Glycemia-lowering effect of cobalt chloride in the
diabetic rat: role of decreased gluconeogenesis
FIRAS SAKER,1 JUAN YBARRA,2 PATRICK LEAHY,3 RICHARD W. HANSON,3
SATISH C. KALHAN,1 AND FARAMARZ ISMAIL-BEIGI2
Departments of 1Pediatrics, 2Medicine, and 3Biochemistry,
Case Western Reserve University, Cleveland, Ohio 44106-4951
glucose uptake; phosphoenolpyruvate carboxykinase mRNA;
adenosine 38,58-cyclic monophosphate
CELLULAR UPTAKE and metabolism of glucose are of
critical importance in the maintenance of homeostasis
and energy production. The first step in glucose metabolism is its transport across the plasma membranes, a
function that is ‘‘rate limiting’’ in many cells and tissues
(5, 12). Transmembrane transport of glucose is mediated by a family of Na1-independent glycoprotein glucose transporter (GLUT) molecules that are expressed
in a tissue-specific manner (20, 23). Of these, GLUT-4
and GLUT-1 are especially relevant to diabetes because
they mediate insulin-stimulated and basal (insulinindependent) uptake of glucose, respectively (20, 23).
Results of our previous studies in Clone 9 cells, C2C12
myoblasts, and 3T3-L1 preadipocytes demonstrated
that inhibition of oxidative phosphorylation by hypoxia
and by inhibitors of oxidative phosphorylation such as
cyanide and azide causes a stimulation of glucose
transport and an induction of GLUT-1 gene expression
(2, 19, 25, 26). We have also shown in the above cells
E984
that, similar to other ‘‘hypoxia-responsive genes’’ (8),
GLUT-1 gene expression is augmented by cobalt chloride (CoCl2 ) employed under normoxic conditions (2).
On the basis of these findings, we recently explored the
possibility that exposure of diabetic and normal rats to
CoCl2 will similarly enhance tissue GLUT-1 expression
and tissue glucose uptake and thereby lead to a lowering of blood glucose (31). The results showed that
treatment of diabetic rats [induced by streptozotocin
(STZ)] with 2 mM CoCl2 added to the drinking water
resulted in a lowering of nonfasting serum glucose
levels from 35 6 2 to 21 6 2 mM; a smaller decrement in
glycemia of normal rats treated with CoCl2 was not
significant. Treatment with CoCl2 was associated with
a 1.3- to 2.9-fold increase in GLUT-1 mRNA content of
ventricular myocardium, renal cortex, skeletal muscle,
cerebrum, and liver of both diabetic and normal rats.
Although GLUT-1 and glucose transport were not
measured in that study (31), the results are consistent
with the possibility that the glycemia-lowering effect of
CoCl2 may be mediated by enhanced expression of
GLUT-1 mRNA, GLUT-1 protein, and glucose uptake.
However, treatment with CoCl2 could also have reduced
hepatic glucose output and thus lowered blood glucose.
The present study was therefore conducted to determine the potential role of enhanced tissue glucose
uptake vs. decreased systemic glucose production in
mediating the glycemia-lowering effect of CoCl2 in
diabetic rats. The rate of systemic appearance (Ra ) of
glucose was measured under fasting conditions by
tracer dilution in diabetic and normal rats treated or
not treated with CoCl2. We also determined the effect of
CoCl2 on the relative abundance of hepatic phosphoenolpyruvate carboxykinase (PEPCK) mRNA and on the
concentration of cAMP in the liver as a measure of
changes in hepatic gluconeogenesis. The results indicate that the dominant effect of CoCl2 in reducing the
glycemia of diabetic rats results from a reduction in the
Ra of glucose and decreased hepatic gluconeogenesis.
MATERIALS AND METHODS
Materials. CoCl2 and other standard chemicals were obtained from Sigma Chemical (St. Louis, MO). Reagent-grade
ether was obtained from Fisher (Pittsburgh, PA). The cAMP
kit was obtained from Amersham (Arlington Heights, IL).
[3-3H]glucose (10.4 Ci/mmol), 2-deoxy-[U-14C]glucose (294
Ci/mmol), and deoxy-[a-32P]cytidine (3,000 Ci/mmol) were
obtained from DuPont-NEN Research Products (Boston, MA).
PE-50 polyethylene nontoxic tubing (ID, 0.58 mm; OD, 0.965
mm) was obtained from Becton-Dickinson (Sparks, MD). The
rodent sling jacket was obtained from Harvard Bioscience
(South Natick, MA). QuickPrep total RNA extraction kit and
Quikhyb were purchased from Pharmacia Biotech (Piscat-
0193-1849/98 $5.00 Copyright r 1998 the American Physiological Society
Downloaded from http://ajpendo.physiology.org/ by 10.220.32.247 on June 18, 2017
Saker, Firas, Juan Ybarra, Patrick Leahy, Richard W.
Hanson, Satish C. Kalhan, and Faramarz Ismail-Beigi.
Glycemia-lowering effect of cobalt chloride in the diabetic rat:
role of decreased gluconeogenesis. Am. J. Physiol. 274 (Endocrinol. Metab. 37): E984–E991, 1998.—Results of previous
studies indicated that treatment of diabetic rats (induced by
streptozotocin) with cobalt chloride (CoCl2 ) resulted in a
significant decrement in serum glucose concentration. The
present study was designed to determine the potential role of
enhanced glucose uptake vs. decreased glucose production in
the above response. The rate of systemic appearance of
glucose, measured under fasting conditions using [3-3H]glucose tracer, was reduced from 35.5 6 2.5 to 17.5 6 1.8
µmol · kg21 · min21 in diabetic rats treated with 2 mM CoCl2
added to the drinking water for 10–14 days (P , 0.01). Tissue
accumulation of intravenously administered 2-deoxy-[14C]glucose was significantly reduced in kidney and eye of diabetic
rats treated with CoCl2, whereas the uptake remained unchanged in several other tissues including cerebrum, red and
white skeletal muscle, heart, and liver. The relative content of
phosphoenolpyruvate carboxykinase (PEPCK) mRNA was
increased 3.1-fold in livers of diabetic compared with normal
rats (P , 0.001), and treatment of diabetic rats with CoCl2
decreased hepatic PEPCK mRNA levels to normal. The
content of PEPCK mRNA in the liver was decreased by 33% in
CoCl2-treated normal rats (P , 0.05). Treatment with CoCl2
resulted in no change in cAMP levels in the livers of either
diabetic or normal rats. These results suggest that the
glycemia-lowering effect of CoCl2 is mediated by reductions in
the rate of systemic appearance of glucose and hepatic
gluconeogenesis.
COCL2 INHIBITION OF SYSTEMIC GLUCOSE PRODUCTION
Table 1. Effect of treatment with CoCl2 on serum
concentration of certain constituents in diabetic
and normal rats
Diabetic
(n 5 6)
Glucose, mM
Na1 , meq/l
HCO2
3 , meq/l
Ca21, mM
Mg21, mM
Phosphate,
mg/dl
Creatinine,
mg/dl
Albumin, g/dl
Cholesterol,
mg/dl
Triglyceride,
mg/dl
Insulin, ng/ml
Glucagon, pg/ml
Diabetic 1
CoCl2
(n 5 6)
Control
(n 5 4)
Control 1
CoCl2
(n 5 4)
35.4 6 2.8 21.2 6 2.2* 8.6 6 0.3
8.0 6 0.3
123 6 1
127 6 4
128 6 1
131 6 1
27 6 1
23 6 1*
22 6 1
22 6 1
2.48 6 0.03 2.38 6 0.05 2.50 6 0.05 2.50 6 0.05
1.63 6 0.04 1.56 6 0.04 1.90 6 0.06 1.84 6 0.04
8.7 6 0.2
7.9 6 0.3*
9.9 6 0.4
9.7 6 0.2
0.30 6 0.04 0.37 6 0.03 0.28 6 0.03 0.28 6 0.03
3.4 6 0.1
3.8 6 0.2
4.4 6 0.1
4.4 6 0.1
108 6 30
61 6 6
57 6 4
53 6 5
760 6 406
0.4 6 0.1
236 6 20
117 6 31
0.3 6 0.1
300 6 30
88 6 17
2.5 6 0.4
248 6 25
59 6 12
2.8 6 0.4
ND
Results are expressed as means 6 SE; n 5 no. of rats. Nonfasting
diabetic rats were killed at 9:00 AM, and blood was obtained for
analysis. Rats in both control (nondiabetic) groups were fasted for 24
h before death. Measurement of serum glucose was performed,
however, on 200 µl of blood obtained from tail vein 2 days before
death. Nonfasting insulin and glucagon levels are from a previous
study using different but similar groups of rats (31). ND, not done.
* P , 0.05 compared with respective CoCl2-untreated group.
ment with CoCl2 were as indicated above. Indwelling arterial
and venous catheters were placed 3–4 days before the tracer
study. The animals were anesthetized with an intramuscular
injection (0.1 ml/100 g body wt) of ketamine-acepromazine
mixture (90 mg ketamine and 1 mg acepromazine per ml).
PE-50 catheters were inserted in the right external jugular
and left carotid artery under sterile conditions. The catheters
were filled with an anticoagulant mixture of polyvinylpyrrolidone and heparin in isotonic saline (polyvinylpyrrolidone,
0.75 g; heparin, 25 units in 1 ml of isotonic saline). This
polymer allows catheter patency for a week or more. The
catheters were tunneled subcutaneously, and the distal end of
the catheter was sutured to the dorsum of the rat’s neck. The
free end of the catheter was sealed, and the rat was placed in
a Harvard rodent jacket sling, thus allowing the animal to
move freely during the postoperative period. After surgery,
rats were housed in individual cages and had free access to a
standard diet (Purina rat chow) and water or CoCl2 solution
as indicated.
Twenty-four to thirty hours before the infusion, food was
removed from the animal cages. [3-3H]glucose in isotonic
saline was given as a prime constant-rate infusion using a
Harvard pump, starting at 11:00 AM. The priming dose was
1.5–2.0 µCi, and tracer was infused at the rate of 0.05–0.07
µCi · kg body wt21 · min21. Four arterial samples (two 0.5- and
two 0.2-ml samples) were obtained at 10-min intervals between 80 and 120 min. Two hours after the [3-3H]glucose
infusion, a bolus dose (20 µCi) of 2-deoxy-[U-14C]glucose was
administered. Heparinized blood samples of 0.2 ml were
obtained at 1, 3, 5, 7, 10, 15, 20, 30, and 50 min. Total volume
of blood drawn from each animal was ,2.8 ml and was
replaced with fresh blood from a donor rat. At the end of the
study, the animals were anesthetized using intravenous
pentobarbital sodium, and tissues including cerebrum, heart
(ventricle), kidney (cortex), skeletal muscle (red and white
gastrocnemius), and eye orbits were rapidly obtained, rinsed
in isotonic cold saline to wash off blood, blotted, and frozen in
liquid nitrogen.
For measurement of specific activity of glucose, plasma
samples were deproteinated with 0.3 M barium hydroxide
and 0.3 M zinc sulfate. The neutralized supernatant was
evaporated to dryness to remove any radioactive water
formed during glycolysis. The dry samples were reconstituted
with water and passed through a mixed-bed ion-exchange
column (16). Glucose was eluted using 5 ml of water, evaporated to dryness, and used to measure glucose concentration
(Beckman glucose analyzer) as well as radioactivity (Packard
Instruments, Downers Grove, IL). The amount of 2-deoxy[14C]glucose 6-phosphate in various tissues was measured as
described by Ferré et al. (6). A weighed amount of tissue was
homogenized in 1.0 ml of 1.0 M sodium hydroxide and heated
at 60°C for 2 h or until total digestion. The hydrolyzed
homogenate was neutralized with 1 ml of 1.0 M HCl. The
accumulation of 2-deoxy-[14C]glucose 6-phosphate was measured from the difference between total radioactivity (2-deoxy[14C]glucose 1 2-deoxy-[14C]glucose 6-phosphate) and that
due to 2-deoxy-[14C]glucose (6).
The Ra of glucose was calculated during isotopic steady
state using the tracer dilution equation Ra (µmol/min) 5 I/(sp
act), where I is the rate of infusion of tracer isotope (dpm/min)
and sp act is the specific activity of plasma glucose (dpm/
µmol). Tissue uptake of glucose (nmol · g tissue21 · min21 ) was
calculated as the ratio of tissue 2-deoxy-[14C]glucose
6-phosphate content to the integral of the specific activity of
2-deoxy-[14C]glucose in the plasma from 0 to 50 min, where
the numerator represents the 2-deoxyglucose 6-phosphate
radioactivity present in each tissue sample (dpm/g tissue) (6).
Downloaded from http://ajpendo.physiology.org/ by 10.220.32.247 on June 18, 2017
away, NJ) and Stratagene (La Jolla, CA), respectively. GeneScreen Plus hybridization membrane was obtained from
DuPont-NEN Life Sciences. The cDNA probe for the cytosolic
form of PEPCK was a 1.1-kb Pst I/Pst I fragment from the 38
end of the rat PEPCK cDNA, pPCK 10 (32). Kits for measurement of cAMP were obtained from Amersham.
Animals. Male Sprague-Dawley rats weighing 225–250 g
were obtained from Zivic Miller (Zelienople, PA). The animals
were kept in a controlled environment that meets the requirements of the American Association for the Accreditation of
Laboratory Animals. They had free access to rat chow (Purina), and water was administered on a 12:12-h light-dark
cycle. The protocol was approved by the Institutional Animal
Care and Use Committee.
Diabetes was induced by injection of a freshly prepared
solution of STZ in saline at 60 mg/kg body wt in the tail vein
(31). After 1 wk, a sample of blood was obtained from the tail
vein to ensure the presence of diabetes (serum glucose .25
mM). One-half of the group of normal rats and one-half of the
diabetic group were then placed on 2 mM CoCl2 in the
drinking water for 12–16 days. During the ,2-wk period,
normal and diabetic rats gained ,30 and ,15 g of weight,
respectively, and treatment with CoCl2 resulted in no change
in weight gained by diabetic rats. In a previous study
employing various concentrations of CoCl2, we found that a
similar group of diabetic rats treated with up to 4 mM CoCl2
for 7 wk gained ,100 g in body weight, a rate that was
identical to that in diabetic rats not treated with CoCl2 (31).
On the basis of the daily water intake, we estimate that
diabetic and normal rats received a total dose of ,2 and ,1
mmol of CoCl2, respectively. Values for serum glucose, electrolytes, and other constituents summarized in Table 1 were
measured in the hospital laboratory.
Measurement of Ra of glucose and of glucose uptake by
tissues. These experiments were performed in different sets of
animals from the ones described above; diabetes and treat-
E985
E986
COCL2 INHIBITION OF SYSTEMIC GLUCOSE PRODUCTION
RESULTS
Effect of CoCl2 on the concentration of glucose and
selected other constituents in the blood of diabetic and
normal rats. Normal and diabetic rats gained ,30 and
,15 g, respectively, during the ,2 wk of study, and
treatment with CoCl2 resulted in no change in the rate
of weight gain by diabetic rats. Table 1 summarizes the
effect of 10 days of treatment with CoCl2 on the
concentration of several serum constituents in the four
groups of rats. Rats in both diabetic groups were not
fasted before obtaining blood samples. In both groups of
normal rats, with the exception of serum glucose and
insulin, values are from rats after a 24-h period of food
deprivation. Blood glucose was determined under nonfasting conditions, because it was found in preliminary
studies that blood glucose concentration in diabetic
rats decreased to normal levels after a 24- to 30-h
period of food deprivation. In accordance with our
previous observations (31), treatment with CoCl2 resulted in a dramatic reduction of nonfasting glycemia of
diabetic rats from 35.4 6 2.8 to 21.2 6 2.2 mM glucose
(P , 0.01), whereas the agent resulted in no significant
change in the serum glucose concentration of normal
rats. In diabetic rats treated with CoCl2, there was a
significant decrease in the concentration of bicarbonate
and phosphate compared with diabetic rats not treated
with the agent; the decrements in serum cholesterol
and triglyceride in CoCl2-treated diabetic rats were not
significant, although two diabetic rats that had not
been treated with CoCl2 had lipemic serum. Treatment
of normal rats with CoCl2 resulted in no measurable
change in the concentration of any of the serum constituents. The concentrations of insulin and glucagon in the
serum of diabetic rats treated or not treated with CoCl2
[both reported previously (31)] have also been included
in Table 1. After 10 days of treatment with CoCl2,
hematocrit values in both normal and diabetic rats rose
equally from 37 6 1 to 40 6 1%.
Effect of CoCl2 on the Ra of glucose in diabetic and
normal rats. The Ra was determined in all four experimental groups after 24–30 h of food deprivation; the
fasting period resulted in a normalization of serum
glucose concentration in both groups of diabetic rats.
Serum glucose concentrations during measurement of
Ra were 6.37 6 0.22 and 6.04 6 0.42 mM in diabetic and
CoCl2-treated diabetic rats, respectively. Ra, expressed
either as micromoles per minute (data not shown) or as
micromoles per minute per kilogram body weight, was
,50% lower in diabetic animals treated with CoCl2
(35.5 6 2.5 vs. 17.5 6 1.8 µmol · kg21 · min21; P , 0.01;
Fig. 1). Ra also was lower by ,35% in normal rats
treated with CoCl2 (from 47.5 6 8.5 to 33.0 6 3.0
µmol · kg21 · min21 ), but the change was not significant
(Fig. 1). Comparison of glucose Ra values in diabetic vs.
normal rats [both groups not treated with CoCl2]
reveals no significant difference between the two groups.
Because the concentration of glucose in the blood was
constant during the 3-h period of measurement of Ra,
the experimental results can be used to calculate the
clearance of glucose from the circulation. Glucose clearance was 0.89 6 0.22 and 0.59 6 0.07 ml · 100 g body
wt21 · min21 in normal and CoCl2-treated normal rats,
respectively (P . 0.1). It was 0.53 6 0.04 and 0.26 6
0.04 ml · 100 g body wt21 · min21 in diabetic and CoCl2treated diabetic rats, respectively (P , 0.01).
Effect of CoCl2 on the rate of glucose uptake by selected
tissues in diabetic and normal rats. After determination of Ra, glucose uptake by several tissues of the same
Downloaded from http://ajpendo.physiology.org/ by 10.220.32.247 on June 18, 2017
The integral of the specific activity of 2-deoxy-[14C]glucose in
the plasma was calculated by establishing the best-fit exponential curve for the plasma specific activity measurements. No
correction was made for lump constant, the correction factor
for the discrimination against 2-deoxy-[14C]glucose in glucose
transport and phosphorylation pathways (6). It was assumed
that the lump constant will be the same in control and
experimental animals.
Measurement of cAMP levels in the liver. Groups of CoCl2treated and untreated normal and diabetic rats separate from
those used above were employed. Rats were decapitated after
CO2-inhalation anesthesia. Livers were removed promptly
and frozen immediately using liquid nitrogen or dry ice.
cAMP determinations are based on a competition enzyme
immunoassay as described by the manufacturer (Amersham).
Approximately 100 mg of frozen liver were homogenized in 10
volumes of ice-cold 6% (wt/vol) trichloroacetic acid by means
of a Teflon-glass homogenizer. After centrifugation at 2,000 g
for 15 min at 4°C, a 250-µl aliquot of the supernatant was
extracted five times with 1 ml of water-saturated diethyl
ether. Two-hundred microliters of the aqueous phase were
lyophilized and resuspended in 1 ml of the assay buffer
supplied with the kit. cAMP determinations were carried out
using 100-µl aliquots of samples, with standards ranging
from 12.5 to 3,200 fmol/well.
Isolation of RNA and Northern blot analysis. Total RNA
was extracted from livers of experimental animals using the
QuickPrep total RNA extraction kit. Seventy-five to onehundred milligrams of frozen liver (obtained as described
above) were homogenized and treated as per the manufacturer’s instructions. RNA samples (10 µg) were fractionated on
formaldehyde-agarose gels and transferred to GeneScreen
Plus membrane by capillary action. The membrane was
hybridized with a randomly primed 32P-labeled rat PEPCK
cDNA probe (32) and washed, and the radioactivity in each
band was measured by video densitometry using a Molecular
Dynamics phosphorimager. To ensure equal RNA loading of
the lanes and to control for completeness of RNA transfer,
ethidium bromide staining of ribosomal 28S and 18S bands
on the gels and on the nitrocellulose paper was monitored
throughout.
To compare the relative abundance of hepatic PEPCK
mRNA among different experimental groups, the average
densitometric measurement of PEPCK mRNA in normal rats
not treated with CoCl2 was calculated and set to 1.0. Values
obtained for all samples were then normalized against the
mean value obtained for the control group.
Analysis of data and statistical methods. Data on all four
experimental groups, i.e., diabetic rats, diabetic rats treated
with CoCl2, normal rats, and normal rats treated with CoCl2,
are presented throughout. However, analysis of the potential
effects of CoCl2 on any given parameter was performed in
diabetic rats and normal rats treated (vs. not treated) with
the agent. This was done because diabetes itself is associated
with changes in some of the parameters being examined.
All experimental results are expressed as means 6 SE.
Unpaired Student’s t-test was employed, and P , 0.05 was
considered significant (27).
COCL2 INHIBITION OF SYSTEMIC GLUCOSE PRODUCTION
E987
rats was measured by previously described methods
(6). Figure 2 summarizes the amount of 2-deoxy[14C]glucose 6-phosphate accumulated in tissues of
diabetic rats not treated or treated with CoCl2. It
should be noted that accumulation of glucose in tissues
is a composite function of transport and phosphorylation steps acting sequentially. Among the tissues examined in the diabetic rat, heart and cerebrum exhibited
high rates of uptake (expressed as nmol · g tissue
wt21 · min21 ). Because only a small fraction of the eye
represents the metabolically highly active cells of the
retina, this tissue probably manifests the highest rate
of glucose uptake and metabolism. The kidney had an
intermediate rate of uptake, whereas the red and white
gastrocnemius muscle exhibited lower rates. Treatment of diabetic rats with CoCl2 tended to increase the
rate of uptake of 2-deoxy-[14C]glucose by heart and
liver, whereas the uptake by cerebrum, kidney, muscle,
Fig. 2. Effect of CoCl2 on 2-deoxy-[14C]glucose uptake in several
tissues of diabetic rats. Glucose uptake was estimated by accumulation of 2-deoxy-[14C]glucose 6-phosphate in various tissues. Experimental animals are same as those in Fig. 1. Results are expressed as
means 6 SE; n 5 5 rats/group. Ventr, ventricle; Gast R, red
gastrocnemius; Gast W, white gastrocnemius. * Significant decreases
in uptake in kidney and eye (P , 0.05 compared with respective
CoCl2-untreated group).
and eye was decreased; of the aforementioned changes,
only the decrements in uptake by the kidney and eye
were significant.
Results of measurement of 2-deoxy-[14C]glucose uptake by the indicated tissues in normal rats not treated
or treated with CoCl2 are summarized in Fig. 3. Compared with diabetic rats, rates of glucose uptake in
normal rats are somewhat higher in cerebrum, liver,
and muscle and lower in heart, kidney, and eye; none of
the changes are significant. Treatment of normal rats
with CoCl2 resulted in a slight to moderate decrease in
the rate of glucose uptake by all the tissues examined,
although none of the changes reached significance. It is
probable, however, that the sum of uptake values by all
the tissues combined is reduced as a result of treatment
with CoCl2.
Effect of CoCl2 on the content of PEPCK mRNA in the
livers of diabetic and normal rats. We next examined
the possibility that the activity of PEPCK, the key
enzyme in the control of gluconeogenesis, might be
suppressed as a result of exposure to CoCl2. Because
PEPCK activity closely parallels PEPCK mRNA con-
Fig. 3. Effect of CoCl2 on 2-deoxy-[14C]glucose uptake in several
tissues of normal rats. Glucose uptake was measured as described in
legend to Fig. 2. Experimental animals are same as those in Fig. 1.
Downloaded from http://ajpendo.physiology.org/ by 10.220.32.247 on June 18, 2017
Fig. 1. Effect of cobalt chloride (CoCl2 ) on rate of systemic appearance (Ra ) of glucose in diabetic and normal rats.
Jugular vein and carotid artery indwelling catheters were placed 72–96 h before study. Rats were deprived of food
for 24–30 h before experimentation. Ra, measured as described under MATERIALS AND METHODS, is expressed as µmol
glucose · kg body wt21 · min21. Results are expressed as means 6 SE; n 5 5 rats/group. Left: diabetic rats weighed
270 6 10 g, and CoCl2-treated diabetic rats weighed 270 6 8 g; serum glucose concentrations during measurement
of Ra were 6.37 6 0.22 and 6.04 6 0.42 mM, respectively. Nonfasting serum glucose concentrations before placement
of catheters were 37.5 6 3.0 and 20.6 6 2.6 mM in diabetic rats and CoCl2-treated diabetic rats, respectively. * P ,
0.05 compared with respective CoCl2-untreated group. Right: normal rats weighed 290 6 20 g, and CoCl2-treated
normal rats weighed 260 6 5 g; serum glucose concentrations during measurement of Ra were 5.54 6 0.18 and
5.64 6 0.49 mM, respectively. Changes in Ra resulting from CoCl2 treatment are not significant.
E988
COCL2 INHIBITION OF SYSTEMIC GLUCOSE PRODUCTION
Fig. 4. Effect of CoCl2 on relative abundance of phosphoenolpyruvate
carboxykinase (PEPCK) mRNA in livers of diabetic (Diab) and
normal (N) rats. Diabetic or normal rats were treated or not treated
with CoCl2 for 12 days before assay. Rats were killed, and livers were
rapidly removed, frozen in dry ice, and assayed within 2 wk.
A: animals were not fasted. B: animals were deprived of food for 24 h
before experiment. In each group, results were normalized against
average value in normal rats. Results are expressed as means 6 SE;
n 5 5–6 rats in each group. ** P , 0.001, * P , 0.05 compared with
respective CoCl2-untreated group.
Fig. 5. Effect of CoCl2 on concentration of cAMP in livers of diabetic
and normal rats. Experimental animals were same as those in Fig. 4.
Animals were either not fasted (A) or deprived of food for 24 h (B)
before experiment. Results are expressed as means 6 SE; n 5 5–6
rats in each group. None of changes induced by CoCl2 are significant.
Concentration of cAMP in liver of food-deprived normal rats is
significantly higher than that in nonfasted normal rats (P , 0.05).
in liver of fasted normal rats and decreased to below
detectable levels in CoCl2-treated rats (data not shown).
Effect of CoCl2 on the levels of cAMP in the livers of
diabetic and normal rats. It is well established that the
expression of the gene for the cytosolic form of PEPCK
in liver is highly regulated by several hormones, including glucagon, insulin, growth hormone, and cortisol,
and by the nutritional status of the animal (10). Hence
the reduction in the levels of PEPCK mRNA in livers of
CoCl2-treated diabetic and normal animals summarized above may well be mediated by alterations in one
or a combination of the above regulators. Because of the
dominant and opposing roles of glucagon (acting through
cAMP) and insulin on the regulation of PEPCK expression (9, 17) and because of our finding that the concentration of serum insulin is not changed as a result of
treatment with CoCl2 in either diabetic or normal rats
(Table 1), we examined the possibility that the concentration of hepatic cAMP is reduced in CoCl2-treated
rats (Fig. 5). Under nonfasting conditions, the concentration of cAMP is slightly (but not significantly) lower
in the livers of diabetic compared with normal rats (Fig.
5A) and decreases slightly as a result of CoCl2 treatment. Similarly, there was no significant change in
cAMP levels in 24-h food-deprived normal rats treated
with CoCl2 (Fig. 5B). Food deprivation in normal rats,
however, was associated with a significant increase in
hepatic cAMP levels from 625 6 77 to 823 6 40 pmol/g
wet wt (P , 0.05).
DISCUSSION
The present study was prompted from the earlier
observation that treatment of STZ-induced diabetic
rats with CoCl2 results in a significant reduction in the
serum glucose concentration (31). In principle, the
Downloaded from http://ajpendo.physiology.org/ by 10.220.32.247 on June 18, 2017
tent (13), we measured the effect of CoCl2 treatment on
the concentration of PEPCK mRNA in the liver. Two
experimental protocols were employed. In the first, the
relative abundance of PEPCK mRNA was measured in
livers of nonfasted normal rats, diabetic rats, and
diabetic rats treated with CoCl2 (Fig. 4A). Rats were
not fasted because food deprivation increases PEPCK
expression in normal rats. The content of PEPCK
mRNA in the liver of diabetic rats was increased to
3.1-fold over that found in the liver of nontreated rats
(P , 0.001). Treatment of diabetic rats with CoCl2
resulted in a significant reduction in the content of
PEPCK mRNA in the liver to levels similar to those
found in livers of normal rats not treated with CoCl2
(P . 0.4). In a separate set of experiments, we found
that treatment of diabetic rats with CoCl2 under nonfasting conditions also decreased liver PEPCK mRNA
content by approximately threefold (P , 0.05; data not
shown). In the second protocol, the effect of CoCl2 on the
level of PEPCK mRNA in the liver of nondiabetic
animals was determined. In this experiment, normal
and CoCl2-treated normal rats were deprived of food for
24 h to elicit an upregulation of PEPCK gene expression before study (Fig. 4B). The relative abundance of
hepatic PEPCK mRNA was decreased by 33% in normal rats treated with CoCl2 compared with the control
group (P , 0.05). The effect of CoCl2 was also determined in nonfasted normal and CoCl2-treated normal
rats. PEPCK mRNA content was present at low levels
COCL2 INHIBITION OF SYSTEMIC GLUCOSE PRODUCTION
the organism as a whole, is apt to be reduced in
CoCl2-treated animals. It should also be noted that the
decreased clearance of glucose from the circulation of
diabetic rats treated with CoCl2 might help explain the
finding that the glycemia of these rats is not lower than
that of diabetic rats not treated with CoCl2.
Because the measurements of Ra and 2-deoxy[14C]glucose uptake in various tissues were performed
after a fasting period of 24–30 h, it is highly likely that
Ra of glucose closely reflects gluconeogenesis rather
than glycogenolysis. We therefore explored the possibility that treatment with CoCl2 reduces gluconeogenesis
by the liver, a process that is controlled in part by the
level of PEPCK activity (29). Because PEPCK activity
closely parallels the level of its mRNA (13), we measured the relative abundance of PEPCK mRNA in
livers of diabetic and normal rats treated with CoCl2. In
accordance with previous findings, the abundance of
PEPCK mRNA was increased in livers of diabetic rats
compared with normal controls under nonfasting conditions (7). In addition, treatment with CoCl2 significantly reduced hepatic PEPCK mRNA levels in diabetic
rats under fasting or nonfasting conditions. CoCl2
treatment also significantly reduced PEPCK mRNA
levels in the livers of normal rats. However, a direct
correspondence between PEPCK mRNA and hepatic
glucose production may not always exist, especially
under differing experimental and nutritional conditions. It is worth emphasizing that the decrease in
PEPCK mRNA content does not represent a nonspecific
or toxic effect of the agent because diabetic rats treated
with up to 4 mM CoCl2 for 7 wk demonstrated no
reduction in weight gain compared with diabetic controls not treated with CoCl2 (31), and the abundances of
other mRNAs such as those encoding erythropoietin (8,
30) and GLUT-1 and GLUT-2 (31) are increased in
CoCl2-treated rats.
The mechanism by which CoCl2 modifies the level of
PEPCK mRNA in the livers of diabetic rats is of
interest. When CoCl2 is added to cells in the presence of
oxygen, many of its effects on gene expression mimic
the effects noted in response to lowered oxygen concentration (8). For example, transcription of the gene
coding for GLUT-1 is stimulated in response to hypoxia
by oxygen-sensing molecules that can also be activated
by cobalt (2). Cobalt is thought to alter gene transcription by increasing the level of hypoxia-induced factor
(HIF)-1, a transcription factor that binds to a regulatory element (CGTGCTG) in the promoter of a number
of genes, most notably erythropoietin and vascular
endothelial growth factor genes (30). The steady-state
concentration of HIF-1 is induced by hypoxia or CoCl2
by a mechanism that involves the stabilization of the
protein against degradation, resulting in an accumulation of the transcription factor (24). The gene for
PEPCK has also been shown to respond to changes in
the redox state of liver cells in culture (11). This gene is
expressed in the liver in a decreasing gradient from the
periportal to the pericentral region (14), presumably
because of higher levels of oxygen and nutrients supplied to hepatocytes in the periportal region. Previous
Downloaded from http://ajpendo.physiology.org/ by 10.220.32.247 on June 18, 2017
glycemia-lowering effect of CoCl2 could be secondary to
decreased systemic glucose production, increased tissue glucose uptake, or a combination of the two mechanisms. The observed induction of GLUT-1 mRNA in
several tissues of CoCl2-treated rats suggested that
enhanced glucose uptake may well play a dominant
role in the above effect, although GLUT-1 expression
and glucose transport were not measured in that study
(31). The results of studies reported herein, however,
indicate that the glycemia-lowering effect of CoCl2 is
mediated by a reduction in the Ra of glucose and
possibly gluconeogenesis. It is worth emphasizing that
the results of the present study are internally consistent, i.e., the decrease in glycemia of CoCl2-treated
diabetic rats is associated with a significant reduction
in Ra and a decrease in liver PEPCK mRNA content. A
reduction in calorie intake as the explanation for the
reduction in glucose production is excluded by the
observation that, in accordance with previous results
(31), diabetic rats treated or not treated with CoCl2
gained weight at an equal rate during the 2-wk period
of treatment.
Previous reports indicate that glucose Ra values in
non-insulin-dependent diabetic subjects are higher than
in normal controls (22), although the higher Ra values
in diabetic subjects are documented under conditions in
which the blood glucose concentrations are significantly higher than for nondiabetic controls (28). Ra
values measured in the present study, whether expressed per animal or per unit body weight, were
somewhat but not significantly lower in the diabetic
animals. The reasons for this apparent discrepancy are
unknown but may reflect the fact that the model
employed in the present study is one of insulindependent diabetes. Moreover, unlike previous protocols, diabetic and normal rats in this study were
deprived of food for 24–30 h before measurement of
glucose Ra, a condition which served to markedly
reduce the serum glucose concentration of the diabetic
group to levels approximating those in food-deprived
normal rats. A profound depressing effect of a prolonged
period of food deprivation (72 h) on hepatic glucose
production in the rat has been reported previously (1).
Treatment of diabetic rats with CoCl2 resulted in ,50%
reduction in glucose Ra. If glucose Ra is decreased by a
similar extent under nonfasting conditions, then this
change alone would be expected to result in a significant decrease in the glycemia of diabetic rats.
There was a tendency toward lower 2-deoxy-[14C]glucose uptake in tissues of CoCl2-treated rats, although
only in the case of kidney and eye in diabetic rats was
the decrease significant. It is possible that the marked
reduction in glycemia of diabetic rats during the period
of food deprivation results in a lower rate of tissue
glucose uptake and thereby leads to a masking of any
further decrement in glucose uptake by treatment with
CoCl2. In keeping with this premise, it has been reported that food deprivation for 72 h in normal rats
results in a decrease in sensitivity of peripheral tissues
to the actions of insulin (1). Nevertheless, the summation of tissue uptake values, reflecting the uptake by
E989
E990
COCL2 INHIBITION OF SYSTEMIC GLUCOSE PRODUCTION
This study was supported, in part, by grants from the Diabetes
Association of Greater Cleveland and the National Institutes of
Health (DK-45945 to F. Ismail-Beigi, DK-25541 to R. W. Hanson, and
HD-11089 to S. C. Kalhan). P. Leahy and F. Saker were trainees on
the National Institute of Diabetes and Digestive and Kidney Diseases
Metabolism Training Grant DK-07319.
F. Saker and J. Ybarra contributed equally to this study.
Address for reprint requests: F. Ismail-Beigi, Clinical and Molecular Endocrinology, Case Western Reserve Univ., Cleveland, OH
44106-4951.
Received 22 October 1997; accepted in final form 20 February 1998.
REFERENCES
1. Barzilai, N., D. Massiillon, and L. Rossetti. Effects of fasting
on hepatic and peripheral glucose metabolism in conscious rats
with near-total fat depletion. Biochem. J. 310: 819–826, 1995.
2. Behrooz, A., and F. Ismail-Beigi. Dual control of GLUT1
glucose transporter gene expression by hypoxia and by inhibition
of oxidative phosphorylation. J. Biol. Chem. 272: 5555–5562,
1997.
3. Eaton, R. P. Cobalt chloride-induced hyperlipemia in the rat:
effects on intermediary metabolism. Am. J. Physiol. 222: 1550–
1557, 1972.
4. Eaton, R. P. Glucagon secretion and activity in the cobalt
chloride-treated rat. Am. J. Physiol. 225: 67–72, 1973.
5. Elbrink, J., and I. Bihler. Membrane transport: its relation to
cellular metabolic rates. Science 188: 1177–1184, 1975.
6. Ferré, P., A. Leturque, A.-F. Burnol, L. Penicaud, and J.
Girard. A method to quantify glucose utilization in vivo in
skeletal muscle and white adipose tissue of the anaesthetized
rat. Biochem. J. 228: 103–110, 1985.
7. Friedman, J. E., J. S. Yun, Y. M. Patel, M. M. McGrane, and
R. W. Hanson. Glucocorticoids regulate the induction of
P-enolpyruvate carboxykinase (GTP) gene transcription during
diabetes. J. Biol. Chem. 268: 12952–12957, 1993.
8. Goldberg, M. A., S. P. Dunning, and H. F. Bunn. Regulation of
the erythropoietin gene: evidence that the oxygen sensor is a
heme protein. Science 242: 1412–1415, 1988.
9. Granner, D. K., T. Andreone, K. Sasaki, and E. Beale.
Inhibition of transcription of the P-enolpyruvate carboxykinase
gene by insulin. Nature 305: 549–551, 1983.
10. Hanson, R. W., and Y. M. Patel. Phosphoenolpyruvate carboxykinase (GTP): the gene and the enzyme. Adv. Enzymol.
Relat. Areas Mol. Biol. 56: 203–281, 1994.
11. Hellkamp, J., B. Christ, H. Bastian, and K. Jungermann.
Modulation by oxygen of the glucagon-dependent activation of
the P-enolpyruvate carboxykinase gene in rat hepatocyte cultures. Eur. J. Biochem. 198: 635–639, 1991.
12. Ismail-Beigi, F. Metabolic regulation of glucose transport. J.
Membr. Biol. 135: 1–10, 1993.
13. Iynedjian, P. B., and R. W. Hanson. Increase in the level of
functional mRNA coding for P-enolpyruvate carboxykinase (GTP)
during induction by cAMP. J. Biol. Chem. 252: 655–662, 1977.
14. Jungermann, K., and N. Katz. Functional hepatocellular
heterogeneity. Hepatology 2: 385–395, 1982.
15. Jungermann, K., and T. Kietzmann. Zonation of parenchymal
and nonparenchymal metabolism in liver. Annu. Rev. Nutr. 16:
179–203, 1996.
16. Kalhan, S. C., K.-Y. Tserng, C. Gilfillan, and L. J. Dierker.
Metabolism of urea and glucose in normal and diabetic pregnancy. Metabolism 31: 824–833, 1982.
17. Lamers, W. H., R. W. Hanson, and H. M. Meisner. Cyclic AMP
stimulates transcription of the gene for cytosolic phosphoenolpyruvate carboxykinase in rat liver nuclei. Proc. Natl. Acad. Sci.
USA 79: 5137–5141, 1982.
18. Liu, J., E. A. Park, A. L. Gurney, W. J. Roesler, and R. W.
Hanson. Cyclic AMP induction of P-enolpyruvate carboxykinase
gene transcription is mediated by multiple promoter elements. J.
Biol. Chem. 266: 19095–19102, 1991.
19. Mercado, C. L., J. N. Loeb, and F. Ismail-Beigi. Enhanced
glucose transport in response to inhibition of respiration in Clone
9 cells. Am. J. Physiol. 257 (Cell Physiol. 26): C19–C28, 1989.
20. Mueckler, M. Family of glucose-transporter genes. Implication
for glucose homeostasis and diabetes. Diabetes 39: 6–11, 1990.
21. Patel, Y. M., J. S. Yun, J. Liu, M. M. McGrane, and R. W.
Hanson. An analysis of regulatory elements in the PEPCK gene
Downloaded from http://ajpendo.physiology.org/ by 10.220.32.247 on June 18, 2017
studies by Hellkamp et al. (11) have also demonstrated
that glucagon-induced transcription from the PEPCK
promoter is inhibited by reducing the concentration of
oxygen from 16 to 8%. The hypoxia-inducible DNA
element noted in the 38-flanking region of the erythropoietin gene is also present in the PEPCK promoter at
2129 to 2121 kb (from the transcription start site), a
position that is immediately 58 to a control region that
includes the cAMP and the nuclear factor-1 regulatory
elements (2120 to 280 kb), which are critical for both
basal and cAMP-induced transcription of the PEPCK
gene (18, 21). However, T. Kietzmann (personal communication) has implicated a site in the PEPCK gene
between 2277 and 2174 kb as being required for the
negative effect of redox state on transcription from the
PEPCK promoter in primary hepatocytes in culture
(15). Although it is possible that the effect of CoCl2
noted in the present study is due to an inhibition of
transcription of the PEPCK gene in the liver via a
mechanism involving HIF-1, it is likely that there are
multiple elements in the PEPCK promoter involved in
the effect of cobalt on PEPCK gene expression.
Cytosolic PEPCK gene expression in the liver is
highly controlled by a number of hormones and physiological conditions and most importantly by glucagon
and insulin mediation of positive and negative regulation, respectively. Results of previous studies indicate
that treatment with CoCl2 does not alter the concentration of insulin in the blood of normal rats (3, 31) and
does not increase the extremely low levels of insulin in
the blood of STZ-induced diabetic rats (31). It has also
been reported that livers of CoCl2-treated nondiabetic
rats are relatively ‘‘insensitive’’ to glucagon action,
since the release of glucose by the liver in response to
glucagon is decreased both in vivo and in vitro (4); the
lower release of glucose is present, although the livers
of CoCl2-treated rats contain higher levels of glycogen
(3). Because of the critical regulation of PEPCK gene
transcription by glucagon and cAMP, we elected to
measure the levels of this nucleotide in liver of diabetic
and normal rats treated or not treated with CoCl2. The
results showed no systematic change among any of the
groups examined except for the finding that food deprivation of normal rats resulted in a significant increase
in the level of cAMP in the liver. It has been reported
that CoCl2-treated rats appear to be more ‘‘sensitive’’ to
the action of insulin in enhancing glucose disposal from
the blood after a glucose load (3). If the liver of
CoCl2-treated rats is also more sensitive to actions of
insulin, then such a finding might help explain the
lower glucose Ra in such animals. This explanation,
however, cannot be extended to diabetic rats, given the
extremely low levels of insulin in these animals. Further studies are required to gain a better understanding of the mechanisms mediating the apparent reduction of gluconeogenesis in CoCl2-treated diabetic and
normal rats.
COCL2 INHIBITION OF SYSTEMIC GLUCOSE PRODUCTION
22.
23.
24.
25.
27.
28.
29.
30.
31.
32.
lism: pre- and posttranslational mechanisms. Am. J. Physiol. 262
(Cell Physiol. 31): C527–C532, 1992.
Snedecor, G. W., and W. G. Cochran. Statistical Methods.
Ames: Iowa Univ. Press, 1976.
Tayek, J. A., and J. Katz. Glucose production, recycling, and
gluconeogenesis in normals and diabetics: a mass isotopomer
[U-13C]glucose study. Am. J. Physiol. 270 (Endocrinol. Metab.
33): E709–E717, 1996.
Valera, A., A. Pujol, M. Pelegrin, and F. Bosch. Transgenic
mice overexpressing PEPCK develop non-insulin-dependent diabetes. Proc. Natl. Acad. Sci. USA 91: 9151–9154, 1994.
Wang, G. L., B. H. Jiang, E. A. Rue, and G. L. Semenza.
Hypoxia-inducible factor 1 is a basic-helix-loop-helix-PAS heterodimer regulated by cellular O2 tension. Proc. Natl. Acad. Sci.
USA 92: 5510–5514, 1995.
Ybarra, J., A. Behrooz, A. Gabriel, M. H. Koseoglu, and F.
Ismail Beigi. Glycemia-lowering effect of cobalt chloride in the
diabetic rat. Mol. Cell. Endocrinol. 133: 151–160, 1997.
Yoo-Warren, H., J. E. Monahan, J. Short, H. Short, A.
Bruzel, A. Wynshaw-Boris, H. M. Meisner, D. Samols, and
R. W. Hanson. Isolation and characterization of the gene for
cytosolic P-enolpyruvate carboxykinase (GTP) from the rat. Proc.
Natl. Acad. Sci. USA 80: 3656–3660, 1983.
Downloaded from http://ajpendo.physiology.org/ by 10.220.32.247 on June 18, 2017
26.
which are responsible for its tissue-specific expression and
metabolic control in transgenic mice. J. Biol. Chem. 269: 5619–
5628, 1994.
Perriello, G., S. Pampanelli, P. Del Sindaco, C. Lalli, M.
Ciofetta, E. Volpi, F. Santeusanio, P. Brunetti, and G. B.
Bolli. Evidence of increased systemic glucose production and
gluconeogenesis in an early stage of NIDDM. Diabetes 46:
1010–1016, 1997.
Pessin, J. E., and G. I. Bell. Mammalian facilitative glucose
transporter family: structure and molecular regulation. Annu.
Rev. Physiol. 54: 911–930, 1992.
Salceda, S., and J. Caro. Hypoxia-inducible factor 1a (HIF-1a)
protein is rapidly degraded by the ubiquitin-proteasome system
under normoxic conditions: its stabilization by hypoxia depends
on redox-induced changes. J. Biol. Chem. 272: 22642–22647,
1997.
Shetty, M., N. Ismail-Beigi, J. N. Loeb, and F. Ismail-Beigi.
Induction of GLUT1 mRNA in response to inhibition of oxidative
phosphorylation. Am. J. Physiol. 265 (Cell Physiol. 34): C1224–
C1229, 1993.
Shetty, M., J. N. Loeb, and F. Ismail-Beigi. Enhancement of
glucose transport in response to inhibition of oxidative metabo-
E991