Arch Toxicol (1996) 70:164-173
M . R . L o v a t i 9 C. M a n z o n i 9 M . D a l d o s s i
S. Spolti 9 C.R. Sirtori
Effects of sub-chronic exposure to SO2 on lipid
and carbohydrate metabolism in rats
Received: 7 February 1995/Accepted: 13 July 1995
A b s t r a c t Sulfur dioxide (SO2) is a ubiquitous air pollu-
tant, present in low concentrations in the urban air, and
in higher concentrations in the working environment.
While toxicological reports on SO2 have extensively
dealt with the pulmonary system, essentially no data
are available on the effects of chronic exposure to this
pollutant on intermediary metabolism, although some
biochemical changes in lipid metabolism have been
detected. The present investigation was aimed at evalu-
ating the effects of sub-chronic exposure to SO2 on
concentrations of serum lipids/lipoproteins and on glu-
cose metabolism, in animal models of hypercholes-
terolemia and diabetes. A specially designed control-
inert atmosphere chamber was used, where male
Sprague-Dawley rats fed on either standard or choles-
terol enriched (HC) diets, as well as streptozotocin
diabetics, were exposed to SO2 at 5 and 10 ppm, 24 h
per day for 14 days. In rats, both on a standard diet and
on a HC regimen, SO2 exposure determined a signifi-
cant close-dependent increase in plasma triglycerides,
up to + 363% in the 10 ppm HC exposed animals. This
same gas concentration significantly reduced H D L
cholesterol levels. In contrast, exposure of diabetic ani-
mals to 10ppm SO2 resulted in a fall ( - 4 1 % ) of
plasma and liver triglycerides and in a concomitant
increase ( + 6 2 % ) of plasma H D L cholesterol. This
discrepancy could apparently be related to diverging
effects of SO2 exposure on plasma insulin levels in
the different animal groups. Kinetic analyses of trig-
lyceride synthesis carried out in rats on a standard
diet revealed, in exposed animals, a significant reduc-
tion in the secretory rate, in spite of the concomitant
hypertriglyceridemia. These findings suggest that SO2
M.R. Lovati (1~)- C. Manzoni- M. Daldossi
S. Spolti. C.R. Sirtori
Institute of Pharmacological Sciences, Via Balzaretti 9,
20133 Milano, Italy
9 Springer-Verlag 1996
exposure can markedly modify major lipid and glycemic
indices, also indicating a differential response in
normo/hyperlipidemic versus diabetic animals.
Key words Sulfur dioxide" Apolipoproteins"
Hyperlipidemia 9Atherosclerosis 9Diabetes
Introduction
A wide variety of compounds are being increasingly
spread in the environment as a consequence of techno-
logical advances. While the problem of ingestion of
toxic materials in food and water has received wide
attention, the insidious nature of man's exposure via
the respiratory tract is less clearly understood, parti-
cularly as related to the possible impact on metabolic
diseases linked to atherosclerosis (Wolinsky et al. 1985).
Sulfur dioxide (SO2) is a ubiquitous air pollutant,
present in low concentrations in the urban air, and in
higher concentrations in the working environment.
SO2 pollution is produced by combustion and process-
ing of sulfur-containing fossil fuels (Amdur 1969); expo-
sure to SO2 is prevalent in many occupational settings,
e.g. paper pulp factories, smelters, steel works, and
chemical industries. Exposure to SO2 in indoor and
outdoor ambient air and in the workplace can lead
both to direct bronchoconstriction and also to an in-
creased likelihood of bronchoconstrictor responses to
allergens (Sheppard 1988), thus acting as a major envir-
onmental toxin for asthmatics (Riedel et al. 1988). Re-
cent data suggest that in subjects with mild atopic
asthma, exposure to a combination of SO2 and NO2, in
concentrations that could be encountered in heavy
traffic, or NO2, at concentrations encountered in the
home environment, enhances the airway response to
inhaled allergen (Devalia et al. 1994; Tunnicliffe et al.
1994). In addition to gaseous SO2 pollution, other
sulphur compounds such as sulfite anions are ubiqui-
tously present in foods, beverages, and pharmaceuticals.

Even though most toxicological reports on SOz have
dealt with the pulmonary system, extra-pulmonary
effects have also been described. Increased values of
methemoglobin and sulphemoglobin have been re-
corded in people living in a polluted town compared to
those of a control group (Medeiros et al. 1983). Higher
hematocrit and sulfhemoglobin values were also detec-
ted in rats, following acute exposure to 0.87 ppm SOz,
when compared to the control groups (Baskurt 1988).
The presence of sulfhemoglobin, following SO2 expo-
sure, may be considered as an indicator of oxidative
damage in erythrocytes, a process that accompanies the
formation of hydrogen peroxide and oxygen radicals.
Because of their high chemical reactivity, these oxygen
species may induce lipid peroxidation, for instance in
erythrocyte membranes (Snyder et al. 1985). These
effects on blood cells, following SOz exposure, may
also be reflected in variations of the oxidative status of
lipids in other tissues and organs. Changes in lipid
peroxidation of various tissues have been described by
Haider et al. (1985) in guinea pigs, after chronic
SO2 exposure. Recently, an increased production of
O~ and H202 have been recorded in human neu-
trophils following in vitro SO2 exposure (Beck-Speier
et al. 1994).
Lipid components of plasma lipoproteins, especially
in the low density lipoproteins (LDL), are susceptible,
in particular conditions, to oxidation (Esterbauer
1990). Oxidatively altered LDL are considered more
atherogenic than native particles, possibly due to the
cytotoxic properties of the oxidized products (Bruck-
dorfer 1990). The effects of oxidizing air pollution,
promoted by nitrogen oxides and ozone on plasma and
tissue lipid levels and lipid peroxidation, have been
extensively reported (Donovan et al. 1977; Sagai et al.
1981). The question therefore can be posed whether
exposure of animals to SOz, which because of its chem-
ical nature have been held responsible for "reducing
type pollution" (Committee on Sulfur Oxides 1978;
Schimmel 1978), can affect lipid/lipoprotein metabol-
ism in vivo. Since lipoprotein levels and their physical
status are influenced by xenobiotic chemicals (Maliwal
and Guthrie 1982), the contribution of urban air pollu-
tion might therefore be listed among the cardiac risk
factors (Kannel et al. 1971). Epidemiological studies
point, indeed, to an increased risk of atherogenesis and
cancer linked to chronic exposure to air pollutants
(Hackney 1976; Rosenman 1979; Xu et al. 1994). Vari-
ous reports have suggested an atherogenic potential,
e.g. of CO exposure, secondary to its capacity to form
complexes with hemoproteins (Ayres et al. 1979), lead-
ing to hypoxia and enhanced cholesterol deposition on
the intimal surface of arteries (Topping 1977). These
data, however, have been recently partially rejected,
based on the re-examination of animal and epi-
demiological findings, concluding that the evidence
linking CO exposure to atherogenesis is only partially
tenable (Smith and Steichen 1993).
165
The effects of air pollutants on lipid/lipoprotein
metabolism may be more relevant and/or prominent in
subjects with preexisting alterations, for example
hyperlipidemic or diabetic patients. Therefore, also
in view of the limited information on the relationship
between SO2 pollution and plasma lipoproteins,
experiments were carried out in animal models of
hypercholesterolemia and diabetes, as well as in
normal animals. The aim of the study was to evaluate
whether sub-chronic SO2 exposure might affect
serum lipid/lipoprotein profiles and glucose metabol-
ism. All experiments were carried out in a specially
designed controlled-atmosphere chamber where
animals were allocated, 24 h a day, for all experimental
periods.
Materials and methods
Materials
Materials were purchased from the following sources: total and
HDL cholesterol, triglyceride, glucose kits from Boehringer Mann-
heim (Germany); free fatty acids (FFA) kit from Wako Chemicals,
Neuss (Germany); ampholytes from Pharmacia-LKB, Bromma
(Sweden); insulin radioimmunoassay kit from Sorin, Saluggia (Italy);
streptozotocin, Triton WR 1339 from Sigma-Aldrich (Milano, Italy);
Superspher Si 60, solid phase cartridge Supelclean L-C18 from
Supelco, Bellafonte (Pa., USA); synthetic standards of LTB4 and
LTC4 from Cascade Biochem (Reading, UK). SO 2 in air was from
SIAD, Bergamo (Italy). Organic solvents, analytical grade, were
purchased from Merck (Bracco, Italy); all other chemicals and re-
agents used were of the highest quality available.
Animals and treatments
Male, Sprague-Dawley CD rats, 250-275 g, (Charles River, Calco,
Italy) were acclimated in our animal facility and were fed, during the
first week, a pelleted chemical nonpurified diet (standard diet, Pic-
cioni, Gessate, Italy). Animals were then divided into groups on
the basis of plasma cholesterol concentrations, so that the distribu-
tion among groups was similar. One group of animals was then
transferred to a commercial hypercholesterolemic (HC) purified
diet containing 1.0% cholesterol and 0.5% cholic acid (Nath et al.
1959) for 2 weeks, whereas the other group continued the standard
diet. Animals had free access to water and to the diets (standard
or HC).
Streptozotocin diabetes was induced in 50 male, Sprague-Dawley
CD rats (250-275 g) by intravenous injection of 42 mg/kg strep-
tozotocin. This dose was selected from our previous experiments, in
which we established that this dose gave reproducible insulin defi-
ciency, without significant ketosis. Streptozotocin (STZ) was dis-
solved in citrate buffer, pH 4.5 (Reaven and Greenfield 1981) and
then immediately injected into a tail vein after overnight fast under
light ethex anesthesia. Control rats were simultaneously injected
with citrate buffer alone. The rats received sugar as a 5% solution in
drinking water for the following 24 h, in order to prevent STZ
hypoglycemia (Bar-On et al. 1976). Plasma glucose levels
were monitored in overnight fasted rats 8 days after STZ injec-
tion, and animals with plasma glucose levels <250 mg/dl were
excluded.
The Institutional Guidelines for the care and use of laboratory
animals were followed, and experiments were supervised by the
Laboratory Animal Welfare Service at our Institute.
166
Exposure system
The gas generation and controlling system was composed of the
following modules: Monitor Labs 8850 analyser for sulfur dioxide,
Rancon Instruments, Milano, Italy); MPS 2000, a portable data
system, with a central acquisition and processing module (UCS) and
a static RAM Memory Module (Micros Systems, Treviso, Italy); PC
M10 (Olivetti, Milano, Italy). Monitor Labs 8850 calibration was
obtained by a permeation tube (Wafer Device-ML Technology,
VICI Metronics, Santa Clara, Calif., USA) at rate of 840 ng/
min + 5% at 50~ The emission wavelength for SO z detection
was 254 nm.
Sub-chronic experiments were carried out in a specially designed
controlled atmosphere chamber (4m 3) with controlled lighting
(12 h/day), constant temperature (18-20~ and relative humidity
(55-60%), where animals were allocated in wire grid stainless-
steel/plastic cages suspended in racks at a density of no greater than
three rats/cage, except for the diabetic animals, which were distrib-
uted individually. The flow rate through the chamber was
3,200 l/min. Atmospheres of SO z at nominal concentrations of 5 and
10ppm were generated by metering sulfur dioxide in air (97%
purity) into a stream of filtered, conditioned air. SO 2 sensors, present
in the exposure chamber, permitted continuous monitoring of gas
and controlled the administration of UCS, when concentrations
went above or below the selected values. The central acquisition and
processing module obtained data from the exposure chamber ac-
cording to the instructions of the operator and then processed,
displayed and stored them. Chamber concentrations (ppm) of SO:
during a 15-day exposure period were 5.11 + 0.25 and 9.92 + 0.49
(mean + SD of daily measurements) for the groups at 5 ppm SO 2
and at 10 ppm SO 2, respectively.
Experimental protocol
A total of 81 rats were used in the experimental protocol and divided
into three groups. Within each group, three subgroups were formed,
of nine animals each, exposed to conditioned filtered air or condi-
tioned filtered air plus SO 2 at different concentrations, as stated
below:
1. Rats fed on a standard diet exposed to filtered air; rats on
standard diet exposed to filtered air plus 5 ppm SO2; rats on
standard diet exposed to filtered air plus 10 ppm SO2;
2. Rats fed on HC diet exposed to filtered air; HC plus 5 ppm SO2;
HC plus 10ppm SO2;
3. STZ diabetic rats (D) exposed to filtered air; D exposed to filtered
air plus 5 ppm SO2; D exposed to filtered air plus 10 ppm SO 2.
Appropriate controls for rats given standard or HC diets, or
belonging to the D group exposed to filtered air, were allocated in
a control chamber, close to the exposure chamber, with identical
flow of conditioned filtered air, temperature, humidity and day/light
cycles. All experiments, carried out in control and SO 2 exposed
animals, lasted 15 days and the animals were exposed either to
conditioned filtered air + SO z or to conditioned filtered air alone
24 h per day (including weekends). Animals both in the exposure or
control chambers were cleaned only every second day in order to
minimize changes in gas concentrations, occurring during opening
of the chamber (these usually return to normal within 30 min from
the end of the cleaning). Food and water consumptions were re-
corded daily during all experimental periods. Body weight measure-
ments were recorded weekly. Control animals were handled sim-
ilarly to rats exposed to SO 2.
At the end of the exposures (filtered air and filtered air + SO2)
animals, fasted overnight, were anesthetized with diethyl ether
and blood samples, drawn from the abdominal aorta, were col-
lected in the presence of E D T A (1 mg/ml). Bronchoalveolar
lavages were performed in three rats (for both control and exposed
groups).
Triglyceride production studies
Triglyceride secretion rate was measured in rats on a standard diet,
not belonging to the above experimental groups, after 15 days of
exposure either to filtered air or to filtered air + 10 ppm SO 2. Chow
was removed 16 h before the measurement, but free access to drink-
ing water allowed until the end of the experiment. Triglyceride (TG)
production rate was determined as described by Schurr et al. (1971).
Triton WR-1339 was injected via a foot vein into 18 rats (nine
exposed and nine controls) under light ether anesthesia at a dosage
of 300 mg/kg, blocking plasma TG removal completely and produ-
cing a linear increase of T G (Yoshino et al. 1992). Serial blood
samples from the caudal vein were obtained prior to Triton injec-
tion, at 40-, 80-, and 120-min intervals thereafter, and immediately
placed at 4~ T G concentrations were determined by an enzyme
method (Bucolo and David 1973). TG secretion rate was calculated
by the formula (Simonelli and Eaton 1978):
TGSR(mg/min) = TG slope (mg/dl/min) • PV(dl),
where TGSR = triglyceride secretion rate; TG slope = rate of
change of plasma triglyceride concentration by regression analysis
using a least squares determination of the linear resolution of the
data (Y = mx + b); PV = plasma volume I-expressed by the for.
mula: 0.03 • weight (g) + 6.21].
Plasma lipid and apolipoprotein determinations
All blood samples were centrifuged at 4~ and sera stored at
-20~ until assayed. Cholesterol, triglyceride, glucose and free
fatty acid (FFA) concentrations in plasma were determined enzym-
atically using commercially available kits.
Pools of plasmas from three rats in the same experimental group
were used for total lipoprotein (d <__1.21 g/ml) separation, per-
formed in a TL-100 Tabletop ultracentrifuge (Beckman, Pa., USA)
at 10~ Total lipoproteins were delipidated with ethanol-die-
thylether (Scanu and Edelstein 1971) and the protein content was
measured (Lowry et al. 1951). One-dimensional minislab sodium
dodecylsulfate-polyacrylamide gel electrophoresis (SDS-PAGE) was
carried out on 10% acrylamide gels (Weber and Osborn 1969). After
staining of the gels in Coomassie Brilliant Blue-methanol-acetic acid
solution, apoprotein bands were identified on the basis of their
molecular weights, by comparison with the mobility of a mixture of
standard proteins. One-dimensional minislab isoelectric focusing
(IEF), pH 4-6, was performed in homogeneous polyacrylamide gels
containing ampholines (Bar-On et al. 1976). Apoproteins were quan-
titated by densitometry of the stained gels using a computerized
Cellomatic 2A densitometer (CGA, Firenze, Italy).
Plasma immunoreactive insulin levels
Plasma insulin levels were determined in rats on standard and HC
diets, as well as in STZ-diabetic animals, exposed and non-exposed
to SO 2, by radioimmunoassay (Wilson and Miles 1977).
Liver lipid analyses
Separation of free and esterified cholesterol in livers of exposed
and control rats was carried out by thin-layer chromatography

(hexane: diethylether : acetic acid, 70: 30: 1, vot/vol/vol), following
tissue lipid extraction (Folch et al. 1957). The cholesterol content of
the spots was determined according to Zlatkis et al. (1953).
Bronehoalveolar lavage
Bronchoalveolar lavage (BAL) was performed essentially as de-
scribed by Dethloff et al. (1987). After killing, tracheas were can-
nulated with flexible plastic tubes attached to a 10-ml syringe barrel.
Ice-cold 0.9% NaC1 was added to the syringe barrel, and lungs filled
to capacity. The lavage was repeated twice with fresh saline. Lavage
effluents were centrifuged at 580 9 for 10 min at 4~ in a refrigerated
centrifuge to sediment the ceils. Supernatants were stored frozen at
-20~ until biochemical analyses were performed. Free cells ob-
tained by centrifugation of the BAL effluent were resuspended in an
appropriate volume of phosphate-buffered saline. Total cell count
was determined after dilution with a Unopette microcollection sys-
tem (Becton-Dickinson) using an "improved" Neiibauer counting
chamber and phase contrast microscopy. A total of 1 x 1 0 6 cells/ml
were challenged with 2 mM Ca z + ionophore A 23187 and processed
in order to determine, after stimulation, leukotriene production.
Frozen supernatants ( - 8 0 ~ were added with 10ng PGB 2 as
internal standard for high pressure liquid chromatography (HPLC).
Proteins were precipitated in cold ethanol followed by centrifu-
gation in a refrigerated centrifuge. Supernatants were extracted
using a solid phase cartridge. Ninety percent aqueous methanol
eluates were taken to dryness using a Speedvac evaporating cen-
trifuge (Savant Instruments, Farmingdale, N.Y.), reconstituted
in the HPLC solvent A (methanol/acetonitrile/water/acetic acid,
10: 10: 80:0.02, pH 5.5 with ammonium hydroxide) and analyzed
using a gradient pump-system (Beckman model 126) connected to
a diode array detector (Beckman model 126). UV absorbanee was
monitored at 235 nm and 280 nm, and full UV spectra (210-340 nm)
were acquired at a rate of 0.5 Hz. A multilinear gradient from solvent
A to solvent B (50% methanol, 50% acetonitrile) was used. A Super-
spher Si60 15 x 0.2 cm, 4 mm particle size column (Bracco, Milano,
Italy) was used at a flow rate of 0.25 ml/min, in order to increase
sensitivity. This method allows easy separation of LTB 4 from both
5S, 12S-dihydroxy eicosatetraenoic acid (5S, 12S-diHETE) and D6
trans LTB 4 epimers, with a lower detection limit for positive identi-
fication through UV spectral analysis of chromatographic peaks, of
3 picomol LTB 4. Synthetic standards were used for standard curves
and determining daily retention times of arachidonic acid metab-
olites.
Statistical analyses
Statistical analysis of the differences was determined with Duncan's
multiple range test (Bliss 1967). Values in the text are means + S E M .
Results
The exposure chamber (24 h/day) has provided excel-
lent reliability during all experimental periods. The
well-being of the exposed animals was directly recorded
through a port-hole placed in the door of the exposure
chamber. A pilot study was carried out in rats fed on
a standard diet exposed to 10 ppm SO2 for 48 h, in
order to assess the possibility to carry out experiments
for longer periods of time without overt signs of toxic-
ity. At the end of the exposure the animals showed
good health; no alterations in behaviour were observed.
167
In sub-chronic experiments, exposed animals did not
show signs of toxicity to both SOz concentrations.
Effect of SOz exposure in rats fed a standard diet
As shown in Table 1, the body weight gain, recorded at
the end of the experimental period, was reduced in rats
fed a standard diet and exposed to SO2, compared to
respective controls. In these animals an increasing feed
consumption, reaching statistical significance, follow-
ing 10 ppm SO2 exposure, was detected. Concentra-
tions of plasma lipids, measured following 15 days of
SO2 (5 and 10 ppm) exposure, are presented in Table 2.
In rats on a standard diet, increasing SO2 concentra-
tions produced statistically significant 46% and 73%
increases of plasma triglycerides, respectively, in 5 and
10 ppm SO2 exposed rats. Neither total nor H D L cho-
lesterol levels were influenced by SO2 exposure.
The effect of SO2 exposure on apolipoprotein com-
position (d < 1.21 g/ml) was investigated in the differ-
ent experimental groups (Table 3). Rats on a standard
diet showed a dose-dependent decrease of apo C pro-
teins, following SO2 exposure. No changes in the rela-
tive abundance of the apo C isoforms in exposed (5 and
10ppm SO2) animals were detected by one-dimen-
sional IEF gel electrophoresis, as shown by the apo
CII/CIII ratio (Table 3). Plasma immunoreactive insu-
lin levels, determined in rats on a standard diet, showed
a statistically significant dose-dependent reduction of
46 and 64%, respectively, after 5 and 10 ppm SOz
exposure (Fig. 1). No differences in liver weights were
detected in rats on standard diet after SO2 exposure
compared to values recorded in controls (Table 4).
Liver lipid analyses revealed a marked dose-related
increase of the TG content in rats on a standard diet,
reaching statistical significance after 10 ppm SO/expo-
sure (Table 4). A 2-fold increase in esterified cholesterol
content was detected in livers of 10 ppm SOz exposed
rats on a standard diet, eliciting a marked drop in the
free/esterified cholesterol ratio.
Bronchoalveolar lavage was carried out in all experi-
mental groups and the humoral and cell contents were
examined in exposed and control rats. Total cell counts
did not vary in exposed (5 and 10ppm SO2) and
control animals on a standard diet (Table 5). LTB4 and
L T C 4 quantitative analyses, performed by HPLC, were
not influenced by SO2 exposure (5 and 10 ppm) as
shown in Table 5.
Effect of SO2 exposure in rats fed HC diet
As shown in Table 1, the body weight gain, recorded at
the end of the experimental period, was reduced in HC
rats exposed to SO2, compared to respective controls.
In these animals an increasing feed consumption,
reaching statistical significance, following 10 ppm SO2
168

Table 1 Effect of S O / e x p o s u r e
on rat b o d y weight a n d feed Body weight b F o o d intake b
consumption a
Exposure Exposure
Group Pre Post A Pre Post A
(ppm S O / )

Standard diet
0 257 4- 3 366 + 9 109 4- 7 24.7 _4- 0.8 25.9 4- 0.9 1.23 4- 0.15
5 262 _+ 3 378 4- 8 116 ___9 26.3 + 0.3 28.6 4- 0.1 2.31 +__0.81
I0 287 4- 2 346 _ 4 59 ___3 c 26.1 ___0.9 28.5 4- 0.7 2.39 __+0.32 c
H C diet
0 237 4- 2 336 _ 5 89 + 6 22.6 4- 0.7 23.9 4- 0.5 1.29 + 0.75
5 243 -t- 2 323 +__7 80 4- 7 23.1 4- 0.3 24.6 4- 0.7 1.46 _+ 0.33
10 291 4- 3 353 __ 4 62 4- 6 Ca 22.9 4- 0.7 25.2 + 0.9 2.32 + 0.46 cd
Diabetes
0 274 4- 3 294 +_ 6 20 -4-_3 33.5 _+ 0.4 32.2 _+ 0.4 1.30 + 0.51
5 265 _+ 6 241 4- 5 c - 24 _+ 2 c 32.8 4- 0.7 33.5 + 0.6 0.72 + 0.37
10 270 4- 4 219 __ 9 cd - 51 __ 8 ca 34.3 ___0.5 37.5 -t- 1.I 3.20 _+ 0.68 ~

a See M e t h o d s for exposure regimen, dietary treatment and diabetes induction

b All data, expressed as g/rat for b o d y weight and as g/day per rat for feed consumption, are
,X + S E M of 9 animals per group
r Significantly different from controls (0 p p m SOz), p < 0.001
d Significantly different from 5 p p m SOz exposed rats, p <_ 0.01

exposure, was also detected. Concentrations of plasma Table 2 Effect of SO2 exposure on plasma lipids in rats fed standard
lipids, measured following 15 days of SO2 (5 and or H C diets a
10 ppm) exposure, are shown in Table 2. The choles-
Group b Chol HDL-Chol TG
terol-cholic acid diet (HC diet), as expected, while in- (ppm SO2)
ducing a marked increase of total plasma cholesterol,
determined a substantial reduction in the percent cho- Standard diet
lesterol content in the H D L fraction (Table 2). Expo- 0 66+4 41___2 58+2
5 65 4- 3 43 _ 4 67 4- I c
sure to 10 ppm SO2 produced, in HC rats, a marked 10 62 4- 5 40 + 2 100 4- 2 r
reduction, approximately 60%, of H D L cholesterol
HC diet
levels. Exposure to lower gas concentrations did not 0 247 + 41 90 + 7 65 + 8
produce any alteration in total or HDL cholesterol 5 280 + 34 80 + 9 82 4- 7 c
levels. Plasma TG in HC rats at the end of the exposure 10 272 +__24 36 _ 3 c 302 + 29 ca
period showed instead 26% and 363% rises, respective-
a See Table 1
ly, in the 5 ppm and 10 ppm SO2 exposed animals. b All data (expressed as mg/dl) are X +_ S E M of 9 rats per group
The effect of SOE exposure on apolipoprotein com- c Significantly different from controls (0 p p m SOz), p < 0.001
position (d < 1.21 g/ml) was investigated in HC rats a Significantly different from 5 p p m SO2 exposed rats, p < 0.05
following SO2 exposure (Table 3). Changes in apo
E protein, expressed as the percentage content, were
detected in HC rats (+38 and + 5 0 % , respectively,
after 5 and 10 ppm SO2 exposure). A lower content of mals, reaching statistical significance after 10 ppm SO2
C apoproteins was instead detected in 10ppm SO2 exposure (Table 4). Total cell counts and leukotriene
exposed HC animals. Analysis of the different apo analyses carried out on effluents of BAL did not vary in
C isoforms, carried out by IEF gel electrophoresis, did exposed (5 and 10 ppm SO2) versus control animals
not show changes in apo CII/CIII ratios in exposed (5 (Table 5).
and 10 ppm SO2) animals compared to their respective
controls (Table 3). Plasma immunoreactive insulin
levels, determined in rats on the HC diet, showed a Effect of the exposure to SOz of standard and HC diets
statistically significant dose-dependent reductions
( - 2 1 and - 3 3 % , respectively, after 5 and 10ppm In order to assess whether the exposure of the diets
SO2 exposure) Fig. 1. (standard and HC) to SO2 could alter them, thus being
Statistically significant increases in liver weight were responsible for the changes in the examined plasma
detected in 10 ppm SO2 exposed HC rats, as shown parameters, an experiment was performed where con-
in Table 4. Liver lipid analyses revealed a marked trol rats, not belonging to the main experiment, were
dose-related increase in the TG content in these ani- given standard and HC diets pre-exposed for 3 days to
 169

Table 3 Effect of 502 exposure

on total plasma apolipoprotein Group b AIV E AI AII Cs CII/CIIV
composition a (ppm SO2) % composi-
tion

Standarddiet
0 10.2 4- 1.4 24.5 _ 2.1 45.2 4- 2.3 6.2 4- 0.8 14.8 + 0.6 0.32 +__0.07
5 11.7 __+0.7 24.1 _ 2.3 48.1 + 1.7 5.8 4- 0.7 9.3 4- 1.5 a 0.30 4- 0.08
10 12.2 + 1.3 25.6 _ 1.8 46.7 __+ 1.9 6.4 + 1.6 8.4 4- 1.2 a 0.34 + 0.09
HCdiet
0 10.7 4- 0.6 20.7 4- 2.3 51.1 + 2.6 1.94 __+0.1 15.0 4- 0.8 0.29 ___0.06
5 7.2 ___0.8 28.6 +__0.7 a 50.6 + 2.6 1.8 4- 0.9 12.6 4- 1.1 0.31 4- 0.10
10 7.8 4- 0.9 31.1 _ 0.8 d= 50.4 + 3.1 1.7 4- 0.7 9.5 _ 0.7 a 0.30 4- 0.09
Diabetes
0 14.3 4- 0.6 7.3 _ 0.8 59.2 ___ 1.1 1.1 _ 0.7 18.8 _ 1.3 0.25 4- 0.03
5 12.8 -4- 1.2 7.3 + 0.9 60.3 4- 2.8 0.9 -4- 0.3 19,3 __+2.1 0.24 4- 0.07
10 14.1 _ 1.9 6.9 + 1.2 60.8 ___2.5 0.8 + 0.2 18.1 ___ 1.7 0.26 + 0.09

a Percentage distribution of total (d < 1.21 g/ml) a p o l i p o p r o t e i n p a t t e r n following S D S - P A G E

b All data are X + S E M of 3 pools of 3 animals each
CII/CIII isoform ratio determined following one-dimensional I E F
d Significantly different from controls (0 p p m SO2), p < 0.001
Significantly different from rats exposed to 5 p p m SO2, p _< 0.01

40 Although body weight losses are a typical feature

HC diet of animals with STZ-induced diabetes (Yoshino et al.
1989), SOz exposed diabetic rats are characterized
_A 30 by a dose-dependent body weight reduction. In these
E
animals a positive trend in feed consumption versus
diabetic controls was observed, which reached statist-
N 2o Std. diet ical significance in 10 ppm SO: exposed rats (Table 1).

E
n
~ Diabete4
5 10
ppm SO 2
Fig. 1 Plasma insulin levels o f rats fed s t a n d a r d and high choles-
terol(HC) diets and of streptozotocin (STZ)-diabetic animals, non-
exposed and exposed to 5 and 1 0 p p m SO~. Asterisks indicate
significant differences from the non-exposed control values
(p < 0.001). Verticallines represent S E M of the m e a n s
10 ppm SOz. The test diets (standard and HC), given to
control animals, were replaced twice during each week,
in order to ensure a continued supply of SOa exposed
feed. Total and HDL cholesterol and triglyceride levels
were evaluated at the end of this pilot experiment,
which lasted 2 weeks. No variations in the evaluated
parameters (Table 6) of non exposed diets (Table 2)
were detected, thus suggesting that the exposure of the
diets to SO2 was not responsible for the described
plasma and tissue effects of exposed rats.
Effect of SO2 exposure in STZ-diabetic rats
As shown in Table i, dose-dependent decreases in body
weights were observed in diabetic rats exposed to SO2.
Concentrations of plasma lipids, measured following
15 days of 5 and 10 ppm SOa exposure, are reported
in Table 7. STZ-diabetic rats exposed to 10 ppm
SOz, were characterized by reduced plasma TG
concentrations compared to non-exposed diabetics and
to diabetics exposed to 5 p p m SOz (Table 7). A
30% reduction of plasma FFA versus appropriate con-
trois was also detected in 10 ppm SOa exposed diabetic
rats. No difference in the reported plasma parameters,
was observed in 5 ppm SOa exposed diabetics versus
diabetic controls. The effect of SO2 exposure on apol-
ipoprotein composition (d < 1.21 g/ml) was investi-
gated in the different experimental groups. Increased
amounts of total apo AIV, AI and C proteins and
a decreased content of apo E, typical of altered carbo-
hydrate metabolism in rats (Bar-On et al. 1976), were
detected in STZ-diabetics (Table 3). This abnormal
apolipoprotein pattern in diabetic rats was not modi-
fied by SO2 exposure. No changes in the relative
abundance of the apo C isoforms, as shown by the
apoCII/CIII ratio, in exposed (5 and 10 ppm SO2)
animals were detected by one-dimension IEF gel
electrophoresis (Table 3). Plasma immunoreactive
insulin (IRI) levels were finally determined in STZ-
diabetics after SO2 exposure. In these animals, SO2
exposure, while markedly decreasing plasma TG con-
centrations, did not significantly affect plasma IRI
levels Fig. 1.
Diabetic rats showed a marked liver weight reduc-
tion, respectively - 2 1 and 27%, following 5 and
170

Table 4 Effect of SO: exposure on rat liver weights and liver lipids a

Group u Liver weight TG FC EC FC/EC

(ppm SOz) g mg/g mg/g mg/g

Standard diet
0 12.22 • 0.86 4.58 4- 0.78 2.06 + 0.16 0.42 • 0.06 5.56 _ 0.90
5 13.52 -I- 1.32 6.87 ___ 1.32 1.76 4- 0.19 0.45 _ 0.08 4.95 4- 0.86
10 11.28 • 0.82 12.44 4- 0.55 r 2.41 • 0.37 0.83 _ 0.03 ~ 2.94 _ 0.24~
HC diet
0 15.43 + 0.86 33.83 • 3.46 2.66 _ 0.12 22.94 _ 1.5 0.12 4- 0.01
5 16.01 • 0.85 37.08 4- 3.75 3.81 4- 0.23 29.75 -t- 1.9 0.11 4- 0.01
10 18.47 • 0.55 ca 65.12 • 3.66 ~ 2.33 4- 0.i6 23.71 • 1.3 0.10 4- 0.01
Diabetes
0 12.14 • 0.50 30.72 _ 0.50 1.76 • 0,ll 0.31 _ 0.09 5.67 __+0.12
5 9.53 4- 0.39 ~ 8.83 • 0.39 ~ 1.78 4- 0.05 0.15 +_ 0.07 ~ 11.86 + 1.t4 ~
10 8.83 4- 0.50 ~ 4.21 +_ 0.50 *a 1.84 4- 0.19 0.20 _+ 0.0Y 9.20 4- 1.17~

"See Tablel
b All data are X + SEM of 9 animals per group. Free cholesterol (FC), ester• cholesterol (EC)
Significantly different from controls (0 ppm SOz), p < 0.001
d Significantly different from rats exposed to 5 ppm SO2, p _< 0.01

Table 5 Effect of SO2 exposure on leukotriene LTB4 and LTC4 Table 7 Effect of SO2 exposure on plasma lipid and glucose meta-
production in cells recovered by bronchoalveolar lavage" bolism markers in streptozotocin-diabetic rats a

Group Cell number b LTB4 LTC4 ppm SO2

(ppm SO2) • l06 rig/106 cells 0 5 10

Standard diet Chol b 86 _ 10 78 _ 8 75 + 6

0 1.1 + 0.5 78.2 • 6.8 26.3 • 5.1 HDL-Chol 47 _+ 2 51 _+ 5 53 + 1
5 1.6 _ 0.2 73.1 • 2.4 25.9 4- 7.3 TG 187 _ 6 195 4- 11 86 -I- 7 ~
10 1.4 _ 0.6 77.8 • 5.6 28,8 _ 9.2 Glucose 345 • 14 398 _+ 19 325 _+ 10
HC diet FFA 1.774 _+ 0.43 1.857 4- 0.7 1.268 _ 0.3~
0 1.2 • 0.4 70.2 • 3.2 25.6 ___4.1
5 1.4 _ 0.6 75.2 + 6.4 27.2 _ 6.3 "See Table 1
10 1.5 + 0.3 70.8 • 4.2 26.4 4- 4.1 bAll data are i~ _ SEM of 9 rats per group and are expressed as
mg/dl for cholesterol (Chol), HDL-Chol, triglycerides (TG) and
Diabetes glucose, and as gEq/ml for free fatty acids (FFA)
0 1.5 • 0.3 72.6 • 3.4 27.8 ___6.3 r Significantly different from controls and 5 ppm SOz exposed rats,
5 1.4 + 0.4 68.9 ___7.l 24.9 ___8.1 p _< 0.005
10 1.6 • 0.8 77.9 • 9.2 26.1 + 4.6

See Methods for bronchoalveolar lavage and leukotriene deter-

mination 10 ppm SO2 (Table 4); a marked dose-dependent de-
ball data, expressed as x 106 for cell number and as ng/106 cells crease in liver triglyceride content was detected in ex-
for leukotriene determination, are ,X • SEM of 3 animals per
posed STZ-diabetic animals. Statistically significant,
group
but not dose-dependent, reductions of the liver ester•
fled cholesterol content were finally detected in 5
and 10 ppm SO2 exposed diabetics. Bronchoalveolar
lavage was examined in STZ-diabetic exposed and con-
Table 6 Effect of the exposure to SO2 of standard and HC diets on trol rats. Total cell counts did not vary in exposed (5
plasma lipids in control rats a
and 10 ppm SO2) and control animals (Table 5). LTB4
Group b Chol HDL-Chol TG and LTC4 quantitative analyses, performed by HPLC,
(ppm SO:) were not influenced by SO2 exposure (5 and 10 ppm).
LBT4 and LTC4 concentrations (rig/10 6 cells) did not,
SOz exposed
in fact, differ from those recorded in control animals
standard diet
59+6 38+4 57__+4 (Table 5).
SO: exposed
HC diet
236 • 27 87 __ 9 63 + 7 Triglyceride production studies
a Control rats fed standard or HC diets pre-exposed for 3 days to
10 ppm SO2. The question was next addressed, as to whether the
bAll data (expressed as mg/dl) are X _+ SEM of 9 rats per group increased plasma and liver TG content were related to

Controls~,m
6~176
3~176
1 TG levels.
~k J . ~ ' * -- TGSR (met~rain)
. . . . . . .
150-/, , , . j ~ Cont.." 0.81 + 0.021
g ,
0 40 80
Fig. 2 Plasma trigIyceride time curves after injection of Triton WR
1339 in rats fed a standard diet, non-exposed and exposed to 10 ppm
SO2. Values in the bottom of the figure represent the triglyceride
secretion rates expressed as mg/min. Asterisks indicate significant
differences from the non-exposed control values (p < 0.00l)
a specific alteration of lipoprotein metabolism. The in
vivo synthesis of TG-rich lipoproteins was evaluated by
the Triton WR-1339 test, in rats on a standard diet
after 2 weeks on 10 ppm SO2 exposure. Triton adminis-
tration to rats produced a linear rise of plasma TG
concentrations, up to a maximum of 720 _+ 15 mg/dl in
controls versus 560_ 12 mg/dl in SO2 exposed ani-
mals, 120 rain post-injection. In control rats the TG
production averaged 0.814mg/min (slope r = 0.95,
p < 0.001) (Fig. 2). In contrast, 10 ppm SO2 exposed
rats, while showing higher basal plasma TG concentra-
tions compared to controls, were characterized by
a 24% reduction in the TG secretory rate (Fig. 2).
Discussion
The results of the present study point out to a systemic
effect, of SO2 exposure, on lipid metabolism in rats fed
diets with different lipid content as well as in STZ-
diabetics. These effects are observed at SO2 concentra-
tions not inducing perceptible alterations of lung func-
tion, as assessed by analysis of cellular and humoral
content in the effluent of BAL. SO2 exposure appears
to affect mainly TG metabolism, with a statistically
significant, dose-dependent rise of plasma and liver TG
contents both in rats on standard and HC diets.
Since increased TG levels may reflect alterations in
the synthesis and/or catabolism of TG-rich lipoprotein
particles, TG production rate was studied in rats fed
a standard diet at the end of 15 days of 10 ppm SO2
exposure. Results of the Triton WR-1339 test are con-
sistent with a reduced liver TG secretion in exposed
rats versus standard diet controls, suggesting that SO2
exposure might act by interfering with peripheral TG
catabolism (Nilsson-Ehle et al. 1980). Plasma apolipo-
protein analyses revealed in exposed rats, a consider-
able fall in the apo C content, but quantitative analysis
171
of apoC isoforms, i.e. comlSrehensive of the lipase
activator apo CII (Kinnunen et al. 1977) and of
the antagonist apo CIII (Wang et al. 1985), after
IEF gel electrophoresis, did not differ in exposed and
control animals. This rules out the hypothesis that
variations in the apo CII content (Carlson and Ballan-
tyne 1976), may be involved in the rise of plasma
Since lipoprotein lipase is actively regulated by hor-
monal factors (Garfinkel et al. 1976), among them insu-
lin, our attention was directed to the effects of SO2
exposure on plasma insulin levels. These were mark-
120
edly decreased in exposed rats, independent of dietary
treatment, in a more pronounced way following
10 ppm SO2. The reduced plasma insulin levels could
determine a decrease in lipoprotein lipase activity re-
sulting, in exposed animals, in a progressive rise of
plasma TG concentrations. Moreover, since insulin
putatively raises the uptake of apo B/E containing
lipoproteins by the LDL receptor pathway (Chute et al.
1985), decreased insulinemia in SO2 exposed animals
on both dietary regimens might be responsible for the
increased levels of TG-rieh lipoproteins. The increase of
liver TG content in exposed rats could result from an
impaired secretion (demonstrated by the Triton test)
possibly consequent to a decreased total plasma apo-
protein production in exposed animals, fed either stan-
dard or HC diets.
The data obtained in the present experiment are
difficult to explain because no obvious correlation
seems to be present between SO2 inhalation and cha-
nges in lipid metabolism. Moreover, since in these
experiments both the diets and the animals are exposed
to SO2, the question of whether gas exposure itself or
alterations in lipid/protein components of the diets,
could be responsible of the observed changes was next
addressed. A pilot experiment was therefore performed
in control animals, which were not a part of the main
study, with diets (standard or HC) for three days ex-
posed to 5 and 10 ppm SO2. No variations in the
biochemical parameters were detected in rats fed these
diets for 2 weeks, thus suggesting that the described
results are fully ascribable to gas exposure of the ani-
mals.
The following hypothesis might be proposed for in-
terpretation of these data. Inhaled sulfur dioxide can be
easily hydrated to yield sulfurous acid on the surface of
the respiratory tract, which subsequent dissociates to
form sulfite/bisulfite (SO~-/HSO3) and hydrogen
ions (H +) (Gunnison 1981). These other species of
SOz metabolism could reach the circulation during
the breathing, thus influencing the metabolic state
of perfused tissues. Even though numerous feeding
studies with sulfite in different mammalian species
(Gunnison 1981) have failed to demonstrate clear-
cut sulfite toxicity, due to the efficiencyof sulfite detoxi-
fication (oxidation to sulfate, catalyzed by sulfite
oxidase) (Gunnison et al. 1987), available data do not
172
permit to exclude in our experimental model a possible
interference of sulfite concentrations with protein and
non-protein cysteine-rich compounds, such as insulin
and glutathione.
Diabetic rats showed, at the end of the exposure
period, marked body and liver weight losses, despite
increased food intake. Decreasing growth rate was also
observed in pigs fed sulfite, although this effect was
ascribable to reduced feed consumption (Til et al. 1972).
Plasma and liver triglyceride content was also lower in
exposed animals versus non-exposed controls. Since
low plasma insulin levels impair glucose transport into
cells, a compensatory energy supply could rise from
triglyceride catabolism, apparently enhanced in the
present study, in SO2 exposed diabetic rats. Interpreta-
tion of these findings, diverging from those found in
non diabetic animals is, of course, only tentative. It is
known that the mechanism of STZ-diabetes is of the
oxidative type, leading to cell death in pancreatic islet
fl-cells (Like and Rossini 1976), accompanied by re-
duced tissue glutathione (GSH) content (Loven et al.
1986). STZ-diabetic rats, as shown by Magni et al.
(1992), are also characterized by an increased percent-
age of the thiolic dephosphorylated form of 3-hydroxy-
3-methyl glutaryl coenzyme A reductase due to an
increase of glutathione disulfide (GSSG), the oxidized
form of glutathione. SO2 exposure in diabetic rats
might thus possibly interfere with the redox state of
glutathione, thus influencing the cellular oxidative
stress induced by STZ injection.
In conclusion, the reported data point out to a sys-
temic effect of SO2 in animal models of frequent human
pathologies, such as hypercholesterolemia and dia-
betes. In standard and HC rats, SO2 exposure deter-
mined increased feed consumption, high plasma and
tissue TG content and decreased insulin levels, not
associated to changes in behaviour or overt disease.
Diabetic rats, on the contrary, showed, at the highest
SO2 dose, decreased plasma and tissue lipid contents.
Available data do not permit, as of now, transposition
of data from experiments carried out in animals to
man, since people living in urban areas are exposed to
different air pollutants, the effect of which on health
may be combined or superimposed. On the other hand,
the results obtained in the present study raise the
necessity to evaluate, also in the case of gaseous com-
pounds, the involvement of different body compart-
ments, which normally are not believed to be the eli-
gible target site of these air pollutants.
Acknowledgements These experiments were supported by a grant
from Ente Nazionale Energia Elettrica, Progetto Interazione Ambi-
ente e Salute, awarded from the CNR (Consiglio Nazionale delle
Ricerche) of Italy. The authors would like to thank Professor
G. Foleo's staff for technical assistance in eicosanoid metabolism
studies and Professor Claudio Galli, Chair of Experimental Phar-
macology, Institute of Pharmacological Sciences, University of
Milano, for reading the manuscript and providing helpful comments
and suggestions.
References
Amdur MO (1969) Toxicological appraisal of particulate matter,
oxides of sulfur, and sulfuric acid J Air Pollut Control Assoc 19:
638-644
Ayres SH, Evans RG, Buehler ME (1979) The prevalence of car-
boxyhemoglobinemia in New Yorkers and its effects on the
coronary and systemic circulation. Prey Med 8:323
Bar-On H, Roheim, PS, Eder HA (1976) Serum lipoproteins and
apolipoproteins in rats with streptozotocin-induced diabetes.
J Clin Invest 57:714-721
Baskurt OK (1988) Acute hematologic and hemorheologic effects of
sulfur dioxide inhalation. Arch Environ Health 43:344-348
Beck-Speier I, Lenz AG, Godleski JJ (1994) Responses of human
neutrophils to sulfite. J Toxicol Environ Health 41:285-297
Bliss CI (1967) Comparison of several groups. In: Bliss CI (ed)
Statistics in Biology. Me Graw-Hill, New York, N.Y.,
pp 232-237
Bruckdorfer KR (1990) Free radicals, lipid peroxidation and
atherosclerosis. Curr Opin Lipid 1:529-535
Bucolo G, David H (1973) Quantitative determination of serum
triglyceride. Clin Chem 19:476-478
Carlson LA, Ballantyne D (1976) Changing relative proportion of
apolipoprotein C-II and C-III of very low density lipoproteins in
hypertriglyceridemia. Atherosclerosis 23:563-568
Chute EP, Barbosa JJ, Rupp WM, Rohde RD, Goldenberg FJ,
Hitchcock CR, Buchwald H (1985) Reduction of plasma choles-
terol and LDL-cholesterol by continuous intravenous insulin
infusion. Surgery 98:656-661
Committee on Sulfur Oxides (1978) Epidemiological studies of
health effects of sulfur oxides. In: Sulfur oxides. National Acad.
emy of Science, Washington DC, pp 180-209
Dethloff LA, Gladen BC, Gilmore LB, Hook GER (1987) Quantitat-
ion of cellular and extracellular constituents of the pulmonary
lining by using bronchoalveolar lavage. Am Rev Respir Dis 136:
899-907
Devalia JL, Rusznak C, Herdman M J, Trigg CJ, Tarraf H, Davies
RJ (1994) Effect of nitrogen dioxide and sulfur dioxide on airway
response of mild asthmatic patients to allergen inhalation. Lan-
cet 344:1668-1671
Donovan DH, Williams SJ, Charles JM, Menzel DB (1977) Ozone
toxicity: effect of dietary vitamin E and polyunsaturated fatty
acids. Toxicol Lett 1:135 139
Esterbauer H, Dieber-Rotheneder M, Waeg G, Striegl G, Ashy A,
Sattler W, Jurgens G (1990) Biochemical, structural and func-
tional properties of oxidized low-density lipoproteins. Chem Res
Toxieol 3:77-92
Folch J, Less M, Sloan-Stanley GA (1957) A simple method for
isolation and purification of total lipids from animal tissues.
J Biol Chem 226:497-502
Garfinkel AS, Nilsson-Ehle P, Schotz MC (1976) Regulation of
lipoprotein lipase induction by insulin. Biochem Biophys Acta
124:264-273
Gunnison AF (1981) Sulphite toxicity: a critical review of in vitro
and in vivo data. Food Cosmet Toxicol 19:667-682
Gunnison AF, Sellakumar A, Currie D, Snyder EA (1987) Distribu-
tion, metabolism and toxicity of inhaled sulfur dioxide and endo-
genously generated sulfite in the respiratory tract of normal and
sulfite oxidase-deficient rats. J Toxicol Environ Health 21:
141-162
Hackney JD (1976) Relationship between air pollution and cardio-
vascular disease; a review. In: Finkel AJ, Duel WC (eds) Clinical
implications of air pollution research. Publishing Sciences
Group, Acton, pp 89-106
Haider SS (1985) Effects of exhaust pollutant sulfur dioxide on lipid
metabolism of guinea pig organs. Ind Health 23:81-87
Kannel WB, Castelli VP, Gordon T, McNamara PM (1971) Serum
cholesterol, lipoproteins and the risk of coronary heart disease:
the Framingham Study. Ann Int Med 74:1-12

Kinnunen PKJ, Jackson RL, Smith LC, Gotto AM Jr, Sparrow JT
(1977) Activation of lipoprotein lipase by native and synthetic
fragments of human plasma apolipoprotein C-II. Proc Natl Acad
Sci USA 74:4848-4853
Like AA, Rossini AA (1976) Streptozotoein-induced pancreatic insu-
litis: new model of diabetes mellitus. Science 193:415-417
Loven D, Schedl H, Wilson H, Daabees TT, Steigink LD, Diekus M,
Oberley L (1986) Effect of insulin and oral glutathione
on glutathione levels and superoxide dismutase activity in organs
of rats with streptozotocin-induced diabetes. Diabetes 35:
503-507
Lowry OH, Rosebrough NJ, Farr AL, Randall RJ (1951) Protein
measurement with the Folin phenol reagent. J Biol Chem 193:
265-275
Magni F, Cancellieri M, Del Puppo M, Galli-Kienle M (1992)
Diabetes-induced alteration of HMGCoA reductase forms in rat
livers. Acta Diabetol 28:211-214
Maliwal BI, Guthrie FE (1982) In vitro uptake and transfer of
chlorinated hydrocarbons among human lipoproteins. J Lipid
Res 23:474-479
Medeiros MHG, Bechara EJH, Naoum PC (1983) Oxygen toxicity
and hemoglobinemia in subjects from a highly polluted town.
Arch Environ Health 38:11-16
Nath N, Wiener R, Harper AE, Elvehjiem CA (1959) Diet and
cholesterolemia, part 1 (development of a diet for the study of
nutritional factors affecting cholesterolemia in rats). J Nutr 67:
289-296
Nilsson-Ehlep, Garfinkel AS, Schotz MC (1980) Lipolytic enzymes
and plasma lipoprotein metabolism. Annu Rev Biochem 49:
667-693
Reaven GM, Greenfield MS (1981) Diabetic hypertriglyceridemia.
Evidence for three clinical syndromes. Diabetes 30:66-75
Riedel F, Kr~imer M, Scheibenbogen C, Rieger CHL (1988) Effects of
SO2 exposure on allergic sensitization in the guinea pig. JAller
Clin Immunol 82:527-534
Rosenman KD (1979) Cardiovascular disease and environmental
exposure. Br J Ind Med 36:85-97
Sagai M, Ichinose T, Oda H, Kubota K (1981) Studies on the
biochemical effects of nitrogen dioxide. Lipids 16:64-67
Scann AM, Edelstein C (1971) Solubility in acqueous solutions of
ethanol of the small molecular weight peptides of the serum very
low density and high density lipoproteins: relevance to the recov-
ery problem during delipidation of serum lipoproteins. Anal
Biochem 44:576-588
Schimmel H (1978) Evidence for possible acute health effects of
ambient air pollution from time series analysis. Methodological
questions and some new results based on New York city daily
mortality: 1963-1976. Bull NY Acad Med [Second Series] 54:
1052-1108
173
Schurr PE, Schultz JR, Parkinson TM (1971) Triton induced hyper-
lipidemia in rats as an animal model for screening hypolipidemic
drugs. Lipids 7:68-74
Sheppard D (1988) Sulfur dioxide and asthma - a double-edged
sword? JAller Clin Immunol 82:961-964
Simonelli C, Eaton P (1978) Effect of elofibrate on in vivo triglycer-
ide production and clearance in genetically hyperlipidemic rats.
Atherosclerosis 29:269-275
Smith CJ, Steichen TJ (1993) The atherogenic potential of carbon
monoxide. Atherosclerosis 99:137-147
Snyder LM, Fortier NL, Trainor J, Jacobs J, Leb L, Lubin S, Chiu
D, Shohet S, Mohandas N (1985) Effect of hydrogen peroxide
exposure on normal human erythrocyte deformability, morpho-
logy, surface characteristics and spectrin-hemoglobin cross-link-
ing. J Clin Invest 76" 1971-1977
Til HP, Feron VJ, de Groot A, Van der Wal P (1972) The toxicity of
sulphite. II. Short- and long-term feeding studies in pigs. Food
Cosmet Toxicol 10:463-467
Topping DL (1977) Metabolic effects of carbon monoxide in relation
to atherogenesis. Atherosclerosis 26:129-137
Tunnicliffe WS, Burge PS, Ayres JG (1994) Effect of domestic con-
centrations of nitrogen dioxide on airway responses to inhaled
allergen in asthmatic patients. Lancet 344:1733-1736
Wang CS, McConathy WJ, Kloer HU, Alaupovic P (1985) Modula-
tion of lipoprotein lipase activity by apolipoproteins. Effect of
apolipoprotein C-III. J Clin Invest 75:384-390
Weber K, Osborn M (1969) The reliability of molecular weight
determinations by sodium dodecyl sulphate-poly acrylamide gel
electrophoresis. J Biol Chem 244:4406-4411
Wilson MA, Miles LEM (1977) Radioimmunoassay of insulin. In:
Abraham GE (ed) Handbood of radioimmunoassay. M Dekker,
New York, pp 275-293
Wolinsky H (1985) Atherosclerosis. In: Wyngaarden JB, Smith LH
(eds) Textbook of medicine. Sandeers Company, Philadelphia, pp
281-284
Xu X, Gao J, Dockery DW, Chert Y (1994) Air pollution and daily
mortality in residential areas of Beijing, China. Arch Environ
Health 49:216-222
Yoshino J, Iwai M, Kazumi T, Matsushita M, Morita M, Matsuba
K, Iwatani t, Baba S (1989) Effect of dietary fructose on triglycer-
ide turnover in streptozotocin-diabetic rats. Atherosclerosis 79:
41-46
Yoshino G, Matsushita M, Maeda E, Naka Y, Nagata K, Morita M,
Matsuba K, Kazumi T, Kasuga M (1992) Effect of long-term
insulin deficiency and insulin treatment on the composition of
triglyceride-rieh lipoproteins and triglyceride turnover in rats.
Atherosclerosis 92:243-250
Zlatkis A, Zak B, Boyle AJ (1953) A new method for direct
determination of serum cholesterol. J Lab Clin Med 41:486-496