Autonomic influences on the production of cerebrospinal fluid

MISSING IMAGE

Material Information

Title:
Autonomic influences on the production of cerebrospinal fluid
Physical Description:
xii, 108 leaves : ill. ; 29 cm.
Language:
English
Creator:
Haywood, Joseph Roscoe, 1949-
Publication Date:

Subjects

Subjects / Keywords:
Cerebrospinal Fluid   ( mesh )
Autonomic Agents   ( mesh )
Pharmacology thesis Ph.D   ( mesh )
Dissertations, Academic -- Pharmacology -- UF   ( mesh )
Genre:
bibliography   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph.D.)--University of Florida, 1976.
Bibliography:
Includes bibliographical references (leaves 99-107).
Statement of Responsibility:
by Joseph Roscoe Haywood.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 000898805
oclc - 25764493
notis - AEK7516
System ID:
AA00004919:00001

Full Text














AUTONOMIC INFLUENCES ON THE PRODUCTION
OF CEREBROSPINAL FLUID












By

JOSEPH ROSCOE HAYWOOD


A DISSERTATION PRESENTED TO THE
GRADUATE COUNCIL OF THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY












UNIVERSITY OF FLORIDA
1976































To Carol

for her understanding, patience, and love



To Becca and Alyson

for giving me so much happiness















ACKNOWLEDGMENTS


I wish to express my sincere gratitude to Dr.

Betty P. Vogh for-her guidance throughout this course of

study. She has been most helpful as a teacher, as a

researcher, and as an advisor, but especially as a friend

whose interest in her students extends beyond science.

I am also grateful to the other members of my

supervisory committee, Dr. C.Y. Chiou, Dr. Carl Feldherr,

Dr. David Silverman, and Dr. David Travis for their

contributions. The advice of Dr. Thomas Maren, Dr. Peter

Lalley, Dr. Sidney Cassin, and the faculty of the

Department of Pharmacology and Therapeutics is also

greatly appreciated.

I also thank Mr. Mark Loveland for his friendship,

sense of humor, and technical assistance and Ms. Joyce

Simon for her support and contribution in the preparation

of this manuscript.

My sincere thanks are also extended to my fellow

graduate students for their encouragement and friendship

during these years.

This research was supported by a traineeship from

the National Institutes of Health (GM-00760) and a research

grant from the National Institute of General Medical

Sciences (GM-16934).


















TABLE OF CONTENTS


ACKNOWLEDGMENTS . .

LIST OF TABLES . .

LIST OF FIGURES . .

KEY TO ABBREVIATIONS . .

ABSTRACT . . .

INTRODUCTION . .

BACKGROUND . .

Autonomic Innervation and Function in
Cerebral Blood Vessels .

Innervation of the Choroid Plexus .

Blood Flow to the Choroid Plexus .

Autonomic Nervous System Control of
Fluid Secretions . .

Rationale for Study . .

METHODS . .


Page

iii

vi

viii




ix



3


3

5

8


8
. 12






* 12

. 13


Tests of Autonomic Nervous System Influence
on CSF Production . .

Tests of Autonomic Nervous System Influence
on Blood Flow . .

Evaluation of CSF Production Data .

Statistical Analysis .. .

RESULTS . . .

Ventriculocisternal Perfusion Studies
for Estimation of CSF Production Rate .










Changes in Blood Pressure and Heart Rate
during Ventriculocisternal Perfusion Studies

Effect of Autonomic Agonists on Blood Flow

DISCUSSION . . .

Modification of CSF Production by
Autonomic Influences . .

Mechanism of Changes in CSF Production
by Autonomic Influences . .

Practical Significance . .

APPENDICES

I Autonomic Innervation and Function of
Cerebral Blood Vessels . .

II Composition of Artificial Cerebrospinal
Fluid . . .

REFERENCES . . .

BIOGRAPHICAL SKETCH . .


. 53

67

. 70


. 70


. 82

. 91




93


. 98

99

108

















LIST OF TABLES


TABLE Page


1 CONTROL PRODUCTION OF CEREBROSPINAL FLUID .27

2 CARBACHOL DOSE-RESPONSE: CHANGES IN CSF
PRODUCTION . ... 31

3 EFFECTS OF AUTONOMIC BLOCKING AGENTS ON CSF
PRODUCTION . .. 37

4 PHENYLEPHRINE DOSE-RESPONSE: CHANGES IN
CSF PRODUCTION .. . 38

5 THE EFFECT OF HEMICHOLINIUM-3 (HC-3) ON CSF
PRODUCTION DURING INTRAVENOUS INFUSION OF
PHENYLEPHRINE (Phe) . .. 44

6 BETA.ADRENERGIC AGONISTS DOSE-RESPONSE:
CHANGES IN CSF PRODUCTION . 46

7 EFFECTS OF THEOPHYLLINE ON CSF PRODUCTION,
BLOOD PRESSURE, AND HEART RATE 51

8 EFFECTS OF CARBONIC ANHYDRASE INHIBITION
ALONE AND WITH AUTONOMIC BLOCKING AGENTS
ON CSF PRODUCTION . .. .52

9 CARBACHOL DOSE-RESPONSE: CHANGES IN
BLOOD PRESSURE AND HEART RATE . 54

10 EFFECTS OF AUTONOMIC BLOCKING AGENTS ON
BLOOD PRESSURE AND HEART RATE .58

11 PHENYLEPHRINE DOSE-RESPONSE: CHANGES IN
BLOOD PRESSURE AND HEART RATE .. .59

12 THE EFFECT OF HEMICHOLINIUM-3 (HC-3) ON BLOOD
PRESSURE AND HEART RATE DURING INTRAVENOUS
INFUSION OF PHENYLEPHRINE (Phe) 63

13 BETA ADRENERGIC AGONISTS DOSE-RESPONSE:
CHANGES IN BLOOD PRESSURE AND HEART RATE 64











14 EFFECTS OF PHENYLEPHRINE, CARBACHOL, AND
SALBUTAMOL ON BLOOD FLOW TO THE CHOROID
PLEXI, BRAIN, SKELETAL MUSCLE, AND KIDNEY 68


vii















LIST OF FIGURES


FIGURE Page


1 Changes in the production of CSF with
time in control experiments ... .29

2 Carbachol dose-response for CSF production 33

3 95% confidence interval on probit analysis
of the effect of carbachol on CSF
production . . 35

4 Phenylephrine dose-response for CSF
production . . 40

5 95% confidence interval on probit analysis
of the effect of phenylephrine on CSF
production . ... 42

6 Salbutamol dose-response for CSF production. 48

7 95% confidence interval on probit analysis
of the effect of salbutamol on CSF
production . ... .50

8 Carbachol dose-response for blood pressure .56

9 Phenylephrine dose-response for blood
pressure . . 61

10 Salbutamol dose-response for blood pressure 66

11 Autonomic pathways to the choroid plexus I 73

12 Autonomic pathways to the choroid plexus II 77

13 Autonomic pathways to the choroid plexus III 80


viii
















KEY TO ABBREVIATIONS


CBF Cerebral Blood Flow

CSF Cerebrospinal Fluid

cpm Counts per minute

cAMP Adenosine 3',5' cyclic monophosphate

cGMP Guanosine 3',5' cyclic monophosphate

ED50 Median effective dose (dose to
produce 50% effect)

gm Gram

HC-3 Hemicholinium-3

ITP l-isopropylamino-3(2 thiazoloxy)-
2 propanol HCI

kg Kilogram

Pg Microgram

p1 Microliter

mg Milligram

ml Milliliter

mm Millimeter

mM Millimolar

msec Millisecond

min Minute

M Molar

M.W. Molecular weight

sec Second

Phe Phenylephrine










Abstract of Dissertation Presented to the Graduate Council
of the University of Florida in Partial Fullment of the
Requirements for the Degree of Doctor of Philosophy



AUTONOMIC INFLUENCES ON THE PRODUCTION
OF CEREBROSPINAL FLUID

By

Joseph Roscoe Haywood

December, 1976

Chairperson: Betty P. Vogh
Major Department: Pharmacology and Therapeutics

For the first time, treatments which stimulate or

inhibit autonomic nervous system receptors were tested for

their effects on cerebrospinal fluid (CSF) production

using the dilution of blue dextran during ventriculo-

cisternal perfusion in cats. Initial control CSF formation

ranged from 15 to 25 pl/min and gradually decreased at the

rate of 0.015 pl/min each minute between 60 and 300 minutes.

The cholinergic agonist, carbachol, in single

intravenous injections caused a dose-dependent increase

in the production of new fluid. The maximum increase was

10.6 2.1 pl/min which was blocked by atropine.

Single intravenous injections of the alpha adrenergic

agonist, phenylephrine, also increased CSF production to

a maximum of 10.6 1.5 pl/min over control. This dose-

dependent effect was prevented by phentolamine and by

atropine. An increase in CSF formation during infusion

of phenylephrine was blocked by administration of hemi-

cholinium-3 through a reduction of endogenous








acetylcholine. The ability of atropine and hemicholinium

to inhibit the increase in CSF production by phenylephrine

demonstrates that the alpha adrenergic agonist probably

activates a cholinergic pathway. Atropine or phentolamine

alone reduced CSF formation 2.6 0.5 and 3.7 0.8 pl/min,

indicating that the cholinergic pathway exerts some control

over normal fluid production. Neither atropine nor phen-

tolamine was able to add to the maximal reduction in CSF

formation caused by the carbonic anhydrase inhibitor

methazolamide.

The beta2 adrenergic agonist, salbutamol (albuterol),

also increased the rate of CSF formation, but it was only

half as effective as carbachol and phenylephrine. The

betal adrenergic agonist, l-isopropylamino-3(2 thiazoloxy)-

2 propanol HC1 (ITP), had no effect on CSF production.

Propranolol did not affect fluid formation but did reduce

the effect of salbutamol.

Bilateral stimulation of the cervical-sympathetic

trunk reduced CSF production 3.0 0.6 l/min. This

observation indicated that alpha adrenergic agonists

probably act at the choroid plexus to decrease CSF

formation; however, any direct action of circulating

alpha agonists is masked by activating the cholinergic

pathway.

Experiments were also performed to determine whether

autonomic agents influenced CSF production through hemo-

dynamic changes or a direct action on the choroid plexus

epithelium. Blood pressure data for carbachol,

xi









phenylephrine, and salbutamol did not correlate with the

changes in CSF production. Blood flow studies showed

that none of the agonists caused statistically significant

changes in cerebral or choroid plexus blood flow. These

findings suggest that autonomic drugs probably act directly

on the choroid plexus epithelium to change the rate of CSF

formation.

Verapamil, an agent that blocks calcium channels, did

not alter a carbachol-induced increase in fluid production

indicating that the autonomic agents probably did not exert

their effect by influencing calcium fluxes. The phospho-

diesterase inhibitor, theophylline, on the other hand,

increased the rate of CSF production. This effect

suggested that the autonomic treatments might alter fluid

formation by increasing or decreasing levels of the cyclic

nucleotides guanosine 3',5' cyclic monophosphate (cGMP) or

adenosine 3',5' cyclic monophosphate (cAMP).

Although the exact mechanisms remain to be elucidated,

it is clear that cholinergic and beta2 adrenergic receptors

can be stimulated to increase CSF formation significantly

and that the cholinergic pathway which predominates is

also activated by alpha adrenergic agonists.


xii















INTRODUCTION


Epithelia structurally similar to the choroid

plexus epithelium are found in the ciliary body of the

eye, the salivary gland, and the pancreas. Secretory

transport in these tissues is known to be controlled to

some extent by the autonomic nervous system. Therefore,

it is of interest to consider the influence of the

autonomic nervous system on the formation of cerebrospinal

fluid (CSF).

The columnar epithelial cells of the choroid plexus

are considered to be a primary site for the secretion of

CSF. This structure is an evagination of the walls of

the lateral and third ventricles of the brain and forms

the roof of the fourth ventricle. The choroid plexus

receives some innervation from the autonomic nervous

system. On the ventricular side, the outermost structure

of the choroid plexus is a single layer of epithelial

cells supported by connective tissue richly supplied with

blood vessels.

The autonomic nervous system is comprised of

sympathetic and parasympathetic components. The sympa-

thetic system has short fibers from the central nervous

system to the sympathetic chain ganglia, then long









post-ganglionic fibers which innervate the tissues.

Sympathetic effects are classified into alpha, betal, and

beta2 adrenergic actions and are mediated endogenously by

the adrenergic neurotransmitters norepinephrine and

epinephrine. Other agents can selectively stimulate alpha,

betal, or beta2 adrenergic activity if they have the

necessary structural modifications required for each

receptor. Phenylephrine is an alpha agonist; 1-isopropyl-

amino-3(2 thiazoloxy)-2 propanol HC1 (ITP) is a beta1

agonist; and salbutamol is a beta2 agonist. The parasym-

pathetic system has long fibers from the central nervous

system to ganglia close to the tissue being supplied,

then short post-ganglionic fibers which innervate the

tissue. The cholinergic neurotransmitter acetylcholine

mediates parasympathetic actions. Chemically related

agents such as carbachol can initiate cholinergic

activity when applied to the effector site.















BACKGROUND


In all systems in which fluid formation can be

altered by agents that affect the autonomic nervous system,

one must consider two basic means by which these effects

may be mediated. Alterations in fluid production may occur

in direct response to autonomic agents acting on receptors

located in the secretary cells. Also changes in fluid

production may possibly result from changes in hemodynamic

relationships increasing or decreasing pressure and flow

in the vessels of the choroid plexus. Functionally,

changes in hemodynamic relationships may interact with

direct changes in the secretary function of the epithelium

to alter fluid formation.



Autonomic Innervation and Function in Cerebral Blood Vessels


By following the nerve trunks passing into the

brain, anatomists identified nerves supplying the large

pial vessels more than a century ago. Microscopists

subsequently described the presence of nerve fibers in

the finer cerebral vasculature, but classification of the

nerves as sympathetic and parasympathetic did not come

until the 1920's. Since the blood vessels of the choroid









plexus arise from cerebral arteries, a brief summary of

general cerebrovascular innervation and autonomic function

will be given before considering the choroid plexus

specifically. A more detailed presentation of the

autonomic innervation and responses of cerebral blood

vessels is given in Appendix I.

Sympathetic innervation of the large pial blood

vessels is supplied by the cervical sympathetic trunks.

This has been demonstrated by showing the disappearance

of adrenergic nerves after cervical sympathetic ganglionec-

tomy. However, smaller parenchymal vessels within the

brain receive adrenergic nerves from the brainstem. These

nerves survive bilateral cervical sympathectomy. Parasym-

pathetic innervation of cerebral blood vessels originates

primarily through the facial nerve. Recent experiments

have shown that cholinergic nerves also come from sources

other than the cranial nerves.

A decrease in cerebral blood flow in dogs after

electrical stimulation of the cervical sympathetic trunk

or administration of norepinephrine has been demonstrated

by some investigators; however, others have been unable to

elicit these effects. On the other hand, stimulation of

the facial nerve or exogenous administration of acetyl-

choline reproducibly and significantly increases cerebral

blood flow.

Cerebral blood vessels have shown more consistent

adrenergic responses in vitro than in the whole animal





5



experiments. Both electrical stimulation and alpha

adrenergic agents contract cerebral arteries in vitro,

but the maximum force of contraction is less than in

arteries from other tissues. Beta adrenergic responses

in cerebral vessels so far have been found only in the

cat. In this species, beta2 stimulation relaxes vessels

contracted with serotonin.

The significant observation from both the in vivo

and in vitro studies is that regardless of the nature of

the stimulation (i.e., electrical or pharmacological) the

adrenergic response of the cerebral vessel is small or

absent. In contrast, cholinergic stimulation has a strong

influence on cerebral blood vessels. In vitro acetyl-

choline has been shown to relax contracted vessels.



Innervation of the Choroid Plexus


Since the choroid plexus is composed of an epithelium

resting on connective tissue containing blood vessels, it

is not feasible to discuss separately innervation of the

epithelium and innervation of the vessels with regard to

what is presently known about the anatomy of the tissue.

For this reason, evidence for nerves supplying both the

epithelia and the blood vessels of this secretary tissue

will be presented together.

Benedikt (1874) first observed nerve fibers around

blood vessels and near epithelial cells of the fourth





6



ventricle plexus. These fibers arose from a vagal nucleus

in the medulla which he called the XIII Cranial Nerve.

Later, innervation in the lateral plexus was described as

being mostly around the vessels, and it was suggested that

these nerves might be involved in vasomotor function

(Findlay, 1899). Stohr (1922) confirmed the observation

of Benedikt in the fourth ventricle plexus and also

demonstrated a nerve supply to the third ventricle plexus.

He suggested the nerve fibers were divided into "vascular

nerve fibers" entering with the arterial vessels and

choroidd plexus proper" fibers originating in a vagal

nucleus, in pontine nuclei, and in cerebral peduncles.

Junet (1926) also described nerves supplying the epithelial

cells of the plexus. Stohr's observation of separate

vascular and epithelial innervation to the fourth ventricle

plexus originating from the medulla was confirmed by Clark

(1928, 1934), and he extended the theory to include the

lateral plexus. Clark described the fibers as being

mostly in the connective tissue base of the choroid plexus

rather than around the epithelial cells. Some myelinated

nerves and a large number of unmyelinated nerves were

found in association with blood vessels and epithelial

cells without any of them ending directly on capillaries.

With electron microscopy, unmyelinated nerve fibers have

been identified near the choroid plexus epithelium

(Millen and Rogers, 1956) and around some blood vessels









supplying the plexus (Maxwell and Pease, 1956). Tennyson

(1975) has described a nerve process adjacent to an

epithelial cell. The cell membranes of both the nerve

and epithelium were thickened, indicating the presence of

a possible cell junction.

To determine the extent of adrenergic innervation,

two groups of investigators did cervical sympathetic

ganglionectomies. In the first study, a unilateral sympa-

thectomy resulted in a patchy degeneration of nerve fibers

in the connective tissue of the ipsilateral and third

ventricle plexi of dogs and cats with most of the inner-

vation remaining intact (Tsuker, 1947). The other study

demonstrated by fluorescence microscopy the presence of

an interconnected ground plexus of adrenergic nerves

between the vascular walls and the epithelial cells

(Edvinsson et al., 1974). These observations indicated

the lateral and third ventricle plexi had the highest

content of adrenergic innervation and the fourth ventricle

plexus content was much lower. After a bilateral sympa-

thectomy, the fluorescence disappeared in all the choroid

plexus and the norepinephrine content of the plexus

decreased significantly (Edvinsson et al., 1972b). Cholin-

ergic innervation of the choroid plexus epithelium and

blood vessels has also been described (Edvinsson et al.,

1973b). In cats and rabbits, cholinergic fibers were

observed for all the plexus in greater density than









adrenergic fibers. The network of cholinergic nerve

fibers remained intact after sympathetic ganglionectomy.



Blood Flow to the Choroid Plexus


Blood flow to the choroid plexus has been measured

in the intact whole animal in only one study. Alm and

Bill (1973), using radioactive microspheres, obtained a

control flow of 3.01 ml/gm min. Upon stimulation of the

cervical sympathetic chain, the blood flow to the plexus

did not change. Two groups have made in situ determin-

ations of plexus blood flow. Welch (1963), while

measuring CSF production at the lateral choroid plexus in

the rabbit, observed a blood flow of approximately 2.86

ml/gm min. Pollay et al. (1972) perfused the choroid

plexus vessels of sheep and related blood flow to the

production of CSF. For a normal rate of formation they

concluded choroid plexus blood flow must be about 2.56

ml/gm min. Pharmacological agents affecting blood flow

were not used in any of these studies; however, Macri

et al. (1966) and Politoff and Macri (1966) showed that

serotonin and 1-norepinephrine reduced choroidal blood

flow in rabbits.



Autonomic Nervous System Control of Fluid Secretions


Fluid secretion occurs in several specialized

epithelial tissues. Sympathetic or parasympathetic










agonists have been shown to influence secretion in nearly

all of the secretary epithelia, but there has not been a

consistent pattern of effects. The salivary gland is

probably the most thoroughly studied of the tissues in

which there is autonomic control over fluid production.

In this system, both cholinergic and adrenergic agonists

increase secretion (Koelle, 1970). Cholinergic agonists

can produce approximately ten times as much fluid as beta

adrenergic agonists (Mangos et-al., 1973). Petersen (1972)

has suggested that acetylcholine-activated secretion occurs

as a result of an uptake of extracellular calcium which

Douglas and Poisner (1963) showed was necessary for fluid

movement. The calcium, in turn, may cause an efflux of

potassium and an uptake of sodium which results in water

transfer.

The eye is another site in which fluid movement is

actively influenced by autonomic agonists. In this system

both production of fluid by the ciliary body and drainage

of aqueous humor through the trabecular meshwork are

under autonomic control (Chiou and Zimmerman, 1975). It

has been difficult to determine whether the cholinergic

system or the adrenergic system really predominates in

either mechanism, but it is certain that both are

influential. Paterson and Paterson (1972) have shown

that alpha adrenergic agonists cause a slight increase in

aqueous humor production and outflow facility. In contrast,









beta2 agonists (such as salbutamol) cause a decrease in

production and a large increase in flow through the

trabecular meshwork which can be blocked by propranolol

(Langham and Diggs, 1974 and Langham, 1976). The effects

of cholinergic agents on aqueous humor dynamics is more

complicated. The major effect in the whole animal is an

increased drainage of fluid (Ellis, 1971 and Macri and

Cevario, 1973). When acetylcholine and eserine, an anti-

cholinesterase agent, are given intra-arterially in an

isolated perfused cat eye, the formation of aqueous humor

increases 10 to 15 pl/min (Macri and Cevario, 1973). The

action of acetylcholine can be blocked by atropine;

however, it can also be prevented with alpha adrenergic

receptor blocking agents or guanethidine, an adrenergic

neuron blocking agent, indicating the effect is probably

mediated by alpha receptors.

The acinar cells of the pancreas are also under

autonomic control (Thomas, 1967). Either vagal st: !.tion

or the administration of cholinomimetic drugs can elicit

moderate increases in fluid secretion and substantial

increases in the release of digestive enzymes. Norepi-

nephrine and epinephrine decrease secretion possibly by

constricting the blood vessels which supply the tissue.

Fluid transfer in the toad bladder and frog skin

epithelia can be increased by beta adrenergic stimulation

(Foster, 1974). This presumably occurs through changes










in adenosine 3',5' cyclic monophosphate (cAMP) levels

since it can be mimicked by dibutyryl cAMP and potentiated

by phosphodiesterase inhibitors (Jard, 1974). On the

other hand, alpha adrenergic activation reduces water

movement stimulated by theophylline or antidiuretic

hormone (Foster, 1974 and Jard, 1974) but has no effect

in the absence of these agents. No observations have been

made concerning possible cholinergic actions.

Very little investigation has been made into the

role of autonomic agents in the production of CSF. Becht

and Gunnar (1921) set a needle in the cisterna magna of

cats and noted that epinephrine and pilocarpine increased

fluid outflow. They suggested that this was not true CSF

production since some of the collected fluid was drawn up

into the tube again after the action of the drug was

finished. Bhattacharya and Feldberg (1958) observed the

same effect during ventriculocisternal perfusion with

epinephrine, histamine, and tubocurare. In these exper-

iments only volumes were measured, and it was suggested

that the increased flow during the perfusion of the

ventricles was due to a change in cerebral blood flow.

Vates et al. (1964) observed that norepinephrine

(2.5 X 106 M in the perfusate) caused a decrease in

production of about 24% during ventriculocisternal

perfusion.









Rationale for Study


The evidence presented satisfies two major criteria

for concluding that the autonomic nervous system may exert

an influence on CSF production. First, it has been shown

that the blood vessels and epithelia of the choroid plexus

receive both adrenergic and cholinergic innervation. The

adrenergic nerves extend from the cervical sympathetic

trunk whereas the cholinergic nerves, comprising most of

the tissue's innervation, originate from undefined

locations. Second, it is clear that both adrenergic and

cholinergic agents alter secretion in other epithelial

systems.

Therefore, it seemed reasonable to investigate CSF

production during pharmacological alteration of the adren-

ergic and cholinergic nervous system with agonists and

antagonists and to examine the role of cerebral blood flow

as a possible mediator of effects of these agents on CSF

production. Such an investigation is the subject of this

dissertation.















METHODS


Cats have been used extensively for studies of

neurological and secretary functions in the brain and for

this reason were chosen for investigation of the role of

the autonomic nervous system on CSF production. Animals

weighing 2-4 kg were anesthetized with an intrahepatic

injection of 30 mg/kg sodium pentobarbital and further

maintenance doses were given intravenously as needed

throughout the experiment. Surgical procedures for the

insertion of a tracheal cannula and femoral arterial and

venous catheters were performed. Blood pressure and heart

rate were monitored with a Statham pressure transducer

connected to the femoral artery catheter and recorded on

a Grass polygraph. Body temperature was maintained at

370 C. Throughout each experiment, the animal was arti-

ficially ventilated at a rate set to maintain arterial pH

at approximately 7.35, the normal control val'.::I a cat.

By measuring arterial pH and adjusting the ventilatory

rate as needed, the correct rate was established and the

corresponding end-tidal CO2 was recorded and subsequently

monitored to maintain pH at the desired level.

For ventriculocisternal perfusion, the cat's head

was placed in a standard stereotaxic apparatus. A midline









incision was made from the top of the head to the neck

and a burr hole was drilled at a point 2 mm lateral to

the saggital suture and 2 mm caudal to the coronal suture.

A 22-gauge spinal needle, attached through a Statham

pressure transducer to a Harvard infusion pump, was set in

the right lateral ventricle. Placement was determined by

a sharp drop in the pressure recording as the needle passed

beyond the brain parenchyma into the ventricle. This

needle was used to deliver perfusate and to monitor intra-

ventricular pressure which was recorded throughout the

experiment. The tip of a 19-gauge needle was then placed

in the cisterna magna by separating the neck muscles at

the midline and puncturing the dura mater just below the

occipital condyle of the skull, and effluent was collected

through a short tube attached to the needle. An artificial

CSF (see Appendix II for the composition) was used for the

perfusion. pH of the perfusion fluid was adjusted to 7.4

by bubbling 5% CO2 in 02 through it until the desired pH

was achieved. The perfusion rate was 191 pl/min during

the first 15 minutes and 76.4 pl/min thereafter.

Following a 30 minute washout of the original CSF in the

ventricular system, four to eight control periods of 15

minutes each were collected prior to tests with drugs or

stimulation.

Production of CSF was calculated from the dilution

of blue dextran, M.W. 2 X 106 daltons, (Pharmacia). The









collected effluent of each sampling period and the original

artificial perfusate were diluted (100 pl in a total

volume of 1.0 ml) for measurement of optical densities

(Gilford 2400 spectrophotometer). This dilution was

necessary to obtain a large enough volume to assay during

short sampling periods. In most experiments, perfusion

was continued six hours; none went beyond nine hours.

Fluid formation was calculated with the equation

developed by Heisey et al. (1962) for dilution of non-

diffusible substances:


(Cin Cout) (r)
Vf =
Cout


The difference between the concentration of blue dextran

in the infused buffer solution (Cin) and the effluent

(Cout) divided by the concentration of the effluent is the

dilution factor of the perfusate. The product of the

dilution factor and the rate of infusion (r) then provides

a value for the rate of formation of CSF.



Tests of Autonomic Nervous System
Influence on CSF Production


Stimulation of Cervical Sympathetic Trunks

Bilateral electrical stimulation of the cervical

sympathetic trunk was applied during ventriculocisternal

perfusion in some experiments. In addition to the surgical

procedure described for perfusion experiments, the cervical









sympathetic ganglion was exposed and the two preganglionic

sympathetic fibers were isolated from the vagi. The

preganglionic fibers were then passed through a platinum

electrode which was connected to a stimulator. The nerve

fibers were stimulated with 10 volts at a frequency of 10

shocks per second; each shock lasted for 0.5 msec. Nicti-

tating membrane responses were also monitored during these

experiments to insure that adequate stimulation was

bilaterally applied. This was done by gently pulling each

membrane away from the eye, placing a suture in it and

attaching the suture to a Myograph-B (Narco Biosystems)

transducer which permitted the recording of membrane

responses. The experimental sampling period was 15

minutes in duration, with stimulation continuing throughout

the period. Control samples were collected during two

hours preceding and two hours following stimulation.

Modification of CSF Production by Application of
Pharmacological Agents

Drugs tested to determine a possible autonomic

influence on CSF production were carbamylcholine chloride

(carbachol) (Sigma), phenylephrine HC1 (Sigma), salbutamol

(albuterol in the U.S.) (Schering), atropine methyl

nitrate (Sigma), phentolamine mesylate (Ciba-Geigy),

propranolol HC1 (Ayerst), l-isopropylamino-3(2 thiazoloxy)-

2 propanol HC1 (ITP) (Syntex), aminophylline (theophylline)

(Parke-Davis), verapamil (Knoll Pharmaceutical), and

methazolamide (Lederle). All autonomic agonists were










administered in volumes less than 1 ml; all other drugs

were given in volumes less than 3 ml. These volumes

were considered small enough not to alter significantly

the total volume of blood.

At least one and one-half hours elapsed between

administrations of autonomic agonists when the drugs were

given to test the individual effects. No agonist was

tested during the first 30 minutes after an autonomic

blocking agent was given so as to test the effect of the

blocking agent on CSF formation. Subsequently, agonists

were injected at approximately 30 minute intervals.

The doses of the autonomic agonists used in these

experiments were selected on the basis of the blood

pressure dose-response curves. Doses on either side of

the ED50 for blood pressure and the dose which gave the

maximal blood pressure response were tested for their

effect on CSF production. Salbutamol, however, required

a dose higher than that giving maximal blood pressure

effect to elicit a maximal effect on CSF formation. Doses

of the autonomic antagonists used were chosen for their

effectiveness in blocking peripheral actions of the

agonists.

Cholinergic agonist and antagonist. Carbachol was

chosen as the cholinergic agonist for its specificity at

the cholinergic post-ganglionic effector site and prolonged

half-life in the whole animal compared to acetylcholine.









It was dissolved in normal saline in concentrations up

to 10 pg/ml and administered intravenously in doses of

0.03, 0.1, 0.3, or 3.0 pg/kg body weight. As in all

experiments involving agonists, three 3-minute effluent

samples were collected beginning immediately after

injection of the drug.

Atropine (0.15 mg/kg) was given to test for a direct

effect on CSF formation. During atropine treatment,

carbachol was given in doses of 1.0, 3.0, and 30.0 ug/kg

to test for blockade of cholinergic effects on blood

pressure and CSF production. Phenylephrine (30 pg/kg)

and salbutamol (3 pg/kg) were also administered during

atropine treatment.

Alpha adrenergic agonist and antagonist. Phenyl-

ephrine was chosen as the agonist for these experiments

because of its specificity for alpha adrenergic receptors.

It was dissolved in normal saline in concentrations up

to 300 pg/ml and administered intravenously in single

doses of 3.0, 10.0, 30.0, and 100.0 pg/kg body weight

or infused over a nine minute period (30-35 pg/kg min)

intravenously. Single injections were followed by three

3-minute samples. During infusions 3-minute samples were

also collected. Phentolamine was chosen as the alpha

receptor-blocking agent for its specificity and reversi-

bility. The direct action of phentolamine on CSF formation

was tested at the dose of 2 mg/kg body weight. During









alpha adrenergic blockade, single injections of 30 pg/kg

phenylephrine, 3 ig/kg salbutamol, and 3 pg/kg carbachol

were tested. To further examine the mechanism of action

of phenylephrine, another group of experiments was done

in which the nine minute phenylephrine infusion was

preceded three minutes earlier by an intravenous injection

of 50 pg/kg of hemicholinium-3.

Beta adrenergic agonists and antagonist. To

ascertain specificity of beta adrenergic function both a

betal and a beta2 agonist were tested to determine their

effect on CSF production. The betal agent selected was

l-isopropylamino-3(2 thiazoloxy)-2 propanol HC1 (ITP).

This compound has been shown to be specific for cardio-

acceleration and stimulation of lipolysis without affecting

beta2 actions (Lockwood and Lum, 1974). The drug was

dissolved in normal saline less than five minutes before

use to be certain that the drug was still active, and doses

of 100 and 300 pg/kg body weight were administered intra-

venously. Beta2 activity was determined with salbutamol.

Salbutamol has been shown to have bronchodilator activity

and to increase blood flow to the dog hindlimb more

effectively than isoproterenol, and these effects are

reduced in the presence of propranolol (Cullum et al.,

1969). This drug was dissolved in normal saline in

concentrations up to 200 pg/ml and was given intravenously

in doses of 0.1, 1.0, 3.0, 10.0, and 30.0 pg/kg body

weight. For both agents, cisternal effluent was collected









over three 3-minute periods after drug administration.

The general beta blocking agent propranolol (2.0 mg/kg)

was tested to determine its effect on CSF production and

was given to antagonize beta2 adrenergic effects. Phenyl-

ephrine and carbachol were also tested in the presence of

propranolol.

Non-autonomic agents. Three non-autonomic drugs

were studied with regard to their influence on CSF

production: theophylline, verapamil, and methazolamide.

Theophylline has been shown to stimulate the respiratory

rate, increase urine flow, and stimulate the rate and force

of contraction of the heart. Some of these actions of

theophylline are thought to result from the ability of the

agent to inhibit phosphodiesterase, the enzyme responsible

for cyclic nucleotide hydrolysis (Amer and Kreighbaum,

1975). To test whether autonomic agents might be acting

through changes in cyclic nucleotide levels to affect

CSF production, theophylline was administered intravenously

in the form of aminophylline in doses equivalent to 10 and

20 mg/kg theophylline. Again, CSF perfusate was collected

over three 3-minute periods following drug injection.

Verapamil has been demonstrated to block inward calcium

channels in cardiac muscle (Kohlhardt et al., 1972).

Since calcium influx is important in acetylcholine-

stimulated salivary secretion, verapamil was given via

the perfusate at a concentration of 10-5 M to test its










effect on carbachol activity; three 3-minute sampling

periods followed the administration of carbachol.

Carbonic anhydrase inhibitors, acetazolamide and benzol-

amide, have been shown to reduce CSF formation between 50

and 75% (Davson and Segal, 1970; Oppelt et al., 1964).

The carbonic anhydrase inhibitor, methazolamide was chosen

for its high lipid solubility providing ready access to

the secretary tissues. In these experiments methazolamide

(30 mg/kg) was given intravenously over a period of 30

minutes. CSF production in methazolamide treated cats was

also studied in the presence of atropine (0.15 mg/kg) or

phentolamine (2.0 mg/kg). Samples were collected in ten

minute periods for at least two hours after both drugs

had been administered.



Tests of Autonomic Nervous System
Influence on Blood Flow


Radioactive microspheres were used to measure blood

flow to the choroid plexus, brain, kidney, and biceps

brachii muscle. The kidney and muscle flows were used as

reference data. Briefly, the measurement depends on the

immediate lodging of spheres in capillary.beds propor-

tional to the capillary flow of the tissue. After delivery

of the spheres to the left side of the heart, the spheres

will have the same distribution as the blood leaving the

heart. Simultaneously withdrawing blood and microspheres









into a syringe at a known rate then provides the basis for

estimation of flow in tissues (Wagner et al., 1969). Both

femoral arteries and a femoral vein were catheterized, and

a tracheal cannula was inserted. One of the arterial

catheters was connected to a pump set to withdraw blood

at 2.72 ml/min. The chest was opened at the midline and

the pericardium incised. A catheter was placed in the

left atrium and secured. The animal was then given 3500

units of sodium heparin to prevent clotting of blood. End-

tidal CO2 was carefully monitored during the experiment to

insure the same pCO2 during each injection of microspheres

since brain blood flow is highly sensitive to pCO2 changes.

The microspheres were suspended by the 3 M Company

in 20% dextran in Tween 80@ for injection. Microspheres

15 5 microns in diameter were chosen because this size

has been shown to be optimal for cerebral blood flow

measurements, providing high reproducibility and less than

2% shunting of spheres past the capillary beds (Marcus et

al., 1976). Buckberg et al. (1971) showed a minimum of

400 spheres must be present to measure reliably the blood

flow of a tissue. Therefore, the number of counts per

minute (cpm) per microsphere was determined to estimate

a dose that would satisfy this requirement. In all exper-

iments the cpm in each tissue represented more than the

minimum required number of spheres, thus validating the

calculation of flow from the radioactivity in the tissue.

In each experiment injection of 85Strontium spheres ten










seconds after beginning blood withdrawal gave control

blood flows, and injection of 141Cerium spheres later gave

blood flows during the peak influence of 3 pg/kg carbachol,

30 pg/kg phenylephrine, or 3 pg/kg salbutamol on CSF

production. The drugs were given about five seconds

before the blood withdrawal began. Blood withdrawal was

continued 90 seconds to guarantee at least one circulation

time. The animal was sacrificed and the choroid plexus

of the lateral and fourth ventricles was removed and

weighed. The brain, kidney, and muscle were also taken,

weighed, and ashed. The radioactivity was counted in a

Beckman Biogamma Scintillation Counter.

Blood flow to each of the organs was calculated

from the relation


Q F
d d'

whence,
(F)(d)
(d')


where Q is the blood flow (ml/min), F is the rate at

which the arterial blood is sampled, d is the cpm of the

tissue being determined, and d' is the cpm of the isotope

in the arterial blood sample. To standardize the values,

the blood flows are expressed as ml/gm min considering

tissue weight.









Evaluation of CSF Production Data


The changes in CSF production caused by the various

treatments were calculated in the following manner: For

agonists, a least squares linear regression was applied

to the control CSF productions of each experiment. This

regression was based on CSF formation during one or more

control periods before any treatment and at least three

post-treatment control periods as determined by the return

of blood pressure and heart rate to pre-drug levels. No

values for post-treatment controls were taken sooner than

30 minutes after administration of short-acting agonists

such as carbachol, phenylephrine, and salbutamol. This

line was used to determine the theoretical control CSF

production at any given time over a two or three hour

span of the experiment. The production rate measured over

a three minute period after the agonist was given was

compared with the theoretical control for that time, and

the difference taken as the change in production caused

by the drug. Since the autonomic blocking agents and

methazolamide have relatively long active half-lives,

.control periods were not available after administration of

these drugs. For these experiments, four or more periods

prior to the time the agent was given were used to determine

a least squares regression line and calculate theoretical

control CSF productions. Changes in production were

determined in these experiments as for the agonists.










Statistical Analysis


All the changes in CSF production are expressed as

mean standard error of the mean (SE). Statistical

significance of changes in production caused by several

treatments was established by pairing theoretical control

and experimental data for each experiment and using the

paired one-tailed t-test (Goldstein, 1964). Significance

of changes in CSF production before and after treatment

with blocking agents or methazolamide was determined by

using the unpaired two-tailed t-test (Goldstein, 1964).

To ascertain whether increasing doses of autonomic

agonists yielded a dose-response relationship, the indi-

vidual responses for each experiment were converted to

probits, a measure of the per cent of maximum response

obtained. A linear regression was then performed and

a plot of log dose vs. probit was produced with a 95%

confidence interval on the regression line. Significance

of the line was established by testing its slope against

a null hypothesis (Snedecor and Cochran, 1967). In all

cases, significance was assumed when the t-value exceeded

the minimum value for p<.05. Non-significance is denoted

by N.S. in the data.















RESULTS


Ventriculocisternal Perfusion Studies
for Estimation of CSF Production Rate


Control Experiments

Control data were collected in four experiments in

which no drugs were given. Control data from 24 exper-

iments in which only short-acting autonomic agonists such

as carbachol, phenylephrine, and salbutamol were given

were added to these, avoiding data collected within 30

minutes after drug. The two groups of data were combined

since the slopes of the 4 untreated (-0.042 ul/min each

minute) and 24 treated animals (-0.015 pl/min each minute)

were not statistically significant from each other or

different from zero when tested against a null hypothesis.

These combined data, shown in Table 1 and Figure 1,

indicate a steep increase in the optical density of the

CSF perfusate reflected as a decrease in CSF production

during the first 60 minutes of the experiment according

to a regression line with a slope of -0.143 il/min per

minute. This initial apparent decline in fluid formation

was probably due to an incomplete washout of the original

CSF in the ventricular system. A more gradual decline

follows ranging from 19.7 1.0 to 15.8 0.7 pl/min over









TABLE 1
CONTROL PRODUCTION OF CEREBROSPINAL FLUID

Time (min)a n Production (pl/min) SE

30 24 24.0 1.5

45 21 20.7 1.3

60 27 19.7 1.0

75 10 17.5 1.2

90 8 19.7 1.2

105 6 16.9 1.6

120 17 17.0 0.8

135 22 16.2 0.7

150 27 16.8 0.7

165 24 16.5 0.7

180 28 16.7 0.6

195 11 17.2 0.7

210 7 15.6 1.0

225 14 15.1 0.9

240 24 15.9 0.7

255 20 15.1 0.7

270 25 14.9 0.6

285 22 15.6 0.5

300 19 15.8 0.7

Regression lin6 for 0 to 60 minutes
y = -0.143 x + 27.9
r = -0.96

Regression line for 60 to 300 minutes
y = -0.015 x + 19.3
r = -0.81

aDuration of perfusion after washout of original CSF.




























r.d


o0)1
4- 4) *H00)1
4-) l u
p4--i r (1
-H 0)OH-I-3



-P rd r

to i (0 a) 0 r:
rq C: -14E: a



r4 0 U) 0
0 0 4-4 loa)


W ) 4J 0 0-
0 C: U) r () 4-) -
-H E-1 '-
*4J +1 0 C)
04 () %-0 -H
rrd U) ~-d 0-i



04 (1) W
F4 M En~






a)--H Z 4J () 0) >-i
X: $-4 0 (0d4-'4-'
4-P0) -H $
04 -P (1O : :V
X 0 4-i -H- -H 0)
-H- Q) : r. EF 4.)
U) 05 ~cO 0)
m 'A 0 r. o o (1)








0) 0 -4 -H- -H D U)

tP r4 04 4i (1
U QU a0)0 p
(TJ : FZ4 0 E -W 04







0 X'r- (1
4J C ) r












r4~
FT4m














-4-r


(u!w/lId) uoflonpOld JSO


ie


















E
In
9-
E

"4
3

3)








the next 240 minutes in which the slope of the regression

line is -0.015 pl/min each minute. These data compare

favorably with the accepted control CSF production value

of approximately 20 pl/min (Davson, 1967) and with the

slope of -0.025 pl/min per minute for a regression line

on CSF formation in the monkey (Martins et al., 1974).

Cervical Sympathetic Stimulation

Only those experiments in which successful bilateral

stimulation occurred were included in the study. This was

determined by recording the contraction of the nictitating

membrane. After an initial decrease in the response of the

membrane during stimulation, the tension caused by the

contraction remained steady, and no fatigue was observed

during the experiment. Preganglionic stimulation of 10

volts at a frequency of 10 shocks/sec each with a duration

of 0.5 msec decreased CSF production 3.0 0.6 pl/min in

five animals (p<.05 in paired t-test).

Cholinergic Agonist and Antagonist

Carbachol significantly increased the rate of CSF'

formation to 10.6 2.1 pl/min over control (Table 2,

Figure 2). Over a dosage range of 0.3 pg/kg to 3.0 pg/kg

probit analysis (Figure 3) yielded a significant slope of

1.52 (p<.05). These relationships among the CSF

production data were taken to indicate a carbachol dose-

response. The ED50 for carbachol taken from the probit

plot was 0.12 pg/kg. High doses of this cholinergic

agonist occasionally produced slight increases in intra-

ventricular pressure.

























C,4
0

~U)
*d









-4
O




O -
- C

H4




4-)
- +l




























0i
u-
O




O

C-


0)
On
-
Ot
U

c






0
04-










0
^ +
CL|
h'?
CO >-
u
r- -
0 3


S H -. C. 4 o ,
D0 in Ln (nm LO

o C 4f r-i C )o L


+1 +1 +1 +1 +1 +1 +1 +1

o co oD kD m r- 'm
(N N 0 0 0o N t
+ + + + I + H-
+ +















(N r-- r.-4 o Lr) -
o o .
(N ) 4 r H N 0H 4 C
+1 +1 +1 +1 +1 +1 +1 +1

om o- I co co H o o

H N Hr- i r












CO 0 0 .0 X) .0 0
o C m C o A) C)

o o 0 CO H m 0 n
CO


H
r-4 a)

0
04 0 (
0
0 0 H.

Sr
0 0



SU)
i 0 (
0 4
E U 0










m)

H 04 (




4-1



U C ) 4
4 9
o e e








0 i

> I V 0
*H 0 )


0
OH
H
U









H
0








rn

.

u



W
U)

0










u)
0
CQ







0





U


,-1
0
O4
4-P

C u

UO
(U

0
0 4J
)4-






0 --
40
l Ut
r. c

0 Q)





O0 (U
-H 4-4 4-1
4J4-4 0
oU u
'.0 0
0 U ()

044- 4-

U) r
Ua


-HQU)
Ci 4-)
UU n

UULrd

























0)
E-4 0 () LW 4-1
-H4- 0 F0
S000-
S*00 00
UOQ) oO0

0 + 0 -4
0 04 U)rd *H
ro pq* +1 4-) 4+) -

SU 0) M 0 0
04 rd r-i 4


C0 0r0 C -
U r-i -H U d
0 a) IQC -

Q) r 0 Q) >i
S*r- 0 t3 m
- E-4 0 O


0) -H (U c H
k 4- r-4\Nl -


0 ) C (0C
ro 0) ,3 -i -Hi 0 fd





,Q -r4 4 kQ-M Q Ln
-d 4C1 0 H M n 0
r4 aJ r r E r1

u d 40 r0- U -H- U g
rda c cdk-4C(
A-r u- (dC ) (d ~4r

U 0000 -400


CN




-1





































Q.
-
O


UW lD UQ d S


(uLILU/Ir) uo!,nPojd jSO ul aseaj0ul


N 4




























.d rd
4- 0 E-4 (

0 U r 4-d Hr.


Ef) 0 (1) a) 0 -H- r,

r-4 4- IJ 4J 4- 4-H U
noO ( 0 r.
M rl C wi C4 -I


0 4)-4() M "4


) H 04LOC 4-4 U)
,Q En til 0 4 r4
0 NH-Hq 0
rtn OW ~

(d 0 -rq (Ij )l l




00 q)r ( -
4 0
~H-E-1Q a)N0
(0 I0 c Ln

0O *.( ro0-H



-t4 -14 0U
U 4- a4 0) (1)k


z 0 rd ( U)

4 m0U E-r

rO 4J "4 J--H~ 0
-HO U EI 0 $24
4-41) a) 4-r4 En
S4-4 4-)4J UM 0
0 4-4 rd p -H- cn tn

HU (0 $-4 0
O-r Q) (U t C ty) v U-)

dp~r a ~ i-nv
un C'4 k- O() 4C
) 4-) 00 rl- -









r14















































sj!qoJd









Atropine caused a decrease of 2.6 0.5 pl/min in

CSF production which was statistically significant

(Table 3). The cholinergic blocking agent also prevented

the increase stimulated by 3 pg/kg carbachol (Table 2).

The 30 pg/kg carbachol dose increased CSF production

during atropine blockade by 2.7 1.0 pl/min. The 3 pg/kg

dose of carbachol was administered during phentolamine

and propranolol blockade of adrenergic receptors in one

or two experiments. No alteration in the effect of

carbachol was evident. When verapamil (10-5 M) was

present in the artificial CSF perfusate, there was also

no change in the response to this dose of carbachol

(Table 2). Verapamil alone did not alter normal CSF

production.

Alpha Adrenergic Agonist and Antagonist

Phenylephrine caused a dose-dependent increase in

CSF production (Table 4, Figure 4). The change in response

with increasing amounts of drug was subjected to probit

analysis (Figure 5). A slope of 3.41 was obtained from

the plot indicating significance (p<.05). The ED50 for

phenylephrine was 12 pg/kg. A slight increase in intra-

ventricular pressure sometimes occurred with high doses

of phenylephrine.

Phentolamine decreased production 3.7 0.8 pl/min

(p<.05) (Table 3). This alpha adrenergic blocking agent

also significantly reduced the effect of 30 jg/kg phenyl-

ephrine, i.e., from 10.0 1.6 pl/min to -0.8 0.7 pl/min

















O
*r-
0

-1

o

C~
*(4


0




- -
r14 +1









U)






-i
u -







0 -




4-)
U





O ;3
(0







u
S-p
0
C
0











U *
U


z
0
H

13

U
0







O

E-4








0
0
U







O
U





E


0



E
M






U
U
1-1


0~
O 3 0
C 4
l 0
) 4-1-4
0 ,Q
io







Uo O
4- 4 0
Sr-C )
3 (U U








0- 0
CU Qp
O (






U) 0
r. 4


C O


M u



U a -4
(ri


-4-)

41
0)
4-J





C)
-H
a






o

v

-p
LA




.0

















0
*H
-P
0-. ro rd
rO -) L- %l* )
in in i-n i
0
r4 U) + N D I'D LHO r-r- N
F14 +1 14 C-

O U- +1 +1 +1 +1 +1 +1
H
E-4 r- o kD co t
U *l .
03 \ O oo o o
0 0) r- + + r- r- I +
0 ti)A + +

r T



) U
CO








0




20 0 m co 0 C
Z WW+1
OA1 "r 3W




1 ct +1 +1 + 1 +i + I +I

Sn r-I
S r-iH -i .0 'N 0 in Ln
( 0- H IN L H H Hr
O k
0 O
S 4-)

w 0
Z U
H








Z4 tp

r-H m C m (Yr
i-q
U)
0
C)


0
0

) a0)




0 0 04 0)
0 4) D 4
(0 4J

0- C

)C >r am

0)0) W 4J

-r-i 40 4 in -


l ,Q.. o -4
o *0


-lj O E r
OO3 U

rC J .P 4- -H






,Ct o a 0
U 0. ~ O
0 0)(1) Q04
- 4 4- 0 C0 o



U d HC -C 0

oH a) H-H

) -4 0 0 0

S0) 0) 0)
M0 ,4'U > > v
,C X C -4 -
Cd 0 4 0 '
































Figure 4. Phenylephrine dose-response for CSF production.
The ordinate is the mean increase in CSF pro-
duction-over the theoretical control value
SE plotted on a linear scale. The abcissa is
the dose of phenylephrine in pg/kg body weight
plotted on a logarithmic scale. (----- -)
represents phenylephrine alone. (0) is
30 pg/kg phenylephrine during 2.0 mg/kg phen-
tolamine. (A) is 30 pg/kg phenylephrine
during 0.15 mg/kg atropine.





40













12



10




fO



-r





c2
I





--Atropine
0










SPhentolamine








Phenylephrine (pg/kg)
































Figure 5.


95% confidence interval on probit analysis of
the effect of phenylephrine on CSF production.
The ordinate is probits. The abcissa is the dose
of phenylephrine in pg/kg body weight plotted on
a logarithmic scale. The slope of the probit
regression line is 3.41 which is significant
(p<.05). The dose of phenylephrine producing an
ED50 response is 12 pg/kg.












































Phenylephrine (pg/kg)









(p<.01) (Table 4, Figure 4). Atropine, the cholinergic

blocking agent, also effectively prevented the increase

in CSF production caused by phenylephrine (Table 4,

Figure 4). In the presence of atropine, 30 pg/kg

phenylephrine increased CSF formation only 0.4 1.2

pl/min which differed significantly (p<.01) from the

increase due to the alpha agonist alone. Propranolol

did not influence the effect of phenylephrine in the only

experiment in which it was tested.

Phenylephrine was also administered in an intravenous

infusion (approximately 30 pg/kg min) over nine minutes.

In the first three minutes, CSF production increased

14.1 4.0 pl/min, 24.1 4.5 pl/min in the next three

minutes, and 18.6 2.0 pl/min in the final three minutes

(Table 5). Administration of 50 pg/kg hemicholinium-3

to the animal three minutes prior to the phenylephrine

infusion significantly reduced the increase in production

in four of six cats. In these four animals, the increase

in production due to phenylephrine was limited to 2.2

1.0 pl/min (p<.05 compared with phenylephrine alone) in

the first period, 2.5 1.6 pl/min (p<.01) in the next

three minutes, and 0.8 1.8 pl/min (p<.01) in the final

three minutes (Table 5). When the two non-responsive

animals are included, the effect of hemicholinium

pretreatment appears to cause no significant change in

the effect of phenylephrine (Table 5). Hemicholinium

alone did not influence the rate of CSF production.

















4C)
3
*r-l



z
0
H -

E-




H





Z
z ^
0 .H
---


I 4
U
0-0
Lf
SZ
M I 0

m D U)


HZ

0

HO
Z




0 E-H
U
MH

M
wa
m
E-


+I
o
0


+



















4-1
+I











+1
0

+


-I-




4) C 0

o -H


4- 10
r O -H



C r-4

0 "
4-J o- a)
C) Q4 0 0
r r-o v- (
S0
4 ) 0
< -4 En C r-i

0CHO E n 0
o "-i 0- 0
4-1 -H 4 4-4 4-14 *H

) 0 C) C C)

0 u 0 C -

4- > O


or 3 o o
v) r0 > ( 0) r-V
U a e *H r r >-
-d 0( rd ( )


4-J 4J ty) 41 4C 04
0i 0. A *d 4
*r-l r, t 4J)
U C ) Q a) 3



r0 0 l in (f (d ra

0 r, -AH Y) 0 0 Cd
0 1 u 0 Q4

44- -l %






M 4-J ) v v
xi ) C)
(0 ,Q T0 ) 4-


0C d
Or Q) U
0 I 0+ 4- 0X ,I 4-

u -rA U O I N
4- C 4-) C) 4- Q)

rcj -q U 5 r M a -+i r+


04 C) C u


U 0 U 0 U -H


C)-HO C 4* H CU-H
0 0 a) 4- 0 ) m


U 0 Q) ) 0

r-4 rdW 4|4 rCdr
Uc c: a u a-1


U-4 U) r: U 4r










Beta Adrenergic Agonists and Antagonist

The beta2 agonist, salbutamol, increased CSF

production about half as much as carbachol or phenylephrine.

(Table 6, Figure 6). A dose-response effect was demon-

strated by a probit analysis of the changes in CSF

formation (Figure 7). The regression line had a signif-

icant slope of 0.96 (p<.05). The ED50 for salbutamol

was 1.5 pg/kg.

Beta blockade with propranolol did not change CSF

production (Table 3). Propranolol did appear to reduce

the increase observed with 3.0 pg/kg salbutamol, but the

difference was not significant (Table 6, Figure 6).

Atropine (four experiments) and phentolamine (one exper-

iment) did not appear to alter the salbutamol response.

The betal agonist, ITP, did not significantly change the

production of CSF in either 100 or 300 pg/kg doses.

Theophylline

Theophylline significantly increased CSF formation

(Table 7). Doses of 10 and 20 mg/kg increased production

3.3 0.8 pl/min and 9.2 1.4 pl/min, respectively. The

difference (p<.01) in the increases with the two doses

suggest the effect was probably dose-dependent.

Carbonic Anhydrase Inhibition

The carbonic anhydrase inhibitor, methazolamide,

significantly reduced the rate of CSF formation (Table 8).

The maximal decrease was 11.2 1.4 pl/min; over several





















o
0
-IH

0-





U)
r C








.-)


2
H
E-
U

OC
0



OU























0
U













0

o)

H


z




pqr
1,

CO






:









Qi


0o


Q )



LI) I 'D (N -V N 0

C
+1 +1 + + +1 +1 +1

Lo m CC o Nr r, 0N

+ ++ + + + +





























m( 4
U C H n E- X N o m o

(N (N H- H H H













o o a 0 o C
'-4 0 0 0 D0 0 0
0 .0
o r-i m 0 m m
Hr- (


H H
o l


+1 +1



I +







0 0











+1 +t

o L
S.O
















Co C
o o


I (U
kC)
a ^!
04
x ra

Cd

rOH

CdO
4-)
00


rJcl 1
tRO




r(C)

Cdc
a)

C4J 4


w-P
^ aa
ro o)










04-H
ro 4-1








2 0
rn (
U 4-U







U )1

t ,
t C C
C)- C)H


U- 04^




cd
U (U Q)
0 ^


Ho 0
r-H C) U (d
0 r _

04 4 r- o




0 0 0 .
S 04
v n


tn >14
o -H a *

0 e0
N
oC) Xtn


S 0(
S,-I 4-4
4J* 4d


$ I C)


(N CN P fa
r-l r-l 04



04 E



> -,Ia U


U U H Z


r.
0

4-)


O
0

0









O r
kU
r-\ 1-
0 3
H -
-P
























.0 tO
E-1 5: O in u) x



-00 d0 0 m.


: 04 Q a) a, n I
0 r14 +1 4) -
4 J *Ha,
4 E 0 --H

U > )0 EC 4 0
a, cnA 0
a) E n *t*
4 U) r-f H n ri .
>0 0 0n m 0rl
$ ,- 4 4 ., rJr 1 '
S 4- (d, 04.- )
U, >4 0


0 0)0 'r
06 .C -4 0
co ro o -4 Co

I) Q) a, C 0 rd nr -
0, O O O\ .

4 ) r4C r-4 C t

0 a, 0a 0) 0 4
ro -H co r tI
0 a 0 0 fr4 0 0)



i-i (^ 0 < f) *Q r i
I 4l ) r -I (I) 7> C-A

0 O ul O -m Ul) &o



to



44 4
,-4
-H































O
-a
0
.--
0 C-
%- 0
4 ------- 0


E

.0
=3
-o


(u!w/Irf) uo! pnpoJd ASO u! aseaJuli


1 1
























4J
a -H
0 .Q
4- -4 0 C
0 +-I -4J d
.c 0
o -H ((-
-HO 4-.J
Mo (d tP Hc r,
P o a a) u c4



0 -Hi 0 I0

r- U 4 n 0
C TO m 4 o 0


O~C rO -HI0



r 04 0 t

0 U2 0 ( -



OUQ H .0- MO
oH N. On

OU 0 C -)o 0








d0 W0 o
> (d cH 4-1
C 4- 4 H Cdt 0 0 *O


Hc99 O 0u0
CO OH n 4a


0 Cd 0 0 '
.4-J 4 rH) M O
Z4r-d 0 0 ,
(a r:, r0) 0 cl

4-1 f H .n 0
> ( 4-4' 0 C U.-W

r- 0 0a) UL 4 -
oQ) 4?.IH- 4v Lr.

4-1 0 -,i *-14 o
C 44 O H 0l M-^ (
0 44- 0 r-l U) E-I C
) 0 0 C O
om OOMm tOn iv
M 4J P0 (4-O








r4
Q)e




























































. 40 in 4 V -4


sliqoJd












4 -H



0 LO m
U) +1 +1
tn +1 In in
-P -I
c nm m
(l04 + +



O^
00 -:
r4 r JJ (I

0 0
Sa) Q ,


^ ir5 .' i*
4 *H 3 U) +1 +1 -
w + r0J 4
I I I U r o


oQ) l: ul -
z W) U) -- P
o o ,-I
0. 0 a) 4J
N H 'N, .HL4 -1 I|
H -- 1 4- 4 '4
0 1 to. o o
w UUT O 0 W C Q )

H 0OM *-4 O
ED r 4J r0+ Ll
. c -u 4. m
0 -UU 0 a) Q) O. U4


1 C 0 o+ +1 + E) -4 o

U 4 U 44 0
OH -- U*D H) 0 ( 4




O ,- *' m *HQ
4 r + + E a)
M fH ." 0 0
t tn 7 .

0H 0 0


0 0
H 00+ .
E-4 U 0 +1 CM r--
*--
S-1 4-J +1 +1
0 0 U
4 J -H 00 0)
(n 4-Y rO E
E CO rO t o
U O H H- H
0 4 r- r
u -'



HH (

0 ZTI r















-,-Q

OW

C -
0 M

-4 -H
C U +1










rCd
O---









N m 0

dO

Ut
0 -
(0

C 6







0 -H
e
O e












0











-4
0

U
3 m

O +1




U \

0 -



U


H








OU
E- 1

0 u

SP4
H

H U)
mU
H


rz
Z







UC
>4




HO

C m

u U


E Z

00
U
u
440


(N

r-

+1


















+1

N


I














+I


0'
*H











o
.H




rz







0


N
ru

lc
Cd


4-I

l


*


+1








I





*
r+1







+1

0
*
m
H






















+1
rW








-H -







rVd
a

H H



00
N-)
rdC
Ca

)f


-0 OO
O >




U U
a) o
4- 4-4 4-
4-4 0 4-4 (

.I 4- 0l 4-)
U






(0 a)
a) a) 4-
4- 4-1 4-1










U)0 En0 U

0 0 3
WQ) Q-)J-







a) -Q) o


C)! C C(

0) H 0) 0o
Q) 4J Q) 4-) 0



U 4Q Ua4


u 4 u'q 4
fri IQ


a)
--




C)

.4


0
aU
o










*rc

H,

0
N

0)
a







C)
m*
(1


-A
r4

rd






N
0
r0







a,
H
c0



04


r1
0
N











4d
4-)






4--4






a)

C)


Q)

rl

0
'H
4-)
0
rd






al

C
U)









collection periods the mean drop in production was

9.4 1.2 pl/min. The CSF formation rate during metha-

zolamide treatment was approximately 7 to 9 pl/min. Since

atropine and phentolamine successfully decreased fluid

formation, these drugs were tested in combination with

methazolamide for possible additive effects. From Table 8

it is evident that the autonomic blocking agents caused no

further significant decrease in production. It was also

noted in several experiments that carbachol and phenyl-

ephrine partially override the carbonic anhydrase inhi-

bition.



Changes in Blood Pressure and Heart Rate
during Ventriculocisternal Perfusion Studies


Cervical Sympathetic Stimulation

Bilateral preganglionic cervical sympathetic

stimulation (10 volts, 10 shocks/sec, each shock lasting

0.5 msec) produced an increase in mean arterial blood

pressure of 42 27 mm Hg. Heart rate did not change

significantly.

Cholinergic Agonist and Antagonist

Doses of carbachol ranging from 0.03 to 3.0 ug/kg

were administered. From Table 9 and Figure 8 it is clear

that the drug had a dose-response effect in decreasing

blood pressure (p<.01 by probit analysis). The heart rate

also decreased beginning at the 3 pg/kg dose (Table 9).

Atropine caused an initial drop of 43 5 mm Hg in mean







































C)

pO

0
0



S0



H Z
O











CO




0


0
U


( --o
M u
4-J U)
4 I 't o Ln c T C' Li
r0 +1



-H n






---. --4 e -- oo --n o- t ,-s
N 00M C"o D H 0 -N C-

Q U) I '- i o -
-,--I +I +I +I +I +I +I +I +1


4J I
a 4 -t I I I
C r
cdQ)
U '-










0a
ul
i)n





0l m 0 -'
w 1o m c

+I +I +1 +I +I +i +I +I
= +i
3 co o r ,-I % o 0 CNq 0
(1)- CN in Co i I H iC)
g I I I I + I

C




C-)
U













0r
0
I I


0
a) r-

r--
,-d

O 0 O


Cl -4


4- C)




r- -4 o tr
o 3
m~ o

4 a )

0 E p
Li ...


SCc 0 (



*^ 0
o EI c
) 0 0

E E






I v
o a


C 1 0

Hi ,.Q C) .























Q)

a) 0


~c)C r~ C)
E-1 >1U ca
10 rd Q)

M)E- 1 a

C4 ) ,-4C
04OU rd a
(rn 04 04

mk o ro r -
040 (.) r-: 0 r -r

0 En -H -


o ro -i0 U -
z -li 4-)

(d C) O10



4 0rP U)
Qardr u r04

4-l.4 -P0
k E q 0 I




En) 0 44 0 Q
0 rl rd r-4
IC)c-I cn1E


a) 4J 0 r-: Z
o 0 )40Q) t
I -4 CJ2 m -
rC) 04(L r+ a) -

O W 4- 4 l
UrcEnfl4) a

+ cn, c C) r-
0 (1) 04 In 03
U Mn tT) En 4-)



0-H ty) Q

Cu 0- rco d




C1)



rX4



















































(21H /lWW) ajnssaJd poolI uea8 ul aSueio


,L -L









arterial blood pressure and a decrease in heart rate of

17 3 beats/min (Table 10). Since both of these

parameters had returned to pre-drug levels within 30

minutes after atropine administration, these actions

were thought to be unrelated to the anti-cholinergic

action of the drug. The atropine successfully blocked the

fall in blood pressure seen with both 1 and 3 pg/kg

carbachol and the decrease in heart rate seen with 3 pg/kg

carbachol (Table 9). After atropine, 30 pg/kg carbachol

produced a slight, but insignificant, increase in blood

pressure, possibly through nicotinic stimulation of

sympathetic ganglia. No change in heart rate was observed

with this dose of carbachol during atropine treatment.

The cardiovascular effects of 3 pg/kg were not affected

by the presence of verapamil in the CSF perfusate (Table 9).

Alpha Adrenergic Agonist and Antagonist

Phenylephrine, given in doses between 3 and 100 pg/kg

body weight, increased blood pressure to a peak of 133 11

mm Hg (Table 11 and Figure 9). Probit analysis yielded a

statistically significant slope (p<.01) and an ED50 of 17

pg/kg. Although the lower doses of phenylephrine (3 and

10 pg/kg) caused a significant decrease in heart rate,

the larger amounts of drug caused no significant change

(Table 11). Phentolamine caused a decrease in mean blood

pressure of 40 7 mm Hg and an increase in heart rate

of 24 5 beats/min (Table 10). Both blood pressure and

heart rate returned to control values within 30 minutes.


























H












0




0
0


o 0





0
H a







E -
H

U



u


0
U
H

O

O



0




rI)
Er


rd +1


0U)
*HE


















04


0


S+1





a
c

F-










U)












0
Q)
[/


























-rf
(U'-


-l
m Vi -

+1 +1 +1

--
1 +1










0
.4















,-)
U

a,


4-)






m N (M u)















0 U0


0 0














04 Op
a,

-4 0


ao









4 7
Q) (0 0


C *1
*H TC
S-I-r s-
o c Q,
>-1 (U 0
aPx;k -
< i ^



















Ud'-

4-1


fl)
(0 +i





Oci
Cdr
H e


U









U,
M




0


+1C
Cd





0)

















O
ci)












(Um
Cn


-. -, -o '
co ,--I m r- H, .o N-



r-l

+1 +1 +1 +1 +1 +1


I H + r- I re
I I I















-- -. 0U
S L LO f rO -
O-- --- --**

O HP H H LN





+ + + o r- + +



















n o o o o o
H-i (Y C ( n
r-I


0)


c O
0



4-H 0
0





O 4 049
4 a)
0 1

Q)H 0
o Q4 Q) F-:




On 0 \

o 1 a




04
ON \
tr E


o H r







S-0




o o H
(N ON 0 Cd


-H *H 0) 0

p r Q

o :

0) (U

-r4 -4

Cd .Q U r)























a)


4 U >i 0
In ia 0 r 42;
U) Q) 4r ra)r4
U)l a ) ~rr 4Q
a) r-44a)>1M
-4 rO0>1 -H 4- )r
04 0 a) r a U
0 $: a) a -HH
'roI -r-H. 0) 04 04
O0'-4 04 0 0 -0
0 -14 Un p
A (4-4 Q, -S4 4J
.Qd 0 m )N-

a)r~ -H )N
p 0a) rr l
-rq (1 ro c CL


a) 4IJ a)a) -
U) 4)4 Q)4-) M)
$: 0 -P 0 -H (d )
o zrHq-w0H r-r4
Q4 M .4 fr-4--~0
tn1 Ca P4 W C)


(1) U) 0:
a) U ) 4-J Q)

a) Ml5.1cH rl
(J) Cj+1U) U) 014H
U) 41 U) -j [!.
0 -~-ri Q)4- -. ::
r 5 En 1j) (1) ro r




>1~Ua >1 t) 0
-Hr.0 rH E r-


4 F -, r-4:
04-Ha) Ui QP C)04
P PL4 'rI T



aW AO 4 a)
,-i 4 : r -q

0in a a a)




a))
'-4
Q, ro 0.ka






U-i

-H

































c>

E

C 0
oo
Q --
< S


(RH ww) ajnssa


8 8

Ad po018 ueaAI ul aSue43


"1



a)
. C

CL
a)
' C
a)

0.

a.


I I I I 1









This blocking agent significantly reduced the pressor

action of 30 pg/kg phenylephrine (Table 11). In the

presence of atropine the phenylephrine-induced changes

in blood pressure and heart rate were not altered (Table

11). Phenylephrine administered in an infusion (30 pg/kg

min) over nine minutes increased mean arterial pressure

by about 66 mm Hg (Table 12). Heart rate decreased

significantly only in the first three minute period.

Pretreatment with hemicholinium-3 did not significantly

affect the increase in blood pressure or the change in

heart rate due to phenylephrine (Table 12).

Beta Adrenergic Agonists and Antagonist

Salbutamol was given in doses between 0.1 and

30.0 pg/kg body weight. The blood pressure decreased

significantly in response to all doses tested; however,

the maximum fall occurred after 1.0 pg/kg (Table 13 and

Figure 10). A dose-response increase in heart rate

occurred as determined by probit analysis. Propranolol

dropped the mean arterial blood pressure slightly and

significantly decreased the heart rate immediately after

administration (Table 10). Unlike the other autonomic

blocking agents, these effects of propranolol persisted

for two or more hours. Propranolol successfully blocked

the cardiovascular responses to 3 pg/kg salbutamol

(Table 13). Atropine did not significantly alter the

changes in mean blood pressure and heart rate caused by






63








U N
S+
+ +
+1 0 +1

a U o IO
+J +





a, +1 +1 +4 i
-,a




+ 0
Z -.44-


C c ) 0
O 4 -
O E En
04 En U




m -)n a) 4i-

I O ) +1 +-I r*H




o +Oo + +
P-1 U+ I a)H 4
Q q +






O _____ ___________


HN U4 -H -









O-g c f: E
H C nU U 3
0 L. l r0 C) CU)
c 5 -; : Ln j rd
U )0 ( +1 ++1 %- r. C s
H, + 4- C4 (C 0d
Zw I m CHo 04 4
H C> U .D m a E
Q H + I >-, 0



0H W q0 U
HP m 4 a, c,
r+ 0 + li, +1 'IT a .


0 a4 PC 4 -
0 : + 4 a)

U U


S---a, -

m CL4 O n t




1 0 +41 +1
u> CDoe
0 6
--i
PQ0
























Do ao ) N a


M Y) Lii .o Co -- L

+1 +1 +1 +1 +1 +1 +1

m r- n Om O0 r-I oo0
+ n m cn Ln + H-q
+ + + + +


H





CE








a4





0
0
O












E-i
0








H
0

z
0









Ul
H
Ul












z
E M






Q4














U3









-) U
Cd


nj --
(d +1





fi l
a,






U-







U)



O


O ul
a,














k

O-
0r




uH






C)











0
c!+


LO -'I
O H

,-+
+1 +1

H o
N- 0
+ +


Lin
--s ---


to Li
-- v




+1 +1(

I I



















o o

H m


a
0
,--


0 ,-I r-i
r-H Q) U 0
0 r:4

,-4 0 0 0

0 4 --
o i Cd a,
C CiQl 1d










L O r--I
0 4 a 4 r(

S0 u 0
0-1 i r-H

.X N
t0 E |I i 0
Li >i (d
C0 r-l X 0 4J

N 0 3

C o 0i-l H

4.) .c$l H 0

0 r-
4 IN a, (a
a) 0) n3 4J U)









4> 4> -H v -H
4- -H *
fd 4- 4r1
C C C4


I 04 a,(
o o 0 e $
(N MN r 0 Cd




A >1 .;r:
41 4.] 04i C


040 C)

> > H-i v 0 O


Cd a U '0 a,


a,
0



C L oo r-n Ln Li N H--I
E
D +1 +1 +1 +1 +1 +1 +1

r-l C rn ir) .) C -i.
m
-1 oN o ci t t -


H 0 0 0 0 0 0
H 0 -
,-- (
























r--
0
HO r
0 ro


4 ro >1 C )
In O aj :j -l r-I m
a, 0 C 4 4 rO (0 0
Q) 0W,0 -4( 0E

a, 0d H W u, 2
4r-4 H CJ : 4J .114J)

1 a, -1




O dro 0
o4 -r a,)r r )r Lt
.0 u,0a, o Or
o -P (,) m 0 r-I
44Q) P 1)4J -r4 O0-
a, cH4JC oa, E:
cM04 r (

r.4 U)2 04 0 0 Q

m a, rd k
aC,+Ic( H ty 0 ro
44-) U) 0

c, o o u 3d oX

O rd>1 rd
0 :i.0 to --
aOHl
Hl rld-.dQrd 0.

E*Ha, -

(U O 4 -k4 t



P2 04W- tfl'-'d







rJ4



























z
0
C

0
0
a.
0~


B OP0u


(BH LWW) a~JflSJd P0018 ueInJ U! a~e4


0



C, )









salbutamol. The betaI agonist, ITP, increased heart

rate at 100 and 300 ug/kg doses (Table 13). A small

decrease in mean arterial blood pressure was also

observed.

Non-autonomic Drugs

Theophylline given intravenously significantly

decreased mean arterial blood pressure and increased the

heart rate (Table 7). In both cases the effects appeared

to be dose-related. The dose of 10 mg/kg decreased blood

pressure 37 6 mm Hg and increased heart rate 25 5

beats/min, whereas 20 mg/kg decreased blood pressure

54 7 mm Hg and increased heart rate 35 3 beats/min.

Verapamil in the perfusate did not influence peripheral

blood pressure or heart rate. Methazolamide also had

no significant peripheral cardiovascular effects.



Effect of Autonomic Agonists on Blood Flow


One group of three cats served as an internal

control; i.e., each animal received two control injections

of microspheres sequentially. In these animals, no

changes in blood flow were detected in any of the four

tissues (Table 14). Other groups of cats received

30 pg/kg phenylephrine, 3.0 pg/kg carbachol, or 3.0 pg/kg

salbutamol as the experimental treatment during the second

injection of microspheres. The control blood flows

(ml/gm min) for these animals were 2.26 .28 for the






















0


O


Om
Oz
0



QH
0





E Q
40
0



Uz



mH-
E4 U






O

H
Z



U
Q E-




ZM


H



O -




u
P
S0
A 0




a
E-4Q


4-J
















0U 6 r 00 oo i co o ou
.1 a 00 r* 0 LO
t r-I n r-i O m
0 4-) ) I l I

UC
,-I 4 : rP 're) r-'a ,--i -a rcr







U4-) J C) 00 CH C C CI CI




-,-i

CO
C) ( coi Hr oicn OT H o c
H O-4 H N H
q) 0 00 r 00 r". (N r00 M 00' 00 00
r-I 41 N 0 r-tin r-o N1 CL O
SCO oo oo oo oo o
*e OH *.







-, + -I +
1-1




C)
00 $-l








4 0H
tP-l 041 CO C ,-- D 0 O
r..a) O0






( ) + I I e
SOH




-HC 5 O Hm ro uoe



O N r mm Nm m) N Z
U r










0 O






CH H I co r 0O
5O 0, O HN 4 U
40 0 4 C|1 OI



-O -
S 0 0u
0 4J :1 0) 'T r CN rQ re) N N

u 0 r-I Ci



UU U) r U
^j--q


4.




a,


a


43
In
















0
S












0
--
0







r-O
in













10





69



choroid plexus, 0.46 .04 for the brain, .048 .008 for

the muscle, and 1.64 .19 for the kidney. None of the

drugs caused a significant change in blood flow to any of

the tissues examined (Table 14).















DISCUSSION


Modification of CSF Production by Autonomic Influences


The data show significant increases in CSF production

as a result of stimulation of the cholinergic system by

pharmacological agents that influence the autonomic nervous

system. The cholinergic agonist, carbachol, caused a dose-

dependent increase in the rate of CSF formation up to

10.6 2.1 pl/min that was blocked by atropine. Further

evidence for cholinergic control of secretion is found in

the phenylephrine data. This alpha adrenergic agonist

also increased CSF formation by 10.6 1.5 pl/min; this

increase was blocked by both phentolamine and atropine.

These data suggested that the alpha adrenergic agonist

probably was stimulating a cholinergic pathway to release

acetylcholine to increase CSF production. Since hemi-

cholinium blocks the uptake of choline for the synthesis

of acetylcholine (Gardiner, 1957), hemicholinium was given

prior to an infusion of phenylephrine. The infusion of

phenylephrine without hemicholinium increased the

production rate to more than twice that following a

single injection of the drug. However, when the animal

was pretreated with the hemicholinium, the infusion of

phenylephrine was not able to increase fluid production









in four of six animals. These experiments confirmed that

the increase in CSF formation caused by phenylephrine

depends on the synthesis and release of acetylcholine.

This relationship is depicted in Figure 11 which shows.

alpha adrenergic receptors that may or may not be supplied

by a neuron. The failure of hemicholinium in the other

two animals might be explained by larger stores of acetyl-

choline or particularly high levels of choline. An

inadequate concentration of hemicholinium is probably not

a cause for this lack of effect.

A marked similarity exists among several secretary

systems with respect to cholinergic function and parasym-

pathetic innervation. Cholinergic stimulation of the

salivary glands causes a profuse secretion (Koelle, 1970).

Parasympathetic innervation for these glands originates

from the inferior and superior salivary nuclei of the

brainstem (Mitchell, 1953). Secretion by the pancreas

and by the stomach can also be enhanced by cholinergic

agonists (Koelle, 1970). Both of these tissues derive

their parasympathetic innervation from the dorsal vagal

nucleus which is located immediately adjacent to the

salivary nuclei in the brainstem (Mitchell, 1953). The

production of aqueous humor by the ciliary body is also

stimulated by cholinergic agents (Macri, 1971). The

source of the cholinergic innervation in this tissue is

not defined. Increase in aqueous humor formation is
































Figure 11.


Autonomic pathways to the choroid plexus I.
The effector site has a cholinergic receptor.
(++) indicates the relative effectiveness
of the activated receptor to increase CSF
production. The cholinergic nerve supplying
the choroid plexus, in turn, can be stimulated
by alpha adrenergic agonists. The location of
the alpha adrenergic receptor remains undefined.




























CHOLINERGIC
RECEPTORS












ALPHA
ADRENERGIC
RECEPTOR


EFFECTOR SITE






+ +









apparently mediated by a preganglionic cholinergic

stimulation of an adrenergic effector (Macri, 1971).

The origin of the cholinergic nerve supply to the

choroid plexus is unknown. Early anatomical descriptions

of nerves in the choroid plexus suggested that the inner-

vation might originate in the dorsal vagal nucleus of the

brainstem (Benedikt, 1874; Stohr, 1922; Clark, 1932).

Although there are no data confirming that these are

cholinergic nerves, they arise in the same area responsible

for parasympathetic-stimulated pancreatic and gastric

secretion. This area is immediately adjacent to the

superior and inferior salivary nuclei (Mitchell, 1953).

In spite of the proximity of origin to other secretary

nuclei, one major observation thus far marks the cholin-

ergic-stimulated production of CSF as unique among cholin-

ergic secretary systems: the activation of cholinergic

pathways by alpha adrenergic agonists.

As in most secretary systems, for the choroid plexus

it is unknown whether the cholinergic effector site is

vascular or epithelial. Cerebral blood vessels receive

some cholinergic innervation from the facial, glosso-

pharyngeal, and vagal cranial nerves (Mitchell, 1953)

and from pathways arising within the brain (Edvinsson,

1975). Innervation from within the central nervous

system has also been shown for the blood vessels and the

epithelium of the choroid plexus (Clark, 1928 and 1932).










The role of the sympathetic nervous system in the

production of CSF is complex. The action of the alpha

adrenergic agonist, phenylephrine, in stimulating fluid

formation through a cholinergic pathway (Figure 11) has

already been discussed. The beta2 agonist, salbutamol,

also caused a dose-dependent increase in CSF formation;

however, the maximum effect was an increment of 5.7 1.4

il/min which is approximately half of the cholinergic

response. The action of salbutamol appeared to result from

a direct stimulation of beta receptors at the effector

site (Figure 12) since neither atropine nor phentolamine

influenced its effect, whereas the general beta blocking

agent, propranolol, did reduce the salbutamol effect.

Betal receptors are apparently absent in the choroid plexus,

for ITP did not influence production at all.

Since norepinephrine is present in the choroid plexus

and disappears after cervical sympathectomy (Edvinsson

et al., 1972b; Edvinsson et al., 1974), alpha adrenergic

agonists may act at the choroid plexus to influence CSF

production directly. There are two observations that

suggest alpha adrenergic agonists may decrease CSF

production. In this study, bilateral cervical sympathetic

stimulation reduced the rate of CSF formation 3.0 0.6

pl/min. In addition, when norepinephrine is in the

perfusate during ventriculocisternal perfusion, CSF

production is decreased approximately 24% (Vates et al.,
































Figure 12.


Autonomic pathways to the choroid plexus II.
In addition to the cholinergic pathway, the
effector site also has a beta2 adrenergic
receptor. (+) indicates that activation of
these receptors also increases CSF production,
but to a lesser extent than cholinergic
stimulation.























CHOLINERGIC
RECEPTOR


BETA2
ADRENERGIC
RECEPTOR


ALPHA
ADRENERGIC
RECEPTOR









1964). A proposed alpha adrenergic pathway, completing

the autonomic innervation of the choroid plexus, is

depicted in Figure 13. However, any direct decrease an

alpha adrenergic agonist might cause at the choroid plexus

would probably be masked by the simultaneous stimulation

of production by the cholinergic pathway as described

above. Since atropine itself affects the rate of CSF

formation, no attempt was made to block the cholinergic

receptors and then study the alpha adrenergic system.

Although the sympathetic nervous system has a

greater effect on fluid production in other secretary

systems than it has on CSF formation, the actions are

variable depending on the tissue. Alpha adrenergic

agonists increase aqueous humor production by the ciliary

body through an increase in arterial blood pressure (Macri

et al., 1974). Beta2 agonists, on the other hand,

decrease aqueous humor formation (Langham, 1976). In the

salivary gland, both alpha and beta adrenergic agonists

increase fluid production (Emmelin, 1967). However, the

two systems influence secretion in different ways:

alpha adrenergic stimulation causes a secretion that is

watery and contains high potassium, whereas beta stimu-

lation causes secretion that is viscous and contains

digestive enzymes. In the pancreas both norepinephrine

and epinephrine decrease secretion; specific beta agonists

such as isoproterenol have not been tested (Thomas, 1967).
































Figure 13. Autonomic pathways to the choroid plexus III.
An alpha adrenergic receptor may also be at
the effector site in addition to the
cholinergic receptor and beta2 adrenergic
receptor. (-) indicates that this pathway
decreases the production of CSF.



























CHOLINERGIC
RECEPTOR


ALPHA
ADRENERGIC
RECEPTOR


BETA2
ADRENERGIC
RECEPTOR


ALPHA
ADRENERGIC
RECEPTOR


EFFECTOR SITE




+ + +


O










These secretary tissues derive their alpha adrenergic

innervation through the ganglia of the sympathetic chain

as does the choroid plexus.

The relatively minor physiological role of the

autonomic nervous system in the normal production of CSF

is best demonstrated by observing the effects of the

autonomic blocking agents. Atropine, the cholinergic

blocking agent, decreased the rate of CSF formation

2.6 0.5 l/min indicating the cholinergic system may

continually stimulate fluid secretion but only to the

extent of about 15 to 20% of the fluid produced. Phentol-

amine also reduced CSF formation. This decrease may be

occurring through alpha adrenergic blockade of the

cholinergic pathway since the change in CSF production

is approximately the same as that caused by atropine;

however, this is not confirmed. The beta adrenergic

system may have little direct function in the normal

secretary processes of the choroid plexus since propranolol

did not influence CSF production. Nevertheless, it is

possible that in unanesthetized animals the autonomic

nervous system may exert a greater influence on normal

fluid production than is observed here or may cause

temporary fluctuations in the rate of fluid formation.









Mechanism of Changes in CSF Production
by Autonomic Influences


In two previous investigations, epinephrine and

pilocarpine were shown to cause an increase in the flow of

CSF from the cisterna magna followed shortly by a reduction

in CSF outflow (Becht and Gunnar, 1921; Bhattacharya and

Feldberg, 1958). These authors suggested that epinephrine

and pilocarpine increased cerebral blood flow and the

resulting increase in brain volume forced subarachnoid

CSF out through the cisternal cannula without affecting

true CSF production. Increased volumes were also observed

in response to the autonomic agonists used in the study

presented in this dissertation. The increased rate of

outflow during drug treatment averaged approximately

35 l/min over control for a three minute collection

period. If pressure on the subarachnoid fluid was a

partial cause of this phenomenon, the additional fluid

collected would have come first from the cisterna magna

and basal cisterns which contained the dye marker at the

same dilution as the control perfusate. Hence, any error

in the calculation of production changes as a result of

this would cause an underestimate of the dilution of the

inflowing perfusate and an underestimate of the actual

increase in fluid production. By measuring only volumes,

the early investigators could not reliably show whether

the agents actually increased fluid formation. The









increases in CSF production calculated from dye dilution

in this study indicate that autonomic agents do alter

fluid formation.

The increase in CSF production caused by cholinergic

and adrenergic agonists can be ascribed to two possible

mechanisms of action, depending on the effector site.

Either a change in the hemodynamic relationships or a

direct action on the secretary cells might result in

alterations in CSF formation by these drugs. An increase

in the supply of blood to the brain or choroid plexus

would provide more fluid to produce an ultrafiltrate of

plasma in the CSF. Such a transfer of plasma water would

increase CSF volume.

Peripheral blood pressure changes do not correlate

well with CSF production changes. Carbachol and salbutamol

decrease blood pressure; this could decrease the supply of

blood to the secretary tissue. Phenylephrine, on the

other hand, increases peripheral blood pressure which

might serve to increase blood flow to the brain and

choroid plexus and thus to increase the rate of CSF.

formation. However, this alpha adrenergic agonist also

increases blood pressure during treatment with atropine

and hemicholinium which block the increased CSF production

by phenylephrine.

Since blood pressure changes were measured periph-

erally, choroid plexus and cerebral blood flow data were

considered necessary to determine possible vascular actions









of these agents at the site of CSF formation. The control

blood flow to the choroid plexus measured by microspheres

was 2.26 0.28 ml/gm min. This flow rate is comparable

to 3.01 ml/gm min observed by Aim and Bill (1973) for the

choroid plexus using the same technique. It is also

similar to the estimates of choroid plexus blood flow by

Welch (1963) and Pollay et al. (1972). The flow to the

choroid plexus represents one of the greatest amounts of

blood per tissue weight delivered to any tissue. Carbachol

and phenylephrine increased mean plexus blood flow 0.81

and 0.77 ml/gm min, respectively. Salbutamol increased

blood flow to the choroid plexus 0.35 ml/gm min. The

effects of these drugs on blood flow are in about the

same proportions as their effects on CSF production.

However, none of these changes in choroid plexus blood

flow differ statistically from control flows because of

the wide variation in the changes. These data, along with

the data of Alm and Bill (1973) showing no effect on

plexus blood flow by sympathetic stimulation, seem to

indicate that the autonomic nervous system does not

significantly alter the blood flow to the choroid plexus.

Cerebral blood flow in the control experiments was

0.46 .04 ml/gm min. This value is approximately the

same as the flows measured for the dog (Marcus et al.,

1976) and sheep (Hales, 1973). There was no alteration

in brain blood flow after carbachol, phenylephrine, or









salbutamol. When Salanga and Waltz (1973) stimulated the

facial nerve, vasodilation occurred only if the nerve was

cut where it left the brainstem. This suggests that

sensory nerve fibers detecting cerebral vasodilation might

pass back into the central nervous system through the

facial nerve causing a reflex vasoconstriction. If such a

pathway does exist, it would also explain the ineffec-

tiveness of salbutamol and carbachol to increase cerebral

blood flow. The inability of phenylephrine to alter

cerebral blood flow is probably due to its low efficacy on

alpha adrenergic receptors in cerebral blood vessels

(Duckles and Bevan, 1976). Blood pressure and blood flow

data seem to indicate that the autonomic treatments

performed in this study produce minimal effects on the

production of CSF by way of changes in blood flow. Also,

it is clear that any increase in the rate of outflow seen

during drug treatment was not mediated by a generalized

increase in blood flow to the brain. An increase in

choroid plexus blood flow could have forced additional

perfusate from the ventricles.

Verapamil and theophylline were used to investigate

two possible ways in which autonomic agents could influence

CSF production by direct action on secretary cells. The

presence of calcium is a requirement for cholinergic

stimulation of salivary secretion (Douglas and Poisner,

1963). If calcium influx has a role in increasing CSF










production stimulated by carbachol, then verapamil, which

blocks calcium channels (Kohlhardt et al., 1972), should

prevent the increase in fluid formation caused by the

cholinergic agonist. Verapamil has been shown to be

effective when administered on either side of the epithe-

lium and was given intraventricularly (Bentley, 1974).

Verapamil did not affect either normal CSF formation or

fluid production stimulated by carbachol; therefore,

fluid secretion probably occurs through a mechanism other

than a calcium-induced activation.

Since the effects of both cholinergic and adrenergic

neurotransmitters may be mediated by changes in cyclic

nucleotide levels (Greengard, 1976), an increase in

guanosine 3',5' cyclic monophosphate (cGMP) or adenosine

3',5' cyclic monophosphate (cAMP) might be the means by

which carbachol, phenylephrine, and salbutamol increase

the production of CSF. Theophylline, which inhibits the

destruction of the nucleotides by phosphodiesterase among

other actions, was used to enhance the levels of the

cyclic nucleotides. These experiments showed that

theophylline caused a significant dose-dependent increase

in CSF formation. The higher dose increased fluid

production 9.2 1.4 pl/min which is approximately equal

to the maximum effect of the autonomic agonists. Unfor-

tunately, theophylline is not totally specific for either

cGMP or cAMP phosphodiesterases (Amer and Kreighbaum,









1975), so it is not possible to show whether the effects

mediated by cyclic nucleotide action are cholinergic,

beta2 adrenergic, or both. Whereas the theophylline

experiments suggest that increases in cyclic nucleotide

levels are responsible for the increases in fluid

production elicited by carbachol, phenylephrine, and

salbuatmol administration, they do not provide conclusive-

evidence that enhancement of the cyclic nucleotide

concentrations takes place in choroid plexus epithelia.

However, the relatively large blood flow to the choroid

plexus does not significantly change during treatment

with any of the autonomic agents that increase CSF

production. Apparently, the blood flow only provides

fluid and electrolyte substrate for the secretary

processes of the choroid plexus epithelium. The data

strongly imply that autonomic agents influence CSF

production by causing changes primarily at the choroid

plexus epithelium.

The changes in CSF production mediated by the

epithelial cells are probably a result of alterations in

the transport of electrolytes. Three pathways have been

demonstrated for the secretary transfer of electrolytes

in the CSF: the catalysis of bicarbonate formation by

carbonic anhydrase, the transport of cations by the

ouabain-sensitive sodium-potassium ATPase, and the

probable transport of chloride by an ethacrynic acid-

sensitive pump. Inhibition of any of the three mechanisms









has been shown to decrease the rate of CSF formation.

Inhibition of carbonic anhydrase by acetazolamide,

benzolamide, or methazolamide reduces fluid production

approximately 50 to 70% (Pollay and Davson, 1963; Davson

and Segal, 1970; Broder and Oppelt, 1969; and this study).

That this effect involves the movement of bicarbonate ions

into CSF was first demonstrated by Maren and Broder (1970)

and further quantified by Vogh and Maren (1975). Ouabain

decreases the production of CSF 50 to 60% by its inhibition

of sodium-potassium ATPase (Vates et al., 1964; Davson and

Segal, 1970; Garg and Mathur, 1975). A 50 to 60% reduction

in CSF formation is seen with ethacrynic acid (Domer,

1969; Miner and Reed, 1971) and furosemide (McCarthy and

Reed, 1974). Active transfer .of chloride ions inhibited

by ethacrynic acid and furosemide has been described for

the ascending limb of Henle's loop in the kidney (Burg,

1974), and thus may be presumed to occur in CSF secretion.

The possibility of an active chloride pump in the choroid

plexus is supported by the findings that the chloride

concentration in the CSF can be maintained during reduced

plasma levels of the ion (Abbott et al., 1971; Bourke

et al., 1970).

Assuming that changes in cGMP are associated with

cholinergic responses as in other tissues (Greengard,

1976), levels of cGMP may be increased at the critical

site for CSF production by cholinergic agonists or by

alpha adrenergic stimulation of the cholinergic pathway