INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

LIKE OTHER VISCERAL organs, the kidneys have a profuse sensory innervation. Electrophysiological studies have identified two major classes of sensory receptors in the mammalian kidney: chemoreceptors (CRs), sensitive to ionic composition of renal pelvic perfusate and/or renal ischemia (26, 27), and mechanoreceptors (MRs), sensitive to arterial perfusion (21), venous (2), and ureteral pressure (4). These renal afferent nerves could initiate renorenal reflexes that affect the kidney through its efferent innervation or through circulatory changes (19, 28). In addition, afferent nerves may have a direct effect on the function of visceral organs by releasing neuropeptides (18). Thus these renal sensory receptors symbolize specific messages from the different areas of the intrarenal vascular and interstitial beds and play an important role in the regulation of the body fluid homeostasis (13).

Human beings or animals subjected to long-term hypoxia develop physiological adjustments to adapt the hypoxic stress (29). Despite severe polycythemia and hypervolemia after chronic hypoxia, renal regulation of body fluids is maintained (7, 22). For example, the urinary excretion of water and sodium during and after saline loading (5% of body wt) was not affected in chronically hypoxic rats [high altitude (HA) for 4 wk] compared with that in sea level (SL) control rats. This is despite the fact that the regulatory functions of hormones (7-9, 15, 24, 25) and of renal sympathetic nerves (10, 11, 12, 16) are altered in the chronically hypoxic subjects. Nonetheless, the multifiber recordings of renal efferent nerve activity (RENA) and afferent nerve activity (RANA) in response to volume expansion were different in SL and HA rats (10, 11). During saline loading, RANA in SL rats but not in HA rats increased twofold within a few minutes. The mechanism and the significance of the reduced RANA in hypoxic rats have yet to be determined.

In this study, we performed single-unit recordings of the activity of the peripheral end of the cut nerve filaments to clarify the nature of the increased RANA in response to saline load. Our study showed that, although CRs were not affected, the three types of renal MRs were altered in chronically hypoxic rats. Therefore, in normal control rats, the enhanced MR activation outweighed the decreased CR activation during volume expansion, resulting in an increased multifiber RANA, whereas, in hypoxic rats, lower MR activation merely counterbalanced the decreased CR activation, and the multifiber RANA remained unchanged during volume expansion. Our study provided detailed descriptions of the responsiveness of intrarenal receptors to volume expansion. The method (single-unit recordings) used in this study can be applied in the future to examine the responses of the intrarenal environment to other stimuli.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Induction of chronically hypoxic rats. Female Wistar rats weighing 190-220 g were placed for 15 h/day for 4 wk in a high altitude chamber (HA rats) at a constant temperature and a consistent light cycle (light from 0700 to 1800), whereas control animals were maintained at sea level (SL) at the same temperature and light cycle. A level of 380 Torr (5,500 m) was selected, because it represents the maximal altitude to which most rats can successfully adapt. The animals were exposed to hypoxia from 1700 to 0800 and then returned to room air. The body weight of the animals was measured once a week. Food and water were provided ad libitum.

General surgery. After 4 wk, the rats were anesthetized with pentobarbital sodium (40 mg/kg ip). The trachea was exposed via a midline cervical incision and then intubated. Catheters were placed in the left femoral artery for continuous blood pressure recordings and in the left femoral vein for drug and saline administration. Heart rate was determined using a tachograph triggered by the arterial pulsations. All hemodynamic responses were recorded by a Gould polygraph (RS-3400). The rat was placed on its right side, and the left kidney was exposed via a flank incision and dissection from the surrounding tissue. The kidney surface was continuously bathed with paraffin oil warmed to 38°C and illuminated with a fiber-optic light source. With the help of a stereoscopic dissecting microscope (Olympus, SZ-STU2), the left renal nerves at the angle between the aorta and left renal artery were carefully isolated from the surrounding tissues on each side of the left renal artery for recording of the renal nerve activity.

Recording of renal nerve activity. The recording technique has been reported previously (11). Briefly, recordings from the multifiber preparations were made by placing the intact renal nerve fibers in two pairs of thin bipolar stainless steel electrodes. The renal nerves and electrodes were continuously bathed in a pool of warm paraffin oil (38°C) to prevent drying. The electrical signals were amplified 20,000-fold and filtered (high-frequency cutoff, 3,000Hz; low-frequency cutoff, 30 Hz) by a Grass model P511 AC preamplifier and continuously displayed on a Gould oscilloscope (1604). In the meantime, the amplified signals were fed into a window discriminator (World Precision Instruments model 121) and counted on a Gould integrator (13-4615-70). The neural activity was transformed into a spike number. All recordings were performed under normoxic conditions. The renal nerves were transected, and the multifiber RANA was evaluated by several stimuli to determine its activity. This procedure was performed before the single RANA dissection.

Recording of single RANA. There were 24 SL and 26 HA rats. After general surgery, a short section of the renal artery was carefully cleared of connective tissue to allow complete occlusion by clamping with forceps during the single RANA experiment. A three-way adapter constructed of PE-90 tubing was first connected to three saline-filled lengths of PE-50 tubing and then to a short (1-2 cm) length of PE-50 tubing, which had been previously inserted into the left upper ureter and tied at the ureteropelvic junction. This arrangement enabled continuous measurement of intrapelvic pressure during normal urine flow and during backflow of urine, as described previously by Recordati et al. (27). Renal venous pressure was measured from another catheter (PE-10) put into the renal vein from the ovarian vein. Multifiber RANA was recorded from the peripheral end of the cut nerve filaments. The greater splanchnic and all other visible nerves joining the coeliac plexus were severed to isolate the left kidney from efferent neural influences. The nerve bundle was repeatedly split with fine forceps until a single-unit impulse from a dissected renal nerve filament was apparent on the oscilloscope trace (5). Identification and verification of the renal sensory receptor was then done by renal arterial occlusion (RAO), elevated blood pressure, elevated ureteral pressure, and renal venous occlusion (RVO).

Chemoreceptor type 2. The renal chemoreceptor was activated and identified using renal ischemia for 1 to 2 min. Furthermore, chemoreceptor type 2 (CR2) was examined by spontaneous firing and excited by backflowing nondiuretic urine into the left renal pelvis. This procedure has been shown to identify CR2 (26).

Renal mechanoreceptors. The renal arterial mechanoreceptor (MR_a) was activated by an acute rise in arterial blood pressure (14, 17, 21) with norepinephrine (1 µg/100 g body wt iv) and was severely depressed by the decrease of renal perfusion pressure after aorta ligation or renal ischemia.

The renal ureteropelvic mechanoreceptor (MR_u) was stimulated by elevation of ureteral pressure to 35 ± 1 mmHg with both saline and nondiuretic urine through the long PE-50 catheter inserted into the left ureter.

The renal venous mechanoreceptor (MR_v) was specifically activated and identified by RVO. In each case, excitation was considered to be an increase in activity of at least 100% of basal levels. If the unit did not meet this requirement, dissection of the nerve bundle was continued until an acceptable fiber was identified. Nerve activity and hemodynamic variables were stored on a neurorecorder (DR-890, Neuro Data). All hemodynamic and single RANA responses were displayed on a Gould polygraph (RS3400).

Volume expansion. In this experiment, only one single unit was studied in each rat. After identification of each fiber, the resting activity of the sensory receptor was determined for 4 to 5 min to serve as a control period. Intravenous infusion of isotonic saline of 5% of body weight over a 10-min loading period was then begun. After saline loading, we determined the single RANA during a 10-min recovery period. To compare the hemodynamic changes, we recorded the arterial blood pressure, ureteral pressure, and renal venous pressure simultaneously throughout the experiment.

Statistical analysis. The activity of each unit was displayed graphically as instantaneous frequency of each impulse and as counts of the number of impulses per unit time. Quantitative data are expressed as means ± SE. Differences in absolute unit activity expressed as impulses per unit time were assessed by the Student's paired t-test when appropriate. Data of units between groups were analyzed by the unpaired t-test.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

General observations. After 4 wk of intermittent exposure to 5,500 m hypoxia, all rats were in apparent good health, although the final body weight of the HA rats was significantly less than that of SL rats (204 ± 6 to 233 ± 9 vs. 208 ± 5 to 255 ± 9 g, P < 0.05). The hematocrit level in HA rats was significantly higher than that of SL rats (66 ± 3 vs. 45 ± 1%, P <0.01).

Single RANA. A typical recording of single unit of RANA, arterial pressure, and transformed spikes in a SL rat is illustrated in Fig. 1. The transformed spikes were counted by a Gould integrator. Four types of renal sensory receptors, CR2, MR_a, MR_u, and MR_v, were recorded in both SL and HA rats. In this experiment, all the renal sensory receptors we chose were of the spontaneous type because they could represent the basal firing rate of multifiber RANA. The basal discharge of the four types of sensory receptors was similar in SL rats and HA rats (Table 1). In addition to their responses to specific stimuli, MRs are also different from CR2 in the waveform of recordings (especially the amplitude of the action potential) and higher signal-to-noise ratios, since they might be from a different class of fibers.

View larger version (14K): [in this window] [in a new window]   Fig. 1.   Simultaneous recordings of arterial blood pressure (ABP) and single unit of renal afferent nerve activity (RANA) in one rat at sea level (SL).

Chemoreceptor type 2. At rest, the CR2 was identified by its spontaneous firing and was activated by RAO. This procedure was used to distinguish another type of chemoreceptor, CR1, which is silent at rest (26, 27). The result also showed that CR2 was unresponsive to the elevation of ureteral pressure with saline (Table 1). Meanwhile, CR2 did not respond to the elevation of arterial blood pressure or to RVO. The excitation level of CR2 in response to RAO or back flow of nondiuretic urine was similar in HA and SL rats (Table 1).

Renal mechanoreceptors. As seen in Fig. 2, the MR_a was activated by intravenous administration of noradrenaline (2 µg) in one SL and one HA rat. The excited level of MR_a responding to increasing arterial blood pressure was significantly less in the HA rats than in the SL rats (P < 0.05, Table 1), although the elevated level of arterial blood pressure was similar in HA rats and SL rats (178 ± 4 vs. 183 ± 6 mmHg, respectively). On the other hand, decreased renal perfusion pressure by RAO depressed the firing rate of MR_a similarly between HA and SL rats.

View larger version (20K): [in this window] [in a new window]   Fig. 2.   Responses of ABP, heart rate (HR), and excited activity of renal arterial mechanoreceptor (MRa) to intravenous norepinephrine (NE) in one SL and one rat at high altitude (HA, 380 Torr).

The increase in ureteral pressure to 35 ± 1 mmHg by elevating ureteral catheter with both urine and saline activated the MR_u of HA and SL rats (Fig. 3). The activated level of MR_u in response to elevated ureteral pressure was significantly less in HA rats than in SL rats (P < 0.05, Table 1).

View larger version (22K): [in this window] [in a new window]   Fig. 3.   Response of renal ureteropelvic mechnoreceptor (MRu) to elevated ureteropelvic pressure (UP) in one SL and one HA rat.

RVO activated MR_v in both SL and HA rats (Fig. 4). The activated level of MR_v in response to RVO was also significantly less in HA rats than in SL controls (P < 0.05, Table 1). RAO, increasing ureteral pressure, or arterial blood pressure had no effect on the activity of MR_v.

View larger version (17K): [in this window] [in a new window]   Fig. 4.   Response of renal venous mechanoreceptor (MRv) in renal venous pressure (RVP) to renal venous occlusion (RVO) in one SL and one HA rat.

Volume expansion on renal sensory receptors. A representative recording of one MR_a responding to volume expansion in one SL rat is demonstrated in Fig. 5B. The activity of MR_a increased ~3 min after starting the volume expansion and gradually recovered to the control level ~10 min after the expansion period. Group data for response of the CR2, MR_a, MR_u, and MR_v to saline loading is shown in Fig. 6. During the saline loading and recovery periods, the activity of CR2 was significantly decreased, and the response was similar in both HA rats and SL controls. Conversely, activities of MR_a, MR_u, and MR_v to saline loading were enhanced in both HA and SL rats. However, the peak activity of MR_a during saline loading was less in HA rats than in SL rats (20.0 ± 2.5 vs. 41.3 ± 5.7 spikes/10 s, P < 0.05, respectively). The peak firing of MR_u in response to saline loading was less in HA rats than in SL rats (24.3 ± 6.0 vs. 43.0 ± 5.1 spikes/10 s, P < 0.05, respectively). The peak value of MR_v in response to saline loading was also less in HA rats than in SL rats (17.1 ± 3.3 vs. 33.6 ± 3.7 spikes/10 s, P < 0.05, respectively).

View larger version (17K): [in this window] [in a new window]   Fig. 5.   Demonstration of identification and response of MRa to saline loading in one SL rat. A: activated response of MRa to intravenous NE. B: responses of MRa before, during, and after saline loading.

View larger version (26K): [in this window] [in a new window]   Fig. 6.   Mean changes of chemoreceptor type 2 (CR2), MRa, MRu, and MRv in response to saline loading in SL and HA rats. * P < 0.05 compared with control value, # P < 0.05 compared with SL group.

The responses of arterial blood pressure, ureteropelvic pressure, and renal venous pressure to saline load are shown in Fig. 7. The data showed that ureteropelvic pressure and renal venous pressure were markedly increased by volume expansion in both SL and HA rats. However, arterial blood pressure was not significantly increased in both groups of rats.

View larger version (22K): [in this window] [in a new window]   Fig. 7.   Mean changes of mean ABP, ureteropelvic pressure (UP), and RVP in response to saline loading in SL and HA rats. * P < 0.05 compared with control value.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

The significant findings in this study were summarized as follows. 1) The basal activities of renal sensory receptors were similar in both chronically hypoxic (HA) rats and control (SL) rat, and each renal sensory receptor could be activated and identified by its specific stimulus. 2) The excited responses of the CR2 were similar in both groups of rats. Intravenous saline load depressed CR2 activity in both HA and SL rats to the same degree. 3) The saline loading activated renal arterial, ureteropelvic, and venous MRs in both groups of rats, but the degree of response was significantly less in HA rats than in SL rats.

Renal MRs are nerve endings that monitor hydrostatic pressure changes within the kidney. At rest, systemic and renal hemodynamics were similar in both SL and HA rats (6, 22), and this could explain why the basal discharge of MRs was similar in SL and HA rats. In the present study, three types of renal MRs (i.e., MR_a, MR_u, and MR_v) were excited in response to saline loading in both HA and SL rats. Because MR_a is identified with its activation by acute elevation in arterial blood pressure (14, 17, 21), it is suggested that the increased activity of MR_a during saline load is ascribed to the elevated arterial blood pressure. However, in our experiments, the arterial blood pressure was slightly but insignificantly elevated during saline loading. Similar responses were observed by Astrom and Crafoord (2, 3) when renal vasodilation was produced by infusion of isotonic saline into the renal artery. They suggested that the excitation of RANA was attributed to an increase in interstitial (hydrostatic) pressure rather than to the dilation of vasculature or an elevated arterial pressure (4). In this study, we suggest MR_a could be also possibly stimulated by the increase of blood volume (through volume stretch) and/or blood flow (through shear stress). However, an advanced study is needed to explore the mechanism. On the other hand, both renal ureteral pressure and renal venous pressure significantly increased during volume expansion, which can activate MR_u and MR_v, respectively (3, 4). A similar result, demonstrated by Genovesi et al. (13), showed that RANA increases when diuresis is activated within a moderate physiological range and that this increase is due to the activation of renal MRs. Pelvic pressure and/or interstitial pressure appear to be the major determinants of the neural response. The MR activation in response to volume expansion was less in HA rats than in SL rats, although the responses of arterial, ureteropelvic, and venous pressures during volume expansion were similar between SL and HA rats. This result showed that attenuated response of MRs to volume expansion in HA rats is not dependent on the pressure gradients. A chronic volume adaptation of the increased blood volume (40% higher) may account for to the change (or decrease) in the MR sensitivity of HA rats (1). The less-increased responses of MRs to various elevated pressures in the HA rats could also be attributed to hypervolemia-induced structural adaptation, because chronic hypoxia increases total blood volume (22) and total capillary area (1, 22). However, this mechanism remains to be explored.

Renal CRs respond to alterations in the chemical composition of the renal interstitial or pelvic environment. Two types of renal CRs have been identified (26, 27), CR1 and CR2. CR1 has no basal activity and responds to renal ischemia. CR2 is spontaneously active, with the resting discharge rate being highest in nondiuretic conditions. CR2 activity is related to the hydrated state of the animal. Recordati et al. (26) reported that the multifiber resting discharge of renal chemoafferent activity declines during extracellular fluid volume expansion in normoxic rats. On the other hand, hydropenia leads to concentrated urine and increases the discharge of CR2 (20, 26). In this study, CR2 was activated by RAO (renal ischemia) and backflow of the nondiuretic urine but not activated by increasing renal perfusion pressure, ureteropelvic pressure with saline, and RVO. The similar basal resting discharge of CR2 between SL and HA rats may be due to their similar chemical components of the plasma and urine in these animals (6, 7, 22). During RAO and backflow of nondiuretic urine, the extent of CR2 activation was almost the same in both groups of rats. During and after volume expansion, the CR2 activity was similarly depressed in both HA and SL rats. Some chemical substances, such as endogenous adenosine, bradykinin, and potassium ion, could have been diluted below the threshold concentrations and thus have decreased CR2 discharge during and after volume expansion. Whether the sensitivity of CR is altered in hypoxic rats cannot be determined in test parameters used in this study.

In conclusion, although CR2 activity was depressed by volume expansion, the three types of renal MRs were activated in SL and, to lesser degree, in HA rats. In SL rats, the enhanced MR activation outweighed the decreased CR2 activation during volume expansion, thus increasing multifiber RANA, whereas, in HA rats, lower MR activation counterbalanced the decreased CR2 activation, and thus the multifiber RANA remained unchanged during volume expansion. Based on our results, we conclude that chronic hypoxia attenuates the sensitivity of renal MRs and results in a decreased response to the stimulation of saline loading.