Carcinogenesis vol.21 no.4 pp.783–792, 2000
Calcium phosphate-containing precipitate and the carcinogenicity
of sodium salts in rats
Samuel M.Cohen1,2,5, Lora L.Arnold1, Martin Cano1,
Masahiro Ito1, Emily M.Garland1,4 and R.Anthony Shaw3
1Department
of Pathology and Microbiology, 2The Eppley Institute for
Research on Cancer, University of Nebraska Medical Center, Omaha,
Nebraska 68198–3135 and 3National Research Council of Canada, Institute
for Biodiagnostics, Winnipeg, Manitoba R3B 1Y6, Canada
4Present
address: 8204 Londonberry Road, Nashville, TN 37221, USA
5To
whom correspondence should be addressed
Email: [email protected]
Sodium saccharin, ascorbate and other sodium salts fed at
high doses to rats produce urinary bladder urothelial
cytotoxicity with consequent regenerative hyperplasia. For
sodium salts that have been tested, tumor activity is
enhanced when administered either alone or after a brief
exposure to a known genotoxic bladder carcinogen. These
sodium salts alter urinary composition of rats resulting in
formation of an amorphous precipitate. We examined
the precipitate to ascertain its composition and further
delineate the basis for its formation in rat urine. Using
scanning electron microscopy with attached X-ray energy
dispersive spectroscopy, the principal elements present
were calcium, phosphorus, minor amounts of silicon and
sulfur. Smaller elements are not detectable by this method.
Infrared analyses demonstrated that calcium phosphate
was in the tribasic form and silicon was most likely in the
form of silica. Small amounts of saccharin were present in
the precipitate from rats fed sodium saccharin (<5%), but
ascorbate was not detectable in the precipitate from rats
fed similar doses of sodium ascorbate. Large amounts
of urea and mucopolysaccharide, apparently chondroitin
sulfate, were detected in the precipitate by infrared analysis.
Chemical analyses confirmed the presence of large amounts
of calcium phosphate with variably small amounts of
magnesium, possibly present as magnesium ammonium
phosphate crystals, present in urine even in controls. Small
amounts of protein, including albumin and α2u-globulin,
were also detected (<5% of the precipitate). Calcium
phosphate is an essential ingredient of the medium for
tissue culture of epithelial cells, but when present at
high concentrations (>5 mM) it precipitates and becomes
cytotoxic. The nature of the precipitate reflects the unique
composition of rat urine and helps to explain the basis for
the species specificity of the cytotoxic and proliferative
effects of high doses of these sodium salts.
Introduction
In 1970, sodium saccharin, incorporated into cholesterol pellets
and surgically implanted into mouse bladder, was shown
Abbreviations: BBN, N-butyl-N-(4-hydroxybutyl)nitrosamine; BrdU, bromodeoxyuridine; FANFT, N-[4-(5-nitro-2-furyl)-2-thiazolyl]formamide; HPLC,
high performance liquid chromatography; MNU, N-methyl-N-nitrosourea;
OTS, O-toluene sulfonamide.
© Oxford University Press
to increase the incidence of bladder tumors (1). Although
subsequent studies demonstrated that this tumorigenic effect
was due to the pellet itself rather than the sodium saccharin
(2–4), this study raised concern about the safety of saccharin
as an artificial sweetener and food additive (5) and led to
additional testing of its carcinogenicity. This resulted ultimately
in the findings presented as an abstract in 1974 (6) indicating
that sodium saccharin fed at high doses was a bladder carcinogen in rats when administered over two generations. This
study involved administration of the compound to the male
and female F0 parental generation prior to mating, then to the
dams during gestation and lactation, and then to the weaned
pups for the remainder of their lives for up to 2–2.5 years.
This initial report was confirmed by three subsequent studies
(7–9) all indicating that sodium saccharin fed at high doses in
the two-generation protocol increased bladder cancer incidence
in rats. The tumorigenic response was greater in male than in
female rats. Administration of the same doses of this compound
to rats beginning at 35 days of age or later (traditional 2-year
bioassays), however, did not significantly increase the incidence
of bladder tumors (8,10–15).
In contrast to the lack of carcinogenicity of this compound
when administered in a standard 2-year bioassay beginning at
5 weeks of age or greater, sodium saccharin produced a
significant increase in the incidence of tumors when administered to adult rats after brief exposure to known bladder
carcinogens such as MNU (10,11,16,17), FANFT (12–
14,18,19), BBN (20,21) or cyclophosphamide (14), and after
freeze ulceration (14,15) of the bladder. In addition, sodium
saccharin administered to adult rats produced mild superficial
urothelial cell cytotoxicity with consequent mild regenerative
hyperplasia (22–25). The increase in cell proliferation has
been demonstrated by light microscopy, scanning electron
microscopy and by labeling index studies utilizing either
tritiated thymidine (22,23) or BrdU pulse labels (24,25).
Saccharin is generally considered to be non-genotoxic (26).
In most screens for genotoxicity it is negative. However,
positive results have occasionally been observed in chromosomal rearrangement and related assays, but only at high levels
of exposure (26,27). Other sodium and potassium salts, such
as NaCl or KCl, are also positive in these assays (27), and it
has been suggested that the genetic effect is secondary to the
osmotic changes produced by high concentrations of these
salts rather than being due directly to the chemical itself (28).
In addition, saccharin is a moderately strong organic acid [pKa
~2.0 (29)], it is not metabolized (30), and at intracellular pH
it is present nearly entirely as an anion, which would not be
expected to react with DNA. Lutz and Schlatter (31) found no
evidence of DNA–saccharin binding when administered at
high doses to rats. Zukowski et al. (32) also demonstrated that
high doses of sodium saccharin fed to rats did not produce
unscheduled DNA synthesis in the bladder urothelium in an
in vivo–in vitro assay.
Hasegawa and Cohen (33) found that the salt form of
783
S.M.Cohen et al.
saccharin administered significantly influenced whether the rats
developed a proliferative response to the saccharin, although the
doses were essentially the same with the different forms. The
proliferative response to sodium saccharin was greater than
with potassium saccharin, and the response with calcium
saccharin was marginally, but not statistically significantly,
increased. Acid saccharin was without effect on the bladder
epithelium. Although the urinary concentrations of saccharin
were the same following feeding with any of these forms of
saccharin, the other constituents of the urine varied considerably between the different groups as anticipated given the
physiological response to the large load of either sodium,
potassium, calcium or hydrogen ions. Thus, the proliferative
response following high doses of saccharin in the diet is
not dependent solely on the urinary concentration of the
saccharinate ion.
A possible explanation for the differences in urothelial
response to the oral administration of different saccharin salts
is that the various ionic changes in the urine alter the structure
of the saccharinate ion. However, this was examined by
Williamson et al. (29) using various NMR analyses and was
found not to be the case. With variations in the ionic milieu
comparable to the wide variations in urinary concentrations
following administration of these different saccharin salts,
there was no difference in the structure of the saccharinate
ion. These studies and others suggested that the proliferative
response and tumorigenicity of saccharin were probably due
to an indirect process rather than interaction of saccharin itself
with the bladder epithelium. No evidence of saccharin binding
to a cell receptor has been observed (34).
In the 1970s, the Canadian government undertook a detailed
study of the carcinogenicity of sodium saccharin in a twogeneration protocol in rats, and also evaluated the possibility
that its carcinogenic activity was due to a contaminant, OTS
(8). It was conclusively demonstrated that OTS was not
responsible for the bladder carcinogenicity of saccharin in the
rat. Subsequent studies on the proliferative and carcinogenic
effects of saccharin used sodium saccharin synthesized by the
Maumee procedure, which does not result in OTS as a
contaminant, rather than the Remsen–Fahlberg method originally used, which does yield OTS as a contaminant.
During the course of the Canadian bioassay study by Arnold
et al. (8), it was noted that rats fed high doses of sodium
saccharin in the diet developed cloudy urine. The investigators
were able to show that the cloudiness was due to the formation
of a urinary precipitate, and they demonstrated that the precipitate contained relatively small amounts of protein and small
amounts of saccharin (35). Further analyses were not reported.
The presence of this precipitate in the urine was confirmed by
West and Jackson (36), but no analysis of its composition or
its relationship to the carcinogenicity of saccharin undertaken.
In the mid 1980s we rediscovered the presence of this
precipitate in the urine of animals fed high doses of sodium
saccharin, and have since pursued various analytical and
correlative studies of its relationship to the proliferative and
carcinogenic activities of sodium saccharin in the rat (37,38).
Other sodium salts behave similarly to sodium saccharin
when fed at high doses in the diet to rats. Including similar
induction of superficial urothelial cell cytotoxicity, regenerative
hyperplasia and tumorigenicity when administered following
a short exposure to a genotoxic bladder carcinogen, BBN
being the most frequently used carcinogen (18,39–43). A
precipitate, morphologically similar to that following sodium
784
saccharin feeding, forms in the urine of animals fed these
other sodium salts (38).
Initial analyses of this precipitate in our laboratory suggested
that silicon-containing substances (likely to be silicates) were
a significant component and might be related to the biological
effects of the precipitate (37). It is known that a variety of
silicates concentrated in the urine can produce crystalline
material that is cytotoxic to the bladder epithelium leading to
consequent regeneration, and ultimately, to the development
of bladder tumors in various experimental and domestic animals
(44–46). Although subsequent studies in our laboratory have
confirmed the presence of silicon, they suggest that it is only
a minor component of the precipitate. The contribution of
silicon-containing substances to the cytotoxic effects following
feeding of sodium saccharin or other sodium salts remains
unknown.
Since the presence of this precipitate appears to be critical
for the proliferative and tumorigenic effects of these sodium
salts, we have continued our investigations on its composition
and on the factors that influence its formation in the rat. The
results of these investigations are reported here, together with
a discussion of the supporting data from other studies indicating
the role this precipitate has in producing the urothelial effects
in the rat bladder following feeding of sodium saccharin or
other sodium salts at high doses.
Materials and methods
Animal treatments
F344 rats were obtained from Charles River Breeding Laboratories (Kingston,
NY or Raleigh, NC) at 4 weeks of age, weighing ~40–50 g. They were
maintained and quarantined for ~1 week and then started on treatment with
the respective diets for 4–26 weeks, depending on the specific experiment.
Groups of treated and control animals were maintained continuously in the
laboratory for the acquisition of fresh voided urine for precipitate and urinary
chemical analyses. The treatments and protocols of these experiments were
approved by the University of Nebraska Medical Center Institutional Animal
Care and Use Committee. The rats were maintained in a level 4 barrier facility
with five rats per cage in polycarbonate cages (Lab Products, Maywood, NJ)
on corncob bedding (Anderson, Maumee, OH). Room temperature was
maintained at 71 ⫾ 5°C and a relative humidity of 30–70% on a 12 h light/
12 h dark cycle with the changes occurring at 0600 and 1800 h. The rats
were weighed at the beginning of each experiment and then every 2–6 weeks
thereafter. Food and water consumption was measured at the same intervals.
Prolab 3200 diet (Agway, St Mary’s, OH) was used for all experiments except
where stated, and the diets were mixed and pelleted by Dyets (Bethlehem, PA).
Sodium saccharin was received as a gift from PMC Specialties Group
(Cincinnati, OH). Sodium ascorbate and ammonium chloride were purchased
from Sigma (St Louis, MO). The purity of saccharin was verified by NMR
analysis. Sodium ascorbate and ammonium chloride were used as received,
relying on the company’s chemical analyses as reported with the shipment.
Dietary levels of the chemicals were determined by methods previously
described (47,48).
Processing of tissues
At the end of each experiment the animals were killed with an overdose of
Nembutal (50 mg/kg i.p.). The bladder and stomach were inflated in situ with
Bouin’s fixative, ligated, immediately removed, and placed in the same
fixative, and then washed with 70% ethanol. The kidneys were also removed
and placed in 10% phosphate buffered formalin (pH 7.4). All tissues were
processed for light microscopy. The bladders were also processed for scanning
electron microscopic evaluation as described previously (23). For some studies,
1 h prior to terminal sacrifice the rats received an i.p. injection of BrdU (100
mg/kg) for determination of the labeling index of the epithelial cells. A slice
of stomach, through the limiting ridge, was included in the cassettes with the
slices of bladder tissue to be processed for light microscopy; the stomach
served as a positive marker of labeling with the BrdU.
Urine collection
Because of variability occurring secondary to diurnal variations and changes
that can occur secondary to long term collections (49), fresh voided urine
specimens were used for most of the analyses. The procedure for obtaining
Urinary precipitate and bladder cancer
fresh voided urine could not be performed too frequently as this procedure
itself can produce changes in the urine and in the urothelium of the rat (50).
Fresh voided urine specimens were collected by holding the rats directly over
a 1.5 ml conical plastic test tube and firmly squeezing the back of the neck.
This resulted in urine samples from rats ~90% of the time, with a sample size
of 10–1000 µl. Urine specimens were collected between 0700 and 0900 h.
Depending on the experiment, either whole urine was used for chemical
analyses or the specimen was centrifuged at 7000 g and the supernatant
aspirated for chemical analyses.
Processing of precipitate for analysis
To examine the precipitate morphologically, it was transferred to 22-µm filters
(Millipore, Bedford, MA) which were glued to aluminum specimen stubs and
then processed for scanning electron microscopic observation as described
previously (38). The specimens were examined in a 515 Phillips Scanning
Electron Microscope (Phillips, Eindhoven, The Netherlands) with an attached
Kevex Micro-X 7000 X-ray energy dispersive spectrometer (Kevex, Haywood,
CA) at 20 kV voltage. The specimens were examined uncoated to provide a
true spectrum without a heavy metal peak overlap.
For chemical analyses of the precipitate, samples from individual rats were
used whenever possible although on occasion samples were pooled between
various animals for specific analyses as indicated below. For the chemical
analyses, the precipitate was wicked dry after centrifugation, and then
resuspended in 100 µl 100 mM HCl. Silicon composition was analyzed by
atomic absorption utilizing a Perkin Elmer Model 1100 Atomic Absorption
spectrometer (Norwalk, CT) with a wavelength of 251.6 nm. Calcium (51),
phosphorus (52) and magnesium (53) were analyzed using micro methods
(Sigma). Protein was analyzed utilizing SDS–PAGE for total protein and
western blot analyses for α2u-globulin and albumin. The antibody for α2uglobulin was purchased from Hazelton (Vienna, VA); the antibody for
albumin was purchased from The Binding Site (San Diego, CA). Sulfated
mucopolysaccharides were analyzed using a colorimetric method (54) based
on binding of 1,9-dimethylmethylene blue (Biocolour, Belfast, Ireland).
Analysis for specific mucopolysaccharides was performed using an HPLC
method (55). The HPLC system consisted of a Waters (Milford, MA) 510
pump and WISP 712 injector and a Perkin Elmer LS-5B Luminescence
Spectrometer. The pump was operated at a flow rate of 0.8 ml/min. Separations
were carried out on a Nucleosil SB 5 anion exchange column (4.6⫻150 mm,
5 µm particles) for sulfated glycosaminoglycans and on a Hypersil APS weak
anion exchange column (4.6⫻250 mm, 5 µm particles) connected to the
Hypersil ODS RP-C 18 column for unsulfated glycosaminoglycans. Saccharin
(47) and ascorbate (48) were analyzed as previously described.
Urinary chemistries
Urinary silicon (56) concentration, calcium (57), phosphorus (52), magnesium
(58), sodium (59), potassium (60), chloride (61) and creatinine (62) were
performed on whole urine or supernatant (no differences were observed
between these) utilizing methods previously described. All but silica were
assayed on an Ektachem 700 (Kodak, Rochester, NY). These could all be
done on a relatively small sample volume (10 µl), however, occasionally not
all could be done on the same sample. Osmolality required a larger sample
(100 µl) and was determined by freezing point depression using a 3MO
Advanced Micro-Osmometer (Norwood, MA).
Infrared analysis
Infrared spectra were collected using a Digilab FTS-40A spectrometer (BioRad Laboratories, Cambridge, MA), equipped with a Split-Pea sampling
accessory (Harrick Scientific, Ossining, NY). This accessory measured the
attenuated total reflection spectrum of the sample, pressed into close contact
with the surface of the silicon optical element. The infrared beam probed a
circular area of ~300 µm diameter, penetrated to a depth of ~4 µm. All spectra
were acquired at a resolution of 2 cm–1, co-adding 256 interferograms.
To guide the interpretation of the precipitate infrared spectra, infrared
spectra were also measured for several reference compounds that were
considered as possible constituents. These included urea, silica, sodium
saccharin, sodium ascorbate, mono-, di- and tribasic calcium phosphates,
sodium chondroitin sulfates B and C, sodium hyaluronate, sodium heparin
sulfate, α2u-globulin, albumin, uric acid, ascorbic acid, creatinine, oxalic acid
and citric acid.
Immunoperoxidase staining
Paraffin-embedded sections were immunohistochemically stained utilizing a
Vectastain Elite ABC kit (Vector Laboratories, Burlingame, CA) for BrdU (63).
Antibody to BrdU was obtained from Boehringer Mannheim (Indianapolis, IN)
and was utilized at a dilution of 1 to 50. Anti-α2u-globulin (Hazelton) at a
dilution of 1 to 100 000 was used to stain for α2u-globulin according to the
procedure of Dietrich and Swenberg (64).
Fig. 1. Scanning electron microscopic appearance of amorphous urinary
calcium phosphate-containing precipitate from a rat fed 7.5% sodium
saccharin; ⫻263.
Collection of urine from renal pelvis
Urine was collected from the upper portion of the ureter of rats for
determination of the presence or absence of the amorphous precipitate in the
upper urinary tract, in addition to its presence in urine in the bladder. Rats
were anesthetized with Metofane. Normal saline was infused either through
the jugular vein or femoral artery prior to collection of specimens. The ureters
were dissected, severed and allowed to drain directly onto 22-µm filters.
In vitro effects of calcium phosphate
To evaluate the cytotoxic nature of calcium phosphate, C8 rat bladder epithelial
cells (65) were grown to 70–80% confluency in 25 cm2 flasks in Medium
199 (Gibco, Grand Island, NY) and Ham’s F-12 (1:1) (Gibco) with 5% fetal
bovine serum. C8 cells are immortalized bladder epithelial cells, but are not
tumorigenic when transplanted back to syngeneic rats. The cells do not show
terminal differentiation. This medium contains 0.7 mM calcium and 1.1 mM
phosphorus. The medium was then supplemented with 5 mM calcium
phosphate which led to the formation of precipitate in the medium. Cell
viability was examined microscopically and by trypan blue exclusion at 1, 6
and 24 h after exposure began.
Results
Precipitate analysis
The precipitate is a white, flocculent material that sediments
upon centrifugation. By scanning electron microscopy it has
a typical appearance (Figure 1) regardless of the circumstances
of animal treatment leading to its formation in the urine. The
flocculent material is in addition to the usual magnesium
ammonium phosphate crystal (Figure 2) that is present in
control rat urine. The precipitate is rarely present in the urine
of control rats. It can be distinguished from deteriorated
magnesium ammonium phosphate crystals and from cellular
material partly by its more granular appearance and definitively
by its X-ray energy dispersive spectrum.
By using energy dispersive X-ray spectroscopy, several
elements are detected (Figure 3); the most prominent are
calcium and phosphorus. The phosphorus peak appears to be
related to its presence as a phosphate (see below). Other
elements that are present are magnesium, silicon and sulfur.
Lower atomic weight elements, such as carbon, nitrogen,
oxygen and sodium, are not detectable by this method. The
extent of the presence of the elements detected by energy
dispersive X-ray spectroscopy other than calcium and phosphorus is relatively minor.
The infrared spectra of urinary precipitates from rats fed
sodium saccharin could be reconstructed on the basis of
five major constituents (Figure 4). The clearest signatures
785
S.M.Cohen et al.
Fig. 2. Scanning electron microscopic appearance of urinary magnesium
ammonium phosphate crystals from a control rat; ⫻600.
Fig. 3. X-ray energy dispersive spectroscopy of urinary calcium phosphatecontaining precipitate.
Fig. 4. Infrared spectra of various precipitate specimens and pure chemicals.
correspond to saccharin and urea; each of these compounds
exhibits a unique infrared ‘fingerprint’ that is clearly discernible
in the spectrum of the precipitate. More subtle precipitate
absorptions were best accounted for by postulating the presence
of tribasic phosphate, silicate and sodium chondroitin sulfate.
Both tribasic (but not mono- or dibasic) phosphate and silicate
show very simple absorption spectra that account for the
786
diffuse precipitate absorption in the 950–1200 cm–1. Finally,
these assignments are consistent with (i) the X-ray spectroscopic evidence suggesting high calcium and phosphorus
levels, as well as the presence of silicon, and (ii) other evidence
(Alcian blue staining and chemical analyses by colorimetric
methods) indicating substantial amounts of acid mucopolysaccharide (see below, HPLC analysis).
The infrared spectrum of urinary precipitate from animals
fed sodium ascorbate revealed no ascorbate and lower total
mucopolysaccharide levels than found in precipitate from rats
fed sodium saccharin (Figure 4). In addition, the peak at 995
cm–1, assigned to tribasic phosphate, is much more pronounced
in the precipitate spectra for ascorbate-fed rats than in the
spectra from saccharin-fed rats.
Infrared spectroscopy is unsuitable for the detection of trace
components in the precipitates. Although the five components
discussed here are sufficient to reproduce all major features
of the precipitate spectra, we cannot rule out the presence of
other species in concentrations ⬍5%.
Specific quantitative analyses, for various chemicals in the
precipitate, show that the major inorganic components were
calcium and phosphate. However, the amounts varied considerably. The inorganic components of the precipitate usually
accounted for 10–40% of the material, with ~85% of the
inorganic material being calcium phosphate. Magnesium was
also present to a variable extent. This correlated well with the
presence of magnesium ammonium phosphate crystals in the
urine. Extraction of the precipitate with freon 113 (SPI Supplies,
West Chester, PA) removed the crystals from the sediment and
also tended to eliminate the magnesium peak by X-ray energy
dispersive spectroscopy and by chemical analysis. However,
we cannot completely exclude the possibility that magnesium
is present in the precipitate in addition to the crystals, although
clearly calcium and phosphate are the major inorganic components. No other differences were observed between freonextracted precipitate compared with the precipitate not
extracted. In contrast, the sediment from untreated control rats
contained primarily magnesium and phosphate, representing
the magnesium ammonium phosphate crystals. Calcium was
usually present as ⬍5% of the control sediment, and often
⬍1%.
Silicon-containing material was also consistently detected
in the precipitate as analyzed by atomic absorption. However,
it tended to be a relatively minor component of the precipitate,
and it was at similar levels in the sediment obtained from
control rats and from saccharin-treated rats.
Extracts of the precipitate also demonstrated small amounts
of saccharin (from animals fed saccharin), although this
represented, consistently, a very small percentage of the total
precipitate. Levels of ascorbate detected in the precipitate of
animals fed sodium ascorbate were similar to the levels found
in urinary sediment from control rats.
Although protein did not appear to contribute prominently
to the infrared spectrum, small amounts of protein could be
detected by extraction of the precipitate and chemical analysis.
Also, we were able to specifically identify, by western blot
analyses, that the two major proteins present in the precipitate
were α2u-globulin and albumin.
A large amount of organic material was consistently present
in the precipitate. The amount of sulfur present as saccharin
is unlikely to explain the amount of sulfur that was detected
by X-ray energy dispersive spectroscopy. Also, similar levels
of sulfur were detected by X-ray energy dispersive spectroscopy
Urinary precipitate and bladder cancer
in the precipitate from rats fed high levels of sodium ascorbate.
We therefore evaluated the possibility of the sulfur being due
to mucopolysaccharides in the precipitate. Initially, staining of
the precipitate with Alcian blue (66) at pH 2.5 demonstrated
that there was a large amount of acid mucopolysaccharide
present. HPLC analysis of the precipitate from rats fed sodium
saccharin showed chondroitin sulfate to be the major mucopolysaccharide present. No other mucopolysaccharides were
detected in the precipitate because they were not present, or
were present in amounts below the detection limit of the
methodology (2.5 µg/ml).
Urine analyses
In rats administered either sodium ascorbate or sodium saccharin, the changes in urine were similar to those previously
described, including the decrease in osmolality because of the
increased urinary output. This was associated with decreases in
urinary potassium, chloride, total protein, urea, and creatinine.
However, when the sodium salt was administered, urinary
sodium tended to increase. Against the dilutional effect, the
concentrations of calcium, phosphate, and usually magnesium,
tended to be the same or increased compared to control urine.
When normalized to creatinine concentration, there were
marked increases in the excretion of calcium and phosphate,
and usually magnesium. Silicon concentrations tended to be
similar in the urine of both treated and control rats.
The amount of total mucopolysaccharides in the urine for
control and sodium saccharin-treated rats was 0.2 ⫾ 0.0 and
0.4 ⫾ 0.0 mg/ml, respectively. The major mucopolysaccharide
was chondroitin sulfate (0.1 and 0.3 mg/ml in urine from
control and treated rats, respectively). Small amounts of heparin
sulfate were also detectable.
Total urinary protein tended to be decreased because of the
overall dilution of the urine, but remained at extremely high
levels as is typical of rat urine. The males had significantly
more protein than females, and the difference was largely due
to α2u-globulin. The amounts of α2u-globulin and albumin
were similar proportionally in treated versus control rat urine.
In rats administered sodium saccharin, similar to other forms
of saccharin (33), the urinary concentration of saccharin was
directly proportional to the amount administered in the diet,
reaching levels as high as 300 mM in the urine of rats fed
7.5% sodium saccharin in the diet. In contrast, the urinary
concentration of ascorbate (ascorbate plus dehydroascorbate)
was significantly lower (2.7 mM) than saccharin following
comparable doses of sodium ascorbate in the diet. Nevertheless,
there was a clear dose response for urinary ascorbate following
administration of sodium ascorbate at levels 1.0–7.0% of
the diet.
Male versus female
The concentration of saccharin and the other urinary parameters
that were measured were similar in male compared with female
rats except for total protein (Table I). In contrast, the frequency
and the amount of precipitate appearing in the urine was
significantly less in female rats than in male rats, although it
was present in female rat urine administered high doses of
sodium saccharin.
Dose response
Although we have not been able to quantify the amount of
precipitate in the urine, an estimate was made measuring the
optical density of the urine at 620 nm. This method is not
sensitive but gives a general evaluation of the cloudiness of
the urine that indicates the presence of solids in the urine.
Utilizing that method, a no-effect level occurred at 1% sodium
saccharin in the diet. In addition, at the dose of 1% sodium
saccharin in the diet, there was no precipitate present in the
urine following examination by scanning electron microscopy,
a very sensitive method for detecting precipitate if present at
all. At the dose of 3% sodium saccharin in the diet, there
was occasionally precipitate present, but in relatively small
amounts. The precipitate was consistently present at high
doses, with a significant increase in optical density at a dose
of 7.5% of the diet. Similar observations have been made with
sodium ascorbate, although a small amount of precipitate was
detected in five of seven rats fed sodium ascorbate even at a
dose of 0.91% of the diet and heavy amounts in six of seven
rats fed sodium ascorbate as 2.73% of the diet (67).
In vitro cytotoxicity
Calcium phosphate is an essential ingredient of any tissue
culture medium, including urothelial cells. The medium that
we have routinely used in our laboratory [Medium 199 and
Ham’s F-12 (1:1)] has a calcium concentration of 0.7 mM.
When 5 mM calcium phosphate is added (total concentration
5.7 mM) calcium phosphate precipitates from the medium and
settles onto the urothelial cells. After 1 h there are signs of
cytotoxicity, including decreased cell number. Toxicity is
greatly increased by 6 h, and most of the cells died by 24 h
(Table II). At a concentration of 25 mM, the results are even
more profound. There was no visual change in pH based on
color of the medium. No other constituents of the media were
varied, however, levels were not measured.
Other observations
Calcium phosphate-containing precipitate was observed in the
urine of male rats administered 7.5% sodium saccharin in
the diet beginning sporadically as early as 1 day after the
administration of the chemical. It was more frequently present
by 3 days, and was usually present in the urine by 7 days and
subsequently. The presence and the amount of precipitate
varied between rats at a given point in time and in the same
rat at different times. This continued as long as the chemical
was administered in the diet up to 26 weeks of administration.
The urothelial changes following administration of high
doses of sodium saccharin in the diet are seen in the renal
pelvis, ureters and urinary bladder (68,69). We have repeatedly
been able to demonstrate the presence of the calcium phosphatecontaining precipitate in fresh voided urine of rats administered
sodium saccharin, and we have demonstrated its presence in
the urine taken directly from the urinary bladder of rats. Studies
to determine if a calcium phosphate-containing precipitate was
present in the upper urinary tract of sodium saccharin-treated
rats were inconclusive. We obtained urine from the upper
ureters, presumably representing urine from the upper urinary
tract including the renal pelvis. In the urine we found evidence
of the same calcium phosphate-containing precipitate seen in
the urine voided by treated rats and in the urine obtained from
the urinary bladder in two samples from a sodium saccharintreated rat. However, it was also present in one ureteral sample
obtained from a control rat.
In rats fed 7.5% sodium saccharin in the diet, staining of
the bladders and kidneys for α2u-globulin using an antibody
to the protein showed no evidence of the protein in the
urothelium of the renal pelvis, ureter or urinary bladder. As a
positive control on the same tissue sections, it was present in
787
S.M.Cohen et al.
Table I. Twenty-four hour urine chemistries during week 6 of treatment with 7.5% sodium saccharin
Group
Treatment
Sex
Volume
(ml)
1
2
Control
7.5% Sodium
saccharin
Control
7.5% Sodium
saccharin
M
M
4.3 ⫾ 0.3
117 ⫾ 4
17.4 ⫾ 1.5a 37 ⫾ 3a
%
2377 ⫾ 68 199 ⫾ 10 406 ⫾ 19
165 ⫾ 15a 1161 ⫾ 71a 251 ⫾ 17a 130 ⫾ 8a
F
F
2.9 ⫾ 0.3
113 ⫾ 7
11.3 ⫾ 0.5b,c 40 ⫾ 2b
%
179 ⫾ 5b
3
4
Creatinine
(mg/dl)
Saccharin
(mM)
Osmolality
(mosm/l)
Na
(mEq/l)
K
(mEq/l)
2304 ⫾ 87 222 ⫾ 20 436 ⫾ 39
1318 ⫾ 42b 274 ⫾ 15b 141 ⫾ 4b
Cl
(mg/dl)
Ca
(mg/dl)
Mg
(mg/dl)
242 ⫾ 12 6.8 ⫾ 0.4 29 ⫾ 2
81 ⫾ 6a 12.5 ⫾ 0.6a 44 ⫾ 4a
P
(mg/dl)
Protein
(mg/dl)
223 ⫾ 6
136 ⫾ 9a
1.7 ⫾ 0.1
0.3 ⫾ 0.1a
250 ⫾ 15 0.2 ⫾ 0.1a
264 ⫾ 22 14.8 ⫾ 0.9a 36 ⫾ 5
88 ⫾ 3b 14.8 ⫾ 1.3 56 ⫾ 3b,c 142 ⫾ 7b 0.1 ⫾ 0.0c
different from male control group, P ⬍ 0.05.
different from female control group, P ⬍ 0.05.
cSignificantly different from male 7.5% sodium saccharin group, P ⬍ 0.05.
aSignificantly
bSignificantly
Table II. Cytotoxicity of calcium phosphate precipitate to rat bladder epithelial cells
Calcium
concentration
added (mM)
Total number
of cells
1h
Percent
viable
Total number
of cells
6h
Percent
viable
Total number
of cells
24 h
Percent
viable
Control
1
5
25
1.1⫻106
1.3⫻106
8.0⫻105
1.1⫻106
93.7
95.5
51.9
39.2
2.4⫻106
1.9⫻106
1.4⫻106
2.4⫻106
95.7
96.3
78.0
31.2
1.3⫻106
1.2⫻106
1.1⫻106
2⫻105
95.6
95.9
88.1
22.0
the cells of the proximal tubules, similar to that demonstrated
by numerous investigators previously (64).
Discussion
Administration of high doses of sodium salts produces an
increase in urothelial proliferation in rats (22–25). The proliferation represents a regenerative response to cytotoxicity of
the superficial cell layer of the urothelium (23). As discussed
above, it appears that the anion of the sodium salt, such as
saccharin or ascorbate, does not directly produce the urothelial
toxicity. The structure of the anion is not changed by the other
variations that occur in the urinary milieu (29), such as pH,
and there is no evidence that these anions, such as saccharin,
bind to a cell surface receptor (34). Since sodium salts, such
as ascorbate and chloride, are essential ingredients of the
diet or intermediates of cell metabolism (e.g. glutamate,
bicarbonate, aspartate, succinate), it is not surprising that they
are handled metabolically somewhat differently than saccharin.
Saccharin is not metabolized (30), and most of it is excreted
in the urine within 24 h of administration (30). Concentration
in the urine can reach levels as high as 200–300 mM following
the administration of various saccharin salts at 5 or 7.5% of
the diet. In contrast, administration of sodium ascorbate at
similar doses in the diet produces nearly one to two orders of
magnitude lower concentration in the urine of the corresponding anion, ascorbate (as ascorbate plus dehydroascorbate).
These findings also suggest that the cytotoxicity of the urothelium following administration of high doses of these sodium
salts occur by an indirect mechanism.
Nearly two decades ago Arnold et al. (8) observed that a
precipitate formed in the urine following administration of
high doses of sodium saccharin, and this was corroborated by
observations by West and Jackson (36). We have also observed
this precipitate in the urine of rats fed high doses of sodium
saccharin (37,38), and in the urine of rats fed various other
sodium salts (38,67) that have been found to produce urothelial
proliferation.
788
Formation of this precipitate appears to be necessary for the
urothelial cytotoxicity, hyperplasia and tumorigenicity of these
sodium salts. Any treatments that inhibit the formation of this
precipitate inhibit the induction of the toxicity, proliferative
and tumorigenic effects. For example, one of the necessary
conditions for the formation of this precipitate is a pH of ~6.5
or greater (37,38,67). If the rat is treated in such a way that it
produces acidic urine, no precipitate forms and there is no
toxicity, regeneration, or tumor formation (17–19,38,49,67).
Acidification can be produced either by administration of the
parent acid, such as acid saccharin (17,33) or ascorbic acid
(39), or by the sodium salt co-administered with high doses
of ammonium chloride (38,49,67). Administration of the
sodium salt in AIN-76A, a semi-synthetic diet with casein as
the protein source, also results in acidic urine (49) and
inhibition of the urinary and urothelial effects of sodium
salts (19,70).
An apparent exception to the tumorigenicity of the sodium
salts is sodium hippurate (71). Administration of this salt at
high doses in the diet to male rats does not produce proliferation
or tumor formation. However, the administration of high doses
of sodium hippurate leads to a urine pH of ~6.3 or less, which
is not conducive to the formation of the precipitate.
The formation of the precipitate is also dose dependent. The
precipitate is consistently present at high doses of sodium
saccharin, occasionally present at a dose of 3% of the diet and
absent at a dose of 1%. The tumorigenic effects of sodium
saccharin on the urothelium show a similar dose response with
a significant increase in bladder tumors occurring at doses
艌3% (9).
The present studies also demonstrate that calcium phosphate
precipitate is cytotoxic when added to the medium of urothelial
cells in culture. Calcium phosphate has long been known to
be cytotoxic to epithelial cells (72), and the observations with
urothelial cells adds to the long list of other epithelia that
respond to this precipitate. In contrast to epithelial cells,
calcium phosphate precipitate is not cytotoxic to mesenchymal
Urinary precipitate and bladder cancer
cells, such as fibroblasts, but rather has been used as an agent
to transfect DNA into such cells (72). The present studies
demonstrate that the precipitate is formed soon after sodium
saccharin administration begins, and that it may form high in
the urinary tract, being present in the renal pelvis as well as
the ureter and the urinary bladder.
The major components of the precipitate appear to be
calcium and phosphate, present as Ca3(PO4)2, and the major
organic component appears to be sulfated acid mucopolysaccharides, principally chondroitin sulfate. Significant amounts of
urea are also present. Small amounts of saccharin (in rats fed
sodium saccharin) and protein are also present along with
small amounts of silica. Whether magnesium is a constituent
of the precipitate or only a contaminant representing the
magnesium ammonium phosphate in crystals also present in
the urine could not be definitely ascertained. However, the
more we were able to eliminate crystals from the precipitate,
the less magnesium that was present in the material.
The composition of the precipitate appears to vary considerably between animals and even within the same animal at
different times. Although calcium and phosphate, urea, and
mucopolysaccharide were always present, the proportion of
each varied considerably. Likewise, the other components
varied widely between the rats. The analyses that were performed both qualitatively and quantitatively, using X-ray
energy dispersive spectroscopy, infrared spectroscopy and
chemical analyses, provided means to establish the general
composition of the precipitate.
It is clear that protein is present in the precipitate, but it is
present only in small amounts. The role that the protein plays
in this precipitate is unclear. We have suggested (34,37) that
it might principally be acting as a nidus for the formation of
the urinary precipitate, a common process in the formation of
urinary solids such as precipitate, crystals and calculi (73).
The presence of large amounts of protein in rat urine appears
to explain why the rat is susceptible to the formation of this
precipitate as well as the proliferative and tumorigenic activity
of these sodium salts, and also helps to explain the difference
in response between male and female rats. Although the urine
of the male rat has considerably more protein than the urine
of the female rat, the female rat does have a large amount of
protein, mostly albumin. The difference between the sexes is
due to the presence of large amounts of α2u-globulin in the
urine of male rats (74). It is probable that this contributes
significantly to the biological effects in the urinary tract
following administration of these sodium salts. For example,
administration of sodium saccharin (75,76) or sodium ascorbate
(76) to the NBR male rat, which does not excrete α2u-globulin
in large quantities in the urine, produces significantly less
proliferative effect than does administration to Fischer or
Sprague–Dawley rats, which have large amounts of α2uglobulin. The response in the NBR male rat to these salts is
similar to the response of female rats of other strains.
These proteins are present in normal rat urine, and yet
calcium phosphate does not usually precipitate in the urine of
control animals. The changes that occur in the urine of rats
treated with high doses of these sodium salts that leads to
formation of this precipitate are unclear. The anions of these
sodium salts, such as saccharin or ascorbate, are known to
non-covalently associate with proteins, such as albumin and
to an even greater extent to α2u-globulin, which may alter
their structure in a way that leads to formation of this
precipitation. The role of mucopolysaccharides in this process
is also unclear, but could possibly be related to the interaction
of calcium and phosphate.
The role of the extremely high osmolality of rat urine in
the formation of the precipitate is also unclear. Rat urine, like
that of other rodents, is very concentrated, especially in contrast
to the urine of primates, such as humans (77). Human urinary
osmolality is usually in the range of 50–500 mosm/l. Theoretically it has been calculated that osmolality in human urine may
reach levels as high as 1000–1200 mosm/l. Rat urine routinely
has an osmolality above 1000 mosm/l, even ⬎2000 mosm/l.
This increased osmolality is due partly to the significantly
greater concentrations of urea in rat urine than in human urine
(77). Increased urine volume with distension of the bladder
has also been suggested as contributing to the tumorigenicity
of the sodium salts (78). In contrast, increased water and total
liquid consumption are related to a decreased risk of bladder
cancer in men (79).
It has been suggested that high sodium concentrations along
with pH by itself are adequate to produce the proliferative
response of the sodium salts (80). However, high urinary pH
by itself, clearly, is not adequate to produce these effects. For
example, the feeding of a commonly used rodent chow in
Europe, Altromin 1321 (Altromin GmbH and Co., Germany),
routinely produces an urinary pH of 8.0 or higher, and yet the
bladder epithelium of control animals fed this diet remains
normal (81). Similarly, high sodium levels are produced at
similar levels in male and female rats fed high doses of sodium
saccharin, and yet the response of the urothelium is clearly
greater in the male rat (see Ref. 82 and present studies). Even
higher sodium concentrations can be achieved in mice fed
high doses of these sodium salts, and yet there is no toxic or
proliferative response in their urothelium (83).
An explanation is necessary for the lack of effect in the
urothelium of mice fed high doses of these sodium salts. When
sodium saccharin is fed at concentrations of 5 or 7.5% of the
diet, even higher concentrations of sodium and saccharin occur
in the urine of mice than in rat (83). Also, the mouse has
similar high urinary protein concentrations, and the mouse has
a protein analogous to α2u-globulin, mouse urinary protein
(MUP) (84). MUP associates with anions such as saccharin in
a manner and to an extent similar to that of α2u-globulin
(37,85). Nevertheless, we have not observed formation of the
urinary precipitate (unpublished observations), and there has
been no evidence of increased proliferation or tumorigenicity
in mice administered sodium saccharin (86).
The mouse has urinary concentrations of calcium, phosphate
and magnesium that are significantly lower than in the rat (77).
This is particularly true for calcium, where the concentration is
as much as 10–20 times greater in the rat than in the
mouse. Since the solubility product for calcium phosphate is
a multiplicative rather than an additive function of the concentration of these components, it is likely that the large differences
in concentration is at least a partial explanation for lack of
precipitate formation in the mouse. It also points out an
additional correlation between the presence or absence of the
formation of this precipitate and the presence or absence of a
consequent toxic or proliferative response.
α2u-Globulin has been associated with renal carcinogenesis
for a variety of compounds that reversibly bind to the protein,
are absorbed into the proximal tubules, are essentially undigested and accumulate in these tubular cells, and lead to tubular
cell death and consequent regeneration (87). There is some
indication that α2u-globulin might also contribute to the aging
789
S.M.Cohen et al.
nephropathy common in male rats. It does not appear that
saccharin associates with the α2u-globulin in the same way
that rat renal carcinogens, such as D-limonene (L.LehmanMcKeeman, personal communication). Of interest has been
the repeated observation with the sodium salts that long term
administration at high doses tends to inhibit the formation of
the aging nephropathy that occurs in male rats over time (69).
This inhibitory effect can be at least partially reversed by coadministration with ammonium chloride (48), suggesting that
the interaction of these anions with α2u-globulin occurs within
the renal parenchyma and not initially within the renal pelvis.
In summary, a precipitate forms in the urine of rats administered high doses of sodium salts such as saccharin and
ascorbate, and it appears to be an essential determinant in the
toxicity of the urothelium in rats fed high doses of these
compounds with consequent regeneration and tumor formation.
The precipitate is composed principally of calcium phosphate,
mucopolysaccharides and urea, with protein, anions (such as
saccharin), and silica also being present possibly along with
magnesium. Precipitated calcium phosphate is cytotoxic to
urothelial cells, as is the calcium phosphate-containing urinary
precipitate that forms in these rats. The precipitate appears to
act directly on the superficial cells without being incorporated
into the epithelium. The conditions necessary for the formation
of the precipitate such as, high urinary pH (艌6.5), high
urinary protein and osmolality, large amounts of urinary
mucopolysaccharides, and adequate levels of calcium and
phosphate (and possibly magnesium) suggest that the rat is
the only sensitive species to the urothelial effects of feeding
high doses of these sodium salts. The male rat is more
susceptible than the female, and for reasons discussed above,
mice and primates are not susceptible to the effects of high
doses of these sodium salts because they do not produce the
necessary conditions for the formation of this precipitate. The
data presented, in addition to other observations, suggest that
the tumorigenicity following administration of high doses of
these sodium salts is a high dose, rat specific phenomenon.
Acknowledgements
We gratefully acknowledge the assistance of Dr Angela Mann, Steve Eklund,
Margaret St John, Traci Anderson, Linda Johnson, Cindi Lear and Barbara
Mattson with the performance of these studies, and Denise Miller and Michelle
Moore with the preparation of the manuscript. The research was supported in
part by USPHS grants CA32513 and CA36727 from the National Cancer
Institute and a grant from the International Life Sciences Institute. S.M.C. is
the Wall-Havlik Professor of Oncology.
References
1. Bryan,G.T., Erturk,E. and Yoshida,O. (1970) Production of urinary bladder
carcinomas in mice by sodium saccharin. Science, 16, 1238–1240.
2. Clayson,D.B. (1974) Bladder carcinogenesis in rats and mice: possibility
of artifacts. J. Natl Cancer Inst., 52, 1685–1689.
3. DeSesso,J.M. (1989) Proliferative lesions of the urinary bladder: their
development and interrelationships. Comm. Toxicol., 3, 335–348.
4. Clayson,D.B., Fishbein,L. and Cohen,S.M. (1995) The effect of stones
and other physical factors on the induction of rodent bladder cancer. Fd
Chem. Toxicol., 33, 771–784.
5. Arnold,D.L., Krewski,D. and Munro,I.C. (1983) Saccharin: a toxicological
and historical perspective. Toxicology, 27, 179–256.
6. Tisdel,M., Nees,P.O., Harris,D.L. and Derse,P.H. (1974) Long term feeding
of saccharin in rats. In Inglett,G.E. (ed.) Symposium: Sweeteners. AVI
Publishings, Connecticut, p. 145.
7. Taylor,J.M., Weinberger,M.A. and Friedman,L. (1980) Chronic toxicity
and carcinogenicity to the urinary bladder of sodium saccharin in the in
utero-exposed rat. Toxicol. Appl. Pharmacol., 54, 57–75.
8. Arnold,D.L., Moodie,C.A., Grice,H.C., Charbonneau,S.M., Stavric,B.,
790
Collins, B.T., McGuire,P.F., Zawidzka,Z.Z. and Munro,I.C. (1980) Longterm toxicity of ortho-toluenesulfonamide and sodium saccharin in the rat.
Toxicol. Appl. Pharmacol., 52, 113–152.
9. Schoenig,G.P., Goldenthal,E.I., Geil,R.G., Frith,C.H., Richter,W.R. and
Carlborg,F.W. (1985) Evaluation of the dose response and in utero exposure
to saccharin in the rat. Fd Chem. Toxicol., 23, 475–490.
10. Hicks,R.M., Wakefield,J. and Chowaniec,J. (1973) Co-carcinogenic action
of saccharin in the chemical induction of bladder cancer. Nature, 243,
347–349.
11. Whysner,J. and Williams,G.M. (1996) Saccharin mechanistic data and risk
assessment: urine composition, enhanced cell proliferation and tumor
promotion. Pharmacol. Ther., 71, 225–252.
12. Cohen,S.M., Arai,M., Jacobs,J. B. and Friedell,G.H. (1979) Promoting
effect of saccharin and DL-tryptophan in urinary bladder carcinogenesis.
Cancer Res., 39, 1207–1217.
13. Fukushima,S., Friedell,G.H., Jacobs,J.B. and Cohen,S.M. (1981) Effect of
L-trypotophan and sodium saccharin on urinary tract carcinogenesis
initiated by N-[4-(5-nitro-2-furyl)-2-thiazolyl]formamide. Cancer Res., 41,
3100–3103.
14. Cohen,S.M., Murasaki,G., Fukushima,S. and Greenfield,R. E. (1982) Effect
of regenerative hyperplasia on the urinary bladder: carcinogenicity of
sodium saccharin and N-[4-(5-nitro-2-furyl)-2-thiazolyl]formamide.
Cancer Res., 42, 65–71.
15. Hasegawa,R., Greenfield,R.E., Murasaki,G., Suzuki,T. and Cohen,S.M.
(1985) Initiation of urinary bladder carcinogenesis in rats by freeze
ulceration with sodium saccharin promotion. Cancer Res., 45, 1469–1473.
16. Chowaniec,J. and Hicks,R.M. (1979) Response of the rat to saccharin with
particular reference to the urinary bladder. Br. J. Cancer, 39, 355–375.
17. West,R.W., Sheldon,W.G., Gaylor,D.W., Allen,R.R. and Kadlubar,F.F.
(1994) Study of sodium saccharin in co-carcinogenicity in the rat. Fd
Chem. Toxicol., 32, 207–213.
18. Cohen,S.M., Ellwein,L.B., Okamura,T., Masui,T., Johansson,S.L.,
Smith,R.A., Wehner,J.M., Khachab,M., Chappel,C.I., Schoenig,G.P.,
Emerson,J.L. and Garland,E.M. (1991) Comparative bladder tumor
promoting activity of sodium saccharin, sodium ascorbate and related
acids and calcium salts in rats. Cancer Res., 51, 1766–1777.
19. Okamura,T., Garland,E.M., Masui,T., Sakata,T., St.John,M. and
Cohen,S.M. (1991) Lack of bladder tumor promoting activity in rats fed
sodium saccharin in AIN-76A diet. Cancer Res., 51, 1778–1782.
20. Nakanishi,K., Hirose,M., Ogiso,T., Hasegawa,R., Arai,M. and Ito,N. (1980)
Effects of sodium saccharin and caffeine on the urinary bladder of rats
treated with N-butyl-N-(4-hydroxybutyl)nitrosamine. Jpn. J. Cancer Res.,
71, 490–500.
21. Fukushima,S., Arai,M., Nakanowatari,J., Hibino,T., Okuda,M. and Ito,N.
(1983) Differences in susceptibility to sodium saccharin among various
strains of rats and other animal species. Jpn. J. Cancer Res., 74, 8–20.
22. Fukushima,S. and Cohen,S.M. (1980) Saccharin-induced hyperplasia of
the rat urinary bladder. Cancer Res., 40, 734–736.
23. Cohen,S.M., Fisher,M.J., Sakata,T., Cano,M., Schoenig,G.P., Chappel,C.I.
and Garland,E.M. (1990) Comparative analysis of the proliferative response
of the rat urinary bladder to sodium saccharin by light and scanning
electron microscopy and autoradiography. Scanning Microsc., 4, 135–142.
24. Cohen,S.M., Cano,M., St.John,M.K., Garland,E.M., Khachab,M. and
Ellwein, L.B. (1995) Effect of sodium saccharin in the neonatal rat bladder.
Scanning Microsc., 9, 137–148.
25. Garland,E.M., Shapiro,R., Wehner,J.M., Johnson,L.S., Mattson,B.J.,
Khachab,M., Asamoto,M. and Cohen,S.M. (1993) The effects of dietary
iron and folate supplementation on the physiological changes produced in
weaning rats by sodium saccharin exposure. Fd Chem. Toxicol., 31,
689–699.
26. Arnold,D.L. and Boyes,B.G. (1989) The toxicological effects of saccharin
in short-term genotoxicity assays. Mutat. Res., 221, 69–132.
27. Moore,C.W. and Schmick,A. (1979) Recombinogenicity and mutagenicity
of saccharin in Saccharomyces cerevisiae. Mutat. Res., 67, 215–219.
28. Ashby,J. and Ishidate,M.Jr. (1986) Clastogenicity in vitro of the Na, K,
Ca and Mg salts of saccharin; and of magnesium chloride; consideration
of significance. Mutat. Res., 163, 63–73.
29. Williamson,D.S., Nagel,D.L., Markin,R.S. and Cohen,S.M. (1987) Effect
of pH and ions on the chemical structure of saccharin. Fd Chem. Toxicol.,
25, 211–218.
30. Renwick,A.G. (1989) Saccharin: a toxicological evaluation. Toxicology, 3,
289–305.
31. Lutz,W.K. and Schlatter,C. (1977) Saccharin does not bind to DNA of
liver or bladder in the rat. Chem. Biol. Interact., 19, 253–257.
32. Zukowski,K., Debiec-Rychter,M., VanGoethem,D.L., Sangha,G.K.,
Thyssen,J.H. and Wang,C.Y. (1994) A UDS system for differentiating
Urinary precipitate and bladder cancer
genotoxic and nongenotoxic bladder carcinogens in the rat. Proc. Am.
Assoc. Cancer Res., 35, 684.
33. Hasegawa,R. and Cohen,S.M. (1986) The effect of different salts of
saccharin on the rat urinary bladder. Cancer Lett., 30, 261–268.
34. Ellwein,L.B. and Cohen,S.M. (1990) The health risks of saccharin revisited.
Crit. Rev. Toxicol., 20, 311–326.
35. Saccharin Working Group of the Toxicology Forum, Given Institute, Aspen,
Colorado, July 27, 1980, 67–69, Toxicology Forum, Washington, DC.
36. West,R.W. and Jackson,C.D. (1981) Saccharin effects on the urinary
physiology and urothelium of the rat when administered in diet or drinking
water. Toxicol. Lett., 7, 409–416.
37. Cohen,S.M., Cano,M., Garland,E.M., Earl,R.A. and Carson,S. (1991) A
proposed role for silicates and protein in the proliferative effects of
saccharin on the male rat urothelium. Carcinogenesis, 12, 1551–1555.
38. Cohen,S.M., Cano,M., Garland,E.M., St.John,M. and Arnold,L.L. (1995)
Urinary and urothelial effects of sodium salts in male rats. Carcinogenesis,
16, 343–348.
39. Fukushima,S., Imaida,K., Sakata,T., Okamura,T., Shibata,M. and Ito,N.
(1983) Promoting effects of sodium L-ascorbate on two-stage urinary
bladder carcinogenesis in rats. Cancer Res., 43, 4454–4457.
40. DeGroot,A.P., Feron,V.J. and Immel,H.R. (1988) Induction of hyperplasia
in the bladder epithelium of rats by a dietary excess of acid or base:
Implications for toxicity/carcinogenicity testing. Fd Chem. Toxicol., 26,
425–434.
41. Ito,N. and Fukushima,S. (1989) Promotion of urinary bladder
carcinogenesis in experimental animals. Exp. Pathol., 36, 1–15.
42. Ito,N., Fukushima,S., Shirai,T. and Nakanishi,K. (1983) Effects of
promoters on N-butyl-N-(4-hydroxybutyl)nitrosamine-induced urinary
bladder carcinogenesis in the rat. Environ. Hlth Perspect., 50, 61–69.
43. Lina,B.A.R., Hollanders,V.M.H. and Kuijpers,M.H.M. (1994) The role of
alkalizing and neutral potassium salts in urinary bladder carcinogenesis in
rats. Carcinogenesis, 15, 523–527.
44. Okamura,T., Garland,E.M., Johnson,L.S., Cano,M., Johansson,S.L. and
Cohen, S.M. (1992) Acute urinary tract toxicity of tetraethylorthosilicate
(TES) in rats. Fund. Appl. Toxicol., 18, 425–441.
45. Emerick,R.J. (1986) Animal age and duration of exposure to a urolithic
diet as factors influencing silica urolithiasis incidence in an animal model.
Nutr. Rep. Int., 34, 907–913.
46. Emerick,R.J. and Lu,D. (1987) A possible synergism of dietary phosphate
and urine-acidifying salts in preventing silica urolithiasis in a rat model.
J. Nutr., 117, 1603–1608.
47. Tibbels,T.S., Smith,R.A. and Cohen,S.M. (1988) Determination of
saccharin in diet and biological materials. J. Chromatogr., 441, 448–453.
48. Cohen,S.M. Anderson,T.A., deOliveira,L.M. and Arnold,L.L. (1998)
Tumorigenicity of sodium ascorbate in male rats. Cancer Res., 58,
2557–2561.
49. Fisher,M.J., Sakata,T., Tibbels,T.S., Smith,R.A., Patil,K., Khachab,M.,
Johansson, S.L. and Cohen,S.M. (1989) Effect of sodium saccharin and
calcium saccharin on urinary parameters in rats fed Prolab 3200 or AIN76 diet. Fd Chem. Toxicol., 27, 1–9.
50. Cohen,S.M., Cano,M., Anderson,T. and Garland,E.M. (1996) Extensive
handling of rats leads to mild urinary bladder hyperplasia. Toxicol. Pathol.,
24, 251–257.
51. Moorehead,W.R. and Biggs,H.G. (1974) 2-Amino-2-methyl-1-propanol as
the alkalizing agent in an improved continuous-flow cresolphthalein
complexone procedure for calcium in serum. Clin. Chem., 20, 1458–1460.
52. Fiske,C.H. and Subbarao,Y. (1925) The colorimetric determination of
phosphorus. J. Biol. Chem., 66, 375.
53. Ingman,F. and Ringbom,A. (1966) Spectrophotometric determination of
small amounts of magnesium and calcium employing calmagite.
Microchem. J., 10, 545.
54. Farndale,R.W., Buttle,D.J. and Barrett,A.J. (1986) Improved quantitation
and discrimination of sulphated glycosaminoglycans by use of
dimethylmethylene blue. Biochem. Biophys. Acta, 883, 173–177.
55. Gassler,N., Reibner,C., Janzen,N., Kahnert,H. and Kleesiek,K. (1993) A
high performance liquid chromatography method for the determination of
glycosaminoglycans in human blood. Eur. J. Clin. Chem. Clin. Biochem.,
31, 503–511.
56. Emerick,R.J., Kugel,E.E. and Wallace,V. (1963) Urinary excretion of silica
and the production of siliceous urinary calculi in rats. Am. J. Vet. Res.,
24, 610–613.
57. Sundberg,M.W. and Dappen,G.M. (1979) An assay for total serum calcium
using a coated film. Clin. Chem., 25, 1140.
58. Smith-Lewis,M.J., Babb,B.E., Hilborn,D.A., Mauck,J.C., Norkus,N.S.,
Toner,J.L. and Weaver,M.S. (1986) Thin-film colorimetric assay for
magnesium. Clin. Chem., 32, 1200.
59. Tietz,N.W., Pruden,E.L. and Siggaard-Andersen,O. (1986) Electrolytes,
blood gases and acid-base balance. In Tietz,N.W. (ed.) Textbook of Clinical
Chemistry, W.B. Saunders Co., Philadelphia, PA, pp. 1179–1180.
60. Tietz,N.W., Pruden,E.L. and Siggard-Andersen,O. (1986) Electrolytes,
blood gases and acid-base balance. In Tietz,N.W. (ed.) Textbook of Clinical
Chemistry, W.B. Saunders Co., Philadelphia, PA, pp. 1175–1178.
61. Siggaard-Andersen,O. (1976) Electrochemistry. In Tietz,N.W. (ed.)
Fundamentals of Clinical Chemistry, W.B. Saunders Co., Philadelphia,
PA, pp. 135–137.
62. Mauck,J.C., Mauck,K.L., Novros,J. and Norton,G.E. (1986) Development
of a single Kodak Ectachem® thin-film assay for serum and urine creatinine.
Clin. Chem., 32, 1197–1198.
63. Tatematsu,M., Fukushima,S., Aoki,T., Mera,Y., Inoue,T. and Ito,N. (1987)
Patterns of epithelial proliferation revealed by continuous administration
of bromodeoxyuridine during urinary bladder carcinogenesis in rats. Jpn.
J. Cancer Res., 78, 879–882.
64. Dietrich,D.R. and Swenberg,J.A. (1991) The presence of α2u-globulin is
necessary for d-limonene promotion of male rat kidney tumors. Cancer
Res., 51, 3512–3521.
65. Cohen,S.M., Yang,J.P.S., Jacobs,J.B., Arai,M., Fukushima,S. and
Friedell,G.H. (1981) Transplantation and cell culture of rat urinary bladder
carcinoma. Invest. Urol., 19, 136–141.
66. Luna,L.G. (1968) Manual of Histologic Staining Methods of the Armed
Forces Institute of Pathology. McGraw-Hill Book Co., New York.
67. Cohen,S.M., Garland,E.M., Cano,M., St.John,M., Khachab,M. and
Arnold,L.L. (1995) Effects of sodium ascorbate, sodium saccharin and
ammonium chloride on the male rat urinary bladder. Carcinogenesis, 16,
2743–2750.
68. Miller,T.D. and Cohen,S.M. (1983) Alterations in the rat kidney associated
with sodium saccharin feeding: A scanning electron microscopic study.
Toxicol. Lett., 20, 177–181.
69. Johansson,S.L., Sakata,T., Hasegawa,R., Zenser,T.V., Davis,B.B. and
Cohen,S.M. (1986) The effect of long-term administration of aspirin and
sodium saccharin on the rat kidney. Toxicol. Appl. Pharmacol., 86, 80–92.
70. Garland,E.M. Sakata,T., Fisher,M.J., Masui,T. and Cohen,S.M. (1989)
Influences of diet and strain on the proliferative effect on the rat urinary
bladder induced by sodium saccharin. Cancer Res., 49, 3789–3794.
71. Fukushima,S., Hagiwara,A., Ogiso,T., Shibata,M. and Ito,N. (1983)
Promoting effects of various chemicals in rat urinary bladder carcinogenesis
initiated by N-nitroso-N-butyl-(4-hydroxybutyl)amine. Fd Chem. Toxicol.,
21, 59–68.
72. Wigler,M., Pellicer,S., Silverstein,S., Axel,R. and Urlaub,G. (1979) DNAmediated transfer of the adenine phosphoribosyl transferase locus into
mammalian cells. Proc. Natl Acad. Sci., 76, 1373–1376.
73. Jones,W.T. and Resnick,M.I. (1990) The characterization of soluble matrix
proteins in selected human renal calculi using two-dimensional
polyacrylamide gel electrophoresis. J. Urol., 144, 1010–1014.
74. Olson,M.J., Johnson,J.T. and Reidy,C.A. (1990) A comparison of male rat
and human urinary proteins: Implications for human resistance to hyaline
droplet nephropathy. Toxicol. Appl. Pharmacol., 102, 524–536.
75. Garland,E.M., St.John,M., Asamoto,M., Eklund,S.H., Mattson,B.J.,
Johnson,L.S., Cano,M. and Cohen,S.M. (1994) A comparison of the effects
of sodium saccharin in NBR rats and in intact and castrated male F344
rats. Cancer Lett., 78, 99–107.
76. Uwagawa,S., Saito,K., Okuno,Y., Kawasaki,H., Yoshitake,A., Yamada,H.
and Fukushima,S. (1994) Lack of induction of epithelial cell proliferation
by sodium saccharin and sodium L-ascorbate in the urinary bladder of
NCI-Black-Reiter (NBR) male rats. Toxicol. Appl. Pharmacol., 127,
182–186.
77. Cohen,S.M. (1995) The role of urinary physiology and chemistry in bladder
carcinogenesis. Fd Chem. Toxicol., 33, 715–730.
78. Anderson,R.L. (1988) A hypothesis of the mechanism of urinary bladder
tumorigenesis in rats ingesting sodium saccharin. Fd Chem. Toxciol., 26,
637–644.
79. Michaud,D.S., Spiegelman,D., Clinton,S.K., Rimm,E. B., Curhan,G.C.,
Willett,W. C. and Giovannucci,E.L. (1999) Fluid intake and the risk of
bladder cancer in men. N. Engl. J. Med., 340, 1390–1397.
80. Shioya,S., Nagami-Oguihara,R., Oguihara,S., Kimura,T., Imaida,K. and
Fukushima,S. (1994) Roles of bladder distension, urinary pH and urinary
sodium ion concentration in cell proliferation of urinary bladder epithelium
in rats ingesting sodium salts. Fd Chem. Toxicol., 32, 165–171.
81. Cohen,S.M., Cano,M., Johnson,L.S., St.John, M.K., Asamoto,M.,
Garland,E.M., Thyssen,J.H., Sangha,G.K. and VanGoethem,D.L. (1994)
Mitogenic effects of propoxur on male rat bladder urothelium.
Carcinogenesis, 15, 2593–2597.
82. Garland,E.M., Shapiro,R., Kraft,P.L., Khachab,M., Patil,K., Ellwein, L.B.
and Cohen, S.M. (1991) Effects of in utero and postnatal sodium saccharin
791
S.M.Cohen et al.
exposure on the nutritional status of the young rat. I. Effects at 30 days
postbirth. Fd Chem. Toxicol., 29, 657–667.
83. Arnold,L.L., Anderson,T., Cano,M., St.John,M., Mattson,B., Wehner,J. and
Cohen, S.M. (1995) A comparison of urinary chemistry changes in male
and female rats and mice treated with sodium saccharin. The Toxicologist,
15, 201.
84. Hard,G.C. (1995) Species comparison of the content and composition of
urinary proteins. Fd Chem. Toxicol., 33, 731–746.
85. Eklund,S.H., Garland,E.M. and Cohen,S.M. (1992) Characterization of
proteins of male rat urine which bind saccharin in vitro. Proc. Am. Assoc.
Cancer Res., 33, 189.
792
86. Frederick,C.B., Dooley,K.L., Kodell,R.L., Sheldon,W.G. and Kadlubar,F.F.
(1989) The effect of lifetime sodium saccharin dosing on mice initiated
with the carcinogen 2-acetylaminofluorene. Fund. Appl. Toxicol., 12,
346–357.
87. Hard,G.C., Rodgers,I.S., Baetcke,K.P., Richards,W.L., McGaughy,R.E.
and Valcovic,L.R. (1993) Hazard evaluation of chemicals that cause
accumulation of α2u-globulin, hyaline droplet nephropathy and tubule
neoplasia in the kidneys of male rats. Environ. Hlth Perspect., 99, 313–349.
Received September 22, 1999: revised November 17, 1999;
accepted November 29, 1999