Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Hemoglobin solutions have been investigated as an active oxygen-delivering vehicle since Amberson et al. (1949) first reported an application of Ringer-Locke solution containing hemoglobin on humans. Hemoglobin-based oxygen carriers (HBOC) have rational benefits, including universal compatibility and accompanying immediate availability, freedom from disease transmission, and long-term storage. Despite these virtues of HBOC, conventional red blood cell transfusion is used extensively in the treatment of various acute blood loss situations. This is partially because unmodified hemoglobin solutions were found to be nephrotoxic due to the presence of red cell stromal lipid, as a toxic contaminant (Rabiner et al., 1970). In addition, the use of unmodified hemoglobin solutions has caused untoward clinical reactions of varying intensity and severity, including severe gastrointestinal pain and systemic vasoconstriction (Amberson et al., 1949; Brandt et al., 1951; Miller and McDonald, 1951).

Recently developed HBOC are pure, have prolonged circulation time, and retain oxygen binding properties comparable to that of hemoglobin in intact red blood cells (Chatterjee et al., 1986; Gould and Moss, 1996). Diaspirin cross-linked hemoglobin (DCLHb), a product derived from outdated human red blood cells, has a cross-link between the two alpha chains induced by bis-3,5-dibromosalicyl fumarate (Chatterjee et al., 1986). This cross-link results in tetrameric hemoglobin, which renders DCLHb more stable than unmodified hemoglobin.

A number of studies have investigated the physiological effects of DCLHb, mainly on the cardiovascular system and demonstrated that it has a pressor effect mediated by nitric oxide (NO) scavenging (Schultz et al., 1993; Sharma et al., 1995), endothelin release (Schultz et al., 1993; Gulati et al., 1995), and/or adrenergic pathways (Gulati and Sharma, 1994). In the gastrointestinal tract, other modified hemoglobin solutions alter motor functions of the lower esophageal sphincter (Conklin et al., 1995), sphincter of Oddi (SO) (Cullen et al., 1996), and stomach (Hartman et al., 1998). These effects have been attributed to the scavenging of NO by these modified hemoglobin solutions. Abdominal pain has been reported during the application of various hemoglobin solutions (Amberson et al., 1949; Rabiner et al., 1970; Savitsky et al., 1978), and a risk of acute pancreatitis induced by DCLHb administration was reported recently (O'Hara et al., 1998). As NO has been demonstrated to be a major inhibitory neural mediator in the gastrointestinal and biliary system, DCLHb may alter NO neural transmission and produce significant changes in gastrointestinal and/or biliary motility. Previous studies, however, have only focused on the effects of DCLHb on the vasculature and microcirculation (Gulati et al., 1994; Frankel et al., 1996; Barve et al., 1997; Sen et al., 1997; Van Iterson et al., 1998). The effect of DCLHb on the motility of the gastrointestinal tract has not been reported.

This study was conducted to 1) determine the effects of DCLHb on duodenal, gallbladder, and SO motility and trans-sphincteric flow and 2) investigate the possible mediation of these effects by NO in the Australian Brush-tailed possum in vivo.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animal Preparation. Seventy-seven Australian Brush-tailed possums of either sex (1.2-3.2 kg) were used in this study. Animals were fasted for 18 h, and anesthesia was induced with intramuscular xylazine (Rompun, 5 mg/kg; Bayer Australia Ltd., Botany, NSW, Australia) and ketamine (Ketalar, 20 mg/kg; Parke-Davis Pty. Ltd., Caringbah, NSW, Australia) injections. The left femoral vein was cannulated, and a continuous infusion of sodium pentobarbitone (Nembutal, 15-45 mg/kg/h; Rhone Merieux Pty. Ltd., Pinkenba, QLD, Australia) was used to maintain anesthesia throughout the experimental period. Animals were intubated through a tracheostomy and mechanically ventilated using a small animal respirator (Phipps and Bird Inc., Richmond, VA). A constant infusion of saline (2-4 ml/kg/h) was delivered via the left femoral vein. Blood pressure was measured using a pressure transducer (Transpac IV, Abbott Critical Care Systems, Sligo, Ireland) via a catheter in the left femoral artery. Animal body temperature was maintained at 37°C with a homeothermic blanket (Harvard Apparatus Ltd., Edenbridge, Kent, UK).

Assembly for Measurements of Duodenal and Biliary Motility. The basic experimental assembly for measurements of biliary motility has been described previously (Saccone et al., 1992; Baker et al., 1996). Intraperitoneal access was gained by a mid-abdominal incision. An incision was made in the common bile duct 5 to 10 mm distal to the cystic duct, and three catheters were inserted (see Fig. 1). One end of a bile diversion tube, a polyvinyl chloride single lumen catheter (1.52 mm o.d., 0.86 mm i.d., length 25 cm), was inserted 2 mm toward the liver and secured with a ligature. The other end of this catheter was positioned in the distal duodenum and served to divert hepatic bile into the distal small intestine. The two other catheters, to measure trans-sphincteric flow and SO manometry, were inserted distally (toward the duodenum) and secured in position with a ligature.

View larger version (35K): [in this window] [in a new window]   Fig. 1.   A diagram of the preparation used in this study. The location of a sphincter of Oddi (SO) manometry catheter, trans-sphincteric flow catheter, bile diversion catheter, gallbladder balloon catheter, and duodenal strain gauge are illustrated. The pylorus and the duodenum, 4 cm anal to the SO, are ligated, and the duodenal and gastric drainage tubes are shown.

Data Acquisition. Arterial blood pressure, SO, gallbladder, and duodenal motility and trans-sphincteric flow were recorded using a MacLab analog/digital interface and Chart 3.5 software (ADInstruments Pty Ltd., Castle Hill, NSW, Australia).

Experimental Protocol. After a 30-min equilibration period, a test dose of 200 ng/kg cholecystokinin octapeptide (Auspep Pty. Ltd., Parkville, VIC, Australia) was given intravenously to confirm that the SO, gallbladder, and duodenal recordings were satisfactory. The post-cholecystokinin octapeptide recordings returned to baseline within 15 min and an additional 45-min re-equilibration period was allowed before DCLHb or human serum albumin (HSA) administration.

Nitric Oxide Synthase Inhibition. Inhibition of NO synthase was achieved by N-nitro-L-arginine methyl ester (L-NAME, Sigma Chemical Co., St. Louis, MO) with a 15 mg/kg bolus i.v. followed by 6 mg/kg/h continuous infusion. In five separate animals, the effect of the L-NAME on SO motility and trans-sphincteric flow was examined. Measurements of SO motility and trans-sphincteric were performed as described above. In another 11 animals, DCLHb or HSA (1 ml/kg/min for 5 min) was then administered 45 min after commencement of L-NAME infusion. Mean arterial blood pressure, SO trans-sphincteric flow, and duodenal motility were recorded for the following 180 min.

Data and Statistical Analysis. Data were analyzed for 15-min intervals from the period 30 min before the infusion of DCLHb, HSA, or L-NAME to 360 min postinfusion or 180 min for NO synthase inhibition studies. Mean arterial blood pressure (mm Hg), duodenal contraction frequency (contractions/15 min), trans-sphincteric flow (µl/s), SO basal pressure (mm Hg), and area under the curve of phasic contractions (mm Hg · s, an index of motility) for SO and gallbladder recordings were calculated using the Chart 3.5 software. Group data were expressed as mean ± S.E.M. Statistical comparison of the DCLHb and HSA data utilized repeated measures ANOVA, with the averaged preinfusion responses used as a covariate (SPSS 9.0.1, SPSS Inc., Chicago, IL), and P < 0.05 was considered to be significant. Log transformations were used to stabilize the SO basal pressure variance before conducting the analyses of variance.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Mean Arterial Blood Pressure. DCLHb increased mean arterial blood pressure in a dose-dependent manner (P < 0.05). Mean arterial blood pressure displayed an immediate increase, which usually started within 1 min and persisted over 1 h after DCLHb application (Figs. 2A, 2B, and 3A). The mean increase of blood pressure with 500 mg/kg DCLHb infusion was 43 ± 5 mm Hg. Administration of L-NAME alone caused an increase in mean arterial blood pressure, which peaked immediately with a maximum change of 28 ± 3 mm Hg (P < 0.05) (Fig. 4). In conjunction with L-NAME treatment, DCLHb administration did not display any significant change in mean arterial blood pressure (P > 0.05) compared with the HSA control group (Fig. 4).

View larger version (56K): [in this window] [in a new window]   Fig. 2.   Representative profiles of arterial blood pressure, duodenal motility, trans-sphincteric flow, sphincter of Oddi (SO), and gallbladder motility. All parameters are presented in pairs of DCLHb and HSA recordings from separate animals. The effects over the 6-h period are shown on the left (A) and 20-min periods incorporating 5-min control period, the DCLHb, or HSA administration period and the following 10-min period are shown on the right (B). DCLHb (10%) or HSA were infused at 1 ml/kg/min for 5 min (500 mg/kg), as indicated by the rectangular boxes. The change in trans-sphincteric flow rate is illustrated by the horizontal bars (B). The length of the bar indicates the time required for 1 ml of saline to pass through the SO during the control period. This bar is reproduced under the trans-sphincteric flow output following DCLHb or HSA administration. Following DCLHb, the bar no longer occupies the space beneath the trans-sphincteric flow output, indicating slower trans-sphincteric flow rate. With DCLHb application, arterial pressure displayed a rapid and short-term increase (A, B), whereas the duodenal motility showed a prolonged increase with a delayed onset (A). The trans-sphincteric flow showed a small, rapid, but transient decrease (B). SO motility did not show an obvious change. In this experiment, a short-lived increase in gallbladder motility was also evident.

View larger version (24K): [in this window] [in a new window]   Fig. 3.   Mean arterial pressure (A) and duodenal contraction frequency (B) in response to three doses of DCLHb and HSA. DCLHb or HSA was given at time 0 at 1 ml/kg/min for 1, 2.5, or 5 min to produce doses of 100, 250, or 500 mg/kg. DCLHb produced a dose-dependent increase in mean arterial pressure (P < 0.05). The effect was immediate and short-lived. Duodenal contraction frequency was increased at doses of 250 and 500 mg/kg with a delayed onset (P < 0.05) but was not dose-dependent. The data are presented as mean ± S.E.M. for 15-min intervals. The arrow and broken line indicate the interval commencing with the administration of DCLHb or HSA (0 time). ( = DCLHb,  = HSA; n = 6 except for HSA of dose 500 mg/kg in A and dose 100 mg/kg in B: n = 5).

View larger version (28K): [in this window] [in a new window]   Fig. 4.   Mean arterial pressure, duodenal contraction frequency, and trans-sphincteric flow in response to DCLHb or HSA in the presence of N-nitro-L-arginine methyl ester (L-NAME). L-NAME treatment (15 mg/kg bolus i.v. followed by 6 mg/kg/h continuous infusion) produced increases in mean arterial pressure and duodenal contraction frequency and a decrease in trans-sphincteric flow (P < 0.05). Subsequent application of DCLHb (500 mg/kg) did not display further effects in these parameters compared with the HSA controls. The data are presented as mean ± S.E.M. for 15-min intervals. The arrow indicates the interval commencing with the L-NAME administration, and the arrowhead indicates the interval commencing with the administration of DCLHb or HSA. ( = DCLHb,  = HSA; n = 6 except for DCLHb of duodenal contraction frequency: n = 5).

Duodenal Contraction Frequency. DCLHb increased duodenal contraction frequency (P < 0.05), however this was not dose-dependent with the protocol adopted. Duodenal contraction frequency increased with a 15- to 30-min delay after DCLHb application and was sustained for approximately 270 min postinfusion with the middle and high DCLHb doses (Figs. 2A and 3B). A large degree of variation in the magnitude of the activity in both the HSA control and DCLHb treatment groups was noted. The mean increase in duodenal contraction frequency was 27 ± 7 contractions/15 min with the highest dose of DCLHb. Administration of L-NAME alone caused an increase in duodenal contraction frequency (P < 0.05) by 72 ± 16 contractions/15 min (Fig. 4). DCLHb administration with concurrent L-NAME treatment did not produce any significant change in duodenal contraction frequency compared with the HSA control group (Fig. 4).

Trans-Sphincteric Flow, SO, and Gallbladder Motility. DCLHb produced a small decrease (5 ± 1 µl/s) in trans-sphincteric flow (P < 0.05) at the highest dose, which was rapid in onset, but short-lived (Figs. 2B and 5A). However, DCLHb did not show a dose-dependent effect on trans-sphincteric flow. The administration of L-NAME alone caused a decrease in trans-sphincteric flow (P < 0.05) (Fig. 4), whereas the subsequent administration of DCLHb during NO synthase inhibition did not result in any significant change in trans-sphincteric flow (P > 0.05) compared with the HSA control group (Fig. 4).

View larger version (46K): [in this window] [in a new window]   Fig. 5.   Shown are trans-sphincteric flow, sphincter of Oddi (SO) basal pressure, area under the curve of SO, and area under the curve of gallbladder motility in response to three doses of DCLHb and HSA. DCLHb or HSA was given at time 0. DCLHb induced a small but immediate and short-term decrease in trans-sphincteric flow (P < 0.05, A), however, no dose-response relationship was observed. In the other parameters, the DCLHb group did not produce significant change compared with the HSA control (B, C, and D). The data are presented as mean ± S.E.M. for 15-min intervals. The arrow indicates the interval commencing with the administration of DCLHb or HSA. ( = DCLHb,  = HSA; n = 6 except for HSA of dose 100 and 500 mg/kg in A, dose 500 mg/kg in B, dose 100 and 500 mg/kg in C and dose 100 and 500 mg/kg in D: n = 5).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

This study demonstrates for the first time that DCLHb causes an increase in duodenal contraction frequency and a minor decrease in SO trans-sphincteric flow. As expected, mean arterial blood pressure was elevated following DCLHb administration. The increase in duodenal contraction frequency and the decrease in trans-sphincteric flow were not dose-dependent, however, the increase in mean arterial blood pressure was. Inhibition of NO synthase with L-NAME produced similar changes in these three parameters, and no further changes occurred with subsequent DCLHb administration, implicating a major role for NO in these effects.

The DCLHb effect on mean arterial blood pressure demonstrated in this study is in agreement with a number of previous studies in other species (Sharma et al., 1995; Barve et al., 1997; Sen et al., 1997). Our data indicate that the dose-dependent response of blood pressure to DCLHb in the Australian possum is similar to other animals (such as rats and pigs) and humans. Scavenging of NO by DCLHb is proposed as a major underlying mechanism of this change in blood pressure (Schultz et al., 1993; Sharma et al., 1995), whereas release of endothelin and/or catecholamines have also been suggested (Gulati and Sharma, 1994; Gulati et al., 1995).

The effects of DCLHb on duodenal, SO, and gallbladder motility were determined for the first time in this study. Previous studies of DCLHb have focused on the regional blood flow and vascular resistance in the gastrointestinal tract and showed an increase in blood flow in the stomach and small intestine (Przybelski and Daily, 1994; Sharma et al., 1995). Although the increase in regional blood flow may affect duodenal motility indirectly, it is unlikely that this is the major mechanism whereby DCLHb increases duodenal motility. Previous reports in other species have shown that the increase in regional gastrointestinal blood flow occurred only for 30 min after DCLHb administration. This time period of increased regional blood flow is far shorter than the duration of changes in motility we observed in the possum duodenum. The increase in duodenal contraction frequency after DCLHb administration was not observed during L-NAME treatment and, moreover, the L-NAME treatment itself produced a similar increase in duodenal contraction frequency. These findings are consistent with DCLHb acting predominantly as an NO scavenger.

The effects of DCLHb on the biliary system were minor. The trans-sphincteric flow showed a transient decrease of approximately 20% after DCLHb administration. This change was not reflected by significant changes in SO basal pressure or SO motility (area under the curve). Our previous studies have shown that cumulative changes of several SO motility parameters, particularly SO basal pressure and contraction frequency, contribute to changes in trans-sphincteric flow (Liu et al., 1992; Saccone et al., 1992). Consequently, significant changes in individual SO motility parameters, including SO basal pressure, may not occur, as seen in this study, yet the combined changes may be responsible for significant changes in trans-sphincteric flow. The effect of DCLHb on biliary motility has not been examined in other species and is smaller than that reported with another HBOC, recombinant human hemoglobin (rHb1.1) (Cullen et al., 1996). This recombinant molecule (rHb1.1) increased SO motility in the American opossum, particularly elevating the sphincter basal tone. The effects of DCLHb on the biliary system are weaker than those observed with rHb1.1 and are not likely to be physiologically significant. These differences in the magnitude of responses may reflect differences in the nitric oxide binding capacity of DCLHb and the recombinant molecule, or species differences. Similarly, the effects of DCLHb on trans-sphincteric flow and sphincter of Oddi basal pressure were weaker than the small changes in these parameters produced by L-NAME administration alone. These differences could be due to a greater reduction in the NO concentration within the SO following NO synthase inhibition compared with the reduction in the NO concentration following DCLHb administration. The small decrease in the trans-sphincteric flow and increase in SO basal pressure demonstrated by the L-NAME treatment alone and the lack of any further decrease after DCLHb application is also consistent with DCLHb acting mainly as a NO scavenger.

Although our findings in the cardiovascular system agree with other studies suggesting that DCLHb is acting as an NO scavenger (Schultz et al., 1993; Sharma et al., 1995), other mechanisms may be involved. Some studies have implicated endothelin (Schultz et al., 1993; Gulati et al., 1995) and/or the adrenergic pathway (Gulati and Sharma, 1994) in some of the actions of DCLHb in the cardiovascular system. These mechanisms could also operate in the gastrointestinal and biliary systems but may not be evident against the background of similar changes induced by L-NAME treatment.

Clinical investigations of DCLHb have suggested that, in some settings, this molecule may induce acute pancreatitis (O'Hara et al., 1998). However, other studies using an ischemic reperfusion model of acute pancreatitis in the rat demonstrated that DCLHb does not exacerbate the inflammatory response or induce microcirculatory pancreatic damage (Von Dobschuetz et al., 1999). We have reported preliminary studies suggesting that DCLHb decreased pancreatic exocrine secretion and did not change serum amylase levels for 6 h after DCLHb application (Irvine et al., 1999). In addition, histological examination of pancreatic tissue following 6 h of DCLHb administration showed minimal changes, which were no different from the changes seen in pancreatic tissue from the HSA control group (D. E. Gordon, unpublished data). Moreover, the present study has shown no significant changes in SO motility that can be associated with acute pancreatitis. Consequently, these findings in normal possums suggest that DCLHb would not induce acute pancreatitis. We have shown that the effects of DCLHb in the possum are similar to those induced by L-NAME. We do not know if L-NAME induces pancreatitis in the possum, but L-NNA, another NO synthase inhibitor, has been tested in rats and shown to induce edematous pancreatitis (Konturek et al., 1994). L-NAME has been shown to exacerbate cerulein and intraductal glycodeoxycholic acid-induced pancreatitis (Molero et al., 1995; Werner et al., 1997, 1998), but, conversely, it reduced biochemical indices of acute hemorrhagic pancreatitis (Dabrowski and Gabryelewicz, 1994).

In humans, however, it is still unclear if DCLHb increases the risk of acute pancreatitis. The evidence implicating a role for NO in acute pancreatitis is conflicting, with some studies suggesting that NO may ameliorate acute pancreatitis by increasing blood flow (Werner et al., 1998) and/or secretion (Molero et al., 1995), whereas other studies suggest that NO potentiates pancreatic oxidative stress (Dabrowski and Gabryelewicz, 1994; Al-Mufti et al., 1998). A cause of DCLHb-induced acute pancreatitis could be related to the NO scavenging property of this molecule. Recently, new HBOC (rHb3011 and rHb4), which have reduced NO scavenging properties, have been developed (Doherty et al., 1998; Hartman et al., 1998). These new products were reported to produce smaller effects on lower esophageal sphincter function and gastric emptying than previous recombinant hemoglobin molecules.

It is well established that acute pancreatitis is associated with gallstones. In this study we noted that DCLHb induced gallbladder contraction in some animals. If a similar response occurs in humans with gallstones, one might speculate that gallbladder contraction could result in the migration of gallstones into the ampulla, leading to obstruction and the onset of acute pancreatitis. On the other hand, gallbladder contraction in a normal gallbladder may be beneficial, if the SO resistance to flow is low, as this would decrease gallbladder stasis and consequently decrease the likelihood of stone formation.

In conclusion, our results showed that DCLHb increases mean blood pressure and duodenal motility, with minor changes in biliary motility. These DCLHb-induced changes were not evident during NO synthase inhibition, suggesting that the actions of DCLHb are primarily due to NO scavenging. This study does not provide any evidence to support the hypothesis that DCLHb could increase the risk of inducing acute pancreatitis.