Molecular Physiology And Metabolism Of The

Nervous System A Clinical Perspective 1st

Edition Gary A Rosenberg download

https://ebookbell.com/product/molecular-physiology-and-
metabolism-of-the-nervous-system-a-clinical-perspective-1st-
edition-gary-a-rosenberg-6638980

Explore and download more ebooks at ebookbell.com

Here are some recommended products that we believe you will be
interested in. You can click the link to download.

Metabolism And Molecular Physiology Of Saccharomyces Cerevisiae 2nd

Edition 2nd Edition J Richard Dickinson

https://ebookbell.com/product/metabolism-and-molecular-physiology-of-
saccharomyces-cerevisiae-2nd-edition-2nd-edition-j-richard-
dickinson-1308196

Glioma Imaging Physiologic Metabolic And Molecular Approaches Whitney

B Pope

https://ebookbell.com/product/glioma-imaging-physiologic-metabolic-
and-molecular-approaches-whitney-b-pope-10666378

Molecular Physiology And Evolution Of Insect Digestive Systems 1st Ed

2023 Walter R Terra

https://ebookbell.com/product/molecular-physiology-and-evolution-of-
insect-digestive-systems-1st-ed-2023-walter-r-terra-52092460

Molecular Physiology And Ecophysiology Of Sulfur 1st Edition Luit J De

Kok

https://ebookbell.com/product/molecular-physiology-and-ecophysiology-
of-sulfur-1st-edition-luit-j-de-kok-5353802
Legumes Physiology And Molecular Biology Of Abiotic Stress Tolerance
1st Edition Prakash Muthu Arjuna Samy

https://ebookbell.com/product/legumes-physiology-and-molecular-
biology-of-abiotic-stress-tolerance-1st-edition-prakash-muthu-arjuna-
samy-49062022

Physiology And Molecular Biology Of Stress Tolerance In Plants 1st

Edition Kv Madhava Rao

https://ebookbell.com/product/physiology-and-molecular-biology-of-
stress-tolerance-in-plants-1st-edition-kv-madhava-rao-1222990

Molecular Mechanisms And Physiology Of Disease Implications For

Epigenetics And Health 1st Edition Nilanjana Maulik

https://ebookbell.com/product/molecular-mechanisms-and-physiology-of-
disease-implications-for-epigenetics-and-health-1st-edition-nilanjana-
maulik-4696478

Chlamydomonas Molecular Genetics And Physiology 1st Edition Michael

Hippler Eds

https://ebookbell.com/product/chlamydomonas-molecular-genetics-and-
physiology-1st-edition-michael-hippler-eds-6841604

Molecular And Cellular Physiology Of Neurons 2nd Edition Gordon L Fain

https://ebookbell.com/product/molecular-and-cellular-physiology-of-
neurons-2nd-edition-gordon-l-fain-51502248
MOLECULAR PHYSIOLOGY AND METABOLISM
OF THE NERVOUS SYSTEM
SERIES EDITOR
Sid Gilman, MD, FRCP
William J. Herdman Distinguished University Professor of Neurology
University of Michigan

Contemporary Neurology Series

61 HIV NEUROLOGY 73 NEUROLOGIC COMPLICATIONS

Bruce James Brew, MBBS, MD, FRACP OF CANCER,
62 ISCHEMIC CEREBROVASCULAR Second Edition
DISEASE Lisa M. DeAngelis, MD and
Harold P. Adams, Jr., MD, Jerome B. Posner, MD
Vladimir Hachinski, MD, and 74 NEUROLOGIC COMPLICATIONS
John W. Norris, MD OF CRITICAL ILLNESS,
65 MIGRAINE: MANIFESTATIONS, Third Edition
PATHOGENESIS, AND Eelco F.M. Wijdicks, MD, PhD, FACP
MANAGEMENT, 75 CLINICAL NEUROPHYSIOLOGY,
Second Edition
Third Edition
Robert A. Davidoff, MD
67 THE CLINICAL SCIENCE OF Jasper R. Daube, MD and
Devon I Rubin, MD, Editors
NEUROLOGIC REHABILITATION,
Second Edition 76 PERIPHERAL NEUROPATHIES IN
Bruce H. Dobkin, MD CLINICAL PRACTICE
68 NEUROLOGY OF COGNITIVE AND Steven Herskovitz, MD,
BEHAVIORAL DISORDERS Stephen N. Scelsa, MD, and
Orrin Devinsky, MD and Herbert H. Schaumburg, MD
Mark D’Esposito, MD 77 CLINICAL NEUROPHYSIOLIOGY OF
69 PALLIATIVE CARE IN THE VESTIBULAR SYSTEM,
NEUROLOGY Fourth Edition
Raymond Voltz, MD, Robert W. Baloh, MD, FAAN and
James L. Bernat, MD, Kevin A. Kerber, MD
Gian Domenico Borasio, MD, DipPallMed, 78 THE NEURONAL CEROID
Ian Maddocks, MD, LIPOFUSCINOSES
David Oliver, FRCGP, and (BATTEN DISEASE),
Russell K. Portenoy, MD Second Edition
70 THE NEUROLOGY OF EYE Sara E. Mole, PhD, Ruth D. Williams, MD,
MOVEMENTS, and Hans H Goebel, MD, Editors
Fourth Edition
79 PARANEOPLASTIC SYNDROMES
R. John Leigh, MD, FRCP and
David S. Zee, MD Robert B. Darnell, MD, PhD, and
71 PLUM AND POSNER’S DIAGNOSIS OF Jerome B. Posner, MD
STUPOR AND COMA, 80 JASPER’S BASIC MECHANISMS OF
Fourth Edition THE EPILEPSIES
Jerome B. Posner, MD, Jeffrey L. Noebels, MD, PhD,
Clifford B. Saper, MD, PhD, Massimo Avoli, MD, PhD,
Nicholas D. Schiff, MD, and Michael A. Rogawski, MD, PhD,
Fred Plum, MD Richard W. Olsen, PhD, and
72 PRINCIPLES OF DRUG THERAPY IN Antonio V. Delgado-Escueta, MD, Editors
NEUROLOGY, 81 MYASTHENIA GRAVIS AND
Second Edition MYASTEHNIC DISORDERS,
Michael V. Johnston, MD and Second Edition
Robert A. Gross, MD, PhD, Editors Andrew G. Engel, MD, Editor
MOLECULAR PHYSIOLOGY
AND METABOLISM OF
THE NERVOUS SYSTEM

Gary A. Rosenberg, MD
Chairman of Neurology
Professor of Neurology, Neurosciences, Cell Biology
and Physiology, and Mathematics and Statistics
University of New Mexico Health Sciences Center
Albuquerque, NM

1
1
Oxford University Press, Inc., publishes works that further
Oxford University’s objective of excellence
in research, scholarship, and education.

Oxford New York

Auckland Cape Town Dar es Salaam Hong Kong Karachi

Kuala Lumpur Madrid Melbourne Mexico City Nairobi
New Delhi Shanghai Taipei Toronto

With offices in
Argentina Austria Brazil Chile Czech Republic France Greece
Guatemala Hungary Italy Japan Poland Portugal Singapore
South Korea Switzerland Thailand Turkey Ukraine Vietnam

Copyright © 2012 by Oxford University Press

Published by Oxford University Press, Inc.

198 Madison Avenue, New York, New York 10016
www.oup.com

Oxford is a registered trademark of Oxford University Press

All rights reserved. No part of this publication may be reproduced,

stored in a retrieval system, or transmitted, in any form or by any means,
electronic, mechanical, photocopying, recording, or otherwise,
without the prior permission of Oxford University Press.

——————————————————————–
Library of Congress Cataloging-in-Publication Data

Rosenberg, Gary A.
Molecular physiology and metabolism of the nervous system : a clinical perspective / Gary A. Rosenberg.
p. ; cm. — (Contemporary neurology series ; 82)
Includes bibliographical references and index.
ISBN 978-0-19–539427-6 (hardcover : alk. paper)
I. Title. II. Series: Contemporary neurology series ; 82. 0069–9446
[DNLM: 1. Cerebrospinal Fluid—physiology. 2. Blood-Brain Barrier—physiology.
3. Brain Diseases—physiopathology. 4. Cerebrospinal Fluid—metabolism. 5. Cerebrovascular
Circulation—physiology. W1 CO769N v. 82 2012 / WL 203]

612.8’042—dc23 2011044062
——————————————————————–

The science of medicine is a rapidly changing field. As new research and clinical experience broaden our knowledge,
changes in treatment and drug therapy occur. The author and publisher of this work have checked with sources believed
to be reliable in their efforts to provide information that is accurate and complete, and in accordance with the standards
accepted at the time of publication. However, in light of the possibility of human error or changes in the practice
of medicine, neither the author, nor the publisher, nor any other party who has been involved in the preparation or
publication of this work warrants that the information contained herein is in every respect accurate or complete. Readers
are encouraged to confirm the information contained herein with other reliable sources, and are strongly advised to check
the product information sheet provided by the pharmaceutical company for each drug they plan to administer.

987654321
Printed in the United States of America
on acid-free paper
To Evelyn
This page intentionally left blank
Preface

The neurosciences and clinical neurology have undergone dramatic changes in the past 25 years
brought about by major advances in molecular biology and neuroimaging. Clinical practice has
remained grounded in ideas and concepts that were first enunciated decades ago, but the advances
made in the laboratory are beginning to impact the clinician. In the 1970s, the invention of math-
ematical equations for tomography opened an era of neuroimaging using x-rays for computed
tomography (CT), radionuclear isotopes for positron emission tomography (PET), and magnetic
resonance for magnetic resonance imaging (MRI). This ability to visualize brain pathology prior
to autopsy profoundly changed the practice of neurology. At around that same time, advances in
molecular biology began to penetrate the neurosciences, and have now exploded with the elucida-
tion of the human genome, gene chip technology, and, more recently, the findings of proteomics
and metabolomics.
Clinicians and neuroscientists beginning to grapple with this profusion of information are faced
with the need to learn older physiological concepts that are relevant to patient care. But knowing
the physiology, which used to be sufficient, is no longer adequate. It must be combined with the
molecular biology to form a new science of molecular physiology. To be successful in clinical care
as well as in clinical or basic neurosciences, multiple concepts and techniques need to be mastered.
No longer is it sufficient to be well versed only in one of the major branches of neuroscience,
such as neuroanatomy, neurophysiology, neuropathology or neurochemistry; it is now necessary to
combine and use them all at some point. To do this successfully, scientists and clinicians need to
work as a team; each person in the collaboration brings a unique skill to the project. Each person
on the team has a set of skills and a group of words that he or she understands best, but it remains
mainly one person’s work, and for that individual to lead the effort, an understanding of the others’
areas of expertise is required. Learning to do that person’s part of the project usually is possible,
and having a common language is the key to true teamwork. This is true not only for the complex
scientific project but even more so for the clinician, who on a daily, even hourly, basis is involved
in a large team.
Working with residents and graduate students over the past years has taught me the importance
of incorporating the newer molecular insights into the care of patients. While we now know the
patterns of most neurological diseases from intensive work of neuroimagers, along with many of
the genes involved, we remain far behind in developing treatments. The challenge we now face is
to relate the imaging to the molecular studies and to understand the underlying physiological role
of the specific molecules in the injury cascades. Once that knowledge is available, we will need to
translate it into novel therapies.
Translational research attempts to accelerate the movement of information from the basic sci-
ences to the clinic and to take the insights gained from caring for patients back to the laboratory for
further study. Drug screening can be done with high-throughput systems; blood tests can identify
arrays of genes; clinical trials can be done by large consortia; information learned on one continent
can be quickly conveyed to another group far away by the Internet. This acceleration of informa-
tion transfer has resulted in remarkable advances in treatment. For that to occur, a new type of
investigator is needed, one who is equally comfortable working in the worlds of brain physiology
and molecular neurochemistry. My goal in writing this book is to combine the important insights
into brain physiology gained by early investigators with the new knowledge being obtained on
genes and proteins in order to understand the impact of these substances in the living animal.
The goal of this book is explain the basic physiological concepts about the brain fluids, cerebral
blood flow, and the blood-brain barrier and the quantitative approaches to their study. This is the
topic of the first part of the book. The second part is more concerned with metabolic pathways and
aspects of transport. Pathological aspects of the brain fluids and metabolism are introduced where

vii
viii Preface

appropriate in both of these parts but are more extensively discussed in the final part on hypoxia,
ischemia, brain edema, and inflammation. I have tried to emphasize the commonalities among the
various aspects of fluid balance and metabolism in the different diseases.
The information in this book will aid students, trainees in neurology and neurosurgery and
research neuroscientists in the understanding of the basic concepts of physiology and molecular
biology that apply to clinical practice and translational medicine.

— G. A. R
Albuquerque, New Mexico 2011
Acknowledgments

Paul Akmajian skillfully made the original drawings in the book. Craig Panner of Oxford University
Press and Sid Gilman of the Contemporary Neurology Series provided much appreciated assistance.
The American Heart Association and the National Institute of Neurological Disorders and Stroke
provided research support.

ix
This page intentionally left blank
Contents

PART 1 PHYSIOLOGY OF BRAIN FLUIDS

AND THE BLOOD-BRAIN BARRIER

1. ANATOMY OF FLUID INTERFACES THAT PROTECT

THE MICROENVIRONMENT 3

HISTORICAL PERSPECTIVE 3

CEREBRAL MICROENVIRONMENT 4

DEVELOPMENT OF THE BRAIN-FLUID INTERFACES 6

Neural Tube, Ependymal Cells, and Stem Cells • Cilated Ependymal Cells and CSF
Movement • Choroid Plexuses, Arachnoid, and Capillaries

EXTRACELLULAR SPACE AND EXTRACELLULAR MATRIX 10

BRAIN-FLUID INTERFACES 11
Anatomy of the Cerebral Blood Vessels • Brain Cell Interfaces with CSF at Ependyma and Pia

DURA, ARACHNOID, AND PIAL LAYERS 15

WHAT ARE THE SOURCES OF ENERGY? 16

2. PHYSIOLOGY OF THE CEREBROSPINAL AND INTERSTITIAL FLUIDS 18

INTRODUCTION 18

PROTEINS IN THE CSF 19

CSF PRESSURE REFLECTS VENOUS PRESSURE IN THE RIGHT HEART 20

FORMATION, CIRCULATION, AND ABSORPTION OF CSF 21

Formation of CSF by Choroid Plexuses • Choroid Plexus and Disease Biomarkers in
CSF • Absorption of CSF at the Arachnoid Villi

ELECTROLYTE BALANCE IN THE CSF 25

MENINGES AND SITES OF MASSES AND INFECTION 26

xi
xii Contents

INTERSTITIAL FLUID 27

LYPHATIC DRAINAGE 28

WATER DIFFUSION, BULK FLOW OF ISF, AND DIFFUSION

TENSOR IMAGING 28

NEUROPEPTIDES AND FLUID HOMEOSTASIS 29

AQUAPORINS AND WATER TRANSPORT IN THE CENTRAL

NERVOUS SYSTEM 30

3. NEUROVASCULAR UNIT 34

EARLY EXPERIMENTS ON THE BLOOD-BRAIN BARRIER 34

THE NEUROVASCULAR UNIT AND TIGHT JUNCTION PROTEINS 34

INTEGRINS, SELECTINS, AND ENDOTHELIAL CELL ADHESION 37

ASTROCYTES, PERICYTES, AND BASAL LAMINA 38

MOVEMENT OF SUBSTANCES INTO AND OUT OF BRAIN 40

GLUCOSE AND AMINO ACID TRANSPORT 42

PROTEASES AND THE NEUROVASCULAR UNIT 44

MATRIX METALLOPROTEINASES (MMPS) 45

A DISINTEGRIN AND METALLOPROTEINASE (ADAM) 48

BARRIER SYSTEMS EVOLVED TO AN ENDOTHELIAL BARRIER 49

PART 2 METABOLISM, DISORDERS OF BRAIN FLUIDS,

AND MATHEMATICS OF TRANSPORT

4. GLUCOSE, AMINO ACID, AND LIPID METABOLISM 55

GLUCOSE METABOLISM 55

AMINO ACID NEUROTRANSMITTERS 56

LIPID METABOLISM 60

EICOSANOID METABOLISM 61

HEPATIC ENCEPHALOPATHY 62
 Contents xiii

HYPOGLYCEMIA 63

HYPONATREMIA, OSMOTIC DEMYELINATION, AND ACID BALANCE 65

Hyponatremia • Hyperglycemia • Acidosis

5. DISORDERS OF CEREBROSPINAL CIRCULATION: IDIOPATHIC

INTRACRANIAL HYPERTENSION AND HYDROCEPHALUS 68

INTRODUCTION 68

CLINICAL FEATURES OF IIH 68

TREATMENT OF IIH 72

HYDROCEPHALUS 72

HYDROCEPHALUS IN CHILDREN 73

ADULT-ONSET HYDROCEPHALUS 74
Obstructive Hydrocephalus • Normal Pressure Hydrocephalus

6. QUANTIFICATION OF CEREBRAL BLOOD FLOW AND BLOOD-BRAIN

BARRIER TRANSPORT BY NUCLEAR MAGNETIC RESONANCE AND
POSITRON EMISSION TOMOGRAPHY 79

INTRODUCTION 79

MATHEMATICAL APPROACH TO CBF AND TRANSPORT 80

CBF: The Schmidt-Kety Approach • Regional Blood Flow • Transport Between
Blood and Brain

POSITRON EMISSION TOMOGRAPHY 84

Single-Injection External Registration • Patlak Graphical BBB Method for
Autoradiography and MRI

MRI IN CBF AND TRANSPORT MEASUREMENT 88

MRI AND SPECTROSCOPY 88

31
Multinuclear NMR • The Relaxation Phenomenon and the Rotating Frame • P-MRS •
13
C-MRS • 1H-MRS

PART 3 ISCHEMIA, EDEMA, AND INFLAMMATION

7. MECHANISMS OF ISCHEMIC/HYPOXIC BRAIN INJURY 101

EPIDEMIOLOGY, RISK FACTORS, AND PREVENTION OF STROKE 101

xiv Contents

MOLECULAR CASCADES IN ISCHEMIC TISSUE RESULTS

FROM ENERGY FAILURE 102

EXCITATORY AND INHIBITORY NEUROTRANSMITTERS 104

NEUROINFLAMMATION IN STROKE 107

PROTEASES IN HYPOXIA/ISCHEMIA 108

CASPASES AND CELL DEATH 110

TISSUE INHIBITORS OF METALLOPROTEINASES AND APOPTOSIS 111

TIGHT JUNCTION PROTEINS AND MMPS 113

MMPS AND TPA-INDUCED BLEEDING 113

ANIMAL MODELS IN STROKE 116

ARTERIOVENOUS MALFORMATIONS AND CAVERNOUS HEMANGIOMAS 117

MAGNETIC RESONANCE IMAGING, POSITRON EMISSION TOMOGRAPHY,

AND ELECTRON PARAMAGNETIC RESONANCE IN HYPOXIA/ISCHEMIA 118
Magnetic Resonance Imaging and Magnetic Resonance Spectroscopy • Positron Emission
Tomography • Electron Paramagnetic Resonance

8. VASCULAR COGNITIVE IMPAIRMENT AND ALZHEIMER’S DISEASE 124

REGULATION OF CEREBRAL BLOOD FLOW 124

HYPOXIA/ISCHEMIA IN CARDIAC ARREST 127

Prognosis for Recovery After Cardiac Arrest • Cardiac Surgery and Memory Loss • Delayed
Postanoxic Leukoencephalopathy

HYPOXIA-INDUCIBLE FACTORS AND GENE EXPRESSION 129

INTERMITTENT HYPOXIA IS A STRONG STIMULUS FOR HIF 131

VASCULAR COGNITIVE IMPAIRMENT 132

WHITE MATTER HYPERINTENSITIES ON MRI AND

BINSWANGER’S DISEASE 133

AD, VASCULAR DISEASE, AND THE AMYLOID HYPOTHESIS 138

9. EFFECTS OF ALTITUDE ON THE BRAIN 144

INTRODUCTION 144
 Contents xv

GENETIC TOLERANCE TO ALTITUDE 144

AMS AND HIGH-ALTITUDE PULMONARY EDEMA 146

HIGH-ALTITUDE CEREBRAL EDEMA 146

COGNITIVE CONSEQUENCES OF HYPOBARIC HYPOXIA 148

IMAGING OF THE BRAIN AT HIGH ALTITUDE 148

HIF AND SLEEP DISORDERS IN AMS 149

TREATMENT OF ALTITUDE ILLNESSES 150

10. BRAIN EDEMA 152

INTRODUCTION 152

ROLE OF AQUAPORINS IN BRAIN EDEMA 155

ROLE OF NEUROINFLAMMATION IN THE FORMATION

OF VASOGENIC EDEMA 157
Oxidative Stress and Brain Edema • Arachidonic Acid and Brain Edema •
Vascular Endothelial Growth Factor and Angiopoietins

CLINICAL CONDITIONS ASSOCIATED WITH BRAIN EDEMA 159

IMAGING BRAIN EDEMA 160

TREATMENT OF BRAIN EDEMA AND

HYPOXIC/ISCHEMIC INJURY 162

MULTIPLE DRUGS FOR TREATMENT OF ISCHEMIA 164

11. INTRACEREBRAL HEMORRHAGE 169

INTRODUCTION 169

HISTORY OF ICH 170

MOLECULAR MECHANISMS IN ICH 171

CLINICAL ASPECTS OF INTRACRANIAL BLEEDING 172

PATHOPHYSIOLOGY OF ICH: EVIDENCE FROM ANIMAL STUDIES 176

EXTRAPOLATION OF EXPERIMENTAL RESULTS TO

TREATMENTS FOR ICH 177
xvi Contents

12. AUTOIMMUNITY, HYPOXIA, AND INFLAMMATION

IN DEMYELINATING DISEASES 182

INTRODUCTION 182

HETEROGENEITY OF THE PATHOLOGICAL FINDINGS IN MS 184

PROTEASES IMPLICATED IN MS PATHOLOGY 188

BBB DISRUPTION IN MS 189

DEVIC’S NEUROMYELITIS OPTICA 190

NONIMMUNOLOGICAL PROCESSES IN DEMYELINATION 192

EAE AND THE PATHOGENESIS OF MS 193

MODERN APPROACHES TO THE TREATMENT OF MS 194

EPILOGUE: SYNTHESIS AND FUTURE DIRECTIONS 195

INDEX 203
PART 1

Physiology of Brain Fluids

and the Blood-Brain Barrier
This page intentionally left blank
Chapter 1

Anatomy of Fluid Interfaces that

Protect the Microenvironment

HISTORICAL PERSPECTIVE EXTRACELLULAR SPACE AND

EXTRACELLULAR MATRIX
CEREBRAL MICROENVIRONMENT
BRAIN-FLUID INTERFACES
DEVELOPMENT OF THE BRAIN-FLUID
Anatomy of the Cerebral Blood Vessels
INTERFACES
Brain Cell Interfaces with CSF
Neural Tube, Ependymal Cells,
at Ependyma and Pia
and Stem Cells
Cilated Ependymal Cells and CSF Movement DURA, ARACHNOID, AND PIAL LAYERS
Choroid Plexuses, Arachnoid, and WHAT ARE THE SOURCES OF ENERGY?
Capillaries

HISTORICAL PERSPECTIVE Camillo Golgi discovered the method

of silver staining. Cajal used this method
Brain fluid studies entered the modern era to describe the close relationship between
with the work begun at the beginning of the astrocytes and cerebral blood vessels in the
twentieth century by the physiologist Louis human brain. Silver staining shows only a
Weed and the neurosurgeon Harvey Cushing. small population of astrocytes, which for-
They understood that the brain, lacking lym- tuitously provided a clearer picture of the
phatics, used the cerebrospinal and interstitial connections between glial cells and blood
fluids (CSF and ISF) as lymph, calling this a vessels, showing that the astrocytic foot pro-
third circulation1 (Figure 1–1). At around the cesses expanded as they contacted the cere-
same time, there was growing awareness that bral vessels. This prescient observation lost
the brain was sequestered from the systemic favor for many years as the endothelium
circulation. Ehrlich was the first to show this assumed the dominant role as the site of the
in 1885; he injected a blue dye into the blood, BBB. Recent studies have revived the con-
which stained all organs except the brain. In cept of a functional unit including vessels
1913, Goldmann observed that trypan blue and astrocytes as the regulator of perme-
dye injected into the CSF stained brain tis- ability. Golgi’s illustrations were remarkably
sue. These two experiments were a dramatic accurate considering the state of neuroanat-
demonstration that proved that the brain was omy at the beginning of the twentieth cen-
isolated from the blood but that, once inside tury (Figure 1–2). Comparison of a recent
the skull, fluids had free access to brain paren- confocal micrograph of astrocytes attached
chyma via CSF/ISF. This isolation became to capillaries and one of Golgi’s hand-drawn
known as the blood-brain barrier (BBB). figures demonstrates this accuracy.2

3
4 Molecular Physiology and Metabolism of the Nervous System

be done after careful examination of the fundi

to rule out papilledema and close observation
after the test to assess for possible herniation.
Examination of the CSF should always be con-
Cephalic Circulation sidered in the diagnostic workup when infec-
tion, subarachnoid hemorrhage or idiopathic
increased intracranial pressure is suspected
because it is cost-effective and ultimately can
contain costs by reducing the need for more
expensive tests.
Cerebrospinal fluid and ISF are separated by
the ependyma in the ventricles and by the pia
in the subarachnoid space. Substances in the
A. Ca

Ventricular Arachnoidal
System System nt.
si
CSF can enter the brain across these cell lay-
ers because they have gap junctions rather than
rotus

ari

Diagram of
gul

tight junctions. Continuity of the CSF and ISF

cerebro−spinal
comm

means that substances that would be excluded

V. J

circulatory sector
from the brain after injection into the blood can
unis

be injected intrathecally where they can access

brain cells. Drainage of metabolic products
Figure 1–1. Drawing of the third circulation. Blood and from the brain into the CSF via the ISF pro-
lymph are the first and second circulations. Fluid leaving the vides a lymph-like function, which solves the
capillaries and the choroid plexuses moves into the brain’s problem created when the brain tissues were
extracellular space and the cerebral ventricles. Removal of
CSF and ISF is across the arachnoid granulations. (From isolated from the body during evolution. Fluids
Ref. 1.) of the third circulation are formed at the cere-
bral endothelial cells and the ependymal cells
of the choroid plexus. At both sites, adenosine
triphosphatase (ATPase) pumps provide the
CEREBRAL MICROENVIRONMENT energy to create the osmotic gradients that pull
water into brain extracellular space and cere-
Normal cellular function depends on the micro- bral ventricles. Newly formed ISF from the
environment created around brain cells by the capillaries and CSF from the choroid plexuses
ISF, which is contiguous with CSF (Figure 1–3). join together in the ventricles before draining
Cerebrospinal fluid fills the cerebral ventricles, back into the blood.
the subarachnoid space over the brain, and the Proteins in the blood circulate into tissues
fluid spaces around spinal cord and pools in through fenestrations in systemic capillaries.
the lumbar sac below the ending of the cord Acting as passive sieves, pores in the capillar-
at L1-L2. Cerebral ventricles contain about ies permit these large molecules to move into
20 mL, with the remainder of the 140 mL over the tissue to be drained back to the blood by
the brain, around the spinal cord, and in the the lymphatic vessels. This essential lymphatic
lumbar sac, where it can be removed by lum- function prevents protein accumulation in the
bar puncture. Several important neurological tissues, which would create oncotic pressure,
illnesses require examination of the CSF for resulting in edema.
diagnosis. Infection, Guillain-Barré syndrome, By contrast, cerebral vessels restrict trans-
central nervous system vasculitis, idiopathic port between the blood and the brain of
intracranial hypertension, multiple sclerosis, most substances except lipophilic molecules.
and, more recently, Alzheimer’s disease are Specialized proteins that self-assemble to form
examples of diseases in which the CSF is an tight junctions fill the clefts between cerebral
important diagnostic aid (Table 1–1). An imag- capillaries. Tight junction proteins restrict the
ing study should be performed prior to lumbar movement of charged or large protein mole-
puncture to rule out a mass lesion or obstruc- cules. Another unique feature of the brain cap-
tive hydrocephalus. Occasionally, in critically illary is the requirement for an energy source
ill patients suspected of having meningitis or to power sodium-potassium ATPase exchange
when imaging is difficult, lumbar puncture can pumps, which maintain a constant flow of ISF
 (A) (B)

(C)

Figure 1–2. Camillo Golgi (1843–1926) described the close relationship between astrocytes and cerebral blood vessels in
the human cerebellum as follows: “the connection between glia and vessels is either direct, the cell bodies being applied on
the vessel walls, of which they seem to be part of, or occurs through protrusions more or less pronounced, which exhibit a
small expansion at the point of contact.” (A) Drawing from table XII of Golgi’s book. (B) Enlargement of (A) showing the
relationships between astrocytes and blood vessels. (C) Confocal image in which astrocytes were double-labeled with aqua-
porin 4 and glial fibrillary acidic protein (GFAP). The similarities between Golgi’s drawing and the confocal image obtained
more than a century later are striking. (From Ref. 2.)

BLOOD-CSF CSF-BRAIN BLOOD-BRAIN CSF-BLOOD

SAS

Blood Stroma Ependyma Sinus Skull

CHOROID PLEXUS EPENDYMA NEUROPIL PIA ARACHNOID VILLI
(15-20% ECS)
Fenestrated
capillaries Gap junctions

Tight junctions

Figure 1–3. Illustration of the third circulation with the regions of tight and gap junctions. In the choroid plexus, substanc-
es in the blood leave fenestrated capillaries to enter the stroma beneath the ependyma, where apical tight junctions prevent
them from entering the CSF. Once in the CSF, molecules can pass through the gap junctions of the nonchoroid plexus
ependyma. Extracellular space makes up 15% to 20% of the brain, and ISF delivers substances to the cells in the neuropil.
Interstitial fluid leaves the neuropil to enter the subarachnoid space before returning to the blood across the arachnoid villa.
The arachnoid has tight junctions. Within the brain, the capillaries also have tight junctions.

5
6 Molecular Physiology and Metabolism of the Nervous System

Table 1–1 Cerebrospinal Fluid Fluid formed by brain capillaries and choroid
Examinations Essential in the Diagnosis plexus flows through interstitial spaces, deliv-
of Several Neurological Disorders ering nutrients and removing waste, eventually
draining into the ventricles for removal over
CSF test for central nervous system infection the convexities via the arachnoid granulations
Cellular and protein content of CSF are helpful in in the superior sagittal sinus and down along
distinguishing the various types of meningitis the spinal cord, where arachnoid granulations
Protein elevation without cells is diagnostic of located in the nerve root sleeves perform a
Guillain-Barré syndrome similar function (Table 1–2).
CSF pressure is diagnostic test for idiopathic
intracranial hypertension
Large number of cells in central nervous system DEVELOPMENT OF THE
vasculitis, which can separate it from multiple
sclerosis BRAIN-FLUID INTERFACES
Demyelinating profile in multiple sclerosis
diagnosis (myelin basic protein, oligoclonal Neural Tube, Ependymal Cells,
bands, IgG synthesis) and Stem Cells
The nervous system develops from a region in
that leaves the capillaries along osmotic gradi- the middorsal line of the embryo. A thickened
ents created by ion pumps on the abluminal plate of ectoderm folds in to form the neural
surface. Similar mechanisms of fluid formation groove, which, once closed, becomes the neu-
are found at the choroid plexus. ral tube (Figure 1–4). The cephalic part begins
Water can readily cross membranes, which to dilate to form the brain and the ventricular
is anomalous behavior that was poorly under- system, while the caudal segment that will be
stood until the recent discovery of a family of the spinal cord maintains a uniform diameter.
pore-forming molecules called aquaporins. An internal limiting membrane on the inner
These molecules form channels through which surface forms next to the cells that will become
water molecules move passively along pressure the ependyma. At the outer surface is mesen-
and osmotic gradients.3 Astrocytic endfeet are chyma that is separated from the ectoderm
rich in aquaporin molecules, and when edema by an external limiting membrane. Germinal
fluid forms from movement of water out of cells are found between the inner and outer
the capillaries into the extracellular space, the membranes. The mantle layer becomes the
astrocytic endfeet swell. The flow of ISF that is gray matter, composed of glia and neurons,
formed by capillaries occurs through the extra- and the marginal layer becomes white matter.
cellular spaces by either passive diffusion or Ciliated epithelial cells line the neural tube,
bulk flow. White matter tracts permit the uni- and cilia persist in some regions of the adult
directional movement of ISF, while gray mat- human ependyma.
ter has random flow through a dense neuropil. The neural tube is formed from neuroepi-
thelial cells that extend from the internal to the
external limiting membranes.4 Nuclei synthe-
Table 1–2 Circulation of Brain Fluids sizing DNA are found near the external limit-
Controlled by Types of Junctions ing membrane and migrate toward the inner
Between Cells limiting membrane. Once DNA synthesis is
complete, these cells become the neuroblasts
Blood and lymph form the first and second that form the mantle layer. When neuroblasts
circulations
mature into the neuronal cells of the adult, they
CSF and ISF form the third circulation lose their ability to divide. Neurons establish
CSF and ISF are formed at the choroid plexuses synapses with specific nuclear groups, probably
and cerebral capillaries
on the basis of chemical affinities (Figure 1–5).
CSF and ISF are contiguous across ependymal and After neuroblast differentiation has ceased,
pial surfaces
future glial cells are formed from neuroepi-
Drainage back into the blood occurs at arachnoid thelial cells that have differentiated with
granulations
glioblasts. Ependymal cells are formed along
 Neural groove
Amnion

(A)
Mesoderm Notochordal
plate
Neural groove Neurao-ectodermal
Embryonic junction
ectoderm

(B)

Somite
Neural crest
Embryonic
ectoderm

(C)
Neural tube Somite

Ependyma
Dermatome
(D) Dorsal root
Myotome ganglion
Ant. spinal
root

Ependyma
Dorsal root Mantle layer
ganglion
Marginal
(E)
layer
Dermatome
Ant. horn
cells
Myotome Sympathetic
cells

Alar plate
Dorsal root (sensory)
(F) ganglion
Sulcus limitans

Preganglionic Basal plate

ramus Ant. horn (motor)
Postganglionic Muscle
ramus
Sympathetic
ganglion
Visceral
ramus

Figure 1–4. Stages in the development of the neural tube from the neural plate with subsequent formation of the spinal
cord. (From Ref. 29.)

7
8 Molecular Physiology and Metabolism of the Nervous System

Ependymal
Neuroepithelial
cell
cells

Glioblast

Apolar neuroblast

Oligodendrocyte

Bipolar neuroblast

Protoplasmic Fibrillar
astrocyte astrocyte

Multipolar
neuroblast Mesenchyme cell Microglia

Figure 1–5. Diagram of the histogenesis of neurons and neuroglial cells. Neuroblasts, glioblasts, and ependymal cells orig-
inate from neuroepithelial cells. The origin of the oligodendrocyte is obscure, but both protoplasmic and fibrillary astrocytes
are derived from glioblasts. The microglia are considered to arise from mesenchyme. (From Ref. 30.).

with glioblasts. Ependymal and subependymal during the S phase. Using immunofluorescent
cells form a separate unit loosely attached to labeling for BrdU and for one of the neuronal
the outer limiting membrane. antigens, that is, a marker for mature neurons—
Cells in the subependymal zone of the lateral NeuN (neuronal nuclei), calbindin, or neuron
wall of the lateral ventricles continue to divide specific enolase—the researchers demonstrated
throughout life and proliferate after an injury that new neurons, as defined by these markers,
to participate in the repair process. Stem cells are generated from dividing progenitor cells in
differentiate into a wide variety of cells but the dentate gyrus of adult humans, showing that
mainly form glial cell types including astrocytes human hippocampus retains its ability to gener-
and oligodendroglial cells.5 The hippocampus ate neurons throughout life.6 The field of stem
is another region that has plasticity based on cell biology has grown dramatically since this
stem cells. The discovery that there is contin- seminal observation.7,8
ued growth of brain cells in adults was made by
studies of patients dying of terminal cancer who
had been injected with a molecule incorporated Cilated Ependymal Cells
into dividing cells prior to death; their brains and CSF Movement
were studied after death. Human brain tissue
was obtained postmortem from patients who Ependymal cells that line the walls of the ven-
had been treated with the thymidine analog tricular system in the adult brain are ciliated epi-
bromodeoxyuridine (BrdU), which labels DNA thelial cells. These polarized epithelial cells are
 1 Anatomy of Fluid Interfaces 9

thought to propel CSF through the ventricles Blood vessels in the choroid plexus lack tight
by the action of the cilia. Early in neurodevel- junctions, allowing substances from the blood
opment, the embryonic ventricles are lined by to pass through the vessels’ pores, as in the
a germinal epithelium. This embryonic neu- systemic circulation. Ependymal cells pro-
roepithelium has planar polarity that drives vide the barrier, and the tight junctions have
morphogenetic movements essential for neu- shifted to the apical surface of the ependymal
ral tube closure. Radial glial cells contain both cells that line the choroid plexus. For reasons
spatial and temporal patterning that determines that are unclear, there is a sharp demarcation
the fate and position of the cells in the devel- between the type of ependymal cell that lines
oping brain, and a subpopulation of radial glia the ventricles and those that form the choroid
transform into ependymal cells.9 Ependymal plexuses.
cells extend multiple motile cilia from their api- Mesenchyme outside the outer limiting
cal surface into the ventricles. Planar-polarized membrane condenses to become the outer
beating of these cilia generates directed CSF covering layers of the brain, namely, perios-
flow and helps maintain CSF homeostasis. teum, dura, and arachnoid. Fine trabeculi
Ependymal-generated CSF flow establishes join the arachnoid to the pia, and CSF fills the
gradients of chemorepellents that guide the space between the two membranes. The dura
migration of young neurons in the adult mam- thickens into a tough connective tissue, and
malian subventricular zone.10 Radial glia in the space between the dura and arachnoid is
the embryo have a translational polarity that absent except in pathological situations, such as
predicts the orientation of mature ependymal a subdural hematoma or empyema.
cells, which suggests that ependymal planar cell The mechanisms involved in the expan-
polarity is a multistep process initially organized sion of the ventricles are poorly understood.
by primary cilia in radial glia and then refined It is suggested that the large amount of pro-
by motile cilia in ependymal cells.11 tein, albumin, in the fluid in the ventricles of
the newborn, which is formed by the choroid
plexus, may in fact participate in the expansion
Choroid Plexuses, Arachnoid, of the ventricles by creating high oncotic pres-
and Capillaries sure inside the ventricles of the newborn.12
Development of the cerebral capillaries
The choroid plexuses are formed in specialized depends on trophic factors secreted by brain
regions where underlying blood vessels grow tissue. Proof that the brain has trophic factors
and push out the ependyma (Figure 1–6). that determine the type of vessel formed comes

(A) (B)

(C)

Figure 1–6. (A) Scanning electron micrograph of the ventricular system of the adult cat. (B) Choroid plexus with microvilla.
(C) Third ventricle with cilia. (Courtesy of Dr. Linda Saland.)
10 Molecular Physiology and Metabolism of the Nervous System

from elegant transplant experiments between extensive malformations of the cerebellum and
quail and chick embryos. Transplantation brainstem, which are generally incompatible
of quail brain into embryonic chick cultures with life (Table 1–3).
resulted in the formation of systemic vessels,
while nonbrain tissue transplanted into embry-
onic brain produced capillaries with tight
junctions.13 EXTRACELLULAR SPACE AND
Failure of neural tube closure in the first EXTRACELLULAR MATRIX
trimester of embryonic life results in congen-
ital anomalies. Dysraphism refers to a group Physiological studies demonstrated the pres-
of congenital malformations in which the pos- ence of a significant extracellular space that
terior part of the neural tube fails to close.14,15 is necessary for flow of the CSF/ISF through
Failure of posterior closure produces develop- brain tissue. Rall and colleagues performed
mental disorders that range from spina bifida ventriculocisternal perfusions with an inert
occulta, an incidental finding discovered on substance, inulin, that remained in the extra-
routine spinal x-ray, to myelodysplasia, a severe cellular space. By sampling tissues surround-
deformation that involves failure of closure of ing the ventricles, they found an extracellular
both the midline structures in the posterior space of 15% to 20%, which was consistent with
fossa of the brain and the central canal of the a third circulation.16 In other organs, the cells
spinal cord and can lead to death. Commonly are embedded in a connective tissue matrix that
encountered dysraphic syndromes include contains collagen fibers. However, the extracel-
absence of cerebral hemisphere development lular matrix of the brain has very little collagen,
(anencephaly), failure of vertebrae and skull to and the neurons are embedded in a matrix of
close (spina bifida and cranium bifidum), and extracellular matrix molecules with glial cells.
the combined spinal and nervous tissue abnor- Better definition of the glial cell membranes
malities of the Chiari malformations. In Chiari has shown the presence of a space that contains
type I malformation there is a protrusion of an complex carbohydrates, such as the glycosamin-
elongated cerebellar tonsil into the foramen oglycans, heparin sulfate, chondroitin sulfate,
magnum. When these patients begin to have dermatan sulfate, and hyaluronic acid.17
symptoms, usually in adult life, they have signs Extracellular matrix can inhibit neural cell
of lower brainstem dysfunction. Patients with migration. Schwann cell migration is integrin-
type II Chiari malformation have meningomy- dependent and is inhibited by astrocyte-
elocele; hydrocephalus is often present at birth produced aggrecan. Transplantation of
or becomes manifest when the spinal defect is Schwann cells is suggested as a potential treat-
repaired. Chiari type III and type IV are more ment for spinal cord injury. However, following

Table 1-3 Common Dysraphic Syndromes Involving Developmental

Malformations of the Cerebellum, Fourth Ventricle, and Posterior Fossa
Posterior Fossa
Malformations Anatomical Features Clinical Features
Chiari Type I Protrusion of an elongated Symptoms usually begin in adult life with
cerebellar tonsil into the signs of lower brainstem dysfunction
foramen magnum
Chiari Type II Meningomyelocele Hydrocephalus is often present at birth
or becomes manifest when the spinal
defect is repaired
Chiari Types III and IV Extensive malformations of the Generally incompatible with life
cerebellum and brainstem
Dandy-Walker Syndrome Agenesis of the cerebellar vermis, Hydrocephalus and other anomalies of the
cystic dilatation of the fourth brain
ventricle, and enlargement of
the posterior fossa
 1 Anatomy of Fluid Interfaces 11

transplantation Schwann cells show limited Table 1-4 Types of Cell Junctions at
migratory ability, and they are unable to inter- Major Sites of Brain Interfaces
mingle with the host astrocytes. Aggrecan
produced by astrocytes is involved in the inhi- Tight Junctions
bition of Schwann cell motility on astrocytic Choroid plexus ependymal cells (apex)
monolayers. Knockdown of this proteoglycan Arachnoid
in astrocytes using interfering RNA (RNAi) Cerebral endothelial cells
that blocks the action of the normal RNA mol- Gap Junctions
ecules or digestion of glycosaminglycan chains Ependymal cells
on aggrecan improves Schwann cell migration. Pia cells
Aggrecan acts by disrupting integrins on
Schwann cells.18
Injury to the adult central nervous system
increases the levels of extracellular matrix mol- the endothelial cells, as well as at the inter-
ecules, which inhibit repair of injured axons. face of the choroid plexus and the arachnoid
Chondroitin sulfate chains on proteoglycans (Table 1–4). At the ependymal and pial cells,
and enzymes necessary for their synthesis are gap junctions separate the CSF and ISF.
expressed after an injury. Microglial cells at Blood vessels have evolved to internalize the
the injury sites express both keratan sulfate delivery of nutrients to cells. In single and sim-
and chondroitin sulfate. Transforming growth ple multicell organisms the cell surface was ade-
factor-β (TGF-β) induces the expression of the quate, but as the complexity increased, another
enzymes involved in the synthesis of keratan solution was needed. A system of tubes formed
sulfate and chondroitin sulfate as well as the inside the organs that need the nutrients and a
expression of the chondroitin/keratan sulfate fluid to carry them gradually evolved. The pro-
proteoglycan aggrecan. Transforming growth cess of blood vessel formation, which is called
factor-β induces basic fibroblast growth fac- vasculogenesis for new vessel formation and
tor (bFGF) expression in microglia, and bFGF angiogenesis for sprouting from existing vessels,
induces TGF-β expression in astrocytes. Thus, is critical for normal development and for repair
the biosynthesis of keratan sulfate and chon- of tissues after an injury. The end result of these
droitin sulfate is upregulated in common by processes is a densely packed network of arter-
TGF-β in microglia.19 ies, arterioles, capillaries, venules, and veins.
Success of the stem cell transplantation pro- Observations of the vessels on the brain’s
cedures depends to a large extent on the ability surface suggested that there were anastomo-
of the transplanted cells to disrupt the extra- ses between the arteries and the veins. Ernst
cellular matrix and move toward the site of Scharrer made the important observation that
injury. Under normal conditions, movement of arteries rarely form anastomoses with veins
cells within the extracellular matrix is difficult, and that the arteries are more numerous than
but when injury is present and there is secre- the veins.20 Arteries enter the brain separately
tion of extracellular molecules by the reactive from the surfaces, giving off branches in the
astrocytes and microglia, the task becomes layers of the cortex, joining the draining veins
almost impossible. Altering the conditions of through a network of capillaries. This was a
the extracellular matrix by enzymes, such as seminal observation based on a relatively simple
hyaluronidase and chondroitinase, facilitates technique that reversed decades of erroneous
cellular mobility. observations. The finding that cerebral arteries
end in capillaries without anastomoses elimi-
nated the confusion about the vulnerability of
the brain to ischemia since it was now clear that
BRAIN-FLUID INTERFACES the brain had end arteries with poor collater-
alizations. These early observations have been
Anatomy of the Cerebral confirmed with studies that utilize colored
Blood Vessels plastics to show arteries (red plastic) and veins
(blue plastic). Plastic casts of the blood vessels
Specialized cell-to-cell junctions occur at each revealed dense networks of vessels that were
of the sites of contact between brain fluids. Tight necessary to ensure adequate perfusion of the
junctions are present at the major interface of entire brain (Figure 1–7). Elegant images of
12 Molecular Physiology and Metabolism of the Nervous System

(A)

(B)

Figure 1–7. Blood vessels in the human brain. (A) Brain vasculature on the surface of the brain. Red plastic fills the arter-
ies and blue plastic fills the veins. There are no anastomoses of the arteries and the veins since the arteries are end arteries.
(B) This schematic drawing demonstrates that the arteries pass from the surface to the deep white matter, traversing the six
layers of the cortex. Arterioles give off branches as the arteries pass through the cortex. The capillaries that join the arteri-
oles and the veins are not seen. (From Ref. 31; See also the color insert.)

the surfaces of the brain confirmed the rela- structures, the deep white matter, which is at the
tionships of the arteries to the veins shown end of the arterial supply, is vulnerable to either
earlier, with the added benefit of indicating changes in blood flow or oxygenation of the
the extensive branching of the arteries in the blood. Patients with hypoxic/ischemic injury due
layers of the cortex. Layers of the cortex with to loss of cerebral blood flow often have damage
large neurons received the greatest number of to the white matter with death of the oligoden-
arterial branches, as would be expected due to drocytes. Deep white matter is a frequent site
their greater metabolic need. for hypoxic/ischemic damage in the newborn.
As a consequence of the poor collateralization Increased cerebral blood flow to the inner
and the flow of blood from the cortex to deeper layers of the cortex can be seen with functional
 1 Anatomy of Fluid Interfaces 13

magnetic resonance imaging, which shows the a large number of mitochondria in the cyto-
subtle changes in blood oxygenation related to plasm, and a high energy requirement. Tight
metabolic activity. When a region in the cor- junctions join their apical surfaces. These cells
tex is activated, a fall in the oxygenation of the have features in common with those in the kid-
blood vessels produces a change in the nuclear ney that are also involved in active transport.21
magnetic resonance signal that can be visual- Cerebrospinal fluid has two major sources:
ized when very fast scans are made. the choroid plexuses in the ventricles and the
Endothelial cells line the cerebral blood ves- cerebral capillaries. Choroid plexuses are out-
sels. Except in a few specialized regions, these pocketings that contain fenestrated blood ves-
vessels are joined together by self-assembling sels, a stroma, and a layer of epithelial cells
proteins that form the tight junctions. Around that have tight junctions and actively secrete
the vessels is a basal lamina with pericytes. CSF (Figure 1–9). The ependymal cells of the
Astrocytic endfeet surround the capillaries. choroid plexus have microvilla but lack cilia.
Close by, but separate from the vessels, are the Cerebrospinal fluid absorption occurs at the
neurons and microglia (Figure 1–8). arachnoid through arachnoid granulations in
the sagittal sinus. The arachnoid is one of the
tight junction sites except in the region of the
arachnoid granulations where CSF absorption
Brain Cell Interfaces with CSF takes place across one-way valve structures.
at Ependyma and Pia The collapsing of the one-way valves when
blood pressure increases prevents the backflow
A heterogeneous layer of epithelial cells lines of red blood cells into the CSF.
the ventricular surface and choroid plexuses. In the region of the floor of the third ventri-
Over the choroid plexuses the epithelial cells cle, called the median eminence or infundibu-
are cuboidal in shape and have microvilli on the lum, specialized cells called tanycytes connect
apical surface next to the CSF; the microvilli are the hypothalamic nuclei with the ventricu-
short protrusions from the surface of the cells lar surface. On electron micrographs these
that increase the surface area. Choroid plexus tanycytes appear to extend to the surface of
epithelial cells have nuclei in the basal region, the ependyma. The ventricular surface of the

Pericyte

NUC

Mitochondria
Tight Junction
NUC
ENDOTHELIAL CELL

Basal Lamina

Astrocyte
Process END END

BL
PER AS

Figure 1–8. Schematic from electron micrograph of an endothelial cell. Note the large number of mitochondria, the tight
junctions, and the surrounding basal lamina and astrocyte foot processes. Nucleus (NUC), Inset: higher magnification of
two endothelial cells (END) with a basal lamina (BL) containing an embedded pericyte (PER) and an adjacent astrocyte
foot process (AS). (Adapted from Ref. 32.)
14 Molecular Physiology and Metabolism of the Nervous System

(A)

(B)

Figure 1–9. (A) Choroid plexuses showing the ependymal cells with dark nuclei and a cuboidal shape. Beneath the
ependymal cells are the stroma with embedded blood vessels. A rare calcereous deposit is seen in the cores (arrow).
(B) Higher power image. (Courtesy of Dr. Mario Kornfeld.)

tanycytes has microvilli rather than cilia. The third ventricle has tight junctions that limit the
tanycytes are connected by tight junctions movement of substances between the hypo-
forming a diffusion barrier at the ependyma thalamic nuclei and the CSF. Thus, substances
that restricts the movement of molecules that enter the brain in the hypothalamic region
from the CSF to median eminence structures. are restricted from moving into the CSF and
As the tanycyte processes pass through the confined within the brain.
median eminence, they end on capillaries. The The surfaces of the choroid plexuses are cov-
anterior region of the third ventricle contains ered with microvilli. Clefts between cells are
the circumventricular organs, including the seen to extend from the basal surface up to the
median eminence, organum vasculorum of the apical tight junctions. The choroid plexuses are
lamina terminalis, subfornical organ, subcom- similar to other secretory epithelia. The fully
missural organ, neural lobe, pineal gland, and developed choroid plexus cell has numerous
area postrema. The BBB is lacking here, expos- mitochondria, a Golgi complex, an endoplas-
ing hypothalamic cells to the circulating blood, mic reticulum, and small vesicles.21 Occasional
which is important since nuclei in these areas cilia protrude from between the microvilli on
can act as chemical sensors.22 The extracellu- the surface, which may expand the secretory
lar space of the median eminence is exposed components.
to substances in the blood that can modulate The final site where the blood and the CSF
release of the hypothalamic releasing factors. come into contact is at the arachnoid villi. As
To compensate for the absence of the BBB, the at other interfaces, the arachnoid cells cov-
ependyma over the hypothalamic region of the ering the brain’s surface are joined by tight
 1 Anatomy of Fluid Interfaces 15

junctions.23 Over the sagittal sinus, the arach- collagen fibers. Finally, a very thin layer of
noid cells form villi that protrude into the dural leptomeningeal cells is found. Traversing the
sinuses. Electron microscopy of the arachnoid subarachnoid space are sheets of trabeculae
villi suggests that there are continuous chan- that are formed from collagen fibers and con-
nels through them.24 An important function tain small blood vessels. The collagen bundles
of the arachnoid villi is to prevent blood from of the trabeculae are continuous with those in
the venous sinus from entering the CSF. The the subpial space.
valve-like channels that collapse when pressure
is applied from the blood side and open when
the CSF pressure increases accomplish this.
Even when the sinus pressure exceeds that in DURA, ARACHNOID,
the CSF, there is no reversal of flow. Thus, the AND PIAL LAYERS
arachnoid villi act as one-way valves that open
with pressure to allow CSF to drain into the Directly beneath the skull is the dura, a tough
sinuses but close when sinus pressure exceeds membrane and an important structure that
that in the CSF to prevent backward flow of prevents the spread of infection from the skull
blood. An unresolved issue is whether these into the brain and contains the CSF. Tears in
villi are actual channels or merely a series of the dura can occur when the skull is fractured
vesicles that can coalesce to form pseudochan- or during surgery. Once the dura is damaged,
nels. Arachnoid cells with the capacity to drain the CSF can leak out; this leads to symptoms
into veins have been found along the spinal of headache from low CSF pressure or from
cord at the interface of the arachnoid with the the introduction of infection into the central
spinal roots.25 nervous system with meningeal irritation.27
Electron microscopic studies of the arach- Beneath the dura is the arachnoid, whose
noid reveal a multilayered structure in cells are joined together by tight junctions.
humans.26 Five or six layers of cells form Subarachnoid spaces are filled with CSF. Over
the subdural mesothelium. Directly below the surface of the brain is a layer of pial cells
this layer is the central portion formed from that are bathed by CSF. The pia is joined
closely opposed polygonal cells joined by des- together by gap junctions, and similar to the
mosomes and tight junctions; this is the barrier ependymal cells, substances injected into the
layer. The inner layers consist of more loosely CSF can cross the pial surface and enter brain
packed cells that are separated by bundles of tissue.

BLOOD BRAIN CSF

Glucose Glucose Pyruvate Lactate

OXALOACETATE ACETYL CoA

ADP NAD
TCA
O2 Glutamine
NADH Cycle
2 K+ NH3
ATP ATP SUCCINATE α - KETOGLUTARATE
ase Glutamate
3 Na+
GABA
H2O

CAPILLARY EPENDYMA

Figure 1–10. Chemiosmotic work that converts energy sources, such as glucose and oxygen, to ATP and other essential
molecules. Glucose forms pyruvate and lactate (under anaerobic conditions). Pyruvate enters the TCA cycle to form ATP and
amino acids, including glutamate, glutamine, and gamma-aminobutyric acid (GABA). The formation of ISF is accomplished
by an ATPase pump that exchanges 3 Na+ for 2 K+, creating an osmotic gradient along which water passively flows.
16 Molecular Physiology and Metabolism of the Nervous System

Glucose

Glycolysis 2 ATP
O2

2 NADH 2 Pyruvate

Cytosol

TCA
NADH Cycle

Respiratory Chain 26 ATP

Mitochondria 12 H2O

Figure 1–11. Conversion of oxygen and glucose to energy by glycolysis and the respiratory chain. Glycolysis yields only 2
ATP in the absence of oxygen. When oxygen is added, an additional 26 ATP are formed through the TCA. NADH is nico-
tinamide adenine dinucleotide, an electron donor essential for metabolism. ATP is adenosine triphosphate.

WHAT ARE THE SOURCES dinucleotide (NADH), which, in turn, provide

OF ENERGY? the energy to run the membrane pumps of the
chemiosmotic machinery (Figure 1–10). Oliver
The brain is the most metabolically active organ H. Lowry developed the analytical methods to
in the body. Accounting for only 2% of the weight measure small amounts of energy metabolites
of the body, the brain requires approximately and used these methods to describe the loss of
20% of the body’s energy supply. To maintain ATP and other energy-generating molecules
this very large supply of energy, a mechanism with acute ischemia.28
of autoregulation of cerebral blood flow pre- Glucose conversion to ATP is the source of
serves flow through a wide range of blood pres- the energy used for chemiosmotic work, such
sures, breaking down only at the extremes of as membrane pumping, constriction of blood
very high blood pressure, when disruption of vessels, and secretion of CSF and ISF. Under
the blood vessels leads to vasogenic edema, or aerobic conditions, glycolysis of glucose cre-
very low pressure, when syncope occurs. This ates 2 molecules each of ATP, NADH, and
autoregulated system guarantees a steady sup- pyruvate. Pyruvate enters the tricarboxylic acid
ply of oxygen and glucose for all of the brain’s (TCA) cycle (citric acid cycle or Krebs cycle) to
metabolic needs under a wide variety of poten- form 26 molecules of ATP (Figure 1–11).
tially damaging circumstances. Because of the
great need for glucose by brain cells, special- REFERENCES
ized transport mechanisms at the endothelial
cell level carry glucose into the brain across the 1. Cushing H. The third circulation and its channels.
cerebral blood vessels. Lancet. 1925;209:851.
Glucose can be converted into the neu- 2. Iadecola C, Nedergaard M. Glial regulation of the
cerebral microvasculature. Nat Neurosci. 2007;10:
rotransmitter glutamate, which is the most 1369–1376.
abundant amino acid in the brain Along with 3. Agre P. Homer W. Smith award lecture. Aquaporin
glucose, blood delivers oxygen; both substrates water channels in kidney. J Am Soc Nephrol. 2000;11:
are needed for energy generation, which 764–777.
occurs during oxidative respiration. Glucose 4. Langman J, Guerrant RL, Freeman BG. Behavior
of neuro-epithelial cells during closure of the neural
and oxygen are the major fuels for the meta- tube. J Comp Neurol. 1966;127:399–411.
bolic pathways that form adenosine triphos- 5. Kazanis I. The subependymal zone neurogenic niche:
phate (ATP) and reduced nicotinamide adenine a beating heart in the centre of the brain: How plastic
 1 Anatomy of Fluid Interfaces 17

is adult neurogenesis? Opportunities for therapy 19. Yin J, Sakamoto K, Zhang H, et al. Transforming
and questions to be addressed. Brain. 2009;132: growth factor-beta1 upregulates keratan sulfate and
2909–2921. chondroitin sulfate biosynthesis in microglias after
6. Eriksson PS, Perfilieva E, Bjork-Eriksson T, et al. brain injury. Brain Res. 2009;1263:10–22.
Neurogenesis in the adult human hippocampus. Nat 20. Scharrer E. A technique for the demonstration of the
Med. 1998;4:1313–1317. blood vessels in the developing central nervous system.
7. Robel S, Berninger B, Gotz M. The stem cell potential Anat Rec. 1950;107:319–327.
of glia: lessons from reactive gliosis. Nat Rev Neurosci. 21. Tennyson VM, Pappas GD. The fine structures of the
2011;12:88–104. choroid plexus adult and developmental stages. Prog
8. Lazarov O, Mattson MP, Peterson DA, et al. When Brain Res. 1968;29:63–85.
neurogenesis encounters aging and disease. Trends 22. Broadwell RD, Oliver C, Brightman MW. Localization
Neurosci. 2010;33:569–579. of neurophysin within organelles associated with pro-
9. Spassky N, Merkle FT, Flames N, et al. Adult ependy- tein synthesis and packaging in the hypothalamoneu-
mal cells are postmitotic and are derived from radial rohypophysial system: an immunocytochemical study.
glial cells during embryogenesis. J Neurosci. 2005;25: Proc Natl Acad Sci USA. 1979;76:5999–6003.
10–18. 23. Nabeshima S, Reese TS, Landis DM, et al. Junctions
10. Sawamoto K, Wichterle H, Gonzalez-Perez O, et al. in the meninges and marginal glia. J Comp Neurol.
New neurons follow the flow of cerebrospinal fluid in 1975;164:127–169.
the adult brain. Science. 2006;311:629–632. 24. Tripathi BJ, Tripathi RC. Vacuolar transcellular chan-
11. Mirzadeh Z, Han Y-G, Soriano-Navarro M, et al. nels as a drainage pathway for cerebrospinal fluid.
Cilia organize ependymal planar polarity. J Neurosci. J Physiol (Lond). 1974;239:195–206.
2010;30:2600–2610. 25. Welch K, Pollay M. Perfusion of particles through
12. Knott GW, Dziegielewska KM, Habgood MD, arachnoid villi of the monkey. Am J Physiol. 1961;201:
et al. Albumin transfer across the choroid plexus of 651–654.
South American opossum (Monodelphis domestica). 26. Alcolado R, Weller RO, Parrish EP, et al. The cranial
J Physiol. 1997;499(pt 1):179–194. arachnoid and pia mater in man: anatomical and ultra-
13. Stewart PA, Wiley MJ. Developing nervous tissue structural observations. Neuropathol Appl Neurobiol.
induces formation of blood-brain barrier characteris- 1988;14:1–17.
tics in invading endothelial cells: a study using quail- 27. Schievink WI. Spontaneous spinal cerebrospinal fluid
chick transplantation chimeras. Dev Biol. 1981;84: leaks and intracranial hypotension. JAMA. 2006;295:
183–192. 2286–2296.
14. Juranek J, Salman MS. Anomalous development of 28. Goldberg ND, Passonneau JV, Lowry OH. Effects
brain structure and function in spina bifida myelom- of changes in brain metabolism on the levels of cit-
eningocele. Dev Disabil Res Rev. 2010;16:23–30. ric acid cycle intermediates. J Biol Chem. 1966;241:
15. Bejjani GK. Definition of the adult chiari malforma- 3997–4003.
tion: a brief historical overview. Neurosurg Focus. 29. Schade JP, Ford DH. Basic Neurology. Amsterdam:
2001;11:1–8. Elsevier; 1965.
16. Rall DP, Oppelt WW, Patlak CS. Extracellular space 30. Carpenter MB, Sutin J. Human Neuroanatomy.
of brain as determined by diffusion of inulin from the Baltimore: Williams & Wilkins; 1983.
ventricular system. Life Sci. 1962;1:43–48. 31. Duvernoy HM, Delon S, Vannson JL. Cortical blood
17. Margolis RU, Margolis RK. Nervous tissue proteogly- vessels of the human brain. Brain Res Bull. 1981;7:
cans. Dev Neurosci. 1989;11:276–288. 519–579.
18. Afshari FT, Kwok JC, White L, et al. Schwann 32. Peters A, Palay SL, deF Webster H. The Fine
cell migration is integrin-dependent and inhibi- Structure of the Nervous System: Neurons and Their
ted by astrocyte-produced aggrecan. Glia. 2010;58: Supporting Cells (3rd Edition). Oxford University
857–869. Press, New York, 1991. Pp. 349–351.
Chapter 2

Physiology of the Cerebrospinal

and Interstitial Fluids

INTRODUCTION MENINGES AND SITES OF MASSES

PROTEINS IN THE CSF AND INFECTION

CSF PRESSURE REFLECTS VENOUS INTERSTITIAL FLUID

PRESSURE IN THE RIGHT HEART LYPHATIC DRAINAGE
Formation, Circulation, and
WATER DIFFUSION, BULK FLOW OF ISH,
Absorption of CSF
AND DIFFUSION TENSOR IMAGING
Formation of CSF by Choroid Plexuses
Choroid Plexus and Disease Biomarkers NEUROPEPTIDES AND FLUID
in CSF HOMEOSTASIS
Absorption of CSF at the Arachnoid Villi AQUAPORINS AND WATER TRANSPORT
ELECTROLYTE BALANCE IN THE CSF IN THE CENTRAL NERVOUS SYSTEM

INTRODUCTION care. Lumbar puncture is performed with a hol-

low needle containing a stylet, which is inserted
Lumbar puncture for measurement of cerebro- into fluid space containing the spinal nerve
spinal fluid (CSF) pressure and analysis of the roots below the termination of the spinal cord
fluid is one of the most commonly employed around L1. Generally, the needle is inserted in
diagnostic tests in neurology. Measurement the L3-L4 space, which is the location of the
of intracranial pressure and examination of pooled fluid. Ideally, the patient lies in the lat-
the protein and cellular content of the fluids eral recumbent position, and the test is done
provide invaluable information for diagnosis under sterile conditions at the bedside. Once
of serious diseases, such as brain infection, the needle is determined to be in the lumbar
increased intracranial pressure, multiple scle- sac, a manometer is placed on the needle and
rosis (MS), Alzheimer’s disease (AD), vascu- the CSF pressure is measured. After the sam-
lar dementia, and subarachnoid hemorrhage. ple is collected, it is sent for analysis of cells,
When removal of CSF is done carefully with- protein, and other factors that may be useful
out complications and the results are properly in diagnosis.
interpreted, lumbar puncture yields valuable Normal CSF is clear and colorless and resem-
information at a relatively low cost. Therefore, bles water. Three to four lymphocytes may be
it is important to understand the dynamics and present, but generally there are no neutrophils.1
impact of disease on the CSF. The protein content is 40 to 60 mg%, in con-
The method of removing fluid has changed trast to the 4 g of protein in the blood. The CSF
little since it was first introduced into clinical pressure is between 80 and 180 mm H2O in the
18
 2 Cerebrospinal and Interstitial Fluids 19

majority of patients. Low CSF pressure is asso- An IgG index that is higher than 0.6 is
ciated with headaches and may be indicative of abnormal. Any increase in the index is a reflec-
a CSF leak. It is important to obtain either a tion of IgG production in the CNS.
computed tomography (CT) or magnetic res- Measurement of other proteins in the CSF
onance imaging (MRI) scan to rule out a mass is an important diagnostic test in MS, and
lesion or obstructive hydrocephalus prior to CSF analysis is indicated in the workup of
performing a lumbar puncture. Otherwise, patients with suspected MS. In addition to
there is a risk of herniation. Emergency lumbar the IgG index, fragments of degraded mye-
puncture may be indicated when imaging is not lin in the form of myelin basic protein can be
available in patients with suspected bacterial measured. Detection of oligoclonal bands that
meningitis. differ from those present in the blood is highly
suggestive of an immunological reaction in
the central nervous system. Myelin basic pro-
PROTEINS IN THE CSF tein can be elevated in other acute neurologi-
cal disorders, and oligoclonal bands appear in
Protein entry into the CSF is restricted by the other neuroinflammatory conditions, but they
tight junctions at the interfaces between blood, are extremely useful to support the diagno-
CSF, and brain tissue. In spite of the presence sis when clinical and imaging studies are also
of a blood-brain barrier (BBB), small quanti- positive. The presence of myelin basic pro-
ties of albumin and even smaller amounts of tein is only found during the acute attacks,
other proteins are found in the CSF. An ele- while oligoclonal bands are more indicative
vated level of protein is a marker of an abnor- of a chronic process and remain elevated.
mal BBB. Albumin is produced by the liver The pattern of expression of the IgG index,
and circulates in large quantities in the blood. myelin basic protein, and oligoclonal bands is
While the level of albumin in the blood is important to determine the stage of the MS
around 4000 mg%, the normal amount in the attack. In the acute attack early in the illness,
CSF is 40 to 60 mg%. High levels of protein the IgG index and myelin basic protein are
in the CSF are suggestive of disruption of the elevated without oligoclonal bands. In the
BBB, but alterations of the albumin content in chronic stage there may be an elevated IgG
the blood can influence the amount of albumin index and oligoclonal bands without mye-
in the CSF. Therefore, it is more accurate to lin basic protein elevation.2 Once the oligo-
form the ratio of albumin in the CSF to that in clonal bands are increased in the CSF they
the blood. This is called the albumin index, and remain elevated, in contrast to myelin basic
it is a better indicator of the status of the BBB. protein, which falls after the acute process is
Small amounts of other proteins are found in over. Generally, in MS, there are fewer than
the CSF. These include immunoglobulins, par- 50 mononuclear cells; when the cell count
ticularly IgG, which is also at a much lower is higher, other causes, such as vasculitis or
level in the CSF than in the serum. An increase infection, should be suspected.
in the intrathecal production of IgG suggests Very high levels of CSF proteins can occur
an immunological or infectious process in the in meningitis, particularly due to fungal organ-
CSF. The IgG index is elevated in autoim- isms, and in the Guillain-Barré syndrome.
mune diseases such as MS. When infection or Brain and spinal cord tumors can some-
autoimmune reactions occur, there can be an times increase CSF protein levels, but often
increase in IgG. Abnormal permeability of the these levels are normal. While CSF studies
BBB allows IgG from the blood to enter the were once routinely done in patients with
CSF. Separation of IgG formed within the cen- brain tumors, the use of multimodal imag-
tral nervous system from the blood is critical in ing has replaced lumbar puncture. Generally,
diagnostic testing. Therefore, it is common to removal of CSF is not recommended in
calculate the IgG index, which is determined patients with brain tumors because of the
from the albumin in both compartments, using threat of herniation. Rarely, markedly ele-
the formula vated CSF protein will obstruct the outflow of
CSF, resulting in raised intracranial pressure
IgG index = (IgGcsf /IgGserum)/(Albcsf /Albserum) with papilledema.
20 Molecular Physiology and Metabolism of the Nervous System

CSF PRESSURE REFLECTS VENOUS which represents about 2% of the total volume.
PRESSURE IN THE RIGHT HEART When blood volume increases, there is a con-
comitant increase in CSF pressure. Any cause
The CSF is secreted by an energy-dependent of vasodilatation will increase CSF pressure.
process by choroid plexuses and capillaries. For example, lung disease causes vasodilata-
Measurements of CSF production indicate tion secondary to hypoxia, which often occurs
a rate of 0.3 mL/min. There is an estimated during the night, resulting in headaches, and
120 mL of total CSF in the ventricles, the on rare occasions the increase in pressure is
subarchnoid space, and the spinal region; sufficient to produce papilledema. Brain tissue
20 mL is found within the ventricles. Since constitutes the largest compartment. Masses,
about 500 mL of CSF is produced daily in such as hemorrhages and tumors, ischemic
humans, steady drainage is essential to avoid insults, and diffuse brain edema may lead to an
excess accumulation and increased pressure. increase in intracranial pressure.
Drainage of CSF occurs primarily across the Normal CSF pressure is between 80 and
arachnoid villi. 180 mm H2O and is dependent on the pres-
Cerebrospinal fluid, blood, and brain tis- sure in the superior sagittal sinus, which drains
sue contribute to the intracranial pressure. into the jugular veins. Since the pressure in the
Expansion of any of these components cre- right side of the heart is generally low, pres-
ates a life-threatening situation because of sure measured in the recumbent position at
the rigid skull encasing the brain. Alexander the time of lumbar puncture reflects the pres-
Monro noted in 1783 that since the brain was sure in the venous system rather than arterial
incompressible and the skull rigid, the amount pressure, which is measured in millimeters of
of blood leaving by the veins had to be the mercury3 (Figure 2–2). Another factor involved
same as that entering by the arteries. In 1824, in the venous control of CSF pressure is the
George Kellie confirmed Monro’s concept transmission of pressure through the thin-
experimentally when he observed that ani- walled veins, which occurs more readily than
mals killed by exsanguination had blood in the through the arterial system, the muscular wall
skull except where bone was removed with a of which exerts an opposing force. Thus, CSF
trephin. George Burrows extended their obser- pressure reflects the venous pressure and not
vation in 1846 to include the CSF volume, and the arterial pressure. When the patient takes a
the limitation to expansion is referred to as the deep breath, the venous pressure in the chest
Monro-Kellie doctrine (Figure 2–1). is reduced. This pressure is transmitted to the
Measurement of CSF pressure with a CSF through the venous return from the brain
manometer while the patient is in the lateral and the manometer pressure falls, which is an
recumbent position reflects the sum of sev- excellent way to determine that the spinal nee-
eral compartments. Within the cranial vault dle is correctly positioned.
the compartments are blood, CSF, and brain Elevated intracranial pressure is transmit-
tissue. The smallest compartment is the blood, ted through the CSF along the optic nerve to

Blood-2% CSF - 140 mL

Carbon dioxide retention Hydocephalus
Dural sinus thrombosis
Pulmonary hypertension

Brain Tissue
Brain tumors
Cerebral Infacts
Abscess
Brain cell toxin

Figure 2–1. The Monroe-Kellie doctrine states that three compartments determine the pressure in the CSF: blood, CSF,
and brain tissue.
 2 Cerebrospinal and Interstitial Fluids 21

100 system, which is called Batson’s plexus; because

spinal veins lack valves, they are a route for the
280
spread of cancer metastases and infection in
the lung to the spinal cord and the brain.
240 CSF pressure (mm saline)

200

FORMATION, CIRCULATION,
Millimeters

Venous pressure at torcula

160 ( mm saline)
AND ABSORPTION OF CSF
120
Arterial pressure
(mm Hg) Formation of CSF by Choroid
80 Plexuses
40
Choroid plexuses in the lateral, third, and
fourth ventricles secrete 30% to 60% of the
20
A B CSF. Formation of CSF by the choroid plexus
20 40 60 80 120 140 180 epithelial cells involves active secretion of
Minutes
fluid by ion pumps and enzymes. Ames and
Figure 2–2. Cerebrospinal fluid pressure is determined his colleagues developed methods for col-
by the pressure in the thin-walled veins rather than in the lecting the freshly formed CSF in animals by
muscular arteries. This is shown by the compression of the filling the ventricles with oil and trapping the
jugular vein in the neck at time A. The rise in CSF pres- newly secreted CSF for measurements of ion
sure is seen in comparison to the venous pressure in the
torcula. When the pressure is released, it returns to below content.5 In a series of elegant experiments,
the precompression value (B). Arterial pressure during the they measured the CSF content of potassium
maneuver remains stable. Note that CSF pressure is mea- and sodium, showing that it was regulated
sured in millimeters of water, while blood pressure is mea- within a defined range different from that of
sured in millimeters of mercury. (Adapted from Ref. 3.)
the blood. Since the ionic content of CSF dif-
fered from that expected for an ultrafiltrate of
the optic disc and can be seen as papilledema. plasma, they showed that CSF was formed by
Both eyes are generally involved in papille- an active secretory process, which kept potas-
dema except in unusual circumstances where sium levels in CSF within a very narrow range
one of the optic nerves is compressed. Foster- and allowed the sodium content to fluctuate
Kennedy syndrome is a rare condition due to more closely with serum levels.
compression of one optic nerve by a mass, such Sodium-potassium adenosine triphospha-
as a meningioma, causing optic atrophy in the tase (ATPase) and carbonic anhydrase are the
optic disc of the compressed nerve and papille- major enzymes that control CSF secretion
dema in the other eye. Although there is no (Figure 2–3). Secretion of fluid is relatively
exact number, pressures over 200 mm H2O are independent of outside influences, which leads
considered abnormal and often are associated to clinical problems. Few drugs reduce the
with signs of acute or chronic bilateral papille- secretion of CSF, leading to increased intra-
dema. Papilledema is not always present when cranial pressure with or without hydrocepha-
CSF pressure is increased. Large brain tumors lus when the outflow pathways are obstructed.
may be present without signs of papilledema Inhibition of ATPase by ouabin reduces
other than decreased venous pulsations in the CSF formation, but it is highly toxic to the
eyes. One study using cisternography suggested heart and can only be used in animal studies.
that tumors that interfered with outflow of CSF Acetazolamide, a carbonic anhydrase inhibitor,
were more likely to cause papilledema.4 also reduces CSF production and is used clin-
The spinal cord has a unique system of venous ically to treat idiopathic increased intracranial
drainage. There is an epidural space around the pressure. The osmotic agent mannitol reduces
spinal cord that is more extensive than the epi- CSF production and can be used for short peri-
dural space of the brain. The veins around the ods to reduce intracranial pressure.
spinal cord form a network of valveless veins Histochemical studies show that the sodium-
that connect the deep pelvic veins and thoracic potassium-dependent ATPase is located on the
veins to the internal vertebral venous plexuses microvilli of the apical (CSF-facing) surface of
22 Molecular Physiology and Metabolism of the Nervous System

Blood vessels in the stroma of the choroid

plexus are similar to systemic vessels since they
are fenestrated, permitting fluid, protein, and
electrolytes to escape into the stroma underlying
the ependymal cells of the choroid plexus. Tight
junctions at the epithelial cell-like ependymal
cells are located on the apical surface. Sodium
enters the epithelial cell in exchange for hydro-
3
Na Net 3Na gen. Sodium is removed from the cell by an
ATP
exchange at the apical surface of three intra-
cellular sodium ions for two potassium ions in
K 2K
the CSF; the extra ion pumped by ATPase into
the CSF increases the osmotic pressure on the
surface, which results in the formation of CSF.
Tight Sodium ions accumulate on the apical surface
junction as a result of the imbalance in their exchange
with potassium. Excess sodium increases the
osmolality at the CSF surface of the choroid
Water cell. Water is removed from the cell by the
osmotic pressure.
CO2 + H2O
While the exact mechanism of CSF forma-
– tion remains to be resolved, it is well accepted
H+ HCO3 that an excess of ATPase molecules found on
the apical end of the cleft creates an osmotic
gradient with an excess of sodium ions on
Figure 2–3. Mechanism of CSF formation at the choroid the apical surface. As part of CSF formation,
plexus. Ependymal cells are shown with tight junctions at
the apical surface. An ATPase pump creates an osmotic HCO3- and Cl- enter the CSF from plasma.
gradient by importing more sodium ions into the CSF than Bicarbonate plays an important role in CSF
potassium ions are removed. Another source of CSF is the formation by carbonic anhydrase. To balance
conversion of CO2 to HCO3 by carbonic anhydrase. the movement of HCO3−, there is a Cl-HCO3
exchange that is driven by the need to maintain
the epithelial cells of the choroid plexus and a steady-state intracellular HCO3- level.9
that carbonic anhydrase is within the cells.6 Although the primary driving force for CSF
Cyclic nucleotides appear to be involved in CSF formation is Na+/K+ exchange at the apical sur-
secretion; injection of cholera toxin into the face, the HCO3− ion is linked to CSF secretion
ventricle, which stimulates adenylate cyclase, by cyclic adenosine monophosphate (cyclic
increases the CSF formation rate.7 Adenylate AMP).10 Few substances increase the rate of
cyclase is located along the basal plasmalemma CSF formation. In underdeveloped countries
of the choroid plexus.6 with contaminated water supplies, cholera
Although the choroid plexus and the ependy- toxin, which is a potent stimulus of cyclic AMP,
mal cells lining the ventricles have a common is a cause of an increase in CSF formation.7
epithelial cell origin, they differ in their secre- The CSF formation rate varies for differ-
tory properties and enzymatic makeup. The ent species.11 In humans, the rate of CSF pro-
ATPase located on the apical surface of the duction is 0.3 mL/min. However, formation of
choroid plexuses appears on the basal (brain- CSF as a function of choroid plexus weight is
facing) surface of the ventricular epithelium. remarkably constant over a range of species
Adenylate cyclase is found on the apical rather (Table 2–1). Although several drugs can inhibit
than the basal surface of the ependymal cells. CSF production for a short period, they have
Ependymal cells outside the choroid plexus proven to be of limited clinical use because of
have gap junctions and they restrict free water either their time frame of action or their tox-
movement, which was shown when vasopres- icity. Acetazolamide, an inhibitor of carbonic
sin infused into the CSF in cats caused an anhydrase, reduces the CSF formation rate by
increase in water transport across the ependy- up to 40%. Inhibitors of ATPase, such as digi-
mal surface.8 talis, also reduce CSF formation; however, the
 2 Cerebrospinal and Interstitial Fluids 23

Table 2–1 Rate of CSF Formation neoplastic, and systemic diseases. Amyloid beta,
in Different Species Aβ, accumulates in the AD choroid plexus.16 In
MS, the choroid plexus could represent a site
mL/min/mg for lymphocyte entry in the CSF and brain and
Species mL/min Choroid Plexus for presentation of antigens.17 Measurement of
biomarkers is important in the diagnosis not
Rabbit 10 0.43 only of MS, but also of AD and vascular cogni-
Cat 20 0.5 tive impairment (VCI). Measurement of CSF
Dog 50 0.63 biomarkers, hyperphosphorylated tau (P-tau),
Goat 154 0.36 total tau (T-tau), amyloid β1–42 (Aβ (42)) and
Human 350 0.18 neurofilament light polypeptide (NF-L) in
Source: From Ref. 11. patients with mild cognitive impairment (MCI),
the early stage of AD, predicted VCI and AD
at follow-up. Increased baseline concentra-
effect is too short-lived to be of clinical impor- tions of NF-L significantly separated MCI-VCI
tance. Increased CSF pressure reduces CSF from stable MCI.18 In another study of CSF
formation only slightly.12 biomarkers in advanced AD, T-tau, P-tau, and
Table 2.2 lists factors that influence the rate Aβ(42) could predict cognitive progression, the
of CSF formation. Hyperosmolality produced outcome of cholinesterase inhibitor treatment,
by intravenous mannitol reduces CSF produc- and mortality in AD. A subgroup of patients
tion by 50%,13,14 and this drug is used clinically with AD with extremely high levels of CSF bio-
in patients with raised intracranial pressure markers exhibits worse clinical outcomes over
to temporarily lower the pressure.15 Mannitol time, including faster progression of cognitive
is effective in doses of 0.25 mg/kg, which is deficits, no response to cholinesterase inhibitor
below the amount needed to make a significant treatment, and higher mortality.19
change in plasma osmolality, suggesting that it
is working by other mechanisms. Hypothermia
influences CSF production by reducing cere-
bral metabolism.
Absorption of CSF at the
Arachnoid Villi

Choroid Plexus and Disease The absorption of CSF at the arachnoid villi
is pressure sensitive (Figure 2–4); as the CSF
Biomarkers in CSF pressure increases, so does the amount of CSF
The choroid plexus is involved in a variety of
neurological disorders, including neurodegen- 1.2
erative, inflammatory, infectious, traumatic,
CSF Formation (mL/mm)

0.8
Table 2–2 Factors That Influence CSF
Formation
0.4
Substance Site of Action
Normal CFS pressure
Increased Cholera toxin Adenylate
0
Production Adrenergic cyclase
stimulation Adenylate
0 68 100 200
cyclase
CSF Pressure (mm H2O)
Decreased Ouabain/digitalis Na+/K+-ATPase
Production Acetazolamide Carbonic Figure 2–4. Absorption of CSF is dependent on its pres-
Hyperosmolality anhydrase sure. The formation rate is constant at 0.34 mL/min, and
Hypothermia Choroid plexus the rate of absorption increases above a threshold shown
capillaries here at 68 mm H20 as an example. The point where CSF
Decreased formation and absorption cross determines the CSF pres-
metabolism sure, measured by lumbar puncture with the patient lying
on the side.
24 Molecular Physiology and Metabolism of the Nervous System

absorbed. Several explanations have been pro- absorbed into the blood across the arachnoid
posed to describe the absorption process. The villi; alternatively, it can mix with the CSF in
arachnoid villi appear to act as one-way valves, the spinal sac for subsequent removal into the
which open with raised pressure and close as vascular structures around the spinal cord or
the pressure falls.20 Larger particles are trapped transport up over the hemispheres.
by the arachnoid villi. Red blood cells from an Cisternography was developed at a time
intracranial bleed and white blood cells from when normal pressure hydrocephalus (NPH)
an infection can be seen in the villi. Clogging of was initially described and predicted to be a
the absorption channels by cells or high protein cure for dementia.22 This was prior to the real-
levels impedes absorption of CSF, causing an ization that AD was a major cause of demen-
increase in CSF pressure resulting in papille- tia, and enthusiasm for surgery to treat NPH
dema, and may lead under certain circum- waned. The CSF flow patterns can be deter-
stances to communicating hydrocephalus.21 mined clinically by the use of radioisotope cis-
Circulation of CSF begins in the cerebral ven- ternography, which involves the injection of
tricles with the fluid exiting through the foram- radioactive substances into the lumbar spinal
ina of Luschka and Magendie into the cisterna fluid.23 Cisternography is done with the gamma
magna. Flow of CSF within the subarachnoid emitter, technicium, labeled to diethylenetri-
space can follow two patterns (Figure 2–5). aminepentaacetic acid, a large inert molecule.
It can move up over the convexities to be After injection of the isotope into the lumbar
sac, nuclear brain scans are done at 2, 24, and
48 hours; the radioactive substances are trans-
(A) ported slowly toward the head. If the molecular
weight is large, the tracer ascends toward the
head but remains in the subarachnoid space,
where it exits through the sagittal sinus, which
Lat vent occurs within 24 hours under normal circum-
III stances. Smaller molecular weight substances
diffuse from the spinal canal through the vas-
cular plexuses surrounding the cord.24 When
NPH is present, the tracer enters the ventricles
IV and remains for up to 72 hours. If there is ven-
tricular enlargement due to atrophy (hydro-
Luschka & Magendie cephalus ex vacuo), the tracer can temporarily
enter the ventricle; however, it does not remain
as long as in NPH when there is transependy-
(B) mal absorption.25
Subarachnoid Several other uses remain for cisternography,
Space including determination of CSF leaks in patients
with difficult-to-diagnose presentations of spon-
taneous intracranial hypotension. Patients with
low CSF opening pressure often show contrast
enhancement on MRI. Radioisotope cister-
nography is an additional diagnostic method
to detect CSF leaks or pathological kinetics of
radioisotope movement, particularly in cases
with normal MRI findings.26 Low-pressure
Basal headaches can occur after lumbar puncture but
Cistern are usually transient. In an occasional patient,
a persistent CSF leak requires a blood patch,
Figure 2–5. Pathways of CSF drainage from the cere- which involves the injection of the patient’s
bral ventricles to the arachnoid villi. (A) The CSF moves blood into the site of the lumbar puncture to
from the ventricles through the foramina of Luschka and
Magendie into the subarachnoid space. (B) From the sub-
speed the healing of the tear. Trauma to the
arachnoid space it moves over the convexity to exit at the nasal area can lead to CSF leaks. Basilar skull
arachnoid villi. fractures are another cause.
 2 Cerebrospinal and Interstitial Fluids 25

Absorption of CSF is dependent on the state electrical resistance is regulated by the Na+/K+
of the valve-like mechanism in the arachnoid ATPase pump. Epithelial sheets have a high
villa. Several diseases lead to blockage of the electrical resistance due to the presence of tight
absorption pathways with increased intracra- junctions between the cells, and those with
nial pressure and papilledema. Subarachnoid leaky junctions have a lower resistance.32 Thus,
hemorrhage leads to the appearance of red measurement of the electrical resistance across
blood cells in the CSF. These cells are large the secreting epithelium gives an indication of
enough to clog the outflow passages. White the tightness of the junctions. The greater per-
blood cells are another potential blocker of the meability of the choroid plexus epithelium is
absorption mechanism. Rarely, CSF protein due in part to leaky intercellular junctions with
will be increased sufficiently to interfere with an electrical resistance of 26 ohm/cm2, which
absorption. is similar to that of other leaky epithelium. For
Resistance of CSF outflow across the sagittal comparison, the electrical resistance of frog
sinus can be measured by the infusion of arti- brain capillary is 1900 ohm/cm2 and that of
ficial CSF into the lumbar sac. This is another tight epithelium, such as toad bladder, is over
test developed for use in the diagnosis of NPH 4000 ohm/cm2.33
that is rarely performed. To measure the out- The potassium concentration is maintained
flow resistance, a spinal needle is inserted by within a very narrow range in the CSF.5 The
lumbar puncture and connected to a manom- normal CSF potassium level is approximately
eter. While the patient is in a recumbent posi- 3 mEq/L. Changes in plasma potassium have
tion, artificial CSF is infused slowly and the little effect on CSF potassium.34 Even at very
rise in pressure during the infusion is used to high plasma levels, CSF potassium remains
measure the outflow. Normal individuals can within the normal range.35 Transport across the
tolerate infusion of CSF at rates twice those BBB is limited, and the half-time of exchange
of production without an increase in pres- for potassium is 24 hours. When potassium
sure.27 When the CSF absorptive mechanism levels in the CSF are increased, sodium is
is impaired, there is a rise in pressure as the exchanged for potassium by an active transport
fluid is infused. mechanism. Potassium is critical for neuronal
function and affects the release of neurotrans-
mitters, making it important for it to be main-
tained at a constant level in the extracellular
ELECTROLYTE BALANCE fluid.
IN THE CSF Calcium in the CSF normally ranges
between 2 and 3 mEq/L in CSF compared to
Sodium is the most abundant ion in the CSF, plasma levels of 4 to 5.5 mEq/L.36 Calcium is
and it is important in transport and osmoregu- secreted from the choroid plexus and has a
lation. Tracer studies with 24Na have shown that similar value in various CSF spaces. The rate
the CSF and plasma levels are closely relat- of calcium entry from blood to CSF is relatively
ed.28 Acetazolamide, an inhibitor of carbonic independent of the serum calcium level. The
anhydrase, slows the entrance of radiolabeled ratio of CSF to serum Ca2+ in humans is around
sodium into the CSF.29 Vasopressin enhances 0.50.37 The low CSF levels of calcium are main-
the movement of sodium from blood to brain.30 tained by transport mechanisms between blood
The exchange time for 24Na transport from and CSF.
blood to brain is about 2 hours and depends on Both acute and chronic changes in plasma
the region sampled. calcium have little effect on brain calcium
Cerebral capillaries have a membrane per- levels. Fluctuations of plasma calcium from
meability to sodium of l.4 × l0−7 cm/s, which 1 to 7 mmol/L in dogs change CSF calcium
is similar to their permeability to mannitol and levels from 1 to 2 mmol/L; similarly, brain cal-
in the same range as tight-junctioned epithelial cium remains constant during acute changes.38
membranes.31 Both hypertonic saline and man- Young rats fed diets low or high in calcium
nitol are used to temporarily control increased showed a 40% fall or a 30% rise, respectively,
CSF pressure. The electrical resistance across in total plasma calcium; brain levels remained
a membrane is determined by the distribution within 10% of those in controls.39 Although cal-
of charged ions. In epithelial membranes the cium enters the brain at the various interfaces
26 Molecular Physiology and Metabolism of the Nervous System

comprising the BBB, transport across the cho- of cellular membranes and formation of prod-
roid plexus is the dominant route for calcium ucts of inflammation.43
entry from blood to brain.40
Regulation of calcium is essential for normal
brain function. Marked increases in brain cal-
cium produce impairment in thinking and can MENINGES AND SITES
lead to coma, while very low levels of calcium OF MASSES AND INFECTION
cause seizures.36 In order to maintain calcium
homeostasis, active transport of calcium at the Three layers of meninges surround the brain
BBB barrier is necessary. Both the cerebro- and spinal cord. Dura mater is the tough
vascular endothelium and the choroid plexus fibrous layer beneath the skull that forms the
participate in this process.39 inner layer of the cranial periosteum and tightly
Extracellular levels of unbound calcium are adheres to bone. Below the foramen magnum
higher than intracellular ones. Calcium within periosteum is separated from dura forming
the cell is sequestered in mitochondria and the epidural space, which is filled with fat.
smooth endoplasmic reticulum. Entry of cal- Arachnoid is the middle layer that is pressed
cium into the cell occurs either by a change in against the dura. Cerebrospinal fluid fills the
the voltage across the membrane that accom- subarachnoid space and is contiguous with the
panies depolarization or via agonist-operated glia limitans and pia mater covering the brain
channels activated by excitatory neurotrans- surface. Virchow-Robin spaces are lined with
mitters. During pathological changes such as pia mater as they surround the blood vessels
anoxia, the potassium concentration rises in entering the brain from the surface.
the extracellular space and the calcium level Understanding the layers overlying the brain
falls.41 The extracellular calcium enters the cell provides a rational explanation for the sites of
and leads to a cascade of molecular events that infections and mass lesions. Arteries are pre-
result in permanent cell damage. sent between the dura mater and the skull.
Postsynaptic calcium channels are activated When a fracture to the temporal bone tears the
by glutamate and aspartate. Both amino acids middle meningeal artery, a collection of blood
are excitatory neurotransmitters ubiquitously accumulates rapidly under pressure in an epi-
distributed in brain tissue. Two of the glutamate dural hematoma. If the injury results in tearing
receptors are transmembrane channels that are of the veins between the dura mater and the
named according to the dominant molecule that arachnoid, a subdural hematoma forms in the
excites them: N-methly-d-aspartate (NMDA) potential space.
and α-amino-3-hydroxy-5-methyl-4-isoxazole Infections in the sinuses, particularly the
propionic acid (AMPA) are ionotropic calcium ethmoid and the sphenoid, can spread into the
channels. A metabotropic receptor coupled subdural space to form a subdural empyema.
to a G-protein increases intracellular calcium. Because the pus forms a layer of fluid beneath
Glutamate excitatory receptors are mainly the dura mater, visualization by CT is not pos-
active in the synapses but have also been iden- sible and MRI is the best diagnostic test. While
tified on axons. Found in high concentrations subdural hematomas generally grow slowly
in the hippocampus and other regions sensi- and, if small, can be treated conservatively,
tive to ischemic-anoxic injury, they are active subdural empyemas cause seizures and brain
in consolidation of memory through long-term edema, and require high doses of antibiotics
potentiation. Glutamate ionotropic receptors and urgent surgical treatment to drain the pus.
open a sodium channel that allows sodium and Below the arachnoid is the subarachnoid
chloride to enter the cell and another chan- space that is crisscrossed with trabeculae and
nel that permits calcium to pass. The calcium filled with CSF. Meningitis is a collection
channel is strongly antagonized by magnesium, of cells in the subarachnoid space that can
which may be important in the therapeutic enter via blood or spread from the contiguous
action of magnesium.42 Although calcium entry sinuses. Rupture of aneurysms on arteries in
is a normal consequence of cell excitation by the subarachnoid space, commonly around the
glutamate, excess calcium within the cell can circle of Willis, results in severe headache with
lead to the activation of cellular processes that nuchal rigidity. Either blood or pus in the CSF
are detrimental to the cell, such as breakdown can interfere with absorption of the CSF by
 2 Cerebrospinal and Interstitial Fluids 27

blocking the channels in the arachnoid granu- the spinal cord, leading to paralysis. Suspected
lations, resulting in an increase in CSF outflow epidural abscesses can be seen on MRI and, if
resistance that raises the CSF pressure. detected, require surgical removal and drainage
Infections can spread into the brain along to prevent paraplegia (Figure 2–7).
the Virchow-Robin spaces that contain the
arteries entering the brain from the surface.
If the cells enter the Virchow-Robin spaces,
meningoencephalitis results. Once infection INTERSTITIAL FLUID
has begun in the cerebral tissues, the cells
form a cerebritis, which is localized. When The composition of the interstitial spinal fluid
the region of cellular accumulation forms a (ISF) is thought to be similar to that of the CSF
capsule, it is referred to as an abscess. Finally, because of the continuity of the two fluids across
when the cells are scattered more diffusely the ependymal and pial surfaces. Formation of
throughout the brain and accumulate around ISF occurs by active transport processes at the
blood vessels, it is called an encephalitis. The cerebral capillary, utilizing the high density of
sites of infection and masses in the meninges mitochondria that allows the capillary to act as a
are shown in Figure 2–6. secretory epithelium. Estimates of the amount
The epidural space around the spinal cord of CSF coming from ISF production, which is
is used as a site for injection of drugs for pain driven by Na+/K+ ATPase pumps on the ablu-
control. Infection can spread into the epidural minal side, range from 30% to 60%, depend-
space of the spinal cord from sites in the abdo- ing on the species studied and the method of
men and lungs. Once an infection begins in the measurement.44
epidural space it can spread to multiple verte- The extracellular matrix in brain contains
bral levels, but most dangerously it can compress complex carbohydrates and glycoproteins.

L
SKUL Epidura
l Absce
ss
ural
Subd ema
RA Em py
DU AR
AC
S HNO
SA Me
nin ID
git
is
A
PI
Meningoencephalitis

Cerebritis

Capillary
Abscess (blood-borne Encephalitis
infection)

Figure 2–6. This drawing shows sites of potential brain infections and masses. Epidural hematomas occur beneath the skull
and above the dura; these are ruptured blood vessels. Subdural empyemas are infections beneath the dura and over the surface
of the brain. Meningitis indicates an infection limited to the subarachnoid space that contains the CSF. Meningoencephalitis
indicates that the infection in the meninges has moved into the brain, often along the spaces formed as the blood vessels pen-
etrate the surface (Virchow-Robin spaces). An infection within the brain begins as a cerebritis, and when it forms a wall, it is
called an abscess. Encephalitis due to viral infections, such as herpes simplex, spreads within the brain.
28 Molecular Physiology and Metabolism of the Nervous System

Posterior greater drainage of the tracer into the ipsilat-

Spinous process eral lymphatics of the neck. When the radiola-
Epidural beled albumin was injected into CSF, allowing
abscess it to move evenly throughout the subarachnoid
Dura mater
space, the tracer appeared in both the ipsilat-
Arachnoid
eral and contralateral lymph.46 This suggests
Pia mater Epidural that the drainage follows a route within brain
space tissue that is lateralized and unlikely to involve
passage into the CSF in large amounts before
Vertebral it leaves the head.
body
Subarachnoid Brain ISF flows via bulk flow along preferen-
space Subdural space tial pathways through the brain, contributing to
Anterior (potential)
the formation of CSF.13,48,49 The evidence indi-
Figure 2–7. Transverse section of a vertebral body show- cates an ISF bulk flow rate of 0.1–0.3 μL/min/g
ing the locations of the epidural and subdural spaces. in rat brain along preferential pathways, espe-
Epidural abscesses (arrow) arise from infections in the tho- cially perivascular spaces and axon tracts. Brain
racic regions and lead to compression of the spinal cord. capillary endothelium is the main source of the
(Adapted from ref. 84)
fluid; capillaries have the necessary ion trans-
porters, channels, and water permeability to
The main extracellular molecules are the gly- generate fluid at a slow rate, which is much
cosaminoglycans: hyaluronic acid, heparin sul- less than CSF secretion across choroid plexus
fate, dermatan sulfate, and chondroitin sulfate. epithelium. Some CSF may recycle from the
Hyaluronic acid is highly hydroscopic and is subarachnoid space into arterial perivascu-
found in larger amounts in the gray matter than lar spaces on the ventral surface of the brain
in the white matter.45 and join the circulating ISF, draining back via
venous perivascular spaces and axon tracts into
CSF compartments, and out of both through
arachnoid granulations and along cranial nerves
LYMPHATIC DRAINAGE to the lymphatics of the neck. The bulk flow of
ISF has implications for nonsynaptic cell-cell
Molecules injected into the caudate nucleus communication (volume transmission); for
in several animal species drain into the ipsilat- drug delivery, distribution, and clearance; for
eral cervical lymphatics.46 In rabbits, dogs, and brain ionic homeostasis and its disturbance in
sheep, it is suggested that the substances move brain edema; for the immune function of the
through the subarachnoid space and across brain; for the clearance of beta-amyloid depos-
the cribriform plate into the nasal mucosa. its; and for the migration of cells (malignant
However, the significance of this route of ISF cells, stem cells).44
drainage in humans is unknown. Dogs nor-
mally drain CSF out of the brain into the nasal
region.24 Cats and sheep also have significant
drainage by this route. WATER DIFFUSION, BULK FLOW
Lymphatic drainage into the cervical nodes OF ISF, AND DIFFUSION TENSOR
is a potential route for antigenic fragments of IMAGING
brain tissue to reach the peripheral lymphat-
ics where antibody formation could take place, Magnetic resonance imaging has the unique
which may be of importance in autoimmune ability to show the diffusion of water. As MRI
diseases involving the brain.47 Fragments of technology advanced, with faster scans and
myelin released by traumatic injury or other better computational programs, novel uses of
pathological mechanisms could enter the MRI emerged. A particularly useful develop-
lymph nodes by this route, but the role of ment was visualization of water movement in
lymphatic drainage in autoimmune diseases three dimensions. In 1990, Michael Moseley
remains uncertain. reported that water diffusion in white mat-
Direct injection of 131I-labeled human ter was anisotropic, which means that the
albumin into the caudate nucleus resulted in effect of diffusion on proton relaxation varied,
 2 Cerebrospinal and Interstitial Fluids 29

depending on the orientation of fiber tracts There are numerous stimuli that lead to
relative to the orientation of the diffusion gra- the release of AVP into the peripheral circu-
dient applied by the imaging scanner. He also lation. Hemorrhage, water deprivation, hyper-
pointed out that this should best be described tonic sodium, and hypoxia cause an increase
by a tensor, which is a vector that can be bro- in plasma levels of vasopressin, while hemor-
ken down into three directions of movement rhage, hypoxia, and water deprivation lead to
in Cartesian space.50 Diffusion tensor imaging increased levels of the hormone in the CSF.
(DTI) is important when a tissue, such as the There is a threshold effect with graded hyp-
neural axons of white matter in the brain, has oxia, with release of hormones occurring only
an internal fibrous structure analogous to the when the oxygen level is lowered to 10% of
anisotropy of some crystals. Water will then that of inspired air.60 The stimuli that produce
diffuse more rapidly in the direction aligned release into the periphery are not necessarily
with the internal structure, and more slowly the same ones involved in release into brain tis-
as it moves perpendicular to the preferred sue, which is reflected in the levels in the CSF.
direction. More extended DTI scans derive The CSF has been proposed as the conduit for
neural tract directional information from the transport of AVP from the site of release at the
data using three-dimensional or multidimen- median eminence to other regions; however,
sional vector algorithms based on six or more the release into brain tissue with drainage into
gradient directions, sufficient to compute the CSF is also possible.
diffusion tensor. From the diffusion tensor, Atrial natriuretic peptide (ANP) is released
diffusion anisotropy measures such as frac- from cardiac atrial cells; it acts on the kidney
tional anisotropy can be computed. Moreover, and other peripheral organs to counteract the
the principal direction of the diffusion tensor effect of AVP. Choroid plexus cells have recep-
can be used to infer the white matter connec- tors for atriopeptin, and infusion of ANP into
tivity of the brain (i.e., tractography, trying to the CSF reduced the rate of CSF production
see which part of the brain is connected to by 35%.61 Atrial natriuretic peptide acts by
which other part).51 Studies in complex neu- stimulating the production of cyclic guanosine
rological and psychiatric disorders, such as monophosphate, a second messenger. In iso-
traumatic brain injury, stroke, MS, dementia, lated microvesse1s, ANP activated guanylate
autism, and schizophrenia, have shown disor- cyclase activity.
dered patterns of white matter fiber tracts and Increased intracranial pressure, such as in
have inferred that these changes in fiber tracts pseudotumor cerebri and subarachnoid hem-
impact connectivity. orrhage, increases CSF vasopressin levels.62
Hypoxia in animals increases vasopressin lev-
els in both blood and CSF.60 Two vasopressin
receptors have been identified, namely, a V1
NEUROPEPTIDES AND FLUID and a V2 receptor. In rat brain, a V1-type recep-
HOMEOSTASIS tor has been localized to the lateral septum and
dorsal hippocampus. Isolated brain capillaries
Neuropeptides, such as arginine vasopres- have a V1-type receptor, and the pial arteries
sin (AVP) and atrial natriuretic factor (ANF), have vasopressin immunoreactive fibers.63 The
influence water movement in the brain.8,52–56 V1 receptor is coupled to a phosphoinositol sec-
Vasopressin crosses the BBB very slowly, and ond messenger system, which in other organs
AVP in the CSF is produced within the central is found near the alpha-adrenergic receptor.
nervous system. When injected into the CSF of Activation of phosphoinositol results in inositol
rabbits, AVP lowered CSF pressure by increas- triphosphate phosphate (IP3) and diacylglyc-
ing the transport of water across the arach- erol (DAG) formation. These, in turn, acti-
noid villi.57 Increased intracranial pressure in vate protein kinase C and the phospholipases.
humans results in an increase in the level of the Calcium plays a key role in this process since it
hormone in the CSF, suggesting that it may be is activated by DAG and augments phospholi-
important in brain edema.58 Further evidence pase activation. Phospholipases C and A2 can
for a role in brain edema comes from study- release membrane fatty acids, particularly ara-
ing cold-injury edema, which is worsened after chidonic acid. Once arachidonic acid and other
vasopressin injection into the CSF.59 free fatty acids are formed, they have a series of
30 Molecular Physiology and Metabolism of the Nervous System

deleterious effects on cellular function. They side effects. Arginine vasopressin receptor
inhibit Na+/K+-ATPase and lead to the forma- antagonists represent a new approach to the
tion of free radicals and leukotrienes, both of treatment of hyponatremia; they block tubular
which are potent mediators of the inflammatory reabsorption of water by binding to V2 recep-
response. Edema occurs with intraventricular tors in the renal collecting ducts, resulting in
or intracerebral injection of AVP, suggesting aquaresis. Initial clinical experience with AVP
that excessive stimulation of V1 receptors on receptor antagonists for hyponatremia has
brain cells or blood vessels mediates volume shown that these agents augment free water
regulation by AVP.59,64 clearance, decrease urine osmolality, and
Receptors for vasopressin are present in neu- correct serum sodium and serum osmolality.
rons, astrocytes, endothelial and smooth mus- Controlled clinical trials now underway will
cle cells of blood vessels, and choroid plexuses. help elucidate the role of AVP receptor antag-
A number of studies have shown increased onism in the treatment of hyponatremia.67
release of vasopressin and expression of vaso-
pressin receptors in the brain following ische-
mia, trauma, or subarachnoid hemorrhage, and
antagonists of vasopressin V1 receptors reduce AQUAPORINS AND WATER
brain edema. Vasopressin is also implicated in TRANSPORT IN THE CENTRAL
brain edema and in impairment of cerebral vas- NERVOUS SYSTEM
culature in hypo-osmotic states. Vasopressin V1
receptor antagonists are being tested in exper- Early experiments demonstrating that erythro-
imental studies for treatment of cerebrovascu- cyte membranes are more permeable to water
lar pathology.65 than expected from water diffusion through a
Hyponatremia is frequently associated lipid bilayer provided the first experimental evi-
with neurological disease, neurosurgical pro- dence of the existence of water pores.68 Many
cedures, and the use of psychoactive drugs. years passed before the discovery of pore-form-
Arginine vasopressin is the principal phys- ing molecules, aquaporins (AQPs), for which
iological regulator of water and electrolyte Agre was awarded the Nobel Prize.69 The aqua-
balance at the kidney, and disruption of the porins are a family of at least 13 members of
normal AVP response to osmotic stimuli by small membrane-spanning proteins that assem-
brain diseases is a common cause of dilu- ble in cell membranes as homotetramers.70–72
tional hyponatremia in neurological disorders. Each monomer is approximately 30 kDa, and
Hyponatremia due to self-induced water six α-helical domains with cytosolically ori-
intoxication, symptomatic hospital-acquired ented amino and carboxy termini surround
hyponatremia, and hyponatremia associated the water pore.73 The AQPs transport water
with intracranial pathology demand prompt in both directions, depending on the pressure
intervention. The hyponatremia-induced shift gradients from hydrostatic or osmotic forc-
in water from the extracellular to the intracel- es.74 The principal AQP in mammalian brain is
lular compartment can lead to cerebral edema AQP4; this molecule is heavily expressed at the
and serious neurological complications, espe- borders between brain parenchyma and major
cially if the decrease in serum sodium con- fluid compartments, including astrocytic foot
centration is large or rapid. Overly rapid processes, glia limitans, ependymal cells, and
correction of the serum sodium level may subependymal astrocytes.75–77 This distribution
lead to osmotic demyelination and irrevers- pattern indicates that AQP4 controls water
ible brain injury.66 Central pontine myelinoly- flow into and out of the brain.74 Aquaporin 1 is
sis was initially described as a consequence of expressed in the apical membrane of the cho-
the damage to the white matter in the pons, roid plexus and plays an important role in CSF
but it is now recognized that more extensive formation.74,78,79 There was controversy about
damage to the white matter may occur. Fluid whether AQP9 is expressed in the brain.78,80
restriction is considered the first-line treat- However, one study using mice with targeted
ment; the pharmacological agents currently deletion of the AQP9 gene has provided con-
used in the treatment of hyponatremia are clusive evidence for expression of AQP9 in
limited by inconsistent responses and adverse neurons.81
 2 Cerebrospinal and Interstitial Fluids 31

(A) REFERENCES
1. Fishman RA. Cerebrospinal Fluid in Diseases of the
Nervous System. Philadelphia: W. B. Saunders; 1992.
2. Cohen SR, Herndon RM, McKhann GM. Radioimmu-
noassay of myelin basic protein in spinal fluid. An index
of active demyelination. N Engl J Med. 1976;295:
Lc 1455–1457.
3. Davson H. Physiology of the Cerebrospinal Fluid.
London: J. & A. Churchill; 1967.
4. van Crevel H. Rihsa cisternography in cerebral
E
tumours. Neuroradiology. 1979;18:133–138.
Endfoot 5. Ames A III, Sakanoue M, Endo S. Na, K, Ca, Mg and
Cl concentrations in choroid plexus fluid and cisternal
fluid compared with plasma ultrfiltrate. J Neurophysiol.
1964;27:672–681.
6. Masuzawa T, Sato F. The enzyme histochemistry of
the choroid plexus. Brain. 1983;106(1):P55-P99.
7. Epstein MH, Feldman AM, Brusilow SW. Cerebro-
(B) spinal fluid production: stimulation by cholera toxin.
Science. 1977;196:1012–1013.
8. Rosenberg GA, Kyner WT, Fenstermacher JD, et al.
Lc Effect of vasopressin on ependymal and capillary
permeability to tritiated water in cat. Am J Physiol.
1986;251:F485–F489.
E 9. Johanson CE, Parandoosh Z, Smith QR. Cl-hco3
* exchange in choroid plexus: analysis by the dmo method
for cell pH. Am J Physiol. 1985;249:F478–F484.
10. Saito Y, Wright EM. Bicarbonate transport across the
frog choroid plexus and its control by cyclic nucle-
otides. J Physiol. 1983;336:635–648.
11. Cserr HF. Physiology of the choroid plexus. Physiol
Figure 2–8. Drawing showing the expression pattern Rev. 1971;51:273–311.
of AQP4 around brain microvessels. The concentration 12. Heisey SR, Held D, Pappenheimer JR. Bulk flow and
of AQP4 is high in the perivascular endfeet membrane diffusion in the cerebrospinal fluid system of the goat.
domains that face the vessels and drops to low levels as Am J Physiol. 1962;203:775–781.
soon as the membrane loses contact with the basal lamina. 13. Rosenberg GA, Kyner WT, Estrada E. Bulk flow of
Endothelial cells express AQP4 at their adluminal as well brain interstitial fluid under normal and hyperosmolar
as abluminal membranes but at much lower levels than conditions. Am J Physiol. 1980;238:42–49.
the astrocytic endfeet. (B) An electron micrograph (corre- 14. Sahar A, Tsipstein E. Effects of mannitol and furose-
sponding to the box in (A) shows strong AQP4 immunogold mide on the rate of formation of cerebrospinal fluid.
labeling in the perivascular membrane of an astrocytic end- Exp Neurol. 1978;60:584–591.
foot. (The endothelial AQP4 was not evident at this level of 15. Wise BL, Chater N. The value of hypertonic man-
labeling sensitivity.) Asterisk, basal lamina; E, endothelium; nitol solution in decreasing brain mass and lowering
Endfoot, astrocyte endfoot; Lc, capillary lumen. The arrow cerebrospinal fluid pressure. J Neurosurg. 1962;19:
indicates the two membrane domains of the endfoot. Scale 1038–1043.
bar, 0.5 m. (From Ref. 83; See also the color insert.) 16. Vargas T, Ugalde C, Spuch C, et al. Abeta accumulation in
choroid plexus is associated with mitochondrial-induced
apoptosis. Neurobiol Aging. 2010;31: 1569–1581.
17. Wolburg H, Paulus W. Choroid plexus: biology and
Water moving from the blood into the brain pathology. Acta Neuropathol. 2010;119:75–88
through an intact BBB has to cross three mem- 18. Bjerke M, Andreasson U, Rolstad S, et al. Subcortical
branes: luminal and abluminal endothelial cell vascular dementia biomarker pattern in mild cognitive
membranes and the membrane of the astrocyte impairment. Dement Geriatr Cogn Disord. 2009;28:
foot processes.74,82 A high density of AQP4 is pre- 348–356.
19. Wallin AK, Blennow K, Zetterberg H, et al. Csf
sent in the vascular-facing astrocytic membranes biomarkers predict a more malignant outcome in
along with a lower density of AQP4 in endothe- Alzheimer disease. Neurology. 2010;74:1531–1537.
lial cell membranes. Because of the close appo- 20. Welch K, Friedman V. The cerebrospinal fluid values.
sition of the astrocytic foot processes and their Brain. 1960;83:454–469.
21. Alksne JF, Lovings ET. The role of the arachnoid villus
high density of AQP4, water that crosses the in the removal of red blood cells from the subarachnoid
BBB will rapidly and preferentially end up in space: an electron microscopic study in the dogs.
the perivascular astrocyte82 (Figure 2–8). J Neurosurg. 1972;36:192–200.
32 Molecular Physiology and Metabolism of the Nervous System

22. Adams RD, Fisher CM, Hakim S. Symptomatic 42. Nowak L, Bregestovski P, Ascher P, et al. Magnesium
occult hydrocephalus with “normal” cerebrospinal gates glutamate-activated channels in mouse central
fluid pressure: a treatable syndrome. N Engl J Med. neurones. Nature. 1984;307:462–465.
1965;273:117–126. 43. Rothman SM, Olney JW. Glutamate and the patho-
23. Di Chiro G, Reames PM, Matthews WB Jr. Risa- physiology of hypoxic-ischemic brain damage. Ann
ventriculography and risa-cisternography. Neurology. Neurol. 1986;19:105–111.
1964;14:185–191. 44. Abbott NJ. Evidence for bulk flow of brain intersti-
24. Di Chiro G, Stein SC, Harrington T. Spontaneous tial fluid: significance for physiology and pathology.
cerebrospinal fluid rhinorrhea in normal dogs: radio- Neurochem Int. 2004;45:545–552.
isotope studies of an alternate pathway of csf drainage. 45. Margolis RU, Aquino DA, Klinger MM, et al. Structure
J Neuropathol Exp Neurol. 1972;31:447–453. and localization of nervous tissue proteoglycans. Ann
25. Benson DF, LeMay M, Patten DH, et al. Diagnosis NY Acad Sci. 1986;481:46–54.
of normal-pressure hydrocephalus. N Engl J Med. 46. Bradbury MW, Cserr HF, Westrop RJ. Drainage of
1970;283:609–615. cerebral interstitial fluid into deep cervical lymph of
26. Wiesemann E, Berding G, Goetz F, et al. Spontaneous the rabbit. Am J Physiol. 1981;240:329–336.
intracranial hypotension: correlation of imaging findings 47. Knopf PM, Cserr HF, Nolan SC, et al. Physiology
with clinical features. Eur Neurol. 2006;56:204–210. and immunology of lymphatic drainage of intersti-
27. Hussey F, Schanzer B, Katzman R. A simple constant- tial and cerebrospinal fluid from the brain [review].
infusion manometric test for measurement of csf Neuropathol Appl Neurobiol. 1995;21:175–180.
absorption. II. Clinical studies. Neurology. 1970;20: 48. Cserr HF, Ostrach LH. Bulk flow of interstitial fluid
665–680. after intracranial injection of blue dextran 2000. Exp
28. Davson H, Pollay M. Influence of various drugs on Neurol. 1974;45:50–60.
the transport of 131-I and pah across the cerebrospi- 49. Rosenberg GA, Wolfson LI, Katzman R. Pressure-
nal fluid-blood barrier. J Physiol (Lond). 1963;167: dependent bulk flow of cerebrospinal fluid into brain.
239–246. Exp Neurol. 1978;60:267–276.
29. Davson H, Luck CP. The effect of acetazolamide on 50. Moseley ME, Cohen Y, Kucharczyk J, et al. Diffusion-
the chemical composition of the aqueous humour and weighted mr imaging of anisotropic water diffusion in cat
cerebrospinal fluid and the brain. J Physiol (Lond). central nervous system. Radiology. 1990;176:439–445.
1957;137:279–293. 51. Le Bihan D, Mangin JF, Poupon C, et al. Diffusion
30. Fishman RA. Factors influencing the exchange of tensor imaging: concepts and applications. J Magn
sodium between plasma and cerebrospinal fluid. J Clin Reson Imaging. 2001;13:534–546.
Invest. 1959;38:1698–1708. 52. Raichle ME, Grubb RL Jr. Regulation of brain water
31. Rapoport SI. Blood-Brain Barrier in Physiology and permeability by centrally-released vasopressin. Brain
Medicine. New York: Raven Press; 1976. Res. 1978;143:191–194.
32. Crone C, Christensen O. Electrical resistance of a cap- 53. Doczi T, Szerdahelyi P, Gulya K, et al. Brain water
illary endothelium. J Gen Physiol. 1981;77:349–371. accumulation after the central administration of
33. Olesen SP, Crone C. Substances that rapidly augment vasopressin. Neurosurgery. 1982;11:402–407.
ionic conductance of endothelium in cerebral venules. 54. Doczi T, Joo F, Szerdahelyi P, et al. Regulation of brain
Acta Physiol Scand. 1986;127:233–241. water and electrolyte contents: the opposite actions of
34. Bradbury MWB, Stubbs J, Hughes IE, et al. The dis- central vasopressin and atrial natriuretic factor (anf).
tribution of potassium, sodium, chloride and urea Acta Neurochir Suppl (Wien). 1988;43:186–188.
between lumbar cerebrospinal fluid and blood serum 55. Rosenberg GA, Estrada E, Kyner WT. The effect of
in human subjects. Clin Sci. 1963;25:97–105. arginine vasopressin and v1 receptor antagonist on
35. Katzman R, Graziani L, Ginsburg S. Cation exchange brain water in cat. Neurosci Lett. 1988;95:241–245.
in blood, brain and csf. Prog Brain Res. 1968;29: 56. Rosenberg GA, Scremin O, Estrada E, et al. Arginine
283–296. vasopressin v1-antagonist and atrial natriuretic pep-
36. Katzman R, Pappius HM. Brain Electrolytes and Fluid tide reduce hemorrhagic brain edema in rats. Stroke.
Metabolism. Baltimore: Williams & Wilkins; 1973. 1992;23:1767–1773; discussion 1773–1774.
37. Woodbury J, Lyons K, Carretta R, et al. Cerebrospinal 57. Noto T, Nakajima T, Saji Y, et al. Effect of vasopres-
fluid and serum levels of magnesium, zinc, and cal- sin on intracranial pressure of rabbit. Endocrinol Jpn.
cium in man. Neurology. 1968;18:700–705. 1978;25:591–596.
38. Wong RP, Bradbury MW. Permeability of the blood- 58. Sorensen PS, Gjerris F, Hammer M. Cerebrospinal
brain barrier to calcium in adrenal insufficiency. Brain fluid vasopressin and increased intracranial pressure.
Res. 1975;84:361–364. Ann Neurol. 1984;15:435–440.
39. Murphy VA, Smith QR, Rapoport SI. Homeostasis of 59. Reeder RF, Nattie EE, North WG. Effect of vasopres-
brain and cerebrospinal fluid calcium concentrations sin on cold-induced brain edema in cats. J Neurosurg.
during chronic hypo- and hypercalcemia. J Neurochem. 1986;64:941–950.
1986;47:1735–1741. 60. Wang BC, Sundet WD, Goetz KL. Vasopressin in
40. Tai CY, Smith QR, Rapoport SI. Calcium influxes plasma and cerebrospinal fluid of dogs during hypoxia
into brain and cerebrospinal fluid are linearly related or acidosis. Am J Physiol. 1984;247(1):E449-E455.
to plasma ionized calcium concentration. Brain Res. 61. Steardo L, Nathanson JA. Brain barrier tissues: end
1986;385:227–236. organs for atriopeptins. Science. 1987;235:470–473.
41. Nicholson C, Bruggencate GT, Steinberg R, et al. 62. Sorensen PS, Thomsen C, Gjerris F, et al. Increased
Calcium modulation in brain extracellular microen- brain water content in pseudotumour cerebri mea-
vironment demonstrated with ion-selective micropi- sured by magnetic resonance imaging of brain water
pette. Proc Natl Acad Sci USA. 1977;74:1287–1290. self diffusion. Neurol Res. 1989;11:160–164.
 2 Cerebrospinal and Interstitial Fluids 33

63. Itakura T, Okuno T, Ueno M, et al. Immunohisto- 74. Tait MJ, Saadoun S, Bell BA, et al. Water movements
chemical demonstration of vasopressin nerve fibers in the brain: role of aquaporins. Trends Neurosci.
in the cerebral artery. J Cereb Blood Flow Metab. 2008;31:37–43.
1988;8:606–608. 75. Rash JE, Yasumura T, Hudson CS, et al. Direct immu-
64. Doczi T, Szerdahelyi P, Gulya K, et al. Brain water nogold labeling of aquaporin-4 in square arrays of
accumulation after the central administration of astrocyte and ependymocyte plasma membranes in
vasopressin. Neurosurgery. 1982;11:402–407. rat brain and spinal cord. Proc Natl Acad Sci USA.
65. Kozniewska E, Romaniuk K. Vasopressin in vascu- 1998;95:11981–11986.
lar regulation and water homeostasis in the brain. 76. Nielsen S, Nagelhus EA, Amiry-Moghaddam M, et al.
J Physiol Pharmacol. 2008;59(suppl 8):109–116. Specialized membrane domains for water transport in
66. Sterns RH, Nigwekar SU, Hix JK. The treatment of glial cells: high-resolution immunogold cytochemistry of
hyponatremia. Semin Nephrol. 2009;29:282–299. aquaporin-4 in rat brain. J Neurosci. 1997;17: 171–180.
67. Bhardwaj A. Neurological impact of vasopres- 77. Badaut J, Lasbennes F, Magistretti PJ, et al. Aquaporins
sin dysregulation and hyponatremia. Ann Neurol. in brain: distribution, physiology, and pathophysiology.
2006;59:229–236. J Cereb Blood Flow Metab. 2002;22:367–378.
68. Sidel VW, Solomon AK. Entrance of water into human 78. Zador Z, Bloch O, Yao X, et al. Aquaporins: role in
red cells under an osmotic pressure gradient. J Gen cerebral edema and brain water balance. Prog Brain
Physiol. 1957;41:243–257. Res. 2007;161:185–194.
69. Gomes D, Agasse A, Thiebaud P, et al. Aquaporins are 79. Boassa D, Stamer WD, Yool AJ. Ion channel function
multifunctional water and solute transporters highly of aquaporin-1 natively expressed in choroid plexus.
divergent in living organisms. Biochim Biophys Acta. J Neurosci. 2006;26:7811–7819.
2009;1788:1213–1228. 80. Saadoun S, Tait MJ, Reza A, et al. Aqp4 gene deletion
70. Verkman AS. More than just water channels: unex- in mice does not alter blood-brain barrier integrity or
pected cellular roles of aquaporins. J Cell Sci. 2005;118: brain morphology. Neuroscience. 2009;161:764–772.
3225–3232. 81. Mylonakou MN, Petersen PH, Rinvik E, et al. Analysis
71. Agre P, King LS, Yasui M, et al. Aquaporin water of mice with targeted deletion of aqp9 gene provides
channels—from atomic structure to clinical medicine. conclusive evidence for expression of aqp9 in neurons.
J Physiol. 2002;542:3–16. J Neurosci Res. 2009;87:1310–1322.
72. Verkman AS. Physiological importance of aquaporins: 82. Kimelberg HK. Water homeostasis in the brain: basic
lessons from knockout mice. Curr Opin Nephrol concepts. Neuroscience. 2004;129:851–860.
Hypertens. 2000;9:517–522. 83. Amiry-Moghaddam M, Ottersen OP. The molecular
73. Erkman AS, Yang B, Song Y, et al. Role of water chan- basis of water transport in the brain. Nat Rev Neurosci.
nels in fluid transport studied by phenotype analy- 2003;4:991–1001.
sis of aquaporin knockout mice. Exp Physiol. 2000; 84. Baker AS, Ojemann RG. Spinal epidural abscess.
85(spec no: 233S–241S). NEJM. 1975;293:463–468.
Exploring the Variety of Random
Documents with Different Content
 Dès le commencement d'août, sous le prétexte des noces
prochaines, l'armée des Guises est entrée dans Paris, je veux dire les
bandes nombreuses que cette riche maison, du revenu de ses quinze
évêchés, et dans ses terres, ses fiefs, ses innombrables seigneuries,
nourrissait et gardait en armes. Quelques-uns étaient des bravi,
comme Maurevert et Attin, pensionnés pour tuer Coligny et son
frère. La grande masse étaient de pauvres gentilshommes, gueux
nobles et mendiants bien nés, que les cardinaux de Lorraine et de
Guise, les princes de la famille, Henri de Guise, Aumale, Elbeuf, etc.,
tenaient en meutes, avec leurs dogues, pour les lâcher au jour utile.
Ajoutez une grande clientèle de serviteurs volontaires et
désintéressés de la famille, de gros corps de noblesse picarde et
autre, qui venaient d'amitié accompagner MM. de Guise et les
garder. Un seul gentilhomme, Fervaques, un furieux Picard
catholique, leur amenait de son pays un renfort de vingt ou trente
épées.

Tout cela logé autour des Guises, ou chez le clergé de Paris, les
uns chez les chanoines, aux cloîtres Notre-Dame, Saint-Germain-
l'Auxerrois; les autres chez les moines, dans les grands bâtiments
des abbés-princes, chez les curés enfin, où ils se trouvaient en
rapport avec les gros bourgeois et les meneurs des confréries.

Ils se trouvaient ainsi groupés d'avance, ayant appui dans la

population.

Au contraire, les protestants, gens du Midi et de l'Ouest, logeaient

où ils trouvaient logis, étaient fort dispersés, comme perdus dans la
grande ville. Quelques-uns cependant s'obstinèrent à rester dehors,
au faubourg Saint-Germain.

Dans une situation si menaçante, Coligny oserait-il exiger de son

jeune roi la chose redoutée des catholiques, la chose épouvantable
qui marquait la victoire du protestantisme, les noces de Navarre, le
premier mariage mixte entre les deux religions, la solennelle
reconnaissance qu'un protestant est homme, et non un monstre,
l'introduction hardie du petit prince de montagne, semi-paysan
béarnais, dans l'alcôve du Louvre, dans le lit de la Marguerite, qui
affichait très-haut son mépris, son dégoût?

Rien n'arrêta l'homme de bronze. Il somma le roi de sa parole, et

la lui fit tenir.

Les simples fiançailles (17 août) produisirent déjà une explosion

dans Paris. Avec des hurlements terribles, l'armée des aboyeurs,
déchaînée dans toutes les chaires, cria que Dieu ne souffrirait pas
cet exécrable accouplement, que la colère du ciel allait tomber, qu'on
verrait des torrents de sang.

Quels étaient ces prédicateurs de la Saint-Barthélemy? La

première place entre eux est due certainement à l'évêque Sorbin, à
l'évêque Vigor, qui la prêchaient depuis douze ans. La seconde aux
jésuites, le vrai poignard de Rome; Auger, l'un d'eux, fit, à lui seul, la
Saint-Barthélemy de Bordeaux.

Mais le plus véhément de tous, un prêcheur de grande éloquence,

plein de feu, plein d'esprit, puissant acteur, brûlant parleur, fut le
cordelier Panigarola, dont nous avons les œuvres. C'était un jeune
Milanais, un mondain effréné, connu par un duel douteux et fort
sinistre d'où il sortit peu net, en ceignant le cordon de Saint-
François. Pie V, le plus violent des papes, le plus fixe au massacre, et
qui en suit l'idée dans toutes ses lettres, ayant entendu Panigarola,
crut que ce comédien terrible était l'homme même de la chose. Il fit
pour lui ce que jadis on avait fait pour Loyola. Il l'envoya, comme
étudiant, à Paris. L'étudiant ne fit qu'enseigner; sa chaire tonnante
enseigna le massacre et professa l'œuvre de sang.

Les voix bruyantes de ces enfants perdus ne donnent pas le

dessous des choses. Quels étaient ceux qui travaillaient Paris, qui
informaient Bruxelles, qui donnèrent à l'Espagne la première
nouvelle du massacre? Sans nul doute, ceux qui, dès 1560,
sollicitaient l'assistance de Philippe II (V. plus haut). Parti riche, à lui
seul énormément plus riche que le roi, la cour et le gouvernement,
et qui les emportait légers comme une paille, qui entraînait tout par
l'argent, par la force d'un patronage immense. Parti qui précipitait
Guise et l'animait par la concurrence d'Henri d'Anjou; parti qui
rassurait le duc d'Albe et lui promettait le massacre au plus tard pour
le 24 août. (Morillon, lettre du 10.)

Le roi même était menacé. Sorbin disait en chaire que, s'il faisait
les noces, il en serait de lui comme d'Ésaü, que Dieu dépouilla de
son droit d'aînesse pour le transférer à Jacob.

D'autre part, Coligny le tenait, ne lâchait pas prise. Il agissait sur

lui par l'honneur, par la confiance excessive et illimitée. Ayant rendu
les places de sûreté, il avait tiré sur le roi (si le roi était
gentilhomme) une lettre de change qu'il fallait payer ou mourir.

On disait de tous les côtés à Coligny qu'il se perdait en exigeant

cela. Il répondait froidement: «Je suis assez accompagné, si je n'ai
affaire qu'à MM. de Guise.»

Charles IX, alarmé, fit venir au Louvre le chef de la famille, Henri

de Guise, et, Coligny présent, pria et somma le jeune homme de se
réconcilier sincèrement avec cet illustre vieillard, ce grand homme en
cheveux blancs, qui toujours avait protesté qu'il n'avait pas fait tuer
son père. Henri, sans hésiter, donna la main à Coligny, et prouva ce
jour-là sa descendance maternelle, la parenté des Borgia.

On disait dans le peuple «que les noces seraient vermeilles,»

qu'elles n'auraient pas lieu, ou seraient marquées d'un combat. Elles
se firent paisiblement à Notre-Dame.

Charles IX affirma que le pape donnait la dispense, qu'elle allait

arriver, et le cardinal de Bourbon n'osa plus résister. La cérémonie se
fit sous le ciel, sur un échafaud magnifique qu'on avait dressé au
Parvis. Marguerite, qui appartenait de cœur aux Guises et à son
frère Anjou, s'obstina (dit-on) à ne pas dire: Oui, et ce fut Charles IX
qui, d'un mouvement brusque, lui fit baisser la tête et consentir en
apparence. Pendant la messe, Coligny et le roi de Navarre restèrent
à l'Évêché. Après, ils entrèrent dans l'église. De Thou, alors enfant,
vit et entendit Coligny, qui, voyant aux murailles les drapeaux de
Jarnac et de Montcontour, disait: «Nous en mettrons d'autres à la
place, plus agréables à voir,» parlant des drapeaux espagnols.

Le miracle infaisable s'était fait cependant, et l'on s'était passé du

pape. Le parti papal, espagnol, était poussé à bout. Dans son
exaltation furieuse, la coterie des futurs Ligueurs dit le jour même à
Notre-Dame, aux protestants restés hors de l'église: «Vous y
entrerez bientôt malgré vous.»

Le massacre était arrêté certainement, que la cour le voulût ou

non. Du reste, la reine mère ne refusait nul acte préalable. Le soir
des noces, on fit signer au roi une lettre aux gouverneurs, pour
arrêter tout courrier ou tout autre qui passerait les monts avant six
jours. Calipuli affirme que cette lettre fut envoyée à tous les
gouverneurs, dans toutes les directions. On dut faire croire à Charles
IX, à l'amiral peut-être, qu'il était important que don Juan d'Autriche,
l'Espagne, l'armée espagnole, qui d'Italie nous menaçait, ignorassent
le départ de nos troupes pour les Pays-Bas.

Le massacre pouvait-il se faire, sans le roi, malgré lui, par l'audace

des Guises, appuyé d'un si fort parti? Je dis hardiment oui, on
pouvait soulever Paris et tenir le roi dans son Louvre. Coligny avait
peu de monde, six cents épées, le reste des valets.

Mais les Guises n'avaient de chef que ce jeune homme de vingt

ans qui avait si peu brillé à la guerre. Le très-prudent cardinal de
Lorraine avait pris le chemin de Rome. La vraie tête des Guises était
une femme italienne, Anne d'Este, la mère d'Henri de Guise,
hésitante certainement par instinct maternel.

Parti de feu, tête de glace. Pour suivre son parti et hasarder

l'exécution, le jeune Guise voulut un ordre de l'autorité, sinon du roi,
au moins du lieutenant du roi, qui était le duc d'Anjou.
 Jamais Anjou, jamais sa mère, n'auraient pris ce courage. Ce fut
Coligny qui le leur donna, en les poussant au désespoir.

Nos envoyés dans le Levant et autres avaient écrit de longue date

que le trône de Pologne allait vaquer. Ouverture vivement saisie de
Charles IX pour éloigner Anjou. Catherine aussi, pour gagner du
temps, fit semblant de le désirer. Mais, en juillet, voici la vacance de
Pologne, voici une ambassade polonaise, voici l'insistance de Coligny
qui veut chasser Anjou ou le faire expliquer. La chose est poussée à
l'extrême par un mot fort et décisif de l'amiral: «Si Monsieur, qui n'a
pas voulu de l'Angleterre par un mariage, ne veut pas non plus de la
Pologne par élection, décidément qu'il déclare donc qu'il ne veut pas
sortir de France.»

Henri d'Anjou était mis en demeure de résister en face à Charles

IX, de dire franchement qu'il aimait mieux sa situation d'héritier
qu'aucun trône du monde; héritier d'un frère de son âge; héritier
futur, improbable, d'autant plus menaçant, pouvant être tenté de
faire du futur un présent, de se garnir les mains, d'abréger ce frère
éternel et de le mettre à Saint-Denis.

Charles IX sentait tout cela. Il pénétrait fort bien ce mignon de

Catherine, avec ses airs de femme, bracelets, boucles d'oreilles et
senteurs italiennes. Un trop juste instinct lui disait qu'en ce cadet,
docile, doux et respectueux, il avait son danger, sa perte. Et c'était
trop vrai en effet.

Dans un récit très-vraisemblable, attribué au duc d'Anjou, il dit:

«Comme j'entrai un jour dans la chambre du roi, sans me rien dire il
se promena furieusement à grands pas, me regardant souvent de
travers et mettant la main à sa dague, de façon si animeuse, que je
m'attendois à être poignardé. Je fis si dextrement, que, lui se
promenant et me tournant le dos, je me retirai vers la porte que
j'ouvris, et, avec une courte révérence, je fis ma sortie, qui ne fut
quasi aperçue que quand je fus dehors, et toutefois pas assez vite
qu'il ne me lançât encore deux ou trois fâcheuses œillades. Je crus
l'avoir échappé belle.»

Cette frayeur du fils passa augmentée à la mère. Dans le récit que

j'ai cité, le progrès de leur peur est marqué admirablement. Elle alla
jusqu'à leur faire faire la démarche qui autrement leur eût été la plus
antipathique, une alliance avec les Guises.

Ceux-ci avaient besoin extrêmement de l'assassinat. Pourquoi?

Parce que, Henri de Guise, leur héros, ayant tellement échoué à la
guerre, il leur fallait un coup pour se relever.

Le crime fut débattu entre deux femmes. Catherine fit venir la

veuve de François de Guise (alors duchesse de Nemours), la mère
de Henri de Guise. Il n'y eut, avec le duc d'Anjou, que deux témoins,
probablement Gondi (Retz) et Birague. On demanda à la veuve de
Guise si elle ne voulait pas, ayant si belle occasion, exécuter enfin
cette vengeance dont elle faisait bruit, qu'elle affichait depuis dix
ans.

Mais maintenant que la question était vue de si près, la mère de

Henri de Guise eût bien voulu que l'affaire se fît par les hommes du
roi, ou de Henri d'Anjou. Elle proposa un Gascon, épée connue et
sûre. On le fit venir et causer. Mais le duc d'Anjou n'eut garde de le
prendre. Il insista pour que cette vengeance de famille se fît par la
famille, par l'homme qu'elle nourrissait exprès, l'assassin patenté,
Maurevert. En d'autres termes, sa prudence laissait tout sur le dos
des Guises.

Ceux-ci réfléchirent qu'après tout, ayant à commandement, outre

leurs bandes personnelles, cette grosse ville, sa milice de cinquante
à soixante mille hommes contre les six cents gentilshommes de
Coligny; ayant, par le duc d'Anjou, lieutenant général du roi, les
Suisses royaux, tous catholiques, et la garde royale, ils étaient plus
de cent contre un; que, d'ailleurs, très-probablement, il n'y aurait
point de bataille; que, Coligny tué, tout se disperserait.
 Donc ils prirent tout sur eux: ils fournirent l'assassin; ils fournirent
le logis d'où l'on devait tirer; ils fournirent le cheval qui devait sauver
l'assassin. L'intendant de Guise, Chailly, alla chercher Maurevert et le
logea chez le chanoine Villemur, ex-percepteur de Guise, au cloître
Saint-Germain-l'Auxerrois. Ce fut des écuries des Guises qu'on tira
un cheval d'Espagne, qui, sellé, bridé, attendit dans l'arrière-cour,
près de la porte de derrière. Trois jours durant, derrière un treillis de
fenêtre masqué de vieux drapeaux, se tint patiemment l'assassin,
l'arquebuse chargée de balles de cuivre, appuyée et couchant en
joue.

Cependant les noces de Navarre et de Condé, qu'on maria aussi,

continuaient. Des bals, des farces plus ou moins indécentes,
remplissaient toutes les nuits, et le jour on dormait; toute affaire
ajournée, le roi perdu dans les amusement avec sa furie ordinaire;
protestants, catholiques, tout mêlé et dansant ensemble. Cependant,
dans ces fêtes folles, on distingue fort bien la malice du duc d'Anjou
et sa griffe de chat. C'est lui, sa mère, les Italiens, qui, sans nul
doute, se donnèrent le plaisir de ridiculiser le jeune paysan béarnais,
d'en faire un sot devant sa femme, de faire jouer aux dupes mêmes
une comédie du futur crime, de rire avant d'assassiner.

Ce fut, en mascarade, le Mystère des trois mondes, comme on fit

jadis à Florence au pont de l'Arno. Au paradis, rempli de nymphes,
voulaient entrer des chevaliers (Condé, Navarre); mais il était gardé
par d'autres chevaliers, par le roi et ses frères, qui rompaient la
pique avec eux et finissaient par les traîner du côté de l'enfer, où les
diables les enfermaient. Cependant les vainqueurs allèrent chercher
les nymphes et dansèrent avec elles toute une grande heure,
longueur impertinente, ennuyeuse pour les vaincus. Navarre dut
rester en enfer pendant qu'on fit danser sa femme. Le combat reprit
ensuite, et des traînées de poudre qui éclatèrent de tous côtés,
remplissant le palais de fumée, d'odeur sulfureuse, mirent en fuite
toute l'assistance.
 Damnés, vaincus et ridicules, ce fut le sort des deux maris. Le jour
suivant, on les fit Turcs, c'est-à-dire vaincus encore; les Turcs
venaient de l'être à la bataille de Lépante. Dans un tournoi en
mascarade, le roi de Navarre avec les siens, parurent vêtus en Turcs,
avec des turbans verts. Ces Turcs de carnaval furent battus par deux
femmes, deux amazones, qui n'étaient autres que le roi et son frère.

La majesté royale en jupe courte! Spectacle honteux, baroque!

Mais plus choquant encore était Anjou, impudique figure qui se
complaisait dans ce rôle et dans sa grâce infâme, couvrant de
honteuses folies les apprêts de l'assassinat (jeudi 21 août 1572).

CHAPITRE XXIII
BLESSURE DE COLIGNY.—CHARLES IX CONSENT À SA MORT
22-23 Août 1572

Coligny, quoique malade, croyait partir la semaine qui suivrait le

mariage. Il l'écrit ainsi à sa femme, dans une lettre infiniment
tendre, fort touchante, qui ferait croire qu'il sentait sa situation et
pensait bien que c'étaient les dernières paroles qu'ils dussent
échanger dans ce monde.

Dans un sombre petit hôtel, voisin du Louvre, tout près du cloître

Saint-Germain-l'Auxerrois, il recevait coup sur coup de mauvaises
nouvelles. L'édit de pacification devenait une risée; un enfant qu'on
portait au prêche pour le baptiser fut tué dans les bras de sa mère.
Les Guises grossissaient dans Paris, et Montmorency en sortait.

Ce chef futur des politiques, en abandonnant ainsi Coligny, fut une

des causes du massacre. S'il fût resté avec les siens, avec la
nombreuse noblesse attachée à sa famille, on eût regardé à deux
fois avant de tirer l'épée.
 Il crut acquitter sa conscience en avertissant Coligny de pourvoir à
sa sûreté.

Le devoir clouait celui-ci au fatal séjour de Paris; s'il eût bougé, il

perdait tout. La seule chance qu'il eût qu'on fît droit aux plaintes des
protestants, et qu'on aidât d'un secours l'invasion du prince
d'Orange, était dans sa persévérance, dans l'ascendant qu'il avait
pris sur l'esprit du jeune roi. Partir, c'était rompre avec lui, c'était
tout abandonner, recommencer la guerre civile. Dût-il mourir à Paris,
cela valait encore mieux.

Sentinelle infortunée du grand parti protestant qui ne lui donnait

nul appui, ni d'Angleterre, ni d'Allemagne, il périssait abandonné. On
le voit parfaitement par une lettre de Catherine (21 août). Au
moment où l'assassin attendait déjà Coligny, la reine mère est si
convaincue de l'indifférence d'Élisabeth à cet événement qu'elle suit
avec confiance l'affaire du mariage, et propose une entrevue entre
son fils Alençon et la reine d'Angleterre «sur mer, par un beau jour
calme, entre Douvres, Boulogne et Calais.»

On savait parfaitement qu'Élisabeth, alarmée des grands projets

de Coligny, ne vengerait nullement sa mort et prendrait fort en
patience un événement qui allait fermer aux armes françaises la
conquête des Pays-Bas.

Lui seul était la pierre d'achoppement. Il inquiétait l'Europe,

surtout ses prétendus amis.

Le vendredi 22 août, comme il rentrait lentement chez lui,

revenant du conseil et lisant une requête, il passe devant la fenêtre
fatale, il est tiré... Une balle lui emporte l'index de la main droite,
une autre traverse le bras gauche.

Maurevert avait tiré, comme Poltrot, de manière à blesser son

homme, lors même qu'il serait cuirassé. Son arme était appuyée et
pouvait tirer bien mieux. Mais la main du fanatique était restée
ferme, et la main du coquin trembla.
 Sans s'émouvoir, Coligny montre la fenêtre d'où l'on a tiré et dit:
«Avertissez le roi.»

Le roi jouait à la paume avec Guise et Téligny. Il jeta sa raquette,

parut tout bouleversé et rentra brusquement, puis fit trois choses qui
prouvaient sa bonne foi. Il ordonna l'enquête, il défendit aux
bourgeois de s'armer (Registres de la ville), et il fit dire à tous les
catholiques logés autour de l'amiral d'aller ailleurs, afin qu'on pût y
concentrer des protestants.

On a dit qu'il voulait faire massacrer ceux-ci, qu'il les réunissait

pour les envelopper. Cependant, quand on songe à la vaillance
connue de cette noblesse, à sa fermeté éprouvée, on sentira que la
réunir ainsi, c'était la fortifier, c'était rendre le meurtre infiniment
plus difficile, préparer un combat à mort.

Je ne vois pas que Coligny ait profité de l'autorisation. Il voulut lier

Charles IX, comme il avait fait en lui rendant les places de sûreté.
Pourquoi eût-il voulu plus de garantie pour lui-même qu'il n'en
gardait pour son parti? Beaucoup de protestants venaient. Mais il
n'eut, à poste fixe, que des gardes du roi. Anjou eut soin d'y mettre
un capitaine ennemi de l'amiral.

L'illustre chirurgien Ambroise Paré coupa le doigt du blessé et fit à

l'autre bras de profondes incisions. Ses amis pleuraient. Lui,
merveilleusement patient: «Ce sont là des bienfaits de Dieu.»—
Quelqu'un dit: «Oui, monsieur, remercions-le. Il a épargné la tête et
l'entendement.»

Il y avait là un saint homme, le ministre Merlin, le même, je crois,

qui sauva le coupable père de Rubens et obtint sa grâce du prince
d'Orange. Merlin dit à l'amiral: «Vous faites bien, monsieur, de ne
penser qu'à Dieu et d'oublier les assassins.»

Le calme et l'extraordinaire force d'âme de l'amiral parut à deux

choses:
 Dans l'opération très-douloureuse, et qu'Ambroise Paré ne fit
qu'en trois fois, ayant un mauvais instrument, le patient ne sourcilla
point et dit seulement à l'oreille d'un de ceux qui le soutenaient que
Merlin donnât cent écus d'or aux pauvres de l'Église de Paris.

D'autre part, malgré tant de vraisemblances, de preuves même et

d'aveux des gens de la maison fatale, comme on parlait des
coupables, il dit: «Je n'ai d'ennemis que MM. de Guise. Toutefois je
n'affirme point qu'ils aient fait le coup.»

Quelques hommes déterminés offrirent à l'amiral d'aller poignarder

les Guises à la tête de leurs bandes. Mais il le leur défendit.

Les maréchaux Damville, Villars et Cossé vinrent le voir. Ils le

trouvèrent gai et calme. Il dit à Cossé «Vous souvenez-vous de l'avis
que je vous donnais il y a quelques heures?... Il faut prendre vos
sûretés.»

Damville, avec Téligny, alla de sa part prier le roi de venir. Il vint à

deux heures et demie; mais sa mère, son frère Anjou, Gondi, son ex-
gouverneur, ne le laissèrent pas aller seul; ils le suivirent, inquiets de
ce que dirait le blessé. Ils trouvèrent la petite rue, le petit hôtel,
combles de protestants armés qui les regardaient de travers et se
parlaient à l'oreille, témoignaient peu de respect, croyant voir dans
la mère et son fils Anjou les vrais assassins.

Charles IX dit ces propres paroles: «Mon père, la blessure est pour
vous, la douleur pour moi, et pour moi l'outrage... Mais j'en ferai
telle vengeance qu'on se souviendra à jamais.» Et il en fit avec
fureur le plus terrible serment.

Coligny parla comme un homme qui se sent près de la mort. Parmi

les plaintes des Églises, il articula deux accusations.

«Pourquoi ne peut-on dire un mot dans votre conseil privé que le

duc d'Albe n'en soit averti au moment même?»
 Puis il lui dit à l'oreille (ce que de Thou a supprimé par respect
pour Catherine et pour Henri III): «Souvenez-vous des
avertissements que je vous ai donnés sur ceux qui trament contre
vous. Si Votre Majesté tient à la vie, elle doit être sur ses gardes.»

«Vous vous échauffez trop, dit la reine. Il n'y pas d'apparence de

faire parler si longtemps un malade.» Et elle emmena le roi. Le seul
Henri d'Anjou, dont la maligne nature jouissait dans le mensonge,
resta un moment de plus pour dire un mot d'amitié à celui qu'il
assassinait.

Cette hypocrisie pouvait-elle donner le change à Charles IX? On

peut en douter; il rentra profondément triste et rêveur. Sa mère
cependant l'obsédait pour tirer de lui ce que l'amiral avait dit si bas.
Il refusa quelque temps, puis éclata tout à coup: «Ce qu'il me disoit,
madame? Si vous voulez le savoir, il disoit que tout le pouvoir s'est
écoulé dans vos mains, et qu'il m'en adviendra mal.» Il sortit et
s'enferma. «Nous vîmes bien dès lors, dit lui-même Henri d'Anjou,
qu'il n'y avoit pas de temps à perdre pour dépêcher l'amiral.»

Cependant le roi de Navarre et le prince de Condé, qui avaient

demandé en vain permission de se retirer, délibéraient chez Coligny
avec quelques protestants sur ce qu'il convenait de faire. L'un d'eux
dit: «Partir à l'instant. Mais le blessé eût été difficile à transporter, et
Téligny répondait de la sincérité du roi.»

Marguerite nous apprend ici un fait essentiel. On voit que les

protestants ne se fiaient pas beaucoup à son mari, le roi de Navarre;
qu'ils le voyaient apprivoisé par les caresses catholiques, qu'un
pressentiment leur révélait dans le petit Béarnais ce leste sauteur qui
dit: «Je vais faire le saut périlleux.» Et: «Paris vaut bien messe.» Ils
lui firent signer, à lui, au prince de Condé et sans doute aux
courtisans protestants de Charles IX, une obligation écrite de venger
l'attentat fait sur Coligny.
 Le bruit s'en répandit sans doute. On sema par tout Paris la
nouvelle lamentable que ces furieux protestants avaient juré
d'égorger le pauvre jeune Henri de Guise. Malgré les défenses du
roi, les capitaines de quartier, les meneurs des confréries, avaient
fait prendre les armes. L'immensité du mouvement dépassait tout ce
qu'avaient attendu Catherine et le duc d'Anjou, mouvement donné
par le clergé et tout au profit de Guise (samedi 23 août).

Henri d'Anjou, qui s'était retiré si habilement derrière Guise pour

lui faire frapper le premier coup sur l'amiral, perdait toute son
importance, toute faveur des catholiques, tout son renom de Jarnac
et de Montcontour, s'il restait toujours derrière. Il se hasarda dans
Paris, non à cheval, mais à demi caché dans un coche, menant avec
lui son frère bâtard, Henri d'Angoulême, à qui il promettait la place
d'amiral de France s'il achevait Coligny. Sur leur route par la ville,
trouvant tout le peuple armé, ému, mais trop lent encore, ils
semèrent habilement une panique (le même moyen qui fit faire en
93 les massacres de septembre): ils dirent, ce que disaient les
protestants, que Montmorency avait été chercher un grand corps de
cavalerie pour tomber sur Paris. L'effet désiré fut atteint. On trouva
dans la peur des forces inouïes de courage; d'officieux avertisseurs
dirent qu'il fallait se hâter d'égorger les protestants.

Un petit conseil secret de la reine et des Italiens avait eu lieu à

l'écart, non au Louvre, mais aux Tuileries, par-devant le roi. Leur
avis, original et singulier, était qu'il fallait profiter du mouvement,
laisser les Guises égorger les chefs protestants; le roi surviendrait
alors, tomberait sur les Guises affaiblis, se trouverait débarrassé des
uns et des autres, de tous les grands, et vraiment roi.

Conseil italien et classique, d'après les modèles célèbres que les

petits princes italiens avaient laissés en ce genre, mais ici
inapplicable. Le roi était loin de pouvoir se débarrasser des Guises,
étant en réalité plutôt dans leurs mains.
 Il paraît du reste avoir goûté très-peu ces conseils. Un domestique
des Guises ayant été arrêté, ils vinrent hypocritement dire à Charles
IX qu'accablés par la calomnie et dans la disgrâce du roi, ils
demandaient la permission de se retirer. Le roi dit: «Vous pouvez
partir. Je saurai bien vous retrouver, s'il faut faire justice.» Ils se
mirent seulement en route et s'arrêtèrent dans les faubourgs.

C'était le samedi soir (23 août). La reine mère fit un effort décisif
près de son fils. Elle lui montra qu'il était seul, avec son petit
régiment des gardes; que les protestants allaient appeler à eux des
renforts, soulever toutes les villes; que les catholiques eux-mêmes,
s'il n'agissait pas, agiraient sans lui, nommeraient un capitaine
général. C'était lui dire précisément ce qui se fit dans la Ligue.

Elle lui dit: «Vous n'aurez pas une seule ville en France où vous
retirer.

Ce qui me prouve que le récit attribué au duc d'Anjou est vraiment

de lui ou d'un homme à lui, c'est qu'à ce moment il dissimule la
situation honteuse où se trouvèrent les coupables (lui, sa mère et
Retz), et suppose que Catherine réussit auprès du roi. Tavannes
(homme du duc d'Anjou) suit la même tradition, la moins humiliante
pour le fils et la mère.

Mais voici le grand, le véritable, le naïf historien de la Saint-

Barthélemy, Marguerite de Valois, qui nous apprend que le fils et la
mère, repoussés apparemment par Charles IX, dans leur peur et
dans leur danger, lui envoyèrent un homme qui pleurât pour eux et
le décidât au massacre qui seul pouvait les sauver. Cet homme était
Retz (Gondi), ex-gouverneur de Charles IX.

Marguerite nous apprend que, le lendemain dimanche, les

huguenots en corps devaient venir au corps accuser Guise
solennellement devant le roi. Guise, contre qui tant de preuves se
réunissaient, n'eût pu ni voulu nier un coup qui le mettait si haut
dans la faveur des catholiques; mais il eût dit qu'il n'avait rien fait
que sur l'ordre de l'autorité légitime, l'ordre de monseigneur le duc
d'Anjou, lieutenant général du royaume.

Ainsi, tout se fût dévoilé à la face du monde.

Anjou et Catherine allaient être convaincus d'avoir voulu tuer

Coligny, parce que Coligny poussait le roi à mettre hors de France
son dangereux héritier. Cela était trop évident. Avec un homme
soudain et violent comme Charles IX, Anjou eût fort bien pu périr, et
Catherine, menacée tant de fois d'être renvoyée en Italie, eût
probablement, à ce coup, repris le chemin de Florence.

Donc, le samedi 23 août à dix heures du soir, les deux coupables,

la mère et le fils, firent avouer leur cas honteux, en tâchant de
donner le change sur leurs vrais motifs. Retz dit au roi, dit
Marguerite: «Que le coup n'avoit été par M. de Guise, mais que mon
frère le roi de Pologne et la reine ma mère avoient été de la partie.»

Pourquoi: «Parce que la reine mère avoit voulu se venger de la

mort de Charny.» Bourde grossière, qu'on dut faire difficilement
avaler à Charles IX. Il connaissait trop sa mère, qui n'avait ni cœur
ni âme, ni amour ni haine, nulle vendetta, à coup sûr.

À l'appui de cette sottise qui ne prenait pas, Retz ajoutait tout

doucement que: «Si le roi continuoit en la résolution qu'il avoit de
faire justice de M. de Guise, il était en danger lui-même, puisque sa
famille était accusée.»

Mais Charles IX faisant apparemment la sourde oreille, Retz

ajoutait: «Que les huguenots étoient en tel désespoir, qu'ils s'en
prenoient non-seulement à M. de Guise, à la reine, à M. d'Anjou,
mais qu'ils croyaient aussi que le roi en fût consentant et avoient
résolu de recourir aux armes la nuit même. De sorte qu'il voyoit Sa
Majesté dans un très-grand danger, soit du côté des huguenots, soit
des catholiques par M. de Guise.»
 C'était le samedi 23 à dix heures du soir, on voulait agir à minuit.
Pour être en mesure, il fallait tirer un ordre immédiat. Ainsi, pas un
moment de délibération; il lui fallut se décider sur l'heure et sans
remise, trancher en un moment sur la résolution suprême qui allait,
à partir de cette minute, retenir à jamais, emporter sa mémoire dans
l'exécration éternelle!

La peur est contagieuse. Il est probable que la peur visible de ce

lâche Italien, sa pâleur, sa mine basse, courbée, son frissonnement,
gagnèrent Charles IX. Sur son attitude hautaine, et sur sa colère au
retour de Meaux, on l'avait cru brave. Mais il était, tous les récits
l'attestent, d'un tempérament nerveux, d'une imagination infiniment
impressionnable. La nuit, la situation imprévue, la pensée surtout
d'avoir dans le Louvre même trente ou quarante protestants des plus
redoutés, un Pardaillan, un de Piles, les premières épées de France,
tout concourut à la terreur.

Ajoutons une circonstance, la première que je vais emprunter aux

récits protestants (jusqu'ici je n'ai rien tiré que des sources
catholiques). On apprit à Charles IX que le peuple était armé!—Et
comment cela? dit-il étonné.—Votre Majesté elle-même avait
ordonné que chacun fût à son quartier.—Oui, mais j'avais défendu
que personne prît les armes.

Cet étonnement du roi ne se trouve que dans la Relation

protestante. Fait grave déjà prouvé par les Registres de la ville.
D'autant plus grave et naïf ici, qu'il échappe à l'auteur de la Relation
contre son propre système, et dément la longue préméditation qu'il
attribue à Charles IX.

Retz n'a point écrit de mémoires malheureusement. Nous ne

savons pas par quel moyen décisif il gagna sa cause.

Seulement il faut se rappeler qu'on parlait à un homme de tête

bien peu solide, poète et fort imaginatif. L'Italien dut l'emporter, non
en atténuant la chose, mais plutôt en la grandissant, en rappelant
les massacres illustres de l'histoire, comme les Vêpres siciliennes,
mystérieuse et soudaine extermination d'un grand peuple en une
nuit, saignée immense, vastes ruisseaux de sang...

Charles IX, dans sa visite à Coligny, avait demandé et vu la

manche de son habit encore trempée de sang et de rouge. Une très-
mauvaise vue pour un fou. Il s'était fort exalté, regardant toujours
cette manche: «Quoi! c'est là, répétait-il, le sang, le véritable sang
de ce fameux amiral!»

Il paraît qu'au beau milieu de l'animation il lui revint une terreur.

Mais si les protestants se vengent, s'ils se soulèvent par toute la
France, s'ils ont des armées étrangères, etc.

À cela, le doux Italien eut une réponse facile: c'est que MM. de
Guise prenaient tout sur eux, qu'ils en faisaient une affaire de
vendetta, de famille, une querelle personnelle, et nullement une
affaire générale de religion. La chose resterait ainsi comme ces
vieilles querelles de villes italiennes, comme les meurtres de La
Scala, comme les vengeances mutuelles des Montaigu, des Capulet.

Le roi pouvait dormir sur les deux oreilles. Le dimanche soir, tout
serait fini, Guise partirait de Paris. Et en même temps une lettre du
roi pour toute la France: «Les Guises et les Châtillons se sont battus;
on n'a pu les en empêcher; le roi le déplore, mais il s'en lave les
mains.»

Lâche et bas conseil d'un cruel poltron, mais qui trouva le roi à
son niveau.

Ce ne fut guère qu'entre onze heures et minuit que Charles IX,

après ces deux longues conversations, entamé par sa mère d'abord,
achevé par Retz, fasciné et magnétisé par la peur de ce misérable,
défaillit et consentit...

On était si peu sûr de ses résolutions, qu'en envoyant l'ordre à

Guise et à Marcel, ex-prévôt des marchands, la reine mère décida
que le signal sonnerait, non pas d'abord à l'horloge du Palais, assez
éloignée, mais à l'église même du Louvre, à Saint-Germain-
l'Auxerrois.

Chose bizarre, mais très-naturelle, l'ayant enfin emporté, elle

commença à avoir peur de sa propre résolution. Tavannes et le duc
d'Anjou l'avouent unanimement. «Elle se serait désistée, dit
Tavannes, si elle avait pu.»

«Nous allasmes, dit le duc d'Anjou, au portail du Louvre joignant

le jeu de paulme, en une chambre qui regarde sur la place de la
basse-cour, pour voir le commencement de l'exécution. Où nous ne
fûmes pas longtemps, ainsi que nous considérions les événements et
la conséquence d'une si grande entreprise (à laquelle, pour dire vray,
nous n'avions jusques alors guères bien pensé), nous entendismes à
l'instant tirer un coup de pistolet. Et ne sçaurois dire en quel
endroict, ni s'il offensa quelqu'un: bien sçay-je que le son seulement
nous blessa si avant en l'esprit, qu'il offensa nos sens et notre
jugement, esprit de terreur et d'appréhension des grands désordres
qui s'alloient alors commettre. Et pour y obvier, envoyasmes
soudainement et en toute diligence un gentilhomme vers M. de
Guise, pour lui dire et espressément commander qu'il se retirât en
son logis, et qu'il se gardât bien de rien entreprendre sur l'admiral,
ce seul commandement faisant cesser tout le reste. Mais tôt après,
le gentilhomme retournant nous dit que M. de Guise lui avoit
respondu que le commandement étoit venu trop tard et que
l'admiral étoit mort.»

CHAPITRE XXIV
MORT DE COLIGNY ET MASSACRE DU LOUVRE
22-26 Août 1572
 Si le coup de pistolet fit tressaillir la reine mère et son fils, on peut
bien croire que le blessé, dans sa triste insomnie, ne fut pas sans
l'entendre. Il n'avait pas grand monde autour de lui. Beaucoup
étaient au Louvre, chez le roi de Navarre, pour qui on craignait
encore plus. Mais il avait, dans deux maisons voisines de son hôtel,
deux postes de gardes du roi. Il se sentait gardé par la parole
royale, par les promesses et les traités faits avec les princes
étrangers, par tout ce qu'il y a de respecté parmi les hommes. Il
venait de recevoir une visite aimable, la plus rassurante de toutes.
La nouvelle mariée, Marguerite de Navarre, dans ces moments
sacrés où, femme et fille encore, oscillant d'un état à l'autre, la
jeune épouse est si touchante, était venue le voir, et comme
chercher la bénédiction du vieillard.

Fallait-il croire qu'elle fût un espion? Une envoyée d'Anjou? Et ce

frère, trop aimé, usa-t-il de sa petite Margot (ils appelaient ainsi leur
sœur) pour cette commission scélérate? On en croira ce qu'on
voudra.

Le blessé, sur son lit, était dans ses pensées. Quelles? La famille
peut-être qu'il ne devait jamais revoir, cette femme admirable qu'il
avait laissée enceinte et qui le rappelait en vain? Ou bien plutôt
encore cette grande famille de l'Église, si divisée, si hasardée,
orpheline de Dieu, dont la crise suprême était venue par toute la
terre?

Mais ces sombres pensées ne le reportaient-elles pas plus haut,

plus loin encore, à la grande question des déchirements du dogme, à
l'écroulement de l'arbre qui couvrit l'humanité de son ombre?
Ramenée à la foi des Suisses qu'adoptait Coligny, rentrée dans la
simple raison, l'eucharistie emporte le christianisme lui-même.

Tout cela pour lui seul. Il avait cependant près de lui dans cette
chambre deux hommes admirables. L'homme de la douleur, le grand
chirurgien du siècle, Ambroise Paré, grand de cœur autant que de
génie. L'homme de la conscience, le saint pasteur Merlin qui, je
crois, avait été envoyé par le prince d'Orange. C'est lui qui fit la
prière à l'heure dernière de Coligny.

Près de la porte de la chambre veillait aussi un bon et fidèle

Allemand qui, à l'armée, lui servait d'interprète. En bas, quelques
serviteurs et cinq ou six Suisses du roi de Navarre.

C'était un peu avant le jour, entre trois et quatre heures

(dimanche 24 août). La cavalerie de Guise arrive aux portes et
remplit la petite rue. À l'instant, les gardes du roi, de gardiens se
font assassins. Cosscins, leur capitaine, frappe au nom du roi. Le
gentilhomme qui avait les clefs ouvre; il est poignardé.

L'amiral se lève au bruit, et, couvert d'une robe de chambre, dit au

ministre: «Monsieur Merlin, faites-moi la prière.» Et lui-même
ajouta: «Je remets mon âme au Sauveur.»

«Alors celui qui a été témoin et qui a rapporté ces choses entra
dans la chambre, et, étant interrogé par Ambroise Paré que voulait
dire ce tumulte, il dit, en se tournant vers l'amiral: «Monseigneur,
c'est Dieu qui nous appelle à luy.» Il répondit: «Il y a longtemps que
je me suis disposé à mourir... Mais sauvez-vous, vous autres, s'il est
possible.» Les témoins affirment qu'il ne fut pas plus troublé de la
mort que s'il n'y eût eu bruit quelconque. Tous montèrent et
échappèrent la plupart par le toit; l'Allemand, Nicolas Muss, resta
seul avec l'amiral. (Relation.)

Cependant on avait rompu la porte de l'escalier. Cosscins marchait

en tête avec les Suisses du duc d'Anjou, sous ses couleurs (blanc,
noir et vert). Ces Suisses, voyant sur l'escalier les Suisses du roi de
Navarre, ne tiraient pas. Mais Cosscins fit tirer les gardes.

On força alors la porte de la chambre, et deux hommes entrèrent

les premiers, deux serviteurs des Guises: l'un, le Picard Attin, qui
était au duc d'Aumale, nourri chez lui longtemps pour tuer le frère
de l'amiral; l'autre était un Allemand, Behme, attaché à la personne
de Henri de Guise, qui passait pour aimer beaucoup le jeune prince
et le gouvernait entièrement. Il fut récompensé plus tard par un
riche mariage avec une bâtarde du cardinal de Lorraine qui avait été
élevée en Espagne près de la reine Élisabeth. Behme fut comblé des
dons du roi d'Espagne, mais finit misérablement.

Avec ces deux meurtriers, se trouvaient Sarlabous, le gouverneur

du Havre, ex-capitaine de Coligny, qui venait tuer son chef pour
constater sa foi de renégat.

Attin a raconté plus tard qu'ils avaient été interdits de trouver si

extraordinairement tranquille un homme qui avait la mort devant les
yeux. L'impression fut telle sur Attin que, revenu chez lui, plusieurs
jours après, il restait blême et dans une sorte de frayeur.

L'Allemand Behme, qui s'était animé à lever la porte avec un épieu

(et qui, sans doute, avait pris du cœur dans le vin), fut plus résolu
que les autres. Il avança et osa dire un mot; il demanda ce qu'il
savait très-bien: «N'es-tu pas l'amiral?»

Coligny lui dit posément: «Jeune homme, tu viens contre un

blessé et un vieillard... Du reste, tu n'abrégeras rien.» Faisant
entendre que, malade, frappé de la nature, il était mort déjà, hors
de la main des hommes.

Behme, avec un juron horrible, en reniant Dieu, lui poussa dans le

ventre cette bûche pointue, ce gros épieu qu'il avait dans la main.
On dit que Coligny, assommé de la sorte par cette lourde bête,
n'ayant pas même un coup d'épée, sentit son cœur de gentilhomme,
et, tombant, lui lança ce mot: «Si c'était un homme, du moins!...
C'est un goujat!...»

Alors Behme frappa, refrappa sur la tête. Et les autres, enhardis,

vinrent lui donner chacun son coup.

Guise était en bas à cheval dans la cour avec le bâtard

d'Angoulême. Il cria: «Behme, as-tu fini?—C'est fait!—Mais M.
d'Angoulême n'en veut rien croire, s'il ne le voit.»
 Behme alors, avec Sarlabous, prirent le corps par-dessous pour le
jeter par la fenêtre. Était-il, n'était-il pas mort? On ne le sait. Il se
trouva par le trouble des meurtriers, ou par je ne sais quel réveil de
vie et de résistance, que le corps s'accrocha un moment à la fenêtre;
cependant il tomba.

Ces assommeurs savaient si mal leur métier, que, frappant à tort,

à travers, ils avaient justement gâté ce qu'eût le mieux gardé tout
sage bourreau, ce qu'on expose, le visage et la tête. Les deux
grands seigneurs, descendus de leurs chevaux, avaient beau
regarder. Cependant le bâtard «lui torcha la face,» et, écartant le
sang, dit: «Ma foi, c'est bien lui.» Et il lui donna un coup de pied.
Certains disent que Guise en fit autant et lui donna du pied dans le
visage.

Il y avait là aussi un Italien de Sienne, Petrucci, qui appartenait à

Gonzague, duc de Nevers. Il coupa proprement la tête, et la porta
au roi et à la reine, au duc d'Anjou. On l'embauma avec soin pour
l'envoyer à Rome qui, depuis si longtemps et si instamment, l'avait
demandée.

Au moment où l'assassinat fut su au Louvre, l'affaire étant lancée

et toute hésitation désormais impossible, la cloche du signal sonna à
la paroisse du Louvre, Saint-Germain-l'Auxerrois. Ce ne fut que
longtemps après, lorsqu'il était grand jour, qu'on sonna la cloche du
Palais au coin du quai de l'Horloge, pour convier la ville au massacre.

Mais la ville était déjà avertie d'une autre manière. Coligny tué, la
tête coupée, et «ce morceau de roi» ayant été porté au Louvre, on
avait généreusement donné à la canaille les reliefs du festin.

Des enfants et des misérables, qui ne sont ni enfants, ni hommes,

sans barbe, sans âge et qu'on croirait sans sexe, femmes-hommes et
hommes-femmes, les fils naturels du ruisseau, fondirent, à travers
les soldats, dans la cour de l'amiral, et trouvant là ce corps, furent
ravis de s'en emparer. Si la tête manquait, il y avait encore autre
chose, assez pour le régal; les couteaux travaillèrent, on coupa les
mains pâles qui avaient tenu si longtemps l'épée de la France, la
sainte épée de Dieu; on coupa les parties naturelles, et on les porta
dans Paris.

Au tronc, les enfants attachèrent une corde, et le tirèrent par les

ruisseaux rougis jusqu'au bord de la Seine, et il y resta quelque
temps. Mais d'autres amateurs survinrent, qui s'en emparèrent à leur
tour, le suspendirent à Montfaucon. On l'y mit de façon outrageante
et bizarre, le dos sur une poutre, le cou, les pieds, chacun de leur
côté, flottant, ballant, le ventre en l'air.

D'autres, qui arrivaient tard, n'y surent plus que faire, sinon
d'allumer du feu dessous, pour le noircir du moins, le griller comme
un porc. Quelques-uns s'en tenaient les côtes.

Dans cette nuit fatale, du samedi 23 au dimanche 24, les heures

se marquent ainsi. La reine parle au roi le soir (sept ou huit heures?)
Retz vient lui faire l'aveu de sa mère et de son frère (dix heures?)
Ordre donné à Guise (onze heures?) par la reine et le duc d'Anjou.
La ville avertie d'armer à minuit. Long intervalle de quatre heures,
les Guises attendant que la ville soit armée, avant d'attaquer Coligny.
À l'aube, un peu avant quatre heures, signal du coup de pistolet;
Coligny tué.

Marguerite dit qu'au petit jour son mari se leva, sortit, qu'elle
dormit une heure, puis fut éveillée par le massacre du Louvre qui
dut commencer entre cinq et six.

Pourquoi ce dangereux retard après la mort de Coligny qui, su au

Louvre, pouvait faire mettre en défense les protestants du roi de
Navarre? Le duc d'Anjou l'explique peut-être en disant qu'il y eut un
moment d'hésitation, que sa mère et lui eurent frayeur et eussent
voulu tout arrêter, mais que Guise dit qu'il était trop tard.

Qu'allait-on faire de ces gentilshommes qui étaient dans le Louvre,

sous le toit du roi? Grande et cruelle question.
 Si la reine mère, si Retz avaient eu le soir tant de peine à décider
Charles IX sur la question générale, il est peu probable qu'ils
l'eussent encore compliquée de cette difficulté terrible.

Ce fut, je crois, le matin, et, Coligny tué, ce fut vers cinq heures
qu'on apporta à Charles IX ce breuvage amer et qu'on le lui fit
avaler.

C'était lui-même qui, le jour de la blessure de l'amiral, avait

engagé Navarre et Condé à faire entrer leurs gentilshommes pour se
garder des entreprises de Guise, qu'il appelait «un mauvais garçon.»
Tous s'étaient offerts, empressés, sur une telle assurance; ils étaient
trente ou quarante, outre les gouverneurs, précepteurs, valets de
chambre et domestiques des deux jeunes princes. Depuis trois jours,
Charles IX vivait avec eux, les avait aux tables royales, mêlés avec
sa maison. Exécrable fatalité. Il fallait que ce couteau qui leur
coupait le pain du roi, on le leur mît dans le cœur; que, de
commensaux et convives qu'ils avaient été le soir, les serviteurs,
officiers ou capitaines des gardes se trouvassent au matin
bourreaux? La parole du roi de France, révérée chez les infidèles et
jusqu'au bout de la terre! la parole de gentilhomme, de l'hôte féodal,
la sécurité complète avec laquelle on quittait ou on déchargeait ses
armes en passant le pont-levis! Toutes ces vieilles religions de la
France brisées et détruites, et l'honneur même assassiné!... Pour en
venir là, il fallut une grande peur, une crainte extrême de ces
hommes et l'attente d'un combat sanglant.

Dans ce Louvre si bien fermé, au fond même du filet de mort où

personne n'aurait vu, nous trouvons pourtant un témoin, la jeune
reine de Navarre:

«Le soir, étant au coucher de la reine ma mère, assise sur un

coffre auprès ma sœur de Lorraine que je voyois fort triste, la reine
m'aperçut et me dit que je m'en allasse coucher. Comme je faisois la
révérence, ma sœur, se prenant à pleurer, me dit: «Mon Dieu, ma
sœur, n'y allez pas!» Ce qui m'effraya extrêmement. La reine se
courrouça fort et lui défendit de me rien dire. Ma sœur lui dit qu'il
n'y avoit point d'apparence de m'envoyer sacrifier comme cela, et
que, sans doute, s'ils découvroient quelque chose, ils se vengeroient
sur moi. La reine mère me commanda encore rudement que je m'en
allasse coucher. Ma sœur, fondant en larmes, me dit bonsoir sans
m'oser dire autre chose. Et moi je m'en allai toute transie et
éperdue.

«Je trouvai le lit du roi, mon mari, entouré de trente ou quarante

huguenots que je ne connaissois point encore, et qui parlèrent toute
la nuit de l'accident de l'amiral. La nuit se passa sans fermer l'œil. Au
point du jour, le roi, mon mari, dit qu'il vouloit aller jouer à la
paume, attendant que le roi Charles fût éveillé, se résolvant de lui
demander justice. Il sort de ma chambre et tous ses gentilshommes
aussi.

«Moi, voyant qu'il étoit jour, estimant le danger passé, vaincu du

sommeil, je dis à ma nourrice qu'elle fermât la porte pour pouvoir
dormir. Une heure après, comme j'étois le plus endormie, voici un
homme frappant des pieds et des mains à la porte, et criant:
«Navarre! Navarre!» Ma nourrice ouvre, pensant que ce fût mon
mari. C'étoit un gentilhomme nommé M. de Téjan, qui avoit un coup
d'épée dans le coude et un coup de hallebarde dans le bras, et étoit
encore poursuivi de quatre archers qui entrèrent tous après lui. Il se
jeta dessus mon lit. Moi, sentant ces hommes qui me tenoient, je me
jette à la ruelle, et lui après moi, me tenant toujours à travers du
corps. Je ne connoissois point cet homme, et ne savois s'il venoit là
pour m'offenser, ou si les archers en vouloient à lui ou à moi. Nous
criions tous deux et étions aussi effrayés l'un que l'autre. Enfin Dieu
voulut que M. de Nançay, capitaine des gardes, y vînt, qui, me
trouvant en cet état, encore qu'il y eût de la compassion, ne se put
tenir de rire et se courrouça fort aux archers de cette indiscrétion,
les fit sortir et me donna la vie de ce pauvre homme qui me tenoit,
lequel je fis coucher et panser dans mon cabinet jusqu'à ce qu'il fût
guéri.
 «Je changeai de chemise, parce qu'il m'avoit toute couverte de
sang. M. de Nançay me conta ce qui se passoit, et m'assura que
mon mari étoit dans la chambre du roi et qu'il n'auroit nul mal. Et,
me faisant jeter un manteau de nuit sur moi, il m'emmena chez ma
sœur, où j'arrivai plus morte que vive. Entrant dans l'antichambre,
un gentilhomme, se sauvant des archers qui le poursuivoient, fut
percé à trois pas de moi. Je tombai de l'autre côté presque évanouie
entre les bras de M. de Nançay, et pensai que ce coup nous eût
percés tous deux.»

Rien ne manque à ce récit, ni la dureté incroyable de la mère, qui

aventure ainsi sa fille et la remet au hasard, à la générosité
improbable de ceux qu'on va assassiner; ni, d'autre part, la
confiance, l'imprévoyante légèreté des gentilshommes protestants,
qui s'en vont jouer à la paume dans ces sombres circonstances, se
divisent, comme pour rendre l'exécution plus facile. Car les uns
allèrent jouer, les autres restèrent en haut; le capitaine des gardes
désarma ceux-ci un à un. Pour les joueurs, on leur ôta le roi de
Navarre, que Charles fit appeler, avec le prince de Condé. La mort de
ces deux princes avait été mise en discussion, et ils n'avaient été
sauvés que par le duc de Nevers, et sans doute aussi par l'idée qu'en
les tuant on eût rendu trop forts les Guises. On fit remarquer à
Charles IX qu'en réalité ces jeunes princes n'avaient guère de
religion que les femmes et l'amusement; non plus que trois ou
quatre autres protestants de cour qu'on sauva et qui se donnèrent
au roi. Navarre et Condé mandés, Charles IX leur aurait dit, selon
quelques-uns: «La messe! ou la mort!» Parole non probable dans la
bouche du royal acteur, qui décidément avait pris son rôle, et le joua
à faire croire qu'il l'avait toujours médité.

Mais les autres, qui n'étaient pas princes, que devenaient-ils? Les
archers, comme on a vu, les piquaient de chambre en chambre pour
qu'ils se précipitassent par les escaliers ou par les fenêtres dans la
cour, où les massacreurs, en rang, les piques serrées, les recevaient,
les achevaient.
 Le premier qui fut tué dans la cour fut un gentilhomme qui, voyant
toutes ces troupes, s'avisa de demander pourquoi elles étaient là
rangées si matin. On avait dit au dehors qu'on les réunissait de nuit
pour une fête, un combat simulé. Celui à qui il parlait (c'était un
Gascon) pour réponse lui passa l'épée au travers du corps.

Mais la boucherie générale se fit par les Suisses. On voit alors

combien ces Allemands étaient utiles; ne sachant pas le français,
étant catholiques, des petits cantons qui ont l'exécration du
protestantisme, ils frappaient comme des ours ou des assommeurs
de bœufs. Ivres d'ailleurs probablement, ils tuaient sans regarder,
des gens désarmés, n'importe.

Il paraît cependant qu'on doutait de l'obéissance. Car on décida le

roi à se montrer à une fenêtre de la cour. Les amis des Guises sans
doute, Anjou et sa mère, voulurent qu'il fût bien constaté qu'il était
de la tuerie, qu'il la voulait et l'ordonnait.

Le plus vaillant de ces vaillants, Pardaillan, que la plupart

n'auraient pas regardé en face, amené là, sans épée, à l'abattoir, fut
saigné comme un mouton. Le propre gouverneur du roi de Navarre,
Beauvais, sans la moindre considération de son élève, fut égorgé.
Ces malheureux, de la cour, adressaient à cette fenêtre les appels les
plus pathétiques, et ne trouvaient dans le roi, dans leur hôte, dans
ce magistrat de la justice commune, que l'œil sauvage, égaré,
furieux, d'un misérable fou.

Il y avait dans cette foule un homme que Charles IX devait entre

tous épargner, c'était lui qui l'avait arrêté trois mois au siége de
Saint-Jean-d'Angély, le capitaine de Piles; c'était comme un
adversaire, un ennemi personnel. À ce titre, il était sacré. De Piles le
sentait, et, dans la cour, devant ce monceau de morts sur lequel il
devait tomber, il lança au balcon du roi un cri foudroyant, le
sommant de sa parole, à faire trembler la cour du Louvre.
Welcome to our website – the perfect destination for book lovers and
knowledge seekers. We believe that every book holds a new world,
offering opportunities for learning, discovery, and personal growth.
That’s why we are dedicated to bringing you a diverse collection of
books, ranging from classic literature and specialized publications to
self-development guides and children's books.

More than just a book-buying platform, we strive to be a bridge

connecting you with timeless cultural and intellectual values. With an
elegant, user-friendly interface and a smart search system, you can
quickly find the books that best suit your interests. Additionally,
our special promotions and home delivery services help you save time
and fully enjoy the joy of reading.

Join us on a journey of knowledge exploration, passion nurturing, and

personal growth every day!

ebookbell.com