Background

Obese individuals have an expanded interstitium in the renal inner medulla (IM), which stains positively with periodic acid-Schiff and Alcian blue. In obese dogs, the IM is also expanded, with hyaluronan (HA) content being 2.4 times control.

Methods

We determined the anatomic pattern of renal HA deposition following weight gain, using an animal model of obesity consisting of young rabbits (N = 10), representing animals entering into the study, lean rabbits (N = 19), fed a control diet, and obese rabbits (N = 19), fed a high-fat diet (15% fat, by fortifying with corn oil and lard, in a ratio of 2:1) for two to three months. Tissue was papain digested, and HA was recovered in a phosphate or a Tris buffer and detected by an indirect immunoabsorbent competition assay.

Results

Rabbits fed a high-fat diet for 8 to 12 weeks gained weight (37%) and became mildly hypertensive (10 mm Hg). In lean rabbits, HA was low in the renal cortex (6 30 g/g tissue), increased steadily across the outer medulla (OM; 79 28 g/g tissue) and was uniformly high in the IM (192 28 g/g tissue) when recovered in a Tris buffer; these levels of tissue HA did not change during the three-month period of dietary intervention. In obese rabbits, the renal medullary interstitium was expanded and stained intensely with periodic acid Schiff and Alcian blue, and tissue HA was elevated in the IM (448 25 g/g tissue) but not the cortex (5 25 g/g tissue) or the OM (85 25 g/g tissue). The significant difference was due to those IM samples taken from the renal papilla; IM samples from the body of the kidney did not significantly differ among the lean, obese, and young rabbits.

Conclusion

The elevated renal HA associated with weight gain is limited to the IM and occurs most consistently in the papilla, which is the region of the kidney that is most vulnerable to distention caused by elevated renal interstitial hydrostatic pressure.

METHODS

Animals

The experimental protocols for this study were approved by the Institutional Animal Care and Use Committee of the University of Mississippi Medical Center and were carried out according to the Guide for the Care and Use of Laboratory Animals from the National Institutes of Health and according to the guidelines of the Animal Welfare Act.

Female New Zealand white rabbits were purchased in groups of 24 when they were approximately 12-weeks-old and when they weighed seven to eight pounds (Myrtle's Rabbitry, Thompson Station, TN, USA). They were immediately housed in a humidity- and temperature-controlled room with a 12-hour light cycle and fed 100 to 120 g of standard rabbit chow daily (#5321C; Purina, Richmond, IN, USA). After at least two weeks of acclimation, they were randomly divided in to two groups—lean and obese—or were tested and sacrificed immediately in the case of the series II young rabbits. The lean group was fed the standard rabbit chow, a complete maintenance diet for the nonlactating adult rabbit¹². The approximate chemical composition of the control diet was 2.5% fat, 16.2% protein, 50.5% carbohydrate, 14.0% fiber, and 7.3% ash; the mineral content included 1.1% calcium, 1.2% potassium, and 0.25% sodium. The obese group was fed a high-fat diet ad libitum, consisting of standard rabbit chow fortified with 15% added fat. The extra fat in the diet was corn oil and lard in the ratio of 2:1 by weight. During the final week of feeding, there were no significant weight changes in the lean rabbits (-0.05 0.05 kg/week, P > 0.05 by paired t-test), while the obese rabbits continued their average weight gain of 0.17 0.05 kg/week (N = 17, P < 0.05 by paired t-test)¹³. Studies were begun after the rabbits had been on their diets for 12 (series I) or 8 (series II) weeks, and animals were sacrificed at the rate of one per day over the next four weeks.

The HA determinations in this article were done progressively in two series of experiments. Series I rabbits provided random samples of IM and OM (8 lean and 6 obese rabbits). Series II rabbits provided renal samples taken sequentially from the surface cortex to the tip of the papilla (11 lean, 13 obese, and 10 young rabbits).

By necessity, kidneys were perfused with fixative for histologic examination in rabbits other than those chosen for chemical determinations, but were raised according to the same protocols as the series I and II rabbits.

Blood pressure

On the day of the experiments, the rabbit was brought to the laboratory and restrained in a Plexiglas holder (Plas Labs, Lansing, MI, USA). The main ear artery was cannulated using 0.25% bupivocaine as a local anesthetic. The rear and neck restraints were then loosened and the rabbit was allowed to sit quietly for 60 to 90 minutes. The blood pressure was recorded directly from the arterial line using a disposable pressure transducer¹³. The reported values for blood pressure are the mean of the values obtained during the last half hour when the rabbit was in a quiet, resting state. Subsequently, the rabbits were anesthetized with isoflurane gas and sacrificed by cardioplegic arrest with cold Euro-Collins solution.

Chemicals

Unless otherwise stated, all chemicals were from Sigma-Aldrich Chemicals (St. Louis, MO, USA). The phosphate-buffered saline (PBS) was 0.2 mol/L phosphate buffer with 0.15 mol/L NaCl, pH 6.9. All bovine serum albumin (BSA) was fatty acid free, radioimmunoassay grade (#105033; ICN Chemicals, Costa Mesa, CA, USA); BSA-PBS was PBS with 5% BSA.

Tissue preparation for histology

At time of sacrifice, the renal artery was cannulated and perfused with warmed normal saline at a pressure equal to the mean arterial pressure of that animal. Once the blood was washed from the kidney, the perfusate was switched to Zamboni's fixative (abstract; Zamboni and DeMartino, J Cell Biol 35:148A, 1967), also under the same pressure. Perfusion continued until the ureter developed a yellow coloration. The papilla was then dissected from the kidney and stored in Zamboni's fixative. The tissue samples were prepared according to standard procedures by the Clinical Pathology laboratory; the lean and obese samples were stained and mounted at the same time. Photographs were made on the same frame of film by blanking half and photographing the lean tissue and then blanking the other half and photographing the obese tissue.

Tissue preparation for HA assay

Kidneys were removed from the rabbit, allowed to drain of blood, and dissected free of retroperitoneal fat. Extrarenal vessels and fat were trimmed to the outline of the renal hilum. A coronal section was cut that included the hilum of the kidney; such a section contains the entire papilla. Two cuts were then made up through the calyces, parallel to the long axis of the papilla, yielding a renal fragment that extended from the surface cortex to the tip of the papilla. Tissue samples were taken at random from the outer medulla (OM) or inner medulla (IM) for the series I rabbits. For the series II rabbits, sequential tissue samples were taken: two from the cortex, three from the OM, and five from the IM. The edges were trimmed so as to exclude any recognizable tissue from the adjoining regions. Each piece was numbered sequentially and weighed to the nearest 0.1 mg. The tissue was then flash frozen in liquid nitrogen for later analysis.

Tissue digestion

All tissue samples were digested in papain at 60°C, and the HA was recovered by hexadecyl (cetyl) trimethyl ammonium bromide (CTAB) precipitation.

The tissue samples from the series I rabbits were approximately 50 mg and were homogenized in 20 mmol/L sodium phosphate, pH 6.8, at 4°C for four minutes (Tenbroeck homogenizer, #885000-0002; Kontes Glassware, Vineland, NJ, USA). This homogenizing buffer also contained 20 mmol/L ethylenediaminetetraacetic acid (EDTA), 50 mmol/L benzamidine, 100 mmol/L -amino caproic acid, and 0.5 mmol/L n-ethylmaleimide. The volume was brought up to 1.5 mL, 1 mL of which was transferred to a 16 125 screw top glass test tube. Papain was added (15 L or approximately 6 units; Sigma #P3125), and the mixture was incubated at 60°C for two hours in a shaking water bath, after which time the papain was denatured by heating the tube to 100°C for five minutes. The contents of the tube were then transferred to a microcentrifuge tube (1.5 mL; Eppendorf, Westbury, NY, USA) and centrifuged for three minutes at 7000 g, and the supernatant was collected for HA recovery.

The series II tissues were 20 to 50 mg in size and were homogenized in 10 mmol/L Tris buffer, pH 6.8, with added inhibitors as described previously in this article. The volume was brought up to 30 mg wet weight per mL homogenate, and 0.5 or 1.0 mL was transferred to a microcentrifuge tube (1.5 mL; Eppendorf), 15 L of papain added, and the mixture incubated at 60°C for three to four hours in a thermomixer (model 5436; Eppendorf). Samples were then centrifuged for one minute at 7000 g, and the supernatant was transferred to a new microcentrifuge tube for HA recovery. Pellets obtained from Tris buffer were more easily redissolved than those obtained from phosphate-buffered solutions and gave more reproducible results.

Hyaluronan recovery

Polyanions were precipitated from the papain digest by adding 5 to 10 mg CTAB and centrifuging at 7000 g for one minute. The pellet was washed by being resuspended and then centrifuged three times in 0.5 mL potassium thiocyanate-saturated ethanol in order to remove all CTAB. The pellets were then washed twice in ethanol alone and redissolved in 1.0 mL of distilled water for series I or 1.0 mL of BSA-PBS for series II.

Hyaluronan assay

Hyaluronan was measured according to an indirect competition assay¹⁴. The HA sample was quantitatively adsorbed with a solution that contained a slight excess of proteoglycan monomer (aggrecan), a specific HA-binding protein derived from cartilage, and then the remaining free monomer was measured immunochemically.

The following solutions were required for the assay. Proteoglycan monomer 10-times concentrated buffer (PG-10 buffer) was 5% BSA, 0.05% Tween-20, 0.05% sodium azide, 50 mmol/L phosphate, and 1.5 mol/L NaCl. Wash buffer was 0.05% Tween-20 in 0.1 mol/L phosphate, 0.1 mol/L NaCl buffer, pH 7.4; plates were washed three times by a Scanwash 200 (Molecular Devices, Sunnydale, CA, USA). Stock solutions were stored at -30°C: proteoglycan monomer (0.4 mg/mL in BSA-PBS, 0.05% Tween-20; ICN #191486), mouse antikeratan antibody (1:2000 dilution in BSA-PBS, 0.05% Tween-20, ICN #696252), horseradish peroxidase-conjugated anti-mouse antibody (HRP-anti-mouse antibody, 1:500 dilution in BSA-PBS, 0.05% Tween-20, ICN #612041); grade I, highly polymerized HA for standards (0.5 mg/mL distilled water, ICN #100735 or Sigma #H1751), grade III HA for plate coating (0.2 mg/mL distilled water, Sigma #H1504 or ICN #151268). Working solutions made daily were the BSA-PBS and tetramethyl benzamidine dihydrochloride (TMB, 0.25 mol/L in dimethylsulfoxide; Research Organics, Cleveland, OH, USA). Standards were 10 in number, made from the grade I HA stock, and were diluted with BSA-PBS in a geometric distribution from 10 ng/mL to 10 g/mL.

Hyaluronan-coated titer plates were made fresh weekly by adding 200 L of the HA coating solution (a 1:1 dilution of the grade III HA stock with 0.2 mol/L sodium carbonate, pH 9.2) to each well of a 96-well polystyrene enzyme-linked immunosorbent assay (ELISA) plate (#3590; Costar, Cambridge, MA, USA), sealing the plate (Corning #430454; Corning, NY, USA) and incubating it overnight in a refrigerator. The plate was then emptied and blocked with 200 L of BSA-PBS for six hours at room temperature, washed twice with PBS, filled with 200 L of PBS plus 0.02% sodium azide per well, sealed, and then stored refrigerated until used.

Test solutions, blanks, and standards (150 L) were added to deep well polypropylene plates (650 L capacity, 96-well format, Abgene #AB-0765; Marsh Biomedical Products, Rochester NY, USA), along with 15 L PG-10 buffer and 105 L BSA-PBS, agitated (Labline titer plate shaker, Melrose Park, IL, USA), sealed (Abgene #AB0566; Marsh Biomedical Products), refrigerated, and incubated overnight. The resulting solutions were transferred to a HA-coated titer plate, sealed, refrigerated, and incubated overnight in order to bind the remaining free monomer to the plate. The plate was then washed and filled with 200 L of mouse antikeratan antibody stock in order to label the bound monomer. After one hour of incubation at 37°C, the plate was washed, and 200 L of HRP anti-mouse antibody stock was added, incubated at 37°C, and washed. The colorimetric determination was by TMB (1:16 dilution of TMB stock in 10 mmol/L acetate buffer, pH 5.0, with 0.02% hydrogen peroxide), stopped by the addition of 20 L of 25% H₂SO₄, and the resulting optical density read at a wavelength of 450 nm (SLT Rainbow titer plate reader; Tecan U.S., Durham, NC, USA).

Statistical analyses

Values are given as mean SEM. Differences between independent observations were determined by the two-tailed unpaired Student's t-test¹⁵. P < 0.05 was accepted as significant; otherwise, the results were deemed not significant (P > 0.05). The test for multiple comparisons among the results obtained for protocols performed simultaneously was Tukey–Kramer HSD (honestly significant difference) test¹⁵. Critical values were adjusted according to Bonferrroni when more than one statistical test was performed on a data set.

RESULTS

Body weights and arterial blood pressures

The body weights of the obese rabbits were significantly greater than the lean in both series I and II, averaging 37% more Table 1. The mean arterial pressure observed in the obese rabbits was significantly greater than the lean rabbits in series I (P < 0.05 by Student's t-test; Table 2), but the difference did not reach significance when comparing lean, young, and obese rabbits in series II (P > 0.05 by Tukey–Kramer HSD test; Table 2).

Table 1 - Body weights.

Full table

Table 2 - Arterial blood pressure.

Full table

Obesity and the renal medullary extracellular matrix

The extracellular matrix of obese animals appeared expanded, with increased staining by periodic acid-Schiff (PAS) and Alcian blue, in the IM but not in the cortex or the OM. The photomicrograph of Figure 1a was taken of tissue near the tip of the renal papilla of a lean (left) and an obese (right) rabbit (100); the uroepithelium rounding the tip can be seen at the bottom of Figure 1. At this level, the PAS staining appeared more intense in the interstitial spaces. Figure 1b was photographed at a level midway along the papilla, and the Alcian blue staining highlights an expanded interstitium. Together, the increased PAS and Alcian blue staining indicate the presence of additional proteoglycan (acid mucopolysaccharide) in obese rabbits' IM.

Figure 1.

Histology of the renal medullary interstitial expansion associated with obesity. Samples of pressure-perfused kidneys were sectioned and stained for PAS (A) or Alcian blue (B). All sections were cut parallel to the long axis of the papilla, and the pairs of microphotomicrographs were obtained from equivalent positions along the axis of the papilla. The tip of the papilla (a) showed a thickened uroepithelium and more prominent PAS staining of the medullary interstitium. The tissue sections from both the lean and the obese rabbits' kidneys were from the lateral aspect of the papilla, where the collecting ducts were coursing toward the center to join the ducts of Bellini, and so appeared in cross-section. Midway up the papilla (b), the thickening of the uroepithelium of the obese rabbit was less thickened than at the tip, but the expansion of the medullary interstitial matrix was more prominent. The tissue section for the obese rabbit was taken near the centerline of the papilla, where the renal tubules were aligned in a more axial direction; the tissue section for the lean rabbit was from a more lateral site in the papilla. (Calibration bar = 100 .)

Full figure and legend (429K)

Obesity elevates inner medullary HA

Tissue HA was determined in random samples of tissue taken from the IM or OM of lean and obese rabbits (phosphate buffer used; Figure 2). The IM of obese rabbits (981 80 g/g tissue, N = 6) contained significantly more HA than the obese rabbits' OM (533 37 g/g tissue, N = 6, P < 0.05 by Tukey–Kramer HSD test) as well as more than the lean rabbits' IM or OM (618 55 and 425 83 g/g tissue, N = 8,8, P < 0.05 by Tukey–Kramer HSD test). The difference between the IM and the OM of the lean rabbits did not reach significance (P > 0.05 by Tukey–Kramer HSD test).

Figure 2.

Hyaluronan (HA) concentration of the inner medulla (IM) and outer medulla (OM) of lean () and obese () rabbits. Random samples of renal tissue were taken from either lean or obese rabbits' renal IM or OM in order to determine HA content, here plotted as g HA per gram wet weight of tissue. The IM from the obese rabbit had 59% more HA than IM from lean rabbits (P < 0.05 by Tukey–Kramer HSD test).

Full figure and legend (90K)

When the determinations made in series II rabbits were grouped by region, the obese rabbits' inner medullary HA (448 25 g/g tissue, N = 14, Tris buffer used) was again significantly greater than the obese rabbits' OM (85 25 g/g tissue, N = 14) as well as all other regions tested (P < 0.05 by Tukey–Kramer HSD test with Bonferroni's correction). As with series I, the series II lean rabbits' IM HA (192 28 g/g tissue, N = 11) was not significantly greater than the OM (79 28 g/g tissue, N = 11), but did differ from cortical HA in lean, young, and obese rabbits (6 30, 2 36, and 5 25 g/g tissue, N = 10, 7, and 14, respectively, P < 0.05 by Tukey–Kramer HSD test with Bonferroni's correction).

HA is most consistently elevated in the papilla

Tissue HA was next determined in ten tissue samples taken sequentially from the surface cortex to the tip of the papilla. Samples taken from the young rabbits did not differ from the respective samples from the lean controls (P > 0.05 by Tukey–Kramer HSD test with Bonferroni's correction), indicating that no significant differences developed during the 8 to 12 weeks during which the rabbits were maintained on their control diet.

Only three of the ten kidney samples from the obese rabbits differed significantly from the lean and young counterparts: the final three locations, closest to the tip of the papilla (P < 0.05 by Tukey–Kramer HSD test for young vs. lean vs. obese, with Bonferroni's correction; Figure 3).

Figure 3.

Renal hyaluronan (HA) distribution. Ten renal samples were taken in order, from the capsular edge of the renal cortex to the tip of the renal papilla, with two pieces from capsule, three from OM, and five from IM, in order to determine HA content of these three regions of the kidney. Eleven lean adults (), 14 obese adults (), and 7 young () rabbits were used in the study. Pair-wise comparisons of the HA content in each of the 10 locations showed significant differences between the obese tissues and the lean and young samples only in the three sites nearest the tip of the papilla (*P < 0.05 by Tukey-Kramer HSD test, with Bonferroni's correction). In addition, the obese samples differed from the lean, but not the young, in the IM site one distant from the OM (†P < 0.05 by Tukey-Kramer HSD test, with Bonferroni's correction).

Full figure and legend (41K)

Measurements of IM and papilla were made from camera lucida tracings of coronal sections of the kidney. The extent of the IM was measured along the center axis that extended from the outer edge of the cortex to the tip of the papilla and was compared with the length of the papilla. In tissue from lean rabbits, 67 4% of the IM was in the papilla, and in tissue from obese rabbits, 64 2% of the IM was in the papilla (P > 0.05 by t-test, N = 3, 2). Thus, the three tissue locations that were closest to the papillary tip and that showed the most consistent increase in tissue HA were situated in the renal papilla.

DISCUSSION

A weight gain of 50% results in a fivefold increased risk of hypertension in humans¹⁶. The obese female rabbits used in this study showed a 37% weight gain and did have a significant increase in blood pressure (13%) when simply compared with the lean controls. This modest degree of hypertension is similar to that for a population of humans who are approximately 15% overweight and who are thus at increased risk for cardiovascular disease¹⁷. This hypertension is present even though the lipid profile of the obese rabbit is not identical to the obese human, as the obese rabbit is not hypercholesterolemic. However, many other traits are shared with young obese humans, individuals in whom diabetes and marked kidney damage has not yet occurred. For instance, rabbits are insulin resistant, with elevated fasting plasma insulin and elevated plasma triglycerides. The animals also have an elevated plasma renin activity, increased hematocrit, increased plasma norepinephrine, and increased renal blood flow¹³.

In healthy kidneys, HA is limited to the medulla^18,19, being very low in the cortex²⁰. In this study, the cortex was uniformly low in HA in all groups of animals. HA content increased across the width of the OM, and thus, measurements of HA in the OM naturally have a large variability, depending on which side of the OM the sample is taken. The IM of the lean and young rabbits was uniformly rich in HA.

Obese rabbits had elevated HA in the IM, relative to the lean controls, averaging an increase of 84 and 133% for series I and II, which is similar to the increase of 140% reported in a model of canine obesity⁸. Histologically, the extracellular matrix was expanded, with increased PAS and Alcian blue staining, consistent with the quantitative increase in HA and with the histologic changes observed in obese humans¹⁰ and in a canine model of obesity (abstract; Hall et al, Hypertension 18:395, 1991)²¹.

The most consistent increase in obese rabbits' kidney HA was found in the renal papilla. When individual renal samples from the obese rabbits were compared with their anatomic counterparts from lean and young animals, the only significant differences found were in the distal three samples, in the region of the IM that was situated in the papilla. This statistical observation reflects the larger variance of the HA determination in the IM samples taken in the region adjacent to the OM, samples that were in the body of the kidney. Thus, it is the renal papilla that is most susceptible to the stimuli that result in interstitial HA accretion.

Renal HA

The normal pattern of HA distribution can be altered by pathological events that elevate cortical HA but leave renal medullary HA unchanged. Ischemia caused by a 30- or 60-minute ligation of the renal artery resulted in a 20- to 40-fold elevation in cortex²². Post-transplantation reactions found in allogenic kidney grafts performed in rats were associated with 40-fold elevation of HA in the cortex and elevated HA in the OM²³. In both models, the HA content of the IM did not change significantly.

Conversely, medullary HA can change independently from cortical HA. For instance, unilateral ureteral ligation causes a specific and prompt increase in medullary HA in both the ligated kidney and the contralateral, control kidney in rats²⁴. Hydration status also alters HA content, with water-replete rats having more renal medullary HA than water-deprived rats^25,26.

Three potential mechanisms could contribute to the increase in renal HA content: increased HA synthesis, increased HA-proteins, and decreased HA degradation.

HA synthesis.

Hyaluronan is synthesized by renal medullary interstitial cells²⁷ and is under physiological control. An early study demonstrated that antidiuretic hormone stimulates the incorporation of the radiolabeled precursor into mucopolysaccharides within the extracellular matrix of rabbit renal medullae²⁸. More recently, a glycoprotein has been identified that is secreted by fetal kidney, which stimulates the production of HA by renal medullary interstitial cells in culture. The same glycoprotein is secreted by Wilms' tumors, a renal tumor that is rich in HA and that is associated with high levels of serum and urinary HA²⁹. However, the contribution of this mechanism is not known because to our knowledge no studies have measured renal HA synthesis in obesity.

HA binding proteins.

The HA present in the renal papilla is not in solution, but is in the insoluble particulate residue remaining after homogenization⁸. Two classes of HA binding proteins (HBP or hyaladherins) have been described: The first is the Link module superfamily, characterized by two -helices, two triple-stranded antiparallel -sheets, and a very large hydrophobic core, as exemplified by CD44, a membrane HA receptor, and aggrecan and versican, two closely related extracellular matrix proteins³⁰. The second family is characterized by a basic amino acid motif, best exemplified by RAMM (receptor for HA-mediated motility), a cytoplasmic and membrane protein³¹. Such HA binding proteins typically have a high affinity for HA (kD of 10^-9 mol/L), and renal HA is tightly bound to the matrix in both normal and obese animals, being insoluble in normal saline buffer as well as 3 mol/L guanidine HCl⁸. Therefore, renal medullary HBP either exhibits an excess of HA-binding capacity, or the renal medulla increases in HBP content during the development of experimental obesity.

HA degradation by resident enzymes.

Hyaluronan digestion is also under physiological control. A single gene product, Hyal-1, is the source of the hyaluronidase activity found in the renal parenchyma, serum, and urine³². While the resulting proteins show genetic polymorphism³³ as well as post-translational modifications³⁴, all forms are highly lipid soluble and are active under acid conditions, potentially explaining the molecule's transport across the tubular wall and its enzymatic activity in the urine.

Antidiuretic hormone (ADH) releases vesicular stores of hyaluronidase from rat medullary cells, along with two other glycan hydrolases: -glucuronidase and N-acetyl--D-glucosaminidase³⁵. Ginetzinsky initially proposed that such a hyaluronidase was released from the renal medulla and was the source of the urinary hyaluronidase observed during antidiuresis as well as the cause of the reduction in renal medullary mucopolysaccharides observed in hydropenic states²⁵. The activity of this hyaluronidase may have a functional consequence, since antiserum against renal hyaluronidase blunts the antidiuretic response of rats to water restriction or arginine vasopressin, both in the rate of urine excretion and the maximum concentrating ability of the kidneys²⁶, as well as reduces the degree to which the lateral spaces in the medullary collecting duct dilate during water restriction³⁶. Thus, while renal medullary HA is regulated during changes in hydration in part by the release of catabolic enzymes, the activity of renal hyaluronidase during the development of obesity is not known.

HA and obesity-related hypertension

Renovascular dynamics: HA deposition in response to injury.

The elevation in medullary HA follows ureteral ligation²⁴ and occurs under circumstances similar to obesity, as there is some degree of renal outflow obstruction in both cases. Ureteral ligation totally blocks urine outflow and almost immediately causes filtration to cease as renal hydrostatic pressures increase. The bulk of the kidney parenchyma is a solid mass, surrounded by a tough fibrous capsule that resists expansion. However, the renal papilla is unique to the kidney because it is adjacent to the renal calyces, blood, and lymphatic vessels, along with their associated adipose and connective tissue. Because the contents of the calyces and vessels can be displaced, the papilla is particularly vulnerable to volume expansion caused by an elevation in the interstitial hydrostatic pressure. Thus, the increased HA may be a local response to the damage caused by overexpansion of this tissue.

During the development of obesity, urine, lymphatic, and venous outflow from the kidney can be increasingly obstructed by the accretion of adipose tissue about the kidney²¹ and by the deposition of fat within the renal sinus itself³⁷. These alterations may be a cause of the very high interstitial hydrostatic pressure that accompanies obesity²¹, and this increase in interstitial hydrostatic pressure may be sufficient to expand the papilla and cause damage to the renal medulla⁷, as in the case of ureteral ligation.

However, renal medullary HA is also elevated in the contralateral kidney following unilateral ureteral ligation, implicating a mechanism independent of the elevated renal hydrostatic pressure in the ligated kidney. One such change is the consequent hyperfiltration in the unobstructed kidney after unilateral ureteral ligation. Increases in renal blood flow and glomerular filtration rate are also seen in canine and rabbit models of obesity^7,38, and thus may be a cause of the elevated renal medullary HA in these two animal models. Alternatively, a neural or humoral signal may be a mechanism for the elevated renal medullary HA.

Physical forces: HA is a space-filling molecule.

The anatomic consequence of the obesity-related elevation in renal medullary HA is that the renal medullary interstitium is greatly expanded, a change that has the potential to slow the flow of blood and urine through the papilla, contributing to the increased tubular sodium reabsorption characteristic of obesity hypertension. Thus, along with activation of the sympathetic nervous system and the renin-angiotensin system, an expanded renal medullary interstitium may contribute to the shift of the renal pressure natriuresis curve that underlies the hypertension of obesity⁷.

Hypertension: A vicious cycle.

To the extent that both causal circumstances hold true, the increased HA would alter renovascular dynamics in such a way as to elevate systemic blood pressure. In turn, the elevated blood pressure would lead to increased papillary dilation and additional HA deposition. This self-reinforcing cycle could be a potent force to overcome homeostatic adjustments and could result in long-term hypertension.

Summary

Hyaluronan, a large, space-filling molecule, is specifically elevated in the renal papilla when rabbits become obese. This additional glycosaminoglycan content expands the volume of the extracellular matrix and may alter renal fluid dynamics in such a way so as to contribute to the arterial hypertension associated with weight gain.

References

1. Van Itallie, TB: Health implications of overweight and obesity in the United States. Ann Intern Med 1985 103: 983–988,  | PubMed | ChemPort |
2. Dyer, AR, Elliott, P: The INTERSALT study: Relations of body mass index to blood pressure. J Hum Hypertens 1989 3: 299–308,  | PubMed | ISI | ChemPort |
3. Kaplan, N: The deadly quartet: Upper body obesity, glucose intolerance, hypertriglyceridemia, and hypertension. Arch Intern Med 1989 149: 1514–1520,  | Article | PubMed | ISI | ChemPort |
4. Rexrode, KM, Hennekens, CH, Willett, WC, Colditz, GA, Stampfer, MJ, Rich-Edwards, JW, Speizer, FE, Manson, JE: A prospective study of body mass index, weight change, and risk of stroke in women. JAMA 1997 277: 1539–1545,  | Article | PubMed | ISI | ChemPort |
5. Weindruch, R, Sohal, RS: Caloric intake and aging. N Engl J Med 1997 337: 986–994,  | Article | PubMed | ISI | ChemPort |
6. The Trials of Hypertension Prevention Collaborative Research Group: Effects of weight loss and sodium reduction intervention on blood pressure and hypertension incidence in overweight people with high-normal blood pressure. Arch Intern Med 1997 157: 657–667,
7. Hall, JE, Brands, MW, Henegar, JR, Shek, EW: Abnormal kidney function as a cause and a consequence of obesity hypertension. Clin Exp Pharmacol Physiol 1998 25: 58–64,  | PubMed | ISI | ChemPort |
8. Alonso-Galicia, M, Dwyer, TM, Herrera, GA, Hall, JE: Increased hyaluronic acid in the inner renal medulla of obese dogs. Hypertension 1995 4: 888–892,
9. Guyton, AC: The surprising kidney–fluid mechanism for pressure control—its infinite gain. Hypertension 1990 16: 725–730,  | PubMed | ISI | ChemPort |
10. Deleted in proof
11. Banks, SA: The Effects of Obesity on the Extracellular Matrix of the Kidney. 1997 Masters Thesis, Department of Biology, Jackson State University, Jackson, Mississippi,
12. Harkness, JE: Rabbit husbandry and medicine. Vet Clin North Am 1987 17: 1019–1044,  | ISI | ChemPort |
13. Carroll, JF, Dwyer, TM, Grady, AW, Reinhart, GA, Montani, JP, Cockrell, K, Meydrech, EF, Mizelle, HL: Hypertension, cardiac hypertrophy, and neurohumoral activity in a new animal model of obesity. Am J Physiol 1996 271: H373–H378,  | PubMed | ChemPort |
14. Zar, JH: Biostatistical Analysis 1999 (4th ed). Englewood Cliffs, Simon and Schuster,
15. Goldberg, RL: Enzyme-linked immunosorbent assay for hyaluronate using cartilage proteoglycan and an antibody to keratan sulfate. Anal Biochem 1988 174: 448–458,  | Article | PubMed | ISI | ChemPort |
16. Huang, Z, Willett, WC, Manson, JE, Rosner, B, Stampfer, MJ, Speizer, FE, Coldiz, GA: Body weight, weight change and risk for hypertension in women. Ann Intern Med 1998 128: 81–88,  | PubMed | ISI | ChemPort |
17. Ashley, FW, Kannel, WB: Relationship of weight changes to changes in atherogenic traits: The Framingham study. J Chronic Dis 1974 27: 103–114,  | Article | PubMed | ISI |
18. Farber, SJ, Vanpraag, D: Composition of glycosaminoglycans (mucopolysaccharides) in rabbit renal papillae. Biochem Biophys Acta 1970 208: 219–226,  | PubMed | ChemPort |
19. Kresse, H, Grossmann, A: Vergleichende Untersuchung des Mucopolysaccharid-und Kollagengehaltes in verschiedenen topographischen Zonen der Niere von Ratte, Hund und Schwein. Z Klin Chem Klin Biochem 1970 8: 420–424,  | PubMed | ISI | ChemPort |
20. Wells, AF, Larsson, E, Tengblad, A, Fellström, B, Tufvesonn, G, Klareskog, L, Laurent, TC: The localization of hyaluronan in normal and rejected human kidneys. Transplantation 1990 50: 240–243,  | PubMed | ISI | ChemPort |
21. Deleted in proof
22. Johnsson, C, Tufveson, G, Wahlberg, J, Hällgren, R: Experimentally-induced warm renal ischemia induces cortical accumulation of hyaluronan in the kidney. Kidney Int 1996 50: 1224–1229,  | PubMed | ISI | ChemPort |
23. Hällgren, R, Gerdin, B, Tufveson, G: Hyaluronic acid accumulation and redisposition in rejecting rat kidney graft: Relationship to the transplantation edema. J Exp Med 1990 171: 2063–2076,  | Article | PubMed | ISI | ChemPort |
24. Johnsson, C, Hällgren, R, Wahlberg, J, Tufveson, G: Renal accumulation and distribution of hyaluronan after ureteral obstruction. Scand J Urol Nephrol 1997 31: 327–331,  | PubMed | ISI | ChemPort |
25. Ginetzinsky, AG: Role of hyaluronidase in the reabsorption of water in renal tubules: The mechanism of action of the antidiuretic hormone. Nature 1958 182: 1218–1219,  | PubMed | ISI | ChemPort |
26. Law, RO, Rowen, D: The influence of hyaluronidase on urinary and renal medullary composition following antidiuretic stimulus in the rat. J Physiol (Lond) 1981 311: 341–354,  | PubMed | ISI | ChemPort |
27. Pitcock, JA, Lyons, H, Rightsel, PS, Muirhead, EE: Glycosaminoglycans of the rat renomedullary interstitium: Ultrastructural and biochemical observations. Exp Mol Pathol 1988 49: 373–387,  | Article | PubMed | ISI | ChemPort |
28. Darnton, SJ: The conversion of injected glucose into renal glycogen and mucopolysaccharides: An autoradiographic study of rabbits in various states of hydration. Z Zellforsch 1969 102: 273–282,  | Article | PubMed | ISI | ChemPort |
29. Longaker, MT, Adzick, NS, Sadigh, D, Hendlin, B, Stair, SE, Duncan, BW, Harrison, MR, Spendlove, R, Stern, R: Hyaluronic acid-stimulating activity in the pathophysiology of Wilms' tumors. J Natl Cancer Inst 1990 17: 973–974,
30. Day, AJ: The structure and regulation of hyaluronan-binding proteins. Biochem Soc Trans 1999 27: 115–121,  | PubMed | ISI | ChemPort |
31. Cheung, W-F, Cruz, TF, Turley, EA: Receptor for hyaluronan-mediated motility (RHAMM), a hyaladherin that regulates cell responses to growth factors. Biochem Soc Trans 1999 27: 135–142,  | PubMed | ISI | ChemPort |
32. Frost, GI, Csoka, TB, Wong, T, Stern, R: Purification, cloning and expression of human plasma hyaluronidase. Biochem Biophys Res Commun 1997 236: 10–15,  | Article | PubMed | ISI | ChemPort |
33. Fiszer-Szafarz, E, Demaeyer, B: Hyal-1, a locus determining serum hyaluronidase polymorphism, on chromosome 9 in mice. Somat Cell Mol Genet 1989 15: 79–83,  | Article | PubMed |
34. Csoka, TB, Frost, GI, Wong, T, Stern, R: Purification and microsequencing of hyaluronidase isozymes from human urine. FEBS Lett 1997 417: 307–310,  | PubMed | ISI | ChemPort |
35. Nikiforovskaia, LF, Tischenko, NI, Ivanova, LN: The glycosaminoglycanohydrolases of the rat kidney tissue lacking antidiuretic hormone. Sechenov Physiol J USSR 1987 73: 978–985,  | ChemPort |
36. Rowen, D, Law, RO: The effect of antidiuretic stimuli on the morphology of the lateral intercellular spaces in the medullary collecting duct of the rat. J Anat 1981 133: 197–203,  | PubMed | ISI | ChemPort |
37. Dwyer, TM, Mizelle, HL, Cockrell, K, Diveki, K, Buhner, P: Renal sinus lipomatosis and body composition in hypertensive, obese rabbits. Int J Obes 1995 19: 869–874,  | ISI | ChemPort |
38. Carroll, JF, Huang, M, Hester, RL, Cockrell, K, Mizelle, HL: Hemodynamic alterations in obesity-associated hypertension in a new rabbit model of obesity. Hypertension 1995 26: 465–470,  | PubMed | ISI | ChemPort |

Acknowledgments

This study was supported by Mississippi chapter of the American Heart Association and National Institutes of Health (HL-51971). We thank Dr. H. Leland Mizelle for tissue samples and for his support and numerous discussions. We thank Mr. Lester Donald for his skilled technical assistance. We are grateful to Drs. Joey Granger, Michael Hughson, Jeffrey Henegar, and Marcy Petrini for reading earlier versions of the manuscript.