"a”
i

9
D
.
I
U
0
.
O
\
5
'I
C
I
.
‘
.V
Q
s.
,.
‘I
I;
'-
l

V
s
I
-
o
‘.
.
. -
A
.
-
.
'9
t
‘
'l
.

A

W”. O ‘9;

' 'Llfl'f'l-M‘. v 8'

JP ,"
a. ._"

._
"I
h-

“’1' .9}

~01.

."0NO'O' 0°. 4-0.:

 

o‘.—.o

"O’.’ on

..,,... .

“ "‘ 'r' "‘ ' ' "0 r0. 0 -. - - I v I O own. '0‘ght..9'.».' o'uu'o'co I. my.,o.uo.. .I‘U‘Q‘o'o 'O'WOWOO‘. Qu‘vv"

CLEARANCE 0E ‘ ' _ _.
SAL‘AONELLATVPHWURIUM[1ND : .

CARBON BY THE RETTCULOENDOTH‘LTAL

SYSTEM NNORMALAND .. . _
S!MBA-TREATED.MICE];; ' - ;. " ,

a
- I
‘ -
s
o . t .
V I
\ I - .
o
‘ o
I
o .. ¢
0
A' E
.
O c
.
a
I
‘ v
o
' t I ‘
'
o 0 0|
.
t ‘1
v o
.u .
a I
a
. ‘.
T ". ‘ .. o
- . -
I
' \
. u
I O
.- ‘ . O, o
‘ ‘ O. V
‘ .' ' o
\
O ‘ .
‘ o
I ' ‘ a .
. ' . u. - . 5
'. Q
' a

Theas for the Degree 13111 S f - . _. 1. }
MPBIGAN STATE UNIVERSITY , _ '
RICHARD LEE FREEMAN . : .. A -.
1976  .11,,3..

9
O
I
‘ .1 ‘ -o.
'.‘I1..'.‘P‘.'v I'

t
n
c
.
B
V
' UV
I1 v.
V I‘. .
.
IM.’

-I - ' . . 4.:1‘ * ‘c..- .r ' "
.- t:'.‘l‘.z 0.;‘Vl' JI‘O " D' .I .1 .gzl I'.‘.-'.‘ I..‘..‘:' I'," D, “V.'Ooo-o¢.‘-‘«D'O.‘.od

i; 0 ‘ ‘ . Z,

 
  

 
    

  

 

 

 

ABSTRACT
CLEARANCE OF SALMONELLA TYPHIMURIUM AND CARBON

BY THE RETICULOENDOTHELIAL SYSTEM IN
NORMAL AND SILICA-TREATED MICE

BY

Richard Lee Friedman

Dorénturp crystalline silica was used as an experi-
mental tool to studv bacterial trapping in the isolated
perfused mouse liver. Scanning electron micrographs
(SEM) of livers from silica-treated mice revealed destruc-
tion of Kupffer cells but no other deleterious effects
on the liver. Silica treatment decreased bacterial
trapping of 1010 Salmonella typhimurium by the perfused
liver from 63.5% in normal mice to 31.3% in silica-
treated mice. While silica treatment significantly
decreased bacterial trapping by the perfused liver, SBM
showed that sinusoidal trapping of S. typhimurium still
occurred.

In vivo bacterial clearance<ny. typhimurium from

the blood was also depressed with the phagocytic index

being decreased from 0.092 in normal controls to 0.058
in silica-treated animals. After intravenous injection

of S. typhimurium into normal mice, the majority of the

Richard Lee Friedman
bacteria remaining viable in the host were recovered in
the liver, while in silica—treated mice most were recovered
in the carcass. Silica treatment enhanced susceptibility
to S. typhimurium infection in mice over 100-fold.. These
experiments are the first demonstration of enhancement
of gram negative infection by silica. The results obtained
from in vivo experiments were consistent with the data from
the isolated perfused liver and suggested that this in
vitro technique is an accurate indicator of the state of

bacterial clearance in’the whole animal.

CLEARANCE OF SALMONELLA TYPHIMURIUM AND CARBON
BY THE RETICULOENDOTHELIAL SYSTEM IN

NORMAL AND SILICA-TREATED MICE

BY

Richard Lee Friedman

A THESIS

Submitted to .
Michigan State University
in partial fulfillment of the requirementS'
for the degree of

MASTER OF SCIENCE

Department of Microbiology and Public Health

1976

DEDICATION

To my wife, Ellen Kay Friedman, who has suffered twice
as much through this ordeal and who has put Up with
me during these times
To my parents, Sherman and Dorthy Friedman
To my sister, Cynthia Friedman
To my research advisor, Robert J. Moon

ii

ACKNOWLEDGEMENTS

I wish to express my deepest gratitude and apprecia-
tion to Dr. Robert Moon, my research advisor, for his
-patience, always having time to talk, understanding and
advice during my study.

I also wish to thank Dr. Stanley Fleger and Dr. Gary
Hooper of the Michigan State University Center for Electron
Optics for their technical assistance in preparation of

the scanning electron micrographs.

iii

TABLE OF CONTENTS

Page
INTRODUCTION. . . . . . . . . . . . . . . . . . . . 1
LITERATURE REVIEW . . . . . . . . . .‘. . . . . . . 3
MATERIALS AND METHODS . . . . . . . . . . . .,. . . 29
Animals. . . . . . . . . . . . . . . . . . . 29
Bacteria . . . . . . . . . . . . . . . . . . 29
Silica . . . ‘ . . . . . . . . . . . 30
Mouse Liver Perfusion. . . . . . 30
Carbon Clearance in the Whole Animal . . . . 33
Carbon Clearance in the Perfused Liver . . . 34
Whole Animal Bacterial Clearance . . . . . . 35
Whole Animal Bacterial Killing . . . 35
White Blood Cell (WBC) and Differential
Blood Counts of DQ12 Silica- Treated Mice . 36
Effect of DQ12 Silica on S. typhimurium
Infection in Mice. . . . . 36
Scanning Electron Microscopy (SEM) of
Mouse Liver. . . . . . . 36
Statistics . . . . . . . . . . . . . . . . . 37
RESULTS . . . . . . . . . . . . . . . . . . . . . . 38

Effects of Various Silica Preparations

on Carbon Clearance. . . . . . 38
Scanning Electron Microscopy (SEM) of

Livers from Normal and Silica- Treated

Mice . . . . . 38
Carbon Clearance and Bacterial Trapping

by Perfused Livers of Normal and Silica-

Treated Mice . . . . . . . . . . . . . . . 42
Clearance of S. typhimurium from the Blood
of Normal and Silica-Treated Mice. . . . . 43

The Effect of Crystalline Silica on
Bacterial Killing and the Distribution
of Viable S. typhimurium after I.V.

Injection into Mice. . . . . . 48
Susceptibility to S. typhimurium Infec-
tion in DQ12 Silica- Treated Mice . . . 48

Total White Blood Cell (WBC) and Differen-
tial Counts in Silica— Treated Mouse Blood. 52

iv

Page
DISCUSSION. . . . . . . . . . . . . . . . . . . . . S4

LITERATURE CITED. . . . . . . . . . . . . . . . . . S8

Table

LIST OF TABLES

Effect of different preparations of silica
on RES carbon clearance.

Carbon clearance in the perfused liver from
normal and silica-treated mice

Clearance of viable S. typhimurium by
perfused mouse livers from normal and
silica-treated animals

Survival of S. typhimurium 15 min (4a)
and 30 min (4b) after intravenous injec-
tion into normal and silica-treated mice

White blood cell (WBC) and differential
counts of silica- treated mice. .

vi

Page

39

42

43

49

S3

Figure

LIST OF FIGURES

Scanning electron micrographs of Kupffer
cells from mouse livers. .

Scanning electron micrographs of silica-
treated livers perfused with 1-2 x 1010
S. typhimurium . . .

Rate of clearance of S. typhimurium from
the blood of normal and silica-treated
mice

Susceptibility of normal and silica-

treated mice to infection by 2 x 105
(4a) and 2 x 103 (4b) S. typhimurium

vii

Page

41

45

47

51

INTRODUCTION

Crystalline silica is specifically toxic for macro-
phages (l,3,4,19,44,54). Despite this well established
observation, relatively little work has been done on
the ability of crystalline silica to enhance infections.
Silica, in vitr0,Lp0tentiates growth of Mycobacterium
tuberculosis in macrophage cultures (2) and increases
susceptibility to tuberculosis infection in vivo (15,86,
87).‘ Crystalline silica also enhances susceptibility to
viral (48,95-97) and parasitic infections (46), presumably
by suppression of RES activity. To date, data on the
effect of silica on enhancement of gram negative infec-
tions does not exist.

Three different forms of crystalline silica were
initially tested to determine what dose and treatment
period was most effective in suppressing carbon clearance
'from the blood. Carbon clearance is a common method used
to study the functional state of the RES (9). Dorénturp
silica (10 mg over 3 days) was the most toxic and was used
in subsequent experiments.

The primary objectives of this study were to deter-
mine whether silica treatment decreases the trapping
ability of the perfused liver and whether such treatment

1 .

2
causes decreased clearance and/or enhanced susceptibility
to gram negative infection in vivo. The effects of silica
on the perfused liver were studied using colloidal carbon
clearance, bacterial trapping and by scanning electron
microscopy of liver Kupffer cells. To determine if data
from the perfused liver was an accurate indicator of the
state of bacterial clearance from the blood in vivo, the
effects of silica treatment on in vivo clearance of
colloidal carbon and Salmonella typhimurium were also
performed. Additionally the effect of silica on the

mortality to Salmonella infection was examined.

LITERATURE REVIEW

The reticuloendothelial system (RES) is made up of
fixed phagocytic cells which include liver Kupffer cells,
splenic macrophages, microglial cells of the brain,
pulmonary alveolar macrophages and lymph node macrOphages.
Free or wandering macrbphages in the blood are also
included in this system. The polymorphonuclear leuko-
cytes (PMN) are not included as part of the RES. One
of the major roles played by the RES involves protecting
the host from infection by clearing and killing organisms
that enter the blood (36,75). Other physiological func-
tions carried out by the RES include hemoglobin Uptake
and breakdown (30), protein transport (37), uptake and
metabolism of steroids and other lipids (6,8,22,61),
uptake and degradation of red blood cells (6), phagocytic
clearance of autologous tissue debris (88), endotoxin
uptake and detoxification (27), phagocytosis of foreign
colloidal and particulate material from the blood (9,81),
antigen processing for lymphocytes (28) and destruction
and immunity to tumor cells (7,24,43).

With respect to the role of the RES in host defense,
-Rogers reviewed host mechanisms involved in clearing
bacteria from the blood stream (75). When large numbers

3

4
of viable bacteria were injected into the blood, three -
clearance phases could be distinguished. Ninety to
99.9% of the bacteria were cleared in the first 10 min
to 5 h, the rate depending on the bacteria under study.
This first phase was followed by a period when micro-
organisms persisted in the circulation due to slower
rates of removal. This second phase lasted for a few
'hours to seVeral days. During the final phase, bacterial
numbers in the blood rose and either disappeared com-
pletely or increased to high levels until death occurred.

During the early clearance phase, Rogers stated
that the fixed phagocytic cells of the RES sequestered
the majority of the bacteria. The cells most active in
this process were the liver Kupffer cells and the splenic
macrophages. Other RES cells may also have been involved
to a lesser extent. The liver cleared the largest
numbers of bacteria when clearance was rapid. Sixty
to 95% of bacteria cleared in the first 10 min to 5 h
were sequestered in these two organs.

The degree to which different bacteria were trapped
by the liver and spleen varied from microbe to micrdbe
and was not due to the animal system being used. Staphylo-
cocci were cleared quickly by the RES of dog, rabbit and
rat while E. coli was sequestered less efficiently in
these animals. Without antibody containing serum,
pneumococci were minimally removed by the splanchnic

tissue. The ability of the RES to clear different

5
organisms from the circulation determined the rate at which
bacteria were removed during early periods after intra—
venous injection.

Factors other than the RES were also involved in
maintaining sterility of the peripheral blood, particu—
larly the PMN. Intravenous injection of pneumococci or
staphylococci caused a profound leukopenia with leukocytes
reappearing in the circulation 10 to 20 min later.
Stained smears of blood at this time showed bacteria in
the cytOplasm of the PMN. It seems that these bacteria
caused the leukocytes to stick to endothelial surfaces
in capillary beds. White blood cells lodged in lung
capillaries were very active in clearing the cocci from
the blood. Gram negative organisms, such as E. coli,
caused prolonged leukopenia of 4 to 6 h. Here it seems
more leukocyte damage was caused and that cells that
appeared in the following leukocytosis were new PMN.

With high RES sequestering, the_PMN probably play a
small role in clearance. Polymorphonuclear neutrophils
play a major role when the RES is unable to contain
certain bacteria. The circulating PMN are a major
defense against localized or fixed infections while the
RES protects against circulating organisms.

Howard (36) stated that Wyssokowitsch, in 1886,
demonstrated that bacteria and fungal spores injected
I.V. into rabbits were Cleared by specific endothelial

cells in the liver and spleen. Bull (16), in 1915,

6

studied the fate of typhoid bacilli (Salmonella typhi)
after intravenous injection into rabbits. He found that
after 15 min the number of bacteria per milliliter of
blood dropped from 107 to 40 bacteria/ml. Bull looked
at the distribution of s. typhi in different tissues
21 min after injection and found bacteria mainly in
spleen, liver and lung. The numbers of bacteria were
much lower than were initially in the blood at l min.
He stated that the bacilli accumulated mainly in the
, liver and spleen and wEre taken up by assembled PMN
(actually RES cells) and destroyed.

Manwaring and Coe (52) ran perfusion experiments
on different organs of the rabbit to determine which
organs were involved in clearing bacteria from the blood.
Pneumococci were suspended in Ringer solution with or
without immune serum (1%). The perfused liver, in the
presence of immune serum, completely cleared the per-
fusion media of bacteria. .Without immune serum only
about 25% of the pneumococci were cleared. Intestine,
kidney, and hindquarters were also perfused and few I
bacteria were cleared. The addition of 1% immune serum
did not enhance retention of the baCteria by these organs
as it did in the liver. Manwaring and Coe called the
serum component that causes the pneumococci to adhere
to the liver capillary endothelium, endothelial opsonin.

The "laws” of microbial tissue affinity were further

studied by Manwaring and Fritschen (53) using other types

7
of microorganisms-in perfusion experiments on dogs.
Spleen perfused with serum cleared 60% of the Staphylo-
coccus aureus in the perfusate while the liver cleared
80%. Other bacteria tested had different hepatic
affinities; E. coli 40%, Bacillus biscptius 10%, Bacillus
anthracis 25% and Bacillus lactis aerogenes only 4%.
When immune serum was added to the perfusion, 60% of the
Bacillus lactis aerogenes was cleared by the liver. Heat;
ing immune serum to 600 for 30 min reduced hepatic
affinity for the above microbe from 60% to 15%. This
study showed that bacteria were mainly cleared by the
liver and spleen and that the affinity of bacteria for
these organs was increased by serum factors.

Howard and Wardlaw (21) and Wardlaw and Howard (89),
using the rat liver perfusion technique, studied how
different perfusion fluids affected phagocytosis of
bacteria by liver Kupffer cells. When E. coli was suse
pended in Ringer-Locke solution, 11% were taken up by
the liver. Forty-one percent were cleared when serum
was added.

An attempt was made to delineate what factors were
involved in the ability of serum to enhance phagocytosis
in the perfused liver. Heating normal human serum at
560C for 30 min depressed its opsonic activity for E.
coli but not completely. Serum absorbed with homologous
but not heterologous E. coli depressed the opsonic

activity of the serum. Zymosan absorbed serum also

.8
decreased opsonic activity. By both absorbing with
homologous bacteria and heating serum to 560C for 30
min, opsonic activity was almost completely abolished.
Only 3% of bacteria perfused in this treated serum were
cleared while in untreated serum 41% were cleared by
the liver.

Experiments done on gram positive bacteria (89),
staphylococci and streptococci showed that they were
sequestered in the liver in the absence of serum (85%
cleared). With the addition of serum to the perfusate
liver clearance diminished to between 67 and 54%,
respectively. By these studies Howard and Wardlaw
showed that a heat-labile factor (complement) and a heat-
stable factor (antibody) in serum enhanced phagocytosis
by Kupffer cells in the perfused rat liver. Also they
suggested that properdin may play a role.

Using the isolated perfused rat liver technique,
Bonventre and Oxman studied the phagocytosis and intra-
cellular survival of Staphylococcus aureus and Salmonella
enteritidis (14). Four different experimental conditions
were used: normal rat liver and normal rabbit serum,
normal livers with immune serum, immune livers (livers
from animals immunized with heat killed S. aureus or S.
enteritidis) with normal serum and immune livers and
senmL In the case of S. aureus, the immunological state
of the liver or serum had no effect on phagocytosis or

killing. Clearance and killing were essentially the same

9

under all experimental conditions. Salmonella enteritidis
was found to be sensitive to both immune serum and immune
liver. The presence of immune serum enhanced clearance
and intracellular killing in normal and immune livers.
The rate of clearance and killing was the greatest when
both immune liver and immune serum were used. In this
case only 6% of the perfused Salmonella survived. In
normal serum and liver experiments bacterial levels
increased 113% during the 2 h perfusion.'

Jeunet et al. (40;4l) did liver perfusion experi-

125I-labeled heat-killed Brucella melitensis,

ments using
Salmonella typhosa and Brucella abortus. Salmonella
typhosa and B. melitensis were perfused with either
non-immune rat blood plus Tyrodes solution, immune rat
blood plus Tyrodes or just Tyrodes solution. In all
cases clearance of the bacteria was similar. No clearance
of B. abortus occurred unless specific agglutinating
antibody was present in the perfusate. In vivo experi-
ments showed that S.typhoaaand B. abortus were both
efficiently cleared from the circulation after I.V.
injection. Salmonella typhosa was found mainly in the
liver. The distribution of Brucella abortus was vastly
different, the majority being cleared by the spleen.
Recent studies have shown that the RES was also
involved in clearing fungi from the circulation. Baine

et al. (5) found that Candida albicans, intravenously

injected into rabbits, was cleared quickly from the blood

10

stream. Distribution of the fungi varied depending on
the portal of entry. -If Candida were injected via the
peripheral vein, most organisms were trapped in the lungs.
Injection of the mesenteric vein found clearance pre-
dominantly by the liver. Clearance of C. albicans in
isolated rabbit liver was performed. Ninety percent of
the yeasts were sequestered when perfused in buffer.
Addition of 5% normal rabbit serum increased clearance
to 98%. In their work Baine et al. made no mention of
any killing Of the C. albicans. Sawyer (77), doing rat
liver perfusion and whole animal studies, likewise found
no killing of C. albicans.

Aspergillus fumigatus spores (10) were injected
I.V. into normal rats to study experimental aspergillosis
by Turner et al. (84). This treatment.resulted in a high
incidence of lethal intracranial aspergillosis within
48 to 72 h. Aspergillus fumigatus spores were found to
be rapidly cleared and destroyed mainly by the lung,
liver and spleen. Even with this high rate of clearance
and destruction, Spores were still able to get into
cerebral tissue and cause lethal intracranial
aspergillosis.

More recently, Moon et al. (57) have done a study
to more clearly define what occurs in the perfused liver.
They showed that clearance of S. typhimurium by the
perfused mouse and rat liver was not synonymous with

bacterial phagocytosis and killing by Kupffer cells.

11
They found that no killing of S. typhimurium occurred in
perfusion of mouse and rat liver when the perfusion
media contained only M—199. Greater than 70% of the
Salmonella perfused in the livers were cleared in one
pass when no blood was in the perfusate. By use of A
phase and electron microscopy, bacteria in mouse and
rat livers were found to be trapped in the liver
sinusoids and at times were stacked in the vessels
giving a ”log-jam" appearance. When whole blood was
added to the bacteria and the perfusate in the rat
liver perfusion, over 50% of the Salmonella were killed.
Blood alone caused no killing of the bacteria. These
experiments showed that bacterial clearance in liver
perfusion did not reflect phagocytosis but actually
trapping of Salmonella in the liver sinusoids. In order
for phagocytosis and destruction of bacteria to occur,
whole blood had to be present.

The ability of the RES to clear bacteria or par-
ticulate matter from the blood depends on two factors:.
the state of the RES cells and the presence or absence
of humoral factors (antibody, complement, properdin and
undefined opsonins). Early studies on the kinetics of
RES phagocytosis or clearance were done with many types
of inert colloidal suspensions. Such colloids used
were: colloidal gold, saccharated iron oxide, heat
aggregated serum proteins, lipid emulsion and most

commonly colloidal carbon. Biozzi et al. (9) determined

12

the kinetic laws governing the phagocytosis of colloidal
carbon and found them to be true for all other test
particles as well. By histological examination, 90%
of the colloidal carbon injected intravenously into rats
”accumulated in the liver Kupffer cells and splenic macro-
phages. Phagocytic activity of these cells could there-
fore be measured by the clearance of carbon from the
blood. The early rate of clearance is independent of
the initial carbon concentration. The rate of phagocy-
‘tosis is directly proportional to the blood concentration
and inversely proportional to the amount of carbon
already phagocytized. Thus, particles are phagocytized
according to an experimental function of their blood
Econcentration in relation to time. Phagocytic index
(K) is the measure of.the rate of phagocytosis of
particles from the blood. The phagocytic index is
inversely preportional to the dose ofparticles injected
(D) or KD = cte; where c is blood concentration of
colloid and t is time. To determine the phagocytic
function of the RES, one must use a dose of carbon
greater than the critical dose: critical dose = KD/Kmax
Under theSe conditions (using saturating dose) the
phagocytic index can be determined. '

,Biozzi et al. (9) and Stiffel et al. (81) stated
that the decrease in phagocytic function resulting from
the uptake of a given dose of colloid after its blood

clearance caused a saturation or blockade of the RES

13
cells. In effect the reduction in phagocytosis was due
to physical overloading of macrophages by colloid. They
stated that colloidal clearance was not influenced by
humoral factors.

Much controversy has surrounded the role of serum
opsonins in colloid clearance. Jenkin and Rowley (38)
demonstrated that clearance of bacteria and colloidal
carbon from the blood of mice requires serum factors in
order to be phagocytized by RES cells. Mice were
'.injected with a very high dose of carbon to produce
RES blockade. Then the rate of clearance of either
serum opsonized or unopsonized carbon was studied in
these mice. Unopsonized carbon clearance was greatly
impaired but the Opsonized carbon was cleared at a
-rate similar to that found in untreated mice. From this
experiment Jenkin and Rowley felt RES blockade was due
to exhaustion of serum opsonins for colloid carbon.

Jeunet and Good (39) found that RES blockade to
colloidal gold, in the perfused rat liver, could be
completely reversed to normal by addition of fresh rat
Splasma to the perfusion. Filkens et al. (25,26)
obtained similar results when colloidal carbon clear-
ance in perfused rat liver was studied. They found that
a plasma factor was required for carbon clearance to
occur normally.

Studies done by Murray (59,60) showed that RES

blockade was alSo due to depletion of plasma factors.

14
He found that blockade to a tracer dose of colloid
occurred only when the surface properties of both
blockading and tracer particles were identical.
Colloidal carbon is kept in suspension by use of fish
glue, 3 type of gelatin, and by regular gelatin. In
effect, then, it can be defined as a gelatin-coated
particle. Thus, the blockading effect of carbon is
dependent on the surface properties of the particle,
i.e., gelatin, and net on the nature of the particle
itself. Murray stated-that the decrease in clearance
in RES blockade was due to gelatin decreasing the level
of Opsonins in the plasma. He found that the degree and.
duration of RES blockade were directly related to the
circulating levels of gelatin agglutinins. Experiments
by Normann et a1. (65) seem to confirm Murray's work.
Normann et al. found that carbon clearance was dependent
on the gelatin concentration in the carbon suspension.
As the ratio of carbon concentration to gelatin became
less (more gelatin present), the phagocytic index
decreased in a linear fashion. He stated gelatin mediates
its effect by direct action on the carbon particle sur-
'face. It either alters its_relationship.with blood
constituents or affinity of RES cells for carbon.
Gelatin injected during the course of carbon clearance
ldid not alter the carbon removal rate, thus supporting

the above conclusion.

15

While many authors have shown that colloidal RES
blockade was due to depletion of opsonic factors, other
workers supported the concept that blockade was due to
saturation of the RES cells. Biozzi et al. (12) did
, experiments similar to those of Jenkin and Row1ey (38).
They found that incubation of carbon with normal sera
.from different animals had no effect on the rate of
carbon clearance in animals previously treated with
blockading levels of carbon. Lireman et a1. (50) were
also unable to find serumfactor depletion involved in
RES blockade in the rabbit.

While there is still disagreement on the role of
serum factor involvement in colloidal particle clearance,
their role in bacterial clearance is well established.
Early liver perfusion experiments showed that for most
bacteria, RES clearance was vastly enhanced by the
presence of normal or immune serum (35,38,40,4l,52,53,89).
Moon et al. demonstrated that without the presence of
whole blood no killing of Salmonella typhimurium occurred
in the perfused rat liver (57). Others have also shown
the role of serum factors in RES bacterial clearance
(11,12,58,76).

To further understand the role of the RES in resistance
to infection, experimental depression of the RES has been
employed. Techniques, besides inert colloidal suspensions,
‘ used to suppress RES functions have been: x-irradiation

(6,31), anti-Kupffer cell serum (70), anti-lymphocyte

16
serum (32,71), cyclophosphamide (80), cortisone (64,83,90),
throtrast (91,92), ethyl palmitate (82), and methyl
palmitate (13,23). It is not known precisely how many
of these agents cause their depression of the RES. With
these methods it is doubtful that the RES cells are the
only cells being affected by the treatment. Thus, a
more specific depressant of the RES is required to study
its role in host resistance to infection. One such com-
pound, crystalline silica, has been found to be toxic
only for maCrophage (1344).

Crystalline silica is just one of many different
forms of silica (silicon dioxide) which exists on the
earth. Other types are amorphous or vitreous silica,
diatomaceous earth quartz, sandstone, flint and asbestos.
All forms of silica are composed of silicon and oxygen
atoms arranged in a tetrahedron structure, with the
silicon atom at the center and the oxygen atoms at the
four corners. Each oxygen atom is shared by an adjacent
silicon atom- With this tetrahedral arrangement,
silica has a three-dimensional lattice macromolecular
structure. Most silicas contain SiOH groups (silanol
groups) on the surface. These phenolic hydroxyl groups
can form hydrogen bonds with water, producing an outer
layer of free or adsorbed water on its surface. The
silanol groups can also react with other hydrogen
acceptors (see below). The spatial arrangement of the

tetrahedra can vary and this accounts for the various

17
types of silica. Vitreous silica has an irregular
pattern; crystalline silicas, Such as DQ12 quartz and
tridymite, have a characteristic regular three-
dimensional tetrahedral structure (3,15).

Interest in the toxicity of silica particles increased
when it was realized that silica dust could cause fibrotic
lesions, similar to tubercule granuloma, when inhaled
into the lung. The disease, silicosis, has been reported
common among miners_and is described as the progressive
development of fibrOUS‘ti55ue, in a nodular or diffuse
form, in response to'a dust. Many theories of how siliCa
causes the pathological lesions of silicosis have developed
over the years.

One early hypothesis to explain silica toxicity was
the mechanical theory.' It was thought that the particles
caused damage by the cutting and laceration of tissue
with which it came in contact. Microscopically, silica
particles have very sharp edges and points. Gardner
(29) disproved this theory by showing that silicon
carbide dust, which also has sharp edges like silica,
does not produce silicosis when inhaled by teSt animals.

A second explanation of silica toxicity was the
solubility or chemical toxicity theory. According to
the theory, silica gradually dissolved as it was bathed
in body fluids, liberating silicic acid which acted as
a tissue toxin and caused fibrosis (15,47). Gye and

Purdy (33) were able to produce fibrotic lesions in the

18

liver of mice after injection of silicic acid or
amorphous silica. Kettle (45), by implanting silica in
a sac subcutaneously in rabbits, obtained local fibrosis.
‘King's in-depth study on the solubility theory of sili—
cosis shed much doubt on this explanation (47). King
compared the solubility of different types of silica
with their pathogenicity. He found that pathogenicity
did not correlate at all with solubility of the test
particles. Sandstone was found to have low solubility
and yet was very toxic‘to animals. Twenty Angstroms
silica had high solubility but low pathogenicity. Quartz
had high solubility and high pathogenicity. King, who
initially was a proponent of the solubility theory, felt
from these experiments that a better explanation must be
sought to explain the fibrogenic action of silica.

- The experimental work of Curran and Rowsell (21)
was also evidence contrary to the solubility theory.
Diffusion chambers made from millipore filters containing
silica were placed subcutaneously into rats. After
several months, the implanted area was checked and no
fibrogenic reactions were found. Scheel et al. (78)
found that silicic acid and its salts were non-toxic
and that silicic acid in body fluids were~rapidly‘
eliminated by the kidney.

The autoimmune theory of silicosis was advanced by

Vigliani and Pernis (85). They suggested that the basic'

lesion in silicosis was started by silica uptake by

19

macrophage where it damages the cell. These damaged
cells would then liberate their contained antigens and
stimulate anantibody response. The resultant antigen-
antibody reaction would precipitate at the site of
antigen release, producing the hyaline substance in the
silicotic nodule. Composition of the acellular part of
human silicotic nodules was studied. It was found that
40% of the hyaline substance was collagen and 60%
gammaglobulin.~

Silica has been shown to have an adjuvant effect
on antibody production (69,85).~ Pernis et al. (69)
found that pretreatment of rats or rabbits with tridymite
I.V., before immunization with ovalbumin or horse serum,
greatly enhanced antibody production to these antigens.
They stated that the adjuvant effect of silica may cause
proliferation of the RES and that this might play a role
in silicosis. Vigliani and Pernis (85) reasoned that
lipopolysaccharide liberated from the cell wall of
destroyed macrophages might act as an autoantigen and
initiate the whole pathological process. To date there
has been no evidence to back this assumption. Scheel
et a1. (79) showed that quartz could absorb proteins
and denature them. They proposed that this may result
in the production of a foreign protein reaction in
tissue.

Razumov et al. (72) proposed a theory of silicosis

based on the piezoelectric effect. Quartz cryStals have

20
no center of symmetry and thus are piezoelectric and
manifest optical rotation. He states:

Individual particles of quartz crystals or

groups of particles, when introduced into the

animal tissues, evidently react intimately

with them, transforming the electrical

energy of the fields arising in the course

of metabolic processes in living tissue into

mechanical energy. On the other hand, the

mechanical energy of the organ when trans-

mitted to the implanted crystalline particle,

may thereby be transformed into electrical

energy. In this way, an alternating electric

field may develop around the crystals, which

in turn disturbs the normal course of tissue

metabolism and energy exchange.

The most recent theory of silicosis is that of
Allison (1,3,4) and Page'et al. (67). This theory is
that silica is specifically cytotoxic for macrophage and
after ingestion, the particles interact with the membranes
of secondary lysosomes and make them permeable. Lyso-
.somal enzymes leak into the cytoplasm causing cell
destruction. With the destruction of the macrophage,
lysosomal enzymes and a fibrogenic substance are released
causing the pathological lesion of silicosis.

Crystalline silica has been shown to be a specific
toxin for'macrophage (1,19,44,54). Marks et al. (54)
studied the in vitro effects of different dusts on
guinea pig peritoneal macrophage. Crystalline silica
was found to reduce dehydrogenase.activity in half by
3 days at a dose of 2—3 micrograms per 3 x 106 macro-
phages. !Other dusts like coal or kaolin-had no toxic

6

effect until 800 micrograms per 3 x 10 cells were Used.

21

Experiments of Kessel et al. (44) studied the effects
of tridymite, a crystalline silica, tridymite coated
with amorphous silica, vitreous silica, hematite and
carbon on viability of guinea pig peritoneal macrophage
in vitro. iViability was determined by the rate of lactic
acid production. The number of macrophage and the amount
and size of the particles used in each eXperiment were
kept constant. Tridymite was found to be the most toxic
compound and completely inhibited lactic acid production
after 7 h incubation. ‘Hematite, carbon and vitreous
silica treated macrophages had normal levels of lactic
acid production during the same time period. Coating
tridymite with amorphous silica vastly decreased tridymite
macrophage cytotoxicity. Lactic acid production was near
normal in these treated cells. Kessel et al. showed by
this eXperiment that the characteristic cytotoxicity of
silica was mainly involved with properties of its surface.
Tridymite was also found to be toxic for macrophage of
rabbit, rat, mouse and man. The other test particles
had no toxic effect.

Other phagocytic cell types were studied for crystal-
line silica cytotoxicity. Polymorphonuclear leukocytes
from guinea pigs and rabbits were unaffected by tridymite
after its phagocytosis. No toxicity was shown for monkey
kidney cells, KB cells, HeLa cells, mouse fibroblasts,
guinea pig fibroblasts or for chick embryo skin, liver,

muscle, or heart cells. All these cell types actively

-Jfl

22

phagocytose crystalline silica with no decrease in lactic
acid production. Kessel and co-workers showed that
crystalline silica with an appropriate surface was
specifically cytotoxic to only one cell type, the
macrophage.

. Effects of silica, diamond dust and carrageenan on
mouse peritoneal macrophages were studied by Allison
et a1. (1), utilizing phase-contrast cine—micrography,

I

electron microscopy, histochemical techniques for lyso-

 

somal enzymes and measurement of the release of lysosomal
enzymes into the culture medium. They found that the
initial phagocytosis of the silica and diamond dust was
the same and that phagosomes developed normally in both
cases. Macrophage that took up diamond dust remained
viable for at least 30 h, while those phagocytizing
silica were dead by 15 h. Histochemical studies were
done on cells taking up either silica or diamond dust.
Acid phosphatase was used as a lysosomal enzyme marker.
In cells that engulfed diamond dust, phosphatase

activity was confined within the phagosomes and only

low levels of lysosomal enzymes were released into the .
culture medium. Cells that phagocytosed silica had
diffuse cytoplasmic phosphatase staining and high levels
of lysosomal enzymes were released into the culture
medium. Cells that phagocytosed silica had diffuse
cytoplasmic phoSphatase staining and high levels of lyso-

somal enzymes in the culture medium. Electron

23
micrographs showed that silica, initially after phago-
cytosis, was within phagosomes. By 14 h silica particles
were seen free in the cytoplasm. Other morphological
changes were swelling and rounding up of mitochondria
and even complete disintegration of cytoplasmic structure.
Allison et al. suggested that crystalline silica was

cytotoxic for macrophages because it was taken up readily'

__'“!

by the cell and silica could then react with the mem-

,IJl“
1a

branes surrounding the Secondary lysosomes, making them.
permeable. Lysosomal enzymes and silica could then get
into the cytoplasm, causing death of the cell, and were
released into the culture media. Silica released from
dead macrophages was just as cytotoxic as the original
silica preparation. In vivo these released particles
could be taken up by other macrophages causing.further
killing. This cycle would continue unless the silica
was somehow sequestered. .

Nadler et al. (62) confirmed this phenomenon by
the diaminobenzidine technique with E.M. They found
that macrophage ingesting silica do release their lyso-
somal enzymes into the cytoplasm of the cell. Scanning
electron microscopy by Hope and Friend (34) on macro;
phages cultured with crytalline silica also confirmed
Allison's results.

Allison et al. (1) and Nash et al. (65) have also
presented evidence that crystalline silica in Suspension,

due to its many phenolic hydroxyl groups on its surface,

24
could act as a powerful hydrogen bonding agent. These
phenolic hydroxyl groups could react with secondary
amide groups of proteins causing denaturation (79).
The interaction with phoSpholipids was even stronger and
by this-action could cause lysosomal membranes to become
permeable. The rigid structure of the silica crystal
was found to be important because silicic acid and other -
non-crystalline forms of silica were much less toxic to [
macrophage. As Allison stated:

It seems that the rigid structure of the quartz,

with many hydrogen bonding groups arranged in

a regular and immovable order on its surface,

must be important in damaging cells. Pre-

sumably, the formation of multiple bonds dis-

torts the membrane structure thereby leading

to the breakdown of the membrane.

Support for the concept that H—bonding is important
in silica cytotoxicity also comes from experiments using
poly-Z-vinylpyridine N-oxide (PVPNO) (1,52,65,73).

Allison et al. (1) found that this polymer, when taken

up by macrophage into their secondary lysosomes, protects
the cell from silica that was later phagocytized. The
cytotoxicity of crystalline silica could also be inhibited
when PVPNO was first mixed directly with the particles

to coat them. Nash et al. (65) pointed out that PVPNO

has oxygen atoms which can hydrogen bond with the silanol
groups on the surface of silica particles. PVPNO pre-

vented silica from interacting with the lysosomal membrane,

and in this manner protected the cell.

25

Silica has been used to shOw the role of macrophage'
in allograft rejection (51,68,72). Pearsall and Weiser
(68) showed that intraperitoneal (I.P.) injection of
silica in mice prolonged skin allograft survival time
if given before or after grafting. Silica had no effect
when given at the time of grafting. Lotzova and Cudkowicz
(26) in experiments with allogeneic bone marrow grafts,
and Rios et al. (73) with allogeneic skin grafts, found
that silica prevented rejection of the allograft. PVPNO
administration to the grafted animals inhibited the
immunosuppressive effect of silica and allowed graft
rejection to occur. Keller (42), by the use of silica,
showed that the macrophage plays a role in restricting
tumor cell proliferation in vitro. Zarling et al. (94)
found the same to be true in in vivo experiments.

Only limited studies are available on enhanced
susceptibility to infections in silicotic animals or
man. Human patients with silicosis have a higher than
normal incidence of tuberculosis (86). Vorwald et al.
(87) found that when BCG was administered to silicotic
guinea pigs a progressive tuberculosis developed, while
in normal animals BCG injection did not cause the:
‘disease. The effect of silica on the growth of Myco—
bacterium tuberculosis in macrophage cultures was studied
by Allison et al. (2). Their experiments showed that the
growth of M. tuberculosis H37Rv in macrophage culture was

increased by the addition of sublethal doses of silica.

26
Silica-treated macrophages had earlier multiplication of
the bacteria and cells died and released the organisms
sooner than cells, either infected without silica or
treated with just silica. If these two factors--faster
multiplication and earlier release of bacteria from
macrophage--occur in vivo as well, it could greatly
enhance the spread of tuberculosis in patients with
silicosis.l

The role of the RES in viral and parasitic diseases
has been studied in experimental animals treated with
silica to depress RES functions (46,48,95-97). Zisman
et al. (95-97) treated mice I.P. or I.V. with silica to
determine the role of the RES in viral infections. When
inormal mice were injected I.P. with herpes simplex virus,
only 18% of the mice died (95). Before death the animals
developed focal meningitis but no other organs showed
histological changes. If mice were injected I.P. with
crystalline silica, 2 h before the virus, mortality was
73%. In the silica-treated mice high viral titers
developed in the liver with severe hepatic necrosis.
Silica treatment alone caused no deaths.

Adult mice were normally not susceptible to yellow
fever virus infection by the intravenous or intraperi-
toneal routes (80-100% resistant). When mice were first
treated with silica I.V. and then injected 2 h later with
virus, over 95% of the mice died of the yellow fever

viral infection (96). Similar experiments were done with

27
coxsackie B-3 virus (97). Mice treated with silica had
a total mortality of 70% to viral inoculation, while only
25% of the normal mice succumbed to the infection.

Kierszenbaum et al. (46), in trying to determine the
role of the RES in resistance to Trypanosoma cruzi,
treated mice intravenously with crystalline silica.
Silica was found to decrease resistance to infection with'
virulent blood forms of the parasite. Animals given
silica had both a higher mortality rate and higher levels
of parasitemia than controls.

Recently Levy and Wheelock studied the effects of
silica on immune and non-immune funCtions of mice (49).
Carbon clearance inintravenously injected silica-
treated mice was significantly depressed 2 h after
injection and this depression persisted for 3 days. These
Idata correlate well with the enhanced susceptibility of
silica—treated animals to viral, parasitic and M. tuber-
culosis infections. Silica had no effect on the induc-
tion of interferOn by chlorite-oxidized-oxyamylose—statolon
or by Newcastle disease virus. Intravenous silica did
not greatly alter the number or composition of peritoneal
exudate cells (PEC). Macrophage isolated and cultured
from PEC were found to have a 40% reduction in their.
phagoCytic activity as determined by uptake of opsonized
sheep red blood cells (SRBC) in a 30 min period. Silica
given 1 to 3 days before I.P. injection of SRBC inhibited

the production of splenic anti-SRBC-plaque forming cells.

 

28

No inhibitory effect was noted when animals were treated
2 h before or after SRBC immunization.

Silica inhibited the cell-mediated immune response
of mice to histoincompatible fibroblasts inoculated I.P.,
'when given 3 days before the fibroblasts. The cytotoxic
activity of spleen cells of silica-treated immunized
mice was less than half of the activity in normal
immunized mice. Concanavalin A (ConA) stimulation of
the uptake of 3H-thymidine by spleen cells of silica-
treated mice was first'stimulated and then depressed.
The-response may be due to the combined consequences of
macrophage depletion, release of lymphocyte stimulatory
factors and the action of suppressor lymphocytes. Miller
and Zarkower (55,56) reported that silica dust inhala-
tion in mice alters the activity of macrophage and also
B and T lymphocytes. They found T cell response to ConA
was increased and B lymphocyte response to lipopoly-
saccharide was depressed. 'Levy and Wheelock feel that
these effects on T and B lymphocytes were not due to
the direct action of silica on the lymphocytes but rather
due to impairment of macrophage directed or dependent

functions.

 

MATERIALS AND METHODS

Animals .
Spartan (HA/ICR) female mice (Haslett, Michigan)
weighing 18 to 25 g, were used in all experiments.
Animals were kept under standard laboratory conditions.

Water and Purina Chow were available ad libitum.

Bacteria

Salmonella typhimurium strain SR~11 was used in all
investigations. Eighteen to 24 h cultures of the organism,
grown in brain heart infusion broth, were centrifuged at
8,000 rpm for 15 min. IThe bacteria were resuspended in
M—199 (Gibco, Grand Island, NY) or sterile saline
(depending on the experiment). For quantitative plate
counts, an aliquot was removed and appropriate dilutions
were made in deionized water to prevent bacterial clump-
ing., Duplicate dilutions were plated on tryptose agar
to quantitate the viable organisms. The bacterial
concentration_used in both liver perfusion and whole

10

animal studies was 152 x 10 bacteria/ml and in whole

animal bacterial clearance and killing studies 1-2 x 109

bacteria/0.1 ml.

29

30

'Three types of crystalline silica were studied for
their effects on suppressing RES activity. -The most
effective for this study was Dorénturp silica, particle
size 5 u or less-(DQ12) (74). DQ12 was kindly provided
by Dr. Gustavo Cudkowicz, Department of Pathology, State
University of New York at Buffalo, NY. ‘Ten milligrams
of DQ12 silica was injected I.V. over a 3 day period
into mice before they were used for experiments on the
fourth day after treatment. Crystalline silica (Spex
Industries, Meterchen, NJ, no. 03741) preparation of
particle size less than 5 u was prepared by the Cummings
sedimentation technique (20). Another silica, 5 u min-u-sil
silica (Wittaker, Clark and Daniels, South Plainsfield,
NJ) was also tested. ‘Injection regimen was the same as for
Dorénturp silica except doses of 4 mg to 9 mg were used.
The Silica was autoclaved in powder form and then suspended
in sterile saline. Before injection the silica solution
was exposed to ultrasonic vibration in a Bronsonic Ultra—
sonic cleaner (no. B220, Branson Co., Sketon, CN) to

resuspend the silica evenly.

Mouse Liver Perfusion

 

Before surgery mice were given 200 U of heparin
(Upjohn Co., Kalamazoo, MI) followed by 1.5 mg of pento-
barbital sodium (Haver Lockhart Laboratories, Shawnee,

KN) intravenously. ’A medial abdominal incision was made

31
and the intestine was moved to expose the portal vein
and the inferior vena cava. A silk suture was introduced
under the vana cava and above the right renal vein.
Cannulas were made from ZO-gauge needles which were filed
until the beveled end was blunt. The cannulas were auto-
claved before using.

The portal vein was held taut with forceps approxi—
mately 1 cm from the liver, and a small cut was made in
the vein between the forceps and the liver. A cannula
filled with sterile M—199 was inserted in the vein and
tied in place with the suture. When the cannula was in
‘place, it was attached to Tygon tubing (inner diameter
0.31 inch; outer diameter 0.93 inch) which was connected
to a three-way valve (Becton—Dickinson and Co., model
no. 3160) via an l8-gauge needle. One port of the valve
held a 1 ml syringe containing sterile M-l99 and the
second port was attached to two 50 ml glass syringes
filled with pre-warmed, sterile M—199. Before the
,perfusion was started, the inferior vena cava was cut
below the right kidney, to avoid swelling of the liver-
once the perfusion started. Perfusion medium then
flowed in through the portal vein and out the inferior
vena cava.

The thoracic cavity was then excised and a ligature
was placed under the superior vena cava above the liver.
A small incision was made in the right atrium, and the

efferent metal cannula was inserted and tied in place.~

_ 32
A 10-20 cm piece of Tygon tubing (inner diameter 0.31
inch; outer diameter 0.93 inch) was then attached to the
cannula.

With the second cannula in place, the suture above
the right renal vein was tied, which diverts the flow
of M-l99 into the efferent cannula. During the experiment
the liver was kept moist by bathing it with sterile M—199
and keeping the specimen covered with a petri dish to
slow evaporation.

The liver was then perfused with M-199 until the
effluent was clear of blood. Flow rate during the first
few minutes of the perfusion was carefully adjusted until
the equilibrium of the circulation had been established
and a constant flow rate was maintained. Effluent was
tested for sterility, and any liver effluent having >10
colony-forming units (CFU) of bacteria per milliliter
at this point was omitted from the final tabulation of
data.

After washing, the l-ml syringe on the 3-way valve
which contained M-199 was replaced with a syringe filled
with 1 m1 of S. typhimurium. The bacteria were slowly
and steadily infused and followed immediately by 50 ml of
M-199. Effluent was collected from the efferent cannula
into a sterile 50 ml graduated cylinder. A 1 ml portion
after the completion of the perfusion was obtained. This
aliquot was plated and the data showed that <0.001% of

the infused bacteria continued to be washed out of the

33
liver. Perfusions lasted 30 min or less and were done at
room temperature. Liver oxygenation was not specifically
maintained.
The effluent was kept on ice until quantitative
tryptose agar pour plates were made. The percent of
non-trapped bacteria was calculated by the formula:

number of CFU recovered in effluent

number of CPU infused X 100'

 

The difference between the percentage of recovery in the
effluent and the total infused (100%) suggested the per-
centage of bacteria trapped by the liver.

The liver was disconnected-from the perfusion apparatus
and homogenized in 9 ml sterile deionized water in a
glass homogenizing tube with a teflon pestle and a
TRIR—STIR-R. Appropriate dilutions were made and subse-
quent quantitative tryptose agar pour plates were used to
determine the number of trapped viable bacteria still
remaining in the liver. The percentage of viable bacteria
remaining in the liver plus the percent recovered in the
effluent could then be subtracted from the percentage of

bacteria infused-(100%) to give the percent killing:

a killing = 100% - % in homogenate + % in effluent).

Carbon Clearance in the-Whole Animal

 

Normal and silica-treated mice were tested for their

ability to clear carbon from their blood by the Biozzi

34
et a1. technique with modification (9). A carbon prepara-
tion, containing 100 mg carbon/ml, was used (Pelikan
Carbon Suspension Cll/l43/a; Gfinther Wagner, Hanover,
,Germany). The carbon was diluted to 50 mg/ml in saline.
Five milligrams of carbon was injected intravenously into
each mouse. The mice were bled from the retro-orbital-
plexus at 2 min and 15 min after injection. At each
bleeding 0.05 ml of blood was obtained and lysed in 4.0
m1 of 0.1% NaéCOS. The concentration of carbon was
determined photometrically (O.D.) in a Hitachi Perkin
Elmer spectrophotometer (model no. Coleman 111) using

tungsten light at a wavelength of 650 nm. The phagocytic

index K was determined for each mouse by the equation:

log C1 — log C

Tz‘Tl

and C2 represent the blood colloid concentration

K = 2

 

where C1
at time 1 (T1) and time 2 (T2), respectively. The clear-
ance of carbon from the blood can also be eXpressed in

terms of a biological half-life (TE) in the blood:

 

'Carbon Clearance in the Perfused Liver

 

Liver perfusion was set up as described above. The
liver was then infused slowly with l'ml of M-199 containing
5 mg of carbon instead of bacteria. Fifty milliliters
of effluent was collected. Concentration of carbon

remaining in the effluent was determined by O.D. readings

35

at 650 nm. From the O.D. reading, the number of milli-

grams of carbon in the effluent was determined graphically.

The milligrams of carbon cleared by the liver was calcu-
lated by subtracting the milligrams of carbon in the

effluent from 5 mg of total carbon infused.

Whole Animal Bacterial Clearance

 

Mice were injected intravenously with 1.0 x 109

viable S. typhimurium in 0.1 ml. At 2, 5, 10 and 15 min
after injection, 0.1 ml of blood-wasobtained from the
retro-orbital plexus. The blood sample was diluted in
sterile deionized water and quantitative tryptose agar

pour plates were made. K was determined by the equation:

 

= log CPU 2 min ; log CPU 15 min
> 13 min

T% was calculated as previously described.-

Whole Animal Bacterial Killing

 

Mice injected intravenously with 1.0 x 109 S. typhi—
murium were killed either 15 or 30 min later_by cervical
dislocation. The liver and spleen were removed and
homogenized separately with a teflon and glass homogenizer
in 9 ml of sterile deionized water. The carcass, exclud-
ing the stomach, intestinal tract, skin, paws and tail,
was homogenized in 99 ml of deionized water in a Waring
blender for 3 min. Quantitative pour plates were prepared

from the three homogenates.

36

White Blood Cell (WBC) and Differential Blood Counts
of DQ12 Silica-Treated Mice

 

 

The blood picture of DQ12 treated mice was followed
before treatment and during the period of silica treat-
ment. White blood cell counts were done by bleeding the
mice from a lateral tail vein. The blood was drawn by use
of a WBC pipette and then mixed with a 2% acetic acid
solution. Counts were made by use of a hemacytometer.
Differential counts were done by staining blood smears
via the Wright-blood stain method. One hundred white
blood cells were counted to determine the percent PMN

and monocytes.

 

Effect of DQlZ Silica on S. typhimurium Infection in Mice
Mice were treated with 10 mg or 3'mg of DQlZ silica
and divided into three separate groups of 6 mice. One
group served as a control and the others were injected
I.V. with either 2 x 105 or 2 x 10:5 S. typhimurium. Sur-
vival of the treated animals was observed for 2 weeks.
No deaths occurred in the silica control groups during
this time. The LD50 for normal mice was determined using
the above doses of bacteria plus doses of 2 x106 and -

2 x 104.

Scanning Electron Microscopy (SEM) of Mouse Liver

 

Livers were infused with bacteria as previously
described. Following the perfusion, 5—10 ml of gluta-

raldehyde was perfused (2.5% glutaraldehyde in 0.2 M

37
phosphate buffer at pH 7.4), for 10 to 20 min. Fixed
livers were excised and cut into small blocks and dehy-
drated in sequential 15 min steps with 30, 50, 70, 90
and 100% ethanol. The blocks were allowed to stand over-
night at 4°C in fresh change of 100% ethanol. The dehy-
drated blocks were cryofractured by immersingthem in
liquid nitrogen for l min. The tissue was fractured
with a precooled single edge razor blade held by locking
forceps. The fractured tissue was placed in metal baskets
under liquid nitrogen and transferred to the critical
point dryer. The tissue was dried in an Omar SPC 900/EX
critical point dryer using C02 as the carrier gas. ,The I
dried specimens were mounted on stubs using double stick
Scotch Brand Tape and the Stub edge painted with television
Tube Koat (G.C. Electronics) to prevent charging. The
specimens were coated with gold (200-300 A) using the
EMS-41 Minicoater (Film Vac. Inc., Englewood, NJ) and
viewed in an AMR-900 scanning electron microscope. Micro—

graphs were made with Kodak positive/negative (P/N) film.

Statistics

 

Where appropriate, data were evaluated by the White

Rank Order method (93).

RESULTS

Effects of Various Silica Preparations
on Carbon Clearance

 

 

Carbon clearance has been extensively used to monitor
the phagocytic potential of the RES (9). The procedure
was used to determine which of three types of crystalline
silica being tested depressed RES activity the greatest.
Each silica was tested at various doses and for varying
time periods. The particle size of all silicas was 5 u
or less. In normal mice the phagocytic index Was calcu-
lated to be 0.056 and the biological-half-life 5.38 min
(Table l).‘ Ten milligrams of Dorénturp silica (DQlZ)
given I.V. over 3 days significantly decreased the phago-
cytic index (K) to 0.020 and increased the T3 to 15.15
min. All other doses and types of silica had no signifi-
cant effect on carbon clearance. Dorénturp silica (10
mg over 3 days) was used in all succeeding experiments.

Scanning Electron Microscopy (SEM) of Livers
. from Normal and Silica-Treated Mice

 

 

Comparative study of liver SEM from normal and
silica-treated mice showed no indication of detrimental
effects of silica on the portal veins, hepatic veins,

Icentral veins, sinusoids and parenchymal cells of the

38

39

Table 1. Effect of different preparations of silica on
RES carbon clearancea

Phagocytic Biological Half-

 

Type of Silica Index (K) Life (T%) (min) P
Control 0.056 5.38 ---
Dorenturp 0.020 15.15, 0.001

10 mg over 3 days

Spex 0.050 6.02 N.S.
15 mg over 3 days

5 u Min-U-Sil 0.068 4.43 N.S.
9 mg over 3 days ‘

 

aAverage value from at least eight separate
experimental determinations.

liver. The only damaging effect seen was 0n liver
Kupffer cells. Kupffer cells in silica-treated livers:

1. Were not as spread out as normal cells, many
being rounded up (Figure 1B, 1D).

2. Were found engorged with silica and appeared
to have part or all of their outer plasma membrane
destroyed (Figure lC-E).

‘ 3. Had few if any membrane appendages (pseudopodia)
as compared to normal K—cells (Figure lC-E).

4. Had fewer attachment sites to walls of the
sinusoids (Figure lA-E).

The crystalline-like material found in many of the
Kupffer cells looked similar to SEM of Dorenturp silica

(Figure 1F). 'Based on these micrographic comparisons, it

 

40

Figure 1. Scanning electron micrographs of Kupffer
cells from mouse livers. (A) Kupffer cells from normal
controls (x 1,450); (B-E) Liver Kupffer cells from silica-

treated mice (x 5,200); (P) SEM of Dorenturp silica
(x 6,500).

 

 

41

 

Figure 1

42
was concluded that the material seen in Kupffer cells
was crystalline silica. The SEM work shows that RES'
cells are destroyed by DQ12 silica.
Carbon Clearance and Bacterial Trapping by

Perfused Livers of Normal and
Silica-Treated Mice

 

 

Normal and silica-treated livers were perfused with
5 mg of colloidal carbon. A statistically significant
though not dramatic decrease in carbon clearance was

observed in livers from silica-treated mice (Table 2).

Table 2. Carbon clearance in the perfused liver from
normal and silica-treated micea

 

Mg carbon in the:

 

 

Treatment Effluent Liver
Control 2.25 (45%) 2.75 (55%)
Silica treated 2.67 (53.4%)b 2.33 (46.6%)b

 

a . .
Average value from at least 51x experimental
determinations.

bP<o.001.

Thus 55% of the carbon was cleared by normal livers while
46.6% was taken up by silica-treated livers (P<0.001).

In normal livers perfused with 1—2 x 1010 S. typhimurium,
63.5% of the bacteria were trapped in the liver and 42.3%

of the organisms were recovered in the effluent (Table 3).

Silica-treated livers trapped only 31.3% (P<0.001), with

43

Table 3. Clearance of viable S. typhimurium by perfused
mouse livers from normal and silica-treated

 

 

 

animalsa
% in:
Liver % Total
Treatment Effluent HOmogenate Recovery
Control 42.3 63.5 105.8
Dorénturp 65.9 31.3 - 97.2
silica (P<0.05) (P<0.001) (N.S.)

 

3Average of eight separate experimental
determinations.

65.9% appearing in the effluent (P<0.05). Scanning
electron micrographs from livers perfused with bacteria
show that the liver was still able to trap S. typhimurium
in sinusoids, even though many Kupffer cells had been
destroyed (Figure 2). In both silica-treated and normal
livers, approximately 100% of the perfused bacteria were
routinely recovered.

Clearance of S. typhimarium'from the Blood of
Normal and Silica-Treated Mice

 

 

Figure 3 shows that silica treatment significantly
decreased the rate of whole animal bacterial clearance.
The phagocytic index for S. typhimurium clearance in
normal mice was 0.092 with a biological half-life of 3.27
min. The decrease in the clearance of bacteria in DQ12

_treated animals was statistically significant at all time

44

Figure 2. Scanning electron micrographs of silica—
treated livers perfused with 1-2 x 1010 S. typhimurium.
(A) Bacteria trapped in sinusoid, note Kupffer cell
lysed by silica (S) and rounded up Kupffer cell (K)

(x 2,600); (B) Sinusoid bacterial trapping (x 6,500);
(C) Bacteria trapped in sinusoid with rounded up
Kupffer cell at sinusoidal junction (K) (x 5,800).

45

 

Figure 2

46

Figure 3. Rate of clearance of S. typhimurium
from the blood of normal (0) and silica-treated (A)
mice. K = phagocytic index; T% = biological half-life.

-VIABLE BACTERIA/ML OF BLOOD

47

   
    
 

K-o-osa
T l/2=5°20

Kso-oez
T I/2'3'27

 

 

IO

l I V I
I0 '5

(II—1

TIME (MIN)

Figure 3

48
points (P<0.001). The silica treatment suppressed the

phagocytic index to 0.058 and increased T% to 5.20 min.

The Effect of Crystalline Silica on Bacterial
Killing and the Distribution of Viable
S. typhimurium after I.V.
Injection into Mice "

9

 

 

 

 

Fifteen minutes after I.V. injection of 1-2 x 10
S. typhimurium into normal mice, 49.5% of the bacteria
had been killed. Table 4a shows the distribution of the
remaining viable organisms in the liver, spleen and car-
cass. In the liver, 40.2% were recovered while 7.5%
were found in the carcass and 2.8% in the spleen. Silica-
treated mice actually killed a higher percentage of
bacteria (56.8% vs 49.5%) after 15 min, but the differences
were not statistically significant. It should be noted
that the distribution of remaining viable bacteria was-
very different. The majority of the viable bacteria
(23.5%) were found in the carcass. The liver had only
12.7% and the spleen 6.9%.

After 30 min, in both normal and silica-treated mice,
81.5% of the injected bacteria were killed (Table 4b).
Distribution of S. typhimurium in silica-treated animals
was similar to the 15 min data.

Susceptibility_to S. typhimurium Infection
in DQ12 Silica-Treated Mice

 

 

Groups of normal mice and mice treated with either
3 or 10 mg of Dorénturp silica were challenged intra-

venously with either 2 x 105 or 2 x 103 S. typhimurium.

49

 

 

 

Table 4. Survival of S. typhimurium 15 min (4a) and 30
min (4b) after intravenous injection into normal
and silica-treated micea

% Recovery in: Total Re- Killing
‘Treatment Liver Spleen Carcass covery (%) (%)

13

Control 40.2 2.8 7.5 50.5 49.5

Dorenturp 12.7b 6 9C 23 5b 43.2C 56.8C

silica

it).

,Control 8.2 2.8 7.5 18.5 81.5

Dorénturp 3.7b 0.5b 14.3b 18.5C 81.5C

silica

 

— aAverage of at least six separate experimental
determinations.

bp<0.001.

CN.S.

One group of silica-treated animals at each test dose

served as controls. No mice in the control groups died

during the two weeks of the experiment. The mortality

rate of S. typhimurium infected mice was followed for
two weeks. Figure 4a shows that both doses of silica

increased the-rate of mortality slightly when mice were

5

infected with 2 x'lO 'bacteria. A more dramatic dif-

ference was seen when mice were infected with 2 x 103

organisms (Figure 4b), where 83% of the normal mice

50

Figure 4. Susceptibility of normal and silica-
treated mice to infection by 2 x 105 (4a) and 2 x 103
(4b) S. typhimurium. Normal contrbl (II); 3 mg silica
(A); 10 mg silica (Q).

51

N.

m>< o

 

4

E!

I!

p.—

043 03

e ohswfim

filo

TON

1.0.?

low

1.0m

 

Foo.

 

 

 

1.0....

low

"NM/W108 “/o

rlow

 

 

52
survived compared to only 20% in the 3 mg silica group

and none survived in the 10 mg group. The LD50 for S.

typhiMurium in normal mice was calculated to be 8.0 x 10
2

4.
The LD50 for silica-treated mice was less than 1 x 10
.bacteria, indicating silica treatment enhanced suscepti;
bility to S. typhimurium well over lOO-fold.
Total White Blood Cell (WBC) and Differential
Counts in Silica-Treated Mouse Blood

White blood cell counts and differential counts were
made before and during-the course of silica treatment.
Table 5 shows that silica caused a leukocytosis and a
lymphocytosis. Total-PMN tripled from 2,000 before
treatment to 6,000 after treatment. The total number of
lymphocytes doubled from 9,487 to 20,371 after the injec-
tions of silica. Initially after the first injectioncfl?
3 mg of silica, there was a drOp in numbers of PMN and~
lymphocytes. As the treatment proceeded,both cell types
increased. While there were large increases in the
numbers of PMN and lymphocytes, the proportion of each
cell type in the blood stayed relatively constant except

during the first day of silica treatment.

53

Table 5. White blood cell (WBC) and differential counts
of silica— treated mice3

 

 

 

Mg Silica Cell Counts

Injected Lymphocytes PMN Total WBC
0 9,487‘(82.5%) 1,989 (17%) 11,476.
3 —4,094 (69%) 1,957 (31%) 6,051
6 14,292 (85%) ‘ 2,127 (15%) 16,419
10 20,371 (79%) 6,012 (21%) 26,383

 

a ’ a 0
Average value of 51x separate experimental
determinations.

DISCUSSION

In this study crystalline silica was used as an
experimental tool to study the process of bacterial
trapping in the isolated perfused liver. Moon et a1.
(57) recently demonstrated that bacterial clearance by
the perfused liver is not synonymous with phagocytosis
by Kupffer cells. Salmonella typhimurium was trapped
in the liver sinusoids, giving a "log-jam" appearance.

No bacterial killing occurred when Whole blood was omitted
from the perfusion media. In the presence of blood, 50%
of the perfused S. typhimurium were killed. No killing
occurred when bacteria were incubated with whole blood
-alone.

Bacterial trapping by the perfused liver may involve
two distinct mechanisms, namely trapping by the Kupffer
cells and by non-phagocytic endothelial cells. Scanning
electron micrographs of silica-treated livers clearly
demonstrated that DQ12 silica caused damage and destruc-
tion to Kupffer cells (cf. Figure l) but had no other
notable histotoxic effects on the liver. The destruction
of liver Kupffer cells significantly decreased trapping
of perfused S. typhimurium (cf. Table 3), but still,
31.3% of the bacteria were trapped. These data show that

54

55

for maximal bacterial trapping by the perfused liver
Kupffer cells must be viable. HoWever, livers devoid of
most functional Kupffer cells can still trap large numbers
of bacteria in the sinusoids (cf. Figure 2), reinforcing
the concept that bacterial trapping also involves non-
Kupffer cell aspects of the liver.

No bacterial killing occurred in perfused livers
from either normal or silica-treated mice. -At all times
100% of the S. typhimurium were accounted for by total
viable cells found in the effluent plus liver homogenates.
This was expected since bacterial killing occurs only
when humoral factors are present (14”36,40,41,57,75,76).

The liver is the major organ of the RES involved in. "
clearing bacteria from the blood (75,76). The state of
bacterial clearance in viva could be evaluated by use of
the perfused liver. Results of the in vivo experiments
suggest that the isolated perfused liver was a true indi-
cator of the state of bacterial clearance in the whole
animal. Silica treatment significantly decreased carbon
and bacterial clearance from the blood over a 15 min
period. The phagocytic index (K) for bacteria dropped
from 0.092 to 0.058 and the biological half-life increased
from 3.27 min to 5.20 min. The distribution of the
remaining viable bacteria varied greatly between silica-
treated and normal mice, the majority of viable S.
typhimurium recovered in silica-treated animals being in

the carcass, while most bacteria in normal mice were found

56
in the liver (cf. Table 4a,b). Decrease in bacterial
clearance in viva is consistent with results obtained in
the perfused liver since, with depression of bacterial
clearance by the liver in viva, more organisms were found
in other areas of the body. ’

_Silica treatment was found to have a profound effect
on mouse susceptibility to S. typhimurium infection (cf.
Figure 4). Normal mice had an LD50 for S. typhimurium of
8.0 x 104. The LDSO for silica-treated mice was less than
1 x 102 organisms. This differencerepresents at least
a lOO-fold increase-in susceptibility to S.Atyphimurium
infection. These experiments are the first demonstration
of enhancement of gram negative infection by silica.

These data also emphasize the importance of clearing
bacteria from the circulation by the RES, a function
mainly done by the liver. The liver clears bacteria from
the blood, localizing them to prevent spread of an infec-
tion to other areas of the body. As North has shown (66),
blood monocytes accumulate in these infective foci. With
impairment of blood clearance and possibly destruction

of blood monocytes by silica, organisms remain in the
blood longer and could form infective foci in other parts
of the body. Bacteria may then multiply without antagonism
by the RES.

~The percent S. typhimurium killed in viva after 15
and 30 min remained essentially the same in treated and

normal mice. (This similar bactericidal activity may be

57
due to silica treatment causing an increase in numbers
of blood PMN. It was found that total and differential
cell counts from peripheral blood of silica-treated mice
were elevated over the period of silica treatment, the
total PMN tripling. Presumably, this leukocytosis explains
why silica-treated animals were able to kill S. typhi-
murium as efficiently as normal nice.

While silica-treated mice still had normal initial
bactericidal ability, they also had significantly
increased susceptibility to S. typhimurium infection
.(cf. Figure 4a,b). These data strongly suggest that PMN
are not the major defense mechanism against S. typhimurium
infections, an interpretation consistent with Collins'
work (18), suggesting that liver phagocytic cells may
evolve cellular immunity which plays the ultimate role

in hoSt resistance to Salmonella infection.

LITERATURE CITED

LITERATURE CITED

Allison, A. C., J. S. Harington, and M. Birbeck.
1966. An Examination of the Cytotoxic Effects of
Silica on Macrophage. J. Exp. Med. 124: 141-153.

Allison, A. C., and P. D. Hart. 1968. Potentiation
by Silica of the Growth of Mycabaoterium tuberculosis
in Macrophage Cultures. Brit. J. Exp. Path. 49:
465-476.

Allison, A. C. 1971. Lysosomes and the Toxicity of
Particulate Pollutants. Arch. Intern. Med. 128:
131:139. -

Allison, A. C. 1970. On the Role of Macrophage in
Some Pathological Processes, p. 422-444. In: Mona—
nuclear Phagocytes. Van Furth Ed. Blackwell Sci.
Publ., Oxford.

Baine, W. B., M. G. Koenig, and J. S. Goodman. 1974.
Clearance of Candida albioans from the Bloodstream
of Rabbits. Infect. Immun. 10: 1420-1425. '

Benacerraf, B., E. Kivy-Rosenberg, M. M. Sebestyen,

and B. W. Zweifach. 1959. The Effect of High Dose

of X-Irradiation on the Phagocytic Proliferative

and Metabolic Properties of the Reticulo-Endothelial
System. J. Exp. Med. 110: 49—64.

Bennett, B., L. J. Old, and E. A. Boyse. 1964. The
Phagocytosis of Tumor Cells in viva. Transplantation
2: 183-202.

Berliner, D. L., and T. F. Dougherty. 1960. Influence
of Reticuloendothelial and Other Cells on the Meta-
bolic Fate of Steroidsw Ann. N.Y. Acad. Sci. 88:
14-27.

58

10.

11.

12.

13.

14.

15.
16.
17.

18.

19.

59

Biozzi, G., B. Benacerraf, and B. N. Halpern. 1953.
Quantitative Study of the Granulopectic Activity of
the Reticuloendothelial System. 11. A Study of
the Kinetics of the Granulopectic Activity of the
RES in Relation to the Dose of Carbon Injected:
Relationships Between the Weight of the Organs and
Their Activity. Brit. Exp. Path. 8£:-44l-457.

Biozzi, G., J. G. Howard, B. N. Halpern, C. Stiffel,
and D. Mouton. 1960. The Kinetics of Blood Clear-
ance of IsotOpically Labelled Salmonella enteritidis
by the Reticulo- Endothelial System in Mice. Immu-
nology 8: 74- 89.

Biozzi, G., C. Stiffel, B. N. Halpern, L. LeMinor,
and D. Mouton. 1961. Measurement of the Opsonic
Effect of Normal and Immune Sera on the Phagocytosis
of Salmonella typhi by the Reticuloendothelial
System. J. Immunol. 88: 296-300.

Biozzi, G., C. Stiffel, B. N. Halpern, and D. Mouton.
1963. Lack of Action of Serum Opsonins in Phago-
cytosis of Inert Particles by Cells of Reticuloendo-
thelial System. Proc. Exp. Biol. Med. 888: 1017- 1020.

Blickens, D: A., and N. R. DiLuizo. 1965. Metabolism
of Methyl Plamitate, A Phagocytic and Immunologic
Depressant and Its Infernce on Tissue Lipids. J.
Reticuloendothel. Soc. 8: 60- 74.

Bonventre, P. F., and E. OXman. 1965. Phagocytosis
and Intracellular Disposition of Viable BaCteria by
the Isolated Perfused Rat Liver. J. Reticuloendothel.
Soc. 8: 313-325.

Bryson, G., and F. Bischoff. 1967. Silicate-Induced
Neoplasms. Progr. Exp. Tumor Res. 9: 77-164.

Bull, C. G. 1915. The Fate of Typhoid Bacilli when
Injected Intravenously into Rabbits. J. Exp. Med.

3;: 475-483.

Carr, 1. 1973. The Macrophage in Iron and Lipid
Metabolism, p. 99-105. 'In: The Macrophage A Review
of Ultrastructure and Function. Academic Press,

New York.

Collins, F. M. 1974. Vaccines and Cell-Mediated
Immunity. Bact. Rev. 88: 371-402.

Comolli, R., and A. Perin. 1963. In vitro Action of
Silicogenic and Non-Silicogenic Dusts on Macrophage
Metabolism. Proc. Soc. Exp. Biol. Med. 113: 289-293.

20.

21.
22.
23.

24.
.25.
26.

27.
28.
29.

30.

31.

60

Cummings, D. E. 1929. Studies on Experimental

Pneumonokoniosis. .The Separation of Particulate
Matter Smaller than Screen Size into Graded Fractions.
J. Indust, Hyg. 88: 245-256.

Curran, R. C., and E. V. Rowsell. 1958. The Appli-
cation of the Diffusion-Chamber Technique to the
Study of Silicosis. J. Path. Bact. 88: 561-568.

DiLuzio, N. R. 1960. Reticuloendothelial Involve-
ment in Lipid Metabolism. Ann. N.Y. Acad. Sci. 88:
244-251.

DiLuzio, N. R., and W. R. Wooles. 1964. Depression
of Phagocytic Activity and Immune Response by Methyl
Palmitate. Amer. J. Physiol. 206: 939-943.

Evans, R., and P. Alexander. 1972. Role of Macro-
phages in Tumor Immunity. I. Co-Operation Between
Macrophages and Lymphoid Cells in Syngeneic Tumor
Immunity. Immunology 88: 615-626.

Filkins, J. P., R. E. Chase, and J. J. Smith. '1965.

, Characteristics of a Plasma Factor Governing Carbon

Phagocytosis in the Isolated Perfused Rat Liver.
J. Reticuloendothel. Soc. 8: 287-295.

Filkins, J. P., and J. J. Smith. 1965.. Plasma
Factor Influencing Carbon Phagocytosis in the 150-
lated Perfused Rat Liver. Proc. Soc. Exp. Biol.
Med. 888: 1181-1184. '

Filkins, J. P. 1971. Hepatic Lysosomes and the
Inactivation of Endotoxin. J. Reticuloendothel.
Soc; 8: 480-490.

Friedman, H. 1972. The Role of the Reticuloendo-
thelial System in Infection and Immunity - General
Aspects. J. Reticuloendothel. Soc. 88: 445-449.

Gardner, L. U. 1923. Studies on the Relation of

Mineral Dusts to Tuberculosis. Am. Rev. Tuberc.
8: 344-357.

Goldfischer, S., A. B. Novikoff, A. Albala, and L.
Biempica. 1970. Hemoglobin Uptake by Rat Hepato—
cytes and Its Breakdown Within Lysosomes. J. Cell
Biol. 88: 513-529.

Gordon, L. B., D. B. Cooper, and C. P. Miller. 1955.
Clearance of Bacteria from the Blood of Irradiated
Rabbits. Proc. Soc. Exp. Biol. Med. 88: 577-579.

32.

33.

34.

35.

36.
37.

38.

39.

40.

41.

42.

~61

Grogan, J. B. 1969. Alterations in Phagocytic
Function of Rats after.Treatment with Antilymphocyte
Serum. J. Reticuloendothel. Soc. 8: 411-418.

Gye, W. E., and W. J. Purdy. 1922. Poisonous Pro-
perties of Colloidal Silica. I. The Effects of

the Parenteral Administration of Large Doses. Brit.
J. Exp. Path. 8: 75-85.

Hope, J., and J. V. Friend. 1969. Preliminary

Scanning Electron Microscopic Observation on the
Effect of Silica Dusts on Cultured Macrophages.

Micron. 8: 310-324. _

Howard, J. G., and A. C- Wardlaw. 1958. The Opsonic
Effect of Normal Serum on the Uptake of Bacteria by
the Reticulo-Endothelial System. Perfusion Studies
with Isolated Rat Liver. Immunology 8: 338-352.

Howard, J. G. 1961. The Reticuloendothelial System
and Resistance to Bacterial Infection. Scot. Med.
J. 8: 60-82.

Hyman, C., and R. L. Paldino.. 1960. Possible Role
of the Reticuloendothelial System in Protein Trans-
port. Ann. N.Y. Acad. Sci. 88: 232-239.

Jenkin, C. R., and D. Rowley. 1961. The Role of
Opsonins in the Clearance of Living and Inert
Particles by Cells of the Reticuloendothelial
System. J. Exp. Med. 888: 363-374.

Jeunet, F. 8., and R. A. Good. 1967. Reticuloendo-
thelial Function in the Isolated Perqued Liver.

I. Study of Rates of Clearance, Role of a Plasma
Factor, and the Nature of RE Blockade. J. Reticulo-
endothel. Soc. 8: 351-369. ‘

Jeunet, F. S., W. Cain, and R. A. Good. 1968.
Differential Recognition of Brucella Organisms by
Kupffer Cells: Studies with Isolated Perfused
Liver. .Proc. Exp. Biol. Med. 888: 187-190.

‘Jeunet, F. 5., w. A. Cain, and R. A. Good. 1969.

Reticuloendothelial Function in the Isolated Perfused
Liver. III. Phagocytosis of Salmonella typhasa and
Brucella melitensis and the Blockade of the Reticulo-
endothelial System. J. Reticuloendothel. Soc. 8:
391-410.

Keller, R. 1973. Cytostatic Elimination of Syn-
geneic Rat Tumor Cells in vitra by Nonspecifically
Activated Macrophages. J. Exp. Med. 138: 625-644.

43.

44.

45.

46.

47.

48.

49.

~50.

51.

52.

53.

54.

62

Keller, R. 1974. Mechanisms by which Activated
Normal MacrOphages Destroy Syngeneic Rat Tumor Cells
in vitro. Immunology 88: 285-298.

Kessel, R. W. 1., L. Monaco, and M. A. Marchisio.
1963. The Specificity of the Cytotoxic Action of
Silica-A Study in vitro. Brit. J. Exp. Path. 88:

'351-364.

Kettle, E. H. 1926. Experimental Silicosis. J.
Indust. Hyg. 8: 491-495.

Kierszenbaum, F., E. Knecht, D. B. Budzko, and M. C.
Pizzimenti. 1974. Phagocytosis: A Defense Mechanism

Against Infection with Trypanosoma cruzi. J. Immunol.
112: 1839-1844.

King, E. J. 1947. Solubility Theory of Silicosis -
A Critical Study.‘ Occup. Med. 8: 26-49.

Larson, C. L., R. N. Ushijima, R. E. Barker, M. B.
Barker, and C. A. Gillespie. 1972. Effect of
Normal Serum and Anti-Lymphocyte Serum on Friend

Disease in Mice. J. Nat. Cancer Inst. 88: 1403-1407.

I Levy, M. H., and E. F. Wheelock. 1975. Effects of

Intravenous Silica on Immune and Non-Immune Functions
of the Murine Host. J. Immunol. 115: 41-48.

Lirenman, D. S., A. J. Fish, and R. A. Good. 1967.
Lack of a Serum Factor in Preventing Blockade of the
Reticuloendothelial System in the Rabbit. J.
Reticuloendothel. Soc. 8: 34-42.

Lotzova, B., and G. Cudkowicz. 1974. Abrogation of
Resistance to Bone Marrow Grafts by Silica Particles.
Prevention of the Silica Effect by the Macrophage
Stabilizer Poly- 2- Vinylpyridine N- Oxide. J. Immunol.
888: 798- 803.

Manwaring, W. H., and H. C. Coe. 1916; Endothelial
Opsonins. J. Immunol. 8: 401-408.

Manwaring, W. H. and W. Fritsthen. 1923. Study of
Microbic- Tissue Affinity by Perfusion Methods. J.
Immunol. 8: 83- 89.

Marks, J., and G. Nagelschmidt. .1959. Study of the
Toxicity of Dust with Use of the in vitro Dehydrogenase
Technique. Arch. Industr. Hyg. and Occ. Med. 88: ‘
383-389. - ,

55.

56.

57'.

58.

59.

60.

61.

61.

63.

64.

65.

66.

63

Miller, S. D., and A. Zarkower. 1974. Alterations
of Murine Immunologic Responses After Silica Dust
Inhalation. J. Immunol. 113: 1533-1543.

Miller, S. D., and A. Zarkower. 1976. Silica-
Induced Alterations of B Lymphocyte Immunocompetence:

A Possible Mechanism. J. Reticuloendothel. Soc. 19:
47-61. _—

Moon, R. J., R. A. Vrable, and J. A. Broka. 1975.
In situ Separation of Bacterial Trapping and Killing
Functions of the Perfused Liver. Infect. Immun.

88; 411-418. ‘ '

Mouton, D., G. Biozzi, Y. Bouthillier, and C. Stiffel.
1963. Phagocytosis of Salmonella by Reticuloendo-
thelial Cells of New-Born Piglets Lacking Natural
Antibody. Nature 888: 706.

Murray, I. M: .1963. The Mechanism of Blockade of
the Reticuloendothelial System. J. Exp. Med. 117:.
139-147. .

Murray, I. M.‘ 1963. Clearance Rate in Relation to
Agglutinins for Gelatin-Stabilized Colloid in the
Rat. Amer. J. Physiol. 204: 655-659.

Nabors, C. J., Jr., D. L. Berlinger, and T. F.
Dougherty. 1967. The Function of Hepatic Reticulo-
endothelial Cells in Steroid Hormone Biotransforma-
tion. J. Reticuloendothel. Soc. 8: 237-253.

Nadler, S., and S. Goldfischer. 1970. The Intra-
cellular Release of Lysosomal Content in Macrophages
that have Ingested Silica. J. Histochem. Cytochem.
88: 368-371. ‘

Nash, T., A. C. AlliSOn, and J. S. Harington. 1966.
Physicochemical Properties of Silica in Relation
to its Toxicity. Nature 210: 259-261.

Nicol, T., R. S. Snell, and D. L. J. Bilbey. 1956.
Effect of Cortisone on the Defense Mechanism of the
Body. Brit. Med. J. 8: 800-801.

Normann, S. J., and E. P. Benditt. 1965. Function

of the Reticuloendothelial System. I. Study on the
Phenomenon of Carbon Clearance Inhibition. J. Exp.

Med. 888; 693-707.

North, R. J. 1970. ‘The Relative Importance of
Blood Monocytes and Fixed Macrophage to the Expres-
sion of Cell-Mediated Immunity to Infection. J.
Exp. Med. 888: 521-534.

67.

68.

69.

70.

71.

72.

73.

74.

75.
76{

77.
78.

64

Page, R. C., P. Davies, and A. C. Allison. 1974.
Participation of Mononuclear Phagocytes in Chronic
Inflammatory Diseases. J. Reticuloendothel. Soc.
88: 413-438.

Pearsall, N. N., and R. S. Weiser. 1968. The Macro—
phage in Allograft Immunity. I. Effects of Silica

as a Specific Macrophage Toxin. J. Reticuloendothel.
Soc. 8: 107- 120.

Pernis, B., and F. Paronetto. 1962. Adjuvant Effect
of Silica (Tridymite) on Antibody Production., Proc.
Soc. Exp. Biol. Med. 110: 390-393.

Pisano, J.’ C. , and N. R,. DiLuzio. 1970. Effect
of Anti- -Kupffer Cell Serum on Phagocytosis and

Humoral Antibody Formation. Infect. Immun. 8:
448- 452. ,

Pisano, J. C., J. T. Patterson, and N. R. DiLuzio.
1969. Depression of Phagocytic Activity of the
Reticuloendothelial System by Antilymphocytic Serum
(ALS). Proc. Soc. Exp. Biol. Med. 888: 517-523.

Razumov, M. 1., B. K. Skirko, and A. YU. Grubina.
1963. Comparison of the Silicogenic Effects of
Various Quartz Preparations. Fed. Proc. Transl.
Suppl. 88: T502-T506. '

Rios, A., and R. L. Simmons. 1972. Poly-Z-Vinyl-
Pyridine N-Oxide Reverses the Immunosuppressive

Effects of Silica and Carrageenan. Transplantation
88: 343-345.

Robock, K. 1973. Standard Quartz DQ12 <5 uM for
Experimental Pneumoconiosis Research Projects in

the Federal Republic of Germany. Ann. Occ. Hyg.
16: 63- 66.

Rogers, D. E. 1960. Host Mechanisms Which Act to
Remove Bacteria from the Blood Stream. Bact. Rev.
24: 50- 66.

Saba, T. M. 1970. Physiology and Physiopathology
of the Reticuloendothelial System. Arch. Intern.
Med. 126: 1031-1052.

Sawyer, R. 1976. Personal c6mmunication.
Scheel, L. D., E. Fleisher, and F. W. Klemperer.

1953.‘ Toxicity Of-Silica. I. Silica Solutions.
Arch. Indust. Hyg. and Occup. Med. 8: 564-573.

79.

80.

81.

82.

83.

84.

85.

86.

87.

88.

89.

65

Scheel, L. D., B. Smith, J. VanRiper, and E. Fleisher.
1954. Toxicity of Silica. II. Characteristics

of Protein Films Adsorbed by Quartz. Arch. Indust.
Hyg. and Occup. Med. 8: 29-36.

Sharabaugh, R. J., and J. B. Grogan. 1969. Sup-
pression of the Reticuloendothelial Function in the
Rat with Cyclophosphamide. J. Bact. 100: 117-122.

Stiffel, C., D. Mouton, and G. Biozzi. 1970.

Kinetics of Phagocytic Function of Reticuloendothelial
Macrophage in viva, p. 335-381. In: Mononuclear
Phagocytes. Van Furth Ed. BlaCkwell Publ., Oxford.

Stuart, A. E., G. Biozzi, C. Stiffel, B. N. Halpern,
and D. Mouton.g 1960. The Stimulation and Depression
of Reticuloendothelial System Phagocytic Function

by Simple Lipids. Brit. J. Exp. Path. 88: 599—604.

Thomas, L. 1953. Cortisone and Infection. Ann.
N.Y. Acad. Sci. 88: 799-814.

'Turner, K. J., J. Papadimitriou, R. Hackshaw, and

J. D. Wetherall. 1975. Experimental Aspergillosis
in Normal Rats Infected Intravenously. J. Reticulo-
endothel. Soc. 88: 300-312.

Vigliani, E. C., and B. Pernis. 1963.. Immunological
Aspects of Silicosis. Adv. Tuberc. Res. 88: 230-279.

Vorwald, A. J., and A. B. Delahant. 1938. The
Influence of Silica on the Natural and Acquired

Resistance of the Tubercle Bacillus. Amer. Rev.
Tuberc. 88: 347-362.

Vorwald, A. J., M. Dworski, P. C. Pratt, and A. B.
Delahant. 1954. BCG VacCination in Silicosis.

An Experimental Study of the Influence of Inhaled
Quartz Dust Upon Infection by BCG (Aronson), H37Ra
and M. marinum Strains of Tubercle Bacilli. Amer.
Rev. Tuber. 88: 776-789.

Wagner, H. N., Jr., M. 110, and R. B. Hornick. 1963.
Studies on the RES. 11. Changes in the Phagocytic
Capacity of the RES in Patients with Certain Infec-
tions. J. Clin. Invest. 88: 427-434.

Wardlaw, A. C., and J. G.Howard. 1959. A Comparative
Survey of the Phagocytosis of Different Species of
Bacteria by Kupffer Cells. Brit. J. Exp. Path. 88:
113-117.

90.

91.

92.

93.

94.

95.

96.

97.

66

Wiener, J., T. S. Cotterell, W. Margaretten, and
D. Spiro. 1967. An Electron Microsc0pic Study of
Steroid Induced Reticuloendothelial Blockade.
Amer. J. Path. 88: 187-201.

Wiener, J., D. Spiro, and W. Margaretten. 1964.
An Electron Microscopic Study of Reticuloendothelial
System Blockade. Amer. J. Path. 88: 783-803.

Wiener, J. 1967. Fine Structural Aspects of
Reticuloendotehlial Blockade, p. 85-95. In:-
Reticulaendothelial System and Ahterasclerasis.
N. R. DiLuzio and R. Paoletti Ed. Plenum Press,
New York.

Wilcoxon, F., and R. A. Wilcox. 1949. Some Rapid
Approximate Statistical Procedures. American
Cyanamid Co., New York.

Zarling, J. M., and S. S. Tevethia. 1973. Transplan-

tation Immunity to SV40 Transformed Cells in Tumor
Bearing Mice. II. Evidence for Macrophage Partici-
pation at the Effector Level of Tumor Cell Rejection.
J. Nat. Cancer Inst. 88: 149-157.

Zisman, B., M. S. Hirch, and A. C. Allison. 1970.
Selective Effects of Anti-Macrophage Serum, Silica
and Anti-Lymphocyte Serum on Pathogenesis of Herpes
Virus Infection of Young Adult Mice. J. Immunol.
888: 1155-1159.

Zisman, B., E. F. Wheelock, and A. 0. Allison. 1971.
Role of Macrophages and Antibody in Resistance of

Mice Against Yellow Fever Virus. J. Immunol. 107:
236-243.‘

Rager-Zisman, B., and A. C. Allison. 1973. The
Role of Antibody and Host Cells in the Resistance
of Mice Against Infection by Coxsackie B-3 Virus.
J. Gen. Virol. 88: 329-338.

 

(11317111) I 1):)“