G . Ben-Haim, W.M. Armstead Brain Research 884 2000 51 –58
53
identified. Pial arterial vessel diameter was determined assay determines cyclic nucleotide concentration for bind-
every minute for a 10-min exposure period after infusion ing to an antiserum that has a high specificity for the cyclic
onto the exposed parietal cortex of artificial CSF prior to nucleotide. The antibody-bound cyclic nucleotide is then
drug application and after infusion of artificial CSF reacted with an anti-rabbit second antibody bound to
containing a drug. Typically, 2–3 ml of CSF were flushed fluoromicrospheres. Labeled cyclic nucleotide bound to the
through the window over a 30-s period, and excess CSF primary rabbit antibody can then be measured by determin-
was allowed to run off through one of the needle ports. For ing the amount of light emitted by the fluoromicrospheres.
sample collection, 300 ml of the total cranial window All unknowns were assayed at two dilutions, with the
volume of 500 ml was collected by slowly infusing CSF lower limit of detection being 100 fmol ml. The con-
into one side of the window and allowing the CSF to drip centration of the unlabeled cyclic nucleotides is calculated
freely into a collection tube on the opposite side. from the standard curve via linear regression analysis.
Techniques for induction of total cerebral ischemia in the piglet have been well documented [17,19]. Briefly,
2.3. Statistical analysis total cerebral ischemia was accomplished by infusing
artificial CSF into a hollow bolt in the cranium to maintain Pial artery diameter, systemic arterial pressure, and
an intracranial pressure 15 mmHg greater than the numeri- cyclic nucleotide values were analyzed using repeated
cal mean of systolic and diastolic arterial blood [19]. measures of analysis or t-tests where appropriate. If the F
Intracranial pressure was monitored via a sidearm of the value was significant, the Fisher test was performed on all
cranial window. Blood flow in pial arterioles, viewed with data analyzed by repeated measures. An
a level of P,0.05 a microscope and video monitor, stopped completely on
was considered significant in all statistical tests. The n elevation of intracranial pressure and did not resume until
values reflect data for one vessel in each animal. Values are the pressure was lowered [19]. To prevent the arterial
represented as means6S.E.M. of absolute values or as pressure from rising inordinately Cushing response,
percentages of change from control values. venous blood was withdrawn as necessary to maintain
mean arterial pressure no greater than 100 mmHg. As the cerebral ischemic response subsided, the shed blood was
3. Results
returned to the animal. Cerebral ischemia was maintained for 20 min. In combined H1I1R animals, hypoxia pO 5
3.1. Influence of I1R and H1I1R on NOC oFQ pial
2
3563 mmHg was produced for 10 min by decreasing the artery reactivity and release of cAMP
inspired O via inhalation of N , which was followed by
2 2
28 26
the total ischemia protocol as described above. Topical NOC oFQ 10
, 10 M elicited reproducible
Three types of experiments were performed: 1 sham pial small artery 120–160 mm and arteriole 50–70 mm
control bolt inserted, but intracranial pressure not in- dilation over a 12-h period in sham control animals data
creased n514; 2 I1R n514; and 3 H1I1R n5 not shown. NOC oFQ-induced pial small artery dilation
28 26
was diminished within 1 h but returned to control value 14. NOC oFQ 10
, 10 M, Phoenix was topically
with 4 h of reperfusion in I1R animals Fig. 1A. Similar applied before intervention 0 time and at 1 and 4 h of
changes were observed in pial arterioles. Such NOC oFQ- reperfusion in I1R animals or at 1, 4, 8, and 12 h of
induced vasodilation was associated with elevated cortical reperfusion in H1I1R animals. Responses at the same
periarachnoid CSF cAMP concentration Fig. 2A. At 1 h intervals were obtained in sham control animals. In another
of reperfusion, this NOC oFQ-induced increase in CSF series of animals, pial artery responses to 8-Bromo cAMP,
cAMP was attenuated, but such biochemical responses Sp 8-Bromo cAMPs, PACAP 1–27, cromakalim, and
28 26
were restored to control pre-ischemia value within 4 h of NS1619 10
, 10 M all RBI except for cromakalim
reperfusion Fig. 2A. which was obtained from Smith Kline Beecham were
In contrast, NOC oFQ-induced vasodilation was re- obtained in sham control, I1R, and H1I1R animals at the
versed to pial artery vasoconstriction at both 1 and 4 h of 1-h time point only. Appropriate aliquots of the vehicle for
reperfusion after H1I1R Fig. 1B. At 8 h of reperfusion all agents 0.9 saline were added to CSF infused under
such vasoconstriction had returned to modest vasodilation, the window. This CSF vehicle had no effect on pial artery
whereas at 12 h of reperfusion NOC oFQ dilation was not diameter.
different from that observed before the insult Fig. 1B. Similar changes were observed in pial arterioles. NOC
2.2. Cyclic nucleotide analysis oFQ associated elevation in CSF cAMP was blocked and,
in fact, reversed to small stimulus-induced decreases in CSF sample collected after a 10-min exposure to an
CSF cAMP at 1 and 4 h of reperfusion in such animals intervention were analyzed for cAMP concentration using
Fig. 2B. At 8 h of reperfusion, NOC oFQ-induced scintillation proximity assay methods. Commercially avail-
increases in CSF cAMP once again, though attenuated able kits for cAMP Amersham were used. Briefly, this
compared to sham control. Finally, at 12 h of reperfusion,
54 G
28 26
Fig. 1. A Influence of NOC oFQ 10 , 10
M on pial small artery diameter before control and at 1 and 4 h post I1R. B Influence of NOC oFQ on pial small artery diameter before control and at 1, 4, 8, and 12 h post H1I1R, n57. P,0.05 compared to corresponding control.
NOC oFQ-induced elevated CSF cAMP was no different 8-Bromo cAMP and Sp 8-Bromo cAMPs was unchanged
than that observed in sham control animals Fig. 2B. Figs. 3 and 4. However, pial small artery dilation
Baseline CSF cAMP was unchanged by either I1R or induced by the adenylate cyclase activator PACAP and the
H1I1R in the absence of NOC oFQ administration Figs. K
channel activator cromakalim was attenuated within
ATP
2A,B. 1 h of reperfusion in I1R animals Figs. 5 and 6.
NS1619-induced pial small artery dilation, though, was 3.2. Influence of I1R and H1I1R on pial artery
unchanged in I1R animals Fig. 7. Similar observations vasodilation induced by
8-bromo cAMP, Sp 8-Bromo were made for agonist reactivities in pial arterioles.
cAMPs , PACAP, cromakalim, and NS1619
In contrast, pial small artery dilation induced by 8- Bromo cAMP and Sp 8-Bromo cAMPs was blocked in
Topical 8-Bromo cAMP, Sp 8-Bromo cAMPs, PACAP, H1I1R animals within 1 h of reperfusion Figs. 3 and 4.
28 26
cromakalim, and NS1619 10 , 10
M elicited re- PACAP- and cromakalim-induced pial small artery dilation
producible pial small artery and arteriole dilation data not was modestly, though nonsignificantly, attenuated to a
shown. After 1 h of reperfusion, in I1R animals, pial greater extent in H1I1R versus I1R animals at 1 h of
small artery dilation induced by the cAMP analogues reperfusion Figs. 5 and 6. Finally, NS1619-induced pial
28 26
Fig. 2. A Influence of NOC oFQ 10 , 10
M on CSF cAMP fmol ml in sham control animals and in I1R animals at 1 and 4 h of reperfusion. B Influence of NOC oFQ on CSF cAMP in sham control and in H1I1R at 1, 4, 8, and 12 h of reperfusion, n57. P,0.05 compared to corresponding 0
value; †P,0.05 compared to corresponding sham control value.
G . Ben-Haim, W.M. Armstead Brain Research 884 2000 51 –58
55
28 26
Fig. 5. Influence of PACAP 10 , 10
M on pial small artery and
28 26
Fig. 3. Influence of 8-Bromo cAMP 10 , 10
M on pial small artery arteriole diameter in sham control animals and in I1R and H1I1R
and arteriole diameter in sham control animals and in I1R and H1I1R animals at 1 h of reperfusion, n57. P,0.05 compared to corresponding
animals at 1 h of reperfusion, n57. P,0.05 compared to corresponding control value.
control value.
28 26
28 26
Fig. 4. Influence of Sp 8-Bromo cAMPs 10 , 10
M on pial small Fig. 6. Influence of cromakalim 10
, 10 M on pial small artery and
artery and arteriole diameter in sham control animals and in I1R and arteriole diameter in sham control animals and in I1R and H1I1R
H1I1R animals at 1 h of reperfusion, n57. P,0.05 compared to animals at 1 h of reperfusion, n57. P,0.05 compared to corresponding
corresponding control value. control value.
56 G
CSF cAMP concentration in sham control animals similar to previous observations [2], but new data in this study
show that such stimulated cAMP release was attenuated at 1 h but returned to sham control induced release within 4 h
of reperfusion. These data suggest that attenuated ability to elevate CSF cAMP contributes to impaired NOC oFQ-
induced pial artery dilation following ischemia reperfu- sion. This insult, however, did not alter baseline CSF
cAMP concentration.
In contrast, several differences in the observed parame- ters described above were noted when the effects of
ischemia reperfusion were compared to that of hypoxia ischemia reperfusion. For example, NOC oFQ-induced
vasodilation was reversed to pial artery vasoconstriction at both 1 and 4 h of reperfusion following hypoxia ischemia
reperfusion. At 8 h of reperfusion such vasoconstriction was returned to modest vasodilation, whereas at 12 h of
reperfusion NOC oFQ dilation was no different from that observed before the insult. Concomitantly, stimulated CSF
cAMP release by NOC oFQ was blocked, if not reversed to modest decreases in CSF cAMP concentration at 1 and
4 h post hypoxia ischemia reperfusion. At 8 h post reperfusion, NOC oFQ stimulated release of cAMP once
28 26
Fig. 7. Influence of NS1619 10 , 10
M on pial small artery and
again, but such release was less than that in sham control
arteriole diameter in sham control animals and in I1R and H1I1R
animals. NOC oFQ ability to stimulate cAMP release
animals at 1 h of reperfusion, n57. P,0.05 compared to corresponding
comparable to that observed in the sham control animal
control value.
was not fully restored until 12 h of reperfusion. Similar to the
ischemia reperfusion insult,
hypoxia ischemia re- perfusion did not alter baseline non-agonist-stimulated
small artery dilation was attenuated at 1 h of reperfusion in CSF cAMP concentration. Taken together, these data
H1I1R animals Fig. 7. Similar observations were made suggest that the more profound impairment of NOC oFQ-
for reactivities of agonists in pial arterioles. induced pial artery dilation following hypoxia ischemia
versus that observed following ischemia could relate to the 3.3. Blood chemistry
potentiated inability of this agonist to elevate CSF cAMP concentration.
Blood chemistry and mean arterial blood pressure values In order to more fully determine potential contributory
were obtained at the beginning and end of all experiments mechanisms for the observed decrement in NOC oFQ-
as well as during hypoxia. Hypoxia decreased pO to
induced pial vasodilation following hypoxia ischemia, the
2
3563 mmHg, whereas the pH, pCO , and mean arterial effects of such insults on the ability of cAMP analogues,
2
blood pressure values were unchanged. Values for pH, an adenylate cyclase activator, and activators of the K
ATP
pCO , pO , and mean arterial blood pressure were and K
channels to elicit vasodilation were explored.
2 2
ca
7.4560.02, 3663, 9064, and 7065 mmHg at the start of Results of these studies show that pial artery dilation
experiments versus 7.4460.02, 3763, 9165, and 6766 induced by the cAMP analogues, 8-Bromo cAMP and Sp
mmHg, respectively, at the end of experiments. There were 8-Bromo cAMPs, was unchanged by ischemia reperfusion,
no group differences in either blood pressure or blood consistent with the observations of others who showed that
chemistry values. the dilation to another analogue, dibutyryl cAMP, was
similarly unchanged in a piglet global cerebral ischemia model [8]. In contrast, results of the present study show
4. Discussion that hypoxia ischemia produces blunted pial dilation to
