Characterization of the Effects of Adenosine Receptor Agonists on Cerebral Blood Flow in Uninjured and Traumatically Injured Rat Brain using Continuous Arterial Spin-Labeled Magnetic Resonance Imaging

Abstract

Hypoperfusion after traumatic brain injury may exacerbate damage. Adenosine, a vasodilator, regulates cerebral blood flow (CBF). Treatment with adenosine receptor agonists has shown benefit in experimental CNS trauma; however, their effects on CBF after injury remain undefined. We used magnetic resonance imaging to assess CBF in uninjured rats both early and at 24 h after intrahippocampal administration of either the nonselective adenosine receptor agonist 2-chloroadenosine (2-CA, 12 nmol) or the A_2A-receptor agonist 2-p-(2-carboxyethyl)-phenethylamino-5'-N-ethylcarbox-amidoadenosine (CGS 21680, 6 nmol). We also assessed the effects of these agents on cerebral metabolic rate for glucose (CMRglu). We then assessed the effect of 2-CA on CBF at 3.5 to 5 h after controlled cortical impact (CCI). Injection of 2-CA into uninjured rat brain produced marked increases in CBF in ipsilateral hippocampus and cortex versus vehicle (P < 0.05); CBF increases persisted even at 24 h. Measurement of hippocampal levels of 2-CA showed persistent increases to 24 h. CGS 21680 produced even more marked global increases in CBF than seen with 2-CA (2–6-fold versus vehicle, P < 0.05 in 10/12 regions of interest (ROIs)). Neither agonist altered CMRglu versus vehicle. After CCI, 2-CA increased CBF in ipsilateral hippocampal and hemispheric ROIs (P < 0.05 versus vehicle), but the response was attenuated at severe injury levels. We report marked increases in CBF after injection of adenosine receptor agonists into uninjured rat brain despite unaltered CMRglu. 2-Chloroadenosine produced enduring increases in CBF in uninjured brain and attenuated posttraumatic hypoperfusion. Future studies of adenosine-related therapies in CNS injury should address the role of CBF.

Keywords

CGS 21680 2-chloroadenosine perfusion traumatic brain injury

Introduction

Traumatic brain injury (TBI) acutely reduces cerebral blood flow (CBF) in experimental models and humans (Yamakami and McIntosh, 1989; Bouma et al, 1992). This may mediate secondary damage, since posttraumatic metabolic demands are increased (Hovda et al, 1995). Few experimental studies have targeted posttraumatic hypoperfusion. These studies have assessed effects of vasodilators (ethanol, l-arginine) (Kelly et al, 2000; DeWitt et al, 1997; Cherian and Robertson, 2003) or isoflurane (Statler et al, 2000), or agents limiting vasoconstriction—superoxide dismutase (DeWitt et al, 1997) or endothelin antagonists (Armstead, 1999).

Adenosine receptor agonists increase CBF in normal brain. This occurs across species with intracarotid (Joshi et al, 2002), intravenous (Soricelli et al, 1995), topical (Ibayashi et al, 1991), or parenchymal (Van Wylen et al, 1989) administration. However, some investigators have reported differing results depending on the dose administered (Sollevi et al, 1987; Stange et al, 1997). Vasodilation by adenosine occurs through activation of adenosine-2a (A_2A) receptors with adenyl cyclase activation in vascular smooth muscle (Kalaria and Harik, 1988). A_2B receptor activation linked to nitric oxide synthase activity may also participate (Ngai et al, 2001; Shin et al, 2000).

Beneficial effects of the nonselective agonist 2-chloroadenosine (2-CA) on cerebral energy charge and functional outcome were shown after intraventricular or intraparenchymal administration in experimental TBI (Headrick et al, 1994; Varma et al, 2002). Benefit was also observed in experimental stroke (Evans et al, 1987) and epilepsy (Dunwiddie and Worth, 1982). Cerebral blood flow was not assessed in these studies. The A_2A receptor agonist ATL-146e reduced damage in experimental spinal cord injury (Cassada et al, 2002a, b ), but CBF was not reported. Although the benefits of CBF promotion are anticipated, A_2A receptor activation might be either beneficial or detrimental in CNS injury—including stroke (Simpson et al, 1992; Higashi et al, 2002; Chen et al, 1999; Aden et al, 2003), spinal cord injury (Cassada et al, 2002a), Parkinson's (Chen et al, 2001), and Huntington's disease (Popoli et al, 2002). Putative detrimental consequences of A_2A receptor activation include augmented presynaptic glutamate release (Popoli et al, 2002), A₁ receptor desensitization (Dixon et al, 1997) and leukocyte mediated effects (Yu et al, 2004). Since adenosine levels are high early after TBI—with activation of lower-affinity A_2A receptors (Pedata et al, 2001; Bell et al, 1998), effects on CBF deserve investigation. To our knowledge, the effects of adenosine receptor agonists on CBF have not been reported in experimental TBI.

We previously used magnetic resonance imaging (MRI)—with continuous arterial spin labeling—to serially map CBF in traumatically injured rat brain (Forbes et al, 1997; Hendrich et al, 1999) and show that cortical injection of 2-CA acutely increased CBF in normal rat brain (Kochanek et al, 2001).

We currently report MRI measurements of CBF in uninjured sham rats early and at 24 h after intrahippocampal administration of either 2-CA or the selective A_2A receptor agonist 2-p-(2-carboxyethyl)-phenethylamino-5'-N-ethylcarbox-amidoadenosine (CGS 21680). We also studied the effects of these agents on cerebral glucose utilization in uninjured shams using autoradiography. Finally, we studied the effect of 2-CA on CBF in the controlled cortical impact (CCI) model of TBI in rats. We hypothesized that adenosine receptor agonists would increase CBF in uninjured and injured rat brain, and that A_2A receptor activation played an important role.

Materials and methods

This study protocol was approved by the University of Pittsburgh Animal Care and Use Committee. A total of 61 adult, male Sprague–Dawley rats (353 ± 24 g, Harlan, Indianapolis, IN, USA) were studied.

Specific Treatment Protocols

Uninjured rats: Four separate protocols were performed in uninjured rats subjected to craniotomy and intrahippocampal injection of adenosine receptor agonist or vehicle (Table 1). These included assessment of CBF at either ~3.5 and ~5 h or ~24 h after injection by MR or CMRglu at 3 h after injection by autoradiography.

Table 1

Treatment protocols for all studies

Treatment protocol	Outcome parameter	Assessment time	Sample size per group (total sample size for study)
1. 2-CA^a versus saline vehicle in uninjured rats	CBF	~3.5 and ~5h	4 (8)
2. 2-CA versus saline vehicle in uninjured rats	CBF	~24 h	3 (6)
3. 2-CA in uninjured rats	2-CA levels	~3.5 and 24 h	7, 6 (13)
4. CGS 21680^b versus sterile water vehicle in uninjured rats	CBF	~3.5 and ~5 h	3 (6)
5. 2-CA, saline vehicle, CGS21680, and sterile water vehicle in uninjured rats	CMRglu	3 h	4 (16)
6. 2-CA versus saline vehicle after moderate^c or severe^d CCI	CBF	~3.5 and 5 h	3 (12)

2-CA, 2-chloroadenosine; CGS 21680, 2-p-(2-carboxyethyl)-phenethylamino-5'-N-ethylcarbox-amidoadenosine; CBF, cerebral blood flow; CMRglu, cerebral metabolic rate for glucose; CCI, controlled cortical impact.

Doses of 2-CA were 12nmol in 2 µL saline in all studies.

Doses of CGS 21680 were 6 nmol in 2 µL sterile water in all studies.

Moderate CCI represents a deformation of 2.0 mm.

Severe CCI represents a deformation of 2.5 mm.

Controlled cortical impact model: A protocol comprised of four groups (Table 1) was used to study rats treated with either 2-CA or saline vehicle after CCI. Each treatment was assessed in groups subjected to either 2.0 mm deformation (moderate) or 2.5 mm deformation (severe) injury levels, with CBF assessed at ~3.5 and ~5 h after injury. An investigator (KH) masked to treatment performed MR data acquisition and analysis.

Surgical Preparation

For all protocols, male Sprague–Dawley rats had free access to food and water until the time of study. Anesthesia was induced in a plastic jar with 4% isoflurane (Anaquest, Memphis, TN, USA) in oxygen. The trachea was intubated with a 14-gauge angiocath and mechanically ventilated with 2% isoflurane in N₂O:O₂ (2:1). Sterile prep and drape were used for all surgical procedures. Femoral arterial and venous catheters (PE-50) were surgically inserted for continuous monitoring of blood pressure, arterial blood sampling (CCI protocol only), and administration of pancuronium bromide (0.1 mg/kg, Elkins-Sinn, Cherry Hill, NJ, USA). A rectal probe was inserted for continuous monitoring and maintenance of temperature at 37.0°C ± 0.5°C. Bicillin (100,000 U, Upjohn, Kalamazoo, MI, USA) and gentamicin (10 mg/kg, Elkins-Sinn, Cherry Hill, NJ, USA) were administered intramuscularly.

Rats in all protocol groups were randomized to treatment with an adenosine receptor agonist or appropriate vehicle, which was stereotactically injected (immediately after impact for the CCI group), through the craniotomy into the brain to a depth of 3.5 mm (−4.3 mm AP, −2.0 ML, relative to bregma) at a rate of 1 µL/min using a 30-gauge, 10 µL syringe (Hamilton, Reno, NV, USA). All injection volumes were 2 µL. Based on pilot studies, using Evans blue dye, this yielded drug or vehicle delivery to the center of the dorsal hippocampus. After injection, the needle was slowly withdrawn. The craniotomy was closed by replacing the bone flap and sealing the site with dental cement (Koldmount, Vernon Benshoff, Albany, NY, USA). The scalp incision was then closed. After surgery, anesthesia was discontinued, the temperature probe was removed, and the rats were weaned from mechanical ventilation and extubated within 1 h. In acute studies, catheters were fastened externally to the rat using a specially designed thoracic harness and thus maintained for re-use at the NMR center. 2-Chloroadenosine and CGS 21680 were purchased from RBI (Natic, MA, USA) and Sigma (St Louis, MO, USA), respectively. Drug doses are provided in Table 1.

The head was fixed in a stereotactic device (David Kopf, Tujunga, CA, USA). After retraction of the scalp, a craniotomy was made over the left parietal cortex with a dental drill, using the coronal and interparietal sutures as margins. For rats in the CCI model protocol groups, the bone flap was removed to zero the piston: it was then replaced, and the rats were allowed to equilibrate on isoflurane (1.1%) and N₂O:O₂ (2:1) for 30 mins. At 15 mins before CCI, an arterial blood sample was obtained to verify whether arterial blood gas tensions were within physiological limits (PaCO₂ = 30 to 45 mm Hg; PaO₂ > 90 mm Hg). Controlled cortical impact was performed as previously described (Dixon et al, 1999). Briefly, rats were subjected to injury to left parietal cortex (5-mm impactor tip, velocity of 4 m/secs, deformation depth of either 2.0 or 2.5 mm, impact duration of ~50 secs).

MR Studies of Cerebral Blood Flow

At the predefined time after either craniotomy or CCI, rats were transferred to the NMR center. Anesthesia was again induced using 4% isoflurane in O₂. Rats were re-intubated with a 14-gauge angiocath and mechanically ventilated on 1% isoflurane in N₂O:O₂ (1:1), and pancuronium bromide (0.1 mg/kg h) was administered intravenously for immobilization. If rats were in the 24 h postinjection group (Protocol 2, Table 1), femoral arterial and venous catheters were surgically re-inserted at the NMR center. For rats studied in the acute postinjection or postinjury phase, vascular catheters were removed from the harness at the NMR center and reconnected to appropriate infusion or monitoring devices. Rats were placed in a cradle in the prone position. Motion was prevented by gently securing their heads in a custom-built holder with round-tipped ear bars and an adjustable bite bar. After placement in the magnet, the isoflurane and N₂O were discontinued and a gas mixture of O₂ in N₂ (1:1) was initiated. An infusion of pentobarbital (37.5 mg/kg h, Abbot Laboratories, Chicago, IL, USA) was begun and rats were equilibrated on this anesthetic for a minimum of 30 mins before CBF acquisitions were initiated. Rectal temperature was controlled at 37°C ± 0.5°C and target PaCO₂ was between 30 and 45 mm Hg (manipulated by adjusting ventilator rate and tidal volume). Arterial pO₂, pH, and MABP were also measured immediately before and after each perfusion assessment.

Imaging was performed at 4.7 T on a 40-cm horizontal-bore Bruker AVANCE-DBX system, equipped with a 15-cm diameter shielded gradient insert and a 74-mm diameter RF coil. Localized shimming optimized homogeneity within a coronal slice ~3 mm posterior to bregma (field of view = 4 cm, slice thickness = 2 mm). Consistent positioning was confirmed on T₂-weighted images and all subsequent data were acquired from this same plane.

Images were perfusion encoded by arterial spin labeling with continuous flow-induced adiabatic inversion before each phase-encoding step (Detre et al, 1992; Williams et al, 1992). A constant amplitude RF pulse was applied for 2 secs in the presence of an axial B₀ field gradient (1 G/cm). Perfusion data were acquired in duplicate (spin-echo, TR = 2,000 ms, summation of TE = 10, 20, and 30 ms echoes, 2 averages, 128 × 70 zero-filled to 128 × 128). The inversion plane was positioned ± 2 cm from the detection plane for control and labeled images by offsetting the frequency of the labeling pulse ± 8,500 Hz. Spin-labeling efficiency (α) was measured in each study (Zhang et al, 1993). The spatial dependence of the spin-lattice relaxation time of tissue water in the presence of perfusion (T_{1 obs}) was measured within the same slice from a series of spin-echo images acquired with variable TR (range = 100 to 8,000 ms, TE = 8.1 ms, 2 averages). T_{1 obs} maps were generated for each study (Hendrich et al, 1999). At the completion of imaging, and while anesthesia was continued, rats were killed with an intravenous injection of a saturated solution of KCl.

Images were analyzed with the software package ‘Stimulate’ (Strupp, 1996) kindly provided by John Strupp (Center for MR Research, University of Minnesota Medical School, Minneapolis, MN, USA). Anatomical regions of interest (ROIs) within the slice were defined on the control perfusion image. An additional region was also identified that corresponds to the contused area in CCI studies; this ROI encompasses the portion of cortex extending 1 mm from midline to approximately halfway down the rhinal fissure, and is referred to as the medial cortical segment. Pixel-by-pixel maps of (M_C—M_L)/M_C were generated, where M_C is the magnetization intensity from each control image of the perfusion study and M_L is the magnetization intensity from each labeled image. Negative pixel values within the brain were retained for quantification. To visually confirm that CBF calculations were not artifactually affected by potential changes in ¹H density after injection, a map of (M_C—M_L)/M_C was also generated using the first labeled image and the second control image—effectively reversing the acquisition order. The duplicate (M_C—M_L)/M_C maps were then averaged and mean regional values were computed for each ROI within both the (M_C—M_L)/M_C maps and the T_{1 obs} maps. Cerebral blood flow was then calculated for each ROI from (Zhang et al, 1995):

CBF = λ (T_{1 obs} \cdot 2 α)^{- 1} \cdot (M_{C} - M_{L}) \cdot M_{C^{- 1}},

(1)

where λ is the brain–blood partition coefficient of water, with a spatially constant value of 0.9 mL/g assumed. A value of α = 0.7 was used for CBF quantification (mean and standard deviation of α in all MR studies was 0.68 ± 0.03). Pixels outside the brain were assigned to zero intensity for figure presentation.

Assessment of Brain Tissue Levels of 2-Chloroadenosine

Brain tissue levels of 2-CA were assessed in rats decapitated at either 3.5 (n = 7) or 24 h (n = 6) after intrahippocampal injection of 2-CA (12 nmol) using the protocol for injection exactly as described above. In each rat, the brain was rapidly removed and tissue was harvested from the left and right hippocampi and cortices mirroring the ROIs selected for imaging. The tissue was rapidly frozen in liquid nitrogen. Wet brain tissue was weighed (~50 mg), 100 µL of methanol and 10 µL of 200 pg/µL of phenacetin (internal standard) added, and the tissue homogenized. The homogenate was centrifuged and the supernatant collected. An equal volume of water was added to the supernatant, and 50 µL was assayed. A standard curve was performed in brain tissue from rats that had no treatment.

The assay was performed using a Thermofinnigan high-pressure liquid chromatograph system coupled to a Thermofinnigan LCQ Duo ion trap mass spectrometer equipped with an electrospray ionization (ESI) source. The analytical column was a C-18 Varian Microsorb. The mobile phase flow rate was 0.5 mL/min. The mobile phase changed from 100% water to 50% water/50% methanol over 5 mins with a linear gradient. This composition was kept constant for 2 mins, and then changed to 100% methanol over 1 min, and kept at this composition for 2 mins. The mobile phase was acidified with 0.1% formic acid. The mass spectrometer was operated in the ESI-positive ion mode. Nitrogen was used both as sheath and auxiliary gas at a pressure optimized automatically for each of the analytes during tuning of each analyte. The spray voltage and the heated capillary temperature were set at 5 kV and 270°C, respectively. Analysis was performed with two stages of ionization (MSMS), monitoring m/z of 170 for 2-CA (molecular ion m/z of 302 at 26% collision energy), and m/z of 138 for phenacetin (molecular ion m/z of 180 at 34% collision energy).

Deoxyglucose Autoradiography of Cerebral Glucose Utilization

Using anesthesia, surgery, and monitoring methods nearly identical to those in the experiments studying CBF by MR, separate groups of uninjured rats (craniotomy only) were injected with adenosine receptor agonists or respective vehicle and CMRglu was assessed by deoxyglucose autoradiography by the method of Sokoloff et al (1977), as modified by Kelly et al (2000), with minor additional modifications. At 3 h after injection of either 2-CA, CGS 21680, or their respective vehicles, 50 µCi of ¹⁴C-2-deoxy-d-glucose (2DG) was injected into the femoral vein over 30 secs. Timed arterial samples were collected for 45 mins, immediately chilled and centrifuged. In all, 20 µL of plasma was mixed with 20 µL of scintillation fluid, and radioactivity was determined (Tri-Carb 2,100 TR, Packard, Meriden, CT, USA). Studies were terminated by decapitation. Brains were rapidly removed and placed into powdered dry ice—then stored at −70°C. Physiological parameters were measured as described in the CBF studies. Coronal brain sections (20 µm) were cut at −20°C and processed for autoradiography. Sections were incubated with calibrated ¹⁴C standards (Amersham, Piscataway, NJ, USA) and exposed to Kodak Bio-Max film (2 weeks). Tissue radioactivity was determined using computer-based optical densitometry (MCID, Imaging Research, St Catharines, Ontario, Canada). Serum glucose levels were measured (Sigma, St Louis, MO, USA). 2-Deoxy-d-glucose-based CMRglu was determined in µmol 100 g⁻¹ min⁻¹. Coronal sections were taken through the parietal cortex and dorsal hippocampus, approximating the slice used for MRI. ROIs for the analysis of CMRglu included parietal cortex, hippocampus, and thalamus in left and right hemispheres.

Statistical Analysis

Data are presented as mean ± s.d. or median (range), where appropriate. Physiological data were compared between groups using either a two-way (treatment and time) analysis of variance (ANOVA) for repeated measures or Kruskal–Wallis test. In each of the protocols examining uninjured rats with serial assessment of CBF, data were analyzed with a two-way ANOVA for repeated measures. In protocols with analysis of CBF or CMRglu at a single time point, a Kruskal–Wallis test was used. Because the effect of treatment, injury severity, and time after injury were simultaneously assessed in the CCI protocol, CBF data from that study were assessed using a multivariate analysis. Two multivariate models were created, one for absolute CBF values from each group and the second for left (ipsilateral) minus right (contralateral) CBF values. A P-value < 0.05 was considered significant.

Results

Uninjured Rats

Protocol 1. Effect of 2-chloroadenosine on cerebral blood flow in uninjured rats early after injection: Injection of 2-CA produced a marked, localized, and consistent 2–3-fold increase in CBF- versus saline vehicle-injected rats (Table 2, Figure 1). Significant increases in CBF for treated versus vehicle were seen ipsilateral to injection in the hippocampal, cortical, medial cortical segment, and hemispheric ROIs (two-way ANOVA for repeated measures). There was no significant effect of time or time–treatment interaction in any ROI in the injected hemisphere. Controlled and observed physiological parameters did not differ between 2-CA- and saline-injected groups, with two exceptions (Table 2). MABP was lower at both ~3.5 and 5 h in rats injected with 2-CA versus saline (P < 0.05), possibly a result of central A₁ receptor activation (Tseng et al, 1988). However, this reduction was not below a MABP of 75 mm Hg in any rat. In addition, a statistically significant, albeit small (< 0.5°C), difference in rectal temperature was seen between groups. There was also no effect of time or treatment–time interaction for either MABP or temperature.

Figure 1

Acute effects of injection of either saline vehicle (left three vertical panels) or the nonselective adenosine receptor agonist 2-CA(12 nmol, right three vertical panels) into the left dorsal hippocampus of uninjured rats on CBF(mL 100 g⁻¹ min⁻¹) assessed by the continuous arterial spin-labeled MRI. The left hemisphere appears as the left side of image. The top horizontal panels show unlabeled control (Mc) images reflecting anatomy. Pseudocolor images are CBF maps. Numbers to the left of the images are times after injection. After injection of saline, CBF was not affected at either 2 h 52 mins or 4 h 38 mins, but at both 4 h 1 min and 5 h 28 mins after 2-CA administration CBF was increased in the ipsilateral hippocampus and overlying cortex in these representative examples.

Table 2

Regional CBF (mL/100 g min)^a and physiological parameters early after injection of either 2-CA or vehicle in uninjured rats

Treatment	Saline vehicle (n = 4)
Time postinjection (h:min)	3:19 ± 0:43	4:49 ± 0:38
PaCO₂ (mm Hg)	33.3 ± 5.4	34.9 ± 2.4
pH	7.44 ± 0.05	7.43 ± 0.02
PaO₂ (mm Hg)	197.5 ± 45.8	215.2 ± 12.2
BE (mmol/L)	–0.60 ± 3.37	–0.40 ± 2.13
Rectal temperature (°C)	37.4 ± 0.2	37.5 ± 0.2
MABP (mm Hg)	159.5 ± 9.0	145.2 ± 18.1

Region	Ipsilateral	Contralateral	Ipsilateral	Contralateral
Medial cortical segment	79 ± 7	85 ± 24	60 ± 8	67 ± 9
Cortex	81 ± 10	84 ± 17	68 ± 12	62 ± 8
Hippocampus	70 ± 7	68 ± 14	78 ± 16	74 ± 9
Thalamus	94 ± 18	86 ± 5	89 ± 18	90 ± 21
Amygdala/piriform cortex	47 ± 11	55 ± 11	53 ± 15	45 ± 5
Hemisphere	77 ± 10	81 ± 8	76 ± 9	77 ± 7

Treatment	12 nmol 2-CA (n = 4)
Time postinjection (h:min)	3:41 ± 0:22	5:01 ± 0:21
PaCO₂ (mm Hg)	37.9 ± 3.2	35.1 ± 4.3
pH	7.43 ± 0.02	7.42 ± 0.02
PaO₂ (mm Hg)	169.5 ± 45.8	190.8 ± 18.4
BE (mmol/L)	0.77 ± 1.05	–0.52 ± 1.80
Rectal temperature (°C)	37.9 ± 0.2*	37.7 ± 0.3*
MABP (mm Hg)	115.8 ± 19.8*	102.8 ± 23.0*

Region	Ipsilateral	Contralateral	Ipsilateral	Contralateral
Medial cortical segment	167 ± 85*	66 ± 10	157 ± 47*	54 ± 15
Cortex	136 ± 50*	69 ± 16	120 ± 15*	61 ± 6
Hippocampus	217 ± 50*	64 ± 12	210 ± 19*	57 ± 21
Thalamus	92 ± 7	84 ± 17	82 ± 19	77 ± 14
Amygdala/piriform cortex	35 ± 12	36 ± 15	31 ± 13	34 ± 7
Hemisphere	120 ± 26*	79 ± 15	109 ± 6*	69 ± 11

CBF, cerebral blood flow; 2-CA, 2-chloroadenosine; BE, base excess; MABP, mean arterial blood pressure.

Data are mean ± standard deviation.

Standard deviation in CBF represents the intragroup variability in the perfusion (M_C–M_L)/M_C maps.

P < 0.05 treatment effect for 2-CA versus vehicle (two-way ANOVA for repeated measures).

Protocol 2. Effect of 2-chloroadenosine on cerebral blood flow in uninjured rats at ~24 h after injection: Injection of 2-CA produced a remarkably sustained increase in CBF- versus saline vehicle-injected rats (Table 3, Figure 2). As in the studies assessing CBF early after injection, a consistent ~4-fold increase in CBF was seen in the hippocampal ROI ipsilateral to the injection site at ~24 h. Significant increases in CBF were also seen in cortical, medial cortical segment, and hemispheric ROIs ipsilateral to injection (Table 3, Kruskal–Wallis test). At 24 h after injection, significant albeit modest increases in CBF were also seen in three ROIs contralateral to injection (cortical, medial cortical segment, and hemispheric). Controlled and observed physiological parameters did not differ between 2-CA- and saline-injected groups, except again for an ~15% lower MABP at the time of CBF acquisition in 2-CA-versus vehicle-injected rats.

Figure 2

Delayed effects of injection of either saline vehicle (left two vertical panels) or the nonselective adenosine receptor agonist 2-CA(12 nmol, right two vertical panels) into the left dorsal hippocampus of uninjured rats on CBF (mL 100g⁻¹ min⁻¹), assessed by the continuous arterial spin-labeled MRI method. The left hemisphere appears as the left side of image. The top horizontal panels show unlabeled control (Mc) images reflecting anatomy. Pseudocolor images are CBF maps. Numbers to the left of the images are times after injection. After injection of saline, CBF was not affected at 20 h 22 mins, but at 26 h 8 mins after 2-CA administration CBF was increased in both the ipsilateral hippocampus and overlying cortex in these representative examples. This enduring effect on CBF was not anticipated, but is consistent with the sustained increases in brain tissue levels of 2-CA even at 24 h after injection.

Table 3

Regional CBF (mL/100 g min) and physiological parameters at ~24 h after injection of either 2-CA or vehicle in uninjured rats

Treatment	Saline vehicle (n = 3)	12 n mol 2-CA (n = 3)
Time postinjection (h:min)	21:43 (20:22, 27:16)	26:08 (24:52, 26:29)
PaCO₂ (mm Hg)	36.8 (31.8, 41.6)	42.0 (41.3, 43.9)
pH	7.46 (7.38, 7.48)	7.42 (7.38, 7.42)
PaO₂ (mm Hg)	214 (158, 226)	177 (173, 208)
BE (mmol/L⁻¹)	2.0 (0.80, 2.0)	1.1 (−0.6, 2.4)
Rectal temperature (°C)	37.1 (37.0, 37.2)	37.3 (37.2, 37.8)
MABP (mm Hg)	169 (168, 175)	147 (133, 152)*

Region	Ipsilateral	Contralateral	Ipsilateral	Contralateral
Medial cortical segment	66 (43, 82)	63(48, 66)	422 (239, 445)*	86(72, 100)*
Cortex	67 (43, 77)	68(52, 74)	293 (180, 310)*	92(89, 113)*
Hippocampus	67 (43, 85)	76(46, 101)	276 (274, 337)*	100 (90, 104)
Thalamus	74 (41, 117)	63(62, 123)	133 (98, 178)	115 (91, 133)
Amygdala/Piriform Cortex	44 (14, 53)	39(18, 70)	66 (22, 44)	63(49, 79)
Hemisphere	65 (43, 88)	68(53, 95)	176 (174, 215)*	107 (96, 121)*

CBF, cerebral blood flow; 2-CA, 2-chloroadenosine; BE, base excess; MABP, mean arterial blood pressure.

Data are median (range).

P < 0.05 treatment effect (Kruskal–Wallis test) for 2-CA versus vehicle.

Protocol 3. Assessment of 2-chloroadenosine levels in brain tissue at either 3.5 h or 24 h after injection in uninjured rats: Brain tissue levels of 2-CA were 12,567 ± 6,702, 1,381 ± 1,231, 1,604 ± 3,511, and 274 ± 215 pg/mg wet brain tissue in the left hippocampus, right hippocampus, left cortex, and right cortex at 3.5 h after left hippocampal injection, P < 0.05 for left hippocampus versus all other regions. Similarly, brain tissue levels of 2-CA were 2,095 ± 951, 210 ± 148, 262 ± 144, and 44 ± 57 pg/mg wet brain tissue in the left hippocampus, right hippocampus, left cortex, and right cortex at 24 h after left hippocampal injection, P < 0.05 for left hippocampus versus all other regions.

Protocol 4. Effect of CGS 21680 on cerebral blood flow in uninjured rats early after injection: intrahippocampal injection of the selective A_2A receptor agonist CGS 21680 produced even more marked increases in CBF than seen after 2-CA injection. The dramatic increases in CBF in treated versus vehicle-injected rats were nearly global in the slice-reaching significance in 10 of the 12 ROIs that were assessed—with the exception of the ipsilateral and contralateral amygdala (Table 4, Figure 3, two-way ANOVA for repeated measures). For example, four-and five-fold increases in CBF for treatment versus vehicle were observed in ispilateral hippocampal and ipsilateral hemispheric ROIs, respectively. A 2–3-fold increase in CBF for treatment versus vehicle was seen in the hemispheric ROI contralateral to injection. Similarly, a significant effect of time was seen in the ipsilateral hippocampus, thalamus, amygdala, and hemisphere, and contralateral thalamus. A significant treatment–time interaction was also seen in 9 of the 12 ROIs, with higher CBF at ~3.5 versus ~5 h after CGS 21680 injection in these ROIs. Controlled and observed physiological parameters did not differ between CGS 21680- and vehicle-injected groups, except for a statistically significant, albeit small (< 0.5°C), difference in rectal temperature between groups at ~5 h after injection. There were no time effects or treatment–time interactions on physiological parameters.

Figure 3

Acute effects of injection of either sterile water vehicle (left three vertical panels) or the selective adenosine_2A receptor agonist, CGS 21680 (6 nmol, right three vertical panels) into the left dorsal hippocampus of uninjured rats on CBF (mL 100 g⁻¹ min⁻¹) assessed by the continuous arterial spin-labeled MRI method. The left hemisphere appears as the left side of image. The top horizontal panels show unlabeled control (Mc) images reflecting anatomy. Pseudocolor images are CBF maps. Numbers to the left of the images are times after injection. Cerebral blood flow was not affected at either 3 h 7 mins or 4 h 20 mins after injection of vehicle, but at both 3 h 33 mins and 4 h 54 mins after CGS 21680 administration a marked and nearly global increase in CBF is seen in most ROIs within the imaging slice in these representative examples.

Table 4

Regional CBF (mL/100 g min)^a and physiological parameters early after injection of either CGS 21680 or vehicle in uninjured rats

Treatment	Water vehicle (n = 3)
Time postinjection (h:min)	3:44 ± 0:38	5:00 ± 0:41
PaCO₂ (mm Hg)	38.2 ± 4.7	38.8 ± 4.7
pH	7.38 ± 0.02	7.37 ± 0.02
PaO₂ (mm Hg)	194.3 ± 11.9	199.7 ± 5.5
BE (mmol L⁻¹)	–2.5 ± 2.0	–0.25 ± 2.2
Rectal temperature (°C)	37.2 ± 0.2	37.1 ± 0.1
MABP (mm Hg)	130.3 ± 52.3	132.7 ± 41.4

Region	Ipsilateral	Contralateral	Ipsilateral	Contralateral
Medial cortical segment	76 ± 37	83 ± 37	86 ± 29	86 ± 33
Cortex	89 ± 29	91 ± 33	101 ± 27	89 ± 39
Hippocampus	82 ± 7	67 ± 12	93 ± 18	85 ± 27
Thalamus	108 ± 25	103 ± 30	113 ± 33	112 ± 26
Amygdala/piriform cortex	48 ± 15	54 ± 25	60 ± 16	58 ± 23
Hemisphere	83 ± 18	85 ± 21	94 ± 19	91 ± 25

Treatment	6 nmol CGS21680 (n = 3)
Time postinjection (h:min)	3:25 ± 0:23	4:56 ± 0:36
PaCO₂ (mm Hg)	38.6 ± 5.3	39.0 ± 5.3
pH	7.41 ± 0.05	7.44 ± 0.11
PaO₂ (mm Hg)	161.7 ± 50.4	173.0 ± 43.3
BE (mmol L⁻¹)	–0.4 ± 1.0	–0.3 ± 1.4
Rectal temperature (°C)	37.4 ± 0.2*	37.5 ± 0.1*
MABP (mm Hg)	152.0 ± 5.3	152.0 ± 3.6

Region	Ipsilateral	Contralateral	Ipsilateral	Contralateral
Medial cortical segment	433 ± 104*	306 ± 113*	396 ± 54*	206 ± 59*
Cortex	440 ± 139*	230 ± 68*	364 ± 91*	175 ± 34*
Hippocampus	357 ± 33*	332 ± 35*	336 ± 28*	293 ± 22*
Thalamus	579 ± 142*	395 ± 131*	528 ± 130*	269 ± 65*
Amygdala/piriform cortex	129 ± 48	75 ± 22	84 ± 33	68 ± 10
Hemisphere	388 ± 91*	266 ± 67*	331 ± 68*	206 ± 33*

CBF, cerebral blood flow; CGS 21680, 2-p-(2-carboxyethyl)-phenethylamino-5‘-N-ethylcarbox-amidoadenosine; BE, base excess;

MABP, mean arterial blood pressure.

Data are mean ± standard deviation.

Standard deviation in CBF represents the intragroup variability in the perfusion (M_C–M_L)/M_C maps.

P < 0.05 treatment effect for CGS 21680 versus vehicle (two-way ANOVA for repeated measures).

Protocol 5. Effect of 2-chloroadenosine or CGS 21680 on CMRglu in uninjured rats early after injection: Despite marked increases in CBF at ~3 h after injection of either 2-CA or CGS 21680 versus the respective vehicle, neither treatment significantly altered CMRglu in any ROI in pentobarbital anesthetized rats—including ipsilateral hippocampus, cortex, and thalamus (Table 5, Kruskal–Wallis test). Controlled and observed physiological parameters did not differ between groups.

Table 5

Regional CMRglu (µmol/100 g min) and physiological parameters at 3 h after injection of either adenosine agonists or vehicle in uninjured rats

Treatment	Saline vehicle (n = 4)
PaCO₂ (mm Hg)	40 (30, 42)
pH	7.38 (7.36, 7.45)
PaO₂ (mm Hg)	248 (233, 253)
BE (mmol L⁻¹)	–0.8 (−2.2, −0.4)
Rectal temperature (°C)	37.2 (36.3, 37.4)
MABP (mm Hg)	102 (70, 145)

Region	Ipsilateral	Contralateral
Cortex	45.71 (35.36, 54.82)	46.00 (33.59, 56.59)
Hippocampus	36.20 (35.49, 56.13)	35.73 (33.39, 57.86)
Thalamus	44.71 (33.62, 63.09)	45.20 (33.42, 63.76)

Treatment	12 nmol 2-CA (n = 4)
PaCO₂ (mm Hg)	34 (31, 40)
pH	7.44 (7.36, 7.45)
PaO₂ (mm Hg)	232 (212, 257)
BE (mmol L⁻¹)	–1.5 (−1.7, −0.2)
Rectal temperature (°C)	36.8 (36.6, 36.9)
MABP (mm Hg)	138 (135, 155)

Region	Ipsilateral	Contralateral
Cortex	64.57 (33.51, 75.52)	69.26 (35.98, 88.84)
Hippocampus	63.86 (40.01, 74.40)	64.16 (40.83, 81.49)
Thalamus	62.68 (40.88, 84.86)	63.20 (40.30, 89.08)

Treatment	Sterile water vehicle (n = 4)
PaCO₂ (mm Hg)	40 (31, 43)
pH	7.35 (7.34, 7.41)
PaO₂ (mm Hg)	238 (234, 253)
BE (mmol L⁻¹)	–2.4 (−4.0, −1.7)
Rectal temperature (°C)	37.2 (37.1, 37.8)
MABP (mm Hg)	145 (135, 160)

Region	Ipsilateral	Contralateral
Cortex	56.24 (27.68, 72.11)	53.94 (28.65, 71.54)
Hippocampus	46.74 (29.54, 63.47)	45.33 (27.59, 58.95)
Thalamus	50.28 (26.98, 67.08)	51.10 (26.86, 63.75)

Treatment	6 nmol CGS 21680 (n = 4)
PaCO₂ (mm Hg)	37 (34, 38)
pH	7.40 (7.39, 7.41)
PaO₂ (mm Hg)	244 (237, 249)
BE (mmol L⁻¹)	–1.4 (−2.8, −0.4)
Rectal temperature (°C)	37.0 (36.7, 37.4)
MABP (mm Hg)	142 (128, 150)

Region	Ipsilateral	Contralateral
Cortex	36.32 (22.34, 82.06)	34.11 (20.59, 83.57)
Hippocampus	36.15 (21.02, 59.33)	33.67 (21.04, 70.90)
Thalamus	36.34 (21.29, 89.58)	33.94 (19.96, 79.63)

CMRglu, cerebral metabolic rate for glucose; 2-CA, 2-chloroadenosine; CGS 21680, 2-p-(2-carboxyethyl)-phenethylamino-5‘-N-ethylcarbox-amidoade-nosine; BE, base excess; MABP, mean arterial blood pressure.

Data are median (range).

Controlled Cortical Impact Model

Protocol 6. Effect of 2-chloroadenosine on cerebral blood flow in rats after CCI: effect of injury severity and time: Cerebral blood flow in individual ROIs is shown for studies in vehicle- and 2-CA-treated rats after moderate (Table 6) and severe (Table 7) CCI. Representative examples show that the focal increases in CBF produced by 2-CA after CCI (Figures 4 and 5) were more heterogeneous than those previously shown in uninjured rats. Multivariate analysis (Table 8) revealed that posttraumatic increases in CBF mediated by 2-CA were local, reaching significance in the ipsilateral hippocampal and hemispheric ROIs (both P < 0.005 for 2-CA versus vehicle). Multivariate analyses also confirmed that CBF recovery in ipsilateral medial cortical segment, cortex, and hemispheric ROIs after CCI was dependent on the severity of the injury (all P < 0.005 for moderate versus severe injury), with lower flows being observed at the more severe injury level. Similarly, CBF recovery in the ipsilateral hippocampus, medial cortical segment, cortex, and hemispheric ROIs after CCI was also dependent on time after injury (all P < 0.005 for first (~3.5 h) versus second (~5 h) CBF assessment) when assessing this parameter using injured minus noninjured values in respective ROIs. Physiologic parameters were not different between treatment groups, except for a statistically significant, albeit modest difference in temperature between treatment and control groups at the time of CBF acquisition (two-way ANOVA for repeated measures). Mean rectal temperature was never more than 0.7°C different between groups (Tables 6 and 7).

Figure 4

Acute effects of injection of either saline vehicle (left three vertical panels) or the nonselective adenosine receptor agonist 2-CA (12 nmol, right three vertical panels) into the left dorsal hippocampus of rats on CBF (mL 100 g⁻¹ min⁻¹, assessed by the continuous arterial spin-labeled MRI) after CCI at a moderate (2.0 mm deformation) injury level. The left hemisphere appears as the left side of image. The top horizontal panels show unlabeled control (Mc) images reflecting anatomy. Pseudocolor images are CBF maps. Numbers to the left of the images are times after injection, with time after CCI indicated in parentheses. Some local swelling in the Mc images at the injury site is seen after moderate injury. Cerebral blood flow was not affected at either 3 h 12 mins or 5 h 3 mins after injection of saline, but at both 2 h 58 mins and 4 h 9 mins after 2-CA administration CBF was increased in the ipsilateral hippocampus, overlying cortex, and brain regions in proximity to the contusion in these representative examples at the moderate injury level.

Figure 5

Acute effects of injection of either saline vehicle (left three vertical panels) or the nonselective adenosine receptor agonist 2-CA (12 nmol, right three vertical panels) into the left dorsal hippocampus of rats on CBF (assessed by the continuous arterial spin-labeled MRI) after CCI at a severe (2.5 mm deformation) injury level. The left hemisphere appears as the left side of image. The top horizontal panels show unlabeled control (Mc) images reflecting anatomy. Pseudocolor images are CBF maps. Numbers to the left of the images are time after injection, with time after CCI indicated in parentheses. Marked local swelling in the Mc images at the injury site is readily seen after severe injury. Cerebral blood flow was not affected at either 4 h 1 min or 5 h 42 mins after injection of saline, and hypoperfusion is seen at the contusion site. At both 2 h 57 mins and 4 h 6 mins after 2-CA administration, less posttraumatic hypoperfusion is evident in brain regions in proximity to the contusion in these representative examples at the severe injury level.

Table 6

Regional CBF (mL/100 gmin)^a and physiological parameters after CCI (deformation depth = 2.0 mm) of moderate injury level: studies of 2-CA versus vehicle

Treatment	Saline vehicle (n = 3)
Time postinjection (h:min)	3:21 ± 0:10	4:44 ± 0:17
PaCO₂ (mm Hg)	38.3 ± 3.9	36.6 ± 4.0
pH	7.36 ± 0.04	7.35 ± 0.02
PaO₂ (mm Hg)	189.7 ± 19.0	205.3 ± 3.5
BE (mmol L⁻¹)	–3.3 ± 1.4	–3.0 ± 0.8
Rectal temperature (°C)	37.4 ± 0.5	37.5 ± 0.4
MABP (mm Hg)	117.7 ± 36.6	114.3 ± 27.2

Region	Ipsilateral	Contralateral	Ipsilateral	Contralateral
Medial cortical segment	63 ± 20	77 ± 40	66 ± 24	71 ± 13
Cortex	68 ± 31	83 ± 46	76 ± 10	79 ± 10
Hippocampus	43 ± 27	75 ± 49	64 ± 9	65 ± 11
Thalamus	92 ± 35	96 ± 36	94 ± 15	94 ± 16
Amygdala/piriform cortex	42 ± 24	46 ± 18	45 ± 15	46 ± 13
Hemisphere	68 ± 28	77 ± 36	75 ± 9	73 ± 8

Treatment	12 nmol 2-CA (n = 3)
Time postinjection (h:min)	3:33 ± 0:31	4:44 ± 0:31
PaCO₂ (mm Hg)	39.5 ± 2.2	39.3 ± 4.0
pH	7.36 ± 0.01	7.40 ± 0.02
PaO₂ (mm Hg)	185.0 ± 26.1	200.0 ± 7.2
BE (mmol L⁻¹)	–3.3 ± 1.0	–3.9 ± 0.7
Rectal temperature (°C)	36.7 ± 0.3	37.4 ± 0.6
MABP (mm Hg)	151.3 ± 4.0	138.7 ± 9.9

Region	Ipsilateral	Contralateral	Ipsilateral	Contralateral
Medial cortical segment	73 ± 35	84 ± 27	83 ± 44	70 ± 10
Cortex	85 ± 33	88 ± 29	89 ± 34	82 ± 10
Hippocampus	161 ± 75*	89 ± 19	161 ± 69*	87 ± 12
Thalamus	90 ± 21	93 ± 23	96 ± 17	92 ± 14
Amygdala/piriform cortex	40 ± 3	40 ± 8	43 ± 9	46 ± 8
Hemisphere	93 ± 29*	87 ± 22	97 ± 20*	87 ± 11

CBF, cerebral blood flow; 2-CA, 2-chloroadenosine; CCI, controlled cortical impact; BE, base excess; MABP, mean arterial blood pressure.

Data are mean ± standard deviation.

Standard deviation in CBF represents the intragroup variability in the perfusion (M_C–M_L)/M_C maps.

P < 0.05 versus respective vehicle by multivariate analysis.

Table 7

Regional CBF (mL/100 g min)^a and physiological parameters after CCI (deformation depth = 2.5 mm) of severe injury level: studies of 2-CA versus vehicle

Treatment	Saline vehicle (n = 3)
Time postinjection (h:min)	3:17 ± 0:42	4:42 ± 0:55
PaCO₂ (mm Hg)	40.9 ± 3.9	38.1 ± 3.5
pH	7.38 ± 0.03	7.40 ± 0.02
PaO₂ (mm Hg)	188.0 ± 36.5	188.0 ± 28.8
BE (mmol L⁻¹)	–0.8 ± 2.2	–1.0 ± 2.2
Rectal temperature (°C)	37.7 ± 0.2	38.0 ± 0.6
MABP (mm Hg)	136.3 ± 22.1	126.0 ± 15.7

Region	Ipsilateral	Contralateral	Ipsilateral	Contralateral
Medial cortical segment	34 ± 3	63 ± 8	43 ± 5	70 ± 18
Cortex	48 ± 7	69 ± 8	55 ± 2	74 ± 20
Hippocampus	51 ± 12	79 ± 27	60 ± 10	76 ± 22
Thalamus	87 ± 15	86 ± 12	91 ± 16	78 ± 22
Amygdala/piriform cortex	40 ± 19	46 ± 21	38 ± 16	41 ± 22
Hemisphere	60 ± 10	76 ± 9	64 ± 8	75 ± 15

Treatment	12 nmol 2-CA (n = 3)
Time postinjection (h:min)	3:06 ± 0:17	4:24 ± 0:30
PaCO₂ (mm Hg)	42.5 ± 1.4	41.7 ± 4.9
pH	7.38 ± 0.02	7.38 ± 0.01
PaO₂ (mm Hg)	185.7 ± 15.0	191.7 ± 11.8
BE (mmol L⁻¹)	–1.1 ± 3.0	–0.6 ± 1.8
Rectal temperature (°C)	37.0 ± 0.2	37.3 ± 0.4
MABP (mm Hg)	131.0 ± 30.6	125.0 ± 28.8

Region	Ipsilateral	Contralateral	Ipsilateral	Contralateral
Medial cortical segment	56 ± 30	96 ± 31	60 ± 37	71 ± 21
Cortex	69 ± 33	99 ± 33	71 ± 33	80 ± 27
Hippocampus	89 ± 20*	83 ± 36	99 ± 23*	90 ± 44
Thalamus	111 ± 41	102 ± 47	98 ± 38	106 ± 39
Amygdala/piriform cortex	47 ± 39	57 ± 33	41 ± 32	49 ± 45
Hemisphere	83 ± 32*	90 ± 39	82 ± 31*	89 ± 39

CBF, cerebral blood flow; 2-CA, 2-chloroadenosine; CCI, controlled cortical impact; BE, base excess; MABP, mean arterial blood pressure.

Data are mean ± standard deviation.

Standard deviation in CBF represents the intragroup variability in the perfusion (M_C–M_L)/M_C maps.

P < 0.05 versus respective vehicle by multivariate analysis.

Discussion

Effect of 2-Chloroadenosine on Cerebral Blood Flow in Uninjured Rat Brain

The marked increase in regional CBF produced by intrahippocampal injection of 2-CA in uninjured rat brain expands the findings of our prior study using cortical injection (Kochanek et al, 2001). 2-Chloroadenosine is a nonselective adenosine receptor agonist with receptor affinities similar to those of adenosine. Despite the use of identical doses in this and our prior report (Kochanek et al, 2001), intrahippocampal administration of 2-CA produced a more localized distribution of increase in CBF than cortical administration. Early increases in CBF were restricted to the injected hemisphere. This increase in CBF in our study is consistent with the ~2.9-fold increase produced by infusion of a 10^{− 4} mol/L solution of 2-CA into the caudate nucleus in rats (Van Wylen et al, 1989). Our 24 h studies revealed that the increase in CBF is remarkably enduring. Cerebral blood flow was at least as high in the ipsilateral hippocampus and cortex at 24 h after injection as seen early after injection. Prolonged effects of 2-CA on CBF have not received prior study. However, 2-CA is a relatively nonmetabolizable analog and parenchymal injection into caudate nucleus produces anticonvulsant effects that last up to three days (Abdul-Ghani et al, 1997). Consistent with those studies, 2-CA levels were still 2,095 ± 951 pg/mg wet brain tissue in hippocampus ipsilateral to injection at 24 h. This is equal to 6.94 pmol/mg tissue—which is ~1 µL of tissue volume. This represents a 2-CA concentration of 6.94 µmol/L. The equilibrium dissociation constant of 2-CA for A_2A receptors is 0.46 µmol/L. Therefore, the concentration of 2-CA in hippocampus at 24 h after injection is 15-fold greater than the equilibrium dissociation constant and would provide near-saturation levels at A_2A receptors. Even cortical levels were in a range that would activate A_2A receptors at 3.5 and 24 h after injection. The low lipid solubility of 2-CA may also delay parenchymal clearance (De Sarro et al, 1991). Our data, thus, implicate sustained A_2A receptor activation in mediating the sustained increase in CBF. MABP was lower in rats after intrahippocampal injection of 2-CA versus saline vehicle. This finding was surprising since our prior report did not reveal an effect of intracortical injection of 2-CA on MABP (Kochanek et al, 2001). Few prior experiments studied the 12 nmol dose. This effect likely results from central A₁-receptor-mediated effects on sympathetic tone (Tseng et al, 1988). This modest MABP reduction would not be expected to increase CBF in uninjured rat brain. The focal nature of the increases in CBF supports local effects of 2-CA on CBF. Marked increases in CBF, without effects on MABP in the studies of CGS 21680, support potent local effects of adenosine agonists on CBF.

Effect of the A2A Receptor Agonist CGS 21680 on Cerebral Blood Flow in Uninjured Rat Brain

Cerebrovasodilatory effects of adenosine are mediated by actions at A_2A and A_2B receptors located on vascular smooth muscle and endothelium (Ngai et al, 2001; Shin et al, 2000). Before studying 2-CA in experimental TBI, we thought it important to show that local administration of a selective A_2A receptor agonist increases CBF in our paradigm. CGS 21680 exhibits ~140-fold selectivity for the A_2A versus A₁ receptor (Hutchison et al, 1989). Applied to the cortical surface of the rat brain, CGS 21680 increased CBF (Coney and Marshall, 1988). The CGS 21680 dose of 6 nmol was based on pilot studies. Parenchymal injection produced marked increases in CBF in nearly every ROI—in both hemispheres—suggesting rapid diffusion throughout the brain and the dramatic effects of A_2A receptor activation. In isolated rat cerebral arterioles, the A_2A receptor antagonist ZM-241385 only partially reversed vasodilation by adenosine, suggesting a possible contribution of A_2B receptor activation (Ngai et al, 2001). We did not study the effect of A_2B receptor activation because of the lack of availability of a selective agonist. We also did not carry out studies combining A_2A agonists and antagonists in our model. Such studies could contribute additional mechanistic insight. CGS 21680 does not cross the blood–brain barrier (Jacobson and Van Rhee, 1997) and no effect on MABP was seen.

Effect of Adenosine Receptor Agonists on CMRglu in Uninjured Rat Brain

Neither 2-CA nor CGS 21680 affected CMRglu in any ROI in pentobarbital-anesthetized rats. This suggests that the vasodilatory effects of intrahippocampal injection of either 2-CA or CGS 21680 do not result from an increase in CMR. Our CMRglu values in vehicle-treated rats are similar to those reported by Warner et al (1996). Metabolic suppression would be expected after administration of the nonselective agonist 2-CA, based on studies showing potent anticonvulsant actions of 2-CA (Pourgholami et al, 1997; De Sarro et al, 1991). Although studies of the effects of adenosine receptor agonists on CMRglu using a similar paradigm (i.e., intraparenchymal administration, pentobarbital anesthesia) have not, to our knowledge, been reported, McBean et al (1989) reported that a 15-min intracarotid infusion of 2-CA (10⁻⁹ mol/min) had no effect on CMRglu in halothane-anesthetized rats. An increase in CMRglu might be expected after parenchymal injection of CGS 21680, based on its ability to facilitate synaptic transmission by ~14% in rat hippocampal slices (Cunha and Ribeiro, 2000). However, intravenous administration of CGS 21680 reduced CMRglu in unanesthetized rats (Nehlig et al, 1994). We saw no effect of intrahippocampal administration of this agent on CMRglu—despite a marked global increase in CBF.

Effect of 2-Chloroadenosine on Cerebral Blood Flow after Traumatic Brain Injury

We examined 2-CA rather than the selective A_2A receptor agonist CGS 21680 after TBI because 2-CA has been reported to reduce functional deficits in both the fluid percussion and CCI models (Headrick et al, 1994; Varma et al, 2002). We observed that CBF promotion by 2-CA after TBI depends on injury severity. After moderate CCI, 2-CA produced supranormal CBF in the ipsilateral hippocampal ROI—however, this response is less pronounced than that seen after injection of 2-CA in uninjured hippocampus. After severe CCI, mean CBF levels in hippocampus ipsilateral to 2-CA injection were increased versus vehicle-injected groups. However, CBF was similar to uninjured values in our model. This suggests that 2-CA was able to restore CBF only to the normal range early after severe injury. Early posttraumatic hypoperfusion is common in experimental and clinical TBI, and may mediate secondary damage (Yamakami and McIntosh, 1989; Bryan et al, 1995; Bouma et al, 1992); however, the optimal level of CBF early after injury is not known, and changes in metabolic demands are temporally complex. Kelly et al (2000), using a moderate CCI injury level, reported a marked initial increase in CMRglu, followed by metabolic depression by 3 h. Headrick et al (1994) reported that ICV administration of 5 nmol of 2-CA in rats attenuated the loss in phosphorylation potential at 2 to 4 h after fluid-percussion injury. Taken with our study and the work of Varma et al (2002), after experimental TBI, beneficial effects of local administration of 2-CA are seen on CBF, cerebral metabolism, and functional outcome. Multivariate analysis also revealed an effect of time after injury on CBF. Most investigators have reported an initial reduction in CBF, followed by progressive recovery after experimental TBI (Yamakami and McIntosh, 1989; Kochanek et al, 1995). In saline-treated rats in this study, after severe CCI, CBF recovered by ~5% to 50% between ~3.5 and ~5 h in ROIs beneath the impact site. Finally, the spatial pattern of the effect of 2-CA on CBF was less consistent after CCI versus the uninjured condition. This could result from variable vascular responsivity, vascular disruption, or alterations in drug distribution after injury. Brain interstitial levels of adenosine are increased 61-fold after severe CCI in our model in the initial 30 mins, however; they nearly normalize by 40 mins (Bell et al, 1998). The increase in CBF seen after 2-CA administration in our paradigm suggests that, by 3.5 h after injury, A_2A receptors are unlikely to be saturated by endogenous adenosine. However, the time course of changes in the distribution and response of adenosine receptors in rat brain after TBI remain undefined.

Limitations

First, we administered the adenosine receptor agonists by intrahippocampal injection to minimize systemic effects. This may limit the clinical relevance of these findings. Although we are not suggesting clinical testing of this approach based on these initial studies, ventriculostomy drainage is routinely used in clinical TBI. Intraventricular administration of adenosine agonists would be feasible. Second, we did not assess ICP. Some studies of CBF after experimental TBI in rats have assessed ICP (Kroppenstedt et al, 2002; Cherian et al, 1999); most have not (Yamakami and McIntosh, 1989; Bryan et al, 1995; Kelly et al, 2000; Thomale et al, 2002). The use of MRI placed additional limits on monitoring capabilities. We cannot prove that the local increase in CBF by 2-CA after CCI was not limited by intracranial hypertension. Third, CGS 21680 is the most selective commercially available A_2A agonist. However, it also binds to a second high-affinity site in the hippocampus and cortex, distinct from the A_2A receptor (Cunha et al, 1996). Nevertheless, it is used in most studies of the role of this receptor (Ngai et al, 2001; Shin et al, 2000). Fourth, we did not study CBF immediately after injury. Future studies should pursue earlier CBF effects. Fifth, as discussed above, the increase in CBF after administration of 2-CA was attenuated at the more severe injury level. This could importantly limit its efficacy in severely contused brain regions, where the vasculature might be unresponsive. The responsivity of isolated cerebral vessels in rat is directly related to injury severity (Golding et al, 2002). Fluid percussion injury has been shown to blunt both G-protein-mediated vasodilatation and agonist-mediated elevations in cyclic nucleotides (Armstead, 1998, 2003), both of which could be relevant to the attenuation of 2-CA-mediated effects on CBF at the higher injury level. Sixth, we chose the dose of 2-CA based on our prior studies in rats assessing effects on CBF (Kochanek et al, 2001). However, the optimal dose of any adenosine receptor agonist after CCI in our model remains to be determined in studies assessing outcome. Finally, the suggestion to augment CBF after TBI through via A_2A receptor activation must be reconciled with findings that in some studies A_2A receptor antagonists are neuroprotective in models of cerebral ischemia and Parkinson's and Huntington's disease (O'Regan et al, 1992; Higashi et al, 2002; Popoli et al, 2002).

Table 8

Multivariate analyses of^aCBF after CCI; effects of treatment, time after injury, and severity

Brain region	Treatment (2-CA versus vehicle)	Time after CCI	Injury severity (moderate versus severe)^b
Multivariate model using absolute CBF
Medical cortical segment	0.1689^c	0.2907	0.0527
Cortex	0.1261	0.3900	0.0810
Hippocampus	0.0004*	0.0786	0.1071
Thalamus	0.4868	> 0.999	0.7280
Amygdala/piriform cortex	0.8813	0.9595	0.9370
Hemisphere	0.0201*	0.4924	0.2399
Multivariate model using left (traumatized) minus right (non-traumatized) differences in CBF
Medical cortical segment	0.5728	0.0013*	0.0372*
Cortex	0.4610	0.0007*	0.0353*
Hippocampus	0.0022*	0.0108*	0.0642
Thalamus	0.5094	0.7378	0.1325
Amygdala/piriform cortex	0.6636	0.6650	0.4299
Hemisphere	0.0086*	0.0172*	0.0002*

CBF, cerebral blood flow; CCI, controlled cortical impact; 2-CA, 2-chloroadenosine.

Effects of treatment, time, and injury severity shown for CBF in ROIs of the left (injured) hemisphere.

Moderate injury, 2.0 mm deformation; severe injury, 2.5 mm deformation.

P-value.

P < 0.05.

Conclusions

Using MRI, we report marked, regional, and enduring increases in CBF after intrahippocampal injection of the nonselective adenosine receptor agonist 2-CA into uninjured rat brain. Marked global increases in CBF are seen after administration of the selective A_2A receptor agonist CGS 21680. Neither agonist affected CMRglu. 2-Chloroadenosine attenuated posttraumatic hypoperfusion after experimental TBI. Future studies assessing the effect of adenosine-related therapies in CNS injury should address the potential contribution of important effects on CBF.

Footnotes

Acknowledgements

The authors thank Dr Larry Jenkins for helpful review and Marci Provins for assistance with preparation of the manuscript.

References

Abdul-Ghani

Attwell

Bradford

(1997) The protective effect of 2-chloroadenosine against the development of amygdala kindling and on amygdala-kindled seizures. Eur J Pharmacol 326:7–14

Aden

Halldner

Lagercrantz

Dalmau

Ledent

Fredholm

(2003) Aggravated brain damage after hypoxic ischemia in immature adenosine A_2A knockout mice. Stroke 34:739–44

Armstead

(1998) Brain injury impairs prostaglandin cerebrovasodilation. J Neurotrauma 15:721–9

Armstead

(1999) Role of endothelin-1 in age-dependent cerebrovascular hypotensive responses after brain injury. Am J Physiol 277:H1884–94

Armstead

(2003) Protein kinase C activation generates superoxide and contributes to impairment of cerebrovasodilation induced by G protein activation after brain injury. Brain Res 971:153–60

Bell

Kochanek

Carcillo

Schiding

Wisniewski

Clark

RSB

Dixon

Marion

Jackson

(1998) Interstitial adenosine, inosine, and hypoxanthine, are increased after experimental traumatic brain injury in the rat. J Neurotrauma 15:163–70

Bouma

Muizelaar

Stringer

Choi

Fatouros

Young

(1992) Ultra-early evalution of regional cerebral blood flow in severely head-injured patients using xenon-enhanced computerized tomography. J Neurosurg 77:360–8

Bryan

Cherian

Robertson

(1995) Regional cerebral blood flow after controlled cortical impact injury in rats. Anesth Analg 80:687–95

Cassada

Tribble

Long

Kaza

Linden

Rieger

Rosin

Kron

Kern

(2002b) Adenosine A_2A agonist reduces paralysis after spinal cord ischemia: correlation with A_2A receptor expression on motor neurons. Ann Thorac Surg 74:846–50

10.

Cassada

Tribble

Young

Gangemi

Gohari

Butler

Rieger

Kron

Linden

Kern

(2002a) Adenosine A_2A analogue improves neurologic outcome after spinal cord trauma in the rabbit. J Trauma 53:225–9

11.

Chen

Huang

Zhu

Moratalla

Standaert

Moskowitz

Fink

Schwarzschild

(1999) A_2A Adenosine receptor deficiency attenuates brain injury induced by transient focal ischemia in mice. J Neurosci 19:9192–200

12.

Chen

Petzer

Staal

Beilstein

Sonsalla

Castagnoli

Jr Schwarzschild

(2001) Neuroprotection by caffeine and A(2A) adenosine receptor inactivation in a model of Parkinson's disease. J Neurosci 21:RC143

13.

Cherian

Chacko

Goodman

Robertson

(1999) Cerebral hemodynamic effects of phenylephrine and l-arginine after cortical impact injury. Crit Care Med 27:2512–7

14.

Cherian

Robertson

(2003) l-arginine and free radical scavengers increase cerebral blood flow and brain tissue nitric oxide concentrations after controlled cortical impact injury in rats. J Neurotrauma 20:77–85

15.

Coney

Marshall

(1988) Role of adenosine and its receptors in the vasodilatation induced in the cerebral cortex of the rat by systemic hypoxia. J Physiol 509: 507–18

16.

Cunha

Johansson

Constantino

Sebastiao

Fredholm

(1996) Evidence for high-affinity binding sites for the adenosine A_2A receptor agonist [3 H] CGS 21680 in the rat hippocampus and cerebral cortex that are different from striatal A_2A receptors. Naunyn-Schmiedeberg's Arch Pharmacol 353:261–71

17.

Cunha

Ribeiro

(2000) Adenosine A_2A receptor facilitation of synaptic transmission in the CA1 area of the rat hippocampus requires protein kinase C but not protein kinase A activation. Neurosci Lett 289:127–30

18.

De Sarro

Meldrum

(1991) Anti-convulsant action of 2-chloroadenosine injected focally into the inferior colliculus and substantia nigra. Eur J Pharmacol 194:145–52

19.

Detre

Leigh

Williams

Koretsky

(1992) Perfusion imaging. Magn Reson Med 23:37–45

20.

DeWitt

Smith

Deyo

Miller

Uchida

Prough

(1997) l-arginine and superoxide dismutase prevent or reverse cerebral hypoperfusion after fluid-percussion traumatic brain injury. J Neurotrauma 14: 223–33

21.

Dixon

Widdowson

Richardson

(1997) Desensi-tisation of the adenosine A₁ receptor by the A_2A receptor in the rat striatum. J Neurochem 69:315–21

22.

Dixon

Kochanek

Yan

Schiding

Griffith

Baum

Marion

DeKosky

(1999) One-year study of spatial memory performance, brain morphology, and cholinergic markers after moderate controlled cortical impact in rats. J Neurotrauma 16:109–22

23.

Dunwiddie

Worth

(1982) Sedative and anticonvulsant effects of adenosine analogs in mouse and rat. J Pharmacol Exp Ther 220:70–6

24.

Evans

Swan

Meldrum

(1987) An adenosine analogue, 2-chloroadenosine, protects against long term development of ischaemic cell loss in the rat hippocampus. Neurosci Lett 83:287–92

25.

Forbes

Hendrich

Kochanek

Williams

Schiding

Wisniewski

Kelsey

DeKosky

Graham

Marion

(1997) Assessment of cerebral blood flow and CO₂ reactivity after controlled cortical impact by perfusion magnetic resonance imaging using arterial spin labeling in rats. J Cereb Blood Flow Metab 17:865–74

26.

Golding

Bryan

Golding

(2002) The effect of surgery and severe injury on blood vessels. Physiol Meas 23:615–28

27.

Headrick

Bendall

Faden

Vink

(1994) Dissociation of adenosine levels from bioenergetic state in experimental brain trauma: potential role in secondary injury. J Cereb Blood Flow Metab 14:853–61

28.

Hendrich

Kochanek

Williams

Schiding

Marion

(1999) Early perfusion after controlled cortical impact in rats: quantification by arterial spin-labeled MRI and the influence of spin-lattice relaxation time heterogeneity. Magn Reson Med 42: 673–81

29.

Higashi

Meno

Marwha

Winn

(2002) Hippocampal injury and neurobehavioral deficits following hyperglycemic cerebral ischemia: effect of theophylline and ZM 241385. J Neurosurg 96:117–26

30.

Hovda

Lee

Smith

Von Stuck

Bergsneider

Kelly

Shalmon

Martin

Caron

Mazziotta

(1995) The neurochemical and metabolic cascade following brain injury: moving from animal models to man. J Neurotrauma 12:903–6

31.

Hutchison

Webb

Oei

Ghai

Zimmerman

Williams

(1989) CGS 21680C, an A₂ selective adenosine receptor agonist with preferential hypotensive activity. J Pharmacol Exp Ther 251:47–55

32.

Ibayashi

Ngai

Meno

Winn

(1991) Effects of topical adenosine analogs and forskolin on rat pial arterioles in vivo. J Cereb Blood Flow Metab 11:72–6

33.

Jacobson

Van Rhee

(1997) Development of selective purinoceptor agonists and antagonists. In: Purinergic Approaches in Experimental Therapeutics ( Jacobson

Jarvis

, eds), New York, NY: Wiley-Liss, Inc, 103

34.

Joshi

Duong

Mangla

Wang

Libow

Popilskis

Ostapkovich

Wang

Young

Pile-Spellman

(2002) In nonhuman primates intracarotid adenosine, but not sodium nitroprusside, increases cerebral blood flow. Anesth Analg 94:393–9

35.

Kalaria

Harik

(1988) Adenosine receptors and the nucleoside transporter in human brain vasculature. J Cereb Blood Flow Metab 8:32–9

36.

Kelly

Kozlowski

Haddad

Echiverri

Hovda

Lee

(2000) Ethanol reduces metabolic uncoupling following experimental head injury. J Neurotrauma 17:261–72

37.

Kochanek

Hendrich

Robertson

Williams

Melick

Marion

Jackson

(2001) Assessment of the effect of 2-chloroadenosine in normal rat brain using spin-labeled MRI measurement of perfusion. Magn Reson Med 45:924–9

38.

Kochanek

Marion

Zhang

Schiding

White

Palmer

Clark

RSB

O'Malley

Styren

DeKosky

(1995) Severe cortical impact in rats: assessment of cerebral edema, blood flow, and contusion volume. J Neurotrauma 12:1015–25

39.

Kroppenstedt

Sakowitz

Thomale

Unterberg

Stover

(2002) Influence of norepinephrine and dopamine on cortical perfusion, EEG activity, extracellular glutamate, and brain edema in rats after controlled cortical impact injury. J Neurotrauma 19: 1421–32

40.

McBean

Grome

Harper

(1989) A cerebral vasoconstrictive effect of some adenosine analogues. A study of adenosine analogues on local cerebral blood flow and glucose utilization in the rat. J Cereb Blood Flow Metab 9:548–55

41.

Nehlig

Daval

Boyet

(1994) Effects of selective adenosine A₁ and A₂ receptor agonists and antagonists on local rates of energy metabolism in the rat brain. Eur J Pharmacol 258:57–66

42.

Ngai

Coyne

Meno

West

Winn

(2001) Receptor subtypes mediating adenosine-induced dilation of cerebral arterioles. Am J Physiol Heart Circ Physiol 280:H2329–35

43.

O'Regan

Simpson

Perkins

Phillis

(1992) The selective A₂ adenosine receptor agonist CGS 21680 enhances excitatory transmitter amino acid release from the ischemic rat cerebral cortex. Neurosci Lett 138:169–72

44.

Pedata

Corsi

Melani

Bordoni

Latini

(2001) Adenosine extracellular brain concentrations and role of A_2A receptors in ischemia. Ann NY Acad Sci 939:74–84

45.

Popoli

Pintor

Domenici

Frank

Tebano

Pezzola

Scarchilli

Quarta

Reggio

Malchiodi-Albedi

Falchi

Massotti

(2002) Blockade of striatal adenosine A_2A receptor reduces, through a presynaptic mechanism, quinolinic acid-induced excitotoxicity: possible relevance to neuroprotective interventions in neurodegenerative diseases of the striatum. J Neurosci 22:1967–75

46.

Pourgholami

Rostampour

Mirnajafi-Zadeh

Palizvan

(1997) Intra-amygdala infusion of 2-chloroadenosine suppresses amygdala-kindled seizures. Brain Res 775:37–42

47.

Shin

Hong

(2000) Role of adenosine A_2B receptors in vasodilation of rat pial artery and cerebral blood flow autoregulation. Am J Physiol Heart Circ Physiol 278:H339–44

48.

Simpson

O'Regan

Perkins

Phillis

(1992) Excitatory transmitter amino acid release from the ischemic rat cerebral cortex: effects of adenosine receptor agonists and antagonists. J Neurochem 58:1683–90

49.

Sokoloff

Reivich

Kennedy

Des Rosiers

Patlak

Pettigrew

Sakurada

Shinohara

(1977) The [14C]deoxyglucose method for the measurement of local cerebral glucose utilization: theory, procedure, and normal values in the conscious and anesthetized albino rat. J Neurochem 28:897–916

50.

Sollevi

Ericson

Ericksson

Lindqvist

Langerkranser

Stone-Elander

(1987) Effect of adenosine on human cerebral blood flow as determined by positron emission tomography. J Cereb Blood Flow Metab 7:673–8

51.

Soricelli

Postiglione

Cuocolo

De Chiara

Ruocco

Brunetti

Salvatore

Ell

(1995) Effect of adenosine on cerebral blood flow as evaluated by single-photon emission computed tomography in normal subjects and in patients with occlusive carotid disease. A comparison with acetazolamide. Stroke 26:1572–6

52.

Stange

Greitz

Ingvar

Hindmarsh

Sollevi

(1997) Global cerebral blood flow during infusion of adenosine in humans: assessment by magnetic resonance imaging and positron emission tomography. Acta Physiol Scand 160:117–22

53.

Statler

Kochanek

Dixon

Alexander

Warner

Clark

RSB

Wisniewski

Graham

Jenkins

Marion

Safar

(2000) Isoflurane improves long-term neurologic outcome versus fentanyl after traumatic brain injury in rats. J Neurotrauma 17:1179–89

54.

Strupp

(1996) Stimulate: a GUI based fMRI analysis software package. Neuroimage 3:S607

55.

Thomale

Kroppenstedt

Beyer

Schaser

Unterberg

Stover

(2002) Temporal profile of cortical perfusion and microcirculation after controlled cortical impact injury in rats. J Neurotrauma 19:403–13

56.

Tseng

C-H

Biaggioni

Appalsmay

Robertson

(1988) Purinergic receptors in the brainstem mediate hypotension and bradycardia. Hypertension 11:191–7

57.

Van Wylen

Park

Rubio

Berne

(1989) The effect of local infusion of adenosine and adenosine analogues on local cerebral blood flow. J Cereb Blood Flow Metab 9:556–62

58.

Varma

Dixon

Jackson

Peters

Melick

Griffith

Vagni

Clark

RSB

Jenkins

Kochanek

(2002) Administration of adenosine receptor agonists or antagonists after controlled cortical impact in mice: effects on function and histopathology. Brain Res 951:191–201

59.

Warner

Takaoka

Ludwig

Pearlstein

Brinkhous

Dexter

(1996) Electroencephalo-graphic burst suppression is not required to elicit maximal neuroprotection from pentobarbital in a rat model of focal cerebral ischemia. Anesthesiology 84: 1475–84

60.

Williams

Detre

Leigh

Koretsky

(1992) Magnetic resonance imaging of perfusion using spin inversion of arterial water. Proc Natl Acad Sci USA 89: 212–6

61.

Yamakami

McIntosh

(1989) Effects of traumatic brain injury on regional cerebral blood flow in rats as measured with radiolabeled microspheres. J Cereb Blood Flow Metab 9:117–24

62.

Huang

Mariani

Wang

Moskowitz

Chen

(2004) Selective inactivation or reconstitution of adenosine A_2A receptors in bone marrow cells reveals their significant contribution to the development of ischemic brain injury. Nature Med 10:1–7

63.

Zhang

Silva

Williams

Koretsky

(1995) NMR measurement of perfusion using arterial spin labeling without saturation of macromolecular spins. Magn Reson Med 33:370–6

64.

Zhang

Williams

Koretsky

(1993) Measurement of rat brain perfusion by NMR using spin labeling of arterial water: in vivo determination of the degree of spin labeling. Magn Reson Med 29:416–21