION

The results and discussion sections for this thesis
have been combined into one section in which data from
each experiment is presented and discussed as a unit.

Due to the large number of electron micrographs included
herein, it is believed that separated "results" and "dis—
cussion" sections would lead to unnecessary redundancy
and loss of continuity. A brief literature review of the
teleost gill is presented in conjunction with trout ana-
tomical studies for purposes of orientation and clarifi-

cation.

Pathway of Mercury_Uptake

 

Uptake of divalent mercuric ions as well as methyl
mercuric ions occurs primarily by way of the gills as can
be seen in Table 3. Ligation of the esophagous has no
apparent effect on 24 hr accumulation of either form of
mercury by any tissue with the exception of gill inorganic
mercury levels. In this instance gills from the sham
operated fish had Significantly higher levels of mercury
than either the control or ligated fish. The reason for

this is unknown. Surgical trauma might result in an

54

55

$550:

.A<>oz< was wcov Hw>ma mo.o_W m can an o3u Hmnuo on» EOHM ucuuommwo hausmoamasvwm ma macaw macs

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

m h m m m w mGOMHMKerm
4 use to 62
omm.omwmo.ama www.mHva.HvH mro.m HNQ.Hm mmo.HHOh.hH omw.wfimw.aml w0H.Hme.mH woon
mmm.o Hmm.H vnm.o wvm.m mmv.o Hmm.N mHo.0HmH.o mmo.0HmH.o Hmo.0HoH.o wHomsz
hwm.m www.mm mHM.HHHNH.mHH mom.MHHbo.Nm Nmm.0HmN.v h¢¢.0Hmv.m Ohm.0va.v hwcoflm
mvm.m HQH.Hv unh.w Hhm.wv mMHpm “mo.om me.0Hom.m hmm.OHNv.m wo~.owow.m uumom
hhv.m me.MH omH.N Hhh.wa Hmm.N www.mfl mmo.0HNo.N me.OHvH.N omo.meN.NH ocwumouaH
NHH.N wmh.m mom.o Ham.oa mmo.H Hem.m owH.0Hmm.o wmm.oamh.a mmH.Hva.N nonfioum
va.e H ¢.mN mam.v H v.0w vmh.m H m.mm HMM.OHN>.N ovm.owwm.m hmM.OHow.N H0>fiA
m.mmw m.hmm mN.hvw m.omm vw.vmfi w.vov Hmm.Hva.om 5mm.HHHm.nm Mamm.HHmo.mN HHHO

drumqu Bdrm HMMMMMW omumqu Edam HMMWMMU

Aunmflos who m\mm may wufi>fiuom mommwa
.HommMmo Ho NHomm Mwfiuww mm huncumfi

Ham mum mo wrongs H50: HSOMINHCUSD no COHummflH Hmommamomm mo HUUMMMII.M mqmda

56

elevated respiratory frequency, however, this Should be
reflected also by the ligated fish. It is interesting to
note that although not statistically Significant, the con-
centration of either form of mercury is frequently higher
in the sham operated fish than either controls or ligated
groups. This difference could be coincidental or sug-
gestive of possible traumatic effects of the ligation pro-
cedure which suppresses ventilation to some extent. In
any event the uptake differences between the control, sham
and ligated groups are sufficiently small to permit the
conclusion that the gill is the primary pathway for uptake

of either Hg+ or CH Hg+ in nonfeeding rainbow trout.

3

Surgical manipulations used during this experiment
necessarily produce an open wound which can be closed with

wound clips but which might reopen during the experiment

203

and allow Hg to enter the body cavity. As a check on

dry weight

. ratios were
wet welght

 

the "leakiness" of the incision,
calculated for all tissues to measure any water influx and
are given in Table 4. Values were combined from fish
exposed to inorganic and organic mercury, and one way
ANOVA used for statistical comparisons

Only liver and kidney tissues appeared hydrated after
surgical procedures; heart and muscle followed this trend

but the differences were not Significant even at the

57

Table 4.--Effects of surgical incision on tissue hydration.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

dry weight/wet weight x 100%
Control Sham Ligated
Gill 19.17:0.348a 18.97:0.261 l9.22:0.606
LiverC 24.94:0.204 23.03:0.296 22.58:0.296
Stomach 28.84:1.540 30.65:2.837 27.95:4.005
Intestine 31.81:2.904 . 35.14:2.158 32.5613.853
Heart 17.79iO.631 ' l6.71:0.2993 l6.82:0.705
Kidneyb 18.81:0.297 l7.92:0.246 17.5810.353
Muscle 23.66:1.l95 22.71:l.385 20.14ao.446
Blood 13.26t0.572 14.13:0.742 13.4lio.642
3:22:53... .. 1. 13
aMeanrMSE

bOne mean is significantly different from the other
two at the P = 0.05 level (one way ANOVA).

COne mean is Significantly different from the other
two at the P i 0.01 level (one way ANOVA).

58

P = 0.10 level. Blood, which Should provide a good index
of internal hydration, was not affected by the incision.
Lack of hydration throughout the tissues studied suggests
that water influx through the incision either must be
Slight or there is a preferential accumulation of water
by liver and kidney tissues, while other tissues are not
affected. The latter seems unlikely.

Differential uptake rates of the two mercurials
illustrates one important difference between the organic
and inorganic forms of these two compounds. Data for
control fish from Table 3 are plotted in Figure 2 for com—
parison of tissue mercury values. Methyl mercury is taken
up by all tissues at a much faster rate over the twenty-
four hour experimental period. Large concentration
differences observed between the two compounds in gill
tissues are due to greater penetration by the methylated
form. Mucus which continually coats the gill epithelium
can act as an ion binding resin to trap mercury and pre-
vent access into the tissue. Concentrations of both
mercurials were identical for the experiment and if both
compounds were trapped or precipitated by the mucus,
inorganic mercury concentrations at the gill should be no
less than one-half that of the organic form if one assumes
Hg+ ion

that one SH binding Site is occupied by each CH3

59

.Am u av .m.m H msomz .Anomm How mm yam mhmv wusouma
Asmmo . mmmmov oacmmno no Apmammflum .++mmv UHGDmHocw Hmnuwm on
musos em Hum Ummomxm upon» Eoum nonmmwu cfi mc0fiumuucmocoo musoumzll.m mnsmflm

:9

JL
7’

60

 

 

 

 

 

 

(00) mils. g.

 

61

and a maximum of 2 SH Sites per Hg++. Additional binding
of the methylated form must either reflect entry into the
gill tissues where additional Sites are available or
increased mucus secretion from methyl mercury exposure.
Observations on the fish from this experiment and numerous
other experiments have shown, however, that inorganic
mercury is the more potent stimulator of mucus secretion.
Thus gill tissue levels of methyl mercury must reflect an
increased penetration and binding of this compound to gill
tissue.

Comparison of the distribution of the two mercury
compounds within one tissue demonstrates preferential
accumulation of mercury based on its chemical form. The
ratios of methyl mercury to inorganic mercury in the tis-

sues are calculated to be:

 

m
c:
a -H
0 JJ >4 (1)
H o m u o .4 c
u—I (D E a) H c: O O
H > 0 4J «S 'U U) 0
H -A u C m H D H
(D A U) H a: M E: m
Hg as CH3HgCl

14.4 11.3 3.5 1.3 11.8 19.8 15.4 5.3

 

Hg as HgCl2

 

62

These ratios are partially attributable to increased uptake
of methyl mercury by the gill, however, the tissues should
all exhibit approximately the same ratios if there is no
selectivity by the tissues for Specific forms of mercury.
The differences in these ratios show that tissues such as
the kidney, muscle, gill, heart and liver preferentially
accumulate the methylated form and blood, stomach and
intestine do not. Low ratios for the last three tissues
could be due to selective accumulation of the inorganic
form which is masked by higher uptake rates of the organic
compound. Halbhuber §E_gl. (1970) have demonstrated, using

203Hg++ by the

radioautography, selective accumulation of
Paneth cells in the intestine of the rat and guinea pig.
They postulated a divalent heavy metal excretory function
for these cells. Paneth cells in the trout gut might have
a similar function and sequester divalent mercury but not
methylated forms. Blood distribution reflects binding of
the methylated form to red cells and inorganic form to

plasma proteins (see section on blood in results and dis-

cussion).

nylmercuric acetate
mercuric acetate

 

Uptake ratios of phe calculated
from Ellis and Fang (1967) also Show increased tissue
accumulation of the organic mercurial compounds to the

inorganic 24 hours after oral administration. Calculated

63

ratios were highest for blood (44) followed by liver (10),
heart (6), kidney (2.5), and muscle (1.7). These ratios
are not Similar to those calculated for trout. Differences
could be due to either differences between methyl and
phenyl mercury with reSpect to tissue binding or to the
pathway for uptake, i.e. intestinal vs. branchial routes.
Backstrém (1969) administered methyl and inorganic
mercury iv into speckled trout. After 24 hours, distribu-
tion patterns of methyl mercury were identical to that in
the present study with the exception of the gills. Low
gill values reported by Backstrom are likely due to the
mode of administration of the mercurial. Backstrdm also
reported highest inorganic mercury concentrations in the
kidney; blood and gill values were approximately 50% lower.
This discrepancy in distribution as compared to data pre-
sented in this thesis is also probably due to mode of
administration of the mercurial. Hannerz (1968) reported
24 hour tissue mercury distributions in cod exposed to
water borne mercurials which are similar with respect to
relative distributions of the mercurial found in the
present experiment. Hannerz postulates gill uptake as
the principal source of mercury concentrations in the cod
and pike and his postulate is confirmed by the ligation

experiments reported above.

64

Pathway of Mercury_through the Gill

 

Uptake of mercury by the gill could involve one or
both of the two physiological and anatomical pathways found
on the gill. Figure 3 is a sketch of a cross section
through a gill arch. The arch is composed of a cartilagin-
ous base (C) which supports paired filaments (F) and a
gill raker (Gr). The filaments contain a center cartilage
shaft, afferent and efferent arteries and supportive tis-
sues. Projecting at right angles from either side of the
filament are the secondary (respiratory) lamellae (L). In
most teleosts a lamella is eccentrically Shaped and water
flows over its surface as indicated by the large arrows
in Figure 3. A lamella consists of two layers of squamous
epithelial cells sandwiched over pillar cells (Lowenstein,
1962). The pillar cells separate the layers of epithelium
and thus impose a sinus network. Blood flows from the
afferent filamental artery (Fa) around the pillar cells
through the sinus and into the efferent filamental artery
(Fv). Blood flow (small arrows, Figure 3) is counter to
the water flow on the outside of the lamella. Respiratory
EXchange of CO2 and 02 can occur anywhere across the sur-
face of the lamellar epithelium. The base of the lamellae
(Site of attachment to the filament) and the filamental

sPaces between lamellae contain cells responsible for active

65

Figure 3.--Cross section through a gill arch showing the
relationship between the filaments (F), lamellae
(L), and the pathway of water flow (large arrows)
and blood flow (small arrows). Gill rakers (Gr);
filament (F); lamella (L); afferent filamental
artery (Fa); efferent filamental artery (Fv).

66

Gr

 

 

 

 

 

 

67

ion transport. These cells, named "chloride cells" by
Keys and Willmer (1932), are thought to be involved in
active transport of monovalent cations and anions. (For
further description of the structure of the lamellae and
chloride cells, see section on gill ultrastructure on
page 81.)

Whether mercury enters the gill by diffusion through
the respiratory lamellae or is transported by the chloride
cells is unknown. Radioautographs of gills from fish
exposed to either methyl or inorganic mercury exhibited a
diffuse grain pattern which preluded identification of the
uptake pathway. Lack of grain density around the chloride
cell area showed that these cells were probably not
Specifically involved in uptake of either mercurial.
Thinness of the lamellae in cross section and close apposi-
tion of adjacent lamellae (small interlamellar distance)
prevented discrimination of preferential mercury binding
Sites or uptake pathways on the lamella. Dense grain
patterns due to 203Hg++ were found around the filamental
cartilage. Presumably the mercury is associated with the
chondroitin sulfate found in the ground substance of
cartilage (Leeson and Leeson, 1970).

Microprobe analyses of tissues exposed to mercuric

chloride for 24 hours revealed substantial amounts of

68

mercury in the interlamellar space (see Figures 4c and 5).
Figure 4c shows the secondary electron output (SEM image)
from a cross section through several lamella and the fila-
ment. The horizontal line depicts the beam tracing which
corresponds to the line scan in Figure 5. AS the beam
moved from left to right in Figure 4c it traversed the,
filamental sinus (Richards, 1968), entered the filamental
tissue and passed through the basal area of the lamellae
on the filament where the chloride cells are located (see
arrow). The beam then crossed three lamellae (white
areas) and the associated interlamellar spaces. The line
scan (Figure 5) shows the relative distribution of the
natural elements Na, K, Ca and C1 and their relationship
to mercury. The abscissa is divided by segments which
correspond to the large squares in Figure 4c (10 squares
full scale). The ordinate corresponds to the relative
intensity of X radiation emitted by each element. Quan-
titative comparisons of elements are prevented by the use
of different sensitivities for each element and displace-
ment of the base line for graphic display. The left edge
of Figure 5, between the ordinate and the first division,
corresponds to a blood vessel or sinus space which is off
the field of view in Figure 4c. Tissue blood space is

indicated in Figure 5 by the tracings for sodium and

69

Figure 4.—-Electron microprobe: Tissue scan. Figure a,
SEM cross section of filament. Figures b, d,
and e, elemental scans for potassium (K),
calcium (Ca), and phosphorus (P). Figure c,
SEM of longitudinal section of filament (400x).
Figure f, areas scanned by microprobe shown as
dark squares (40x). Arrows in a and e corre-
spond to probable location of filamental sinus.
Arrow in c points to chloride cell region and
arrow in f to filamental cartilage. Black and
white lines in a and c depict beam path for
line scan.

70

n u... \ .
. .3 .
\ {Ant 8).

. .. \ K.
HU‘(-~.F. 41.9

 

71

.ov GH sumo Emma mcon mCOHmH>Hp Uwum o» pcommmuuoo mconH>Hp

mmmflomnd!-.mcaa 09H£3 can roman mm ow madman ca comm on coo doom How

ream Emmm .AHUV wcHHoHno «Amov ESHUHDU “Ammv husoumfi “Amy Esflmmmuom
“Amzv Esaoom .AOe musmflm Eoumv coauomm mmouo HDHHDEDH mo coon mSAQII.m musmam

72

..-._-_ . I '

     
      
    
   
  

 

.1
‘
Err-3.23...
‘oop-ouvv' ‘ ' ‘
COUQO“ .
u "-"5‘321-i‘...
.1:

d

' “ON-11'

.li'

 
 

.-'
.........

- - .. .‘Jl‘

§
.M‘M--.-‘_
--

 

 

 

73

chloride. It was found later that sodium is a very poor
choice for plasma marker because in the vascular space it
becomes very volatile when subjected to an electron beam
and is often lost from the tissue. Why large quantities

of chloride are not found in this Space is unknown. Some
could have been lost by diffusion although this is unlikely
since chloride peaks correspond well to the location of

the lamellae. The strong mercury peak at the left in
Figure 5 suggests that the paucity of other ions in this
area does not reflect tissue separation. Relatively high
levels of mercury in the blood following exposure (see
Table 3) would produce a mercury peak similar to that which
was observed.

As the beam entered the area of the chloride cells,
concentrations of all ions except mercury increased.
Mercury appeared to be inversely related to concentrations
of elements and did not increase in concentration until
the beam had passed through the tissue and entered the
space normally filled by the environmental water. This
confirms results from the radioautographic studies and
supports the concept that inorganic mercury is not
selectively taken up by chloride cells nor concentrated
by them in twenty-four hours. Other line scans of chloride

cell areas on both horizontal and vertical axes substantiate

74

this finding. Figure 4f is a light micrograph of longitudinal
sections of Six filaments (40x). The black squares are

areas examined by the microprobe during a routine experi-
ment. Arrow points to the filamental cartilage which has

been sectioned obliquely.

Interlamellar areas contained definite mercury peaks
and there was an inverse relationship between mercury con-
centrations and concentrations of Na, K, Ca and C1--a11
lamellar markers (Figure 5). The mercury peaks are due
in large part to the mercury which was added one hour prior
to sampling and demonstrated the effectiveness of mucus in
this area to trap ions. Presumably mercury is contained
in the interlamellar mucus from which it can either enter
the gill or be sloughed off with the mucus. McKone eE_§1.
(unpublished data) found 79.3% of the gill mercury could
be removed from the gill by an 80% ethanol wash following
exposure of goldfish to 0.25 ppm HgCl2 for three hours.
They concluded that this fraction represented mercury
trapped by the mucus coating the gill. The above author
did not examine gill tissues after the ethanol wash to
verify that only mucus was removed. Gills are very
fragile and it is quite possible that some epithelium was

also removed in the process.

75

Figure 4a is an SEM image of a filament cut in cross
section. Figures 4b, 4d and 4e are tissue scans for
several elements, two of which give good identification of
the cartilage (elements are identified in the bottom center
of each figure). The filamental sinus can be seen just below
and parallel to the bottom edge of the cartilage (see arrows
in Figures 4a and 4e). The beam path for the line scan
can be seen in Figure 4a and the corresponding element
analysis in Figure 6. Again mercury concentrations are
highest in the cartilage and correspond to calcium levels.
Figure 7 shows a tissue scan of a longitudinal section
through a filament. The cartilage is indicated by an arrow
and lamellae can be seen leaving the filament in the center
of the SEM micrograph. Elemental scans are as indicated
in Figures 7b-7d. Clarity of the lamella in the elemental
scans demonstrates the low degree of ion diffusion during
tissue preparation. Inorganic mercury binding to lamellae
exhibited no preferential accumulation in either micro-
probe or radioautographic studies. Thus it appears that
mercury can enter anywhere through general lamellar sur-
face and active uptake is probably of little consequence
in mercury influx. Once in the filament, inorganic
mercury is accumulated by the cartilage. Other gill cells

contain less mercury than the chondrocytes and chondroblasts

76

.mv SH numm Emma macaw mCOHmw>H© Uflnm ou psommmnuoo macamfl>flp
wmmfiomhm .Amov Esfloamo «Ammo mnsonme «Amzv Endpom .AmcHH muanS
pom roman we musmflm ommv coauowm mmono HopsmEmHHm mo zoom DGHAII.m musmflm

77

‘2‘4‘ ra-
.‘\‘.o- ---:'pv

“539'?

‘“.3-?:. .

-----------
m---------‘--

    

"‘- ......
C. .
11:‘ 31"?"
a.
. A! -‘ fl ‘8‘.

fi‘
_ , - -tff'f'sflu '-

- .‘------..---.----- -.-....”;’
a: ' "" ’3. ”33-.
.-. -2 - s. F ....>
O C- -
’3’... -‘--" WIa...

. ° . a;
mflflic. “:45
a. 21:384- ..

-3. ------‘.‘ '35:;

.‘-...~“..xl '01. 10

c to. ’0' 'ru-nu

 

 

78

.maw>wuoommmu .Amv HSMHSm

pom .AHUV msfiuoHno .AMV Enwmmmuom How msmom Hmucwfiwao .Uln mmusmam

.Axooev mmmEfl Sam .m.mnsmflm

.COHuoom mmouo HDHHmEmH mo zoom msmmHBII.> musmflm

79

 

80

and show no preferential uptake of mercury with respect
to a given cell type, i.e. epithelial cells, pillar cells
or chloride cells. It was not possible to detect the
presence of mercury in gills from trout exposed to methyl
mercury using the microprobe. Since gills from trout

203 203

exposed to CH HgCl contain more label than HgCl

3 2
exposed tissues, it would seem that the microprobe analysis
should also demonstrate more mercury after methyl mercury
exposure. Complete lack of mercury after methyl mercury
exposure could only be due to loss of this element during
tissue preparation. Since methyl mercury crystals will
sublime at room temperature, mercury could be lost during
freeze drying, carbon evaporation or in the vacuum column
of the microprobe. All of these processes require a high
vacuum (around 5 microns pressure). The pathway for methyl
mercury uptake through the gill thus remains obscure.

Mercurial Induced Structural Changes
of’the Gill--Light Microscopy

 

 

Gross structural changes of lamellar epithelium as
described by Schweiger (1957), Amend fl. (1969),
Rucker and Amend (1969) and Miettinen et_§l. (1970) were
not observed in the present study.‘ Following exposure
to methyl mercury H and E stained sections of gill fila-

ments showed cellular hypertrophy in the region of

81

chloride cells. Inorganic mercury produced slight
separation of the lamellar epithelium from the pillar
cells. Sections selectively stained for sulfated muco-
substances revealed heavily stained areas on the lamellar
epithelium, and filamental epithelium of all fish exposed
to inorganic mercury. Exposure to methyl mercury did not
produce any definite increase in the stain density when
compared to control fish. Thus the inorganic form of
mercury appears to act as a local irritant on the gill,
producing increased mucus secretion; whereas methyl

mercury does not have this effect.

Effects of Mercury on Gill Ultrastructure

 

Ultrastructure of the
NormaI'Trout Gill

 

 

Before the mercurial induced morphological changes
of the gill can be discussed it is necessary to consider
the normal gill ultrastructure. Of primary interest are
the secondary lamellae and chloride cells and, for clarity,
these two tissues will be discussed separately.

Gross spatial and functional relationships between
the lamellae and chloride cells have been previously dis-
cussed in this section (see "Pathway of Mercury through

the Gill" and Figure 3). Figure 8 is a scanning electron

82

uoxmuaaflo

.xvm .AAV DHHOEMH “Auuv 05mmHu m>fluomccoo “AHUV
.mucmamaflm mo Mama m mo smoumonoflfi aonuomaw mcflcsmomll.m musmflm

 

84

micrograph of a cross section through a gill arch exposing
two filaments (as in Figure 3). The edge of the 12 mm
diameter glass slide can be seen at the left. A gill raker
(Gr) extends from the cartilaginous arch support in the
lower left. Parallel filaments project from the cartilagi-
nous arch support to the upper right corner of the figure
with the lamellae extending upward, perpendicular to the
filament. Filaments from rainbow trout are connected to
each other by tissue on the inner edge of the filament
(Ct) which extends from one-half to two-thirds of the
length of the filament and prevents separation of the
filamental pairs as well as separation of adjacent fila—
ments along the entire arch. Presence of this tissue
appears highly Species specific (some species such as the
bullhead has no connection between filaments) and it may
greatly affect the resistance characteristics of water
flow across lamellae in this area. Direction of water
flow is from the outside of the filament across the
lamellae and into the space between two filaments, thus
the filamental connection as previously described could
hinder water flow in this area.

Scanning electron microscopic studies of lamellae
were possible only in areas when the lamellar surface

could be positioned perpendicular to the electron

85

detectors. The parallel and closely appositioned nature
of most lamellae hindered observation and it became neces-
sary to find areas on the filament where the lamellar
surfaces were exposed much like the pages of an open book
(L, Figure 8). Figure 9A is a higher SEM magnification
of an open lamellar area on a filament and illustrates
several important features of the filament. The eccentric
shape of the lamella is clearly shown and water would flow
from bottom to top across the lamellae in the figure.
Numerous goblet cells (G) are located on the filament;
mucus secreting cells are also found on the lamella but
they are structually different from goblet cells.

Fixation of tissues in 4% glutaraldehyde for SEM
studies resulted in a coat of mucus over the filament
and lamellae as in Figure 9B. Careful examination with
the SEM of mucus coated lamellae revealed scattered open-
ings through which the lamellar surface could be observed
after 50% glutaraldehyde fixation. The mucus coat in
Figure 9B suggests the importance of this secretion in vivo
in providing a protective coating which acts as a buffer
from gross physical and/or chemical trauma as well as serv-

ing as a precipitator of particulate matter.

86

Figure 9A.--Scanning electron micrograph of a gill filament;
50% glutaraldehyde fixation. Goblet cells (G).
450x. '

Figure 9B.--Scanning electron micrograph of filament fixed
in 4% glutaraldehyde. Mucus covers the filament
and lamellae in this figure: compare to 9A
which is free from mucus. 160x.

 

88

Figure 10 illustrates the structural features of the
lamellar surface. Epithelial cells are outlined by raised
cell junctions and are partially covered by scattered
deposits of mucus (M); direction of water flow is indicated
by the large black arrow. In this figure the lamella is
shaped like a right triangle; the edges adjacent to the
right angle correspond to the free edges of the lamella;
junction of the lamella with the filament is obscured but
would correspond to the hypotenuse of the triangle. "In—
foldings" on the lamellar surface are probably due to
shrinkage and rows of bumps (parallel to the lamellar—
filamental junction) are due to either pillar cells or red
cells underneath the epithelium. The small white and black
arrow points to a mucus cell which appears empty probably
due to loss of mucus during fixation. The white spheres,
seen in the lower left of the figure, do not appear to be
mucus (see Figure 13A for higher magnification). Their
nature is unknown.

Higher magnification of the lamellar surface (Fig—
ures 11A and 118) reveals epithelial cells whose outer
surface consists of a random ridge like network of cellular
membrane extensions. Double ridged characteristics of the
cell junctions are shown in both figures (arrows) and can

be seen in cross section in Figure 15A. Epithelial cells

89

.xooaa
.cBocxcs ma GOHuUSSM Hflmnu “moosfi nos mum mmumnmm muHQB one .mommnsm

HDHHDEDH may umoo sz moose mo monoumm poumuumom .mcflcwmo HHmo
moose m mzonm 3ouum muwzz pom rowan “SCAM umme mo coapomuwp m3osm
Bouum xomam .oommusm HmHHmEmH may mo QQDHmOHOHE soupomam mcflccmomnl.oa musmfim

 

91

Figures 11A and llB.--Scanning electron micrograph of
lamellar surface showing Short (A) and long
(B) ridges. Raised epithelial cell junctions
(black arrows) can be seen in both figures.
Note scattered mucus deposits and also nonmucus
structures (white spheres) on the epithelial
cells. 50% glutaraldehyde fixation. Both
figures 2900x.

. v-
. "4:51 .

[>31

 

93

with short ridges (Figure 11A) generally seemed to be
located near the filament; whereas, the long ridges
appeared on cells which were found nearer the periphery
of the lamella.

A cross section through a lamella (Figure 12) shows
the relationship between the epithelial cells (Ep) and the
underlying pillar cell (P) as described by Hughes and
Wright (1970). (Nuclei for these cells stain dark.)
Pillar cells separate the two layers of epithelial cells
and flanges from these cells form a vascular space (V).
Adjacent pillar cells are connected by interdigitating
cell junctions of the flanges (black arrows). (See
Appendix V for details of pillar cell body.) Within the
vascular space on the right, a normal red blood cell
(RC) is shown. Red cells characteristically have an
amorphous cytoplasm with a darker staining nucleus, how-
ever, the plane of this section did not cut the nuclear
structure (see Figure 14A for other red cells). The
basement membrane or basal lamina (Bl) forms a continuous
layer over all pillar cells and separates them from the
epithelial layer. A small sinus space is present between
the basement membrane and epithelial cells. Small vacuo-
lated cells or cell fragments (C) are often found in this
sinus space, in the vascular space or in the process of

moving from one compartment to the other (me). These

.xoom.eH
.4 .ousamuu Amos opossum amass: osm logo spouses some nooxauumoo ecuo ma
9

“coflumxam mpwampamumusHm we .Hmaos: HHmo mam mum Hamo Homamnuflmm pom
Hamo Hmaaflm ca muommno xnmo .GSOSm mum Amzouuo romanv mmmcmam Hamo
Hmaawm mo mSOHuosSh .AoEV Hamo pmumaosoo> Hanan msfiumumwe «ADV
HHmo pmumaosom> HHmEm “Aamv DSHEDH Amman «Aomv Hams vocab pmu AA>V.
mommm Hmasomm> “Amy HHmo HmHHHm “Ammv Hamo amaamnuflmm .maamfima
may nmsounu c0wuomm mmono m mo smmHmOHOHE souuomam scammHEmsmHBII.mH musmflm

95

 

96

vacuolated cells have not been reported in the literature.
Although their vacuolated nature suggests a phagocytotic
function, this remains speculative. Epithelial cells
(Ep) form the outer boundary of the lamella and are the
thickest barrier to diffusion. Epithelial cell nuclei
are usually located just above pillar cell nuclei (Hughes
and Wright, 1970). The outermost area of the epithelial
cells is composed of finger-like protoplasmic extensions
Similar to intestinal microvilli. Tangential sections
and Transmission electron micrographs from this area have
shown these extensions to be ridge like in the guppy
(Straus, 1963) and in several species of marine teleosts
(Hughes and Wright, 1970); the function of these ridges
as stated by Hughes and Wright (1970) is solely to serve
as an anchor for the mucus film which coats the epithelial
surface. These authors believe that any increase of the
surface area due to the ridges would be negated by the
mucus coat which serves in effect to increase the thick-
ness of the epithelium. The outer edge of mucus, being
continuous with the tips of the ridges, would then appear
as a smooth surfaced barrier to diffusion. The author of
this dissertation does not believe that the resistance to
gas diffusion through mucus is as great as that through

the cell plasma membrane or cytoplasm. The osmotic

97

gradient maintained across the outer epithelial surface
(270 mOsm inside-—10 mOsm outside) suggests a relatively
impermeable membrane at least with respect to water and
ion fluxes. If the plasma membrane is less permeable to
diffusion than the mucus, it would act as the limiting
barrier and any increase in its surface area would cer-
tainly increase the effective surface area for diffusion.
Calculations based on dimensions of ridge height and
width (from TEM studies) and on their length (from SEM)
have shown that these ridges can increase the lamellar
surface area an average of 2.47 times (N = 7). Thus the
contribution of these ridges to total surface area could
be very large.

Figures 13A and 13B are scanning electron micrographs
of lamellar surfaces showing both raised (Figure 13A) and
depressed (Figure 133) cell junctions (black arrows) and
possible pores (Figure 133, black and white arrows).
Sanger (personal communication) reports Similar cell junc-
tions and pores in the cell surface near the cell junction
in preparations of the turkey sinus. Existence of pores
in the epithelial cell surface of trout has not been veri-
fied using the transmission electron microscope. Small
"holes" also were found in the center of epithelial cells
(Figure 13A, black and white arrows) but again they were

not observed using the TEM.

98

Figures 13A and 13B.--Scanning electron micrographs of the
lamellar surface Showing the microridges of
the epithelial cells. Black arrows depict
raised cell junctions (A) and depressed junc-
tions (B). Holes appear on the cell surface
(black and white arrows A and B). The large
white sphere in the upper left corner in A does
not appear to be mucus; its function is unknown.
50% glutaraldehyde fix. 7,300x (A) and 14,500x
(B) .

 

100

Normal red cells and the diffusional distance from
them to the environmental water (W) are shown in Figure 14A.
Note the amorphous nature of the erythrocyte cytoplasm
and the nucleus in the red cell on the left.

Figure 14B shows the outer membrane of the epithelial
cell ridges (arrow). Cell cytoplasm in the ridge area is
agranular and absence of vacuoles suggests little if any
Specialized cellular function.

One type of epithelial cell junction is illustrated
in Figure 15B. This raised junction (note the two lobes
above the junction) consists of a tight junction (tj) and
loose junction (lj)and is similar to that reported by
Hughes and Wright (1970) for several Species of marine
teleosts and the septate junction found in molluscan gill
epithelium (Gilula and Satir, 1971). Interdigitation of
the loose junction is very pronounced. Figure 15A shows
the second type of junction. There is lack of paired
ridges above the tight junction and the loose junction in

this figure is an abutment junction.

Chloride Cells.--Chloride cells located on the fila-

 

mental surface between the lamellae or on the basal area
of the lamellae are believed to be the primary cells
involved in active ion uptake by fresh water fish or active

ion excretion by salt water fish. Numerous investigators

101

Figure 14A.--Transmission electron micrograph of a cross
section through a lamella. Red cells (RC)
are characterized by amorphous cytoplasm.
Environmental water space (w) and barriers
to gas exchange are shown. 4% glutaraldehyde
fix; 1% 0504 postfix. LC and UA stain.
12,500x.

Figure l4B.--Transmission electron micrograph of cross
section through epithelial cell ridges. Arrow
points to plasma membrane. 4% glutaraldehyde
fix; 1% 0304 postfix; LC and UA stain.
56,000x.

 

103

.samsu so out ou Axamumom eOno ma

.xmm oomsooemumosao we room .xooo.mm Ame cosoossfl Hams assumssoao

Immune pom pmmflmu pom Axooo.mn Amv coauocsh ucmEDSQm pom pmmmmummo

.Anav coauocsm mmooH “Anuv sowuossn names .msowuocsn HHmo
Hmflamrpflmm HDHHmEmH mo gmmHmOHoHE conuomam scammflfimcmuall.mma cam «ma mmusmfim

.,1.. ,I . V...
\ . “fi‘” .,».§. 4

.. .a...‘ .~. .13.

U, .\ o.

0

 

105

have shown these cells are characterized by the presence
of many mitochondria and a dense network of smooth endo-
plasmic reticulum (SER) which has been shown to be
continuous with the inner plasma membrane (Ritch and
Philpott, 1969). Conte (see Hoar and Randall, 1969) has
reviewed much of the recent literature on these cells.
In salt water fish there is an involution of the outer
cell membrane forming an apical crypt which is filled
with polyanionic mucus. Below the apical crypt is a
granular cytoplasmic region which is devoid of mito-
chondria and SER.

Chloride cells in fresh water fish also contain
many mitochondria and a dense SER network. An apical
crypt is well defined in the guppy (Straus and Doyle,
1961; Straus, 1963) but absent in the stickelback
(Bierther, 1970) and in fresh water salmon (Threadgold
and Houston, 1964). The apical surface of the salmon and
stickelback chloride cells consists of many microvilli
which are covered by a mucus haze. Adaptation of the
fresh water salmon or stickelback to salt water results
in formation of the apical crypt and loss of most micro—
villi associated with the Cl cell.

Figure 16 shows a section through a chloride cell

(Cl) which is located on the lamella of a rainbow trout

106

v .xooo.NH .CHmum mo tam

DA Axflwumom Omo wH Axflm mp>£moamumpsam we .SOHuomm map so mxumE
mmacx an pmmsmo mum mmcfia xumo .Aomv Hamo pmu “ADV Hamo mpHHoHno
osasmuocoooo Alamo Hams Hmaaocusso llama Haoo ooauoaco .uHHosma

o co HHmo mpwuoano HmEuoc m mo ammnmonoflfi soupomam cowmmHEmcmHBII.oH mnsmflm

 

108

gill. There are numerous mitochondria (m) within this
cell, protoplasmic extensions of the apical cell surface
and a depressed cell junction (abutment type) between it
and other epithelial cells. Chloride cells identical to
this were also seen in great abundance on the filament
between the lamellae. Pillar cells and a red blood cell
(RC) are in the bottom right of the micrograph.

Threadgold and Houston (1964) have identified
degenerating chloride cells by the following criteria:

1) vacuolization of mitochondria, 2) vacuolization of the
golgi apparatus, 3) cytoplasmic degeneration--decreased
electron density and 4) inclusion bodies within the cell.
A chloride cell which fulfills many of these criteria can
be seen at the right of Figure 16. The dark lines across
the tissue are due to knife marks.

The apical portion of a chloride cell and the junc-
tion between this cell and an epithelial cell are Shown
in Figure 17B. The numerous mitochondria (M) and smooth
endoplasmic reticular network (tubular network) are pre-
dominant in the chloride cell. The sparsely granulated
apical portion of this cell, however, lacks both the mito-
chondria and SER. A desmosome (d) of the tight junction
is shown and this junction is depressed below the epithelial

cell surface. Haze, present over the chloride cell

109

.xooo.mm .aamym

«a com me .xamuuoa ecoo

wa .xym mpwampamymysHm

we .Apv mEOmOEmmp “AEV

maupaoaooyyz .aflmmu

mmamaoxm coy am mm myom

mHamaoum HHHy>ouan

may maflum>oo mnmm

.EmmHmoymo may aw

mmasamym cam auo3yma

Esasoyymn OHEmmHm

IOpcm ayOOEm .mmamp

myoz .Hamo mpyuoaao

m yo aoyyuom Hmoymm

may mo ammnmouofla
aouyomam aoammemamHBII.mnH

.xooo.m
.Umaflmymas .xay mpha
nooamumbsam we aloe
yamanuy>cm Hmaumyxm
may oy Hamo moose
mo mafiammo “AEV mymH
Imono aflosz .momeSm
HmHHmEmH may no Hamo
moose m.mo ammumouoye
mnsmym conyomam aonmHEmamHBII.<nH musmam

 

111

microvilli, was specific for these cells and aided in their
identification at low magnification.

Figure 18 shows the relative distribution of chloride
cells on the interlamellar filamental surface as seen by
scanning electron microscopy. Lamellae (parallel rows, L)
were removed by microdissection to expose the filamental
surface. Chloride cells with depressed cell junction
(arrows point to several) are quite numerous in this area.

Higher magnification of the chloride cells (Cl) on
the filament (Figure 19) reveals the microvillar nature
of the protoplasmic extensions of the cell surface. Six
mature and several immature (I) chloride cells in this
micrograph also exhibit depressed cell junctions. Chloride
cells are also located on lamellae and they are similar to
filamental Cl cells except that lamellar cells tend to
bulge out from the general lamellar surface (Figure 20B).

In addition to exhibiting the characteristics
described by Threadgold and Houston (1964), chloride cells
from rainbow trout also showed a loss of microvilli on the
apical surface. A degenerating (dcl) and a normal chloride
cell (ncl) are shown in Figure 20A and Figure 21 (A and B)
shows cross sections through several degenerating C1 cells.
Cell vacuolation, mitochondrial disintegration and loss of
microvilli are all distinct features of the degenerating

cells.

112

.xoos .xam mossooasumosam mom
.maoyyoasm HHmo Ummmmummp ha ammumOHOHE myay ca Umamwsmaflymflp

mum aoaas mHHmo mUHHOHao acme may mo 3mm m haao oy yayom mBOuua
.mmum mflay mmomxm 0y aoyyommmyUOHOHE ma Aav Um>OEmH amma m>ma
mmHHmEma .momwusm HmyamEmHHm may mo ammumouofle aouyomam mafiaamomll.ma musmflm

 

114

.xooo.s .xau monsooamumusao mom .IHV Hams
musymEEH HHmEm m mayUsHoafi ammumouoye mHay ca ammm omam mum mHHmo
mpflHoHao Hmayo .flHHH>onoHE Humans may paw mGOHyossm Hamo pmmmmummo
mymuymaoEmp aofla3 mHHmo mcyuoaao 03y oy yaflom m30uu4 .ycmfimaflm

may no AHUV mHHmo mUHHoHao Hmum>mm mo ammnmouoflfi aonyomam mafiaamomll.ma musmflm

 

116

Figure 20A.--Scanning electron micrograph of chloride cells
on filament. Normal chloride cell (ncl);
degenerating chloride cell (dcl). Note loss of
microvilli during chloride cell degeneration.
50% glutaraldehyde fix. 9,000x.

Figure 20B.--Scanning electron micrograph of chloride cells
on a lamella. Chloride cell (Cl). 50% gluta-
raldehyde fix. 2,900x.

 

...-.-nfl ...; .. .
. . I
cap‘ {I’D . r1...

/ .. .
_. x ..
. . 1....
. v:
'\ £-

3 . ..
6 K. ...L.\ 3: (a
V

. . . . V. . .. «I.» . ... (Aw, _
. I r ‘0 \ I . . I: flird . . AMY .‘ x
v . \.. . .. . z . . .u H»:
r l// d v“ f.‘ . Mk

. . .
(I... I . . ...: ..x: a, . . . .
I . . ' . . . . . i 4|" : .
. ..y o M... I 2', u _
. . . . .. .. .. . I... «in. . .
.. . . .o . .. .
. . .. ... . . ..
. y. . . .. u . ., .7 A. , . ...-k
, ... . ,

. ., , . .

p 7 . u . v \

v . ... .

 

 

118

Figures 21A and 21B.--Transmission electron micrographs
of degenerating chloride cells. Note loss of
mitochondrial membranes, increased cellular
vacuolation and loss of apical microvilli.
4% glutaraldehyde fix, 1% 0504 postfix, LC and
UA stain. Both 19,000x.

 

120

The other major cell type found in the epithelium
is mucus cells. On the filament the mucus cells are goblet-
like and with the SEM appear as in Figure 22A (G). Lamel-
lar mucus cells are not of the goblet type. Figure 223
is a scanning electron micrograph of two such mucus cells
on the lamellar surface in different stages of mucin
secretion (arrows) and Figure 10 (black and white arrow)
shows the Opening of another mucus cell on the lamellar
surface. Figure 17A is a transmission electron micrograph
of a mucus cell containing droplets of mucin (m). Note
also the apical opening of this cell (0).

Effects of Methyl Mercury
on Gill Ultrastructure

 

 

Exposure to 0.05 ppm Hg (as CH HgCl) for 24 hours

3
produced degenerative changes in the chloride cells and
the lamellar surface. 'Figure 23 is a scanning electron
micrograph of the filamental area between lamellae.
Lamellae were removed by microdissection and the black
arrows show numerous red blood cells in the lamellae at
the point of section. Both normal and degenerating
chloride cells (Cl, black and white arrows) are found on
the filament as indicated by the presence or absence of

microvilli of these cells. Figure 243 is a higher magni-

fication of two degenerating chloride cells (C1) on the

121

.xoom.~ syom .xau mossuoamumusao mom syom .lmmm .uzouumo

mHHmo moose HmHHmEmH ha coyymuomm msosE mo mmmmym yamumm

IMHU myoz .mmm CH Am3ouymv MHHmEmH may no cam dmm afl yamEmHHm
may no va maamo msosfi mo ammumouoye aouyomam maycamomll.m- cam 4mm mmusmym

 

123

.xooa.m .xfim mpmampamymysHm wom .aowymxym mayyyp mmmsaym HmaamEMH

may aw pmmmmny mmymooyaymym mymowpsy m3ouym aomam .Hmauoa muoe

ma oy mummmmm ymma ym Hamo mpyuoaau .mawymumammmp ma oy Hmmmmm

can HHHy>ouoye Hmowmm ymoH m>ma Am3oyym mywaz can xomaa .aov mHHmo

mpyuoyao mam: .deoa em you “Hommmmov mm mm Ema mo.o oy pmmomxm
ysoya .mmaamEmH ammsyma yamfimafim mo ammnmoyowa aoyyomam maflaamomll.mm mysmym

 

125

Figures 24A and 24B.--Scanning electron micrograph of
trout exposed to 0.05 ppm Hg as (CH3HgC1) for
24 hours.

A. Lamellar surface Shows some loss of micro-
ridge structure; however, epithelial cell
junctions appear normal. 50% glutaraldehyde
fix. 2,900x.

B. Chloride cells on filament show loss of
microvilli and appear to be degenerating.
Black arrows point to erythrocytes trapped in
the lamella. 50% glutaraldehyde fix. 2,900x.

(‘1

.7 \V ‘I'vl ‘ .'
\fk" \’ 1“ ~ I '\
:1 5...; ». .
‘1‘, .| , \

‘fifi

‘9!" ""1

I

.hd
_ sax

/
’ I

 

127

filament and loss of microvilli is obvious. Ridges on
the lamellar epithelium (Figure 24A) appear to have lost
some of their ordered nature at this mercury exposure.

Prolonged exposure (5 days) to the same mercury
concentration (0.05 ppm) augmented the degeneration of
chloride cells (Figures 25 and 26B) and lamellae (Fig-
ure 26A). Very few chloride cells on the filament in
Figure 25 appear normal; most have lost all of their
microvilli (arrows). Figure 26B shows a normal chloride
cell immediately below the mucus (M) and a degenerating
chloride cell to the left of the patch of mucus. Loss
of ridge-like topography of the lamellar epithelial cells
is also evident (Figure 26A) even though other epithelial
areas were not abnormal. Cellular junctions (Figure 26A,
arrows) appeared normal and indicated optimal beam and
detector angles of the SEM.

No definite structural changes of the lamella were
observed in cross sections after exposure to 0.05 ppm
methyl mercury for either one or five days. Figure 27
shows a lamella from a fish exposed 5 days to 0.05 ppm.
In-this section even the ridges appeared normal. The
white and black arrow points to a highly membranous
structure in the vascular space and the black arrow to a

heavily stained object in the sinus Space. Both of these

128

.xooa.~ .xyu monsooemsmosao mom

.maamo mpyyoHao maaymymammmp Hmym>mm mymoapafl mzoyym myHaB mam

xomHm .nmmo m sch laoommmov mm on Ema mo.o on oomoaxo noose .Aqo
mmHHmEmH may ammzyma yamEmHHw may mo ammymouoye aoyyomam mayaamomll.mm musmam

 

130

Figures 26A and 26B.--Scanning electron micrograph of
trout exposed to 0.05 ppm Hg as (CH3HgC1) for
5 days.

A. Lamellar surface shows degeneration of
microridges; arrows indicate epithelial cell
junctions. 50% glutaraldehyde fix. 2,900x.
B. Chloride cells on the filament. The
chloride cell below the mucus patch (M)
appears normal. The chloride cell to the
left shows loss of microvilli, indicative of
degeneration. 50% glutaraldehyde fix.
7,300x.

 

132

.xoom.NH .aymym «D tam 0a .xymymom

eOmO my .xyy mphampamumySHm we .Ammv mHHmo Hmyamayamm .Amv maamo
ymaaym .amHm Ummomxm mysoyme yo Hoyyaoo Hmayym Eoym mmHHmEmH

mo mmum msaam yo mysymHsomm> may ca pasom aaaoEEoo mysyosuyw
msoamyaEmE maamya m oy myayom 3oyym myya3 can xomHa may now yomflao
mmampaoyyomam pmamyyampya: am 0y myayom 3oyym xomam .mymooasoyymy
m yo m>yymoypay .myypaoaooywe msoymEsa mayoyaoo aQDHmOMOAE may mo
yamyy may Cy mymooyaywym may tam Umymaosom> hayamfiam my ymyamo may
ca myhooyaymym may amsoayam HmEyoa mummmmm maamEmH maB .mwmp m
Mow Aaummmmuv mm mm Ema mo.o 0y ammomxm ysoyy m Bony MHHmEmH

may amsoyay aoflyomm mmoyo m mo ammymoyoyfi aoyyomam aoymmHEmamyBII.nm musmym

 

134

objects were found scattered throughout lamellae of all
fish studied; their function is unknown. The red cell in
the center of the figure (unlabeled) appears abnormal in
that the outer membrane has become somewhat irregular and
the cell cytoplasm has distinct vacuoles. The "ghost"
red cell at the right of this figure contains numerous
mitochondria which is indicative of a reticulocyte.
Increased erythrocyte degeneration, assuming vacuolation
is commensurate with degeneration, would logically increase
circulating reticulocytes due to erythropoietic responses
of the fish.

Low ambient mercury levels (300 ppt, Hg as CH3HgC1)
produce similar changes of the gills and red cells if
exposure times are increased. Four week exposure produced
red cell vacuolation and slight loss of ridge integrity
(Figure 28). Vacuolation of two red cells (RC) is clearly
shown and the vascular space (V) is filled with debris,
possibly contents from ruptured red cells. A chloride
cell (C1) is located above the vascular space and appears
normal; however, ridges on the epithelial cell adjacent
to the chloride cell appear reduced in height and density.
Figure 29A shows two vacuolated red cells which were found
in the afferent filamental artery and indicate.that these
cells are present in circulating blood and not merely

trapped in the lamellar sinus.

135

.xoon.m .aymym
«D can DA .xflmymom eOmo my .me mpmampymymysym we .HMEHOS mummmmm
madman mflay CH AHUV Hamo mpyyoHao mae .A>V mommm Hoysomm> may ca
mflyamp tam Aumv mmymooyaymym may mo aoflymaosom> may myoz .mxmmz v
How Aaummmmuv mm mm yam oom oy ammomxm ysoyy m Eoym mHHmEmH

m amsoyay aoyyomm mmoyo m mo ammumonoye aouyomam aowmmHEmamnell.wm mysmym

136

 

137

Figure 29A.--Transmission electron micrograph of
vacuolated red cells found in the afferent
filamental artery. Trout was exposed to
300 ppt Hg as (CH3HgCl) for 4 weeks. 4%
glutaraldehyde fix, OSO4 postfix, LC and
UA stain. 27,000x.

Figure 29B.--Transmission electron micrograph of a
lamellar cross section from trout exposed
to 300 ppt Hg as (CH3HgCl) for 4 weeks.
A leukocyte (L) has moved into the sinus
space between two degenerating chloride
cells. 4% glutaraldehyde fix, OSO4 postfix,
LC and UA stain. 10,000x.

 

139

Degenerating chloride cells in fish exposed to
methyl mercury (300 ppt, 4 wks) are also present in
greater numbers than in control Specimens. Often a leuko-
cyte is observed in the Sinus space adjacent to degenerat-
ing C1 cells (leukocyte (L) is between two chloride cells
in Figure 29B) and is probably involved in phagocytosis
of either chloride cells or "old" epithelial cells.

After an eight week exposure to 300 ppt methyl
mercury at 5° C, a new cell type appears on the lamellar
surface and is characterized by low electron density,
relatively large cell volume, large spherical nucleus and
reduction of surface ridges. Figure 30 is a micrograph
of a section through two lamellae; two low density cells
are present and have less cellular content than adjacent
cells, one has the large Spherical nucleus (N) (compare
to pillar cell nucleus, B). Three highly vacuolated
chloride cells (cl) are also shown. Erythrocytes (rc)
appear normal in this micrograph; however, other sections
of this same gill did contain vacuolated erythrocytes.
Figure 31A shows large membranous vacuoles (va) in the low
electron dense cells. Presence of many mitochondria and
much SER suggest a secretory function for these cells.
Loss of ridges from the surface of both epithelial cell
types and an absence of intracellular structures in part

of these low density cells were also noticed (Figure 31B).

140

.xoom.h .zflmum «D Ucm DA .xflmumom

Omo wH .xflm mpmcmoamumusHm wv .wpflmcmw couuomam ommmmuomp mo mum

cam mmmcfiuouofle couscou .sz Hmaooc boson cflmucoo SUHQB mmHHoEma

map so UGSOM mum mHHmo Hafiamnuflmm momma .a303m Oman mum Amy maamo

Hmaawm cam Aouv mHHmo own HmEuoz .QQMHOOHOHE may mo uwma mnu

Op comm on coo AHUV maamo mwfluoano mcflumumcmmmc mouse .0 cm um

mxmmB w you Aaummmmuv mm mm umm com on ommomxm psouu mo mmHHmEMH
Hmum>mm cmsounu mCOHuoom mmouo mo nmmumOHOHE aouuomam cofimmfifimcmuall.om musmflm

v

 

142

.xooo.ma um musmflm “xooo.ma "fl musmflm .cflmum «D can oq
.xflmumom «0mo ma .xflu momnmcamumusam we .Axv musuosuum
msocmHQEmE macmfln wmflwwucmcwcs “Am>v mmaosom> “sz
mamaosz .o .m um mxmmz m mow Aaommmmov mm mm pad com
on Ummomxm usouu mo mmaamfima co muflmcmo conuomam 30H

mo maamo HMflHmnuHQm Mo cmmanHOHE coupomam coammHEmcmuBII.mHm can «Hm mwusmflm

 

    

L

.., .u.
am ‘
, aw,

  

3:3.

 

  

144

A highly membranous structure (X) with unknown function is
located in the low density cell of Figure 318. At this
exposure chloride cells become highly vacuolated and were
detached from the subepithelial structures (Figure 32A).
High degree of vacuolation, loss of mitochondrial integrity
and surface microvilli all suggest degeneration of these
cells. Figure 32B shows typical red cell (RC) vacuolation
and presence of vascular debris following mercury exposure.
Exposure to 300 ppt methyl mercury for 8 weeks at
15° C produced similar abnormalities of most chloride cells
shown in Figure 33. The chloride cell in the far right of
this figure appears to be separated from the epithelium
and is completely devoid of microvilli. Loss of lamellar
epithelial ridges was consistently found (Figure 34A).
The chloride cell in Figure 343 shows typical cell vacuola—
tion, mitochondrial degeneration (m) and loss of surface
microvilli. A small vacuolated cell can be seen in the
chloride cell (arrow).

Effects of Inorganic Mercury
on Gill Ultrastructure

 

Exposure of fish to 0.25 ppm Hg as HgCl resulted in

2
distinct abnormalities of epithelial cells. Figure 35 is
a scanning electron micrograph of the lamellar surface

from a fish exposed to HgCl2 for 5 days which illustrates

145

Figure 32A.—-Transmission electron micrograph of a
degenerating chloride cell on the lamella.
Trout exposed to 300 ppt Hg as (CH3HgCl)
for 8 weeks at 5° C. Note extreme vacuola-
tion of this cell. 4% glutaraldehyde fix,
1% 0504 postfix, LC and UA stain. 12,500x.

Figure 32B.--Transmission electron micrograph of a highly
vacuolated red cell (RC) in the lamellar
blood sinus. Trout exposed to 300 ppt Hg
as (CH HgCl) for 8 weeks at 5° C. 4% gluta-

raldehyde fix, l% 0304 postfix, LC and UA
stain. 27,000x.

,.

A

'w-«~Q~o*.
..

 

147

mxoom.e .xflm we»;
umwfimumusfim wom .o ems um mxmzs m Mom Aaomm moo mm mm and com
on ommomxm usoue .psoEmHHm map so Am3onum muwnz cam xomHQv
maamo mofluoHno mafiumumcmmmc mo QQMHmOHowE couuomam mcflccmomll.mm mnsmflm

148
'~ '~ - MW“
“'5 '.?'~.:?”’ ”:9‘r“ ' \f . ‘
n . 3- ’, 0
X 0'

s

s 4r
”-v’

..?

 

149

Figure 34A.--Transmission electron micrograph of an
epithelial cell devoid of ridges. Trout
exposed to 300 ppt Hg as (CH HgCl) for
8 weeks at 15° C. 4% glutaraldehyde fix,
1% 0504 postfix, LC and UA stain. 45,000x.

Figure 34B.--Transmission electron micrograph of a
degenerating chloride cell and amorphous
mitochondria (M). Note small vacuolated
cell on cytoplasmic fragment (black and
white arrow). Trout exposed to 300 ppt Hg
as (CH3HgCl) for 8 weeks at 15° C. Epi-
thelial cell (Ep). 4% glutaraldehyde fix,
l% 0304 postfix, LC and UA stain. 16,000x.

150

 

151

.xoon.m .coHu

Imxwm momnoUHMHMpDHm mom .mHHmEMH map :0 comm on Oman coo maamo

HMHHmanmm buooEm .maamo wUHuoHno omumumcmmmo ou ucflom msouum

muflnz cam MomHm .mmmo m Mom Amaummv mm mm Ema mm.o on Ummomxm
usouu m Eoum ovumusm mHHoEma msu mo QQMHmOHOHE couuomam maficcmomll.mm musmflm

 

153

the propensity of smooth, i.e. non-ridged, epithelial cells
(arrows). These cells, as seen in higher magnification

in Figures 36A and 36B, appear to separate normal epithelial
cell junctions (arrow, Figure 36A) and replace normal
ridged cells on the lamellar surface. (Figure 36B shows
difference between a chloride cell (arrow) and several
smooth cells.) Cross sections of the lamellae (Figure 37)
reveal large epithelial cells with smooth, i.e. non-ridged,
outer cell membranes (S); these cells are not electron
dense. They have a relatively large nucleus, a large cell
body with numerous mitochondria, cytoplasmic vacuoles;
thus, with the exception of the ridged epithelium, they

are similar to epithelial cells from trout exposed to
methyl mercury. No red cell abnormalities were observed

in gills from inorganic mercury exposed fish. Figures 38A
and 38B show the cell junction of smooth epithelial cells
(S) with normal epithelial cells (Ep). The smooth cells
appear to push the normal cells apart (black arrows).

Chloride cells from fish exposed to HgCl have many

2
small vacuoles at the apical edge of the cell (Figure 39).
This is in contrast to methyl mercury exposed fish whose
chloride cells initially become vacuolate from the inner

border of the cell. This difference may reflect different

penetration rates of the two mercurials or different

154

Figures 36A and 36B.--Scanning electron micrographs of
lamellae with smooth epithelial cells. Trout
exposed to 0. 25 ppm Hg as (HgCl2 ) for 5 days.
Arrow in 36A points to separation of normal
epithelial cell junctions and the overlap of
these cells on the smooth cell. 50% gluta-.
raldehyde fix. 5,700x.

Arrow in 36B points to a degenerating chlo-
ride cell. 50% glutaraldehyde fix. 4,700x.

155

-$‘

' ‘.r

r
w.
r

- $

 

156

e .xoom.va

.cflmum <3 cam 0A .xflmumom Omo wH .xflm mcmcmcamnmusam wv

.Aumv HHmo vocab own “Adv Hawo umaaflm .mmmc m Mow Amaommv mm mm

Ema mm.o ou ommomxm usoua .Amv Hamo HMHH¢£UHQ¢ AMOOEm mcflzozm
coauomm mmouo umHHmEma m mo namumouofle couuomam cowmmflEmcmnell.nm musmfim

 

158

Figures 38A and 38B.--Transmission electron micrograph of
the cell junctions (arrows) between a smooth
epithelial cell (S) and a "normal" epithelial
cell (Ep). Trout exposed to 0.25 ppm Hg as
(HgClz) for 5 days. 5% glutaraldehyde fix,
1% 0504 postfix, LC and UA stain (both 38A
and 38B). Figure 38A: 38,000x; Figure 38B:
57,500x.

lg.\‘:\f' ' V
v- .

 

160

.xoom.mH .cwmum fl: can on

.xfiuumom v0mo wH .xflm mossmcamumusam we .mmmo m now Amaommv mm mm

Ema mm.o on ommomxm usoua .mHHmo mums» cflnuwz mmaosom> HHmEm can

mmHmH mo wuwmcwc couuomam muoz .maawama M NO mmmn may no AHUV
maawo mvfluoano mcflumumcmmmv mo ammumonowe couuowam coammwfimcmuell.mm musmflm

 

162

biochemical interactions. Large vessicles found in the
chloride cells from inorganic mercury exposed fish were
not found in those exposed to methyl mercury and may
reflect differential biochemical action of the two mer-
curials.

Long term exposure (4 weeks) to a lower concentra-
tion (0.05 ppm, Hg as HgClZ) produced similar smooth
epithelial cells as shown in Figure 40 (arrows) and
Figure 41B. Areas of the lamella from these fish also
exhibited some changes in the epithelial cell ridges
(Figure 41A); especially noticeable were a decrease in
ridge height and fusion of numerous ridges into a “mat-
like" appearance.

Loss of chloride cell ridges and aberrant forms of
mitochondria in these cells has been shown to occur fol-
lowing a 30 minute exposure of the marine teleost §3g2§_

callarias to 10-8 M rotenone (aberg, 1967). Inhibition

 

of electron transport by rotenone (White et_al., 1968) and
mercury (Webb, 1966) could explain the similarities
between some of the chloride cell abnormalities in the
present study and those found by aberg.

Lindahl and Schwanborn (1971) using a rotary-flow

technique measured the ability of the roach Leuciscus

 

rutilus L. to maintain their position in flowing water

163

.xooa.m

.xfim mcwnmoamnmusam wom .Esaamnuflmm Hmaamama map so maawo nuOOEm

on pcflom mzouud .mxmwz Q How Amaummv mm mm Ema mo.o on ommomxm
usouu m Eoum momwusm umHHoEmH mnu mo nmmnmouofie conuomao mcflccmomll.ov muomwm

 

165

Figures 41A and 4lB.--Scanning electron micrograph of the
lamellar surface from trout exposed to 0.05 ppm
Hg as (HgClz) for 4 weeks. Note loss of micro-
ridge structure (Figure 41A) and appearance of
smooth epithelial cells (Figure 41B). 50%
glutaraldehyde fix. Figure 41A: 2,900x;
Figure 41B: 7,300x.

 

167

("fitness") and correlated this ability with the exposure
history of the fish to environmental mercury. Fish were
taken from two Swedish lakes, one with known high levels
of ambient mercury (muscles of pike in the lake contained
more than 1 ppm Hg) and another with relatively "pure
water." They found that fish from the "pure water" lake
could withstand a significantly greater water flow than
the mercury exposed fish and there was a highly signifi-
cant negative correlation between maximal rate of flow
withstood and muscle mercury concentration. In view of
the mercury induced changes of trout gill ultrastructure
found in the present experiment, impairment of respiratory
efficiency in gas exchange or osmoregulatory capability
by mercurial action on the gill could indeed contribute
to the loss of "fitness" of mercury exposed roach as sug-
gested by Lindahl and Schwanborn.

Uptake and Distribution of Mercury
in the Blood

 

 

Methyl mercury induced erythrocyte vacuolation and
lack thereof following inorganic mercury exposure as noted
in the transmission electron micrographs raised several
questions on the role of blood as the secondary uptake
pathway. In blood, are the two mercurials carried by the

cellular or plasma fraction? Do red cells bind either

168

mercurial stoichiometrically? What are the physiological
consequences of mercury in blood? The following experi-
ments were designed to answer some of these questions.
Uptake Rates and Percent

Incorporation of Mercury
into Red Cells

 

 

The percent uptake into the erythrocyte of both
mercurials as a function of time is plotted in Figure 42.
With the exception of methyl mercury incubated with whole
blood, all incubations attain equilibrium in less than
10 minutes. Methyl mercury did not reach equilibrium when
incubated with whole blood until after 30-40 minutes.
Component analysis (see Appendix II) of this reaction
revealed a two component uptake system, the first component
having a half time of under 1 minute, the second about
11 minutes. This can be explained by initial binding of
a small percent of mercury to plasma proteins or thiols,
the remainder being taken up by the red cells as a function
of the free or non-plasma bound mercury. After the red
cells have removed most of the unbound mercury from the
plasma a new equilibrium (slow component) is established
which is dependent on the dissociation of bound mercury
from "plasma sites" prior to red cell uptake. This postu-

late is further supported by uptake of methyl mercury into

169

.mucflom mump man no umumEMflp

map can“ HmHHmEm mumz >03» mm gamma on» no c3onm no: mum.muonum
pnmoamum .sz msowum>ummno mo M63552 ..Naomm Hmuou Ema vmo.o

cam HumMmmU 8mm Hmuou mmao.o mamumfiflxonmmm macaumuuawocoo whamomxm

.Avv Hmmcwm pmummmsm
mumnmmonm ca noncommsm mHHmo nonmmz suw3 pmumnzocw «Homm "nu

“Avv
moonwm uuonnd cw popsmmmSm maamo pmnmm3 sufl3 cmumnsocfi «Homm "Av

“Awe
“madam uponnfi cw Umpcmmmsm mHHmo pmnmmz Suez pmumnsosfl Hummmmo "AV

:3 683 30:3 an ~88: u.

2ch 683 393 5 63385 Hommmmo v.4

.mEHu mo cowuocsm 6 mm maawo

pooan pom ma musoume ucwam>wp no husonwfi Hmnuma mo mxmums usmonmmlu.mv musmflm

170

2.5 ms:

 

 

     

3r! 1» on» 117 am $10.9 fl 0..
.---:11»--------.o.----i?-----¢.--«T-.O ........... §-------:O-PQL-o.
5...
...On
5..
13

0--._=.-- -- -.Q. ------- b --------- 0\

Q ----- ‘F-I-uo --------------- AW --------- Q- -

 

 

 

088 A8 BMVidfl %

171

cells suspended in Ringer solution. In this instance the
uptake rate is so fast that one component can be determined.
Increased CH3Hg+ binding at equilibrium by washed cells in
Ringer solution compared to cells in plasma (95% compared
to 90%) is probably indicative of some irreversible

mercury binding to plasma protein. Failure of the washed
cells to bind 100% of the mercury could be due to slight
hemolysis coupled with some mercury binding by the incubat-
ing media and glassware.

Inorganic mercury incubated with whole blood binds
almost exclusively to the plasma fraction; less than 9%
was taken up by red cells after 1440 minutes. Such a dis-
tribution could explain why in_!izg, methyl mercury pro-
duced ultrastructural abnormalities in erythrocytes and
inorganic mercury did not.

Washed red cells suspended in Ringer solution take
up 9 to 10 times more inorganic mercury than cells do in
plasma. Removal of plasma proteins undoubtedly is the
critical factor. There is also a significant difference
between percent red cell uptake which is dependent on the
suspending media; cells suspended in Abbott Ringer take
up less Hg++ than cells suspended in phosphate buffered
Ringer solution. Since the only difference between

Abbott and phosphate buffered Ringer (PBR) is the presence

172

of magnesium and buffering capacity of the PBR, either

the magnesium or pH (perhaps both) must enhance equilibrium
binding by red cells. Weed gt_gl. (1962) compared the
effects of buffered to unbuffered solutions of saline on
Hg".+ uptake by human red cells and found no effect even if
the pH of the unbuffered solution was decreased to 6.8.
Thus magnesium or some other unaccounted for parameter

must be the important factor(s).

Less than 100% of the available methyl or inorganic
mercury was taken up by red cells. LeFevre (1947) found
that the effect of mercury on the uptake of sugar by
mammalian red cells was reduced if the cells were allowed
to stand for some time before addition of mercury. He
postulated that there was secretion of a mercury chelator
by the cells which after a period of time reached suf-
ficient levels to bind enough mercury to effectively reduce
the free mercury available for cellular uptake. weed 23_
El. (1962) found similar results with respect to inorganic
mercury uptake, i.e., the longer the cells were incubated
without mercury, the less mercury would be taken up by
them after this ion was added. In the present experiments
percent mercury uptake before and at equilibrium was
calculated for cells to which mercury was added within

1 minute after final suspension or whole blood sampling

173

and for cells which had been allowed to stand in Ringer
solution or whole blood for 45 minutes prior to addition
of mercury. Uptake of either methyl or inorganic mercury
was in all instances independent of the length of the
incubation period which preceded addition of mercury to
the cell suspension. Thus, secretion of a chelator by
trout red cells is not a factor in the failure of these
cells (especially washed cells) to bind all mercury
present.

In an attempt to determine the total number of bind-
ing sites in the red cell, washed cells were incubated for
90 min with increasing concentrations of mercury. Weed
g£_gl. (1962) have shown that the percent of inorganic
mercury bound to the red cell decreases as ambient mercury
increases, which is indicative of saturation of specific
binding sites on or within the red cell. Figure 43 shows
the results of his experiment in which increased levels of
ambient mercury produced a characteristic saturation
curve (Figure 43, open circles) and percent mercury bound
decreased from 97 to 39 percent. These data do not cor-
relate with methyl (open triangles) or inorganic (solid
triangles) mercury binding from the present study as the
present mercury bound remained constant for all levels of

ambient mercury examined. Ambient mercury levels higher

174

.mucfiom pump mo HmumEMHp
swap Hmaamsm mum muouuo pumpcmum .sz mcowum>ummno mo quEdz

Amy mophoounwmum amen: pmnmm3 .Ammmav .Hm um pwmz Eoum pump u o
<
4

Avv moumoounumum upon» Umnmmz .musouma owcmmuo

Avv mmumoounuhum usouu cosmMB .musoume oflsmmuoqfl

.mHm>mH muscuwe ucmHnEm mo cofiuo:5m
6 mm coauomuw HHwo pooHQ con man CH musouwfi mo coflumupchGOUIl.mv ousmflm

175

«0.x ......
c
—

a... ham-084 90..
N

 

«9

cl-
zonx de can 901

I
'0

It

 

176

than those plotted produced excessive red cell hemolysis
in the trout blood experiments; Weed reported essentially
no hemolysis for human erythrocytes irrespective of
ambient mercury concentration.

It appears that saturation of trout erythrocytes
with mercurials either must involve some group (SH, amino,-
acidic sugars, mucopoly saccharides, or other) which is
responsible for cell membrane integrity and is not satu-
rated at the time of hemolysis or mercury levels below total
saturation sufficiently disrupt ionic balance and produce
osmotic hemolysis.

, Consistent RBC unbound/bound ratios for either
mercurial at all ambient mercury concentrations studied
is unexplainable in terms of a specific binding site
theory since at low mercury concentrations a greater per-
centage of the available mercury should bind with the
cell. Why 5% of the methyl mercury and approximately
25% of the inorganic mercury fail to bind to the red cell
irrespective of a one hundred thousand fold increase of
ambient methyl mercury or a three thousand fold increase
of ambient inorganic mercury concentration is unknown.

Sulfhydryl containing compounds such as albumin,
cysteine and reduced glutathione failed to remove either

inorganic or organic mercury previously bound to red cells

177

 

MfiOHHM>H0mQO

 

 

 

 

 

 

mo Honesz
A2 Ioa x hm.av
IIIIIII mm.mumm.vn Hw.HHmm.mn Nv.mnmm.nv no.mnom.aw .m w pwospom +
cowumuomuoocH w
A: IOH x hm.Hv
IIIIIII mm.awnm.nn mn.aflnm.am mn.anmo.vo om.mnwn.mm wcfimummu +
cowumuomuoocH w
AEm-oa x hm.Hv
III-III mm.HHmm.>n mn.aflmm.mn mm.mflmm.mm wn.m«mm.mm GHESQH¢ +
:ofiumuomnoocH w
masonm mm mo powmmm
.m.m H 0mm ous“
mamaaoEms mn.ouam.nh vm.awmm.vn mm.HHmo.>m na.mnmm.mo coaumuomuoocH w
Houucoo

m m
4.83 3.03 3.3 $6 235 33.8 Eu 1 msmw
m o Ema

30.03 38.: «$4 3.0 $0.0 H u u

 

 

 

coaumuuqmocoo mm

 

 

 

mucoumfi owcmmuocH

 

.maamo own ou mcflpcfin mudouwfi so mucwmm HmnphanSm mo muowNMMIl.m manna

178

wcownumusaw u mmw

“.m.m H mammz

 

mGOHum>Hmmno
mo Honfisz

 

HH.OHmH.mm

mH.0H¢m.mm

A: IoH x hm.Hv
.m w cmoswmm +
coaumuomnoocH w

 

mm.owmm.mm

mH.OHmm.wm

AEvloa x hm.Hv
mcflwummo +

cowumuomuoosH w

 

Hm.ofivv.hm

0H.0Hmm.mm

cam-3 x 2.:
£232 +

coaumuomuoocH w
masoum mm mo uuwmmm

 

mo.owmm.mm

HN.OHmH.mm

.m.m H 0mm oucw
cowumuomuoocH w
Houuaou

 

m.mth

hvm.o

AmHHmo 00H H\mmmfiv
Emu

 

 

 

O.COOH

 

 

mmmH-o

 

Hmuou Ema
GOwumuucoocoo mm

 

hHDQHwE Hmnuoz

 

.Umscfipsooll.m UHQMB

179

(see Table 5). In both experiments levels of sulfhydryl
groups exceeded mercury concentrations at the low ambient
mercury levels, yet all SH contributors failed to remove
any mercury from the cells, which suggests strong mercury-
red cell binding.
Effect of Mercurials onngygen Carrying
Capacity of Red Cells
Methyl and inorganic mercury in concentrations up

to 100 ppm (total Hg conc.) had no significant effect on
the total amount of oxygen carried by red cells (Table 6).

Table 6.--Effects of CH3HgCl and HgClz on oxygen carrying
capacity of washed red cells. MeaniSE (N).

 

 

 

 

 

Ambient
mercury Methyl mercury Inorganic mercury
concentration (ml Oz/g Hb) (ml Oz/g Hb)
(ppm)
0 1.42 i 0.114 (4) 0.50 i 0.054 (4)
1 1.45 1 0.119 (4) '0.62 1 0.040 (4)
100 1.48: 0.163 (4) 0.63 1 0.026 (4)

 

 

 

In spite of these results, effects of mercury on oxygen
transport to the tissues cannot be ruled out. Incubation

times of one to two hours, althoughadequate for saturation

180

of mercury binding sites, might not be sufficient to show
mercurial effects on oxygen carrying capacity resulting
from alteration of ionic balances or sugar transport.
Mercury might also affect the oxygen dissociation curve
which is of great physiological importance and would not
be detectable in studies of oxygen carrying capacity.
Long term exposure to mercurials could alter the number
of circulating erythrocytes or their hemoglobin content.
Mercurials such as para-hydroxymercuribenzoate and para-
hydroxymercuriphenylsulphorate have, for example, been
shown to block the uptake of transferrin bound iron by
reticulocytes in humans (Fielding gt_31., 1969; Edwards
and Fielding, 1971).

Mercury in sublethal concentrations has been shown
to enter teleosts via the gill and have definite conse-
quences on the ultrastructure of this organ. Many ques-
tions remain unanswered as to the physiological correlation
with morphological alterations and the precise pathway and
chemical form of mercury entering the gill and vasculature.
Additional studies are certainly needed on the effects of
environmental levels of mercurials on the various organ
systems of aquatic animals. If man is going to establish
meaningful criteria for water quality, he must set standards
which will not only be safe for him but which will also

insure the safety of all his natural resources.

CONCLUSIONS

The gill is the major pathway for uptake of mercury

as either Hg++ of CH3Hg+) in nonfeeding rainbow trout.

Uptake rates of methyl mercury are higher than for
inorganic mercury at 24 hours. The difference is due
to increased permeation of gill tissue by the methyl-

ated form.

Accumulations of methyl mercury by tissues is highest
for gill followed in decreasing order by kidney,
blood, heart, liver, intestine, stomach and muscle.
Inorganic mercury is concentrated primarily in the
gill followed by blood, intestine, kidney, stomach,

liver, heart and muscle.

Inorganic mercury is trapped by the mucus coat cover-
ing the gills. Uptake into the gill tissue appears

to involve the general lamellar and filamental surface.

Inorganic mercury is not concentrated in and therefore
probably not actively transported by the chloride

cells.

181

10.

11.

182

Inorganic mercury has a greater stimulatory action on

mucus secretion by the gills than methyl mercury.

Methyl mercury induced hyperplasia of cells on the
filament between the lamella and inorganic mercury
induced separation of the lamellae epithelial cells.

Gross necrosis of the gills was not observed.

Protoplasmic extensions of the lamellar epithelial
surface form a random network of ridges which increase

surface area 2.45 times.

Chloride cells, characterized by many mitochondria,
dense SER and presence of apical microvilli are in
a state of constant turn over. Degenerating chloride
cells are typified by cellular vacuolation, amorphous
mitochondria, reduction and vacuolation of SER and

loss of apical ridges.

Ultrastructural changes of gill tissue from trout
exposed up to 8 wks to methyl mercury include loss of
epithelial cell ridges, appearance of large highly
vacuolated epithelial cells of low electron density

and enhanced chloride cell degeneration.

Inorganic mercury exposure (up to 4 wks) produced

ultrastructural changes in trout gill characterized

12.

13.

14.

15.

183

by appearance of large epithelial cells whose outer
membrane was smooth, i.e., non-ridged, and chloride

cell degeneration accompanied by atypical vacuolation.

Ultrastructural abnormalities were found in erythro-
cytes after fish were exposed to methyl mercury but
not after inorganic mercury exposure. Methyl mercury
exposed fish also exhibited greater numbers of
reticulocytes (as evidenced by numerous mitochondria

in the red cell) in the gill vasculature.

In whole blood, 89% of the methyl mercury is found in
red cells compared to 8% for inorganic mercury.

Washed red cells exposed to mercury in Ringer solution
take up 95% of the methyl mercury and 55% of the inor-
ganic mercury from the bath. Equilibrium in all
instances is achieved in under 10 minutes except for
methyl mercury in whole blood which achieves equilibrium

in 40 minutes.

Rainbow trout erythrocytes do not secrete a compound

capable of chelating either mercurial.

Percent of either mercurial bound to red cells was
independent of ambient mercury concentration. Satura-
tion of erythrocyte associated SH groups was prevented

by red cell hemolysis.

16.

17.

184
Mercury bound to red cells could not be dissociated
from them even after addition of thiols to the plasma.

Inorganic and organic mercury have no effect on oxygen

carrying capacity of red cells.

REFERENCES

REFERENCES

Amend, D. F., W. T. Yasutake and R. Morgan. 1969. Some
factors influencing susceptibility of rainbow trout
to the acute toxicity of an ethyl mercury phosphate
formulation (Timsan). Trans. Amer. Fish. Soc.
3:419-425.

Ancill, R. J., E. R. Richens and D. A. Norton. 1970. The
reactions of organic mercurials on the cornea. Trans.
Opthalmol. Soc. U.K. 89:863-866.

Anderson, T. F. 1951. Techniques for the preservation of
three-dimensional structure in preparing specimens
for the electron microscope. Trans. N.Y. Acad. Sci.
13:130-134.

Backstr6m, J. 1969. Distribution studies of mercuric
pesticides in quail and some fresh-water fishes.
Acta Pharm. et. Toxicol. 27:5-103.

Bentley, P. J. 1962. Permeability of the skin of the
cyclostome Lampetra fluviatilis to water and electro-
lytes. Comp. BioChem. Physiol. 6:95-97.

Berlin, M., and S. Gibson. 1963. Renal uptake, retention
and excretion of mercury. I. A study in the rabbits
during infusion of mercuric chloride. Arch. Environ.
Health 6:617-622.

Bierther, M. 1970. Die chloridzellen des stichlings.
Z. Zellforsch. 107:421-446.

Bucci, E., and C. Fronticelli. 1965. A new method for
the preparation of a and 8 subunits of human hemo-
globin. J. Biol. Chem. 240:551-552.

Cameron, J. N., and J. C. Davis. 1970. Gas exchange in
rainbow trout (Salmo gairdneri) with varying blood
oxygen capacity. J. Fish. Res. Ed. Canada 27:1069-
1085.

185

186

Davis, J. C., and J. N. Cameron. 1971. Water flow and
gas exchange at the gills of rainbow trout (Salmo
gairdneri). J. Exp. Biol. 54:1—18.

 

D'Itri, F. M. 1972. Mercury in the aquatic ecosystem.
Institute of Water Research. Michigan State Uni-
versity Technical Report #23. 101p.

Edwards, S. A., and J. Fielding. 1971. Studies of the
effect of sulphydryl and other inhibitors on reticu-
locyte uptake of doubly-labelled transferrin. Br.
J. Haematol. 20:405-416.

Ellis, R. W., and S. C. Fang. 1967. Elimination, tissue
accumulation, and cellular incorpoEBEion of mercury
in rats receiving an oral dose of Hg-labeled
phenylmercuric acetate and mercuric acetate. Toxicol.
Appl. Pharm. 11:104-113.

Evans, D. H. 1969. Studies on the permeability to water
of selected marine, freshwater and euryhaline
teleosts. J. Exp. Biol. 50:689-703.

Fielding, J., S. A. Edwards and R. Ryall. 1969. Effect
of sulphydryl inhibition on the uptake of transferrin-
bound iron by reticulocytes. J. Clin. Path. 22:
677-679.

Friberg, L., and J. Vostal. 1972. (Eds.) Mercury in the
Environment: an Epidemiologican and Toxicological
Appraisal. Chemical Rubber Co. Cleveland, Ohio.
215p.

Giardina, B., I. Binotti, G. Amiconi, E. Antonini, M. Brunori
and C. H. McMurry. 1971. Functional properties of
human hemoglobin and isolated hemoglobin chains treated
with organic mercurials. Eur. J. Biochem. 22:327-
330.

Gilula, N. B., and P. Satir. 1971. Septate and gap
junctions in molluscan gill epithelium. J. Cell.

Goldstein, L., and R. P. Forster. 1961. Source of ammonia
excreted by the gills of the marine teleost Myoxo-
cephalus scorpius. Am. J. Physiol. 200:1116-1118.

187

Goldstein, L., R. P. Forster and G. M. Fanelli, Jr. 1964.
Gill blood flow and ammonia excretion in the marine
teleost M oxocephalus scorpius. Comp. Biochem.
Physiol. 12:489-499.

Halbhuber, von K. J., H. J. Stibenz, U. Halbhuber and
G. Geyer. 1970. Autoradiographische untersuchungen
uber die verteilung einiger metallisotope im darm
von laboratoriumstieren. Ein beitrag zur ausschei-
dungsfunken der panethschen kornerzellen. Acta
Histochem. 35:307-319.

Hannerz, L. 1968. Experimental investigations on the
accumulation of mercury in water organisms. Rep.
Inst. Freshwater Res. Sweden 48:120-176.

Hibiya, T., and M. Oguri. 1961. Gill absorption and tissue
distribution of some radionuclides (Cr-51, Hg-203,
Zn-65 and Ag-llOm. 110) in fish. Bull. Japan. Soc.
Sci. Fish. 27:996-1000.

Hoar, W. S., and D. J. Randall. 1969. (Eds.) Fish Physi-
ology. Vol. I: Excretion, Ionic Regulation and
Metabolism. Academic Press, New York. 465p.

Holeton, G. F., and D. J. Randall. 1967. The effect of
hypoxia upon the partial pressure of gasses in the
blood and water afferent and efferent to the gills
of rainbow trout. J. Exp. Biol. 46:317-327.

Horridge, G. A., and S. L. Tamm. 1969. Critical point
drying for scanning electron microscope study of
cillary motion. Science 163:817-818.

Hughes, G. M. 1966. -The dimensions of fish gills in
relation to their function. J. Exp. Biol. 45:177-
195.

Hughes, G. M., and E. R. Weibel. 1972. Similarity of
supporting tissue in fish gills and the mammalian
reticuloendothelium. J. Ultras. Res. 39:106-114.

Hughes, G. M., and D. E. Wright. 1970. A comparative
study of the ultrastructure of the water-blood path-
way in the secondary lamellae of teleost and elasmo-
branch fishes--Bethnic forms. Z. Zellforsch. 104:
478-493.

188

Jarvenpéa, T., M. Tillander and J. K. Miettinen. 1970.
Methyl mercury: Half-time of elimination in
flounder, pike and eel. Suomen Kemistilehti B
43:439-442.

Joyce, C. R. B., H. Moore and M. Weatherall. 1954. The
effects of lead, mercury and gold on the potassium
turnover of rabbit blood cells. Brit. J. Pharmacol.
9:463—470.

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

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.

Klingenmaier, C. H., M. G. Behar and T. C. Smith. 1969.
Blood oxygen content measured by oxygen tension after
release by carbon monoxide. J. Appl. Physiol. 26:
653-655..

Knauf, P. A., and A. Rothstein. 1971. Chemical modifica-
tion of membranes. I. Effects of sulfhydryl and
amino reactive reagents on anion and cation per-
meability of the human red blood cell. J. Gen.
Physiol. 58:190-210.

Knauf, P. A., and A. Rothstein. 1971. Chemical modifica-
tion of membranes. II. Permeation paths for sulf-
hydryl agents. J. Gen. Physiol. 58:211-223.

Knoll, J. 1962. Sodium and chromate ion movement across
isolated frog and fish skins with the effect of
magnetic fields upon such movement. Ph.D. Thesis.
Michigan State University, East Lansing, Michigan.

Kostenko, V. G., and V. I. Iaksuleva. 1971. Morphological
and functional characteristics of changes in mouse
erythrocytes in'vitro in the presence of bivalent
mercury and cadmium ions. Tsitologiia 13:1475-1482.

Krogh, A. 1937. Osmotic regulation in freshwater fishes
by active absorption of chloride ions. Z. Vergl.
Physiol. 24:656-666.

189

Leeson, T. S., and C. R. Leeson. 1970. (Eds.) Histology.
W. B. Saunders Co., Philadelphia. 525p.

LeFevre, P. G. 1947. Evidence of active transfer of cer-
tain nonelectrolytes across the human red cell
membrane. J. Gen. Physiol. 31:505.

Lindahl, P. E., and E. Schwanbom. 1971.. Rotatory—flow
technique as a means of detecting sublethal poison-
ing in fish populations. OIKOS 22:354-357.

Lofroth, G. 1969. Methylmercury. Ecological Research
Committee Bulletin #4. Swedish Natural Science
Research Council, Stockholm, Sweden. 38p.

Lowenstein, O. 1962. (Ed.) Advances in Comparative
Physiology and Biochemistry. Vol. I. Academic
Press, New York. 392p.

Luna, L. G. 1968. (Ed.) Manual of Histological Staining
Methods of the Armed Forces Institute of Pathology.
McGraw-Hill Book Co., New York. 258p.

Lundgren, K. D., A. Swensson and U. Ulfuarson. 1967.
Studies in humans on the distribution of mercury
in the blood and the excretion in urine after
exposure to different mercury compounds. Scand. J.
Clin. Lab. Invest. 20:164-170.

Maetz, J. 1971. Fish gills: mechanisms of salt transfer
in freshwater and sea water. Phil. Trans. Roy. Soc.
Lond. B 262:209—249.

Maetz, J., and E. Skadhauge. 1968. Drinking rates and
gill ionic turnover in relation to external salinities
in the eel. Nature 217:371-373.

Martin, K. 1971. Some properties of an SH group essential
for choline transport in human erythrocytes. J.
Physiol. (Lond.) 213:647-664.

McKone, C. E., R. G. Young, C. A. Bache and D. J. Lisk.
Rapid uptake of mercuric ion by goldfish. (unpub-
lished data).

Mercury in Man's Environment. 1971. Proceedings of the
Symposium. The Royal Society of Canada. Ottawa,
Ontario. 201p.

190

Mercury in the Environment. 1970. Geological Survey
Professional Paper 713. U.S. Government Printing
Office, Washington, D.C.

Meyer, D. K. 1952. Effects of mercuric ion on sodium
movement through the gills of goldfish. Fed. Proc.
11:107-108.

Miettinen, J. K., M. Tillander, K. Rissanen, V. Miettinen
and Y. Ohmomo. 1969. Distribution and excretion
rate of phenyl and methyl mercury nitrate in fish,
mussels, molluscs and crayfish. Reprinted from
Proceedings of the 9th Japan conference on radio-
isotopes. Atomic Energy Society of Japan.

Miettinen, J. K., M. Heyraud and S. Keckes. 1970.
Mercury as a hydrospheric pollutant. II. Biological
half-time of methyl mercury in four Mediterranean
species: a fish, a crab, and two molluscs. FAQ
Technical conference on marine pollution and its
effects on living resources and fishing. ROme, Italy.

Miettinen, V., E. Blankenstein, K. Rissanen, M. Tillander,
J. K. Miettinen and M. Valtonen. 1970. Preliminary
study on the distribution and effects of two chemical
forms of methyl mercury in pike and rainbow trout.
FAO technical conference on marine pollution and

its effects on living resources and fishing. Rome,
Italy.

Motais, R., J. Isaia, J. C. Rankin and J. Maetz. 1969.
Adaptive changes of the water permeability of the
teleostean gill epithelium in relation to external
salinity. J. Exp. Biol. 51:529-546.

Motais, R., and F. Garcia-Romeu. 1972. Transport
mechanisms in the teleostean gill and amphibian skin.
Ann. Rev. Physiol. 34:141-176.

Muir, B. S. 1969. Gill dimensions as a function of fish
size. Fish. Res. Bd. Canada 26:165-170.

Muir, B. S., and G. M. Hughes. 1969. Gill dimensions for
three species of tunny. J. Exp. Biol. 51:271-285.

Norseth, T., and T. W. Clargsgn. 1970. Studies on the
biotransformation of Hg-labeled methyl mercury
chloride in rats. Arch. Environ. Health 21:717-727.

191

Oberg, K. C. 1967. The reversibility of the respiratory
inhibition in gills and the ultrastructural changes
in chloride cells from the rotenone-poisoned marine
teleost, Gadus callarias L. Exp. Cell Res. 45:
590-602. "'

Olson, L. E. 1969. Glucose metabolism and maintenance of
transparency in ocular tissue of rainbow trout.
Ph.D. Thesis. Michigan State University, East Lansing,
Michigan. ' '

Osmer, B. L. 1965. (Ed.) Hawk's Physiological Chemistry.
14th ed. McGraw-Hill Book Co., New York. 1472p.

Paling, J. E. 1968. A method of estimating the relative
volumes of water flowing over the different gills of
a freshwater fish. J. Exp. Biol. 48:533-544.

Potter, G. D., D. R. Mglntyre and G. M. Vottuone. 1972.
Metabolism of Hg administered as HgClz in the
dairy cow and calf. Health Physics 22:103-106.

Prosser, C. L., and F. A. Brown. 1966. Comparative
Animal Physiology. W. B. Saunders Co., Philadelphia.
688p.

Rahola, T., T. Hattula, A. Korolainen and J. K. Miettinen.
1971; The biological half-time of inorganic mercury

) in man. Scand. J. clin. Lab. Invest., Abstr.,
27 (Suppl. 116), 77.

Rasmussen, H. P., V. E. Shull and H. T. Dryer. 1968.
Determination of element localization in plant tissue
with the microprobe. Developments in Applied Spec-
troscopy 16:29-42.

Reynolds, E. S. 1963. The use of lead citrate of high
pH as an electron opaque stain in electron microscopy.
J. Cell Biol. 17:208-212.

Richards, B. D. 1968. Sodium uptake by isolated perfused
gills of rainbow trout Salmo airdneri with special
reference to patterns of 9111 Blood flow. Ph.D.
Thesis. Michigan State University, East Lansing,
Michigan.

Ritch, R., and C. W. Philpott. 1969. Repeating particles
associated with an electrolyte-transport membrane.
Exp. Cell Res. 55: 17- 24.

192

Rucker, R. R., and D. F. Amend. 1969. Absorption and
retention of organic mercurials by rainbow trout
and Chinook and sockeye salmon. The Prog. Fish-
Cult. 31:197-201.

Schweiger, G. 1957. Die toxikologische einwirkung von
schwermetallsalzen auf fische und fischnahrtiere.
Archiv. fur Fishereiwissenschaft 8:54-77.

Selikoff, I. J. 1971. Hazards of mercury. Environ.
Res. 4: 1-69.

Shapiro, B., G. Kollmann and D. Martin. 1970. The
diversity of sulfhydryl groups in the human erythro-
cyte membrane. J. Cell Physiol. 75:281-292.

Shehadeh, Z. H., and M. S. Gordon. 1969. The role of
the intestine in salinity adaptation of the rainbow

trout Salmo gairdneri. Comp. Biochem. Physiol. 30:
397-418.

 

Silberberg, I. 1968. Percutaneous absorption of mercury
in man: I. A study by electronmicroscopy of the
passage of topically applied mercury salts through
stratum corneum of subjects not sensitive to mercury.
J. Invest. Derm. 50: 323-331.

Silberberg, I. 1972. Ultrastructural identification of
mercury in epidermis. Arch. Environ. Health. 24:
129-144.

Smith, H. W. 1929. The excretion of ammonia and urea by
the gills of fish. J. Biol. Chem. 81:727-742.

Smith, H. W. 1930. The absorption and excretion of water
and salts by marine teleosts. Amer. J. Physiol.
93:480-505.

Stevens, E. D., and D. J. Randall. 1967. Changes of gas
concentrations in blood and water during moderate

swimming activity in rainbow trout. J. Exp. Biol.
46:329-337.

Straus, L. P. 1963. A study of the fine structure of the
so-called chloride cell in the gill of the guppy
Lebistes reticulatus P. Physiol. 2001. 36:183-198.

 

193

Straus, L. P., and W. L. Doyle. 1961. Fine structure of
the so-called chloride cells of the gills of the
guppy. Am. Zoologist 1:392.

Sutherland, R. M., A. Rothstein and R. I. Weed. 1967.
Erythrocyte membrane sulfhydryl groups and cation
permeability. J. Cell Physiol. 64:185-195.

Swensson, A., K. D. Lundgren and O. Lindstrom. 1959.
Distribution and excretion of mercury compounds after
single injection. A.M.A. Arch. Ind. Health 20:432-

443.

Swensson, A., K. D. Lundgren and O. Lindstrom. 1959.
Retention of various mercury compounds after subacute
administration. A.M.A. Arch. Ind. Health 20:467-

472.

Threadgold, L. T., and A. H. Houston. 1964. An electron-
microscopic study of the "chloride cell" of Salmo
salar L. Exp. Cell Res. 34:1-23.

VanSteveninck, J., R. I. Weed and A. Rothstein. 1965.
Localization of erythrocyte membrane sulfhydryl
groups essential for glucose transport. J. Gen.
Physiol. 48:617-632.

Vincent, P. C., and C. R. B. Blackburn. 1958. The effects
of heavy metal ions on the human erythrocyte.
I. Comparisons of the action of several heavy metals.
Austral. J. Exp. Biol. 36:471-478.

Watson, M. L. 1958. Staining of tissue sections for
electronmicroscopy with heavy metal. J. Biophys.
Biochem. Cytol. 4:475.

VVebb, J. L. 1966. Enzyme and Metabolic Inhibitors.
Vol. 2. Academic Press, New York. 1237p.

‘Need, R. I., J. Eber and A. Rothstein. 1962. Interaction
of mercury with human erythrocytes. J. Gen. Physiol.
45:395-410.

‘Nhite, A., P. Handler and L. Smith. 1968. Principles of
Biochemistry. 4th ed. McGraw-Hill Book Co., New

York. 1187p.

194

Yamaguchi, S. 1969. Toxicological studies on organo-
mercuric pesticides. Nippon Hoigaku Zasshi (The
Japanese Journal of Legal Medicine). 23:123-138.
Translated by Fish. Res. Bd. Canada series #1477,

1970.

APPENDICES

APPENDIX I

SOLUTIONS

Dietrichs Fixative:
Glacial acetic acid
40% formalin
95% ethanol

tap water

Phosphate Buffered Ringer (PBR):

NaCl

KCl

NaZHPO4 (anhydrous)
KZHPO4 . H20 '
CaCl2

MgCl2 ° 6H20

S¢rensen Buffer :

A. NaHZPO4

HPO ' 7H 0

B. Na 4 2

2

25 ml

75 ml

150 ml

8.00 g/Liter
0.20 g/Liter
1.15 g/Liter
0.20 g/Liter
0.10 g/Liter

0.19 g/Liter

27.8 9 into 1 Liter

53.0 9 into 1 Liter

Just prior to use mix 19 parts A with 81 parts B.

195

 

APPENDIX II

COMPONENT ANALYSIS OF MERCURY UPTAKE

EQUILIBRIUM CURVE

Uptake rate of methyl mercury by red cells in plasma
did not approach equilibrium as rapidly as the methyl
mercury in washed cells (Figure 42), therefore, it was
necessary to determine if the uptake rate had more than
one component.

The present mercury uptake at equilibrium (Peq) was
determined from Figure 44 at the time when the slope of
uptake rate was zero (i.e., at 180 mins). The percent
uptake at all times prior to equilibrium (P) was sub-
tracted from the percent uptake at equilibrium and this
value (Peq-P) plotted on the ordinate of semi-log paper
against the corresponding time on the abscissa (Figure 44).
The graph obtained in this manner showed multicomponent
characteristics. In order to isolate the components, the
linear portion of the curve (Component II) was extrapo-
lated back to zero time. The extrapolated portion of the
Component II was then subtracted from the curve (Peq-P)
at their respective times to give Component I. As Com-

ponent I is linear or nearly so, it appears that the

196

197

Figure 44.-—Component analysis of methyl mercuric uptake
by red blood cells. See text for details.

198

 

 

O. O
o In.

.r u}; b< £502. R I .0350u ...< .9302. R

 

 

OJ

TIME (MIN)

199

uptake of methyl mercury by whole blood is a two-component
system. Mathematical description of the components can be

written according to the general decay equation as follows:

Component I: PtI = Pole
Where: PtI = Peq - Pt at t1me t
PoI = Peq - Pt at t1me o

e = base of natural logarithm
I = rate constant = 0'693/T1/2 = slope
t = time in minutes
T1/2 = t1me required to change Peq - Pt by one-
half
Component II: can be written in a similar
manner substituting the T1/2

for Component II.

The overall increase in percent incorporation of mercury

into red blood cells with time can be written:

Ptotal = Peq - (Pole It + PoIIe IIt)
Where: Ptotal = percent Hg in the RBC at time t
eq = percent Hg in RBC at equilibrium
PoI = percent Hg in RBC due to Component I

at time t

200

P = percent Hg in RBC due to Component II
011 .
at t1me t

e, I, II are described previously.

APPENDIX III

TRANSMISSION ELECTRON MICROSCOPY

TISSUE EMBEDDING PROCEDURE

Fix tissues 3-6 hrs in glutaraldehyde-S¢rensen buffer

at 4° C.

Rinse twice (30 min each) in S¢rensen buffer.

Post fix in 1% 030 2-3 hrs at room temperature.

4

Rinse in S¢rensen buffer 30 mins. Tissues may be

stored at 4° C for 2 weeks at this step only.

Dehydrate in graded concentrations (30-60 min per
step) of ethanol or acetone as follows: % EtOH/HOH
or Acetone/HOH 10, 25, 50, 75, 90, 95, 100, 100
(keep the last two steps in the desiccator when not

changing solutions).

Infiltrate in 50% Spurrs (Polyscience Inc., Warring-
ton, Pa.), 50% ethanol or acetone for 12 hrs (store
Spurrs in freezer when not in use and be sure to

allow Spurrs to reach room temperature before opening).

201

202

Infiltrate in 100% Spurrs for 12 hrs.

Place tissues in plastic embedding mould or other
suitable receptacle and cover with fresh Spurrs.

at 70° C for 8 hrs.

Store blocks in desiccator until ready to cut.

Cure

APPENDIX IV

SCANNING ELECTRON MICROSCOPY

TISSUE PREPARATION

Excise tissues and fix in 50% glutaraldehyde 3-6 hrs.

Rinse 6 times, 30 mins each, with S¢rensen buffer.
Tissues may be stored up to 4 weeks at this step

only.

Freeze drying:

 

3a.

4a.

Quick freeze tissues in isopentane cooled with liquid
nitrogen. The tissues may be pinned to small styro-
foam blocks with insect pins and floated upside down

on the isopentane.

Once frozen, dry in freeze dryer at below -40° C,

5-10u pressure for a minimum of 12 hrs.

Critical point drying:

 

3b.

Dehydrate in ethanol using 60 min steps of the
following: % EtOH/dist. HOH, 10, 25, 50, 75, 90,

95, 100, 100.

203

4b.

5b.

6b.

7b.

8b.

9b.

10b.

204

Infiltrate with 50% amyl acetate (AA) (Mallinckrodt
Chemical Works, St. Louis, Mo.) in EtOH for 1 hr and

then in two changes of 100% AA for 1 hr each.

Place in critical point dryer (DCPl Critical Point
Dryer, Denton Vacuum, Inc.) and flush with liquid
CO2 until the banana odor of amyl acetate is no longer

present in effluent.

Close outlet valve, wait 5 mins and close inlet valve

when sample pressure is at 900 psig.

Heat sample holder with water at 49-50° C. The CO2
will pass through the critical point at 1050 psig.

Allow pressure to reach 1600 psig.

Hold pressure at 1600 psig for 10 min (do not exceed

1700 psig, bleed lines if necessary).

Release pressure over 5 minute duration. Samples

are now dry.

If any amyl acetate odor is noted in effluent during

step 9b, repeat steps 5b-9b.

APPENDIX V

PILLAR CELL STRUCTURE

Figure 45 shows the structure of the pillar cell
body. The large nucleus (N) is surrounded by a nuclear
membrane which is porous (P). White and black arrow
points to the basal lamina which can clearly be seen to con-
nect to the basal lamina on the opposite side of the
pillar cell (bottom of the figure). Connecting elements
of the basal lamina pass through the pillar cells in
several locations (Hughes and Wright, 1970) and are
believed to have contractile properties or supportive
functions similar to those of mammalian reticular cells

in lymph or splenic tissues (Hughes and Weibel, 1972).

205

206

.xooo.mv .cflmum «a can on .xflmumom 3.08 WA .xnm momamoamu
Imvdam we .AHQV MGHEMH Amman “Amv mmuom HmmHosc “sz mDmHosz
.mpon Hamo HMHHHQ may mo nmmumouowe couuomam sowmmHEmcmueul.mv musmflm

207

.Q.

 

‘3

1301-... 1“
. p1 .
.0. _.

APPENDIX VI

SURFACE FEATURES OF LAMELLAE
FROM THE BLACK BULLHEAD

AND NORTHERN PIKE

Lamellar epithelia from various species of teleosts
were examined with the SEM to ascertain whether the ridges
found on rainbow trout epithelia were characteristic of
all fish. Instead of a common ridge pattern, the lamellar
surface from 10 species studied showed great variability,
from no ridges at all in the common carp, or small "bumps"
in the black bullhead (Figure 46A) to comparatively long
ridges from northern pike (Figure 46B). Teleologically,
the purpose of such wide variation of respiratory sur-
faces is unknown; more studies are needed of other species

before sound conclusions can be drawn.

208

209

Figures 46A and 46B.--Scanning electron micrographs of
the lamellar surface from two species of
teleosts. Figure 46A: black bullhead
(14,500x); Figure 46B: northern pike
(7,300x). Arrow in Figure 46B points to a
chloride cell on the lamellar surface.

50% glutaraldehyde fix (both Figure 46A
and Figure 46B).

 

STATE UNIV.

MICHIGAN
WWII
312

LIBRRRIES
4” III I! IIHIHWIHIHWI
9143

9301068