 |
||||
| 2007 JASN IMPACT FACTOR 7.111 | HOME AUTHOR INFO EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP | |||
| CURRENT ISSUE | ARCHIVES | JASN Express | ONLINE SUBMISSION | |
|
||||
| 2007 JASN IMPACT FACTOR 7.111 | HOME AUTHOR INFO EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP | |||
| CURRENT ISSUE | ARCHIVES | JASN Express | ONLINE SUBMISSION | |
*
School of Biological Sciences, University of Manchester, Manchester,
United Kingdom.
University School of Medicine, Manchester Royal Infirmary, Manchester,
United Kingdom.
Correspondence to Dr. Daniela Riccardi, G38 Stopford Building, School of Biological Sciences, Oxford Road, Manchester, M13 9PT, United Kingdom. Phone: 44-161-275-5944; Fax: 44-161-275-5600; E-mail: riccardi{at}man.ac.uk
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Classically, rat renal function is studied by collection of 24-h urine samples in metabolic cages. However, an improvement on this method is to cannulate the rats to allow real-time collection of urine from the bladder and to permit infusion of [3H]inulin for the accurate measurement of GFR. In this model, fluid balance is maintained during the time period in which urine and blood samples are collected for analysis by replacing fluid lost in the urine with a servo-controlled infusion pump (13,14). In addition, using this model, data can be obtained from conscious rats, with the advantage of preventing the depressive effects of both acute surgery and anesthesia on renal hemodynamics. In the first series of experiments, we therefore studied the effect of chronic (2 wk) diabetes on the renal function in conscious catheterized rats.
To determine the molecular mechanisms that could be affected by the diabetic condition and that could therefore explain the altered mineral ion metabolism during diabetes, we looked at changes in the renal expression of proteins involved in renal calcium and water transport in control versus STZ-diabetic rats and/or in diabetic rats on insulin replacement in the second series of experiments. A central protein in divalent cation homeostasis is the cell surface, calcium/polyvalent cation-sensing receptor (CaR) (15,16). In the distal nephron, the CaR maintains normocalcemia by integrating signals that arise from divalent cation excretion and water preservation (17). Calbindin-D28k is an intracellular Ca2+-binding protein that regulates cellular calcium transport in the late distal convoluted tubule (DCT) and connecting tubule (CNT) cells (18). Plasma membrane Ca2+-ATPase (PMCA) is found on the basolateral surface of DCT and CNT cells and can actively extrude Ca2+ against a concentration gradient (18). The thiazide-sensitive NaCl contransporter (NCCT) is expressed on the apical membrane of DCT and early connecting segment cells (19,20), and in the DCT, Na+ reabsorption through NCCT is inversely related to urinary Ca2+ excretion (21). Aquaporin 1 (AQP1) is a constitutively active water channel located in renal proximal tubules and part of the descending thin limbs. AQP2, the vasopressin-regulated water channel, is localized to the apical side of collecting ducts and is specifically inserted into the apical membrane in response to vasopressin (reviewed in reference 22). Accordingly, the expression levels of CaR, NCCT, calbindin-D28k, PMCA, and AQP1 and 2 proteins were investigated in the development of diabetic hypercalciuria. Finally, in the same experimental rats, we measured bone formation (osteocalcin) and resorption (urinary deoxypyridinoline crosslinks) markers, as well as circulating levels of parathyroid hormone (PTH) and 1,25-dihydroxy vitamin D3 (1,25(OH)2D3).
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Animals
Experiments were performed in accordance with the UK Animals (Scientific
Procedures) Act of 1986. Male Sprague-Dawley rats (Charles River Laboratories,
Wilmington, Kent, UK) were individually housed in wire-bottom cages and were
maintained under a 12-h light photoperiod (lights on at 0800 h) and an
environmental temperature of 21 to 23°C (55 ± 10% humidity). Rats
had free access to food (Beekay Rat and Mouse Diet, Bantin & Kingman,
Hull, UK) and tap water throughout the study. Experiments were performed on
two separate series of rats, the first of which was used for the renal
function measurements and the second for the protein expression and for
biochemical determination of 25(OH)D3,
1,25(OH)2D3, PTH, and bone formation and resorption
markers.
Induction of Diabetes Mellitus with STZ
Eight-wk-old Sprague-Dawley rats were rendered diabetic with STZ (60 mg/kg
intraperitoneally in citrate buffer)
(13). Control rats received
citrate buffer alone. Diabetes was confirmed by the development of glycosuria
within 36 h (Uristix; Ames DVN, Miles Ltd., Slough, UK) and hyperglycemia
(blood glucose concentration > 15 mM). Blood for the latter was obtained
from a tail vein (or from the venous cannula in cannulated rats) and assayed
using a blood glucose analyzer (HemoCue, Sheffield, UK). Half of the rats made
diabetic then received one interscapular subcutaneous 14% porcine insulin
implant (Linplant, LinShin Canada Inc., Toronto, Ontario, Canada), under
halothane anesthesia immediately after confirmation of diabetes, to maintain
blood glucose concentration below 10 mM.
In Vivo Measurements of Renal Function
The methodology used for the servo-controlled fluid replacement system is
described extensively elsewhere
(13,14).
Briefly, sterilized cannulae were implanted into the femoral artery (for blood
sampling and measuring BP) and vein (for infusion) of rats under anesthesia.
Custom-made titanium bladder catheters (AstraZeneca Pharmaceuticals, Alderley
Park, UK) were implanted exteriorized through the ventral abdominal wall. The
rats were then left to regain presurgical body weight, which took
approximately 5 d.
Each rat was placed in a restraining cage and allowed to settle for 1 h before administration of a bolus infusion of 6 µCi of [3H]inulin. The mass of voided urine was measured at 5-min intervals, and an adjustable infusion pump then delivered an infusion of 2.5% dextrose solution at a rate matching the spontaneous urine output. Preliminary experiments measuring urinary ion excretion over a 5-h period indicated that the period of the experiment that gave the most stable rates of ion excretion occurred between 3 and 4.5 h after beginning the renal clearance measurements (t0; data not shown). Thus, results shown represent the data obtained during the period 3.5 and 4 h after t0. Urine samples collected over three 30-min intervals during the period 3 to 4.5 h after to were weighed and stored in screw-top vials at 4°C until being analyzed. Blood samples were taken at the midpoint of the urine collection periods (t = 3, 3.75, and 4.5 h). At the end of the experiment, the rats were returned to their cages. On the following day, rats were injected intravenously with or without STZ as described above, and after 1 wk and then again after 2 wk, the above renal function protocol was repeated.
Parameters of renal function were quantified as described previously (12,13). Briefly, GFR was determined by [3H]insulin clearance, urinary calcium and magnesium contents were analyzed using an atomic absorption spectrophotometer, and urinary glucose was determined using a commercially available kit (Ames Sera-Pak, Bayer Diagnostics, Basingstoke, UK). Calculations of renal function were performed as described previously (13,14).
Preparation of Kidney Crude Membranes
After 14 d, the noncatheterized rats were killed and single kidneys from
each were excised and homogenized in buffer containing 12 mM HEPES (pH 7.6),
300 mM mannitol, pepstatin, leupeptin, and phenylmethylsulfonyl fluoride. To
reduce between-group variability, we processed kidneys in groups of three, one
from each experimental condition (i.e., control, diabetic, and
insulin-diabetic rats). The homogenate was centrifuged at 2500 x
g for 15 min, and an aliquot of this postnuclear supernatant was
centrifuged at 100,000 x g for 30 min to give a particulate
protein pellet. Samples were normalized for protein content by assaying
according to the method of Bradford
(23). The protein equivalency
of the subsequent sample loading volumes was demonstrated by staining of a
sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) gel with
Coomassie blue (not shown).
Immunoblotting
Immunoblotting was performed as described previously
(24) with specific conditions
for each protein as follows. All samples, except for CaR, were denatured at
100°C for 5 min in the presence of 143 mM ß-mercaptoethanol before
SDS-PAGE. For CaR, samples were denatured in the absence of reducing agent at
room temperature. We previously showed that CaR resolved under nonreducing
conditions migrates as a 240- to 310-kD dimeric species
(24). Thus, to improve on the
method of quantification of CaR, we resolved the protein under nonreducing
conditions here so as to have only one band to quantify as opposed to three.
The antisera used included affinity-purified anti-CaR (1:800 dilution, raised
in rabbit to amino acids 214 to 236 of the extracellular domain of the rat
kidney CaR [Lofstrand, Inc., Bethesda, MD] and affinity purified as described
elsewhere (25), monoclonal
anti-calbindin-D28k (1:2500 dilution; Sigma-Aldrich), monoclonal
anti-PMCA (1:1500 dilution; Cambridge Bioscience, Cambridge, UK),
affinity-purified polyclonal anti-thiazide-sensitive NCCT (1:5000 dilution; a
gift of Dr. Steven Hebert, Vanderbilt University, Nashville, TN), and
affinity-purified polyclonal anti-AQP1 and AQP2 (1:5000 dilution; Chemicon
Int., Harrow, UK).
Immunofluorescence Microscopy
Assessment of immunoreactivity for both CaR and NCCT proteins in kidneys
from normal versus diabetic rats was performed as described
previously (25). Briefly, rat
kidneys were perfusion fixed with 4% paraformaldehyde and cryoprotected in
sucrose solution. Four-µm cryosections were antigen-retrieved using citrate
buffer, permeabilized with 1% SDS (for CaR immunostaining only), and stained
with affinity-purified anti-CaR or anti-NCCT polyclonal antibodies. As a
secondary fluorescence antibody, an anti-rabbit IgG conjugated with Texas Red
was used according to the manufacturer's instructions. Slides were viewed
using a Zeiss Axioplan 2 microscope with 10 to 40x objectives. Images
were acquired using a Hamamatsu digital camera and processed using the
software package KS300 version 3.0 (Carl Zeiss Ltd., Hertfordshire, UK).
Measurement of Serum 25(OH)D3,
1,25(OH)2D3, PTH, Deoxypyridinoline Cross Links, and
Osteocalcin Levels
Serum levels of 25(OH)D3, and 1,25(OH)2D3
were assayed by in-house methods as previously published
(26). Rat serum PTH was
assayed using a kit supplied by Nicholls Institute Diagnostics (San Juan
Capistrano, CA). Urinary deoxypyridinoline cross links and rat serum
osteocalcin were measured using kits provided by Metra Bio-systems Inc.
(Mountain View, CA). Serum calcium levels in the same experimental group were
determined using atomic absorption spectrophotometry.
Statistical Analyses
Data are presented as means ± SEM, and statistical significance was
determined either by multivariate ANOVA test for the urine and plasma in
vivo data or by unpaired t test for the semiquantitative
immunoblots.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
|
|
|
|
|
|
As shown in Figure 1A, GFR was increased by STZ diabetes from 2510 ± 231 µl/min in control rats to 4189 ± 454 µl/min, and this effect was greatly ameliorated by insulin replacement. The urine output of the STZ-diabetic rats was substantially higher than in control rats (Figure 1B). By day 14, the polyuria was still evident but was lower than at day 7 and was again normalized by insulin replacement. Expressed as percentage of fractional fluid reabsorption, water reuptake was reduced in the STZ-diabetic rats after 14 d from 98.3 ± 0.4% in control to 95.1 ± 1.1%.
There was a sixfold increase in urinary calcium output in the STZ-diabetic
rats by day 7 that remained stable until day 14 (0.167 ± 0.022
versus 0.025 ± 0.003 µmol/min control;
Figure 2). Again, insulin
replacement fully normalized the hypercalciuria. The hypercalciuria was
associated with significantly reduced fractional calcium reabsorption in
STZ-diabetic rats (day 14, 96.9 ± 0.6% versus 99.1 ±
0.2% control). Levels of total plasma Ca2+ were unchanged by STZ
diabetes (day 14, control 2.06 ± 0.05 mM versus STZ-diabetic
2.0 ± 0.02 mM; N 
5, NS). Similarly, the levels of ultrafiltrable
Ca2+ in the blood, i.e., free ionized Ca2+,
were not significantly altered by STZ diabetes (day 14, control 1.31 ±
0.08 mM versus STZ-diabetic 1.11 ± 0.12 mM; N 5, NS).
In the servo-controlled, fluid-replaced STZ-diabetic rats, there was a more
modest increase in urinary magnesium excretion that failed to reach
significance versus day 14 control rats (0.36 ± 0.033
µmol/min STZ-diabetic versus 0.209 ± 0.047 µmol/min
control; N 6, NS). However, the magnesium excretion rate in day 14
STZ-diabetic rats was significantly higher than in day 14 insulin-treated
diabetic rats (0.174 ± 0.056 µmol/min; n = 7; P
< 0.05). Despite this apparent increase in urinary magnesium excretion,
there was no change in fractional magnesium reabsorption (STZ-diabetes, 80.7
± 1.4% versus 79.3 ± 5.2% control; N 5, NS). Also,
there were no significant changes in STZ diabetes in either total plasma
Mg2+ levels (0.58 ± 0.02 mM in day 14 control
versus 0.63 ± 0.02 in day 14 STZ-diabetic; N 5, NS) or
ultrafiltrable Mg2+ levels (0.44 ± 0.02 mM in day 14 control
versus 0.52 ± 0.05 in day 14 STZ-diabetic; N 5, NS).
Urinary sodium excretion did not change significantly in the
servo-controlled rats during diabetes either in the presence or in the absence
of insulin (1.26 ± 0.30 µmol/min control versus 0.58
± 0.19 µmol/min STZ-diabetic, 1.57 ± 0.50 insulintreated
STZ-diabetic; N 6, NS). Similarly, percentage of fractional sodium
reabsorption did not differ between experimental groups (99.65 ± 0.06%
control versus 99.91 ± 0.03% STZ-diabetic, 99.54 ±
0.19% insulin-treated STZ-diabetic; N 5, NS). In contrast, there was a
significant reduction in plasma sodium concentration in STZ-diabetic rats but
not in insulin-treated diabetic rats (148 ± 4.9 mM control, 131.8
± 3.7 mM STZ-diabetic, 141 ± 3 mM insulin-treated STZ-diabetic;
N 5; P < 0.05 control versus STZ-diabetic).
Molecular and Biochemical Characterization of Diabetic
Hypercalciuria
Expression of the Extracellular CaR. To assess the effect of
diabetes on rat renal CaR expression, we used semiquantitative immunoblotting
to measure the whole kidney CaR content. By this method, we determined that
the whole kidney content of CaR is reduced by STZ diabetes to 52% of control
levels (Figure 3, B and C). In
contrast, the mean renal CaR content of the insulin-treated diabetic rats was
normal. No signal was detected in rat kidney membranes when the anti-CaR
antibody was preabsorbed by the antigenic peptide
(Figure 3A).
Previously, we showed that the strongest CaR-specific immunoreactivity is expressed in the proximal tubules, thick ascending limb (TAL), DCT, and the collecting ducts (25). Immunofluorescence microscopy on rat kidney cryosections showed a uniform reduction in CaR staining throughout the kidney section in the control versus diabetic rats (Figure 4). At higher magnification, the reduction in CaR immunostaining in cortical TAL (Figure 4, C and D) seemed to be due to a decrease in both the intensity of the immunoreactivity and the number of cells expressing the receptor. In addition, Figure 4, E and F, show that the reduction did not seem to be region specific, as it could also be observed in proximal tubule, collecting duct, and DCT cells and also in medullary regions (not shown). There was no apparent change in CaR immunoreactivity in control versus insulin-treated diabetic rats, and no signal was detected in slides in which the primary antibody was preabsorbed with the antigenic peptide (not shown).
Expression of Calbindin-D28k and PMCA. The monoclonal anti-calbindin-D28k antibody detected a single protein band (Mr approximately 26 kD) in the postnuclear supernatants of whole kidneys (Figure 5A, upper). The relative abundance of this calbindin-D28k—reactive species was the same as control in both non—insulin-treated and insulin-treated STZ-diabetic rats (Figure 5B, left). The anti-PMCA monoclonal antisera detected a broad band of 128 to 155 kD as well as a smaller band of 89 kD (Figure 5A, lower). There was no change in PMCA protein abundance in the particulate fractions of total kidneys from control versus STZ-diabetic rats and STZ-diabetic rats on insulin replacement (Figure 5B, right).
Expression of the Thiazide-Sensitive NCCT. The anti-NCCT antibody detected a broad immunoreactive band of approximate molecular mass 138 to 181 kD (Figure 6) similar in mass to NCCT immunoreactivity previously reported (19). In the STZ-diabetic rats, NCCT levels were significantly raised in all of the rats tested (+ 192%), and in particular, it was the lower portion of the NCCT band that was most markedly upregulated (Figure 6A and not shown). Figure 6B shows that membrane particulate from kidneys of insulin-treated diabetic rats exhibited levels of NCCT immunoreactivity similar to control.
In agreement with the Western blot data, NCCT immunoreactivity in rat kidney cryosections was enhanced in the DCT of diabetic versus control rats (Figure 6, C and D). No staining was detected in sections in which the primary antibody was omitted, and no difference in NCCT immunoreactivity was observed in kidney cryosections from control versus insulintreated diabetic rats (not shown).
Expression of AQP1 and AQP2. In preliminary experiments (not shown), there was a great deal of kidney-to-kidney variability between total kidney membrane samples when immunoblotted against anti-AQP2 antisera. Accordingly, when we quantified AQP2 abundance in STZ diabetes, only the inner medulla was studied. Anti-AQP1 and anti-AQP2 antisera both detected protein species of approximate molecular masses 29 (nonglycosylated) and 35 to 45 kD (glycosylated) in total kidney membrane proteins and in kidney inner medulla, respectively (Figure 7A). STZ diabetes did not affect renal AQP1 protein expression in total kidney or AQP2 abundance in inner medulla (Figure 7B).
Measurements of 1,25(OH)2D3 and Osteocalcin Levels in STZ Diabetes. 1,25(OH)2D3 levels were significantly lower in sera from STZ-diabetic rats, whereas 25(OH)D3 levels did not change significantly (Table 2). PTH values were extremely variable and did not reach significant differences between the two experimental groups (221.83 ± 200.98 pg/ml STZ-diabetic, n = 6, versus 216 ± 135.24 pg/ml control, n = 4, NS). Given this variability, we cannot be sure that PTH levels were completely unaffected by the diabetes, because a modest change in PTH levels could be masked by the statistical error. Nevertheless, we saw no evidence of even a qualitative trend suggesting altered PTH secretion in STZ diabetes, although further work would be required to confirm this. As for the in vivo measurements, serum Ca2+ levels were not significantly different in diabetic rats (Table 2). The bone formation marker osteocalcin was reduced in STZ diabetes by 53%, whereas levels of the biochemical marker for bone resorption, urinary deoxypyridinoline cross links, was not significantly altered (Table 2).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
When rats were examined under conditions of servo-controlled fluid replacement, it was found that the induction of STZ diabetes increased the glomerular flow rate and decreased fractional fluid reabsorption, producing a polyuria consistent with that seen in early human diabetes. Urinary sodium excretion was not significantly affected during diabetes in either the presence or the absence of insulin. In contrast, there was a significant reduction in plasma sodium concentration in diabetic versus control rats, which was normalized by insulin treatment. These and previous studies (11,12) clearly indicate that urinary sodium output cannot account for the strong hypercalciuria and that the latter is not simply due to a glomerular hyperfiltration, because there was no change in percentage of fractional sodium reabsorption. The most obvious explanation for the hypercalciuria seen in STZ-diabetic rats is that in diabetes, GFR is raised by the glycosuria-mediated osmotic diuresis, increasing the delivery of Ca2+ to the proximal tubule. If renal calcium reabsorption is occurring at a close-to-maximum rate, then the diuresis alone could explain the hypercalciuria. However, previous microperfusion experiments from both this and other laboratories suggest that there is a real Ca2+ handling defect in diabetes that is not localized to the proximal tubule, where Ca2+ reuptake occurs passively, and that rather it seems to be a distal tubular lesion (31,32).
In a previous study, we showed that in 2-wk diabetic rats, urinary calcium output is qualitatively similar to that obtained in the conscious rats on servo-controlled fluid replacement (12). Therefore, in the second series of experiments, we investigated the renal expression of proteins involved in mineral ion handling by the distal nephron in noninfused rats. Under these conditions, we found that after 2 wk of diabetes, rats exhibited an increase in NCCT protein. This could either represent a compensation mechanism for the additional Na+ delivery to the distal nephron caused by the elevated GFR or be a direct consequence on the DCT cells of the hyperglycemia/hypoinsulinemia. This upregulation per se could also account for the apparent decrease in Ca2+ reabsorption by the distal tubule (21). It is known that thiazide diuretics promote calcium reabsorption in the DCT in addition to inhibiting NaCl reabsorption through NCCT. In the NCCT gene knockout mouse, one observes very similar urinary sodium levels between the wild-type and homozygote animals, yet the homozygotes excrete 75% less Ca2+ (33). This is also true in patients with Gitelman's syndrome, who express mutant NCCT protein resulting in limited salt wasting but marked hypocalciuria (34). This suggests that it is not unusual for a change in NCCT activity to produce a relatively small change in urinary Na+ excretion yet a substantial alteration in renal Ca2+ excretion. The upregulation in NCCT would also explain the much greater effect on urinary Ca2+ excretion compared with Mg2+ excretion observed in the in vivo experiments. Whereas urinary excretion of magnesium was raised in the STZ-diabetic rats, there was no significant change in its fractional reabsorption, suggesting that the increased Mg2+ excretion is predominantly an osmotic effect. Hypomagnesemia has long been known to be associated with diabetes mellitus and has been confirmed in nearly one third of diabetic outpatients (35). Despite the hypermagnesiuria in our study, we did not detect a significant decrease in serum Mg2+ in diabetic rats. This could be due to the duration of diabetes studies here, because hypomagnesemia has been previously reported in STZ rats after a longer period of diabetes (36).
In our study, we identified reduced CaR protein expression in diabetes mellitus (approximately 50%), which can be seen throughout the whole kidney section and does not seem to be region specific. The important question is whether the renal CaR downregulation is a cause or a consequence of the increase in renal divalent cation excretion. Ho et al. (37) previously reported that adult CaR gene knockout mice exhibit increased renal reabsorption of Ca2+, leading to reduced Ca2+ clearance, i.e., hypocalciuria. Because the reduced CaR activity in the knockout model is due to a gene dosage effect (37), the uniform CaR downregulation seen throughout the diabetic kidneys should produce hypocalciuria instead of the observed hypercalciuria. Conversely, a downregulation secondary to hypercalciuria could explain reduced CaR expression in regions where the receptor is present at the luminal aspect, i.e., proximal tubules and collecting ducts (25), but it would not explain the clear reduction in CaR immunoreactivity in the TAL, where the receptor is basolateral. This means either that in the STZ-diabetic rat, other changes such as raised GFR or NCCT upregulation have a much greater effect on renal Ca2+ clearance than renal CaR or that the changes in renal CaR reported here are either secondary to the hypercalciuria or even unrelated to it. The diabetic condition could affect CaR expression in other organs involved in extracellular calcium homeostasis that express the CaR (reviewed in reference 38), i.e., the parathyroid glands, the gastrointestinal tract, and the bone. Schwartz et al. (39) reported a mild reduction in the set point for PTH secretion in patients with insulin-dependent diabetes mellitus, although their PTH levels were comparable to control patients'. Because the parathyroid CaR controls PTH secretion, this study suggests that in diabetics, there might be an altered sensitivity to serum calcium levels by the parathyroid CaR. It would be interesting to test CaR expression levels in parathyroid glands from poorly controlled diabetic versus control patients.
Whatever the mechanism for diabetic hypercalciuria, it does not seem to result from altered calbindin-D28k expression. A fall in the levels of the intracellular Ca2+-binding protein theoretically could contribute to reduced Ca2+ reabsorption by impairing transcellular Ca2+ transport, but such a change in expression was not observed. Similarly, there was no change observed in the renal expression of PMCA, indicating that the reduced Ca2+ reabsorption cannot be explained by decreased basolateral active Ca2+ exit. It is perhaps surprising that renal calbindin-D28k expression is unchanged given the fall in 1,25(OH)2D3 levels. However, in a previous study that reported a similar observation (40), it was suggested that either the 1,25(OH)2D3 levels, although reduced, may still have been above threshold levels for inducing gene expression or that local 1,25(OH)2 D3 levels were actually higher than the circulating levels measured. It is also possible that renal calbindin-D28k gene expression may be induced by another calciotropic agent, because vitamin D receptor-ablated mice exhibit only modestly reduced renal calbindin-D28k expression (41).
In several experimental conditions, including chronic lithium treatment (42), hypokalemia (43), low-protein diet (44), and chronic dihydrotachysterol treatment (45), a polyuria that can be explained by downregulation of AQP2 in inner medullary collecting ducts is observed. Therefore, we examined whether the polyuria seen with STZ diabetes also involves a reduction in the expression of either AQP1 or AQP2. As we observed no change in the protein levels of AQP1 in the total kidney or AQP2 in the inner medulla, we conclude that diabetes mellitus-induced diuresis does not involve downregulation of these water channels. The AQP2 result is consistent with data previously reported by Klein et al. (46), who showed that in male Sprague-Dawley rats, STZ treatment failed to alter AQP2 levels in the inner medullary tip. In contrast, a recent study that examined STZ-induced diabetes in female Wistar rats reported an increase in AQP2 levels, presumably representing a compensation mechanism to combat their severe diuresis (47). In addition to strain and gender differences between this study and the two former studies, Nejsum et al. (47) measured AQP2 expression in whole kidney rather than in inner medulla alone and thus may have detected AQP2 upregulation in cortical collecting ducts not measured here. Important, though, is that all three studies consistently observed no downregulation of AQP2.
It is still unclear what precisely is the source of the additional calcium excreted in the diabetic urine. Is osteopenia secondary to a diabetes-induced impairment of renal calcium reabsorption, or does the hypercalciuria follow impaired bone metabolism? A recent study showed that chronic furosemide treatment in rats effected a renal hypercalciuria sufficient to reduce bone mineral content and bone mineral density measured after 7 wk (48). Therefore, hypercalciuria of primarily renal origin does seem to be capable of inducing osteopenia in rats. In the last part of the study, we therefore related the observed functional and molecular changes with circulating levels of calciotropic hormones and bone markers. The bone formation marker osteocalcin was significantly reduced during diabetes, whereas the bone resorption marker deoxypyridinoline was unchanged. Serum levels of 1,25(OH)2D3 were significantly reduced during STZ diabetes, whereas levels of 25(OH)D3 were not, indicating that the fall in 1,25(OH)2D3 was not due to a lack of substrate. Together, these data indicate that the reduced bone formation and mineralization could account for the hypercalciuria with normocalcemia and explain, at least in part, the progressive osteopenia seen in diabetic patients. In addition or alternatively, osmotic diuresis and increased dietary calcium and carbohydrate intake secondary to hyperphagia (12) can account for the hypercalciuria. This would also explain the reduction of circulating levels of 1,25(OH)2D3 during diabetes observed in the current and previous studies (49). However, control osmotic diuresis with insulin therapy and hyperphagia with paired feeding only partly corrects the hypercalciuria (31), which indicates that changes in 1,25(OH)2D3 are specific to the diabetic condition. Further studies using a Ca2+-deprived diet will help understand the contribution of the gastrointestinal tract toward diabetic hypercalciuria.
In summary, we demonstrated that hypercalciuria and impaired bone deposition together with renal NCCT upregulation and CaR downregulation are features of experimental diabetes mellitus. It should therefore be possible to test in vivo whether pharmacologic modulators of NCCT and CaR function exert any corrective effect on diabetic hypercalciuria in either an animal model of diabetes or human patients.
|
Acknowledgments |
|---|
We dedicate this work to the memory of Dr. Hugh Garland (1948 to 1999).
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  S. Tiwari, S. Riazi, and C. A. Ecelbarger Insulin's impact on renal sodium transport and blood pressure in health, obesity, and diabetes Am J Physiol Renal Physiol, October 1, 2007; 293(4): F974 - F984. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Kanetsuna, K. Hirano, M. Nagata, M. A. Gannon, K. Takahashi, R. C. Harris, M. D. Breyer, and T. Takahashi Characterization of diabetic nephropathy in a transgenic model of hypoinsulinemic diabetes Am J Physiol Renal Physiol, December 1, 2006; 291(6): F1315 - F1322. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. M. Thrailkill, C. K. Lumpkin Jr., R. C. Bunn, S. F. Kemp, and J. L. Fowlkes Is insulin an anabolic agent in bone? Dissecting the diabetic bone for clues Am J Physiol Endocrinol Metab, November 1, 2005; 289(5): E735 - E745. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Gamba Molecular Physiology and Pathophysiology of Electroneutral Cation-Chloride Cotransporters Physiol Rev, April 1, 2005; 85(2): 423 - 493. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. M. Dvorak, A. Siddiqua, D. T. Ward, D. H. Carter, S. L. Dallas, E. F. Nemeth, and D. Riccardi Physiological changes in extracellular calcium concentration directly control osteoblast function in the absence of calciotropic hormones PNAS, April 6, 2004; 101(14): 5140 - 5145. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Song, M. A. Knepper, J. G. Verbalis, and C. A. Ecelbarger Increased renal ENaC subunit and sodium transporter abundances in streptozotocin-induced type 1 diabetes Am J Physiol Renal Physiol, December 1, 2003; 285(6): F1125 - F1137. [Abstract] [Full Text]
|
|
|
|
 |
|
 A. Iwama, N. Nishigaki, K. Nakamura, I. Imaizumi, N. Shibata, M. Yamasaki, H. Nakamura, Y. Kameyama, and Y. Kapila The Effect of High Sugar Intake on the Development of Periradicular Lesions in Rats with Type 2 Diabetes J. Dent. Res., April 1, 2003; 82(4): 322 - 325. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. T. Ward, S. J. McLarnon, and D. Riccardi Aminoglycosides Increase Intracellular Calcium Levels and ERK Activity in Proximal Tubular OK Cells Expressing the Extracellular Calcium-Sensing Receptor J. Am. Soc. Nephrol., June 1, 2002; 13(6): 1481 - 1489. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
