Journal of Pharmacological and Toxicological Methods 44 (2000) 489 ± 505 A novel procedure for daily measurements of hemodynamical, hematological, and biochemical parameters in conscious unrestrained rats Andre Blouin, SteÂphanie Molez, Dung Pham, Bilal Ayach, Patrick Dussault, Emmanuel Escher, Arco Y. Jeng, Bruno Battistini* Centre de recherche, HoÃpital Laval, Institut de cardiologie et de pneumologie, Department of Medicine, Laval University, Ste-Foy, QC, Canada G1V 4G5 Institute of Pharmacology, Faculty of Medicine, Universite de Sherbrooke, Sherbrooke, QC, Canada J1H 5N4 Metabolic and Cardiovascular Diseases, Novartis Institute for Biomedical Research, Summit, NJ 07901, USA Received 10 July 2000; accepted 15 December 2000 Abstract Accurate and chronic measurements of various parameters in conscious animals are fundamental for depicting pathological chronic conditions and their etiology in many experimental models, but they are often difficult to achieve. The aim of the present work was to develop and describe step-by-step a reproducible surgical procedure and daily manipulations for continuous, chronic use of conscious rats as models towards a better understanding of various cardiovascular and renal diseases and the testing of novel pharmacological drugs. The complete apparatus involved the use of a series of specialized devices (harness, rotating swivel, revolving arm) supporting a flexible, permanently implanted vascular catheter into the left femoral artery up to the abdominal aorta connected to a miniaturized individual peristaltic pump for delivering fluid at a constant rate. Such a set-up also enabled easy, quick, and reproducible daily blood sampling for the evaluation of more than 20 parameters, including the monitoring of heart rate (HR) and blood pressure in freely moving conscious rats. The overall success and survival rate reached 98% over 14 days and could be extended further. This model represents a much needed and valuable advance in surgical research techniques to evaluate the hemodynamic, hematological, biochemical, pharmacokinetic, and toxicological profile of any new drugs over time in conscious animal models such as rats. What makes this procedure satisfactory is the long-term reliable arterial access and reproducibility of the methodological approach for accurate and continuous measurements, minimizing the stress or invasiveness associated with the use of currently employed systems. D 2001 Elsevier Science Inc. All rights reserved. Keywords: Chronic catheterization; Infusion; Blood sampling; Conscious; Rat; Blood pressure; Bilirubin; Creatinine; Endothelin; Nitric oxide 1. Introduction Advances in molecular biology and cloning have been phenomenal over the past decade and have undoubtedly opened a new era of pharmaceutical discovery and gene therapy in cardiovascular diseases (Feldman, Tahlil, & Steg, 1996; French, 1998; Hajjar, del Monte, Matsui, & Rosenzwig, 2000). Nevertheless, any novel drug emerging from such an approach will have to undergo a series of preclinical pharmacological, pharmacokinetic, and toxicological studies before entering clinical trials. The development of * Corresponding author. Centre de recherche, Laval Hospital, Institut de cardiologie et de pneumologie, 2725 Chemin Ste-Foy, Ste-Foy, QC, Canada G1V 4G5. Tel.: +1-418-656-8711, ext. 2614; fax: +1-418-6564509. E-mail address: bruno.battistini@med.ulaval.ca (B. Battistini). new therapies and/or drugs is a long process that begins with in vitro and in vivo testing. At present, very few key publications fully addressed the technical complexity of creating and using on a daily basis an accurate rat model for the continuous, chronic administration of substances, blood samplings, and hemodynamic measurements in conscious, unrestrained rats. It is known that basal levels of most hormones and cardiovascular parameters are influenced by anaesthesia and surgical trauma (Depocas & Behrens, 1977; Lestage et al., 1985), and that simple handling of animals may have profound effects. About 30 years ago, Buckle and Nathanielsz (1974), and even before that (Still & Whitcomb, 1956), described a dual catheter system inserted into the left carotid artery and jugular vein for simultaneous infusion of drugs and sampling of blood, respectively, in unrestrained rats, but for a very short period. Three years later, a publication by Robert- 1056-8719/01/$ ± see front matter D 2001 Elsevier Science Inc. All rights reserved. PII: S 1 0 5 6 - 8 7 1 9 ( 0 1 ) 0 0 1 0 8 - 3 490 A. Blouin et al. / Journal of Pharmacological and Toxicological Methods 44 (2000) 489±505 Ross (1977) described the measurement of blood pressure in unrestrained rats. The main concern was, and still is today, the rapid loss of cannula patency. These approaches constituted the first attempts using a polyethylene (PE) catheter inserted into the left carotid artery for blood sampling and positioned on the rat's skull with screws and dental cement. When ready, the tip of the PE tubing was cut off and connected to a recording system. That approach involved resealing the tip of the catheter after each single measurement or collection. With such approach, catheters remained patent for more than 20 days. Similarly, a technique from Brandstaetter and Terkel (1977) using a double jugular vein cannulation remained patent for 16 days on average; the success was attributed to daily cannula care and filling the volume of each catheter with heparin. A number of methodologies describing vascular access in the rat was also summarized by Burt, Arbeit, and Brennan (1980). Nevertheless, a review on 21 technical procedures (9 arterial and 12 venous) for vascular access and collection of blood in rats suggested the lack of reliability and reproducibility regarding multiple blood samplings in unrestrained rats (Cocchetto & Bjornsson, 1983). Most described conventional methods involving a direct vascular access described acute short-term studies (e.g., 24-h survival), associated with catheterization using a vascular-access-port surgically implanted subcutaneously (Epstein, Branum, Cucchiaro, & Meyers, 1990; Paulose & Dakshinamurti, 1987). Innovative methods were described during the 1990s involving radiotelemetry (small implants capable of measuring blood pressures directly relayed by microwaves) (Anderson et al., 1999; Balakrishnan, Tatchum-Talom, & McNeill, 1998; Bazil, Krulan, & Webb, 1993; Brockway, Mills, & Azar, 1991; Deveney, Kjellstrom, Forsberg, & Jackson, 1998; Guiol, Ledoussal, & Surge, 1992; Kharidia & Eddington, 1996; Van den Buuse, 1994). To date, more than 325 publications on surgical methods related to monitoring of conscious rats have appeared since 1962. Even though complex and expensive to establish and operate, some procedures enabled the recording of blood pressure, heart rate (HR), and peripheral vascular blood flow for periods of up to 6 months with radiotelemetry. However, none of them assessed blood hematological, electrochemical, and biochemical profiles of these same conscious laboratory rats on a daily basis or at random time postoperatively. Importantly, despite many relatively innovative methods (Burt et al., 1980), very few reported the ability to achieve prolonged period of repeated blood samplings without experiencing clot and/or embolism (Giner, Snyder, & Meguid, 1987). This paper describes in detail a reproducible method for prolonged and continuous infusion of sterile saline, new drugs, or respective vehicles. The femoral artery is used as an insertion site for the positioning of a sterile vascular catheter, avoiding the formation of clotted blood while connected to a microperfusion pump. This protocol allows daily blood collection for measurements of various hema- tological and biochemical parameters, as well as recording the hemodynamic profile for comparison to baseline in freely moving rats. 2. Material and methods 2.1. Presurgical setup, anaesthesia, surgical procedures, and postoperative care Adult male Sprague ± Dawley (Crl:CD(SD)BR) rats (420 ±425 g; 13 ±14 weeks of age) were purchased from Charles Rivers (St. Constant, QC, Canada). Upon arrival, each rat was isolated for 5 days in the same cage (Fig. 1a,b) that will be used for the chronic experiment toward adaptation to its new environment. The animals were housed according to the local (Ethic Committee of Laval University) and national [Canadian Council on Animal Care (CCAC)] guidelines on animal welfare under a 12-h cycle of day/night, with free access to drinking water and fed ad libitum. To insure no infection in these animals for the entire duration of chronic treatment, all materials (surgery tools, catheters, swivels, etc., see Section 2.2) were sterilized by autoclave. All surgical procedures were performed in compliance with the above guidelines in an aseptic environment. 2.1.1. Step 1 Five days after their arrival, the rats were anaesthetized with a mixture of ketamine ±xylazine [freshly prepared ketamine (3.75 ml; Rogar-STB, London, ON, Canada), xylazine (0.50 ml; Bayer, Etobicoke, ON, Canada), sterile water (5.75 ml), giving a final solution of 100 mg/ml of each anaesthetic; 0.200 ml/100 g of rat body weight administered intraperitoneally). This dose was sufficient to allow the complete installation of the catheter into the animal while the other end was already attached to the infusion system. During the surgical procedure, the rats were kept at a constant body temperature of 37°C using a homeothermic blanket controlled by a probe inserted rectally into the rat (Harvard Apparatus, St. Laurent, QC, Canada). 2.1.2. Step 2 The nape of the neck and the anterior left leg regions were shaved and disinfected with 70% alcohol and 10% (v/v) proviodine (Rougier, Chambly, QC, Canada). A 3-mm incision was made between the shoulder blade for the exit of the vascular catheter prior to positioning the animal on its back. The legs were then attached and stretched down. A sterile field was created around the animal, and the coil and swivel, to which the saline-filled sterilized vascular catheter is already attached, were approached to the animal (see Section 2.2). An oblique incision of 5 mm was made in the left inguinal site, in the concavity between the abdomen and the leg (Acland, 1980). Insertion of the closed tip of a scissor separated the fat pad and thin connective tissues over A. Blouin et al. / Journal of Pharmacological and Toxicological Methods 44 (2000) 489±505 the common femoral vessels and nerve. The perivascular sheet was picked up, separating the artery and vein; the nerve was gently pulled aside. The femoral artery was isolated from the vein over 10 mm and prepared for the insertion of the sterile catheter. Two silk 4.0 (Ethicon Suture, Peterborough, ON, Canada) threads were positioned around the artery, one tied distally and the other one was used to tie the catheter after insertion of the catheter in the artery, away from the epigastric vessels nourishing the fat pad. The artery was elevated using a 3  20-mm plastic flat rod prior to arteriotomy created as an insertion point for the flexible vascular catheter. A few drops of lidocaine hydrochloride (2%, AstraPharma, Mississauga, ON, Canada) was used to dilate the vessel. The inner part of the sterile urethane vascular catheters is already coated with an antithrombogenic film that improves blood compatibility (PhysioCath; cat no. 277-0011-002; 0.76-mm outer diameter  80-cm length; Data Sciences International, Saint Paul, MN, USA). The uncut end of the catheter was inserted 40 mm deep inside the femoral artery and positioned in the abdominal aorta. The other end was connected to the swivel and subsequently to a mini-peristaltic pump (see below), for administering heparinized (15 IU/ml; Hepalean, WyethAyerst, Toronto, ON, Canada) sterile saline (0.9% NaCl; Baxter, Chicago, IL, USA). Arterial blood was allowed to flow out, and a brief purge pushed it back into the vessel. The catheter was then permanently attached using the silk threads criss-crossing over the artery and the catheter. Afterwards, it was then firmly anchored with a 5-0 prolene thread directly into the muscle to retain and secure the catheter into that position. The wound was hydrated with sterile saline all along the procedure. 2.1.3. Step 3 The animal was repositioned to show dorsal upper area without moving the back limbs. A 17-cm long semirigid protective coil (3-mm inner diameter  4-mm outer diameter; Advanced Cardiovascular Systems, Mountain View, CA, USA), was tunneled subcutaneously from the dorsal site to the opening made in the left leg femoral surgical site (see above). The vascular catheter was then clamped close to the femoral artery, disconnected from the swivel, and then passed through the semirigid coil up to the scapular incision. The vascular catheter (total length from the wound: 66 cm), while clamped, was pulled upward until leaving a small loop at the inguinal site. The vascular catheter was clamped, the coil was removed, and the catheter reconnected to the swivel. 2.1.4. Step 4 Closure of the wound with sutures constituted a critical step, knowing that the rat can gnaw everything. The inguinal wound was cleaned with sterile saline prior to closure. At the femoral site, the fat pad was reattached with 5-0 vicryl thread (Ethicon Suture) by interrupted sutures, and the subcutaneous tissues were also reconnected. The 491 superficial skin edges were closed by external interrupted mattress sutures with surgical steel monofilament (5.0 Ethicon). The closed wound was cleaned again with sterile saline and ethanol. After drying, a tissue adhesive spray (Vetbond, cat no. 1469, 3M, Saint Paul, MN, USA) was applied only on the wound line. 2.1.5. Step 5 The animal was repositioned on its belly to expose the dorsal upper area. The small opening in the scapular region was closed with one knot of 5-0 prolene; both ends of the same thread were used to anchor the vascular catheter at once. The vascular catheter was then clamped, disconnected from the swivel, inserted through the Covance infusion harness (see below) and passed through the 30cm long stainless steel spring stock protector, and reconnected to the swivel that was attached to the counterbalanced lever arm and pump. The clamp was removed to reestablish the flow from the pump. The harness was then positioned around the rat. 2.1.6. Step 6 The rat was put back into the cage. A heparinized sterile saline solution was constantly flowing from a syringe reservoir through the tubing that connected sequentially to a microperistaltic pump, the swivel, under the skin of the dorsal upper area and into the femoral artery. The animals received butorphanol tartrate (2 mg/kg, rat body weight, sc; 10 mg/ml; Ayerst Labs, Montreal, QC, Canada) as an analgesic following surgery. Rat body weights were monitored every morning before blood collection to assess the growth of the animals. The harness was loosened whenever necessary to insure a minimum level of stress and/or minimal skin irritation. The volume of fluid delivered by the low-flow peristaltic pump was measured each day. 2.2. Technical apparatus Each rat was put into a round metabolic cage (30-cm inner diameter  37.5-cm high) (cat no. 61-0042; Harvard), equipped with a 250-ml water bottle and a side-arm feeder (cat no. 61-0049; Harvard) and topped with a slotted cover and a kidney-shaped opening (cat no. 61-0044; Harvard). The rat was equipped with a flexible Covance infusion harness (model no. CIH95; cat no. 61-0041, Instech Labs, Plymouth Meeting, PA, USA) made of adjustable C-flex tubing (cat no. 6424-62; 1/16 in. inner diameter; 1/8 in. outer diameter; 1/32 in. wall; Cole-Parmer Institute, Vernon Hills, IL, USA) and a central vented dome or tether of soft molded elastomer to protect the vascular catheter (Fig. 1a). The 17-cm C-flex tubes were pulled through a central hole on the dorsal top to tighten the harness around the animal. Front legs were fitted between the two belly bands, crossing over the thorax. Excess tubing was trimmed, but a certain length was preserved for readjustment purpose following the animal growth. 492 A. Blouin et al. / Journal of Pharmacological and Toxicological Methods 44 (2000) 489±505 A. Blouin et al. / Journal of Pharmacological and Toxicological Methods 44 (2000) 489±505 493 Fig. 1. (a) The complete and detailed schematic diagram and (b) photograph of the proposed apparatus used for chronic arterial infusion and daily blood collection in conscious unrestrained rats. The cage, swivel, lever arm, etc. are commercially available. 494 A. Blouin et al. / Journal of Pharmacological and Toxicological Methods 44 (2000) 489±505 Fig. 2. Daily volume of intra-arterial infusion of heparinized saline in conscious unrestrained rats in Group 1 (O, n = 4) and Group 2 (O, n = 5) expressed in milliliter/day. * Significant difference between Groups 1 and 2 on the indicated day; zsignificant variation within Group 1 or 2 over time. The vascular catheter was passed through the dome into the stainless steel spring stock protective cannula (cat no. PS95, Instech) and connected to a 22-gauge single channel SS swivel (0.016 in. inner diameter; cat no. 61-0001, Instech) with a minimum dead volume (10 ml). The swivel was able to rotate 360° to keep the cannula from tangling. It was attached to a light and very responsive lever arm using an adjustable spring as the counterbalance (cat no. 61-0023, Instech). The arm constantly moved with the animal, 360° around, with a 30° maximum downward allowed by the stopper, to prevent slack or disconnection. A tension was created until the desired upward force was reached in order to counterbalance the weight of the apparatus exerted on the rat. A 12-cm sterile, non-radiopaque Tygon tubing (cat no. 95609-18; 0.0200 in. inner diameter  0.0920 in. outer diameter; Cole-Parmer Institute) was connected to the upper part of the swivel, toward a 22G connector attached to the lever arm, as a site for blood collection. From this connection, an upward 20 cm of the Tygon tubing was then connected to a single channel 22G connector (cat no. 22, Instech) and silicon tubing (cat no. 61-0241, Instech) inserted into the three rolling barrels of a low-flow peristaltic pump (model P720; cat no. 61-0098, Instech). The lowflow miniature peristaltic pump was attached via a rod mounting clamp (cat no. 61-0106, Instech) to a pole (cat no. 866, CDMV, Saint Hyacinthe, QC, Canada). A Tygon tubing (20 cm) from the pump was inserted to a blunted 22-gauge needle connected to a three-way stopcock (Medicis, St. JeÂrome, QC, Canada) attached to a 10-ml PE syringe acting as a reservoir for the heparinized saline (eventual drugs or respective vehicles), fitted with a filtered top. The junction between the vascular catheter and the SS swivel, was reinforced with a 10-mm piece of a 5Fr tube (Med-RX, Mississauga, ON, Canada) to secure the connection and prevent any leak that would compromise the integrity of the system. The total volume within the vascular catheter system from the femoral artery insertion point up to the syringe reservoir is 350 ml. Each component of the assembled system was verified for its integrity; any leak may compromise the procedure, such as causing thromboembolism in the vascular catheter. When this happens, the animal and undergoing experiment may nevertheless be salvaged (see below). The volume delivered by these pumps may vary slightly at this very low speed. Due to multiple uses (for example, two to three consecutive sets of experiments of 14 days Fig. 3. (a) MABP (mmHg) and (b) heart rate (bpm) of conscious unrestrained rats upon infusion with saline and blood collections, with or without reinfusion of blood cells, over time. Group 1 (O; n = 5) and Group 2 (O; n = 5). zSignificant variation within Group 1 or 2 over time. A. Blouin et al. / Journal of Pharmacological and Toxicological Methods 44 (2000) 489±505 each), the three rolling barrels stretched the channel tubing and further affected volume delivery. Therefore, for maximal accuracy, each pump must be precalibrated for 24 ±48 h at a speed of 0.250 ml/h (1 drop/4 min) prior to use in the rat. In case of power failure, the pump could operate with a 30-h internal lithium battery. 2.3. Constant perfusion In the present experiment, no drug was administered. A continuous low flow of heparinized (15 IU/ml  6 ml/ day = 90 IU/day) sterile saline was administered to prevent the formation of blood clots and/or embolism within the vascular catheter or at its tip within the femoral artery. Such formation would impair hemodynamic measurements and the collection of blood twice a day. We considered and used this low dose of heparin as long as it did not complicate eventual pharmacological and/or pharmacokinetic studies, 495 as mentioned before (Wood, Shand, & Wood, 1979). If heparin poses problems, such as affecting lipoprotein lipase activity and therefore, plasma levels of TG when high dose is used, another anticoagulant may be used, or simply saline. Since the completion of the present series, we noticed that a continuous low flow of saline through the vascular catheter and within the femoral artery is enough to prevent the formation of blood clots and/or embolism. Overall parameters were not affected in any way (data not shown). 2.4. Blood collection Blood (1.2 ml) was collected twice daily, 12 h apart at 8:00 and 20:00 h, in Groups 1 and 2. In Group 2, blood samples collected at 2000 h were centrifuged, the plasma was removed and the cell pellet was resuspended in sterile saline and reinjected into the animal. During blood collection, perfusion through the arterial catheter was interrupted, Fig. 4. (a) RBC (  1012/l) counts, (b) levels of Hb (g/l) and (c) Hct (l/l) in the blood of conscious unrestrained rats upon infusion with saline and blood collections, with or without reinfusion of blood cells, over time. Group 1 (O; n = 10) and Group 2 (O; n = 5). * Significant difference between Groups 1 and 2 on indicated days; zsignificant variation within Group 1 or 2 over time; {significant difference between Groups 1 and 2 over 8 days. 496 A. Blouin et al. / Journal of Pharmacological and Toxicological Methods 44 (2000) 489±505 and a 1-ml syringe with a 23-gauge blunted needle was inserted in the Tygon tubing connected to the swivel. A volume of saline (100 ml dead-space in the line) was aspirated until blood appeared at the tip of the syringe. The Tygon tubing was clamped. The syringe was replaced with an empty one to obtain 1.2 ml of blood. The Tygon tubing was clamped again. The initial syringe was put back, and a 400-ml volume of sterile saline was injected with short jerked pulses to ensure that no blood remained in the tubing that would cause blood cell activation and/or clot, followed by momentaneous clamping. The Tygon tubing catheter was reconnected to the system, unclamped, and purged for 30 s. The morning collection was split between a heparin ± lithium blood collection tube vacutainer (1.0 ml for electrolyte and biochemical analyses; cat no. 367682; Becton Dickinson Vacutainer Systems, Franklin Lakes, NJ, USA) and a 7.3-mg K2 EDTA vacutainer (0.200 ml for hematological analyses; cat no. 367861; BDVS). The evening collection was performed in sodium citrate (0.050 ml of 3.8% w/v) for the measurements of vasoactive mediators. 2.5. Hemodynamic measurements Each day at 14:00 h, the Tygon tubing catheter adjacent to the SS swivel was connected to a blood pressure transducer (cat no. 60-3002; Harvard). Mean arterial blood pressure (MABP) of unrestrained conscious rats was then continuously monitored for 15 min on a computerized system (Po-ne-mah Acquisition Analysis and Archive Systems, cat no.; Gould Inst. Syst./Po-NeMah; Simsbury, CN, USA; Harvard) precalibrated in millimeters of mercury with a pressure transducer calibrator (cat no. 2900; Stoelting/Ugo Basile, Varese, Italy). HR in beats/minute (bpm) was simultaneously derived from these data and recorded. 2.6. Biochemical analyses The concentration of hemoglobin (Hb; g/l) was measured using an ABL system 625 (Radiometer Copenhagen, Denmark). Blood urea nitrogen (BUN; range: 0.8 ± 66.8 mM) and creatinine (range: 8.8± 1591 mM), as parameters of kidney functions, alkaline phosphatase (ALP; range: 3± 1000 IU/l), alanine aminotransferase (ALT; range: 4 ± 600 IU/l) and bilirubin (range: 0.8± 513 mM), as parameters of hepatic functions and hemolytic anemia, were all measured with a Hitachi 917 (Boehringer Mannheim/Roche Diagnostic, Laval, QC, Canada) via various UV/visible colorimetric assays. The levels of angiotensin-converting enzyme (ACE; range: 3± 120 IU/l) activity were measured with a Cobas Mira (Roche Diagnostic) via a decrease in turbidity. The concentrations of various electrolytes (cations: calcium (Ca2 + ; range: 0.004 ± 5.0 mM), sodium (Na + , range: 80 ±180 mM), potassium (K + , range: 1.5± 10.0 mM); anions: phosphates (PO4 , range: 0.1 ± 6.5 mM), chloride (Cl , range: 60 ±140 mM) were measured with the Hitachi 917 (Boehringer Mannheim/Roche Diagnostic) using specific electrodes directly (Na + , K + , and Cl ) or a colorimetric reaction (Ca2 + and PO4 ). 2.7. Measurements of vasoactive mediators The levels of vasoactive mediators such as endothelin-1 (ET-1; picogram/milliliter), its precursor, big ET-1 (picogram/milliliter), and stable metabolites of the nitric oxide (NO) pathway, combined nitrates (NO3 ) and nitrites (NO2 )(NOx; micromolar), were also determined by EIA, EIA, and chemiluminescence, respectively. Briefly, ET-1 in the rat plasma (and stored at 80°C obtained from the sodium-citrate blood after centrifugation at 12,000  g for 6 min at 4°C) was measured using an EIA Fig. 5. (a) WBCs (  109/l) and (b) platelets (  109/l) counts in the blood of conscious unrestrained rats upon infusion with saline and blood collections, with or without reinfusion of blood cells, over time. Group 1 (O; n = 10) and Group 2 (O; n = 5). zSignificant variation within Group 1 or Group 2 over time. A. Blouin et al. / Journal of Pharmacological and Toxicological Methods 44 (2000) 489±505 Fig. 6. Daily concentrations of sodium (Na + , mM) in the blood of conscious unrestrained rats upon infusion with saline and blood collections, with or without reinfusion of blood cells, over time. Group 1 (O; n = 7 ± 10) and Group 2 (O; n = 5). * Significant difference between Groups 1 and 2 on the indicated day; zsignificant variation within Group 1 or 2 over time; { significant difference between Groups 1 and 2 over 8 days. (cat no. BI-20052; Biomedica, Vienna, AUSTRIA, via American Research Product, Belmont, MA, USA). Plasma samples (500 ml) were precipitated with 750 ml of a solution [solution: HCl (0.2 M) in 80 ml acetone] and centrifuged at 3000  g for 20 min at 4°C. The supernatant was lyophylised in a speed-vac concentrator (Savant Instruments, Holbrook, NY, USA) and reconstituted in the assay buffer (500 ml). Plasma samples or standards (200 ml) and the monoclonal mouse anti-ET antibodies (detection antibodies, 50 ml) were added into each wells precoated with polyclonal rabbit anti-ET antibodies (capture antibodies, second step) 497 and incubated overnight at room temperature. Under these conditions, the ET in the sample was bound to the latter antibody and formed a sandwich with the detection antibody. Nonspecific binding was removed by washing (4  350 ml, third step) and peroxidase-conjugated antibodies [antimouse IgG antibody conjugated with horseradish peroxidase (HRP)] were added (200 ml, fourth step) and incubated for 1 h at 37°C to detect the presence of bound monoclonal antibodies. The unbound conjugate was removed by washing (4  350 ml). Tetramethylbezidine (TMB) was then added (200 ml, sixth step) as a substrate and kept in the dark for 30 min at room temperature followed by addition of a stop solution (50 ml). The absorbance at 450 nm was determined using an EIA plate reader (Molecular Devices, Sunnyvale, CA, USA). The amount of ET present in the sample was calculated after comparing with a standard (six standards of human ET-1; 0 ±10 fmol/ml; 0 ± 24,919 pg/ml). The cross-reactivity of the capture antibodies was 100% with ET-1 and -2, < 5% with ET-3 and < 1% with big ETs. Intra-assay variations were less than 5%. Big ET-11 ± 38 was measured in plasma samples using another EIA (cat no. BI-20072, Biomedica via American Research Product). The plasma samples were extracted and reconstituted as described above. The reconstituted plasma, BALF samples or standards (100 ml) were added at room temperature with the monoclonal mouse anti-big ET-1 antibodies (detection antibodies; 50 ml) into wells precoated with polyclonal rabbit anti-big ET-1 antibodies (capture antibodies, second step) for a 3-h incubation at 37°C in an incubator/shaker. Subsequent steps were the same as those for the detection of ET-1 except that the samples were washed three times each with 300-ml buffer and that the volumes of HRP and TMB used were 100 ml each. Fig. 7. (a) Daily concentrations of BUN (mM) and (b) creatinine (mM) in plasma samples of conscious unrestrained rats upon infusion with saline and blood collections, with or without reinfusion of blood cells, over time. Group 1 (O; n = 10) and Group 2 (O; n = 5). 498 A. Blouin et al. / Journal of Pharmacological and Toxicological Methods 44 (2000) 489±505 The amount of big ET-1 in the samples was calculated using a nonlinear curve (4PL algorithm) of six standards of human big ET-11 ± 38 (0.05 ± 15.6 fmol/ml; 0.21 ± 66.81 pg/ml). The cross-reactivity of the capture antibodies was < 1% for ET1, -2, or -3. Intra-assay variations were less than 5%. The level of NO was measured in plasma samples via the determination of nitrite (NO2 ) and converted nitrate (NO3 ), two of the final metabolites of the nitric oxide synthase (NOS) pathway, by gas-phase chemiluminescence using a nitric oxide analyzer (NOA model 280; Sievers Instruments, Ionics, Boulder, CO, USA). Plasma samples (40 ml) were combined with PBS (40 ml; pH 7.2; Gibco BRL, Grand Island, NY, USA), a solution of cofactors (10 ml; cat no. 780012; Cayman) and a solution of nitrate reductase (10 ml; cat no. 780010; Cayman Chemicals, Ann Arbor, MI, USA) for a 3-h incubation at room temperature, and then precipitated with ethanol (200 ml of 90.5% ice-cold HPLC-grade ethanol; Lab Mat, Quebec, QC, Canada) for 30 min on ice and centrifuged for 5 min at 13,000 rpm at 4°C. The supernatant (100 ml) of each sample was added separately to a sealed radical purger containing 1% v/v of sodium iodide (NaI; > 50 mg/5 ml) and antifoaming agent (100 ml of a 1:30 stock dilution; Sievers) in 5 ml glacial acetic acid. The chemical reducing reaction converted NO2 into gaseous NO that was purged by a flow of inert nitrogen to the NOA for reacting with ozone (O3) to form nitrogen dioxide (NO2) to be detected. Values (areas under the curve) were calculated using a linear standard curve of sodium nitrite (NaNO2; 100 pmol ± 10 nmol). 2.8. Hematological cell counts Hematocrit values (Hct, l/l) were measured following centrifugation of EDTA-collected blood assessed via the Fig. 8. (a) Daily concentrations of ALP (IU/l), (b) ALT (IU/l), and (c) bilirubin (mM) in plasma samples of conscious unrestrained rats upon infusion with saline and blood collections, with or without reinfusion of blood cells, over time. Group 1 (O; n = 7 ± 10) and Group 2 (O; n = 5). * Significant difference between Groups 1 and 2 on the indicated day; zsignificant variation within Group 1 or 2 over time; {significant difference between Groups 1 and 2 over 8 days. A. Blouin et al. / Journal of Pharmacological and Toxicological Methods 44 (2000) 489±505 percentage of red blood cells (RBCs). RBCs (  1012/l), white blood cells (WBCs,  10 9 /l), and platelets (Plts,  109/l) counts in the blood were determined by mean of the volume conductivity scatter (VCS) technology using a Coulter STKS 2A W/RETIC counter (Beckman Coulter, Hialeah, FL, USA). 499 period. After up to 3 weeks, there was no macroscopic or microscopic sign of acute ischemia of the hind limbs or renal and intestinal infarctions. Furthermore, there was no sign of clot in the vascular catheter, and no arterial embolism was found. 3.2. Body weight and perfusion rate 2.9. Macroscopic and histological tissue observations Animals were euthanized and an autopsy followed. A physical external inspection of the rat was performed to note any abnormalities. For instance, the infusion harness around the animal was examined for skin irritations. Pieces of kidney and liver were removed for histological examination. The positioning, integrity, and functionality of the vascular catheter inside the abdominal aorta were also determined. 2.10. Statistical analyses Results of representative measurements were expressed as mean ‹ S.E.M. Because some observations were sometimes missing for a given time point, a repeated measurement analysis of variance was inapplicable to the data to compare measures at different time periods. Thus, for comparisons within groups, a randomized block design was applied using two factors defined for the analysis: the subject effect and the time-period effect. Comparison between groups was performed by a threeway ANOVA with a blocking factor representing subjects. Interaction between the time-period factor and one used to compare groups was added to the model. When interaction was significant for parameters, comparisons at different time periods were analyzed using Student's paired t tests. The normality and variance assumptions were met for almost all data. All analyses were conducted using the statistical package SAS (SAS Institute, Cary, NC, USA). The body weights were 424.8 ‹ 3.6 g for rats in Group 1 (two blood collections/day) and 418.6 ‹ 4.1 g in Group 2 (two blood collections/day, but one was reinfused), respectively. There was no significant difference (NS) between the two groups of rats at any time point over the entire experimental period. Rat body weights decreased by 2.4% and 2.2%, respectively, at 48-h postsurgery, returning to the original weight by Day 8 and increasing by 7.8% on Day 14 (up to 452.1), showing a steady growth over time ( P < .0001) in both groups. The volume of sterile saline that was administered intraarterially varied slightly (NS, except on Day 1) between 5.7 and 6.3 ml/day over 8 days in Group 1 and over 14 days in Group 2. There was no significant difference between the two groups at any time point except on Day 1 (Fig. 2). 3.3. Hemodynamic profile MABP was elevated for the first 3 days postcatheter implantation (range 121± 125 mmHg) but stabilized thereafter (range: 108 ±114 mmHg) over the remaining period of the experiment ( P < .0001). There was no difference between the two groups of rats (NS; Fig. 3a). HR was 3. Results 3.1. Postoperative period No complication was observed during or after surgical procedures. All animals were examined the day after the surgery for physical abnormalities, and several times a day thereafter. No postoperative infection was observed for the entire length of the present set of experiments (14 days; we did not observed problems up to 21 days; data not shown) and rats appeared in nominal health. Skin irritation that could be caused by the infusion harness was minimal. On one occasion, the incision located in the femoral region was reopened by the animal and was immediately repaired with no consequence. The apparatus (joints, swivel) and the vascular catheter remained functional and well positioned, and there was no leakage during the experimental Fig. 9. Daily activities of angiotensin convertase (IU/l) in plasma samples of conscious unrestrained rats upon infusion with saline and blood collections, with or without reinfusion of blood cells, over time. Group 1 (O; n = 7) and Group 2 (O; n = 5). * Significant difference between Groups 1 and 2 on the indicated day; zsignificant variation within Group 1 or 2 over time; { significant difference between Groups 1 and 2 over 8 days. 500 A. Blouin et al. / Journal of Pharmacological and Toxicological Methods 44 (2000) 489±505 almost identical between the two groups, ranging between 347 ‹ 10 and 391 ‹ 29 bpm (NS; Fig. 3b). 3.4. Hematological profile The concentration of RBCs was normal at 8.51 ‹ 1.42 (  1012/l). It decreased by 14% within 24 h after the first blood collection in both groups of rats, and reached a maximal reduction of 34% ( P < .004) in both groups on Day 8 (Fig. 4a). There was no further decrease past after Day 8 in both groups. RBC counts were significantly different ( P < .03) between the two groups on Days 3 ±7. A slower decrease was observed in Group 2. The plasma concentration of Hb was 159.8 ‹ 5.3 (g/l) in normal rats. It decreased by 9% after 1 day and reached a maximal reduction of about 37% in both groups ( P < .0001; Fig. 4b). The maximal decrease was observed on Days 7 and 9 in Groups 1 and 2, respectively. Similar to RBC counts, the profile of the decrease was significantly different ( P < .005) between the two groups. The Hct (0.47 ‹ 0.04 l/l) decreased within 24 h by 11% and stabilized after a maximal reduction of about 38% on Days 6 and 9, in Groups 1 and 2, respectively (Fig. 4c). The profile between the two groups was also different ( P < .008) on Days 2 ±7. WBC counts (15.4 ‹ 2.3  109 WBCs/ml) were not significantly different from baseline in both Groups 1 and 2 (NS; Fig. 5a). The profile between both groups was not different from Days 1 ±8. Platelet counts decreased by 13± 26% within 48-h postsurgery in Groups 2 and 1, respectively (NS; Fig. 5b). From Day 2 onwards, platelet counts in the blood increased steadily ( P < .0001; Fig. 5b) in both groups, by 18% after 8 days in Group 1, and by 35% after 14 days in Group 2. The profile between both groups of rats was not different from Days 1 ± 8. Fig. 10. (a) Daily concentrations of NOx (mM), (b) ET-1 (pg/ml) and (c) big ET-1 (pg/ml) in plasma samples of conscious unrestrained rats upon infusion with saline and blood collections, with or without reinfusion of blood cells, over time. Group 1 (O; n = 7) and Group 2 (O; n = 5). * Significant difference between Groups 1 and 2 on indicated days; zsignificant variation within Group 1 or 2 over time; {significant difference between Groups 1 and 2 over 8 days. A. Blouin et al. / Journal of Pharmacological and Toxicological Methods 44 (2000) 489±505 3.5 Concentrations of electrolytes in the plasma The concentrations of five electrolytes (three cations and two anions) in plasma samples were monitored. Concentrations of calcium did not fluctuate significantly over 14 days and were comparable (except on Day 1) to the baseline of 2.44 ‹ 0.10 mM in either group (NS). Sodium levels were different between Groups 1 and 2 on Days 3 and 8 ( P < .0009; Fig. 6). There were no significant fluctuations in sodium levels within a same group over time ( P < .068). Potassium levels did not fluctuate significantly from baseline (3.8 ‹ 0.2 mM), and were comparable over time in both groups (in Group 1, Day 5 vs. Day 1; a 12% increase; NS). The concentrations of chloride remained stable between 103 and 111 mM over time in both groups (NS). Phosphate levels on a day-by-day basis were not significantly different between the two groups, except on Days 1 and 6, or over time (from 1.8 to 2.1 mM, NS). 3.6. Plasma levels of markers for renal and hepatic functions The basal levels of BUN, a natural product of metabolism filtered by the kidneys, in the plasma of conscious rats over 8 ±14 days of saline infusion and blood collection, did not significantly fluctuate and were not different between the two groups (NS; Fig. 7a). Likewise, creatinine levels in both groups were similar and did not significantly fluctuate over the entire period of experimentation (NS; Fig. 7b). Levels of ALP were not significantly different between the two groups from Days 2± 8 (NS; Fig. 8a). However, within each group, their levels fluctuated over time ( P < .01; Fig. 8a). ALT concentrations were not different between Groups 1 and 2 over Days 1 ± 8, and there was no significant fluctuations within each group over time (NS; Fig. 8b). The levels of bilirubin (baseline: 1.8 ‹ 0.6 mM) remained stable (range: 1.40 ± 2.44 mM) over 8 days in Group 1 (NS; Fig. 8c). Conversely, plasma levels of bilirubin were significantly ( P < .0001) lower in Group 2 compared to Group 1 within the first 5 days of saline-resuspended blood cells reinfusion (Fig. 8c). 3.7. Activity and levels of vasoactive mediators in plasma samples zThe activity of angiotensin convertase increased by 23% in Group 1 over 8 days ( P < .0001), and it was significantly ( P < .0001) elevated in Group 2 (range: 159 ±226 U/l) compared to Group 1 (Fig. 9). The concentrations of nitrate (NO3 ) and nitrite (NO2 ), e.g., NO metabolites (NOx), were not different between Groups 1 and 2 (NS; Fig. 10a). The plasma concentrations of ET1 in plasma remained stable and did not fluctuate over the period of chronic experimentation within each groups but they were significantly ( P < .001) different between the two groups, being higher in Group 2 (Fig. 10b). Plasma 501 levels of big ET-1 followed a similar stable profile of that of ET-1 but were not different between Groups 1 and 2 over time (Fig. 10c). 3.8. Clinical observations The rats did not reveal any sign of unusual behavior during the entire period of the experiment nor any other particular abnormalities. The wound located at the left femoral site was healed. There was minimal skin irritation located at the site where the infusion harness was positioned around the front legs. 4. Discussion The described methodological approach allowed the consecutive uninterrupted daily collection of blood samples and assessment of hemodynamic parameters in conscious, unrestrained rats. The samples enabled the measurements of more than 20 distinct parameters related to blood hematology, renal and hepatic functions, as well as cardiovascular profile of mediators and hemodynamics. In general, the hematological and clinical chemistry values were within the range of nonoperated, sham-operated or noncatheterized SD rats (aged 13± 14 weeks) serving as controls (data not shown), and as reported previously (Archer, Jeffcott, & Lehmann, 1977; Canadian Council of Animal Care, 1993; Charles River, 1984; Hawkey, 1975; Mitruka & Rawnsley, 1977; Taconic Technical Library, 1998). Over time, the animals, which did not received or required any antibiotics, did not show any signs of infection, either locally at the site of catheter implantation or systemically, that could be associated with an increase in WBCs. A sterile methodological procedure is one key element to a successful long-term vascular access in the rat (Popp & Brennan, 1981). Nonsterile catheterisation would result in infection, which might complicate the interpretation of physiological measurements and even death in animals. The flexibility and antithrombogenic properties of the urethane vascular catheter is another key element since it reduced blood cell reactivity and inflammation reaction around the cannula compared to stiffer PE tubing prone to clot formation. Aseptically catheterized rats gained weight steadily, subsequent to their 2.5% weight loss over the first 48-h recovery phase due to the surgical implantation of the vascular catheter. Similar weight loss was reported in the first three postoperative days in Fisher F-344 rats (Burt et al., 1980). In our model, we cannulated the main left femoral artery for continuous infusion and blood sampling for minimal interference with regional blood flow. Other systems used the carotid artery (Buckle & Nathanielsz, 1974), the inferior/superior vena cava (with access, for some, via the jugular vein; Bakar & Niazi, 1983; Burt et al., 1980; 502 A. Blouin et al. / Journal of Pharmacological and Toxicological Methods 44 (2000) 489±505 Kaufman, 1980), the jugular vein (Bakar & Niazi, 1983), the tail artery (Fejes-Toth, Naray-Fejes-Toth, Ratge, & Frolich, 1984; Hagmuller, Liebmann, Porta, & Rinner, 1992), and, most recently, the portal vein (Strubbe, Bruggink, Steffens, 1999) as the point of entry. Most methods selected at the nape of the neck as the point of exit of the cannula, whereas a smaller number chose the positioning of a tail-cuff (Fejes-Toth et al., 1984). Our approach in the hind limb region and dorsal upper area exit was minimally invasive, compared to others where surgery involved a 6.5cm medial ventral incision and unilateral nephrectomy and subsequent cannulation of the stump of the renal artery (Dworkin, Filewich, Da Costa, Eissenberg, & Miller, 1980). Yoburn, Morales, and Inturrisi (1984) reported that implanted vascular cannula in the femoral artery was preferable to jugular vein or carotid artery, both localized in the upper region, in terms of patency, morbidity, and minimal postsurgical weight loss over 14 days. Our cannulated rats steadily received 250 ml/h of heparinized sterile saline (6 ml/day; another study even went to 18 ml/day; Keen, Brands, Smith, & Hall, 1998) that insured a constant flow of fluid toward maintaining the patency of the vascular catheter, as suggested before (DiCara, Pappas, & Pointer, 1969). By doing so, we insure the patency of the cannulation, and we could also enabled the continuous administration of drugs and hormones around the clock over weekly periods. Subsequently, we realized that the infusion of saline (without anticoagulants) was also nominal toward maintaining the patency of the catheter, with no differences in any of the parameters studied (data not shown). Previous models, involving repeated vascular access and collection of blood in conscious rats with daily on ± off vascularaccess-port, showed lack of reliability and reproducibility and were associated with the formation of clot and/or embolism (Cocchetto & Bjornsson, 1983; Epstein et al., 1990; Giner et al., 1987; Paulose & Dakshinamurti, 1987), even when daily flushing with a heparin solution was used (Weinstein & Annau, 1967). In rare cases, only when a mechanical break occurred involving the swivel and/or the lever arm (keep in mind that rats are quite active, especially at night that may inflict some stress on the setup over time), the line may be severed causing the arterial blood to flow out and coagulate rapidly within the vascular catheter, rendering it nonpatent. In such cases, the flow can be reestablish. The rat is anaesthetized and the harness disassembled. The vascular catheter is cut 1 cm above the neck. A bolus of warm, heparinized saline is forcefully injected into the catheter, taking along the clot, which will necessarily enter the arterial blood stream toward the hind limb. A truncated needle is used to reconnect the patent vascular catheter from the animal to a new one inserted into the spring stock up to the swivel. Only, and only if, after monitoring the animal for 48 h, all parameters are normal (no increase in MABP, HR, fluctuations in blood cells and enzymatic markers), the animal is kept into the current ongoing protocol. MABP was elevated (118 ± 125 mmHg) for a period of 3 days in both groups of normotensive rats, i.e., above previously reported for unrestrained conscious rats (110 mmHg, Robert-Ross, 1977; 106 ‹ 5 and 107 ‹ 3 mmHg in SD rats, Fink, Bryan, Mann, Osborn, & Werber, 1981; Muller & Mannesmann,1981; 113 ‹ 4 mmHg, Lestage et al., 1985; 102 ‹ 4 mmHg in WKY rats, Singewald, Kouvelas, Mostafa, Sinner, & Philippu, 2000; 102 ‹ 2 mmHg in Long± Evans rats, Fejes-Toth et al., 1984). We attribute this state to the combined effects of the surgical trauma related to the implantation of the vascular catheter and the adaptation of the animal to its infusion harness. From Days 4 ± 14, and further on (data not shown), baseline values at 14:00 h, thus in between the two blood collections of 8:00 and 20:00 h, remained stable around 110 mmHg, despite the fact that blood samples were withdrawal twice daily, with or without reinfusion of blood cells after the second sampling in Groups 1 and 2, respectively. Thus, future chronic experiments using the present method will proceed only after a minimum of 3± 4 days of recovery. HR remained stable during the entire experimental period, within the published ranges (417 ‹ 20 in SD rats, Muller & Mannesmann, 1981; 386 ‹ 9 in SD rats, Fink et al., 1981; 336 ‹ 18 in W/K rats, Irvine, White, & Chan, 1997; 360 ‹ 11 in Long ±Evans rats, Fejes-Toth et al., 1984; range: 250± 450, Canadian Council of Animal Care, 1993). Thus, the procedure described here provided a stable and reliable profile of MABP and HR that was not different from measurements obtained using telemetry (Anderson et al., 1999; Balakrishnan et al., 1998; Bazil et al., 1993; Deveney et al., 1998; Irvine et al., 1997; Kharidia & Eddington, 1996). It should be noted that restraining the animal, such as in indirect tail-cuff blood pressure readings (Bunag, 1983; Bunag & Butterfield, 1982), alas influenced blood pressure even though this had little effect on HR, in both SHR and their W/K controls (Irvine et al., 1997). In a normal SD rat, the volume of blood is estimated to be 50 ±65 ml/kg (or 21.25 ±27.62 ml total/rat of 425 g, Canadian Council of Animal Care, 1993). Therefore, each blood collection (1.2 ml) corresponded to a 4.2± 5.4% of total blood volume in these rats. Consequently, about 10% (2.4 ml) of the total blood was taken every day and replaced, at least in volume, by the daily continuous, but minimal (the autonomous microperfusion pump cannot go under 250 ml/h), infusion of saline (6 ml/day). Under these conditions, we expected, and results showed, that RBC counts, Hb, and Hct levels would all be reduced in an identical way. Nevertheless, the lowest value of RBCs in our experiments was within 11% of the reported baseline value (range: 5.4 ± 8.5  1012 RBCs/l; Canadian Council of Animal Care, 1993; Charles River, 1984). A similar observation was noted for Hb (range: 115 ±150) but not for Hct values (range: 37 ± 49) (Canadian Council of Animal Care, 1993; Charles River, 1984). The significant differences observed between Groups 1 and 2 from Days 3 ±7 in all three parameters can be explained by the reinfusion of A. Blouin et al. / Journal of Pharmacological and Toxicological Methods 44 (2000) 489±505 cellular elements in Group 2. Noticeably, there was no significant difference in these measurements between the two groups on Day 8, and all three parameters were stabilized thereafter. There was a slight but significant increase in WBC count on Days 4± 5 in both groups, probably due to postsurgical recovery, as previously reported (Hiesmayr et al., 1999). WBC counts returned to published baseline values ( < 10.2  109 WBCs/l) thereafter. Platelet counts rose over time, exceeding reported baseline values (range: 450 ±885  109 Plts/l, Charles River, 1984). This elevation in circulating platelets may be associated to the daily blood collection that might have triggered the hematopoietic process toward producing more platelets from stem cells. There was no difference between both groups regarding any of the five anionic and cationic electrolytes measured. Our values were within the range previously published for calcium (2.5 ± 3.2 mM), sodium (139 ± 150 mM), and potassium (3.6 ± 8.4 mM), slightly more elevated for chloride (84 ± 99 mM), but lower for phosphate (2.4 ± 5.6 mM) (Canadian Council of Animal Care, 1993; Taconic Technical Library, 1998). Thus, the concentrations of electrolytes were not affected by chronic blood samplings. These results also show the absence of nephrotoxicity, as renal insufficiency would influence potassium balance and changes in glomerular filtration and tubular reabsorption that would affect phosphate excretion. The concentrations of BUN and creatinine, as markers of renal functions, were not altered in conscious unrestrained rats infused for 14 days with saline when blood was collected daily. The values for creatinine (35 ± 62 mM) were consistent with published range, but urea were lower than the established values (9 ± 22 mM) in normotensive SD rats with matched sex and age (Charles River, 1984). These results show that there was damage in the kidney and/or the nephron and that blood creatinine, formed by creatinine metabolism in muscle tissues, was filtered normally by the kidney and did not accumulate in the circulation. This observation is very important since altered renal function may have significant effects on other organ systems. The activities of ALP and ALT were found to be within established levels (ALP; range: 142 ± 298 IU/l; ALT: range: 110± 274 IU/l) in rats (Canadian Council of Animal Care, 1993) with no differences between the two groups; but the values in Group 2 did fluctuate over time. An elevation in the activity of ALT, a liver-specific enzyme, would have been indicative of hepatocellular damage. The activity of gamma glutamyltransferase was merely above the detection limit in rat plasma (data not shown), as reported before (Boyd, 1983). Bilirubin is formed from the breakdown of Hb molecules. The concentrations of bilirubin were stable in Group 1, but were significantly lower in Group 2 rats on Days 1 ± 5, suggesting that reinjection of autologous blood cells was 503 not associated with hemolysis. This result exclude the possibility of hemolytic or hepatocellular diseases, or obstructive jaundice. Angiotensin convertase activity was slightly stimulated over time in Group 2 animals. Except for the most potent vasoconstrictor ET-1, levels of other vasoactive mediators, such as the metabolites (NOx: nitrite and nitrate) of the vasodilator NO and the inactive ET-1 precursor (big ET-1), were not different between the two groups and were mostly stable over time. It is likely that reinjection of autologous blood cells in rats of Group 2 activated the sensitive endothelium and increased ET-1 production in that group when compared to Group 1. We obtained concentrations that were lower from those published for NOx (3.9 ± 4.3 mM, Hecker, Denzer, & Wohlfeil, 1995; 7.4 ‹ 0.4 mM, Wu & Yen, 1999), within the range for ET-1 (2.0 ± 4.0 pg/ml, Brooks, Contino, Storer, & Ohlstein, 1991; Burkhardt, Barton, & Sham, 2000; Horio et al., 1991; Saito et al., 1989; Vermulapalli, Chiu, Rivelli, Foster, & Sybertz, 1991) and slightly higher for big ET-1 (1.2 pg/ml, Telemaque, Emoto, deWit, & Yanagisawa, 1998). These discrepancies can be explained by the different biochemical assays used for these measurements. We observed no differences between the levels of these mediators in individually housed and chronically cannulated rats (as for Group 1) versus multi-housed, nonoperated, and/or decapitated rats (data not shown). It was reported by others (Fagin, Shinsako, & Dallman, 1983) that plasma levels of ACTH and corticosterone were also similar in cannulated rats versus controls, further validating the method for obtaining repeated blood samples and allowing multiple intra-animal comparisons of basal and treated animals of vasoactive mediators. Thus, surgery (acute stress within 3± 4 days) and the constant presence of the cannula and jacket (accommodation) did not have any endocrine consequences under the present conditions. The apparatus used in the present protocol has been developed by Instech Labs. The cost of a complete operating system (US$1500/rat) is more affordable than the methods involving radiotelemetry (Anderson et al., 1999; Balakrishnan et al., 1998; Bazil et al., 1993; Brockway et al., 1991; Deveney et al., 1998; Guiol et al., 1992; Kharidia & Eddington, 1996; Van den Buuse, 1994). In addition to the cardiovascular parameters obtained with radiotelemetry, our approach can provide a substantial number of hematological and biochemical measurements that are also related to various organ functions and hemodynamic equilibrium. Furthermore, the present procedure shows that rats are less stressed than those that are restrained or using more invasive methods. Conversely, 24-h monitoring and/or longer period of monitoring (in term of months) may be possible only with a radiotelemetry approach. In the present system, hemodynamic monitoring cannot be repeated several times for periods of 15 min, but not continuously, over 24 h, since arterial blood pushed its way back up, and there would be a problem with catheter patency. Thus, radio- 504 A. Blouin et al. / Journal of Pharmacological and Toxicological Methods 44 (2000) 489±505 telemetry is necessary for getting diurnal blood pressure changes; thus, we could combine both vascular catheter insert (adding the measurements of blood gases, lipid profile, other hormones, glucose levels, etc.) and radiotelemetry in the future. In summary, the present operating system is a reliable model that ensured atraumatic and minimal stress for serial or sequential blood sampling, a long term (in terms of weeks) of reliable arterial access, and reproducibility for accurate daily measurements of various parameters, and for direct infusion of any drugs, over a given period of time in conscious unrestrained rats. We suggest to wait for at least 4 days postsurgery before initiating any studies. Furthermore, we do not believe that it is beneficial to re-administer blood cells since there was no difference between Groups 1 and 2 for RBCs, Hct, and Hb profile after 7 days, and that this action may rather affect the vascular endothelium. Finally, the use of heparinized saline is not necessary, considering the constant microperfusion of a fluid with the pump keeps the cannula patent, therefore eliminating any undesirable effects that might be associated with heparin or any other anticoagulants. In conclusion, such an integrated approach for prolonged and repetitive blood samplings on the same animal enabled the understanding of the pathology associated with diseases affected by genetic and/or environmental factors. It can also be employed to evaluate novel pharmaceutical agents. Acknowledgments The authors wish to thank Guy NoeÈl, Justin Robillard, and Evelyn Vachon for their guidance and expertise in animal care and Dr. Fernand Bertrand, Mrs Jacqueline Paquette (Biochemistry), and Mrs. Helene Arsenault (Hematology), in Hospital Departments, for the numerous assays and analyses, Mr. Lynn Atton for technical assistance, Mr. Serge Simard for statistical analyses, and Mr. Robert Bellemare (audiovisual). BB is a Junior I Research Scholar from the Fonds de la recherche en sante du QueÂbec (FRSQ). This work is supported by grants from the FRSQ, La Fondation des maladies du cúur du QueÂbec (FMCQ), La Fondation de l'Institut de cardiologie de QueÂbec (FICQ), and Les Fonds pour la formation de Chercheurs et l'Aide aÁ la Recherche (FCAR). References Acland, R. D. (1980). Arterial anastomosis: common femoral artery. In: C. V. Mosby (Ed.), Microsurgery: practice manual ( pp. 38 ± 59). Toronto. Anderson, N. H., Devlin, A. M., Graham, D., Morton, J. J., Hamilton, C. A., Reid, J. L., Schork, N. J., & Dominiczak, A. F. (1999). Telemetry for cardiovascular monitoring in a pharmacological study: new approaches to data analysis. Hypertension, 33, 248 ± 255. Archer, R. K., Jeffcott, L. B., & Lehmann, H. (1977). Comparative clinical haematology. Oxford: Blackwell. Bakar, S. K., & Niazi, S. (1983). Simple reliable method for chronic cannulation of the jugular vein for chronic cannulation of the jugular vein for pharmacokinetic studies in rats. Journal of Pharmaceutical Sciences, 72, 1027 ± 1029. Balakrishnan, S., Tatchum-Talom, R., & McNeill, J. R. (1998). Radiotelemetric versus externalized catheter monitoring of blood pressure: effect of vasopressin in spontaneous hypertension. Journal of Pharmacological and Toxicological Methods, 40, 87 ± 93. Bazil, M. K., Krulan, C., & Webb, R. L. (1993). Telemetric monitoring of cardiovascular parameters in conscious spontaneously hypertensive rats. Journal of Cardiovascular Pharmacology, 22, 897 ± 905. Boyd, J. W. (1983). The mechanism relating to increases in plasma enzymes and isoenzymes in diseases of animals. Veterinary Clinical Pathology, 12, 9 ± 24. Brandstaetter, J., & Terkel, J. (1977). Chronic double lumen intravenous cannulation of the unrestrained rats. Laboratory Animal Science, 27, 999 ± 1003. Brockway, B. P., Mills, P. A., & Azar, S. H. (1991). A new method for continuous chronic measurement and recording of blood pressure, heart rate and activity in the rat via radio-telemetry. Clinical and Experimental Hypertension, 13, 885 ± 895. Brooks, D. P., Contino, L. C., Storer, B., & Ohlstein, E. H. (1991). Increased ET excretion in rats with renal failure induced by partial nephrectomy. British Journal of Pharmacology, 104, 987 ± 989. Buckle, J. W., & Nathanielsz, P. W. (1974). Proceedings: a dual catheter system for the simultaneous infusion and sampling of the vascular system of the unrestrained rats. Journal of Physiology (London), 242, 55P ± 56P. Bunag, R. D. (1983). Facts and fallacies about measuring blood pressure in rats. Clinical and Experimental Hypertension, Part A, 5, 1659 ± 1681. Bunag, R. D., & Butterfield, J. (1982). Tail-cuff blood pressure measurement without external preheating in awake rats. Hypertension, 4, 898 ± 903. Burkhardt, M., Barton, M., & Sham, S. G. (2000). Receptor- and nonreceptor-mediated clearance of big ET and ET-1: differential effects of acute and chronic ETA receptor blockade. Journal of Hypertension, 18, 273 ± 279. Burt, M. E., Arbeit, J., & Brennan, M. F. (1980). Chronic arterial and venous access in the unrestrained rats. American Journal of Physiology, 238, H599 ± H603. Canadian Council of Animal Care. (1993). In: E. D. Olfert, B. M. Cross, & A. A. McWilliam (Eds.), Manual for the care and utilization of experimental animals. CCAC, Ottawa, ON, Canada ( pp. 194 ± 200). Charles River. (1993). Baseline haematology and clinical chemistry values for Charles River CD rats as function of sex and age. Charles River Technical Bulletin, 3 (2), 1 ± 47. Cocchetto, D. M., & Bjornsson, T. D. (1983). Methods for vascular access and collection of body fluid from the laboratory rat. Journal of Pharmaceutical Sciences, 72, 465 ± 492. Depocas, F., & Behrens, W. A. (1977). Effects of handling, decapitation, anesthesia and surgery on plasma noradrenaline levels in the white rat. Canadian Journal of Physiology and Pharmacology, 55, 212 ± 219. Deveney, A. M., Kjellstrom, A., Forsberg, T., & Jackson, D. M. (1998). A pharmacological validation of radiotelemetry in conscious, freely moving rats. Journal of Pharmacological and Toxicological Methods, 40, 71 ± 79. DiCara, L. V., Pappas, B. A., & Pointer, F. (1969). A technique for chronic recording of systemic arterial blood pressure in the unrestrained rat. Behavior Research Methods and Instruments, 1, 221 ± 223. Dworkin, B. R., Filewich, R. J., Da Costa, J., Eissenberg, E., & Miller, N. E. (1980). A chronic arterial catheter and low compliance system for recording blood pressure and heart rate from the rat. American Journal of Physiology, 239, H137 ± H141. Epstein, R. E., Branum, G. D., Cucchiaro, G., & Meyers, W. C. (1990). A model for biliary and vascular access in the unanesthetized, unrestrained rat. Physiology & Behavior, 48 (4), 539 ± 542 (October). A. Blouin et al. / Journal of Pharmacological and Toxicological Methods 44 (2000) 489±505 Fagin, K. D., Shinsako, J., & Dallman, M. F. (1983). Effects of housing and chronic cannulation on plasma ACTH and corticosterone in the rat. American Journal of Physiology, 245, E515 ± E520. Fejes-Toth, G., Naray-Fejes-Toth, A., Ratge, D., & Frolich, J. C. (1984). Chronic arterial and venous catheterization of conscious, unrestrained rats. Hypertension, 6, 926 ± 930. Feldman, L. J., Tahlil, O., & Steg, G. (1996). Perspectives of arterial gene therapy for the prevention of restenosis. Cardiovascular Research, 32, 194 ± 207. Fink, G. D., Bryan, W. J., Mann, M., Osborn, J., & Werber, A. (1981). Continuous blood pressure measurement in rats with aortic baroreceptor deafferentation. American Journal of Physiology, 241, H268 ± H272. French, B. A. (1998). Gene therapy and cardiovascular disease. Current Opinion in Cardiology, 13, 205 ± 213. Giner, M., Snyder, K., & Meguid, M. M. (1987). Chronic vascular access for repeated blood sampling in the unrestrained rat. American Journal of Physiology, 253, H992. Guiol, C., Ledoussal, C., & Surge, J. M. (1992). A radiotelemetry system for chronic measurement of blood pressure and heart rate in the unrestrained rat validation of the method. Journal of Pharmacological and Toxicological Methods, 28, 99 ± 105. Hagmuller, K., Liebmann, P., Porta, S., & Rinner, I. (1992). A tail-artery cannulation method for the study of blood parameters in freely moving rats. Journal of Pharmacological and Toxicological Methods, 28, 79 ± 83. Hajjar, R. J., del Monte, F., Matsui, T., & Rosenzweig, A. (2000). Prospects for gene therapy for heart failure. Circulation Research, 31, 616 ± 621 (Mar). Hawkey, C. M. (1975). Comparative mammalian haematology; cellular components and blood coagulation of captive wild animals. London, England, U.K.: Heinemann. Hecker, G., Denzer, D., & Wohlfeil, S. (1995). Elevation of circulating NO: its effects on hemodynamics and vascular smooth muscle cell proliferation in rats. Agents and Actions, 45, 169 ± 176. Hiesmayr, M. J., Spittler, A., Lassnigg, A., Berger, R., Laufer, G., Kocher, A., Artemiou, O., Boltz-Nitulescu, G., & Roth, E. (1999). Alterations in the number of circulating leucocytes, phenotype of monocyte and cytokine production in patients undergoing cardiothoracic surgery. Clinical and Experimental Immunology, 115, 315 ± 323. Horio, T., Kohno, M., Yokokawa, K., Murakawa, K., Yasunari, K., Fujiwara, H., Kurihara, N., & Takeda, T. (1991). Effect of hypoxia on plasma immunoreactive endothelin-1 concentration in anesthetized rats. Metabolism, 40, 999 ± 1001. Irvine, R. J., White, J., & Chan, R. (1997). The influence of restraint on blood pressure in the rat. Journal of Pharmacological and Toxicological Methods, 38, 157 ± 162. Kaufman, S. (1980). Chronic, nonocclusive, and maintenance-free central venous cannula in the rat. American Journal of Physiology, 239, R123 ± R125. Keen, H. L., Brands, M. W., Smith, M. J., & Hall, J. E. (1998). Maintenance of baseline angiotensin II potentiate insulin hypertension in rats. Hypertension, 31, 637 ± 642. Kharidia, J., & Eddington, N. D. (1996). Effects of desethylamiodarone on the electrocardiogram in conscious freely moving animals: pharmacokinetic and pharmacodynamic modeling using computer-assisted radio telemetry. Biopharmaceutics & Drug Disposition, 17, 93 ± 106. Lestage, P., Vitte, P. A., Rolinat, J. P., Minot, R., Broussolle, E., & Bo- 505 billier, P. (1985). A chronic arterial and venous cannulation method for freely moving rats. Journal of Neuroscience Methods, 13, 213 ± 222. Mitruka, B. M., & Rawnsley, H. M. (1977). Clinical, biochemical and hematological reference values in normal experimental animals. New York: Masson. Muller, B., & Mannesmann, G. (1981). Measurement of cardiac output by the thermodilution method in rats: II. Simultaneous measurement of cardiac output and blood pressure in conscious rats. Journal of Pharmacological Methods, 5, 29 ± 34. Paulose, C. S., & Dakshinamurti, K. (1987). Chronic catheterization using vascular-access-port in rats. Journal of Neuroscience Methods, 22, 141 ± 146. Popp, M. B., & Brennan, M. F. (1981). Long-term vascular access in the rat: importance of asepsis. American Journal of Physiology, 241, H606 ± H612. Robert-Ross, A. (1977). Measurement of blood pressure in unrestrained rats. Physiology & Behavior, 19, 327 ± 329. Saito, Y., Nakao, K., Shirakami, G., Jougasaki, M., Yamada, T., Itoh, H., Mukoyama, M., Arai, H., Hosoda, K., Suga, S., Ogawa, Y., & Imura, H. (1989). Detection and characterization of endothelin-1-like immunoreactivity in rat plasma. Biochemical and Biophysical Research Communications, 163, 1512 ± 1516. Singewald, N., Kouvelas, D., Mostafa, A., Sinner, C., & Philippu, A. (2000). Release of glutamate and GABA in the amygdala of conscious rats by acute stress and baroreceptor activation: differences between SHR and WKY rats. Brain Research, 864, 138 ± 141. Still, J. W., & Whitcomb, E. R. (1956). Technique for permanent log-term intubation of rat aorta. Journal of Laboratory and Clinical Medicine, 48, 152 ± 154. Strubbe, J. H., Bruggink, J. E., & Steffens, A. B. (1999). Hepatic portal vein cannulation for infusion and blood sampling in freely moving rats. Physiology & Behavior, 65, 885 ± 887. Taconic Technical Library. (1998). Body weight ratios, hematological and clinical chemistry values for Long ± Evans, Sprague ± Dawley and Wistar ± Kyoto rats. http://www.taconic.com/healthr/hematology/ balbheme.htm. Telemaque, S., Emoto, N., deWit, D., & Yanagisawa, M. (1998). In vivo role of endothelin-converting enzyme-1 as examined by adenovirusmediated overexpression in rats. Journal of Cardiovascular Pharmacology, 31, S548 ± S550. Van den Buuse, M. (1994). Circadian rhythms of blood pressure, heart rate, and locomotor activity in spontaneously hypertensive rats as measured with radio-telemetry. Physiology & Behavior, 55, 783 ± 787. Vermulapalli, S., Chiu, P. J. S., Rivelli, M., Foster, C. J., & Sybertz, E. J. (1991). Modulation of circulating endothelin levels in hypertension and endotoxemia in rats. Journal of Cardiovascular Pharmacology, 18, 895 ± 903. Weinstein, S. A., & Annau, Z. (1967). A chronic cross-circulation technique for rats. Journal of Applied Physiology, 23, 601 ± 604. Wood, M., Shand, D. G., & Wood, A. J. (1979). Altered drug binding due to the use of indwelling heparinized cannulas for sampling. International Journal of Clinical Pharmacology and Therapeutics, 25, 103 ± 107. Wu, C. C., & Yen, M. H. (1999). Higher level of plasma nitric oxide in spontaneously hypertensive rats. American Journal of Hypertension, 12, 476 ± 482. Yoburn, B. C., Morales, R., & Inturrisi, C. E. (1984). Chronic vascular catheterization in the rat: comparison of three techniques. Physiology & Behavior, 33, 89 ± 94.