Effect of Nitric Oxide Synthase Inhibition and Saline Administration on Blood Pressure and Renal Sodium Handling During Experimental Sepsis in Rats

Abstract

Much effort has been made in recent years to clarify metabolic and renal function changes in sepsis. A number of studies performed in different models of sepsis have been described. One such model that is frequently used is cecal ligation and puncture (CLP) in rats. This model resembles human sepsis in several important aspects, such as an early phase of hyperdynamic, hypermetabolic sepsis followed by a late hypodynamic, hypometabolic phase. The present study evaluated the blood pressure (n = 5) and renal function changes during development of CLP renal failure and to determine the effects of NOS inhibition (L-NAME) and 0.15 M NaCl administration on tail blood pressure and renal function in randomly assigned five groups (n = 10 each): (1) Sham-operated, (2) Sham-operated L-NAME-treated, (3) CLP rats, (4) CLP L-NAME-treated, and (5) CLP 0.15 M NaCl-treated rats. The basal tail blood pressure was not significantly different among the four groups. One week later, arterial pressure was significantly increased in sham-operated L-NAME-treated rats (159 ± 12 mmHg) compared with the other groups (118 ± 9.0 mmHg in nontreated rats, p<0.05). Blood pressure shows a slightly and not significant decrease up to 12 h in L-NAME and 0.15 M NaCl treated rats, which in turn was followed by a significant reduced arterial pressure 18 h after CLP in both groups (L-NAME: 96.0 ± 3.6 mmHg, p<0.05) and NaCl: 82.3 ± 2.4 mmHg, p<0.05) compared to sham-operated groups. The glomerular filtration rate estimated by C_Cr decreases significantly in the CLP untreated group (p<0.001) and did not significantly differ from the sham-operated and L-NAME-treated groups (p = 0.4) during the studies of renal tubule sodium handling. On the other hand, subcutaneous 0.15 M NaCl administration prevented C_Cr decreases in CLP rats (p = 0.25). CLP increased the FE_Na in the sham-operated from: 857.2 ± 85.1 Δ% min⁻¹ to CLP: 1197.8 ± 119.0 Δ% min⁻¹. The high FE_Na to CLP was blunted and significantly reduced by previous systemic treatment of animals with L-NAME from sham-operated + L-NAME: 1368.0 ± 72.0 Δ% min⁻¹ to CLP+L-NAME: 1148.0 ± 60.4 Δ% min⁻¹ (p<0.01). The enhanced FE_Na in the CLP group were accompanied by a significant increase in proximal sodium reabsorption rejection. The salient findings of the present study suggest that a decrease in the blood pressure and creatinine clearance caused by CLP may benefit from L-NAME and fluid resuscitation during initial bacteremia (first 12 h) by promoting an additional increase of tubule sodium reabsorption in the post-proximal segments of nephrons, but these therapies could not prevent acute renal failure after established endotoxemia.

• Acute renal failure
• Nitric oxide synthase
• Renal function
• Blood pressure
• CLP and lithium clearance

Introduction

Acute renal failure is a frequent complication in septic patients associated with a high mortality rate.[[1]] An important characteristic is decreased glomerular filtration rate, which is even present in situations where renal blood flow has been maintained.[[2]] The prevailing hypothesis on the pathogenesis of renal failure suggests an inappropriate and uncontrolled release of various inflammatory mediators.[[1]] Overwhelming sepsis, with or without septic shock, can be an extremely difficult clinical problem to treat. Antibiotics clearly are a mainstay of therapy but are often inadequate in the immunocompromised host. Additionally, an adequate nutritional, pressure, fluid and electrolyte management is critical in the care of patients with sepsis.

Much effort has been made in recent years to clarify metabolic and renal function changes in sepsis. A large number of these studies have been performed in experimental animals and a number of different models of sepsis have been described.[[3]], [[4]], [[5]], [[6]], [[7]] One such model that is frequently used is cecal ligation and puncture (CLP) in rats. This model resembles human sepsis in several important aspects, such as an early phase of hyperdynamic, hypermetabolic sepsis followed by a late phase of hypodynamic, hypometabolic sepsis.[[3]], [[4]] An intra-abdominal abscess causes the condition as well as devitalized tissue, both common sources of sepsis in patients.

We have previously shown data indicating that endogenously produced growth factors and drugs may protect the kidneys from acute damage.[[8]], [[9]] In rodents, an increased production of nitric oxide (NO) via the inducible isoform of NO synthase (iNOS) has been incriminated in the pathogenesis of septic shock.[[10]] The endothelium is the major source of nitric oxide (NO), which plays an important role in local circulatory control.[[11]], [[12]] Although the precise mechanism by which continued NO synthesis inhibition induces chronic hypertension remains to be identified, renal control of the fluid and electrolyte balance is thought to play a dominant role in the long-term control of arterial pressure in both normal and pathophysiological states.[[13]], [[14]], [[15]] Renal sympathetic nerve activity increases after inhibition of NO synthesis[[16]], [[17]], [[18]] and the activation of these nerves inhibits renal sodium excretion and promotes renal renin secretion.[[19]] Since the time course of iNOS activity is not precisely known during endotoxic shock in rats and, in many species, including humans, the higher concentrations of several inflammatory mediators required to activate iNOS suggest that involvement of NO pathways may be less important.[[10]]

On the other hand, administration of an adequate volume of fluid seems to be important for the hyperdynamic response after CLP in rats.[[3]] However, the results of this study showing unaltered hematocrit levels do not suggest that the rats were significantly dehydrated. Perhaps aggressive volume resuscitation would improve survival rate and also alter the renal injury response as well. Thus, the protective effect of NOS inhibition or 0.15 M NaCl administration on blood pressure and renal function in CLP model is actually unknown. The sodium and water nephron handling is an important tool to check tubular function, and tubule lithium transport has been shown to be a suitable indirect method that allows the study of proximal sodium reabsorption.[[8]], [[20]], [[21]]

The aims of the present study were to evaluate the blood pressure and renal function changes during development of CLP renal failure utilizing the lithium clearance method and to determine the effects of NOS inhibition (L-NAME) and 0.15 M NaCl administration on tail arterial pressure and renal function in rats.

Material and Methods

The experiments were carried out on five groups of male Wistar-Hannover rats (200–250 g) allowed free access to tap water and standard rat chow containing 0.3–0.5% sodium (Labina-Purina, Campinas, Brazil). The rats were housed in a 12 h light/dark cycle animal facility with controlled humidity (50%) and temperature (25°C). Rats in this study were maintained in accordance with the guidelines established by the Brazilian College of Animal Experimentation (COBEA). Cecal ligation and puncture (CLP) induced sepsis as previously described.[[3]], [[22]] Briefly, rats underwent surgery after an overnight fast while under intraperitoneal pentobarbital sodium (30 mg/kg) anesthesia. The cecum was mobilized from the mesentery and distally ligated to the ileocecal valve; the fecal matter in the most proximal part of the colon was milked back into the cecum before ligation. After that, the cecum was punctured twice with an 18-gauge needle in order to facilitate the spread of bacteria into the peritoneal cavity and planes, then the abdomen was closed with a running 2.0 silk suture.

The assessment of the influence of oral administration of L-NAME or subcutaneous isotonic (0.15 M NaCl) fluid reposition on the septic rats was randomly assigned to one of five groups (n = 10 each): (1) Sham-operated, (2) Sham-operated L-NAME-treated, (3) CLP rats, (4) CLP L-NAME-treated, and (5) CLP 0.15 M NaCl-treated. After surgery, the animals were kept in individual steel metabolic cages for the duration of the study, with free access to chow and tap water.

Experimental Design

The experiments were performed in parallel for each group of sham and CLP-treated animals. The arterial blood pressure was estimated and a renal function study was performed 3, 6, 12, and 18 h after surgery procedures (sham or CLP). Sham-operated L-NAME-treated and CPL L-NAME-treated groups were maintained on the daily oral administration of L-NAME (60 mg/kg/day) or vehicle preceding surgery procedures for seven days. Arterial pressure was measured one day before the renal test in conscious, restrained rats by the tail-cuff method, using an electrosphygmomanometer (Narco Bio-System, Austin, TX, USA). This indirect approach permits repeated measurements with a close correlation (correlation coefficient = 0.975) compared to direct intra-arterial recording.[[23]] In the CLP 0.15 M NaCl-treated group, a volume reposition of 5 mL of sterile 0.15 M NaCl per 100 g of body weight[[3]] was subcutaneously administered on the back immediately after CLP.

Fourteen hours before the renal test, 60 mmol LiCl/100 g body weight was given by gavage. The unanesthetized rats were subsequently housed individually in metabolic cages with free access to tap water but no food. The experiment was performed at the same time in each group of sham-operated, CLP [vehicle or L-NAME-treated] and CLP 0.15 M NaCl-treated rats. At 8:00 a.m., each rat received a tap water load (2% of body weight) by gavage followed by a second load of the same volume 1 h later. Twenty minutes after the second load, spontaneously voided urine began to be collected over a 2 h period. The voided urine passed through the funnel in the bottom of the cage into a graduated centrifuge tube. At the end of the experiment, blood samples were drawn by cardiac puncture and the kidney was immediately removed, decapsulated and weighed.

Biochemical Analysis

Plasma and urine sodium, potassium, and lithium concentrations were measured by flame photometry, while the creatinine concentration and pH, pCO₂, and HCO₃ levels were determined spectrophotometrically by the alkaline picrate method and standard gasometric sensitive electrode, respectively.

Statistics and Calculations

The results are reported as means ± SEM per 100 g body weight. Creatinine clearance was used to estimate glomerular filtration rate (GFR) and lithium clearance (C_Li) was used to assess proximal tubule output.[[8]], [[20]], [[21]] Fractional sodium excretion (FE_Na) was calculated as C_Na/C_Cr, where C_Na is sodium clearance and C_Cr is creatinine clearance. The fractional proximal (FEP_Na) and post-proximal (FEPP_Na) sodium excretion rates were calculated as C_Li/C_Cr × 100 and C_Na/C_Li × 100, respectively. Renal parameter responses to experimental maneuvers were calculated as the area under curve vs. time (AUC, Δ% min⁻¹). The renal data was expressed as a percentage of sham-operated experimental values obtained during the 120-min at the same time and compared with each group of CLP [vehicle or L-NAME-treated] and CLP 0.15 M NaCl-treated rats. Statistical analysis of the data was performed using ANOVA for repeated measurements. Bonferroni's post-hoc analysis was used to determine the extent of the differences. A p value<0.05 was considered significant.

Results

Figure 1 and Table 1 show the effects of cecal ligation and puncture (CLP) with or without seven days L-NAME or 0.15 M NaCl previous treatment on tail arterial pressure (in mmHg) and renal Na⁺ and K⁺ handling, and pH, pCO₂, and plasma levels expressed as means ± SEM per 100 g b.w. All rats survived up to eighteen hours after sepsis induction. The peritoneal cavity contained about 6 mL of brownish, foul-smelling fluid 18 h after CLP in all experimental groups. In all CLP groups, the cecum was engorged and gangrenous with extensive cecal encapsulated accumulation of purulent material. There were no significant differences between serum sodium, potassium, and lithium levels (Table 1) in sham-operated rats compared with the other groups. There was also metabolic acidosis as shown by significant drop in pH, P_aCO₂, and upto 3 h after CLP in all septicemia-induced groups.

Figure 1. Effect of oral administration of L-NAME or subcutaneous isotonic fluid reposition in sham-operated L-NAME-treated (Sh + L-NAME), the cecal ligation and puncture L-NAME-treated (CLP + L-NAME) and CLP 0.15 M NaCl-treated (NaCl) rats on creatinine clearance (C_Cr), fractional excretion of sodium (FE_Na), proximal (FEP_Na) and post-proximal (FEPP_Na) fractional excretion of sodium, fractional excretion of potassium (FE_K), and blood pressure compared to sham-operated (Sh) and CLP rats (CLP) untreated rats. The data were calculated as the area under curve vs. time (AUC, Δ% min⁻¹) and reported as the means ± SEM. *p ≤ 0.05 vs. GI group (ANOVA and Bonferroni's contrast test).

Table 1. Effect of oral administration of L-NAME or subcutaneous isotonic fluid reposition in sham-operated L-NAME-treated (GII), the cecal ligation and puncture (CLP) L-NAME-treated (GIV), and CLP 0.15 M NaCl-treated (GV) rats on body weight, serum pH, P_aCO₂, , sodium, potassium, and lithium levels compared to sham-operated (GI) and CLP rats (GIII) untreated rats. The data are reported as the means ± SEM for 10 rats per group

	Body weight		PaCO2		Na^+	K^+	Li^+
Groups	(g)	pH	(mmHg)	(mM)	(mM)	(mM)	(µM)
GI	300 ± 15	7.46 ± 0.05	42.9 ± 0.9	23.0 ± 2.3	143 ± 0.9	3.7 ± 0.1	107 ± 5.0
GII	289 ± 16	7.01 ± 0.05a	21.8 ± 0.5a	7.0 ± 2.5a	142 ± 1.2	4.5 ± 0.1	117 ± 6.0
GIII	291 ± 19	7.37 ± 0.05	40.7 ± 0.7	22.9 ± 3.0	143 ± 0.4	4.4 ± 0.1	09 ± 8.0
GIV	302 ± 14	6.95 ± 006a	17.8 ± 0.9a	9.5 ± 2.4a	145 ± 0.7	4.3 ± 0.2	110 ± 7.2
GV	298 ± 11	7.00 ± 0.06a	20.1 ± 1.3a	10.1 ± 2.6a	144 ± 1.2	4.3 ± 0.5	109 ± 8.6

^ap ≤ 0.05 vs. GI group (ANOVA and Bonferroni’ s contrast test).

Blood pressure measurements using a tail cuff were performed 3, 6, 12, and 18 h after surgery in randomly selected rats of all experimental groups (n = 5, each). The basal blood pressure was not different among the four groups. One week later, arterial pressure was significantly increased in sham-operated L-NAME-treated rats compared with the other groups and reached 159 ± 12 mm Hg compared to 118 ± ± 9.0 mm Hg in nontreated rats (p<0.05) (Fig. 1). The tail blood pressure remained stable in the CLP nontreated group of rats during 6 h showing an accentuated and significant decrease to 78.1 ± 1.6 mm Hg (p<0.05) and 48.3 ± 2.6 mm Hg (p<0.01) 12 and 18 h, respectively, after CLP (Fig. 1). Blood pressure shows a slightly and not significant decrease up to 12 h in L-NAME and 0.15 M NaCl treated rats, which in turn was followed by a significant reduced arterial pressure 18 h after CLP in both groups (L-NAME: 96.0 ± 3.6 mm Hg, p<0.05) and NaCl: 82.3 ± 2.4 mm Hg, p<0.05) compared to sham-operated groups. These results demonstrated that the arterial pressure decrease was retarded and attenuated regarding CLP untreated animals by previous treatment of the animals with L-NAME and 0.15 M NaCl (Fig. 1).

The glomerular filtration rate estimated by C_Cr decreases significantly in the CLP untreated group (p<0.001) and did not differ from the sham-operated and L-NAME-treated groups (p = 0.4) during the studies of renal tubule sodium handling (Fig. 1). On the other hand, subcutaneous 0.15 M NaCl administration prevented C_Cr decreases in CLP rats (p = 0.25). Cecal ligation and puncture increased the FE_Na from the sham-operated: 857.2 ± 85.1 Δ% min⁻¹ to CLP: 1197.8 ± 119.0 Δ% min⁻¹, and FE_K from the sham-operated: 458.6 ± 28.5 Δ% min⁻¹ to CLP: 894.0 ± 80.5 Δ% min⁻¹. The high FE_Na to CLP was blunted and reduced by previous systemic treatment of animals with 60 mg/kg/day L-NAME from sham-operated + L-NAME: 1368.0 ± 72.0 Δ% min⁻¹ to CLP + L-NAME: 1148.0 ± 60.4 Δ% min⁻¹ p<0.01 (see Fig. 1). This attenuated urinary ion excretion was associated with unchanged proximal and significant decrease in post-proximal sodium rejection (Fig.1). Also, the natriuresis response after CLP + 0.15 M NaCl administration (570.4 ± 69.0 Δ% min⁻¹) was significantly lower than in the sham-operated group (855.6 ± 85.1 Δ% min⁻¹). The enhanced FE_Na and FE_K in the CLP group were accompanied by a significant increase in proximal sodium rejection (from sham-operated: 557.4 ± 44.3 Δ% min⁻¹ to CLP: 685.1 ± 32.1 Δ% min⁻¹) compared with unaltered proximal absorption in all the other groups (Fig. 1). This increase occurred despite a decreased C_Cr and FEPP_Na (Fig. 1). The increased post-proximal sodium reabsorption remained even after L-NAME treatment of CLP rats. The decreased fractional urinary sodium excretion after CLP + 0.15 M NaCl administration was accompanied by a paradoxical and significant increase in post-proximal sodium excretion compared to the sham-operated untreated rats. This increase occurred despite unchanged C_Cr and FEP_Na (Fig. 1).

Discussion

Whether acute renal failure following overwhelming bacterial septicemia is initially a primary consequence of a cytotoxic insult or a deleterious decrease in blood pressure and renal perfusion insufficiency remains unclear. It has been difficult to develop and test sepsis in laboratory animal models that provide adequate analogies to the human forms of sepsis. We chose the CLP model for several reasons: the hemodynamic and metabolic effects of the CLP in rat have been well studied[[3]], [[4]], [[5]], [[6]], [[24]]; in this model, the animals developed bacteremia within 2 h of CLP with early, hyperdynamic stage of sepsis, which deteriorated into a later hypodynamic phase and, this model seems to have clear parallels to the clinical situation of patients.[[3]], [[4]], [[5]], [[6]], [[24]] In the current studies we demonstrated that experimental CLP-induced sepsis produced substantial increase in the urinary output of Na⁺ and K⁺, along with significant decrease in creatinine clearance; and that CLP-induced renal ion excretion is, at least in part, related to changes in peripheral NO-dependent pathways. In addition, we showed that CLP-induced natriuresis occurred by increasing proximal tubule Na⁺ rejection despite a decreased C_Cr (Fig. 1) and the Na⁺ filtered load. Thus, the observed increase in renal FE_Na and FE_K (Fig. 1) may be due to the inability of renal tubules to handle the electrolytes, with a promoted disruption in glomerotubular balance in CLP septic animals. Our findings also demonstrated that maintained arterial pressure stability induced by a 7-day oral administration of L-NAME resulted in transiently sustained glomerular filtration rate, estimated by creatinine clearance, and in the proportion of the filtered sodium load associated with a marked post-proximal antinatriuresis in experimental sepsis induced by CLP in rats (Fig. 1). Previous studies suggest involvement of nitric oxide (NO), generated by inducible NO synthase (iNOS), in the pathogenesis of endotoxin-induced renal failure.[[10]], [[25]] In rodents, the nitric oxide pathways seem to play a key role in the alteration of experimental endotoxic shock. Under these conditions, an overproduction of NO is triggered by endotoxin, mainly through the activation of inducible NO synthase.[[10]], [[25]], [[26]] Furthermore, the endogenous atrial natriuretic peptide increase in the acute phase of endotoxic shock in parallel to NO activity and may play an important role in water and sodium imbalance observed in CLP animals.[[27]]

The inhibition of NO synthesis may influence tubular reabsorption by a direct effect on tubular sodium transport or by hemodynamically mediated mechanisms involving a reduction in medullary blood flow or a rise in renal interstitial hydrostatic pressure.[[28]], [[29]] In support of a direct NO action, the study by Stoos et al.[[30]] showed that NO directly modified sodium transport in cultured cortical collecting duct cells. Mattson and colleagues[[29]] also reported that NOS inhibition decreased papillary plasma flow. These changes in intrarenal hemodynamics may play an important role in mediating the altered sodium handling induced by L-NAME.

Previous studies, which examined the effects of systemic NOS inhibition on sodium excretion, have yielded mixed results perhaps because of differences in the experimental methods and doses of L-NAME used, type of anesthesia, and the activity of endogenous vasoconstrictor systems. Such factors could influence the extent to which renal perfusion pressure and renal vascular resistance responded to inhibition of NO synthesis, thus affecting sodium excretory rate. Lahera et al.[[31]] examined the effect of different doses of L-NAME on natriuresis. In rats receiving 1 mg kg⁻¹ min⁻¹, an anti-natriuresis response occurred with no changes in mean arterial pressure, renal plasma flow or glomerular filtration rate. In contrast, the infusion of 50 mg kg⁻¹ min⁻¹ significantly increased arterial pressure and was accompanied by an early, progressive fall in renal blood flow and glomerular filtration with no change in renal sodium excretion.

These results suggest that the inhibition of NO synthase can directly and transiently enhance sodium reabsorption in post-proximal nephron segments. At larger doses such as those used in our experiments, L-NAME produced increases in arterial pressure that overrode the initial natriuretic effect observed in CLP-untreated animals. In agreement with our findings, Cohen et al. showed that NO synthase inhibitors increase mean arterial pressure and systemic vascular resistance in animal models of sepsis and in humans with septic shock without any significant effects on coronary blood flow and left ventricle function. Additionally, a study by Preiser et al.,[[10]] indicates that septic shock in dogs is associated with an increase in iNOS activity, mainly in the liver and the heart. An increase in iNOS activity implies an increase in NO production, and may also imply an increase in superoxide generation. The ensuing peroxynitrite generation may actually be more deleterious to tissues, including kidneys, than NO and superoxide separately.[[10]], [[32]] The cellular toxic effects of iNOS-derived NO and peroxynitrites can contribute to the detrimental effects of endotoxin on renal tissue function. A previous study[[33]] showed that pretreatment of rats with L-NAME seven days before CLP significantly attenuated the renal and metabolic effects as well as the increases in the plasma levels of TNF alpha caused by endotoxic shock.

On the other hand, Sakuma et al.[[34]] showed that intravenous administration of NO synthesis inhibitor increased renal sympathetic nerve activity. Spinal sectioning but not vagotomy or sinoaortic baroreceptor denervation abolished this effect. The injection of NO synthesis inhibitor intracisternally[[35]] and directly into the nucleus tractus solitarius[[36]] and rostral ventrolateral medulla[[37]] also increases arterial pressure and renal sympathetic nerve activity. Thus, NO may be also involved in brain stem regulation of sympathetic nerve activity and NO inhibitors may influence the renal control of sodium excretion and arterial pressure via this pathway in normal and CLP septicemic model. The present as well as other studies[[32]], [[33]] suggest that iNOS activation may be implicated in the septic renal dysfunction and iNOS inhibition may offer, at least transiently, some advantage to keep adequate kidney function in CLP septic model.

Since cardiac output was not measured in the present study, it cannot be stated whether the rats were in a hyperdynamic or hypodynamic state of sepsis. Decreased blood pressure in CLP animals suggests that the rats were in hypodynamic sepsis after a 12 h period following CLP. Subcutaneous administration of saline fluid before the surgery is important to maintain stable the blood pressure response after CLP in rats. In the present study, 5 mL per 100 g of body weight of 0.15 M NaCl were administered subcutaneously, as previously recommended.[[3]], [[38]] The results suggest that fluid reposition may be required to assure a transient hemodynamic homeostasis in endotoxemic animals. The improvement in blood pressure and renal function in the CLP model after saline administration may be directly associated to improved hemodynamic parameters or indirectly to attenuated increases in plasma TNF alpha levels in these animals.[[38]], [[39]]

The precise mechanism underlying the transient improvement of hemodynamic and renal parameters induced by NOS inhibition and fluid administration has not been identified in the present study. However, its salient findings suggest that a decrease in the blood pressure and creatinine clearance caused by CLP may benefit from L-NAME and fluid resuscitation during initial bacteremia (first 12 h) by promoting an additional increase of tubule sodium reabsorption in the post-proximal segments of nephrons, but these therapies could not prevent acute renal failure after established endotoxemia. A complete understanding of the physiopathological as well as the potentially therapeutical role of fluid support and NOS inhibition on septic hemodynamic and renal dysfunction must await further studies.

Acknowledgment

The authors wish to thank Ms. Adriana R. M. Crété, B.S., for expert technical assistance. This research was supported by CNPq (No. 500868/91-3), PRONEX, CAPES and FAPESP (95/1299-0).