Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Phosphodiesterases (PDEs) are enzymes primarily responsible for the breakdown of the intracellular second messengers cAMP and cGMP. Although these enzymes were initially described many years ago, new families and subclasses of PDEs continued to be discovered (Lowe and Chang, 1992). Presently, there are nine families and 35 subtypes of PDEs, which are classified by their affinity to cAMP and cGMP as well as the presence of inhibitors (cGMP) or activators (calmodulin) (Dousa, 1999). There also exists a wide variation in the species and specific tissue distribution of the types of PDEs. The tissue-specific differences, as well as the discovery of type-specific PDE inhibitors, have led to increased study on each specific PDE type in the pharmacologic literature.

Type IV phosphodiesterase (PDE4), a cAMP-specific family of PDE, has been found to be the predominant PDE type in renal vasculature (Jackson et al., 1997). This finding and the fact that cAMP is a potent vasodilator of the renal vasculature led us to report that Ro 20-1724, a type IV phosphodiesterase inhibitor, prevents endotoxin-induced increases in renal vascular resistance when used as a pretreatment to shock induced by intravenous lipopolysaccharide (Begany et al., 1996). We have since reported that PDE4 inhibition attenuates the increases in renal and superior mesenteric vascular resistance and improved survival in a lethal model of endotoxic shock in the presence and absence of catecholamines (Carcillo et al., 1997). The purpose of this study was to examine the chronic effects of PDE4 inhibition on renal and mesenteric vascular resistance and blood flow, as well as systemic hemodynamics, in a sublethal inflammatory model of MODS.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Male Sprague-Dawley rats (370 ± 5 g, mean ± S.E.M., n = 28) were obtained from Charles River Laboratories (Wilmington, MA) and housed for at least 1 week in the animal care facility, with a 12-h light/dark cycle (lights on from 7:00 AM to 7:00 PM), ambient temperature of 22°C, and relative humidity of 55%. Institutional guidelines for animal welfare were followed. Rats were fed Prolab Isopro RMH 3000 rodent diet (PMI Nutrition Intl., Richmond, IN) and were given water ad libitum. The Institutional Animal Care and Use Committee approved all procedures.

On day 1 of the experiment, rats were anesthetized with halothane. An area of the back was shaved and prepped with alcohol, and a surgical incision was placed inferior to the scapula. After the subcutaneous space was expanded, a mini-infusion osmotic pump (model 2 ML1; Alza Scientific Products, Mountain View, CA) filled with either Ro 20-1724 (no. R111; Sigma, St. Louis, MO) set to deliver at 0.3 or 2.0 µg/kg/min or vehicle (propylene glycol, no. P 4347; Sigma) was inserted. The incision was closed with surgical staples. The rats were then injected with zymosan (0.25 mg/g of body weight; no. Z 4250, Sigma) or vehicle (mineral oil, no. 400-5; Sigma) intraperitoneally and 20 ml/kg of 0.9% normal saline subcutaneously. Rats were allowed to recover from anesthesia and returned to the cages, where they were again allowed access to water and rat chow ad libitum. On day 2 of the experiment, rats were injected with 20 ml/kg of 0.9% normal saline subcutaneously, without anesthesia, and returned to their cages.

On day 3 of the experiment, rats were anesthetized with Inactin (thiobutabarbital sodium, 90 mg/kg intraperitoneally) and placed on a Deltaphase isothermal pad (Braintree Scientific, Braintree, MA). Body temperature was monitored with a rectal temperature probe thermometer (Physiotemp Instrument, Clifton, NJ) and maintained at 37 ± 0.5°C by adjustment of a heat lamp positioned above the rat. Two PE-50 catheters were inserted into the left jugular vein and infusions were administered using a Braintree infusion pump (model BSP99). One PE-50 cannula infused 0.9% normal saline at 80 µl/min, which was begun after a 40-ml/kg bolus of 0.9% normal saline over 30 min. The second PE-50 cannula infused 0.9% normal saline and later norepinephrine (1 and 10 µg/kg/min) at 20 µl/min. The left carotid artery was cannulated with PE-50 tubing that was connected to a digital blood pressure analyzer (Micro-Med, Inc., Louisville, KY) for the continuous measurement of mean arterial blood pressure (MABP) and heart rate (HR). The abdominal cavity was exposed through a midline incision, and the superior mesenteric artery and left renal artery were carefully freed from surrounding tissue. Transit-time blood flow probes (model 1RB for the renal artery and model 2SB for the mesenteric artery; Transonic Systems Inc., Ithaca, NY) were placed around the arteries and connected to a two-channel, small-animal, digital, transit-time blood flowmeter (model T206; Transonic Systems Inc.) and renal blood flow (RBF) and superior mesenteric blood flow (MBF) were continuously displayed. The left ureter was then cannulated with PE-10 tubing for collection of urine.

After completion of surgery, the 80-µl/min jugular infusion of 0.9% normal saline was replaced with [carboxyl-¹⁴C]inulin in 0.9% saline (0.5 µCi bolus followed by 0.035 µCi/min at 80 µl/min). After a 1-h stabilization period, three 20-min periods (periods 1-3) were conducted to collect urine and analyze local and systemic hemodynamics. After the end of the period 3, norepinephrine was begun at 1 µg/kg/min and continued for the next four periods (periods 4-7). At the end of period 7, norepinephrine was increased to 10 µg/kg/min and continued for the last three periods (periods 8-10). At the midpoint of each period, a 150-µl blood sample was collected via the carotid artery, for measurement of glomerular filtration rate (GFR) and hematocrit. Also during each clearance period, MABP and heart rate were time-averaged, and RBF and MBF were taken as the average of four measurements. Renal vascular resistance (RVR) and superior mesenteric vascular (MVR) were calculated as MABP/RBF and MABP/MBF, respectively. GFR was determined as the renal clearance of [carboxyl-¹⁴C]inulin, which was quantified by liquid scintillation counting (Tri-Carb model 2500TR; Packard Instrument Co., Meriden, CT) analysis of serum and urine samples.

Statistical significance for drug effects on hemodynamic and renal parameters was evaluated using repeated measures analysis of variance (two factor) for all animals in all time periods, using Sigma Stat 2.0 (Jandel Scientific, San Rafeal, CA). Statistical significance was considered at p < 0.05.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Table 1 depicts the measured physiologic parameters in rats that did not receive zymosan. Ro 20-1724 at 2.0 µg/kg/min significantly attenuated the increase in RVR (p = 0.04) and decrease in RBF (p < 0.01) that resulted from norepinephrine infusions when compared with control animals over time. Treatment with Ro 20-1724 at either 2.0 or 0.3 µg/kg/min did not result in differences in GFR, SMAVR, SMABF, MABP, or HR when compared with control.

Much of the previous work done with this zymosan MODS model used a dose of 1 mg/g, with a resultant 50% mortality. Because the purpose of our experiment was to examine the changes in organ perfusion and the effect of PDE4 inhibition on these changes, a sublethal dose of 0.25 mg/g was injected. This dose of zymosan caused a 33% decrease in RBF, a 50% decrease in GFR, and a 15% decrease in SMABF without mortality in our study.

In animals that received zymosan, urinary cAMP excretion was significantly increased in rats treated with Ro 20-1724 at 2.0 µg/kg/min (p = 0.05); animals that received PDE inhibition at lower doses had urinary cAMP levels midway between the high dose Ro 20-1724 and control, but not significantly different from controls (Table 2).

Animals that received zymosan and Ro 20-1724 at either dose had attenuated increases in RVR when compared with zymosan alone (p = 0.02 for low dose; p < 0.01 for high dose) (Fig. 1). The decreased RVR becomes more apparent as norepinephrine is added and subsequently increased (Fig. 1). With high-dose norepinephrine (10 µg/kg/min), there is little difference between the control group that received no zymosan and the animals that received zymosan but were protected with high-dose Ro 20-1724. Renal blood flow also reflects the protective effects of PDE4 inhibition in the high-dose Ro 20-1724 group (p < 0.01) when compared with zymosan alone (Fig. 2). Likewise, the decrease in GFR caused by zymosan was attenuated by high-dose Ro 20-1724 (p = 0.05), but not low-dose Ro 20-1724 (Fig. 3).

View larger version (25K): [in this window] [in a new window]   Fig. 1.   Effect of PDE4 inhibition with Ro 20-1724 on renal vascular resistance changes induced by zymosan and norepinephrine. Solid circles represent the control animals that did not receive either zymosan or Ro 20-1724; open triangles represent animals that received zymosan but not Ro 20-1724; solid squares represent animals that received zymosan and low-dose Ro 20-1724 at 0.3 µg/kg/min; open diamonds represent animals that received zymosan and high-dose Ro 20-1724 at 2.0 µg/kg/min.

View larger version (25K): [in this window] [in a new window]   Fig. 2.   Effect of PDE4 inhibition with Ro 20-1724 on renal blood flow changes induced by zymosan and norepinephrine. Solid circles represent the control animals that did not receive either zymosan or Ro 20-1724; open triangles represent animals that received zymosan but not Ro 20-1724; solid squares represent animals that received zymosan and low-dose Ro 20-1724 at 0.3 µg/kg/min; open diamonds represent animals that received zymosan and high-dose Ro 20-1724 at 2.0 µg/kg/min.

View larger version (27K): [in this window] [in a new window]   Fig. 3.   Effect of PDE4 inhibition with Ro 20-1724 on glomerular filtration rate changes induced by zymosan and norepinephrine. Solid circles represent the control animals that did not receive either zymosan or Ro 20-1724; open triangles represent animals that received zymosan but not Ro 20-1724; solid squares represent animals that received zymosan and low-dose Ro 20-1724 at 0.3 µg/kg/min; open diamonds represent animals that received zymosan and high-dose Ro 20-1724 at 2.0 µg/kg/min.

The splanchnic hemodynamics were adversely effected as well by the intraperitoneal zymosan, and PDE4 inhibition protected this flow in a similar manner as the kidney. High-dose Ro 20-1724 blunted the increase in SMAVR caused by zymosan and norepinephrine when compared with zymosan alone (p = 0.02), but low dose did not confer this protection (Fig. 4). SMABF was decreased with zymosan, as well as with the initiation of norepinephrine, and Ro 20-1724 at 2.0 µg/kg/min protected against this decrease (p = 0.05); however, Ro 20-1724 at 0.3 µg/kg/min did not alter the SMABF changes caused by zymosan (Fig. 5).

View larger version (31K): [in this window] [in a new window]   Fig. 4.   Effect of PDE4 inhibition with Ro 20-1724 on superior mesenteric vascular resistance changes induced by zymosan and norepinephrine. Solid circles represent the control animals that did not receive either zymosan or Ro 20-1724; open triangles represent animals that received zymosan but not Ro 20-1724; solid squares represent animals that received zymosan and low-dose Ro 20-1724 at 0.3 µg/kg/min; open diamonds represent animals that received zymosan and high-dose Ro 20-1724 at 2.0 µg/kg/min.

View larger version (27K): [in this window] [in a new window]   Fig. 5.   Effect of PDE4 inhibition with Ro 20-1724 on superior mesenteric blood flow changes induced by zymosan and norepinephrine. Solid circles represent the control animals that did not receive either zymosan or Ro 20-1724; open triangles represent animals that received zymosan but not Ro 20-1724; solid squares represent animals that received zymosan and low-dose Ro 20-1724 at 0.3 µg/kg/min; open diamonds represent animals that received zymosan and high-dose Ro 20-1724 at 2.0 µg/kg/min.

Systemic hemodynamics were also altered by zymosan and PDE4 inhibition. MABP was decrease by zymosan from 140 to 115 mm Hg, and remained relatively decreased when compared with control despite high doses of norepinephrine (Fig. 6). The group of animals that received high-dose Ro 20-1724 had lower MABP when compared with zymosan alone (p = 0.01), but there was no difference between zymosan alone and the low-dose Ro 20-1724 group (Fig. 6). Intraperitoneal zymosan also increased HR from 380 to 430 beats/min, and these changes remained with the addition of norepinephrine (Fig. 7). There was no difference between controls, zymosan alone, and either dose of Ro 20-1724 in regard to HR (Fig. 7).

View larger version (27K): [in this window] [in a new window]   Fig. 6.   Effect of PDE4 inhibition with Ro 20-1724 on mean arterial blood pressure changes induced by zymosan and norepinephrine. Solid circles represent the control animals that did not receive either zymosan or Ro 20-1724; open triangles represent animals that received zymosan but not Ro 20-1724; solid squares represent animals that received zymosan and low-dose Ro 20-1724 at 0.3 µg/kg/min; open diamonds represent animals that received zymosan and high-dose Ro 20-1724 at 2.0 µg/kg/min.

View larger version (26K): [in this window] [in a new window]   Fig. 7.   Effect of PDE4 inhibition with Ro 20-1724 on heart rate changes induced by zymosan and norepinephrine. Solid circles represent the control animals that did not receive either zymosan or Ro 20-1724; open triangles represent animals that received zymosan but not Ro 20-1724; solid squares represent animals that received zymosan and low-dose Ro 20-1724 at 0.3 µg/kg/min; open diamonds represent animals that received zymosan and high-dose Ro 20-1724 at 2.0 µg/kg/min.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The results of this study support the hypothesis that chronic PDE4 inhibition with Ro 20-1724 protects the kidney and mesentery against hypoperfusion from persistent as well as acute inflammation. The observation that urinary cAMP excretion is increased in animals that received Ro 20-1724 supports the hypothesis that PDE4 inhibition increases cAMP, leading to the attenuation of vasoconstriction that occurs in chronic inflammation.

The "gut-origin hypothesis" (Pastores et al., 1996) suggests that MODS occurs as a result of inadequate gut perfusion and leads to ischemia of the gastrointestinal tract, increasing permeability of the intestine and allowing bacteria to translocate to the circulation, thus triggering an uncontrolled inflammatory response that affects distant organ function (Landow and Andersen, 1994). This hypothesis is supported by animal and human studies (Carrico et al., 1986; Deitch, 1990; Barquist et al., 1998). Multiple organ dysfunction associated with acute renal failure has a mortality of over 50% (Wardle, 1994). During overwhelming sepsis, renal blood flow may be decreased (O'Hair et al., 1989; Frey et al., 1990; van Lambalagen et al., 1991). The increase in renal vascular resistance appears to occur as a result of the direct effects of the potent vasoconstrictor endothelin (Voerman et al., 1990). Delayed onset of acute renal failure with MODS is predictive of mortality. It stands to reason that therapy that can decrease the development of acute renal failure in these patients is likely to impact mortality.

Animal research aimed at understanding MODS had been hindered by the lack of a reproducible model. In 1986, Goris et al. (1986) described a bacteria-independent model of MODS by intraperitoneal injection of zymosan A, a major component of the cell wall of the yeast Saccharomyces cerevisiae. This resulted in a severe inflammatory response that led to histopathologic changes in the lung, kidney, and liver. One limitation of this study was that it lacked adequate evaluation of functional organ changes. These data were provided when Steinberg et al. (1989) validated this model by observing hypoxia, decreased creatinine clearance, and changes in hepatic cytochrome P450 content in animals receiving zymosan. This model is now used to help study the process of MODS. It is well known to consistently produce pathologic changes that mimic the changes that occur in human MODS. The model has three defined stages (Jansen et al., 1997; Ferrer et al., 1998). Stage 1 is the acute phase, which manifests as an acute generalized inflammatory response to sterile peritonitis that lasts for the initial 48 h. Stage 2 occurs over days 3 to 7 and represents the convalescent phase where animals improve clinically. Stage 3 is characterized by progression of clinical illness and organ dysfunction, with a second increase in the host's proinflammatory state. Most published data have used the dose of 1 mg of zymosan A per gram of body weight. This dose produces mortality of 20 to 60%, with animals expiring in the initial 48 h (stage 1) or after 8 days (stage 3) (Steinberg et al., 1989; Scalia et al., 1992; Jansen et al., 1996, 1997, 1998; Nieuwenhuijzen et al., 1997; Ferrer et al., 1998). Some authors have reported using various decrements in this dosing, and those who inject 0.5 mg/g report a mortality of 16 to 33% (Rooyackers et al., 1994; Deng et al., 1998). Because the purpose of our experiment was to examine the changes in organ perfusion and the effect of PDE4 inhibition on these changes, a sublethal dose of 0.25 mg/g was injected. This dose of zymosan caused a 33% decrease in RBF, a 50% decrease in GFR, and a 15% decrease in SMABF without mortality in our study.

Previously, we have reported that PDE4 inhibition protects against renal and mesenteric ischemia in an acute, 6-h model of inflammation (Begany et al., 1996; Carcillo et al., 1997). Our present article extends these results to chronic therapy in a 48-h model of inflammation. The mechanism by which PDE4 inhibition protects renal and mesenteric perfusion in our experiment is not completely understood. Increased cAMP levels are a likely mechanism. Two possible functions of cAMP-conferring protection are an inhibition of vasoconstriction and a reduction of inflammation.

Vasoconstriction occurs in renal and mesenteric vascular beds because of activation of phospholipase C. This binding in turn produces the second messengers inositol-1,4,5-trisphosphate and diacylglycerol, which stimulate the release of calcium from the sarcoplasmic reticulum and activate protein kinase C, respectively. Cyclic AMP is the major negative modulator of phospholipase C-induced vasoconstriction. In our experiment, we found that PDE4 inhibition and the subsequent increase in cAMP resulted in a reduction of norepinephrine-induced vasoconstriction. Therefore, Ro 20-1724 may protect the kidney and mesentery through modulation of signal transduction, even in the absence of inflammation. This mechanism may be the reason for the decrease in systemic blood pressure that we have reported, but this hemodynamic change is clinically well tolerated and is coupled with improved end-organ function.

cAMP is also a potent suppressor of inflammatory cell activity, and increases in intracellular cAMP will decrease the inflammatory response released by these cells (Bourne et al., 1974; Kammer, 1988). It is known that inflammatory cells of all species have a predominance of PDE4. Inhibition of PDE4 with specific PDE4 inhibitors decreases tumor necrosis factor and interleukin-6 production, and increases the levels of the anti-inflammatory cytokine interleukin-10 (Semmler et al., 1993; Kambayashi et al., 1995; Sekut et al., 1995; Verghese et al., 1995; Souness et al., 1996). This family of PDE inhibitors also decreases the respiratory burst of eosinophils (Dent et al., 1994) and inhibits formation of oxygen radicals (Chini et al., 1994). In whole-animal models, PDE4 inhibition protected against endotoxin-induced liver injury (Fischer et al., 1993) and improved survival after endotoxin exposure (Sekut et al., 1995). Therefore, it is possible that the effects we have observed in the renal and mesenteric vasculature occurred due to a down-regulation of the inflammatory process.

This study was an attempt to observe the effects of prolonged use of PDE4 inhibition in chronic inflammation. It was also an attempt to determine the range of dosing required in MODS to observe protection of the renal and splanchnic circulation while avoiding the hypotension we have previously observed (Carcillo et al., 1997; Herzer et al., 1998). We continue to see some small decrease in blood pressure with the use of PDE4 inhibition. When lower doses are used in this model to avoid any systemic hemodynamic effects, no protection of the renal or gastrointestinal systems is observed. We have previously shown that this decrease in blood pressure is a result of decreased preload, as opposed to the decrease in afterload observed with PDE3 inhibitors (Herzer et al., 1998). This hypotension should be treatable with fluid therapy to maintain adequate preload. Animal studies are planned to address this theory.

There are limitations to our study. First, in contrast to prior studies, the dose of zymosan chosen was relatively small. Although this was done to ensure that our mortality would approach zero, the decreases in RBF and especially SMABF were moderate. It is conceivable that, with higher doses of zymosan, the protective effect that we observed would be less. Alternatively, it is possible that we would observe an even greater effect if more vasoconstriction and inflammation were present. We also chose to examine the effects of PDE4 inhibition 48 h after the injection of zymosan. No conclusions can be drawn as to the effects of PDE4 inhibition at the later stages of MODS.

In conclusion, we have shown that PDE4 inhibition protects renal and mesenteric circulation during MODS. The mechanism of this protection is most likely 2-fold: PDE4 inhibitor's potent vasodilatory properties in combination with their strong anti-inflammatory effects. Which of these mechanisms is most important remains to be seen. PDE4 inhibitors may have a role in preventing secondary organ failure in systemic inflammatory insults.