- Email: [email protected]
Carotid body chemoreception in the absence and presence of CO2HCO3−
Brain Research, 568 (1991) 253-260 ~) 1991 Elsevier Science Publishers B.V. All fights reserved. 0006-8993/91/$03.50
253
BRES 17310
Carotid body chemoreception in the absence and presence of CO2-HCO 3 Rodrigo Iturriaga and Sukhamay Lahiri Department of Physiology, School of Medtcme, Umverstty of Pennsylvama, Philadelphia, PA 19104-6085 (U.S.A.) (Accepted 13 August 1991)
Key words: Anion exchanger; Chemoreceptor; Hypoxia; Hypercapnia
Carotid body (CB) chemosensory responses to natural and pharmacological stimuli were studied in vitro in the presence and nominal absence of COe-HCOf in the perfusion-superfusion media. The CBs obtained from cats (n = 10), anesthetized with sodium pentobarbltone, were simultaneously perfused and superfused with a modified Tyrode solution at 36.5 --- 0.5°C, equilibrated respectively with POe of 120 and < 20 Torr. The Tyrode, nominally free of COe-HC(Ta(HEPES-NaOH, pH 7.38, 310 mOsm), was used first. Subsequently the Tyrode containing HEPES-HCO-~, equilibrated with PCOe of 36.8 Torr (pH 7.38) was used. Chemosensorydischarges were recorded from the carotid sinus nerve. Both hypoxia (POe = 20--25 Torr) and ischemic hypoxia stimulated the discharge in the absence and presence of COe-HCO;. However, the presence of COe-HCO~ siguifieantly raised the baseline activity, augmented the speed, sensitivity and the maximal responses to both types of hypoxia. Hypercapnic perfusate (PCOe = 65 Torr at pH 7.17) produced a peak response equally promptly in the absence and presence of COe-HCO~ in the ongoing perfusate but generated a larger and more sustained response. Presence of CO2-HCO~ strongly potentiated the responses to cyanide (10-1°-10-7 tool) but less strikingly the responses to nicotine (10-11-10-s mol). Thus, the extraceUular COe-HCO~ siotmitlcantlyimproved the response to hypoxia but was not essential for 02 chemoreception. The underlying mechanisms of the effect of COe-HCO; is likely to be mediated by the C1--HCO3 anion exchanger in the pH regulation of glomus cells.
INTRODUCTION Both in vivo and in vitro studies showed that hypercapnia and application of acid to the carotid body (CB) stimulated the chemosensory discharge 6'14'15'17'2°-22. It has been assumed that the effect was due to a decrease in the cytosolic pH (pHi). There is, however, little in the literature showing the effects of pH~ changes at constant extracellular pH (pHo). The pH, regulation depends on the ion exchangers and ion channels in the plasma membrane of the cells 19'23. In recent years it has become apparent that these exchangers are not equally effective in all cells, and the pH, regulation depends on the net effects of the exchanger functions2'4'11'19'~3. The role of these exchangers in the pH~ regulation of the carotid body cells and the related chemosensory responses has no been adequately appreciated. For example, in their studies to distinguish between a possible effect of a stimulus directly on the sensory endings and an indirect one on an intermediary process (glomus cell) Eyzaguirre and Koyano 6 found that hypercapnia induced a significantly larger chemosensory discharge than without CO2-HCO~ in the superfusate medium. The issue of ion-exchanger function did not arise in their study. In the subsequent
studies on the CB in vitro several investigators used superfusate, free of CO2-HCO31'7'9'14'18, but did not address the questions regarding the role of COe-HCO 3 and ion exchangers in the expression of chemoreception. Our hypothesis was that nominal absence of CO2-HCO 3 in the extracellular medium would diminish the speed and magnitude of the chemosensory responses to the stimuli corresponding to anion exchanger function in the pH~ regulation. We tested the hypothesis by studying the chemosensory responses of the cat CB in vitro to hypoxia, perfusate flow interruption, hypercapnia, and to cyanide and nicotine injections in boluses in the presence and nominal absence of COe-HCO~ in the perfusate and superfusate media. We found that CO2-HCO~ in the external media improved the speed and magnitude of the chemosensory responses to hypoxia and other physiological and pharmacological stimuli, although it is not essential for the hypoxic response. The presence of CO 2HCO 3 in vitro improved and mimicked in vivo carotid body chemosensory responses.
MATERIALS AND METHODS Ten cats of either sex (2.2-3.5 kg) were anesthetized with so-
Correspondence: S. Lahiri, Department of Physiology,B-400, Richard Bldg., School of Me&cane, Umversity of Pennsylvania, Philadelphia, PA 19104-6085, U.S.A. Fax: (1) (215) 573-5851.
254 dlum pentobarbitone (35 mg/kg, 1.p.). The cats were tracheotomized and a femoral vein was cannulated for administration of additional anesthetic. The CB was perfused and superfused as was previously described 14. Briefly, the major arteries originating from the carotid bifurcation were ligated, the CB veins were left open and the carotid sinus nerve (CSN) was cut distally. The common carotid artery was cannulated and the bifurcatmn with the CB and the CSN was excised and placed in a plexiglas chamber. The CB was perfused at 80 Tort (by gravity) with a modified Tyrode solution, equihbrated with varying levels of PO 2 and PCO 2 and simultaneously superfused with the same Tyrode, equilibrated at PO E < 20 Torr. The chamber temperature was maintained at 36.5 -+ 0.5°C. The composition of the modified Tyrode was (in raM) Na + 154, K + 4 7, Ca 2÷ 2.2, Mg 2+ 1.1, C1- 110, glutamate 42.0, glucose 5.0, and HEPES 5.0 and dextran (5.0 g/l). Glutamate was found to be an avid substrate for chemosensory function during prolonged perfusmn 25. The preparataons (n = 13) were perfused first with Tyrode without CO2-HCO 3 for 30 rain (PO 2 = 124 -+ 2.9 Tort, pH = 7.38 -+ 0.01) and then with Tyrode containing CO2-HCO3 (HCO~ = 21.4 raM, replacing the same amount of NaC1, P O 2 -- 123.2 -+ 2.6 Torr. PCO 2 = 36.8 -+ 0.7 Torr, pH = 7.38 -+ 0.01). Dunng perfusion with Tyrode, free of CO2-HCO3, the superfusate was equilibrated with 100% N 2, and during perfusion wzth medmm containing CO2-HCO~, the superfusate was equilibrated with 5% CO 2 m 95% N 2. PO2, PCO2 and pH of the perfusion and superfusmn media were measured at 37.0"C with a blood gas analyzer. The chemosensory discharges were recorded from the whole desheathed CSN. The nerve was placed on bipolar platinum electrodes and lifted into paraffin oil. The effluent was aspzrated with a vacuum system to maintain the fluid level in the chamber, sufficaent to cover the CB. The electrical signals were amplified and recorded with an electrostatic recorder. For quantitatiou of the electroneurograms, we used standard procedures as have been used both in vivo and m vztro6,7'9'14,zs'2z. Action potentials of given amplitudes above the basehne noise were selected wzth an electronic amplitude discriminator and were counted electromcally by a frequency meter and registered on a printer. The raw signals were also stored in a VCR-digltai system for later analyses. The caroud chemosensory responses, during perfusion-superfuslon with media free of or containing CO2-HCO3, were tested as follows. Two types of maneuvers were used for the responses to hypoxia: interruption of the perfusate flow by cross-clamping the mfluent line for 1-3 min (while the superfusate flow remained constant) and perfusion with hypoxic Tyrode (PO2 = 20-30 Torr) for 1-8 rain. For the responses to CO 2, normoxic Tyrode equilibrated wath PCO 2 of 65.6 -+ 3.5 Torr (pH = 7.17 -+ 0.02) was perfused for 2 rain. To test for the dependence of the effects of the pharmacological agents on CO2-HCO ; , several doses of sodium cyanide (10-m-10 -7 mole) and mcotme bitartrate (10-11-10-8 mole) were given in a bolus of 0.2 ml into the perfusate line. These agents were dissolved m Tyrode free of CO2-HCO~. Results are expressed as mean -+ S.E.M. Statistical differences between the paired samples were assessed by the Wilcoxon's signed rank test. To compare the dose-responses in the presence and absence of C O 2 - H C O 3 the chemosensory activity was expressed as percentage of the maximal response evoked by the interruption of the perfusate flow in the presence of CO2-HCO~. The data from cyanide and nicotine dose-response curves were fitted to symmetrical sigmoidal functions5. This method provides a basis for poolmg data from separate experiments, and allows the test of the characteristics which are shared by various curves. The data points were adjusted according to the following logistic expression R = Rm. x + (baS-Rma~)/(1 +
(DIEDso)s)
A
soo
60 s
CSN activity (imp/s)
Impulses
B
500
CSN activity (imp/s) j 0
Impulses C-90-93
Fig. 1. Effects of perfusate flow interruption on carotid chemosensory activity in the presence and nominal absence of CO2-HCO~. A: perfusion-superfusion wzth Tyrode nommaUy free of CO2HCO~. B: perfuszon-superfusion with Tyrode with CO2-HCO~ (PCO 2 = 38 Torr, 21.4 mM of HCO~) pH was 7.39 and PO2 120 Torr in A and B. Bars indicate duration of perfusate flow interruption.
curves were fitted through a computer program based on a simplex algorithm 16 as used previously8. The goodness of each fit was tested using an ANOVA and by the coefficient of correlation calculated by dividing the variance of theoretical values by that of the expertmental values s. The EDso values from different experiments were averaged as geometric mean.
RESULTS
Carotid chemoreceptor responses to interruption of perfusate flow in the presence and absence of C02-HCO ~ Fig. 1 s h o w s t h e e f f e c t s o f i n t e r r u p t i o n o f flow in t h e CB perfused with normoxic Tyrode nominally free of CO2-HCO3
(Fig. 1 A ) , a n d s u b s e q u e n t l y w i t h T y r o d e
c o n t a i n i n g C O 2 - H C O ~ (Fig. 1B). P e r f u s i o n w i t h T y r o d e c o n t a i n i n g C O 2 - H C O ~ a t t h e s a m e p H o o f 7.39 a n d P O 2 o f 120 T o r r (Fig. 1B) r a i s e d t h e b a s e l i n e c h e m o s e n s o r y activity, s h o r t e n e d t h e l a t e n c y o f r e s p o n s e , i n c r e a s e d t h e r a t e o f rise o f a c t i v i t y t o a g r e a t e r m a x i m a l l e v e l d u r i n g flow i n t e r r u p t i o n . T h e p e a k r e s p o n s e w a s f o l l o w e d b y a
where; R = chemosensory response; Rraax = mammal chemosensory response; bas = baseline chemosensory discharge; D = arithmetzc dose; ED5o = median effective dose; S = slope factor that deternunes the steepness of each curve. R, Rraax and bas were expressed as percentage of stop flow response in C O 2 - H C O 3 The
g r a d u a l d e c l i n e o f t h e activity. R e s t o r a t i o n o f flow dim i n i s h e d t h e activity r a p i d l y b u t t h e r a t e o f d e c l i n e was slower with CO2-HCO~
than without. The instant of
flow r e s t o r a t i o n i n b o t h Fig. 1 A a n d B w a s first a c c o m -
255 500-
T
400-
t. ~"
O ~
Q-
O
¢~++~.
-+m,~..
.
300
~v
E_
200
+~
I O0
200100-'
o
g, 0 !
0
-30
0
30
60
90
120
Time (s)
150
180
210
-30
240
C-90-95
Fig. 2. Carotid chemosensory response to hypoxia ( P O 2 = 25 Torr) during perfusion with Tyrode free of CO2-HCO~ (o) and then during perfusion with Tyrode equilibrated with CO2-HCO~ (PCO2 = 38 Torr, 21.4 ram of HCO 3, O). pH was 7.38 and PO 2 120 Torr. Arrows indicate onset and withdrawal of hypoxic perfnsion.
panied by a rise in the discharge rate which could have been due to transient baroreceptor activity. But that was unlikely because a similar p h e n o m e n o n occurred during the transition of the chemical stimulus (Fig. 2). Results obtained in 10 preparations are summarized in Table I. Clearly, C O 2 - H C O 3 improved the chemosensory response to flow interruption. The absolute maximal activity was greater but the net change (maximal minus basal) was the same with and without CO2-HCO3.
Carotid chemosensory responses to hypoxia in the presence or absence o f C 0 2 - H C 0 3 Fig. 2 compares the carotid chemosensory responses to changes of perfusate PO2 from 120 Torr to 25 Torr. In the presence of C O 2 - H C O 3 the baseline activity was greater and the same hypoxic stimulus sharply increased the chemosensory discharge after a shorter latency to a higher maximal level which showed a slight adaptation. Table II summarizes the characteristics of the responses with and without effects of C O 2 - H C O 3. O n the average, presence of CO2-HCO~ reduced the latency of the response to hypoxia by 95 --- 33.0 s (P < 0.05). Thus, the
0
I
I
I
30
60
90
Time (s)
--I
120
150
C-90-95
Fig. 3. Carotid chemosensory response to hypereapnia ( P C O 2 -- 68 Torr, pH 7.20) during perfusion with q~yrode free of CO2-HCO3 (o) and then during perfusion with Tyrode equilibrated with CO2HCO 3 (PCO2 = 38 Torr, 21.4 mM of HCO 3, Q). pH was 7.38 and PO2 120 Torr. Arrows indicate onset and withdrawal of hypercapnic perfusion.
presence of CO2-HCO 3 in the perfusate-superfusate at the same p H o strikingly augmented the speed of the responses. The absolute maximal response to a given hypoxia also increased but this was due in part to an increased baseline activity and to further stimulus interaction between pH, and hypoxia in the presence of CO2HCO~.
Carotid chemosensory response to hypercapnia in the presence or absence o f pre-perfusion with C02-HCO ~ Fig. 3 shows the chemosensory responses to hypercapnic-acidosis which was applied to the CB pre-perfused with and without CO2-HCO~. In both cases hypercapnic perfusion (PCO2 = 68 Torr, p H = 7.20, P O 2 = 118 Torr) augmented the chemosensory activity equally rapidly in a few seconds. In the absence of CO2-HCO~ the perfusate PCO2 rose from about 0.2 to 68 Tort and the absolute peak response was less but adapted more rapidly to a lower steady state level. During perfusion with Tyrode containing CO2-HCO ~, a rise of P C O 2 from 38-68 Torr caused a greater absolute peak response which adapted slowly. However, the initial baseline activity
TABLE I
Effects of C02-HCO ~ on the chemosensory responses to perfusate flow interruption Values are the mean -+ S.E.M. of 10 preparations. The half-time on and off values refer to flow interruption and flow restoration. During perfnsiou with Tyrode free of CO2-HCO~, PO2 was 121.3 -+ 2.5 Torr and pH 7.39 -4- 0.01. During perfusion with Tyrode containing C O 2 - H C O 3 , PCO2 was 37.1 -+ 0.8 Tort, pH 7.39 -+ 0.01 and PO 2 123.2 -+ 2.5 Tort.
CondiUon CO2-HCO3
CO2-HCO3
free
Baseline (imp/s)
Maximal response (imp~s)
Net response (maxtmal-basal)
Half-time on (s)
Half-time off (s)
7.3 +- 1.7 53.9 -+ 9.1"
344.0 _+ 23.8 391.4 +- 25.9*
336.7 -+ 23.3 344.3 -+ 22.4
53.2 -+ 7.8 12.4 -+ 2.2*
6.2 -+ 0.3 12.5 -+ 1.7"
Significant differences from the control were assessed by the Wilcoxon Test (*P < 0.05).
256 TABLE II
Effects of C02-HCO 3 on the chemosensory response to hypoxtc perfuston Values are the mean - S.E.M. of 6 preparations. Half-time refers to the half response from the onset of hypoxia. Dunng perfusion with "l]¢rode free of CO2-HCO ~, PO 2 was changed from 120.5 --- 3.2 to 25 6 -+ 1,7 Torr and pH was kept constant at 7.38 -+ 0.01. Dunng
perfusion with Tyrode containing CO2-HCO~, PO2 was changed from 124.5 -+ 2.5 to 22.8 -+ 1.6 Torr, PCO2 was kept constant at 34.2 -+ 1.4 and pH at 7.39 -+ 0.01. Condinon
Baseline (imp/s)
Maxtmal response (imp/s)
Net response (maximal-baseline)
Half-ume (s)
CO2-HCO ~ free CO2-HCO 3
6.9 -+ 1.4 51.0 -+ 9.8*
247 9 -+ 45.8 380 0 -+ 36.1"
232.8 -+ 43.1 328.2 -+ 34.9*
117.1 -+ 38.8 22.1 -+ 6.9*
Significant differences from the control were assessed by the Wilcoxon Test (*P < 0.05)
was higher and the net peak response was only slightly greater. Withdrawal of hypercapnia was followed by a greater undershoot in the presence of CO2-HCO 3. Table III summarizes the chemosensory responses to hypercapnic-acidosis. Pre-perfusion of the CB with CO 2HCO3 did not modify the speed of response and did not significantly increase the net peak response (maximal minus basal). Thus the initial responses to hypercapnic acidosis were apparently correlated with the applied stimulus strength but the net response during adaptation was greater with CO2-HCO3 pre-perfusion.
the latency of the physiologic response. Fig. 4 shows the effects of several doses of cyanide on the chemosensory activity in the absence (Fig. 4A) and presence of CO 2HCO~ (Fig. 4B). As expected, the magnitude and duration of carotid chemosensory responses increased with the dose in both cases. However, the same dose evoked a larger and a more prolonged increase of chemosensory activity in the presence of CO2-HCO 3 which also raised the baseline activity. Also, presence of CO2-HCO~ reduced the latency of responses. Fig. 5 shows the fitting of the normalized chemosensory activities elicited by cyanide to sigmoidal functions (n = 5). Each data point is expressed as percent of the maximal response elicited by stop flow of the perfusate with CO2-HCO3 (see methods). The parameters defining the dose-response curves show that the presence of CO2-HCO 3 increased the maximal activity from 42.2 12.5% to a 96.0 - 9.2% (P < 0.01). The slope factors
Effects of carotid chemosensory responses to nicotine and cyanide in the presence and absence of C 0 2 - H C O 3 Injections of nicotine or cyanide increased the chemosensory discharge within a few seconds. This delay was partly due to the transit time of the perfusate flow from the point of injection to the CB and partly to
A CSN activity (imp/s)
Impulses
4olr =, ,
.
, = ......
60 s
J .....
, _it . . . . . .
_~
JUt _-S
I ill
~i
,
ii~111
I I
I ~_~,
Illll I J
NaCN (nmol) 6
4 0
6 0
20
40
60
200
400
600
B CSN actwity (imp/s)
Impulses NaCN (nmol) C-90-105 r
Fig. 4. Effects of CO2-HCO 3 on caroUd chemosensory response to several NaCN doses. A: responses dunng normoxic perfuston with Tyrode free of CO2-HCO 3. B: responses during perfuslon with Tyrode equilibrated with CO2-HCO3 (PCO 2 at 36.6 Tort, 21.4 mM of HCO3). pH was 7.39 and PO 2 122 Torr. Note that the doses in B were smaller than those m A.
257 with and without CO2-HCO~ respectively, were 1.6 --. 0.3 and 2.1 --- 0.3 (P > 0.05). The EDso decreased from 2.9 - 0.9 l0 s mol to 7.6 --- 1.5 10-9 tool (P < 0.01). Fig. 6 shows the carotid chemosensory response to several doses of nicotine. Presence of C O 2 - H C O 3 raised the baseline activity, did not significantly modify the peak responses to nicotine ( > 4 - 6 10-9 mol) but prolonged its effect (Fig. 6B). Fig. 7 summarizes the doseresponses to nicotine (n = 5). The maximal peak activities were not different in the absence or presence of CO2-HCO~ (respectively 88.7 - 11.6% vs 85.1 - 12.2%, P > 0.05) and the slope factors were not different with and without CO2-HCO 3 (respectively 2.2 - 0.9 and 1.7 --- 0.3, P > 0.05). H o w e v e r the ED5os decreased from 9.3 --- 1.3 10-~° mol to 6.0 --- 1.6 10-~° mol (P < 0.05) in the presence of C O 2 - H C O 3. This difference was mostly due to an increase in the baseline activity which raised the lower part of the curve upward.
I
0
f ]
• C?2-HC03-
"~ ~ u 0
80.
o
60-
~
(..)
g ~~ m 0
IZ
0 0-
,
+"
1
,0.1
8~ ~
_
0
-9
-8 -7 Log NoCN dose (moo
-6
Fig. 5. Dose-response curves for chemosensory activity elicited by NaCN in the presence or absence of CO2-HCO~. The coefficient of correlatton in the absence and presence of CO2-HCO~, respectively, were 0.997 and 0.995. Chemosensory activity was expressed as percentage of maximal response produced by stop flow during COz-HCOf perfusate in each preparation (n = 5). o, dose response curve with normoxic perfusion with Tyrode free of CO2HCO~; O, dose response curve during perfusion with Tyrode equilibrated with CO2-HCO~ (PCO2 at 36 Torr, 21.4 mM of HCO~). pH was 7.39 and PO e 120 Tort. Note that the doses in B were smaller than those in A. Maximal reactivity was significantly higher (P < 0.01) with CO2-HCO; perfusate.
DISCUSSION Cat carotid body chemosensory fibers responded to the physiological and pharmacological stimuli in the nominal absence of C O 2 - H C O ; in the external medium but the presence of C O 2 - H C O ; significantly augmented these responses. The baseline chemosensory discharge increased in the presence of C O 2 - H C O ~, suggesting a stronger receptor potential. Since the responses of sensory organ receptors to stimuli depend on their background discharge ~3 it is reasonable that the presence of CO2-HCO3 augmented the sensory responses to expect
A
o HEPES
100-
the applied stimuli. However, the improvement of the responses were not uniform. The results showed a large improvement of the chemosensory responses to hypoxia and cyanide but only a small one to nicotine. Thus the increased baseline activity alone does not explain the differences. Also the fact that the CB sensory response to C O 2 strikingly differs from those observed in other
60 s
250
activityi
CSN (Imp/s)
I."1
r
%
4
::q
ot Impulses
Nicotine (nmol)
B CSN activity (,mp/s)
0.~6
•2
I
0.=4
v ,,....
~,+-~
,J '~M.
.
0".6
1".0
4".0
6".0
,
, 'e.+
•
~-,~,
• ...,~,y.
"+~.~-'~41.~,,,'
Impulses Nicotine (nmol)
0.06
0:2
0:4
0".6
2.0
~0
~.0
C-~l>-,osr
Fig. 6. Effects of CO2 -HCO 3 on carotid chemosensory response to several nicoUne doses. A: responses during normoxic perfusion with Tyrode free of CO2-HCO~. B: response during perfusion with Tyrode equilibrated with CO2-HCO; (PCO2 at 36.6 Torr, 21.4 mM of HCO3). pH was 7.39 and PO2 122 Torr.
258 i" 100 -~ o4 u oO (j
0
HEPES
• =co -.co3-
80
.
fects of C O 2 - H C O 3 occurred even though the p H o was the same with and without C O 2 - H C O 3 ( p H 7.38). Accordingly the effect of C O 2 - H C O 3 was m e d i a t e d intracellularly, presumably by intracellular acidification of the glomus cells. I n d e e d , glomus cell acidification was found with m e d i u m containing C O 2 - H C O 3 at the same external p H 4'27. Exposure to C O 2 - H C O 3 increased the cellular acidity not just because of carbonic acid formation m e d i a t e d by carbonic anhydrase 15 but also because of the H C O 3 extrusion through the C I - - H C O ~ exchanger mechanisms, as found in glomus cells4 and some o t h e r cells2't1'19'23. The observation that b l o c k a d e of C1--HCO 3
~c 0 -
60
o
40
~ ° o -,,-, (.) m
20. 0 -11
-10
-9
-8
Log N,cobne dose (tool)
Fig. 7 Dose-response curves for chemosensory activity ehcited by nicotine in the presence or absence of CO2-HCO~. The coefficient of correlation in the absence or presence of CO2-HCO ~, respectively, were 0.996 and 0.995. Chemosensory activity was expressed as percentage of maximal response produced by stop flow dunng CO2-HCO~ perfusion m each preparation (n = 5). o, dose response curve during normoxic perfusion with Tyrode free of CO 2HCO~; O, dose response curve during perfusion with Tyrode equilibrated with COz-HCO~ (PCO 2 at 36 Torr, 21.4 mM of HCO3). pH was 7.39 and PO2 120 Torr, The maximal responses were the same (P > 0.05) during perfusion with or without CO2-HCO3.
sensory organs is relevant: C O 2 suppressed the generator potentials in the sensory endings of the cat muscle spindle 1° and diminished the amplitude of the receptor potential of the barnacle photoreceptor3. Accordingly it is likely that the interaction of CO2 with hypoxia and cyanide involving 0 2 chemotransduction was not the same with nicotine, although the initial generator potential was similar. The current consensus is that the glomus cell which contains both carbonic anhydrase Is and neurotransmitter 7 is the initial site of chemoreception. The glomus cells are therefore equipped to promptly respond to CO2 and to release neurotransmitters which in turn could generate action potentials in the chemosensory nerve terminals 7. There is an interesting and relevant parallelism between glomus cell acidification and chemosensory discharge. The results showed that the augmenting el-
exchanger by eliminating extracellular H C O 3 or C1m a d e the rat (neonatal and adult) glomus cells alkaline even in the presence of CO2-HCO3-4 support that acidification by the anion-exchanger mechanism was p r e d o m inant in these cells. Chloride channels in the glomus cells with large conductance and selectivity for HCO326 presumably also participated in the process because we recently found that the C1- channel b l o c k e r 9-anthracenecarboxylic acid decreased the baseline chemosensory activity (unpublished results). Hypercapnic acidic stimuli. T h e unsteady-state responses to hypercapnic acidosis were not correlated with the final strength of the stimulus. T h e same final stimulus elicited a greater absolute response when the P C O 2 was raised from 35 to 65 Tort than from 0 to 65 Torr. H o w e v e r , for the same p H o stimulus the response was similar, and was s u p e r i m p o s e d on the existing baseline chemosensory activity. The baseline chemosensory activity was greater in the presence of CO2-HCO~, presumably because of a m o r e acidic pH, relative to that in the absence of CO2-HCO3-4. The responses to hypercapnic acidosis are due in part to acid pHo and to intracellular acid, generated by C O 2 hydration. The net effect (maximal minus basal), but not the absolute effect, was correlated with the final stimulus regardless of the initial level of P C O 2. O n the other hand, the net increase in chemosensory activity was correlated with A p H o (PHo
TABLE III Effects of C02-HCO 3 on the chemosensory response to hypercapmc perfuston
Values are the mean -+ S.E.M of 4 preparations. Half-ume refers to the half response from the onset of hypereapmc periuslon. Dunng perfusion with Tyrode free of CO2-HCOf, PO 2 123.0 - 2.4 Torr and pH 7.38 ± 0.01 and dunng perfusion with q~yrode containing CO2HCO~, PCO2 was 36.6 ± 2.8, pH was 7.39 -+ 0.01 and PO 2 123.0 - 1.4 Torr. During hypercapnic-acidic perfusion PCO2 was 65.6 ± 3.5 Torr, pH 7.17 ± 0.02 and PO 2 118.8 - 4.2 Torr. Chemosensory response was measured at its maxtmal (Peak) response and then at 1 (Res at 1 mm) and 2 min (Res at 2 mm) of hypereapnic perfumon. Condition
Baseline (imp/s)
Peak response Omp/s)
Net response (Peak-basal)
Half-time (s)
Res at 1 min (trap~s)
Res at 2 rain (trap~s)
COz-HCO ~ free CO2-HCO ~
8.3 - 1.8 67.8 ± 16 3*
216 4 _ 39.2 314.4 ± 53.7*
207.1 - 38.4 247.1 - 37.1
7.0 ± 1.9 10.0 ± 4.7
79.8 ± 18.3 232.5 ± 14.5"
112.4 +- 17.2 239.9 ± 16.5"
Slgmficant differences from the control were assessed by the Wdcoxon Test (*P < 0.05)
259 before and during hypercapnia) but not with APCO 2. The initial and late responses to PCO2 were strikingly less when the initial PCO 2 was near zero and the cells alkaline 4. The initial pH~ difference with and without CO2-HCO 3 presumably made the difference in the chemosensory discharge during the unsteady-state. Ultimately in the steady state, the pI-I~ and chemosensory responses would be the same. Metabolic acidosis at a constant arterial PCO 2 is known to stimulate carotid chemosensory discharges in vivo21 and in vitro 6. We also found in vitro that perfusion of acid Tyrode (pH 6.8) stimulated carotid chemosensory discharge in the absence of CO2-HCO ~ (unpublished observations). These observations are consistent with the reports that the glomus cells pI-I~ correlates well with pHo 4'27. Also the release of carotid body dopamine and chemosensory discharge correlate during acidic stimulation 22. Despite the parallelism between pH~ changes and chemosensory discharge the mechanisms relating cellular proton concentrations and signal transduction are unclear. H y p o x i c stimulus. T h e mechanisms of 0 2 and CO2 stimulus interaction were not studied here. However, if neurotransmitters are released by these stimuli, as seems to be the case 7, and the chemosensory discharges are any indication of neurotransmitter release, it is possible that the presence of CO2-HCO ~ in the cellular environment augmented the release of neurotransmitter during hypoxia. On the other hand, the studies which used medium free of CO2-HCO ~ to superfuse the carotid body might have underestimated the response to hypoxia and flow interruption because of a possible alkaline pI-I~ of the glomus cells4. Also a brief hypoxic stimulus without CO2-HCO ~ may not elicit a significant chemosensory response and could give an impression that the hypoxic response is lost without CO2-HCO ~ in the medium as Shirihata and Fitzgerald were led to conclude 24. Nonetheless hypoxia has been found to diminish the membrane K + conductance TM and to increase the intracellu-
REFERENCES 1 Biscoe, T.J. and Duchen, M.R., Electrophysiological responses of dissociated type I cells of the rabbit carotid body to cyanide, J. Physiol., 405 (1988) 210-232. 2 Boron, W.F., Boyarski, G. and Ganz, M., Regulation of intracellular pH in mesangfialcells, Ann. N.Y. Acad. Sci., 574 (1989) 321-332. 3 Brown, H.M. and Meech, R.W., Light induced changes of internal pH~ in a barnacle photoreceptor and the effect of internal pH on the receptor potential, J. Physiol., 297 (1979) 73-93. 4 Buckler, K.J., Vaughan-Jones, R.D., Peers, C. and Nye, P.C.G., Intracellular pH and its regulation in the isolated type I carotid body cells of the neonatal rat, J. Phystol., 436 (1991) 107-129. 5 De Lean, A., Munson, P.J. and Rodbard, D., Simultaneous
lar [Ca2+] 1 of the glomus cells in CO2-HCO 3 free medium. These results are consistent with our findings that the hypoxic effect ultimately does not require the participation of external CO2-HCO~. The present findings are also in conformity with the previously published reports of hypoxia stimulated carotid chemosensory discharge and catecholamine release in the nominal absence of CO2-HCO; in the superfusate medium 6'7'9'14. Pharmacological stimuli. T h e augmented response to cyanide, a metabolic inhibitor, in the presence of CO 2H C O 3 can also be explained by the cellular [H +] increase. The in vitro response to cyanide is analogous to the observed CO2-cyanide stimulus interaction in the cat chemosensory responses in vivo2°. The mechanism of this interactive response is not known but the phenomenon resembles CO2-hypoxia interaction. The less interactive response to nicotine, a receptor agonist, also resembles that in in vivo2°. Hypoxia and cyanide are knowt3 to release putative neurotransmitters, acetylcholine and dopamine 7'9A2, whereas nicotine has been reported to elicit a nearly selective release of norepinephrine 12 and also is expected to act on the nicotinic receptors on the nerve endings. These basic differences between the effects of nicotine and cyanide presumably explain their different interaction with CO2-HCOf. In summary, the results showed that 0 2 chemoreception concurred both in the absence and presence of CO2H C O ; in the extracellular medium but CO2-HCO ~ raised the baseline activity and improved the responses to physiological and pharmacological stimuli. The effect of CO 2H C O 3 could be attributed to intracellular acidification through the CI--HCO~ exchanger mechanisms. However, the basic hypoxic mechanisms of chemotransduction is not mediated by the acidification produced by CO 2 and the acid loading exchanger mechanisms.
Acknowledgements. We would like to thank A. Mokashi for his assistance. Supported in part by NIH Grants HL-43413 and HL19737.
analysis of famdies of sigmoidal curves: application to bioassay, radioligand assay, and physiological dose-response curves, Am. J. Physiol., 253 (1978) E97-E102. 6 Eyzaguirre, C. and Koyano, H., Effects of hypoxia, hypercapnia, and pH on the chemoreceptor activity in vitro, J. Physiol., 178 (1965) 385-409. 7 Eyzaguirre, C., Fitzgerald, R.S., Lahiri, S. and Zapata, P., Arterial Chemoreceptors. In J.T. Shepherd and F.M. Abboud (Eds.), Handbook of Physiology. Sect. 2: The Cardiovascular System, Vol. Ili, Peripheral Circulation and Organ Flow, American Physiological Society, Williams and Wilkins, Baltimore, 1983, pp. 557-621. 8 Eugenin, J., Larrain, C. and Zapata, P., Functional recovery of the ventilatory chemoreflexes after partial chronic denervation of the nucleus tractus solitarius, Brain Research, 523 (1990) 263-272.
260 9 Fidone, S.J., Gonzalez, C. and Yosluzaki, K., Effects of low oxygen on the release of dopamine from the rabbit carotid body in vitro, J. Physiol., 333 (1982) 93-110. 10 Fukami, Y., The effect on NI-Ia and CO2 on the sensory ending of mammalian muscle spindles: intracellular pH as a possible mechamsm, Brain Research, 463 (1988) 140-143. 11 Gailard, D.S. and Dupont, J.L., Ionic control of intracellular pH In the rat cerebellum Purkinje cells maintained in culture, J. Physiol., 425 (1990) 71-83. 12 Gomez-Nifio, A., Dinger, B., Gonzalez, C. and Ftdone, S.J., Differential stunulus coupling to dopamine and norepinephrine stores in rabbit carotid body type I cells, Brain Research, 525 (1990) 160-164. 13 Hunt, C.C. and Kuffler, S.W., Further study of efferent smallnerve fibres to mammalian muscle spindles. Multiple spindle innervation and activity during contraction, J. Physiol., 113 (1953) 283-297. 14 Iturriaga, R., Rumsey, W.L., Mokaslu, A., Spergel, D., Wilson, D.E and Lahiri, S., In vitro perfused-superfused cat carotid body for physiological and pharmacological studies, J. Appl. Physiol., 70 (1991) 1393-1400. 15 Iturriaga, R., Lahiri, S. and Mokashi, A., Carbonic anhydrase and chemoreception in the cat carotid body, Am. J. Physiol., 261 (1991) in press. 16 Johnston, A., SIMP: a computer program in BASIC for nonlinear curve fitting, J. Pharmacol. Meth. Progr. Biomed., 28 (1989) 37-42. 17 Lahiri, S. and DeLaney, R.G., Stimulus interaction in the responses of carotid body chemoreceptor single afferent fibers, Respir. Physiol., 24 (1975) 249-266.
18 L6pez-L6pez, J.R., Gonzalez, C,, Urefia, J. and L6pez-Barneo, J., Low PO 2 selectively inhibits K channel acttvlty in chemoreceptor cells of the mammalian carotid body, J Gen. Physiol., 93 (1989) 1001-1015. 19 Madshus, I.H., Regulation of mtracellular pH in eukariotic cells, Biochem. J., 250 (1988) 1-8. 20 Mulligan, E. and Lahln, S., Dependence of carotid chemoreceptor stimulation by metabolic agents on PaO2 and PaCO 2, J Appl. Physiol., 50 (1981) 884-891. 21 Pokorski, M. and Lahirl, S., Aortic and carotid chemoreceptor responses to metabolic acidosis in the cat, Am. J. Physiol., 244 (1983) R652-R658. 22 Riguald, R., L6pez-L6pez, J.R. and Gonzalez, C., Release of dopamine and chemoreceptor discharge induced by low pH and high PCO 2 stimulation of the cat carotid body, J. Physiol., 433 (1991) 519-531 23 Roos, A. and Boron, W.E, Intracellular pH, Physiol. Rev, 61 (1981) 296-434. 24 Shlrahata, M. and Fitzgerald, R.S., The presence of CO2/I-ICOf is essential for hypoxic chemotransduction in the m vivo perfused carotid body, Brain Research, 545 (1991) 297-300. 25 Spergel, D., Metabohc substrate dependence of carotid chemosensory discharge, FASEB J., 5 (1991) Al120. 26 Stea, A. and Nurse, C., Chloride channels in culture glomus cells of the rat carotid body, Am J Physiol., 257 (1989) C174C181. 27 Wllding, T.J., Cheng, B and Roos, A., The relationship between extracellular pH (pHo) and intracellular pH (pHi) in adult rat carotid body glomus cells, Btophys. J., 59 (1991) 184a.
253
BRES 17310
Carotid body chemoreception in the absence and presence of CO2-HCO 3 Rodrigo Iturriaga and Sukhamay Lahiri Department of Physiology, School of Medtcme, Umverstty of Pennsylvama, Philadelphia, PA 19104-6085 (U.S.A.) (Accepted 13 August 1991)
Key words: Anion exchanger; Chemoreceptor; Hypoxia; Hypercapnia
Carotid body (CB) chemosensory responses to natural and pharmacological stimuli were studied in vitro in the presence and nominal absence of COe-HCOf in the perfusion-superfusion media. The CBs obtained from cats (n = 10), anesthetized with sodium pentobarbltone, were simultaneously perfused and superfused with a modified Tyrode solution at 36.5 --- 0.5°C, equilibrated respectively with POe of 120 and < 20 Torr. The Tyrode, nominally free of COe-HC(Ta(HEPES-NaOH, pH 7.38, 310 mOsm), was used first. Subsequently the Tyrode containing HEPES-HCO-~, equilibrated with PCOe of 36.8 Torr (pH 7.38) was used. Chemosensorydischarges were recorded from the carotid sinus nerve. Both hypoxia (POe = 20--25 Torr) and ischemic hypoxia stimulated the discharge in the absence and presence of COe-HCO;. However, the presence of COe-HCO~ siguifieantly raised the baseline activity, augmented the speed, sensitivity and the maximal responses to both types of hypoxia. Hypercapnic perfusate (PCOe = 65 Torr at pH 7.17) produced a peak response equally promptly in the absence and presence of COe-HCO~ in the ongoing perfusate but generated a larger and more sustained response. Presence of CO2-HCO~ strongly potentiated the responses to cyanide (10-1°-10-7 tool) but less strikingly the responses to nicotine (10-11-10-s mol). Thus, the extraceUular COe-HCO~ siotmitlcantlyimproved the response to hypoxia but was not essential for 02 chemoreception. The underlying mechanisms of the effect of COe-HCO; is likely to be mediated by the C1--HCO3 anion exchanger in the pH regulation of glomus cells.
INTRODUCTION Both in vivo and in vitro studies showed that hypercapnia and application of acid to the carotid body (CB) stimulated the chemosensory discharge 6'14'15'17'2°-22. It has been assumed that the effect was due to a decrease in the cytosolic pH (pHi). There is, however, little in the literature showing the effects of pH~ changes at constant extracellular pH (pHo). The pH, regulation depends on the ion exchangers and ion channels in the plasma membrane of the cells 19'23. In recent years it has become apparent that these exchangers are not equally effective in all cells, and the pH, regulation depends on the net effects of the exchanger functions2'4'11'19'~3. The role of these exchangers in the pH~ regulation of the carotid body cells and the related chemosensory responses has no been adequately appreciated. For example, in their studies to distinguish between a possible effect of a stimulus directly on the sensory endings and an indirect one on an intermediary process (glomus cell) Eyzaguirre and Koyano 6 found that hypercapnia induced a significantly larger chemosensory discharge than without CO2-HCO~ in the superfusate medium. The issue of ion-exchanger function did not arise in their study. In the subsequent
studies on the CB in vitro several investigators used superfusate, free of CO2-HCO31'7'9'14'18, but did not address the questions regarding the role of COe-HCO 3 and ion exchangers in the expression of chemoreception. Our hypothesis was that nominal absence of CO2-HCO 3 in the extracellular medium would diminish the speed and magnitude of the chemosensory responses to the stimuli corresponding to anion exchanger function in the pH~ regulation. We tested the hypothesis by studying the chemosensory responses of the cat CB in vitro to hypoxia, perfusate flow interruption, hypercapnia, and to cyanide and nicotine injections in boluses in the presence and nominal absence of COe-HCO~ in the perfusate and superfusate media. We found that CO2-HCO~ in the external media improved the speed and magnitude of the chemosensory responses to hypoxia and other physiological and pharmacological stimuli, although it is not essential for the hypoxic response. The presence of CO 2HCO 3 in vitro improved and mimicked in vivo carotid body chemosensory responses.
MATERIALS AND METHODS Ten cats of either sex (2.2-3.5 kg) were anesthetized with so-
Correspondence: S. Lahiri, Department of Physiology,B-400, Richard Bldg., School of Me&cane, Umversity of Pennsylvania, Philadelphia, PA 19104-6085, U.S.A. Fax: (1) (215) 573-5851.
254 dlum pentobarbitone (35 mg/kg, 1.p.). The cats were tracheotomized and a femoral vein was cannulated for administration of additional anesthetic. The CB was perfused and superfused as was previously described 14. Briefly, the major arteries originating from the carotid bifurcation were ligated, the CB veins were left open and the carotid sinus nerve (CSN) was cut distally. The common carotid artery was cannulated and the bifurcatmn with the CB and the CSN was excised and placed in a plexiglas chamber. The CB was perfused at 80 Tort (by gravity) with a modified Tyrode solution, equihbrated with varying levels of PO 2 and PCO 2 and simultaneously superfused with the same Tyrode, equilibrated at PO E < 20 Torr. The chamber temperature was maintained at 36.5 -+ 0.5°C. The composition of the modified Tyrode was (in raM) Na + 154, K + 4 7, Ca 2÷ 2.2, Mg 2+ 1.1, C1- 110, glutamate 42.0, glucose 5.0, and HEPES 5.0 and dextran (5.0 g/l). Glutamate was found to be an avid substrate for chemosensory function during prolonged perfusmn 25. The preparataons (n = 13) were perfused first with Tyrode without CO2-HCO 3 for 30 rain (PO 2 = 124 -+ 2.9 Tort, pH = 7.38 -+ 0.01) and then with Tyrode containing CO2-HCO3 (HCO~ = 21.4 raM, replacing the same amount of NaC1, P O 2 -- 123.2 -+ 2.6 Torr. PCO 2 = 36.8 -+ 0.7 Torr, pH = 7.38 -+ 0.01). Dunng perfusion with Tyrode, free of CO2-HCO3, the superfusate was equilibrated with 100% N 2, and during perfusion wzth medmm containing CO2-HCO~, the superfusate was equilibrated with 5% CO 2 m 95% N 2. PO2, PCO2 and pH of the perfusion and superfusmn media were measured at 37.0"C with a blood gas analyzer. The chemosensory discharges were recorded from the whole desheathed CSN. The nerve was placed on bipolar platinum electrodes and lifted into paraffin oil. The effluent was aspzrated with a vacuum system to maintain the fluid level in the chamber, sufficaent to cover the CB. The electrical signals were amplified and recorded with an electrostatic recorder. For quantitatiou of the electroneurograms, we used standard procedures as have been used both in vivo and m vztro6,7'9'14,zs'2z. Action potentials of given amplitudes above the basehne noise were selected wzth an electronic amplitude discriminator and were counted electromcally by a frequency meter and registered on a printer. The raw signals were also stored in a VCR-digltai system for later analyses. The caroud chemosensory responses, during perfusion-superfuslon with media free of or containing CO2-HCO3, were tested as follows. Two types of maneuvers were used for the responses to hypoxia: interruption of the perfusate flow by cross-clamping the mfluent line for 1-3 min (while the superfusate flow remained constant) and perfusion with hypoxic Tyrode (PO2 = 20-30 Torr) for 1-8 rain. For the responses to CO 2, normoxic Tyrode equilibrated wath PCO 2 of 65.6 -+ 3.5 Torr (pH = 7.17 -+ 0.02) was perfused for 2 rain. To test for the dependence of the effects of the pharmacological agents on CO2-HCO ; , several doses of sodium cyanide (10-m-10 -7 mole) and mcotme bitartrate (10-11-10-8 mole) were given in a bolus of 0.2 ml into the perfusate line. These agents were dissolved m Tyrode free of CO2-HCO~. Results are expressed as mean -+ S.E.M. Statistical differences between the paired samples were assessed by the Wilcoxon's signed rank test. To compare the dose-responses in the presence and absence of C O 2 - H C O 3 the chemosensory activity was expressed as percentage of the maximal response evoked by the interruption of the perfusate flow in the presence of CO2-HCO~. The data from cyanide and nicotine dose-response curves were fitted to symmetrical sigmoidal functions5. This method provides a basis for poolmg data from separate experiments, and allows the test of the characteristics which are shared by various curves. The data points were adjusted according to the following logistic expression R = Rm. x + (baS-Rma~)/(1 +
(DIEDso)s)
A
soo
60 s
CSN activity (imp/s)
Impulses
B
500
CSN activity (imp/s) j 0
Impulses C-90-93
Fig. 1. Effects of perfusate flow interruption on carotid chemosensory activity in the presence and nominal absence of CO2-HCO~. A: perfusion-superfusion wzth Tyrode nommaUy free of CO2HCO~. B: perfuszon-superfusion with Tyrode with CO2-HCO~ (PCO 2 = 38 Torr, 21.4 mM of HCO~) pH was 7.39 and PO2 120 Torr in A and B. Bars indicate duration of perfusate flow interruption.
curves were fitted through a computer program based on a simplex algorithm 16 as used previously8. The goodness of each fit was tested using an ANOVA and by the coefficient of correlation calculated by dividing the variance of theoretical values by that of the expertmental values s. The EDso values from different experiments were averaged as geometric mean.
RESULTS
Carotid chemoreceptor responses to interruption of perfusate flow in the presence and absence of C02-HCO ~ Fig. 1 s h o w s t h e e f f e c t s o f i n t e r r u p t i o n o f flow in t h e CB perfused with normoxic Tyrode nominally free of CO2-HCO3
(Fig. 1 A ) , a n d s u b s e q u e n t l y w i t h T y r o d e
c o n t a i n i n g C O 2 - H C O ~ (Fig. 1B). P e r f u s i o n w i t h T y r o d e c o n t a i n i n g C O 2 - H C O ~ a t t h e s a m e p H o o f 7.39 a n d P O 2 o f 120 T o r r (Fig. 1B) r a i s e d t h e b a s e l i n e c h e m o s e n s o r y activity, s h o r t e n e d t h e l a t e n c y o f r e s p o n s e , i n c r e a s e d t h e r a t e o f rise o f a c t i v i t y t o a g r e a t e r m a x i m a l l e v e l d u r i n g flow i n t e r r u p t i o n . T h e p e a k r e s p o n s e w a s f o l l o w e d b y a
where; R = chemosensory response; Rraax = mammal chemosensory response; bas = baseline chemosensory discharge; D = arithmetzc dose; ED5o = median effective dose; S = slope factor that deternunes the steepness of each curve. R, Rraax and bas were expressed as percentage of stop flow response in C O 2 - H C O 3 The
g r a d u a l d e c l i n e o f t h e activity. R e s t o r a t i o n o f flow dim i n i s h e d t h e activity r a p i d l y b u t t h e r a t e o f d e c l i n e was slower with CO2-HCO~
than without. The instant of
flow r e s t o r a t i o n i n b o t h Fig. 1 A a n d B w a s first a c c o m -
255 500-
T
400-
t. ~"
O ~
Q-
O
¢~++~.
-+m,~..
.
300
~v
E_
200
+~
I O0
200100-'
o
g, 0 !
0
-30
0
30
60
90
120
Time (s)
150
180
210
-30
240
C-90-95
Fig. 2. Carotid chemosensory response to hypoxia ( P O 2 = 25 Torr) during perfusion with Tyrode free of CO2-HCO~ (o) and then during perfusion with Tyrode equilibrated with CO2-HCO~ (PCO2 = 38 Torr, 21.4 ram of HCO 3, O). pH was 7.38 and PO 2 120 Torr. Arrows indicate onset and withdrawal of hypoxic perfnsion.
panied by a rise in the discharge rate which could have been due to transient baroreceptor activity. But that was unlikely because a similar p h e n o m e n o n occurred during the transition of the chemical stimulus (Fig. 2). Results obtained in 10 preparations are summarized in Table I. Clearly, C O 2 - H C O 3 improved the chemosensory response to flow interruption. The absolute maximal activity was greater but the net change (maximal minus basal) was the same with and without CO2-HCO3.
Carotid chemosensory responses to hypoxia in the presence or absence o f C 0 2 - H C 0 3 Fig. 2 compares the carotid chemosensory responses to changes of perfusate PO2 from 120 Torr to 25 Torr. In the presence of C O 2 - H C O 3 the baseline activity was greater and the same hypoxic stimulus sharply increased the chemosensory discharge after a shorter latency to a higher maximal level which showed a slight adaptation. Table II summarizes the characteristics of the responses with and without effects of C O 2 - H C O 3. O n the average, presence of CO2-HCO~ reduced the latency of the response to hypoxia by 95 --- 33.0 s (P < 0.05). Thus, the
0
I
I
I
30
60
90
Time (s)
--I
120
150
C-90-95
Fig. 3. Carotid chemosensory response to hypereapnia ( P C O 2 -- 68 Torr, pH 7.20) during perfusion with q~yrode free of CO2-HCO3 (o) and then during perfusion with Tyrode equilibrated with CO2HCO 3 (PCO2 = 38 Torr, 21.4 mM of HCO 3, Q). pH was 7.38 and PO2 120 Torr. Arrows indicate onset and withdrawal of hypercapnic perfusion.
presence of CO2-HCO 3 in the perfusate-superfusate at the same p H o strikingly augmented the speed of the responses. The absolute maximal response to a given hypoxia also increased but this was due in part to an increased baseline activity and to further stimulus interaction between pH, and hypoxia in the presence of CO2HCO~.
Carotid chemosensory response to hypercapnia in the presence or absence o f pre-perfusion with C02-HCO ~ Fig. 3 shows the chemosensory responses to hypercapnic-acidosis which was applied to the CB pre-perfused with and without CO2-HCO~. In both cases hypercapnic perfusion (PCO2 = 68 Torr, p H = 7.20, P O 2 = 118 Torr) augmented the chemosensory activity equally rapidly in a few seconds. In the absence of CO2-HCO~ the perfusate PCO2 rose from about 0.2 to 68 Tort and the absolute peak response was less but adapted more rapidly to a lower steady state level. During perfusion with Tyrode containing CO2-HCO ~, a rise of P C O 2 from 38-68 Torr caused a greater absolute peak response which adapted slowly. However, the initial baseline activity
TABLE I
Effects of C02-HCO ~ on the chemosensory responses to perfusate flow interruption Values are the mean -+ S.E.M. of 10 preparations. The half-time on and off values refer to flow interruption and flow restoration. During perfnsiou with Tyrode free of CO2-HCO~, PO2 was 121.3 -+ 2.5 Torr and pH 7.39 -4- 0.01. During perfusion with Tyrode containing C O 2 - H C O 3 , PCO2 was 37.1 -+ 0.8 Tort, pH 7.39 -+ 0.01 and PO 2 123.2 -+ 2.5 Tort.
CondiUon CO2-HCO3
CO2-HCO3
free
Baseline (imp/s)
Maximal response (imp~s)
Net response (maxtmal-basal)
Half-time on (s)
Half-time off (s)
7.3 +- 1.7 53.9 -+ 9.1"
344.0 _+ 23.8 391.4 +- 25.9*
336.7 -+ 23.3 344.3 -+ 22.4
53.2 -+ 7.8 12.4 -+ 2.2*
6.2 -+ 0.3 12.5 -+ 1.7"
Significant differences from the control were assessed by the Wilcoxon Test (*P < 0.05).
256 TABLE II
Effects of C02-HCO 3 on the chemosensory response to hypoxtc perfuston Values are the mean - S.E.M. of 6 preparations. Half-time refers to the half response from the onset of hypoxia. Dunng perfusion with "l]¢rode free of CO2-HCO ~, PO 2 was changed from 120.5 --- 3.2 to 25 6 -+ 1,7 Torr and pH was kept constant at 7.38 -+ 0.01. Dunng
perfusion with Tyrode containing CO2-HCO~, PO2 was changed from 124.5 -+ 2.5 to 22.8 -+ 1.6 Torr, PCO2 was kept constant at 34.2 -+ 1.4 and pH at 7.39 -+ 0.01. Condinon
Baseline (imp/s)
Maxtmal response (imp/s)
Net response (maximal-baseline)
Half-ume (s)
CO2-HCO ~ free CO2-HCO 3
6.9 -+ 1.4 51.0 -+ 9.8*
247 9 -+ 45.8 380 0 -+ 36.1"
232.8 -+ 43.1 328.2 -+ 34.9*
117.1 -+ 38.8 22.1 -+ 6.9*
Significant differences from the control were assessed by the Wilcoxon Test (*P < 0.05)
was higher and the net peak response was only slightly greater. Withdrawal of hypercapnia was followed by a greater undershoot in the presence of CO2-HCO 3. Table III summarizes the chemosensory responses to hypercapnic-acidosis. Pre-perfusion of the CB with CO 2HCO3 did not modify the speed of response and did not significantly increase the net peak response (maximal minus basal). Thus the initial responses to hypercapnic acidosis were apparently correlated with the applied stimulus strength but the net response during adaptation was greater with CO2-HCO3 pre-perfusion.
the latency of the physiologic response. Fig. 4 shows the effects of several doses of cyanide on the chemosensory activity in the absence (Fig. 4A) and presence of CO 2HCO~ (Fig. 4B). As expected, the magnitude and duration of carotid chemosensory responses increased with the dose in both cases. However, the same dose evoked a larger and a more prolonged increase of chemosensory activity in the presence of CO2-HCO 3 which also raised the baseline activity. Also, presence of CO2-HCO~ reduced the latency of responses. Fig. 5 shows the fitting of the normalized chemosensory activities elicited by cyanide to sigmoidal functions (n = 5). Each data point is expressed as percent of the maximal response elicited by stop flow of the perfusate with CO2-HCO3 (see methods). The parameters defining the dose-response curves show that the presence of CO2-HCO 3 increased the maximal activity from 42.2 12.5% to a 96.0 - 9.2% (P < 0.01). The slope factors
Effects of carotid chemosensory responses to nicotine and cyanide in the presence and absence of C 0 2 - H C O 3 Injections of nicotine or cyanide increased the chemosensory discharge within a few seconds. This delay was partly due to the transit time of the perfusate flow from the point of injection to the CB and partly to
A CSN activity (imp/s)
Impulses
4olr =, ,
.
, = ......
60 s
J .....
, _it . . . . . .
_~
JUt _-S
I ill
~i
,
ii~111
I I
I ~_~,
Illll I J
NaCN (nmol) 6
4 0
6 0
20
40
60
200
400
600
B CSN actwity (imp/s)
Impulses NaCN (nmol) C-90-105 r
Fig. 4. Effects of CO2-HCO 3 on caroUd chemosensory response to several NaCN doses. A: responses dunng normoxic perfuston with Tyrode free of CO2-HCO 3. B: responses during perfuslon with Tyrode equilibrated with CO2-HCO3 (PCO 2 at 36.6 Tort, 21.4 mM of HCO3). pH was 7.39 and PO 2 122 Torr. Note that the doses in B were smaller than those m A.
257 with and without CO2-HCO~ respectively, were 1.6 --. 0.3 and 2.1 --- 0.3 (P > 0.05). The EDso decreased from 2.9 - 0.9 l0 s mol to 7.6 --- 1.5 10-9 tool (P < 0.01). Fig. 6 shows the carotid chemosensory response to several doses of nicotine. Presence of C O 2 - H C O 3 raised the baseline activity, did not significantly modify the peak responses to nicotine ( > 4 - 6 10-9 mol) but prolonged its effect (Fig. 6B). Fig. 7 summarizes the doseresponses to nicotine (n = 5). The maximal peak activities were not different in the absence or presence of CO2-HCO~ (respectively 88.7 - 11.6% vs 85.1 - 12.2%, P > 0.05) and the slope factors were not different with and without CO2-HCO 3 (respectively 2.2 - 0.9 and 1.7 --- 0.3, P > 0.05). H o w e v e r the ED5os decreased from 9.3 --- 1.3 10-~° mol to 6.0 --- 1.6 10-~° mol (P < 0.05) in the presence of C O 2 - H C O 3. This difference was mostly due to an increase in the baseline activity which raised the lower part of the curve upward.
I
0
f ]
• C?2-HC03-
"~ ~ u 0
80.
o
60-
~
(..)
g ~~ m 0
IZ
0 0-
,
+"
1
,0.1
8~ ~
_
0
-9
-8 -7 Log NoCN dose (moo
-6
Fig. 5. Dose-response curves for chemosensory activity elicited by NaCN in the presence or absence of CO2-HCO~. The coefficient of correlatton in the absence and presence of CO2-HCO~, respectively, were 0.997 and 0.995. Chemosensory activity was expressed as percentage of maximal response produced by stop flow during COz-HCOf perfusate in each preparation (n = 5). o, dose response curve with normoxic perfusion with Tyrode free of CO2HCO~; O, dose response curve during perfusion with Tyrode equilibrated with CO2-HCO~ (PCO2 at 36 Torr, 21.4 mM of HCO~). pH was 7.39 and PO e 120 Tort. Note that the doses in B were smaller than those in A. Maximal reactivity was significantly higher (P < 0.01) with CO2-HCO; perfusate.
DISCUSSION Cat carotid body chemosensory fibers responded to the physiological and pharmacological stimuli in the nominal absence of C O 2 - H C O ; in the external medium but the presence of C O 2 - H C O ; significantly augmented these responses. The baseline chemosensory discharge increased in the presence of C O 2 - H C O ~, suggesting a stronger receptor potential. Since the responses of sensory organ receptors to stimuli depend on their background discharge ~3 it is reasonable that the presence of CO2-HCO3 augmented the sensory responses to expect
A
o HEPES
100-
the applied stimuli. However, the improvement of the responses were not uniform. The results showed a large improvement of the chemosensory responses to hypoxia and cyanide but only a small one to nicotine. Thus the increased baseline activity alone does not explain the differences. Also the fact that the CB sensory response to C O 2 strikingly differs from those observed in other
60 s
250
activityi
CSN (Imp/s)
I."1
r
%
4
::q
ot Impulses
Nicotine (nmol)
B CSN activity (,mp/s)
0.~6
•2
I
0.=4
v ,,....
~,+-~
,J '~M.
.
0".6
1".0
4".0
6".0
,
, 'e.+
•
~-,~,
• ...,~,y.
"+~.~-'~41.~,,,'
Impulses Nicotine (nmol)
0.06
0:2
0:4
0".6
2.0
~0
~.0
C-~l>-,osr
Fig. 6. Effects of CO2 -HCO 3 on carotid chemosensory response to several nicoUne doses. A: responses during normoxic perfusion with Tyrode free of CO2-HCO~. B: response during perfusion with Tyrode equilibrated with CO2-HCO; (PCO2 at 36.6 Torr, 21.4 mM of HCO3). pH was 7.39 and PO2 122 Torr.
258 i" 100 -~ o4 u oO (j
0
HEPES
• =co -.co3-
80
.
fects of C O 2 - H C O 3 occurred even though the p H o was the same with and without C O 2 - H C O 3 ( p H 7.38). Accordingly the effect of C O 2 - H C O 3 was m e d i a t e d intracellularly, presumably by intracellular acidification of the glomus cells. I n d e e d , glomus cell acidification was found with m e d i u m containing C O 2 - H C O 3 at the same external p H 4'27. Exposure to C O 2 - H C O 3 increased the cellular acidity not just because of carbonic acid formation m e d i a t e d by carbonic anhydrase 15 but also because of the H C O 3 extrusion through the C I - - H C O ~ exchanger mechanisms, as found in glomus cells4 and some o t h e r cells2't1'19'23. The observation that b l o c k a d e of C1--HCO 3
~c 0 -
60
o
40
~ ° o -,,-, (.) m
20. 0 -11
-10
-9
-8
Log N,cobne dose (tool)
Fig. 7 Dose-response curves for chemosensory activity ehcited by nicotine in the presence or absence of CO2-HCO~. The coefficient of correlation in the absence or presence of CO2-HCO ~, respectively, were 0.996 and 0.995. Chemosensory activity was expressed as percentage of maximal response produced by stop flow dunng CO2-HCO~ perfusion m each preparation (n = 5). o, dose response curve during normoxic perfusion with Tyrode free of CO 2HCO~; O, dose response curve during perfusion with Tyrode equilibrated with COz-HCO~ (PCO 2 at 36 Torr, 21.4 mM of HCO3). pH was 7.39 and PO2 120 Torr, The maximal responses were the same (P > 0.05) during perfusion with or without CO2-HCO3.
sensory organs is relevant: C O 2 suppressed the generator potentials in the sensory endings of the cat muscle spindle 1° and diminished the amplitude of the receptor potential of the barnacle photoreceptor3. Accordingly it is likely that the interaction of CO2 with hypoxia and cyanide involving 0 2 chemotransduction was not the same with nicotine, although the initial generator potential was similar. The current consensus is that the glomus cell which contains both carbonic anhydrase Is and neurotransmitter 7 is the initial site of chemoreception. The glomus cells are therefore equipped to promptly respond to CO2 and to release neurotransmitters which in turn could generate action potentials in the chemosensory nerve terminals 7. There is an interesting and relevant parallelism between glomus cell acidification and chemosensory discharge. The results showed that the augmenting el-
exchanger by eliminating extracellular H C O 3 or C1m a d e the rat (neonatal and adult) glomus cells alkaline even in the presence of CO2-HCO3-4 support that acidification by the anion-exchanger mechanism was p r e d o m inant in these cells. Chloride channels in the glomus cells with large conductance and selectivity for HCO326 presumably also participated in the process because we recently found that the C1- channel b l o c k e r 9-anthracenecarboxylic acid decreased the baseline chemosensory activity (unpublished results). Hypercapnic acidic stimuli. T h e unsteady-state responses to hypercapnic acidosis were not correlated with the final strength of the stimulus. T h e same final stimulus elicited a greater absolute response when the P C O 2 was raised from 35 to 65 Tort than from 0 to 65 Torr. H o w e v e r , for the same p H o stimulus the response was similar, and was s u p e r i m p o s e d on the existing baseline chemosensory activity. The baseline chemosensory activity was greater in the presence of CO2-HCO~, presumably because of a m o r e acidic pH, relative to that in the absence of CO2-HCO3-4. The responses to hypercapnic acidosis are due in part to acid pHo and to intracellular acid, generated by C O 2 hydration. The net effect (maximal minus basal), but not the absolute effect, was correlated with the final stimulus regardless of the initial level of P C O 2. O n the other hand, the net increase in chemosensory activity was correlated with A p H o (PHo
TABLE III Effects of C02-HCO 3 on the chemosensory response to hypercapmc perfuston
Values are the mean -+ S.E.M of 4 preparations. Half-ume refers to the half response from the onset of hypereapmc periuslon. Dunng perfusion with Tyrode free of CO2-HCOf, PO 2 123.0 - 2.4 Torr and pH 7.38 ± 0.01 and dunng perfusion with q~yrode containing CO2HCO~, PCO2 was 36.6 ± 2.8, pH was 7.39 -+ 0.01 and PO 2 123.0 - 1.4 Torr. During hypercapnic-acidic perfusion PCO2 was 65.6 ± 3.5 Torr, pH 7.17 ± 0.02 and PO 2 118.8 - 4.2 Torr. Chemosensory response was measured at its maxtmal (Peak) response and then at 1 (Res at 1 mm) and 2 min (Res at 2 mm) of hypereapnic perfumon. Condition
Baseline (imp/s)
Peak response Omp/s)
Net response (Peak-basal)
Half-time (s)
Res at 1 min (trap~s)
Res at 2 rain (trap~s)
COz-HCO ~ free CO2-HCO ~
8.3 - 1.8 67.8 ± 16 3*
216 4 _ 39.2 314.4 ± 53.7*
207.1 - 38.4 247.1 - 37.1
7.0 ± 1.9 10.0 ± 4.7
79.8 ± 18.3 232.5 ± 14.5"
112.4 +- 17.2 239.9 ± 16.5"
Slgmficant differences from the control were assessed by the Wdcoxon Test (*P < 0.05)
259 before and during hypercapnia) but not with APCO 2. The initial and late responses to PCO2 were strikingly less when the initial PCO 2 was near zero and the cells alkaline 4. The initial pH~ difference with and without CO2-HCO 3 presumably made the difference in the chemosensory discharge during the unsteady-state. Ultimately in the steady state, the pI-I~ and chemosensory responses would be the same. Metabolic acidosis at a constant arterial PCO 2 is known to stimulate carotid chemosensory discharges in vivo21 and in vitro 6. We also found in vitro that perfusion of acid Tyrode (pH 6.8) stimulated carotid chemosensory discharge in the absence of CO2-HCO ~ (unpublished observations). These observations are consistent with the reports that the glomus cells pI-I~ correlates well with pHo 4'27. Also the release of carotid body dopamine and chemosensory discharge correlate during acidic stimulation 22. Despite the parallelism between pH~ changes and chemosensory discharge the mechanisms relating cellular proton concentrations and signal transduction are unclear. H y p o x i c stimulus. T h e mechanisms of 0 2 and CO2 stimulus interaction were not studied here. However, if neurotransmitters are released by these stimuli, as seems to be the case 7, and the chemosensory discharges are any indication of neurotransmitter release, it is possible that the presence of CO2-HCO ~ in the cellular environment augmented the release of neurotransmitter during hypoxia. On the other hand, the studies which used medium free of CO2-HCO ~ to superfuse the carotid body might have underestimated the response to hypoxia and flow interruption because of a possible alkaline pI-I~ of the glomus cells4. Also a brief hypoxic stimulus without CO2-HCO ~ may not elicit a significant chemosensory response and could give an impression that the hypoxic response is lost without CO2-HCO ~ in the medium as Shirihata and Fitzgerald were led to conclude 24. Nonetheless hypoxia has been found to diminish the membrane K + conductance TM and to increase the intracellu-
REFERENCES 1 Biscoe, T.J. and Duchen, M.R., Electrophysiological responses of dissociated type I cells of the rabbit carotid body to cyanide, J. Physiol., 405 (1988) 210-232. 2 Boron, W.F., Boyarski, G. and Ganz, M., Regulation of intracellular pH in mesangfialcells, Ann. N.Y. Acad. Sci., 574 (1989) 321-332. 3 Brown, H.M. and Meech, R.W., Light induced changes of internal pH~ in a barnacle photoreceptor and the effect of internal pH on the receptor potential, J. Physiol., 297 (1979) 73-93. 4 Buckler, K.J., Vaughan-Jones, R.D., Peers, C. and Nye, P.C.G., Intracellular pH and its regulation in the isolated type I carotid body cells of the neonatal rat, J. Phystol., 436 (1991) 107-129. 5 De Lean, A., Munson, P.J. and Rodbard, D., Simultaneous
lar [Ca2+] 1 of the glomus cells in CO2-HCO 3 free medium. These results are consistent with our findings that the hypoxic effect ultimately does not require the participation of external CO2-HCO~. The present findings are also in conformity with the previously published reports of hypoxia stimulated carotid chemosensory discharge and catecholamine release in the nominal absence of CO2-HCO; in the superfusate medium 6'7'9'14. Pharmacological stimuli. T h e augmented response to cyanide, a metabolic inhibitor, in the presence of CO 2H C O 3 can also be explained by the cellular [H +] increase. The in vitro response to cyanide is analogous to the observed CO2-cyanide stimulus interaction in the cat chemosensory responses in vivo2°. The mechanism of this interactive response is not known but the phenomenon resembles CO2-hypoxia interaction. The less interactive response to nicotine, a receptor agonist, also resembles that in in vivo2°. Hypoxia and cyanide are knowt3 to release putative neurotransmitters, acetylcholine and dopamine 7'9A2, whereas nicotine has been reported to elicit a nearly selective release of norepinephrine 12 and also is expected to act on the nicotinic receptors on the nerve endings. These basic differences between the effects of nicotine and cyanide presumably explain their different interaction with CO2-HCOf. In summary, the results showed that 0 2 chemoreception concurred both in the absence and presence of CO2H C O ; in the extracellular medium but CO2-HCO ~ raised the baseline activity and improved the responses to physiological and pharmacological stimuli. The effect of CO 2H C O 3 could be attributed to intracellular acidification through the CI--HCO~ exchanger mechanisms. However, the basic hypoxic mechanisms of chemotransduction is not mediated by the acidification produced by CO 2 and the acid loading exchanger mechanisms.
Acknowledgements. We would like to thank A. Mokashi for his assistance. Supported in part by NIH Grants HL-43413 and HL19737.
analysis of famdies of sigmoidal curves: application to bioassay, radioligand assay, and physiological dose-response curves, Am. J. Physiol., 253 (1978) E97-E102. 6 Eyzaguirre, C. and Koyano, H., Effects of hypoxia, hypercapnia, and pH on the chemoreceptor activity in vitro, J. Physiol., 178 (1965) 385-409. 7 Eyzaguirre, C., Fitzgerald, R.S., Lahiri, S. and Zapata, P., Arterial Chemoreceptors. In J.T. Shepherd and F.M. Abboud (Eds.), Handbook of Physiology. Sect. 2: The Cardiovascular System, Vol. Ili, Peripheral Circulation and Organ Flow, American Physiological Society, Williams and Wilkins, Baltimore, 1983, pp. 557-621. 8 Eugenin, J., Larrain, C. and Zapata, P., Functional recovery of the ventilatory chemoreflexes after partial chronic denervation of the nucleus tractus solitarius, Brain Research, 523 (1990) 263-272.
260 9 Fidone, S.J., Gonzalez, C. and Yosluzaki, K., Effects of low oxygen on the release of dopamine from the rabbit carotid body in vitro, J. Physiol., 333 (1982) 93-110. 10 Fukami, Y., The effect on NI-Ia and CO2 on the sensory ending of mammalian muscle spindles: intracellular pH as a possible mechamsm, Brain Research, 463 (1988) 140-143. 11 Gailard, D.S. and Dupont, J.L., Ionic control of intracellular pH In the rat cerebellum Purkinje cells maintained in culture, J. Physiol., 425 (1990) 71-83. 12 Gomez-Nifio, A., Dinger, B., Gonzalez, C. and Ftdone, S.J., Differential stunulus coupling to dopamine and norepinephrine stores in rabbit carotid body type I cells, Brain Research, 525 (1990) 160-164. 13 Hunt, C.C. and Kuffler, S.W., Further study of efferent smallnerve fibres to mammalian muscle spindles. Multiple spindle innervation and activity during contraction, J. Physiol., 113 (1953) 283-297. 14 Iturriaga, R., Rumsey, W.L., Mokaslu, A., Spergel, D., Wilson, D.E and Lahiri, S., In vitro perfused-superfused cat carotid body for physiological and pharmacological studies, J. Appl. Physiol., 70 (1991) 1393-1400. 15 Iturriaga, R., Lahiri, S. and Mokashi, A., Carbonic anhydrase and chemoreception in the cat carotid body, Am. J. Physiol., 261 (1991) in press. 16 Johnston, A., SIMP: a computer program in BASIC for nonlinear curve fitting, J. Pharmacol. Meth. Progr. Biomed., 28 (1989) 37-42. 17 Lahiri, S. and DeLaney, R.G., Stimulus interaction in the responses of carotid body chemoreceptor single afferent fibers, Respir. Physiol., 24 (1975) 249-266.
18 L6pez-L6pez, J.R., Gonzalez, C,, Urefia, J. and L6pez-Barneo, J., Low PO 2 selectively inhibits K channel acttvlty in chemoreceptor cells of the mammalian carotid body, J Gen. Physiol., 93 (1989) 1001-1015. 19 Madshus, I.H., Regulation of mtracellular pH in eukariotic cells, Biochem. J., 250 (1988) 1-8. 20 Mulligan, E. and Lahln, S., Dependence of carotid chemoreceptor stimulation by metabolic agents on PaO2 and PaCO 2, J Appl. Physiol., 50 (1981) 884-891. 21 Pokorski, M. and Lahirl, S., Aortic and carotid chemoreceptor responses to metabolic acidosis in the cat, Am. J. Physiol., 244 (1983) R652-R658. 22 Riguald, R., L6pez-L6pez, J.R. and Gonzalez, C., Release of dopamine and chemoreceptor discharge induced by low pH and high PCO 2 stimulation of the cat carotid body, J. Physiol., 433 (1991) 519-531 23 Roos, A. and Boron, W.E, Intracellular pH, Physiol. Rev, 61 (1981) 296-434. 24 Shlrahata, M. and Fitzgerald, R.S., The presence of CO2/I-ICOf is essential for hypoxic chemotransduction in the m vivo perfused carotid body, Brain Research, 545 (1991) 297-300. 25 Spergel, D., Metabohc substrate dependence of carotid chemosensory discharge, FASEB J., 5 (1991) Al120. 26 Stea, A. and Nurse, C., Chloride channels in culture glomus cells of the rat carotid body, Am J Physiol., 257 (1989) C174C181. 27 Wllding, T.J., Cheng, B and Roos, A., The relationship between extracellular pH (pHo) and intracellular pH (pHi) in adult rat carotid body glomus cells, Btophys. J., 59 (1991) 184a.