Connecting tubule glomerular feedback mediates tubuloglomerular feedback resetting after unilateral nephrectomy

Abstract

Unilaterally nephrectomized rats (UNx) have higher glomerular capillary pressure (P_GC) that can cause significant glomerular injury in the remnant kidney. P_GC is controlled by the ratio of afferent (Af-Art) and efferent arteriole resistance. Af-Art resistance in turn is regulated by two intrinsic feedback mechanisms: 1) tubuloglomerular feedback (TGF) that causes Af-Art constriction in response to increased NaCl in the macula densa; and 2) connecting tubule glomerular feedback (CTGF) that causes Af-Art dilatation in response to an increase in NaCl transport in the connecting tubule via the epithelial sodium channel (ENaC). Resetting of TGF post-UNx can allow systemic pressure to be transmitted to the glomerulus and cause renal damage, but the mechanism behind this resetting is unclear. Since CTGF is an Af-Art dilatory mechanism, we hypothesized that CTGF is increased after UNx and contributes to TGF resetting. To test this hypothesis, we performed UNx in Sprague-Dawley (8) rats. Twenty-four hours after surgery, we performed micropuncture of individual nephrons and measured stop-flow pressure (P_SF). P_SF is an indirect measurement of P_GC. Maximal TGF response at 40 nl/min was 8.9 ± 1.24 mmHg in sham-UNx rats and 1.39 ± 1.02 mmHg in UNx rats, indicating TGF resetting after UNx. When CTGF was inhibited with the ENaC blocker benzamil (1 μM/l), the TGF response was 12.29 ± 2.01 mmHg in UNx rats and 13.03 ± 1.25 mmHg in sham-UNx rats, indicating restoration of the TGF responses in UNx. We conclude that enhanced CTGF contributes to TGF resetting after UNx.

INTRODUCTION

Unilateral nephrectomy (UNx) is followed by complex glomerular hemodynamic alterations in the nephrons of the remaining kidney (13). These hemodynamic changes include increased renal blood flow, glomerular capillary pressure (P_GC), and single nephron glomerular filtration rate (SNGFR) (6, 10). Sustained elevation in P_GC can cause glomerular barotrauma leading to glomerular damage after UNx (10). Enhanced P_GC or glomerular hypertension has been strongly linked to the sclerotic glomerular damage in the remnant glomeruli, which is accompanied by progressive azotemia and proteinuria after UNx (26).

In the kidney, afferent arteriole (Af-Art), glomerular capillaries, and efferent arteriole (Ef-Art) are arranged in series and thus their dynamics are closely interlinked (19). Because of this unique arrangement of two resistance vessels, the Af-Art and Ef-Art regulate inflow and outflow of blood through glomerular capillaries, respectively, and thus consequently control both P_GC and glomerular filtration rate (GFR) (19). Constriction of the Af-Art can reduce P_GC and flow downstream, which in turn decreases glomerular filtration in absence of other changes (19). Likewise, constriction of the Ef-Art would build up pressure upstream and may increase capillary hydrostatic pressure and GFR (19). Af-Art resistance is regulated by two intrinsic feedback mechanisms: 1) tubuloglomerular feedback (TGF) that causes Af-Art constriction in response to increased NaCl in the macula densa, via the sodium-potassium-2-chloride cotransporter-2 (NKCC2); and 2) connecting tubule glomerular feedback (CTGF) that causes Af-Art dilatation and is initiated by the epithelial sodium channels (ENaC) present in the connecting tubule (27).

Previous studies suggest that there is minimal or no change in Af-Art resistance upon increased tubular delivery of sodium chloride at the macula densa after UNx, which indicates attenuation of TGF (TGF resetting) (13, 23). This TGF resetting may lead to enhanced pressure transmission from systemic circulation to the glomerulus leading to increased P_GC, thus causing the kidney to be susceptible to barotrauma and eventual glomerulosclerosis (1). However, the mechanism of TGF resetting following UNx is still undefined.

Since CTGF is an Af-Art vasodilator mechanism, we studied the possible involvement of CTGF in TGF resetting after UNx. We hypothesized that post-UNx, CTGF is enhanced and that, in-turn, contributes to TGF resetting. To test this hypothesis, in vivo renal micropuncture studies were performed in Sprague-Dawley rats using the stop flow technique 24 h after sham-UNx or UNx with and without intratubular blockade of the ENaC with benzamil. TGF was calculated as a decrease in P_SF caused by an increase in nephron perfusion. CTGF was calculated as the difference between P_SF in the tubule perfused with vehicle and benzamil. We have shown earlier that the Ef-Art diameter gets increased upon increment of NaCl concentration at macula densa using the in vitro double microperfusion method (19, 20). We used same experimental approach to measure Ef-Art resistance in both sham and UNx animals. Since in vitro double microperfusion experiments are technically difficult to perform in rats, we used rabbits for these experiments.

METHODS

The experiments were approved by the Henry Ford Health System Institutional Animal Care and Use Committee and were conducted in accordance with the National Institutes of Health Guidelines for the Care and Use of Laboratory Animals.

UNx in rats and in vivo micropuncture studies.

We used Sprague-Dawley rats for our experiments. Average weight for the sham-UNx and UNx was 313 ± 4.73 and 316.94 ± 4.0 g, respectively. All rats were fed a standard chow diet and given tap water ad libitum. Rats were anesthetized using intraperitoneal injections of pentobarbital sodium (40 mg/kg body wt). Once anesthetized, rats were prepared for surgery by shaving the right flank area immediately distal to the ribs and then disinfecting it with 70% ethanol and betadine. An incision of ~1.5 cm was made in the skin, followed by blunt dissection of the underlying muscle. The exposed right kidney was decapsulated. The renal artery, renal vein, and ureter were tied off as a bundle, using 4–0 silk, and the kidney was removed. Abdominal muscles were sutured first followed by the skin using 4–0 silk, and then buprenorphine (0.01–0.05 mg/kg) was administered subcutaneously for analgesia. For the sham-UNx surgery, we opened the skin and abdominal muscles and then gently manipulated renal fat, after which the incision was sutured as described above without removal of the kidney. Twenty-four hours after sham-UNx or UNx surgery, we performed a renal micropuncture study as described previously (12, 25, 28). Briefly, rats were anesthetized with Inactin intraperitoneally (125 mg/kg body wt). The left kidney was overturned and placed in a Lucite cup. Saline-soaked cotton was placed around the kidney to immobilize it, and 30–45 min were allowed for equilibration. Colored dye was injected into one of the kidney surface tubules, permitting detection of tubule loops with a finding pipette. Grease was then injected into an early segment of the proximal tubule causing a tubule blockage, after which two pipettes were inserted inside the same tubule. First, a perfusion pipette was inserted downstream from the grease block and attached to the infusion pump. Second, a pipette for measuring P_SF was inserted upstream of the grease block and was attached to a micropressure system (model 900A; World Precision 99 Instruments, Sarasota, FL). To generate a P_SF curve, the late proximal perfusion (in orthograde manner) rate was increased stepwise from 0 to 10, 20, 30, and 40 nl/min while the P_SF was measured. In orthograde perfusion of tubules, perfusate reaches the macula densa and then ENaC containing nephron segments, i.e., distal tubule, connecting tubule, and collecting duct, respectively. Each of the perfusion rates were maintained until we observed a stable P_SF. We performed two consecutive TGF response measurements by perfusing tubules with vehicle and after that with the ENaC inhibitor benzamil (1 μM). All the renal micropuncture experiments were performed 24 h postsurgery.

UNx in rabbits.

Young New Zealand rabbits were anesthetized using 50 mg/kg im ketamine and 10 mg/kg im xylazine (Cutter Laboratories, Shawnee, KS) and were kept under anesthesia with isoflurane. The right kidney was removed via a flank incision, and muscles and skin were sutured. During surgery, heart rate and respiratory rate were checked every 10 min and isoflurane was adjusted accordingly. After surgery, animals were maintained on a normal diet. For sham nephrectomy, rabbits were anesthetized as described above. The right kidney was exposed via a flank incision, and then renal fat was gently manipulated without removal the kidney. After that, muscles and skin were sutured and animals were kept on a normal chow diet.

In vitro microperfusion.

One day after sham or UNx surgery, rabbits were anesthetized with ketamine (50 mg/kg im) and given heparin intravenously (500 U). The kidneys were removed and sliced along the corticomedullary axis. Slices were placed in ice-cold MEM (GIBCO, Grand Island, NY) containing 5% BSA (Sigma, St. Louis, MO) and dissected under a stereomicroscope (Olympus SZH, Tokyo, Japan). A single superficial Ef-Art and Af-Art with intact glomerulus from each rabbit were microdissected together with adherent tubular segments consisting of portions of the thick ascending limb, macula densa, and early distal tubule. With the use of a pipette, the microdissected complex was transferred to a temperature-regulated chamber mounted on an inverted microscope (model IMT-2; Olympus) with Hoffman modulation. Both the Ef-Art and macula densa were perfused with an array of concentric glass pipettes as described previously (19). The Ef-Art was perfused by an orthograde manner (through the end of Af-Art) with MEM containing 5% BSA, gassed with room air. The Af-Art was cannulated and the perfusion pipette was advanced to the end to eliminate its hemodynamic influence on the Ef-Art. The tip of the pressure pipette was placed just beyond the distal end of the Af-Art, keeping intraluminal pressure constant at 50 mmHg. The macula densa perfusion solution contained the following (in mmol/l): 4 KHCO₃, 10 HEPES, 0.5 Na acetate, 0.5 Na lactate, 0.5 K₂HPO₄, 1.2 MgSO₄, 1 CaCO₃, and 5.5 glucose, and the final NaCl concentration was 0, 20, 40, and 80 mmol/l. The Ef-Art was preconstricted with norepinephrine (10–6 mol/l), and two consecutive concentration-response curves were generated by increasing the macula densa luminal NaCl from 0 to 20, 40, and 80 mmol/l (each NaCl concentration was maintained for 5 min). The Ef-Art diameter was measured in the region of maximal response to norepinephrine at three sites, 3–5 μm apart, and expressed as the average of these three measurements.

Statistics.

Data are expressed as means ± SE. We used Student’s two sample t-tests, and for the measurement of repeated data, we used Student’s paired t-tests. Hochberg’s step-up procedure for adjusting P values for multiple comparisons was used to control the family-wise type 1 error rate, predefined as 0.05.

RESULTS

Time control TGF responses 24 h after UNx.

To determine whether TGF responses varied with time, the late proximal tubule was perfused twice while measuring P_SF. During the experiment, the perfusion rate was increased from 0 to 40 nl/min step-wise at 10 nl/min intervals. We found no difference between the first and second curves of both sham-UNx (Fig. 1A) and UNx rats (Fig. 1B), indicating that this response was reproducible over time. We also observed that, increase in tubule perfusion decreased P_SF only in sham-UNx rats but not in the UNx rats.

Fig. 1.

Time control for sham-unilaterally nephrectomized rats (UNx) rats and UNx rats after 24 h of surgery. Increasing perfusion rates in the late proximal tubule 2 consecutive times affected stop-flow pressure (P_SF) reproducibly (○, first curve; ●, second curve) in both sham-UNx rats (A) and UNx (B).

TGF response (maximum P_SF change) decreases 24 h after UNx.

To determine whether UNx causes a decrease in the TGF response, maximum P_SF change (at 40 nl/min perfusion rate) was measured in sham and UNx rats. Maximum P_SF change was significantly lower (P < 0.05) in the UNx rats compared with the sham UNx rats. This result indicates that UNx causes TGF resetting 24 h after surgery (Fig. 2, A and B).

Fig. 2.

A: tubuloglomerular feedback (TGF) responses in sham-UNx rats were significantly higher compared with the UNx rats, indicating TGF resetting. B: maximum P_SF change in sham UNx and UNx rats 24 h after surgery. *P < 0.05, sham-UNx vs. UNx.

Benzamil reduces P_SF in sham and UNx rats.

To study whether the basal CTGF response occurs after 24 h of sham-UNx surgery, we generated two consecutive P_SF curves from the same nephron. The first P_SF response shows perfusion treatment with vehicle and the second with ENaC blocker benzamil. To calculate CTGF, we subtracted the vehicle treatment P_SF values from benzamil treatment for each perfusion rate. As expected, the P_SF response (decrease in P_SF) was significantly lower (P < 0.05) after vehicle treatment compared with the benzamil treatment at the perfusion rate of 30 and 40 nl/min. This result indicates that basal CTGF occurs in sham-UNx rats 24 h after surgery (Fig. 3A).

Fig. 3.

Effect of benzamil on P_SF change in both sham-UNx (A) and UNx (B) rats. *P < 0.05, vehicle (Veh) vs. benzamil (BZ).

We performed a similar experiment with intratubular vehicle and benzamil in UNx rats to evaluate the CTGF response in these groups of rats. Vehicle treatment did not elicit any change in P_SF response whereas benzamil treatment significantly increased the P_SF response (decrease in P_SF) compared with the vehicle treatment at the 30 and 40 nl/min perfusion rates (Fig. 3B).

CTGF response in sham-UNx and UNX rats.

To compare the CTGF response between sham-UNx rats and UNx rats 24 h after surgery, we calculated the CTGF response in both sham-UNx rats and UNx rats by subtracting the vehicle P_SF values from the P_SF values obtained after intratubular benzamil treatment at each perfusion rate. The CTGF value at 40 nl/min was significantly enhanced (P < 0.05) in the UNx rats compared with sham-UNx rats (Fig. 4).

Fig. 4.

Comparison of maximum connecting tubule glomerular feedback (CTGF) in both sham-UNx and UNx rats. CTGF is enhanced in UNx rats (black bar) compared with sham UNx rats (owhite bar). *P < 0.05, sham UNx vs. UNx.

Absolute TGF response after benzamil treatment in sham-UNx and UNx rats.

In the presence of benzamil, P_SF decreased in response to the increase in nephron perfusion rate in both sham-UNx and UNx rats 24 h after surgery. This decrease in P_SF upon benzamil treatment was similar in both sham-UNx rats and UNx rats. These data also show that upon CTGF inhibition, TGF behaves similarly in both sham-UNx and UNx group (Fig. 5).

Fig. 5.

Effect of benzamil treatment on P_SF in sham and UNx rats. Inhibition of CTGF eliminate the decrease in TGF in the UNX rats (○, sham-UNx benzamil; ●, UNx benzamil).

Ef-Art diameter after UNx in rabbits.

To study if there is any change in the Ef-Art resistance after UNx, we measured the efferent arteriolar diameter after 1 day of UNx. The Ef-Art of both sham UNx as well as UNx rabbits was preconstricted with norepinephrine 1 μM/l and then was observed upon an increase in the concentration of NaCl at macula densa. Both Sham UNx as well as UNx rabbit’s Ef-Art diameter was not found to be different as they dilate similarly.

DISCUSSION

After UNx, the remnant kidney undergoes numerous hemodynamic changes (11). These changes include elevated renal blood flow, P_GC, and SNGFR and are followed by hypertrophy of the remnant kidney (3). Elevated P_GC has been implicated in causing glomerular barotrauma leading to renal damage (2). Post-UNx, there is also an attenuation in the TGF response called TGF resetting (23). TGF is important in regulating P_GC, Af-Art resistance as well as Ef-Art resistance (19). TGF causes constriction of Af-Art while exerting a dilatory effect on Ef-Art upon increment of NaCl concentration at macula densa (19). Attenuation of TGF has been suggested to have damaging effects on the glomerulus (2). TGF resetting decreases Af-Art resistance, allowing the systemic pressure to be transmitted to the glomerular capillaries and causing barotrauma that may eventually lead to glomerular sclerosis. Six months after UNx, Sprague-Dawley rats showed signs of renal damage (15). Various clinical studies also indicate that there are potential adverse long-term effects in the remnant kidney postkidney donation. One of the studies suggested that ~11 yr after kidney donation, 56% of donors developed proteinuria and GFR reduced by 25%; however, the mechanisms of these adverse effects remain elusive (7). Another recent clinical study showed that 15 yr postkidney donation, the risks of renal failure in the United States are 3.5 to 5.3 times higher for kidney donors compared with the projected risks in the absence of donation (8).

Previously, we demonstrated a novel renal autoregulatory mechanism acting opposite of TGF at the single-nephron level, called CTGF (27). Contrary to TGF, CTGF is a vasodilator mechanism initiated in the connecting tubule segment of the nephron by the ENaC. In the present study, we found that compared with sham-UNx rats, TGF in UNx rats is significantly attenuated 24 h after surgery. Our findings are consistent with the studies by Salmond and Seney (23) and Müller-Suur et al. (13) reporting similar TGF resetting after UNX. Our current data show that intratubular inhibition of CTGF with benzamil significantly reduces P_SF in both UNx rats as well as in sham-UNx rats, but this P_SF reduction is significantly higher in the UNx rats, indicating increased CTGF in the UNx rats. These data suggest that TGF resetting after UNx could be at least in part explained by the enhanced CTGF.

In the UNx group, intratubular perfusion of benzamil restored the TGF response to a level like that in the sham-UNx group. Thus these data may suggest that upon blocking CTGF, TGF could become operational after UNx. UNx seems to shift the balance between vasoconstrictive (TGF) and vasodilatory (CTGF) forces in the favor of the latter, resulting in high P_SF. Enhanced CTGF after UNx may be responsible for the increased renal blood flow as well.

The mechanism by which CTGF is enhanced after UNx remains elusive, but it is known that CTGF is initiated by sodium transport in the connecting tubule via ENaC. We recently showed that the effect of CTGF is mediated by prostaglandin E₂ (PGE₂) (18) and epoxyeicosatrienoic acid (EET) (16). Binding of prostaglandin on PGE₂ receptor 4 (EP-4) on the Af-Art causes dilation and thus exerts the phenomenon of CTGF (17). A similar function for prostaglandins and EET may be expected for the mediation of TGF resetting after UNX. Furthermore, it has been shown that pretreatment with indomethacin prevented the reduction in TGF sensitivity after UNx (9). Additionally, renal ablation increased cyclooxygenase-2 expression in the renal cortex, and it also further increased urinary excretion of PGE₂, indicating that increased CTGF may have had vasodilatory effects on the renal Af-Art through increased PGE₂ synthesis (24, 29).

At this point, it is difficult to determine the factors involved in increased renal blood flow after UNx. Elevation in renal blood flow after UNx could result from a combination of impaired myogenic response, enhanced nitric oxide (NO) production, and CTGF. Increased shear stress is required for increased NO production, and an increase in shear stress requires an increase in the renal blood flow. Essentially, enhanced NO production after UNx should be the result, but not the cause, of the increased renal blood flow. Enhanced CTGF along with the impaired myogenic response could explain the mechanism behind elevated renal blood flow after UNx.

Because TGF has a vasodilatory effect on the Ef-Art (at least in vitro), one may speculate that in the UNx model in which TGF is attenuated, the Ef-Art may constrict, resulting in elevation of P_GC (21). However, we did not observe any change in the Ef-Art resistance after UNx (Fig. 6). We measured the efferent arteriolar resistance directly while performing the Ef-Art TGF response after UNx and found it to be similar in both sham UNx and unilaterally nephrectomized rabbits. Brenner and colleagues (6) measured the Ef-Art resistance in Munich Wistar rats after UNx and found it to be decreased.The reason behind difference in our results to theirs could be attributed to either the method to calculate the Ef-Art resistance (Brenner and colleagues determined the resistance indirectly using equation based calculation while we measured it directly) or species difference (Brenner and colleagues used Munich Wistar rats while we used rabbits). Resetting of TGF occurs hours after UNx, and thus there may be a possible involvement of impaired myogenic response or interaction between CTGF and the myogenic response.

Fig. 6.

Effect of UNx on the diameter of efferent arteriole (Ef-Art) in rabbits. There was no difference in the diameter of efferent arteriole diameter in sham-UNx rabbits and UNx rabbits upon increasing sodium concentration at macula densa (MD), and this effect was reproducible as evidenced by first curve (○) and second curve (●). NE, norepinephrine.

ENaC channels are present in the distal tubule, in the connecting tubule, and in the collecting duct (22). When we perfuse the tubule with benzamil in orthograde manner at the late proximal tubule during micropuncture experiments, we assume that Af-Art constriction is mediated by the blockage of ENaC channels in the connecting tubule because the connecting tubule is the only known ENaC-containing segment of the nephron that juxtaposes with the Af-Art (5). Other ENaC-containing segments including the distal tubule and collecting duct do not contact with the Af-Art, and thus the chance of transfer of signal from these segments to the Af-Art is minimal. Additionally, amiloride-sensitive sodium conductance along the cortical collecting duct, including the connecting tubule, has been shown to be increased after UNx (14). In this elegant paper, the authors dissected the cortical collecting duct along with the connecting tubule after UNx and measured the sodium current in the isolated cortical collecting duct that was enhanced after the UNx. This enhanced sodium current after UNx was significantly reduced upon Amiloride treatment, which is an ENaC inhibitor indicating increased ENaC activity (14).

One of the drawbacks in our study is that we have indirectly measured the P_GC as P_SF because Sprague-Dawley rats do not have surface glomeruli (4). In our experiments, we found P_SF to be high at a higher perfusion rate after UNx compared with the sham UNx rats like the other reported findings (13, 23). Direct measurement of P_GC after unilaterally nephrectomized Munich Wistar rats is also shown to be significantly higher when compared with the sham UNx rats (3, 6). Another limitation of the current study is that we have not measured SNGFR in our study. Apart from an increase in the P_GC, increased SNGFR or single nephron plasma flow per se could also play a role in renal damage after UNx. However, in the current study, we have not ruled out the contribution of SNGFR to renal damage after UNx. Both increased P_GC and increased SNGFR could contribute to renal damage post-UNx.

In summary, our studies provide direct evidence of TGF resetting after UNx and that CTGF mediates TGF resetting induced by UNx 24 h after the surgery. UNx rats had almost no TGF response compared with sham-UNx rats. These differences are attributable in part to greater CTGF in UNx rats. These findings may help explain the increase in P_GC and reduced TGF response after UNx.

Perspectives

An increase in CTGF may explain the higher P_GC and renal dysfunction in kidney donors. Blocking CTGF with ENaC inhibitors could decrease P_GC and diminish the risk of renal dysfunction in the remnant kidney of these donors.

GRANTS

Research reported in this publication was supported by National Heart, Lung, and Blood Institute Grant HL-028982. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

S.R.M. and O.A.C. conceived and designed research; S.R.M., Y.R., J.M.J., K.K., and N.K. performed experiments; S.R.M., Y.R., J.M.J., K.K., H.W., N.K., E.L.P., and O.A.C. analyzed data; S.R.M., H.W., and O.A.C. interpreted results of experiments; S.R.M. prepared figures; S.R.M. drafted manuscript; S.R.M. and O.A.C. edited and revised manuscript; O.A.C. approved final version of manuscript.