The arachidonate signaling pathways comprise prostanoids formed by cyclooxygenases, EETs, and HETEs formed by cytochrome P-450 (CYP) enzymes and HETEs and leukotrienes generated by lipoxygenases. Whereas the intrarenal localization of cyclooxygenases and of some CYP enzymes along the nephron has already been determined, the localization of lipoxygenases and leukotriene-forming enzymes together with leukotriene receptors in the kidney is less clear. This study therefore aimed to determine the expression of 5-, 12-, and 15-lipoxygenases as well as the leukotriene receptors along the rat nephron. The kidneys were dissected into cortex and outer and inner medulla, and the microdissected nephron segments were collected after a collagenase digestion. mRNA abundance was determined by RT-PCR and real-time PCR. 15-LOX mRNA showed a characteristic expression pattern along the distal nephron. 12-LOX mRNA was only found in the glomerulus. Similarly, 5-LOX mRNAs together with 5-LOX-activating protein mRNAs were expressed in the glomerulus and also in the vasa recta. The leukotriene A4 hydrolase was found in all nephron segments, whereas leukotriene C4 synthase mRNA could not be found in any nephron segment. The leukotriene receptor B4 and the cysteinyl leukotriene receptor type 1 were selectively expressed in the glomerulus, whereas cysteinyl receptor type 2 was not found in any nephron segment. Our data suggest that the glomerulus is a major source and target for 5- and 12-HETE and for leukotrienes. The collecting duct system, on the other hand, appears to be a major source of 15-HETE.
- nephron segments
arachidonic acid cleaved from phospholipids by phospholipases can be oxidized along three main pathways, namely, cyclooxygenase, lipoxygenase, and cytochrome P-450 (CYP) monoxygenase pathways, all of which appear to be relevant for kidney function. The intrarenal expression sites of cyclooxygenases have already been extensively studied, and there is agreement that cyclooxygenase 1 is expressed in blood vessels including glomeruli and to a major extent in the collecting duct system. Cyclooxygenase 2 is also found in blood vessels, in medullary interstitial cells, and in the macula densa cells of the distal tubule (reviewed in Ref. 16). From the CYP monoxygenase pathway, the intrarenal localization of CYP4A enzymes giving rise to 20-HETE has already been characterized. It turned out that CYP4A2, -A3, and -A8 are rather ubiquitously expressed along the nephron, whereas CYP4A1 was not found in any nephron segment (10). The intrarenal distribution of other enzymes of the CYP monoxygenase pathway family has not yet been determined.
Similarly, the expression of lipoxygenases in the kidney is only poorly known. There exist three classic lipoxygenases, which form 5-, 12-, and 15-HETE. Lipoxygenase-5 can also generate leukotriene A4 (LTA4) in the presence of the 5-LOX-activating protein (FLAP). LTA4, in turn, is the substrate of the LTA4 hydrolase, forming LTB4 binding to LTB4 receptors, and of the leukotriene C4 synthase, forming the so-called cysteinyl (cys) leukotrienes (LTC4, LTD4, and LTE4) binding to cysteinyl leukotriene receptors.
From this group of enzymes and receptors, only the intrarenal distribution of LTA4 hydrolase has so far been determined, which was found to be expressed along the whole rat nephron (23). In vitro data, moreover, suggest that mesangial cells and podocytes can express 12-LOX mRNA (13).
Because HETES derived from lipoxygenase activity and leukotrienes are locally acting mediators, their intrarenal site of production probably also reflects their regions of action. Because HETE and leukotrienes exert a number of effects in the kidney, such as regulation of glomerular filtration or renal vasoconstriction (2), it appeared of interest to us to localize the key elements of the lipoxygenase/leukotriene effector group within the kidney.
Microdissection of nephron segments.
Male Sprague-Dawley rats, weighing 200–260 g, having free access to tap water and standard commercial pellet chow (Altromin C1000, Lage), were used. All animal experiments were conducted in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and German laws on animal protection. The animals were anesthetized, decapitated, and both kidneys were removed. They were immediately cooled in 4°C cold 1× PBS solution, followed by decapsulation and dissection into cortex and outer and inner medulla. These zones were sliced along the tubular direction into pieces of ∼2-mm width. According to a modified collagenase digestion protocol from Schafer et al. (31), the obtained pieces were incubated with 2 ml collagenase solution (250 U/ml type 2 collagenase, Worthington; 50 U/ml DNAse, Sigma; 48 μg/ml soybean trypsin inhibitor, Sigma; and 5 mmol/l glycin, Merck, in MEM, Biochrom). After 20 (cortex/outer medulla) and 25 min (inner medulla), the supernatant was removed and added to a 2-ml BSA (PAA-Laboratories) solution (1% BSA in MEM). The remaining kidney pieces were incubated with fresh collagenase solution, and this procedure was repeated 10–14 times.
After sedimentation for 10 min on ice, the supernatant was removed again and the nephron segments were resuspended in 2 ml BSA. At least 11 mm of each segment and 22 glomeruli were collected on ice in tissue culture dishes. From the cortex we collected glomeruli with (Glo+) and without arterioles (Glo−), proximal and distal convoluted tubules (PCT/DCT), cortical thick ascending limbs (cTAL), and connecting and cortical collecting ducts (CCD). In the outer medulla, we differentiated between proximal straight tubules (PST), descending thin limbs (dTL), medullary thick ascending limbs (mTAL), outer medullary collecting ducts (OMCD), and outer medullary vasa recta (OMVR). In the inner medulla, we collected descending/ascending thin limbs of Henle (TL) and inner medullary collecting ducts (IMCD). These segments were identified by their morphological appearance, and contamination was excluded by marker primers (Table 1, Fig. 1) as shown in the study by Vitzthum et al. (35). Yeast tRNA (12 μg) was added as the carrier, and the collected tubules were stored in 400 μl guanidine thiocyanate solution (4 mol/l) at −80°C until RNA extraction.
Isolation of total RNA and RT-PCR.
RNA extraction was performed according to the protocol of Chomczynski and Sacchi (4). The resulting RNA pellets were dissolved in 9 μl diethylpyrocarbonate-treated water and used for RT. RNA samples from all the collected 11-mm tubules and the 22 glomeruli were diluted to a final concentration of 1-mm tubules/2 μl and 2 glomeruli/2 μl.
PCR products were analyzed after separation by ethidium bromide (Sigma)-stained agarose (Cambrex) gel electrophoresis, and Bio 1D software (Vilber Lourmat Biotechnology, Wasserburg, Germany) was used to capture a digital image of the PCR products. PCRs were performed in a total volume of 20 μl. Either 2 μl cDNA, equivalent to an 1-mm tubule, and dissolutions of cDNA (1:10/1:100) or water and yeast tRNA served as positive or negative controls. Standard PCR protocols were used: denaturing for 30 s at 94°C, annealing at 60°C for 30 s, and elongating at 72°C for 1 min. Actin was run at 30 cycles, and only collections presenting a sufficient actin signal were used. On average three to five collections of nephron segments and glomeruli were tested. The primers used for PCR amplification are listed in Tables 1 and 2. To detect even low levels of LOX/leukotriene expression, these PCRs were run at 36 cycles.
Real-time PCR analysis.
Two micrograms of whole kidney total RNA/dissected kidney zone RNA (cortex, outer medulla, and inner medulla) or the resulting RNA pellets of microdissected tubules were reverse transcribed into cDNA (20 μl) according to standard protocols. In brief, cDNA probes were synthesized in a 20-μl reaction with total RNA, 0.5 μg oligo(dT)12-18 (Sigma), 20 U RNasin (Promega), 4 μl 5× RT buffer, 0.5 mmol/l deoxynucleoside triphosphate (dNTP, Amersham), and 20 U murine Moloney leukemia virus (M-MLV) RT enzyme (GIBCO Life Technologies).
Real-time PCR was performed with a Light Cycler DNA Master SYBR Green I kit (Roche Molecular Biochemicals). Each reaction (20 μl) contained 2 μl cDNA, 3.0 mmol/l MgCl2, 1 pmol of each primer, and 2 μl of Fast Starter Mix (containing buffer, dNTPs, SYBR Green dye, and Taq polymerase). The amplification program consisted of 1 cycle of 95°C with a 10-min hold (“hot start”) followed by 40 cycles of 95°C with a 15-s hold, an annealing temperature of 60°C with a 5-s hold, and 72°C with a 20-s hold. Amplification was followed by a melting curve analysis to verify the accuracy of the amplicon. A negative control with tRNA or with water instead of cDNA was run with every PCR to assess specificity of the reaction. An analysis of the data was performed using Light Cycler software, version 3.5.3. Additionally, a melting curve analysis was carried out. Standard curves for the LOX/leukotriene primers were generated by using cDNA of rat whole kidney total RNA as a template. For every zone, the ratio of the amount of the LOX/leukotriene mRNAs to that of β-actin mRNA was calculated. The exact nephron-specific distribution was determined by RT-PCR and PCR of the mRNA samples of the microdissected nephron segments.
Zonal distribution of 5-, 12-, 15-LOX, FLAP, LTA4 hydrolase, LTC4 synthase, LTB4 receptor, cys-LT1, and cys-LT2 mRNA.
The intrarenal expression of the enzymes and receptors of the lipoxygenase/leukotriene pathway was semiquantitatively analyzed by real-time PCR, using total mRNA isolated from the cortex, outer medulla, and inner medulla of two rat kidneys from two different animals.
5-LOX and FLAP expression in the cortex, outer medulla, and inner medulla was at about the same level in all three zones, but the expression of FLAP mRNAs was higher than 5-LOX expression (Table 3). 12-LOX mRNA abundance was equally low in the cortex, outer medulla, and inner medulla. 15-LOX showed rising abundance from the cortex to the inner medulla. LTA4 hydrolase mRNA expression was remarkably higher than LTC4 synthase mRNA expression, which even decreased from the cortex to the inner medulla.
In view of the leukotriene receptors, the expression of the LTB4 receptor was the strongest, followed by the cys-LT1 receptor, and cys-LT2 receptor expression was the lowest (Table 3). Only the cys-LT2 receptor showed a slightly higher expression in the inner medulla than in the outer medulla and cortex, whereas cys-LT1 and LTB4 distribution showed no zonal preference.
Distribution of 5-, 12-, 15-LOX, FLAP, LTA4 hydrolase, LTC4 synthase, LTB4 receptor, cys-LT1, and cys-LT2 receptor mRNAs in microdissected nephron segments.
The gene expression of enzymes and receptors along the rat nephron was investigated by RT-PCR. For each product, we assayed mRNA expression in at least three pools of nephron segments, each collected from a different animal.
Localization of 15-LOX mRNA.
We found weak expression of 15-LOX mRNA in the distal nephron (Fig. 2A), beginning in the cTAL and the DCT, and strong expression along the whole collecting duct (CCD, OMCD, IMCD) (see Fig. 9). Outer medullary vasa recta did not express significant amounts of 15-LOX mRNA (Fig. 2B).
Localization of 12-LOX mRNA.
We observed 12-LOX mRNA expression in microdissected nephron segments only in the glomerulus (Fig. 3 and see Fig. 9). The same results were obtained with glomeruli that had been flushed with saline before isolation to remove trapped blood cells.
Localization of 5-LOX and of FLAP mRNA.
For 5-LOX, we found mRNA signals in the glomerulus and in vasa recta (Fig. 4).
The mRNA for 5-LOX-activating protein showed the same expression pattern as the mRNA for 5-LOX in microdissected nephron segments, with additional weak expression along the distal nephron from the mTAL to OMCD (Fig. 5).
Localization of LTA4 hydrolase and LTC4 synthase mRNA.
LTA4 hydrolase mRNA showed high abundance in all nephron segments (Fig. 6) except the outer medullary vasa recta, where we found weaker expression. LTC4 synthase mRNA was detected in whole kidney, but not in any of the microdissected nephron segments (Fig. 7).
Localization of LTB4, cys-LT1, and cys-LT2 mRNA.
The mRNAs for cys-LT1 and LTB4 receptors were highly abundant in flushed and nonflushed glomeruli (Fig. 3 and see Fig. 9). We observed varying signals for the LTB4 receptor along the distal nephron, which we considered to be nonspecific as they were not always reproducible. mRNA for the cys-LT2 receptor was not found in any microdissected nephron segment and liver but at (as a positive control) a low level in the whole kidney and brain and at higher levels in heart and lung, according to published data (Fig. 8).
We found 15-LOX mRNAs along the distal nephron, with increasing abundance from the mTAL to CD. These findings are in accordance with increasing mRNA expression from the cortex to the inner medulla in kidney zones without microdissection, probably reflecting the relatively high density of collecting ducts in the inner medulla (see Fig. 10). Small vessels adhering to the collecting duct might have contributed to this expression pattern. We analyzed four different collections of outer medullary vasa recta to assess the expression and possibly exclude contamination of the collecting duct (CCD, OMCD, IMCD). We found no 15-LOX mRNA expression in the OMVR, so contamination of adhering vessels could not be excluded, but localization in the collecting duct was confirmed. Until now, the expression of 15-LOX was observed in blood, endothelium, and lung (22). An inhibitory role in renin release was described for 15-LOX products (1), and Munger et al. (21) observed preserved renal function in experimental anti-glomerular basement membrane antibody-mediated glomerulonephritis due to stimulated 15-LOX-levels. These effects could be mediated by 15-HETE binding to peroxisome proliferator-activated receptor γ (9), which was found to be widely expressed from mesangial cells to medullary collecting ducts (30). From our observation of high expression of 15-LOX in the collecting duct and the data from Munger et al. (21), we deduce the hypothesis that 15-LOX either might exert a cytoprotective effect in the kidney in the presence of high concentrations of possibly harmful substances in the urine or might antagonize proinflammatory effects of 5-LOX (28) in the distal nephron.
We observed 12-LOX mRNAs in microdissected nephron segments only in the glomerulus, but in no tubular compartment (Fig. 9). These results were reproducible with glomeruli that were microdissected after the kidney was flushed with 0.9% NaCl solution. However, mRNA abundance was about the same in the cortex, outer medulla, and inner medulla. Therefore, the mRNA we detected in the kidney zones presumably originates from other cells like macrophages, endothelial cells, or smooth muscle cells, where it already had been detected by other groups (7, 14, 15, 24). Podocytes were shown to express 12-LOX mRNAs (13). Matrix protein formation in mesangial cells, stimulated by high glucose, is mediated by 12-LOX (26). Matrix protein production in other renal/interstitial cells could be influenced by 12-LOX as well.
Oyekan et al. (25) have shown that ANG II-induced vasoconstriction may be mediated, in part, by 12-LOX, and increased levels of 12-HETE, which were measured in the plasma of spontaneously hypertensive rats, are thought to enhance the responsiveness of vascular smooth muscle cells to ANG II (29). Our observation that 12-LOX is highly abundant only in the glomerulus is in accordance with the vasoconstrictory action of 12-LOX in the glomerulus and the vasculature, thereby decreasing renal blood flow and glomerular filtration rate.
5-LOX and FLAP.
We observed the expression of 5-LOX and FLAP mRNAs mainly in the glomerulus and in the OMVR (Fig. 9) and weak expression of FLAP that we could not always reproduce along the distal nephron. 5-LOX mRNAs have been found in leukocytes (36), dendritic cells (8), aortic smooth muscle cells (18), and renal interstitial cells (20), which is in accordance with our results. The expression of FLAP was demonstrated in glomeruli (34), but FLAP expression independent of 5-LOX was also observed in monoclonal T cells and myeloid cell lines (5). An independent role for FLAP in protein handling was proposed by Valdivielso et al. (34), and this might be reflected by the overall higher expression level of FLAP in the kidney zones (Table 3) and the weak expression of FLAP along the distal nephron (Fig. 5).
A regulation of renal blood flow and glomerular filtration rate has been reported for 5-LOX products; i.e., 5-LOX products might impair renal function (6, 37). Furthermore, 5-LOX products are considered to be proinflammatory in glomerulonephritis (3). In addition, leukocytes could easily be attracted from the glomerulus and mediate glomerular but also tubulointerstitial inflammation, as the mediation of inflammation is probably the most thoroughly investigated aspect of 5-LOX products (27).
The leukotriene receptors cys-LT1 and LTB4 that mediate effects of 5-LOX products also showed marked abundance in the glomerulus (Fig. 9). Again, in mRNA that we isolated from kidney zones, the cys-LT1 and LTB4 receptors were almost equally abundant in all three zones. We presume that either large vessels, endothelium, local immunocompetent, or interstitial cells are responsible for this observation. These results reinforce the putative role of 5-LOX products in glomeruli. In contrast to cys-LT1 and LTB4 receptors, we found no cys-LT2 in microdissected renal tissue, suggesting that cys-LT2 is not involved in leukotriene signaling in glomeruli or tubuli. However, we observed mRNA expression of the cys-LT2 receptor in the heart and lung at higher levels, a weaker expression in the brain and kidney, and no expression in the liver. Takasaki et al. (33) found the human cys-LT2 receptor in embryonic kidney cells, heart, and leukocytes, but only at low levels in the lung and not at all in liver tissue. Our results differ mainly concerning the higher expression in the lung. Maybe we detected cys-LT2 mRNAs from leukocytes because the lungs were not flushed before they were removed or they were from the pulmonary vein, where the cys-LT2 receptor had been detected by Labat et al. (17). On the other hand, sheep trachealis muscle seems to express cys-LT2 (12), and our results could reflect these findings. McMahon et al. (19) found evidence for the existence of the cys-LT2 receptor in cultured human mesangial cells, contradicting our results and possibly reflecting a difference between the species.
We found no signal at all for LTC4 synthase in microdissected nephron segments, whereas we detected low abundance in whole kidney, which decreased from the cortex to the inner medulla (Fig. 10). LTC4 synthase has been found in platelets, endothelium, and kidney homogenates (30). Microdissected nephron segments are free of endothelium and platelets, whereas the homogenated kidney zones are not. This might be one reason for our results, and another one could be LTC4 synthase activity in interstitial cells. Thus glomerular and tubular structures in the kidney are clearly not involved in LTC4 synthesis.
In contrast, LTA4 hydrolase showed a high abundance along the whole nephron, except for the OMVR, where we saw weaker signals. Thus we confirmed the results by Nakao et al. (23) and provided the additional information that there is a relatively weak expression of LTA4 hydrolase in OMVR. These data clearly suggest a relatively higher importance of the LTA4 hydrolase pathway in the kidney compared with the LTC4 synthase pathway, especially as we found no reproducible mRNA signals of the LTC4 synthase in microdissected nephron segments.
Taken together, our data suggest that glomeruli and collecting ducts express key enzymes forming HETEs and leukotrienes and express some of their receptors. As a consequence, HETEs and leukotrienes are candidate mediators in the control of renal function.
This study was supported by the ReForM Programm of the Medical Faculty of the University of Regensburg.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2006 the American Physiological Society