1

Investigation of the hyponatremic brain edema, and

the involvement of epithelial sodium channels

in its evolution:

Steier Roy, M.D.

Ph.D Thesis

Department of Neurosurgery

University of Pécs, Medical School

Pécs, Hungary

2012

Supervisors: Prof. Dóczi Tamás, M.D., Ph.D., D.Sc.

Schwarcz Attila, M.D., Ph.D.

2

Table of content:

Acknowledgements 4

Abbreviations 5

I. Introduction: 7

1. Brain edema 7

1.1. Brain edema – General 7

1.2. Diagnosis of brain edema 7

1.3. Cellular brain edema 8

1.4. The Aquaporin water channel family 9

1.5. Treatment of brain edema 9

2. The brain`s response to hyponatremic challenge 10

2.1. Regulatory volume decrease (RVD) 10

2.2. Epithelial sodium channels and Benzamil 10

3. MRI 11

3.1. Magnetic resonance imaging – General 11

3.2. Small animal experiments in a clinical MR machine 11

3.3. Diffusion weighted imaging (DWI) 12

3.4. Magnetic resonance spectroscopy (MRS) - water signal 14

II. Aims 15

III. Benzamil prevents water accumulation in hyponatremic rats 16

1. Materials and methods 16

2. Statistical analysis 17

3. Results 17

4. Discussion 18

IV. Quantitative proton MRI and MRS of the rat brain wi th a 3T

clinical MR scanner 20

1. Materials and methods 20

1.1. Experimental protocol 20

1.2. MRI protocol 20

1.3. Data processing 21

3

1.4. Statistical analysis 22

2. Results 22

3. Discussion 24

V. The influence of Benzamil hydrochloride on the evolution of

hyponatremic brain edema as assessed by in vivo MRI study in rats 26

1. Materials and methods 26

1.1 Experimental protocol 26

1.2 MRI protocol 29

1.3 Data processing 30

1.4 Statistical analysis 30

2. Results 30

3. Discussion 33

VI. A biexponential DWI study in rat brain intracellula r edema 36

1. Materials and methods 36

1.1. Experimental protocol 36

1.2. MRI protocol 36

1.3. Data processing 37

1.4. Statistical analysis 38

2. Results 38

3. Discussion

VII. Limitations of the study 48

1. Limitations of the Benzamil studies 48

2. Limitations of the MRI studies 48

VIII. Future plans 49

IX. Summary of the thesis 50

X. References 51

XI. Bibliography 65

1. Articles grounding the thesis 65

2. Presentations and posters grounding the thesis 65

4

ACKNOWLEDGEMENTS:

I wish to express my deepest gratitude to the persons and institutions that guided me throughout

this research.

First and foremost I would like to express my gratitude to my supervisors, Prof. Dóczi Tamás

and Dr. Schwarcz Attila. I wish to thank Prof. Dóczi Tamás for providing me with the

opportunity to work and learn under him; for his continuous support, efforts, and guidance.

I also wish to thank Dr. Schwarcz Attila for his teachings and tools that will continue to serve me

for many years to come.

Special thanks is indebted to Prof. Sulyok Endre for his continuous support, fertile consultations

and brilliant counsel.

I also wish to thank the scientists at the science laboratory of the Department of Neurosurgery

Pécs, Hungary. Thanks to Prof. Gallyas Ferenc, Dr. Pál József, Dr. Bukovics Peter, and the staff.

I wish to convey my gratitude to the scientists and staff of the Pécs Diagnostic Center, Pécs,

Hungary. Especially I wish to thank my colleagues: Dr. Aradi Mihály, Orsi Gergı, and Perlaki

Gábor for a fruitful common group effort that led us to achieve our goals.

I wish to acknowledge the financial support from the Marie Curie Research training networks –

Aquaglyceroporins (MRTN-CT-2006-03995).

5

Abbreviations:

ADC: Apparent diffusion coefficient.

AVP: Arginine vasopressin.

BBB: Blood brain barrier.

CCD: Cortical collecting duct.

Cho: Choline.

Cr: Creatine.

CSF: Cerebrospinal fluid.

CV: Coefficient of variation.

DWI: Diffusion weighted imaging.

ENaC: Epithelial sodium channel.

EM: Electron microscopy.

EPI: Echo planar imaging.

HN: Hypernatremia.

HNB: Hyponatremia + ICV Benzamil.

HNS: Hyponatremia + ICV Saline.

ICV: Intracerebroventricular.

IP: Intraperitoneal.

MRI: Magnetic resonance imaging.

MRS: Magnetic resonance spectroscopy.

mT/m: Militesla/meter.

NAA: N-acetyl-aspartate.

NN: Normonatremia.

NNB: Normonatremia + ICV Benzamil.

PRESS: Point resolved spectroscopy sequence.

RF: Radio frequency

ROI: Region of interest.

RVD: Regulatory volume decrease.

SNR: Signal to noise ratio.

6

SPAIR: Spectral attenuated inversion recovery.

SD: Standard deviation.

T: Tesla.

7

I. INTRODUCTION:

1. Brain edema:

1.1. Brain edema - General:

Brain water homeostasis is imperative for the normal functioning central nervous system (CNS).

Brain edema involves abnormal water accumulation in the brain1,2. According to the “Monro-

Kellie principle, the brain, cerebrospinal fluid (CSF) and the brain’s blood compartment are

encased within the bony cranium of the skull and relatively rigid dura. An increase in any of

these compartments comes at the “expense” of the others. In the event of brain swelling, due to

brain edema, intracranial pressure (ICP) increases. Initially, the brain tries to counter-act this by

shifting of CSF and decreasing of the blood compartment. In the event that the insult causing the

brain edema is so great, these compensatory mechanisms are overwhelmed, eventually fail and

the ICP continues to rise. This can lead to cerebral herniations, which may result in death. Brain

edema may be caused by a variety of pathologies. These include stroke, brain tumors, traumatic

brain injury, systemic hyponatremia, hydrocephalus, infections, etc.

Brain edema was originally classified into cellular (cytotoxic) and vasogenic subtypes1,3, though

hydrocephalic and osmotic types of brain edema are also known1,4.

Cellular brain edema is viewed in the event of an early stroke and hyponatremia. The

intracellular (IC) compartment swells as water accumulates within the cells2. On the other hand,

in vasogenic brain edema, usually caused by brain tumors or brain abscesses, the extracellular

(EC) compartment swells. This occurs due to increased movement of plasma proteins and water

in to the EC space, permitted by a malfunctioning blood brain barrier (BBB)5. In the clinical

practice however, more than one type or subtypes of brain edema can be present at a given time

in a given pathology3,6.

1.2. Diagnosis of brain edema:

The signs and symptoms of brain edema may vary according to the etiology causing the brain

swelling. The manifestation of symptoms and physical findings direct the diagnostic measures.

These may include: Headache, nausea and vomiting, dizziness, visual changes or loss, memory

loss, inability to walk, stupor, seizures, loss of consciousness, irregular breathing (Chein-Stokes

or hyperventilation), nuchal rigidity and papilla edema.

8

The diagnostic measures taken in order to reach a definitive diagnosis usually include:

Physical examination: head and neck, neurologic examination, and ophthalmologic examination.

Laboratory tests and imaging: blood and spinal tap (if possible) tests are done in order to

determine whether the cause of the edema is infectious or chemical. Computed tomography (CT)

may be utilized to visualize the edematous area. Due to the higher water content of the edema,

the edematous area will appear hypodens. Flattened gyri, filled sulci, midline dislocations and

herniations are but a few of the outcomes of brain edema that can be imaged. MRI provides

better imaging resolution than CT as well as offering us the benefit of measuring other

parameters including: T1 water maps, diffusion weighted imaging that can help differentiate

between the different types of brain edema, and spectroscopy – that can measure the levels of

different metabolites.

1.3. Cellular brain edema:

In cellular brain edema, along with a reduction in the EC space, glial swelling is the chief

finding2, though neurons and dendrites also swell7,8. On top of the fact that astrocytes have the

ability to swell up to five times their original volume, and in humans, they outnumber neurons

20:1, astrocytes and their end-foot processes are predisposed for swelling due to: a/ The brain

specific aquaporin (AQP)-4 water channel is mainly to be found in the astrocytic end-foot

process9,10. b/ The osmotic imbalance, facilitated by K+ and glutamate clearance in astrocytes,

can induce IC water accumulation2.

One of the classical examples for cellular brain edema is the early stages of focal or global

cerebral ischaemia11,12,13. A disruption of the blood flow causes an energy deficit in the supply

area, which in turn causes the adenosine tri-phosphate (ATP) dependent Na+-channels to

malfunction. As IC Na+ concentration rises, according to the concentration gradient, water

diffuses from the EC space into the IC space.

Another type of cellular edema is the one caused by hyponatremia. Hyponatremia is one of the

most common electrolyte disorders in hospitalized patients14. During acute hyponatremia, in

response to the hyponatremic stress, more and more water shifts into the intracellular space (IC)

15-19. This causes cellular cerebral edema, which manifests the most important symptoms of this

pathology. The ambiguous nature of acute hyponatremia and its inappropriate correction can

result in life threatening conditions20.

9

1.4. The Aquaporin water channel family:

The AQPs are a family of integral membrane water channel proteins, originally identified in red

blood cells21. Most members of this family are involved in water transport. In addition to this,

AQP-3, 7 and 9 also have the ability to transport glycerol and other small polar molecules. The

main AQP found in the brain is the AQP-4. As mentioned above it is mainly located in the

astrocytic end-foot process. AQP-4 is hypothesized to play a major role in brain edema and brain

water homeostasis. AQP-4 null mice respond to vasogenic edema with a greater brain swelling

than wild-type mice22, implying that AQP-4 is involved with water elimination in this kind of

edema. In cytotoxic edema AQP-4 null mice demonstrate a lowered permeability of BBB to

water. Such mice had a lower mortality rate compared to their wild type counter-parts when

exposed to severe hyponatremia, meningitis and cerebral ischaemia23,24.

1.5. Treatment of brain edema:

The current treatment for brain edema includes: sedation, hyperventilation, intravenous

administration of hypertonic saline, mannitol, and corticosteroids. Surgical resection may be

done in order to treat peritumoral edema. Decompressive craniotomy can be used in extreme

cases. Many of these therapeutic approaches however, have not been tested in large enough

studies, putting their efficacy in question 25,26.

Due to this lack of effective therapy for brain edema further research is underway to discover

new and better ways to treat these patients. One hope for such future drugs is the AQP-blockers.

One should bear in mind though that modulation of AQP-4 as a means for treating brain edema

is highly challenging considering that AQP-4 facilitates bidirectional water transport27 and it is

involved in the control of perisynaptic K+ homeostasis28. Furthermore, AQP4 is expressed in

tissues other than the brain29; therefore, only brain-selective inhibitors could be applied. Efforts

to modulate the molecular structure of AQP4 or to interfere with its translocation to the

membrane remained unrewarding so far, as no such modalities are available to date30.

10

2. The brain’s response to hyponatremic challenge

2.1. Regulatory volume decrease (RVD):

The initial response of the brain cells to hypo-osmotic stress is swelling which is directed by

water diffusion across cell membranes according to the concentration gradient. This in turn, is

followed by a complex process aiming to restore the cells to their original volume regardless of

the ongoing hypo-osmotic conditions. This process is known as regulatory volume decrease

(RVD). The chain of events making up the RVD takes place in three stages. Initially a sensor

detects a shift in the cell volume. This information then initiates signaling cascades which result

in activation of several membrane channel proteins including swelling-activated Cl- channels31, stretch-

activated, voltage-gated K+ channels, Ca++ -dependent K+ channels and neutral K+/Cl- co-transporters32.

Activation of ion channels induces cellular efflux of K+ and Cl- ions and organic osmolytes. At the

conclusion of this reaction, it is the memory of the original cell volume that brings RVD to its

end33.

2.2. Epithelial sodium channels and Benzamil:

Epithelial Na channels (ENaC) are members of the ENaC/degenerin family of ion channel

proteins34. This family includes channels identified in Caenorhabditis elegans involved in

mechanosensation35,36, mechanosensitive cation channels in cochlear hair cells as well as skin37-

40 and acid sensing ion channels (ASIC)41,42. ENaCs are present in a variety of body tissues.

These include the brain, kidney, urinary bladder, lungs, salivary and sweat gland ducts as well as

in the distal colon43,44. The ENaC is made up out of three subunits that share roughly 30 – 35%

sequence identity: these are α, β and γ-ENaC45. These subunits are present in the brain at the

choroid plexus as well as ependyma of the antero-ventral third ventricle. They are also found in

the subfornical organ, the supraoptic nucleus, and in the paraventricular nucleus43.

Transepithelial sodium transport is a two-step process: initially Na enters the cell via amiloride-

sensitive ENaCs located on the apical membrane of the cell. Then, using the Na-K-ATPases

found at the baso-lateral membrane, the cell extrudes the Na46,47.

Epithelial Na transport may be studied in A6 cells, derived from the kidney of Xenopus

laeves48,49. Niisato et al demonstrated that exposure of A6 cell monolayer to hypotonic stress

increases cellular sodium uptake via translocation of preformed ENaC protein from the cytosolic

store into the membrane and/or via generation of new channel proteins. The hypotonicity-

11

stimulated transport could be completely inhibited by Benzamil, a specific Na channel

blocker50,51. Based on these findings we assume that ENaC and sodium flux into the cells might

be involved in the initiation and progression of brain water accumulation in severe hyponatremia.

3. MRI:

3.1. Magnetic resonance imaging - General:

Magnetic resonance imaging (MRI) is a technique that externally measures the nuclear magnetic

resonance (NMR) signals originating from the subject. It allows internal structural anatomical as

well as chemical information to be examined. MRI has several advantages on other tomographic

equipment: on the one hand, MRI has the ability to produce two and three dimensional images

from a variety of orientations, as well as spatial-spectral four dimensional figures. On the other

hand, unlike other techniques that use ionizing radiation which can be potentially dangerous,

MRI uses electromagnetic waves which are harmless according to our current knowledge. MR

signals visualized by the MRI originate from the subject itself thereby eliminating the need to

expose the subject to harmful radioactive isotopes.

The superiority of MRI is to be attributed to the wide variety of possible contrast manipulations

and its high flexibility in data acquisition. For instance, by manipulating certain parameters like

Repetition time (TR), Echo time (TE) and flip angle (α) the final MR image of a given structure

may alter drastically52.

3.2. Small animal experiments in a clinical MR machine:

In the case of small animal experiments, using the MR machine is superior to classical in-vitro

techniques since the evolution of a given pathology can be studied in real time and in-vivo,

thereby giving us the possibility to recreate and analyze the background behind signal changes in

similar human pathologies53. Large bore Human MRI scanners have previously been used for

small animal experiments with fair quality54-68. Furthermore, it was reported that in a human 3

tesla (T) MR machine such spatial resolution can be achieved, that does not fall from results

obtained with dedicated small animal MRI scanners60,63,69.

In order to use small animals as test subjects in the MR machine, two main obstacles must be

overcome: One must have a reliable and reproducible anesthesia technique, and a means to

absolutely control the temperature of the animal for the duration of the MR examination without

12

taking the animal out of the MR machine, or moving it at all from its original placement inside

the machine.

A variety of anesthesia techniques are available for small animal experiments. These include the

use of intraperitoneal injections of Ketamin and barbiturates 70,71, and inhalatory anesthetics like

halothane and isoflurane72,73. The use of a veterinarian small animal ventilation device and

isofuran as anesthetic gas comply with the requirements of long term anesthesia (in our case no

less than 2.5 hours) without fluctuations of anesthesia depth or the need for repetitive

administration and handling of the animal.

It was demonstrated that a change of 1° Celsius (C) in the animal’s body temperature can lead to

significant changes in measured MR parameters. Le Bihan et al reported that a change of 1° C

can induce a 2.4% change in the measured apparent diffusion coefficient (ADC)74. Means of

controlling the animal’s body temperature include the use of a circulating warm water pad, heat

lamps70, and warm heating pads75.

3.3. Diffusion weighted imaging (DWI):

MRI may be used to investigate tissue microstructures. It can be utilized to investigate

microscopic water motion in the brain76. Abnormalities of ADC have been recorded in several

brain pathologies, including systemic hyponatremia70, seizures77, stroke78 and head trauma79.

Intra and extra-cellular diffusion coefficients as well as modifications in cellular volume

fractions, are believed to influence the ADC80.

In vasogenic edema, an increase of the extracellular volume fraction is implicated in the

elevation of ADC79,81. On the other hand, in cellular edema, it is believed that increased

extracellular tortuosity82 and translocation of protons from a faster diffusing extracellular

compartment into the slower diffusing intracellular space83,84, contribute to the ADC decrease

seen in this type of edema.

In diffusion weighting we measure diffusion by signal loss. We apply 2 identical field gradients

in opposite directions to each other: 1 for dephasing, and a 2nd one for rephasing. If we were to

add these gradients, their sum would be zero. However, if some sort of molecular water

movement takes place in direction of one of these gradients, the gradients will no longer cancel

each other out, resulting in net signal loss.

13

The amount of detectable motion depends on the gradients, which are in turn set by the b-factor.

Finer changes of water motion become detectable as we measure with larger b-factors.

In our mono-exponential diffusion measurement, the b-values range is 0 – 1000 s/mm2. The

resulting mono-exponential ADC values represent only a net change in the diffusion of water

molecules, without giving us any clue as to the components making it up.

By extending the b-value range of the measurement up to 6000 s/mm2, the DWI may be

examined from yet another point of view. At high b-values the ADC becomes multi-exponential.

Using a bi-exponential mathematical model, this decay can be interpreted as the sum of the fast

(ADCfast) and slowly (ADCslow) diffusing components85. Therefore, one can measure the

components making up the mono-exponential ADC: The ADCslow, the ADCfast, and the

percentages (P-slow, P-fast) in which these components contribute to the change in mono-

exponential ADC.

There is much debate in the literature about the origin of “biexponential diffusion”. The

biexponential diffusion paradox is that the identified water populations do not match the

physiological extra and intracellular compartments53,85-87. It was pointed out that even without

compartmentation the biexponential signal decay is still present53 and it can also be observed in a

single cell88. Therefore it would be wise to presume that the fast and slow components do

represent two water populations and it is better to define components rather than compartments.

Many speculations have been put forward about the two components: sol and gel state of water89,

bulk water and membrane bound water90, free water and bound water in the hydration shells of

the macromolecules53. All of these speculations about a two-component model assume two

distinct water populations with two distinct apparent diffusion coefficients (ADC). Since these

speculations could not be proven so far, the biexponential model itself was questioned91 arguing

about a continuous, broad distribution of ADC values92. Thus other mathematical models have

been proposed to describe the diffusion and the consequent signal decay in diffusion weighted

imaging (DWI): diffusional kurtosis imaging (Taylor expansion of the logarithm of signal

decay)91,93,94, inverse Laplace transformation92, and stretched exponential95.

Several in-vivo as well as in-vitro models have been tested using the bi-exponential analysis in

order to gain better insight into this phenomenon. These models include global ischemia and cold

injury in mice, as well as packed human erythrocyte samples examination53, human studies of

extensive peritumoral edema73, osmotic perturbations of human hippocampus slice96, and

14

exposure of rat hippocampus to ouabain97. To this day, it is still not absolutely clear what do the

ADC and its components represent.

3.4. Magnetic resonance spectroscopy (MRS) - water signal:

Our clinical MR machine is tuned to measure the resonance of protons. However not all

hydrogen protons resonate at the same frequency. The reason for this is the fact that H protons

are influenced by the chemical environment of the macromolecules in which they reside. In

normal spectroscopy, because of this phenomenon known as the chemical shift, certain

metabolites are resonating at different characteristic frequencies. Since the proton signal is much

higher than the metabolites signals, we usually must suppress the water signal98. By not

suppressing the water signal, and eliminating other influencing factors, it is hypothesized that the

integral of the water signal will be proportional to the number of protons in it. Thereby

estimating the amount of water present within the measured voxel.

15

II . AIMS:

The aim of our study was to better understand the process taking place during brain edema and

find new and better ways for diagnosing and treating the hyponatremic brain edema in particular.

Our hypothesis was that ENaC and sodium flux into the cells might be involved in the initiation

and propagation of brain water accumulation in severe hyponatremia.

1.) The initial goal was to test whether Benzamil hydrochloride, a specific ENaC blocker, has an

effect on brain edema, and whether it has the power to decrease the hyponatremic edema.

2.) The secondary aim was to develop a modality that would allow us to investigate brain edema

and the effects of Benzamil hydrochloride in an in vivo experiment. For this task MRI was

chosen. MRI has proven to be a highly valuable tool in the diagnosis and treatment of

neurologic pathologies. The advantage of using MRI is that the brain edema can be followed

in the in vivo setting. In order to achieve this goal, we had to develop the methodology that

would allow us to use our clinical human 3T MRI machine to scan small animals as test

subjects.

3.) The tertiary goal of the research was to advance our knowledge about the initiation and

propagation of the hyponatremic brain edema, as well as the dynamic effects of Benzamil

treatment on it.

4.) Another aim of the research was to better understand the background behind MR related

observations of water movements in brain edema. The goal was to develop a new method to

measure the hyponatremic brain edema, by using DWI biexponential analysis in an in vivo

intracellular brain edema model.

16

III. Benzamil prevents water accumulation in hyponatremic rats:

1. MATERIALS AND METHODS:

Male Wistar rats weighing 250-300 g were used. The animals were anaesthetized in a bell jar for

5 minutes under 4% isoflurane and a 70:30 mixture of N2O:O2, endotracheally intubated and

maintained under 1.5-2% isoflurane. The animals were then fixed on to a stereotaxic apparatus.

Rectal and temporal temperatures were monitored and maintained at 37°C with a ventral heating

pad. A midline scalp incision was made and the skull was exposed with a blunt dissection. The

exposed parietal bones were cleaned and dried with bone wax. The animals were randomly

assigned to four experimental groups, each consisting of 13 animals. For intra-cerebro-

ventricular (ICV) administration of various substances, a cannula was inserted into the right

lateral ventricle with the following stereotaxic coordinates referenced to bregma: 0.8mm

posterior, 1.5mm lateral and 4.0mm deep. Proper placement of the tip of the cannula was tested

in preliminary experiments by injecting Evans blue dye into the ventricle. The animals in group I

(normonatremia + 0.9% NaCl) and in group II (hyponatremia + 0.9% NaCl) were injected ICV

with 4µl physiological saline. Animals in group III (hyponatremia + AVP) received 120µg

Arginine vasopressin (AVP) – a Na channel activator as it was used in previous studies71,

whereas those in group IV (hyponatremia + Benzamil) received 4µg Benzamil. The dose of

Benzamil hydrochloride was selected based on a pilot study and according to Nishimura et

al99.Following the ICV treatment, systemic hyponatremia was induced in groups II-IV by

intraperitoneal (IP) administration of 140 mmol/l dextrose solution at a dose of 20% of body

weight. The exact time course of hyponatremia was not determined. However, in the present and

in previous studies, occasional plasma sodium concentration measurements were made, and

severe hyponatremia defined as plasma sodium levels of 125mmol/l or less was already attained

after 30 min of water loading. The time-sensitivity and dynamics of development of

hyponatremic brain edema were addressed previously in essentially the same experimental

setting by using diffusion-weighted magnetic resonance imaging71. After the IP dextrose

injection, the animals were allowed to survive for 120 minutes. At the conclusion of the

experiments the animals were deeply anaesthetized, killed and their brains were removed.

The tissue water content of the total brain was determined by weighing the brain before and after

drying at 95°C for 48 h then at 105°C for further 48 h. The dehydrated samples were digested in

17

5 ml of 1 mol/l hydrochloric acid for 1 week. The sodium and potassium contents of these

solutions were then measured by using a flame photometer. Plasma sodium concentration and

plasma osmolality were determined with an ion-selective electrode, and a vapor pressure

osmometer, respectively.

2. Statistical analysis:

For statistical analysis ANOVA post-hoc kruskal-wallis test was used. Results are presented as

mean ± standard deviation.

3. Results:

As shown in Table III.1, IP administration of isotonic dextrose solution resulted in severe

hyponatremia with a corresponding fall in plasma osmolality by the end of the 2-h-long

hydration protocol. Hyponatremia was associated with a significant decrease of the sodium and

potassium contents in the brain tissue. ICV pretreatment had no effect on these parameters.

By contrast, brain water content responded to severe hyponatremia with a significant increase

from the control value of 77.55±1.00% to 78.45±0.94% (p<0.01) followed by a further rise to

79.35±0.80% (p<0.0005) when ICV AVP was given. Interestingly, the brain water content of the

Benzamil group was 77.61±1.04%. i.e. ICV pretreatment with Benzamil completely abolished

brain water accumulation that occurred in response to severe hyponatremia (Fig. III.1).

Figure III.1: Brain water content.

Brain water content in the different experimental groups: Group I: normonatremia + icv 0.9% NaCl; Group II: hyponatremia + icv 0.9% NaCl. Group III: hyponatremia + icv AVP. Group IV: hyponatremia + icv Benzamil.

18

Table III.1:

Effects of intracerebroventricular injections on osmolality, sodium and potassium content of plasma and brain tissue:

Group Number

Plasma Na mmol/l

Plasma osm. mOsm/l

Tissue Na mmol/l

Tissue K mmol/l

I. (NN + NaCl) 145.82±4.81 315.05±7.47 24.88±1.36 12.26±1.29 II. (HN + NaCl) 110.25±7.8*** 276.15±9.23*** 20.92± 1.32*** 9.25± 0.57*** III. (HN + AVP) 111.56± 11.94*** 275.12±10.94*** 21.08±1.97** 9.37±0.72*** IV. (HN + Banz.) 109.63±12.64*** 274.23±8.21*** 21.28±1.5*** 9.01±0.64*** Effects of intracerebroventricular injections on osmolality, sodium and potassium content of plasma and brain tissue:

Effects of intracerebroventricular injections of: 0.9% NaCl, arginin vasopressin (AVP) and Benzamil (Benz) on Sodium, osmolality and potassium concentrations. NN: normonatremia, HN: hyponatremia. **p<0.001, ***p<0.0005 as compared with normonatremic controls.

4. Discussion

These results show that ICV pretreatment with Benzamil, a specific sodium channel blocker,

completely abolished the hypo-osmolality-related brain edema formation. These observations

imply that cellular sodium uptake via activated sodium channels generates an osmotic gradient

and may mediate osmotically driven passive water flux across AQP4, the brain specific water

channel membrane protein.

A well-documented fact is that Benzamil-blockable sodium channels function as one of the Na+

receptors in the central nervous system. These receptors allow the influx of extracellular sodium

into cells, which in turn causes a release of endogenous AVP.

Enhanced AVP secretion in turn activates sodium channels and contributes to further

accumulation of cellular water99. Convincing experimental evidence, supporting this notion, has

been provided demonstrating that the AVP-stimulated cellular sodium transport operates in the

cortical collecting duct (CCD) cells in the kidney. Namely, AVP has been shown to increase

sodium transport in the whole kidney100, in perfused CCD101-104, in primary CCD cell

suspension105,106 and in several cell lines such as A6 cells107,108 and M-1 cells109 having the

features of CCD. Exposure to AVP stimulation was found to augment the abundance of all three

subunits (α,β,γ) of the epithelial sodium channel, which is assumed to account for the increased

amiloride-sensitive sodium transport capacity100. It is of note that AVP was also demonstrated to

have the ability to redistribute the epithelial sodium channel to the apical plasma membrane in

A6 cells110,111. The stimulatory effects of AVP on amiloride-sensitive sodium transport are

thought to be exerted through the V1b receptor-G-protein-sodium channel pathway112.

19

Considering that V1b receptors113,114 and sodium channels99,115,116 are widely distributed in the

central nervous system, it is reasonable to suggest that AVP may function in an analogous way to

increase cellular sodium and water uptake in the brain. Supporting this notion is the decreased

volume regulatory sodium uptake, in response to hyperosmotic stress, in brain tissue of AVP-

deficient Brattleboro rats117. Furthermore, relevant data indicates that AVP prevents cells from

effectively extruding osmotically active solutes during hypotonic exposure, thereby limiting the

capacity of RVD118. Moreover, the calcium-coupled sodium efflux system is also inhibited by

AVP119.

Nishimura et al. suggested that AVP stimulates insertion of intracellular membrane vesicles

containing activated sodium channels into the cell membrane by increasing intracellular cAMP99.

In this way, AVP is likely to enhance the activity, density and selectivity of the amiloride-

sensitive sodium channel for Na+ in brain cells of spontaneous or DOCA-induced hypertensive

rats. Activated sodium channels in turn may function as one of the brain sodium receptors and

may be involved in the stimulation of AVP release when the brain is exposed to elevated

concentrations of extracellular sodium99. On the basis of these observations, it is tempting to

postulate that prevention of brain water increase in hyponatremic rats may be achieved by

interfering with AVP, with the complex interaction of the V1b receptor-activated sodium

channel, the cellular sodium influx and osmotically driven AQP4-mediated water accumulation.

It is to be considered, however, that simultaneously with the activation of the sodium channel,

the hyponatremic challenge reduces the osmotic water conductivity of AQP4 possibly through

phosphorylation of this water channel protein. As a result, osmotic water flow into the cells is

limited, and development of brain edema is attenuated120,121. It is also to be considered that

hypotonic stress has been shown to directly stimulate sodium channel activity in the absence of

AVP122,51, but this increased transmembrane sodium transport can be completely abolished by

Benzamil.

20

IV. Quantitative proton MRI and MRS of the rat brai n with a 3T clinical MR scanner:

1. Materials and methods:

1.1. Experimental protocol:

Eight Wistar rats (three females, five males) weighing 250-300g were used. The rats were

intubated endotracheally and anesthesia was maintained with 1.5% isofurane at a 70:30 mixture

of N2O:O2. The temperature of the animals was controlled using an isothermal pad. Rectal

temperature was 36.5-37.5°C throughout the experiment. Each animal was fixed in a custom-

built MRI compatible head holder and placed through a loop of radio frequency (RF) coil.

Following the completion of the MRI and MRS experiments, the animals were sacrificed with an

overdose of isofurane, and their brains were rapidly extracted in a humid chamber. The brains

were immediately weighed and dried to constant weight at 90°C. Water content in percentage

(W) of each sample was calculated according to Eq. IV.1. This data was utilized later on for the

calibration of the quantitative MRS measurement.

Eq. IV.1: W = 100 × (wet weight − dry weight)/wet weight

1.2. MRI protocol:

A 3.0 T (tesla) clinical MRI scanner with a field gradient strength of 40 miliTesla/meter (mT/m)

was used. The body coil was used for excitation, and signal acquisition was carried out by

utilizing a commercial loop of RF coil having an inner diameter of 40mm. Following scout

images acquisition of the rat brain, the MRI and MRS measurements were performed.

In a separate experiment, high-resolution T2-weighted MRI was performed on three rats

using a turbo spin echo imaging sequence: TR/TE = 3000/76ms, slice thickness = 0.13mm,

FOV = 33×33mm2, 256×256 pixel matrix, bandwidth = 40Hz/pixel, fat suppression with spectral

attenuated inversion recovery method (SPAIR), number of acquisitions = 1.

Diffusion was determined with a trace-weighted single shot echo planar imaging sequence:

TR/TE = 4000/119ms, slice thickness = 1.9mm, distance factor = 0mm (i.e., no gap),

FOV = 30×95mm2, 40×128 pixel matrix, bandwidth = 830Hz/pixel, number of acquisitions = 4,

number of slices = 14, b values: 0, 500, 1000, 2000, 3000, 4000, 5000 and 6000s/mm2.

21

Prior to the MRS measuring, a 10×10×5mm voxel was designated in the middle of the rat

brain (Fig. IV. 1). Following localized manual shimming and water suppression adjustment, a

fully relaxed short-echo-time proton MR spectra (PRESS, TR/TE = 6000/30ms, 64

accumulations) were acquired. For water suppression, chemical shift selective sequence

[CHESS] pulse was used. At the end of the MRS experiment, water suppression was turned off

in order to record a reference water signal for the metabolite concentration calibration.

Fig. IV. 1: Magnetic resonance spectroscopy.

Magnetic resonance spectroscopy: Positioning of the voxel on the animal`s brain, and the corresponding proton spectrum, showing major peaks of NAA: N-acetyl-aspartate; Cho: Choline; Cr: tot. creatine.

1.3. Data processing:

ADC maps were generated by a monoexponential fit of signal intensities in the low b value

range (i.e., b = 0, 500 and 1000 s/mm2). Using the biexponential fit (Eq. IV. 2) over the extended

b – value range, the ADC values of the rat cortex were broken down into ADCfast and ADCslow.

Eq. IV. 2: M/M0 = ffast exp(−b × ADCfast) + fslow exp(−b × ADCslow)

Where M is the signal in the presence of diffusion sensitization, M0 is that in the absence of

diffusion sensitization, ADCfast and ADCslow are apparent diffusion coefficient values, and ffast

22

and fslow are the contributions to the signal of the fast and slow-diffusing water compartments.

Freehand regions of interests (ROIs) were drawn on the maps to yield mean values.

Using the Siemens Leonardo Workstation, raw spectroscopic data were post-processed

with the built-in Siemens spectroscopy software. We applied a Hanning filter and zero filling to

2048 data points before Fourier transformation. Following manual baseline and phase correction,

an automatic fit was applied to the Fourier-transformed spectra, and the peak integrals of water,

choline (Cho), creatine (Cr) and N-acetyl-aspartate (NAA) were determined. The molar tissue

water content (MWC) was calculated from the tissue water content (W) determined by the

desiccation method and the molar concentration of pure water (55.6 mol/L). Absolute tissue

metabolite concentrations could then be calculated using Eq. IV. 3.

Eq. IV. 3: C = IM × MWC × 2/(n × IW)

where C is the molar concentration of the metabolite, n is the number of resonating protons, IM is

the peak integral of a given metabolite, and IW is the peak integral of water. This formula applies

only for spectra with a long TR and a short TE.

1.4. Statistical analysis:

In every animal, two ROIs were defined on each side of the cortex. Mean value and standard

deviation (SD) were calculated by taking into account pixelwise data of both ROIs. The

coefficient of variation (CV) was calculated (i.e. SD-to-mean ratio) in order to normalize the SD.

2. Results:

The high-resolution T2 weighted image (Fig. IV. 2) demonstrates the capabilities of a 3 T

human MRI machine to achieve a microscopic resolution of 130×130×130µm3 (Signal to noise

ratio (SNR) = 30).

By utilizing only a low b value range (i.e., 0,500 and 1000 s/mm2), ADC maps were created (Fig.

IV. 3). The low spatial resolution of the ADC map is mainly attributed to the gradient limitations

of the scanner and not to the low SNR. In table IV.1, quantitative ADC data are presented. The

large voxel size in DWI produced such a high SNR which allowed us to extend the acquisition of

diffusion images to larger b values. Even at b values of 6000s/mm2 the SNR was higher than 15.

23

Using the biexponential approach throughout the whole b value range, ADCfast, ADCslow and

their corresponding volume fractions were determined with acceptable standard deviation.

Table IV.1:

Quantitative ADC data: ADCmono ADCfast ADCslow Pfast Pslow

Mean 6.02 9.36 2.04 74.3 % 25.7 % CV (in %) 3.8 9.7 16 6.1 17.5 ADC data and coefficient of variations (CV). Units of ADCmono, ADCfast, and ADCslow are presented in (×10-4mm2s-1)

Fig. IV.2: T2-weighted image.

High resolution (130×130×130µm3) T2-weighted image of the rat brain.

Fig. IV.3: ADC map.

ADC map.

24

In Spectroscopy, the localized manual shimming of the voxel was essential in the signal

acquisition. A representative spectrum of the rat brain is shown in Fig. IV.1. The SNR for the

NAA peak was ≈17. This could have allowed for a smaller voxel size, however, due to

limitations in the commercial PRESS sequence, this was impossible. The mean water

concentration (W) in the rat brain was 78.0±0.5%. This was utilized for the calibration of the

water peak as described in Eq. IV.3. The metabolite concentrations of NAA, Cr and Cho are

listed in table IV. 2.

Table IV. 2:

Metabolites concentrations in the rat brain: Choline (mmol/l) Creatine(mmol/l) N-acetyl-aspartate (mmol/l) Mean value 1.77 6.25 8.78 C.V. (in %) 21 25 14 Mean values and coefficient of variations (CV) of metabolites concentrations in the rat brain.

3. Discussion:

High quality small animal imaging on clinical MR scanners has been made feasible thanks to

technological advances54,63,69,123 made in the field. With the development of higher magnetic

field strength scanners as well as custom built coils, the use of small animals as test subjects for

MR spectroscopy has also become a viable option124,125. Recently, mouse brain MRI experiments

using a 2.35 T scanner were made with microscopic resolution126. Similar high resolution

anatomical studies of rat brain were successfully performed using our 3T scanner at the expense

of long measurement time (146 minutes), which is usual in these kinds of experiments69,128.

It was previously demonstrated, mainly in rat brain stoke studies127, that human scanners have

the capability to measure quantitative ADC data in the low b value range. The current set-up at

3T gives a good SNR, but its limitations (mainly in gradients) do not allow better spatial

resolution with the applied EPI sequence. Similar low spatial resolution ADC images obtained

with clinical scanners have been reported in the literature at 3 T63. Such high SNR allowed us to

extend the b values to a greater range. By using the biexponential analysis, we were able to gain

further insight into the components making up the ADCmono. The ADCfast and ADCslow values are

in very good agreement with available data in the literature53,73,85. ADCmono, (×10−4 mm2s−1),

ADCslow (×10−4 mm2s−1) and the volume fractions (%) are practically identical to the results of a

mouse study73 at 9.4 T (ADCmono: 6.02±3.8% vs. 5.90±3.4%; ADCslow: 2.04±16.0% vs.

25

2.02±9.9%; pfast: 74±4.0% vs. 75±4.0%). It is intriguing that despite the markedly increased

diffusion and echo time, the volume fractions do not change. Despite the current debate in the

literature about the origin of the biexponential signal decay of diffusion measurements, our

results show good reproducibility regardless of the field strength and echo time.

Quantitative proton MRS in the rat brain was preformed successfully. The capability of a 3T

clinical scanner to measure MRS of the rat brain was previously demonstrated through the use of

special, non-commercial hardware elements124,125. In our study, the metabolite concentrations are

in overall agreement with literature data128,129 and they are practically identical to values reported

for mouse brain using the same water reference strategy130.

26

V. The influence of Benzamil hydrochloride on the evolution of hyponatremic brain edema

as assessed by in vivo MRI study in rats:

1. Materials and methods:

1.1. Experimental protocol:

Thirty three male Wistar rats weighing 350–450 g were used. The animals were randomly

assigned into one of five groups: The NN group (normonatremic native animals, n=7), the HN

group (hyponatremic animals, n=8), the HNB group (hyponatremic animals treated with ICV

Benzamil, n=8), the HNS group (hyponatremic animals treated with ICV saline, n=5), and the

NNB group (normonatremic animals treated with ICV Benzamil, n=5). The animals were

anesthetized in a bell jar for 5 minutes under a 70:30 mixture of N2O:O2 containing 4%

isofurane. The animals were then intubated endotracheally and anesthesia was maintained

continuously with 1.75–2.25% isofurane (Fig. V.1).

Fig. V. 1: Animal anesthesia and peritoneal cannulation.

Animal anesthesia and peritoneal cannulation: A veterinarian small animal anesthesia device used to maintain constant anesthesia, and commercially available venous cannulas and infusion tubes used for peritoneal cannulation.

The rectal temperature was measured continuously, and the animals were maintained at 36–

37°C. Through the use of an MRI compatible rectal probe and thermometer, and by using a

27

feedback mechanism that controlled a custom-built silicon tube, shaped to form a cylinder

around the animal, connected to a hot/cold water circulator device it was possible to maintain

constant body temperature in the animals (Fig. V.2).

Fig. V. 2: Temperature control.

Temperature control:

An MRI compatible rectal probe and thermometer, and a silicon tube, shaped to form a cylinder around the animal, connected to a hot/cold water circulator device. The head of each animal was fixed in a custom built MRI-compatible head-fixing device, and its head was pulled through a loop of RF coil.

For ICV Benzamil or Benzamil vehicle (saline) administration to the HNB, HNS, and NNB

groups, prior to their placement in the MRI machine, the animals were placed in a stereotaxic

device. A midline scalp incision was made, the skull was exposed, and then drilled at the

following coordinates referred to the bregma: 0.8 mm posterior, 1.5 mm lateral. Water

intoxication was employed in order to induce hyponatremia inside the MRI machine. For this,

the peritoneum of each animal was cannulated with commercially available venous cannules and

infusion tubes (Fig. V.1). Following the preparation, each rat was fixed in a custom built MRI-

compatible head-fixing device, and its head was pulled through a loop of RF coil (Fig. V.2) The

animal was then placed in the MRI machine, and its head was aligned parallel to the axis of the

magnet. From this point on, the animals remained untouched inside the MRI machine up until

the completion of the 2-h MRI protocol. After establishment of the baseline MRI measurement

28

for each animal, induction of acute hyponatremia was achieved by way of water loading. For

induction of hyponatremia, 140 mmol/l dextrose solution in a dose corresponding to

approximately 20% of the animal’s body weight was injected by bolus injection via the

previously prepared peritoneal cannula from outside the machine, while the animal remained

immobile inside the machine. The dextrose solution was maintained at a temperature of

35.8±1°C in order to avoid any dramatic temperature shifts following the IP injection. The

correctness of the IP injection site was verified by visualization with an MRI measurement,

before and directly after the animals in the HN, HNS, and HNB groups had received the injection

(Fig. V.3).

Fig. V. 3: Intraperitoneal injection verification.

Intraperitoneal injection verification: Correct IP (Intraperitoneal) injection verification accomplished by comparing the images. The dextrose solution is apparent in the peritoneal cavity immediately after the IP injection (bottom image, marked with an arrow) as compared to the image taken right before the IP injection (Upper image).

Immediately following verification of the IP injection, 4µg Benzamil hydrochloride in 4µl saline

was injected ICV into the right lateral ventricle of the HNB group animals, at the pre-drilled site

coordinates that referred to the bregma: 0.8 mm posterior, 1.5 mm lateral, and 4 mm deep. The

appropriate location for injection site into the lateral ventricle was established in our earlier

experiments, as well as previous studies71. Immediately after the ICV injection of Benzamil or

saline, the MRI measurement protocol began. Following completion of the MRI measurements,

29

the animals were removed from the machine. They were deeply anesthetized, and blood was

withdrawn via a cardiac puncture for determination of the serum sodium and potassium

concentrations, as well as plasma osmolality. Immediately following this, the animals were

sacrificed with an overdose of thiopental.

1.2. MRI protocol:

MRI measurements were made on a 3.0-T clinical MRI scanner with a field gradient strength of

40mT/m. The body coil was used for excitation, and a commercially available loop of RF coil

with an inner diameter of 40mm for signal acquisition.

For establishment of the baseline for the MRI measurements, following the correct placement of

each animal into the machine, three spectroscopic and diffusion-weighted measurements were

made. After the IP injection, the correct location of the injection was verified via a turbo spin

echo sequence. Following verification, serial diffusion weighted and spectroscopy scans were

recorded every 10 minutes until the completion of the MR protocol 2 hours (h) later. In the

groups receiving ICV injection, Benzamil or saline was injected right after the IP injection, and

immediately before the serial MRI scans.

For IP injection site verification, a turbo spin echo imaging sequence was run, before and

immediately after the IP injection, by using the body coil for signal reception: TR/TE = 3000/78

ms, slice thickness=4 mm, distance factor=0.8 mm. FOV=300 mm, 77x256 pixel matrix,

bandwidth=217 Hz, echo train length=20, number of slices=6 in sagittal orientation; standard fat

suppression was used.

Diffusion was determined with a trace-weighted single-shot echo planar imaging sequence:

repetition time/echo time (TR/TE)=3,000/86 ms, slice thickness=1.9 mm, distance factor=0 mm

(i.e. no gap), field of view (FOV)=29×95 mm, 40×128-pixel matrix, bandwidth=830 Hz, number

of acquisitions=4, number of slices=10, b values: 0, 500, and 1,000 s/mm2.

For water signal acquisition proton spectroscopy was used. The voxel size was adjusted

individually to each animal in order to match the brain size perfectly (the average voxel size was

13×10× 6 mm). After localized manual shimming, only an unsuppressed water signal was

recorded. For this, a fully relaxed short-echo-time proton MR spectrum was run with the PRESS

(point-resolved spectroscopy) sequence: TR/TE=10,000/30 ms, number of acquisitions = 1,

preparation scans=1, vector size=2,048 points, bandwidth=2,500 Hz, no water suppression.

30

1.3. Data processing:

Diffusion analysis was performed with the MATLAB curve-fitting toolbox. For the curve

fitting Eq.V.1. was used:

Eq.V.1: M = M0 × exp( - b × ADC)

Where M is the signal in the presence of diffusion sensitization, M0 is the signal in the absence

of diffusion sensitization, b is the b value describing the diffusion sensitizing gradient used for

the measurement, and ADC is the apparent diffusion coefficient value. Freehand ROIs were

drawn on diffusion-weighted images and parameters were calculated according to the selected

region. ROIs were placed in such a way that they covered both the right and left cortices of the

brain.

For water signal evaluation, standard spectroscopy software was used on a Siemens

Leonardo workstation. For signal processing only Fourier transformation was applied. The

frequency domain of the water signal was fitted according to Lorenzian distribution, and the

integral of the fitted signal was used for comparison between the different time points.

The changes in ADC and water signal were referred to the average of the initial three base line

measurements.

1.4. Statistical analysis:

The data for the experimental groups were compared by using one-way analysis of variance

(ANOVA), and two sample equal variance Student t-tests. The two statistical evaluations gave

practically identical results. Values are expressed as mean±standard deviation.

2. Results:

Body temperature of each animal was maintained very strictly within narrow limits. The IP

injection in the HN, HNB and HNS groups did not cause a significant temperature change

relative to the baseline measurement. The average rectal temperatures of the animals were

maintained within 1°C in each group.

31

By the end of the hydration protocol, roughly 130 minutes from the time of hyponatremia

induction, the IP injected dextrose solution had induced a marked decrease in serum osmolality

from 312.71±8.63 mOsm/l (NN), to 265.85±4.24 mOsm/l, 271.33±4.5 mOsm/l and 268.25±9.75

mOsm/l in the HN, HNS, and the HNB groups, respectively (p<0.0001) (Table V.1).

Table V.1:

Blood parameter analysis: Group Description Na (mmol/l) K (mmol/l) Osmolality (mOsm/l) Gr. I. Native animals 142.11±2.33 6.33±1.8 312.71±8.63 Gr. II. HN animals 109.27±3.42* 6.91±1.58 265.85±4.24* Gr. III. HN + Benzamil animals 108.86+4.13* 7.25±1.55 268.25±9.75* Gr. IV. HN animals + saline 105.7±7.77* 6.86±2.63 271.33±4.5* Gr. V. Native animals + Benzamil 139.33±1.52 5.97±1.01 313.66±1.15 Blood parameter analysis:

Serum sodium, potassium, and osmolality at the end of the 2 hour MR protocol. H.N: hyponatremic. * p<0.0001 as compared to native group.

Similar to the serum osmolality, the serum sodium concentration decreased significantly from

142.11±2.33 mmol/l (NN), to 109.27±3.42 mmol/l, 105.7±7.77 mmol/l and 108.86±4.13 mmol/l

in the HN, HNS and HNB groups, respectively (p<0.0001). Benzamil administered ICV to native

animals had no significant effect on these parameters relative to those in the control group.

The diffusion-weighted scans yielded ADC maps showing a mean ADC range of 6.90–

7.40×10−4 mm2s−1 in the cerebral cortex of the native animals.

32

Fig. V. 4:

Apparent diffusion coefficient values: ADC values measured every 10 minutes, normalized to the average of the initial three baseline measurements. Normonatremic animals (NN); Hyponatremic animals (HN); Benzamil-treated hyponatremic group (HNB); Saline- injected hyponatremic animals (HNS); Benzamil-treated normonatremic group (NNB). * p<0.05; ** p<0.0005 In the HN group ADC started to decline progressively from the baseline immediately after the

induction of hyponatremia. This decrease became significant (p<0.05) at 20 min. The ADC

continued to decrease to reach a minimum of -7.67±3.20% at 90 min (p<0.0005) from the base

line. In Benzamil pretreated animals this ADC decline was absent. In the HNB group there was

no decline in ADC throughout the 2-h hydration protocol and there was no significant difference

between this group and that of the native animals. ADC in animals of the HNS group, treated

with icv saline, did not show significant difference compared with the HN group (Fig.V.4).

In contrast to the ADC, the water signal of the HN group increased markedly relative to the

NN group. This elevation became significant (p<0.05) at 40 min., and reached its maximum at

100 min with a 5.95±2.62% increase (p<0.0005) from the base line (Fig.V.5).

33

Fig. V. 5:

Water signal values: Water signals measured every 10 minutes, normalized to the average of the initial three baseline measurements. Normonatremic animals (NN); Hyponatremic animals (HN); Benzamil-treated hyponatremic group (HNB); Saline- injected hyponatremic animals (HNS); Benzamil-treated normonatremic group (NNB) * p<0.05 ** p<0.0005 The water signal of the HNB group was lower than that of the HN group (p<0.01). Benzamil pretreatment of normonatremic animals did not cause any significant changes in the water signal compared with the native animal group, and pretreatment of hyponatremic animals with ICV saline showed a tendency of water signal elevation similar to the one measured in the HN group. The differences between these groups proved to be non-significant.

3. Discussion:

The present in vivo study provides experimental evidences that serial MRI scans are a reliable

quantitative estimate of brain water accumulation (MR spectroscopy - water signal) and its

cellular distribution (ADC) in hyponatremic rats. Moreover, it clearly shows that ICV

pretreatment with Benzamil hydrochloride, a specific ENaC blocker, promptly and effectively

inhibits the evolution of hyponatremic brain edema. These results suggest that ENaC activation

is intimately involved in the initiation and progression of brain edema formation in rats exposed

to acute hyponatremic stress. It is of practical importance that the MRI-derived in-vivo water

measurements correspond well to our previous in-vitro direct tissue water determination, as seen

34

in chapter III. Benzamil-blockable brain ENaCs have been claimed to play an important role in

the perception of sodium concentration in CSF and brain tissue. They were also implicated in the

mediation of hemodynamic, autonomic nervous and endocrinological responses induced by ICV

administration of hypertonic NaCl. Indeed, acute increase of CSF sodium concentration in

hypertensive rat models induced an elevation in mean arterial pressure, heart rate, norepinephrine

and vasopressin secretion. All these effects could be abolished by ICV pretreatment with

Benzamil99. Similarly, ICV Benzamil administration markedly reduced mRNA and protein

expression of the brain rennin angiotensin system in DOCA-salt hypertensive rats116. Intravenous

Benzamil administration however, had no effect on these parameters, implying that Benzamil

hydrochloride does not readily cross the BBB. The interrelationship of ICV Benzamil, sodium

levels and brain ENaCs is not clearly defined. In our acute study, ICV Benzamil administration

to normonatremic animals did not alter plasma sodium levels, however, tissue sodium was not

measured. Recently, it was shown that ICV Benzamil infusion for 1 or 2 weeks in rats, increased

CSF sodium concentration and lowered hypothalamic tissue sodium significantly. Similar

changes were observed when sodium-rich artificial CSF was infused ICV43. The elevated CSF

sodium in turn enhances the gene and protein expression of ENaC subunits and those of some

related proteins in the brain epithelia43. With respect to the acute nature of our experiments, ICV

administration of Benzamil into hyponatremic rats is unlikely to alter brain ENaC synthesis but

rather it may interfere with surface expression and/or gating mechanisms of this channel

protein131. Our results are in apparent conflict with the traditional concept of RVD. RVD is a

complex process consisting of several distinct and interdependent functional elements by which

the brain adapts to hyponatremia and permits survival without fatal degrees of cerebral edema.

Many details of the molecular and cellular mechanisms of volume adjustment to hypo-osmotic

challenge have been described in individual cells and in the whole brain. It is intriguing that

sodium flux has not been implicated yet in RVD mechanisms. It is to be considered, however,

that hypotonic stress itself has been shown to stimulate directly ENaC activity and several lines

of evidence suggest that in response to hypo-osmolality transmembrane sodium influx is greatly

increased. This increase however, can be completely abolished by Benzamil50,51,132,133. Some

aspects of the underlying mechanism(s) of this process have been explored by demonstrating that

(i) hypotonic activation of cellular sodium uptake is dependent on the protein tyrosine

kinase/tyrosine kinase receptors50,51, and (ii) it can be divided into two phases: during the early

35

phase, translocation of the preformed sodium channels occurs into the apical membrane by way

of a protein synthesis-independent pathway. In the late phase, sodium channels mRNA

expression is stimulated, and new channel protein is synthesized51. The signal for sodium

channel trafficking and synthesis includes the reduction of intracellular osmolality and decrease

of cytosolic Cl- concentration134,135. A further support for the possible role of ENaC in the

development of brain edema is that the elements of V1b receptor-G protein-sodium channels are

widely distributed in the central nervous system and central AVP has been documented to

enhance cellular sodium and water uptake43,113,136,137.

All these observations can be regarded as convincing evidences for an important role of

hypotonicity-induced sodium transport in RVD. One can assume that extrusion of osmolytes

from the hypotonic cells is opposed by sodium influx, ensuring a fine tuning of cell-brain volume

regulation. Moreover, this control mechanism may prevent overshooting of the reduction of

osmolytes and volume when RVD is accomplished and the initial volume of cells is regained.

36

VI. A biexponential DWI study in rat brain intracel lular oedema:

1. Materials and methods:

1.1. Experimental protocol:

Ten male Wistar rats weighing 380–450 g were used. The animals were endotracheally intubated

and anesthesia was maintained at 1.75–2.25% isofurane in a 70:30 mixture of N2O:O2. Rectal

temperature was continuously measured using an MRI compatible rectal probe. The animals

were maintained at 36–37°C. Using a silicon tube, shaped to form a cylinder around the animal,

connected to a hot/cold water circulator device and employing a feedback mechanism, we were

able to heat or cool the animal’s surrounding (Fig. V.2).

For water loading and hyponatremia induction, the animal’s peritoneum was cannulated using a

commercially available venous cannulas and infusion tubes (Fig. V.1). Following this

preparation, each animal was fixed in a custom-built MRI compatible head fixing device, and its

head pulled through a loop of RF coil. The animals were then placed into the MR machine, and

their brains were aligned parallel with the magnet’s axis. Following establishment of a base-line

measurement, the water loading took place. An IP injection of 140mmol/L dextrose solution at a

dose of 20% body weight was given from outside of the MR machine. Correct IP injection site

was visualized with an MR measurement. At the completion of the MR measurements, the

animals were removed from the MR and sacrificed with an over-dose of thiopental. Blood was

acquired by way of cardiac puncture for serum analysis of sodium and potassium as well as

plasma osmolality. The animals then underwent perfusion fixation using physiological saline and

5% glutaraldehyde in Cacodylate buffer. Following 24 h, the brains of the animals were removed

and analyzed for electron microscopy (EM). The exact method of EM analysis is described

previously138.

1.2. MRI protocol:

3.0-T clinical MRI scanner with a field gradient strength of 40mT/m was used. The body coil

was used for excitation and a commercially available loop of RF coil, having an inner diameter

of 40mm was used for the signal acquisition.

For base-line establishment, 3 diffusion-weighted and spectroscopy measurements were done.

Following verification of the correct IP injection (Fig. V.3) using a turbo spin echo sequence, the

37

MR measurement protocol began, which included serial diffusion weighted and spectroscopy

scans every 10 minutes for 2 hours.

Intraperitoneal injection site verification was achieved with a turbo spin echo imaging

sequence before and immediately after the IP injection. By using the body coil for signal

reception: TR/TE = 3000/78ms, slice thickness = 4mm, distance factor = 0.8mm. FOV = 300mm,

77x256 pixel matrix, bandwidth = 217Hz, echo train length = 20, number of slices = 6 in sagittal

orientation; standard fat suppression was used.

Diffusion was determined with a trace-weighted single shot spin echo EPI sequence:

TR/TE = 3000/115ms, slice thickness = 1.9mm, distance factor = 0mm (i.e. no gap), FOV =

29mm×95mm, 40×128 pixel matrix, bandwidth = 830Hz, number of acquisitions = 6, number of

slices = 10, b values: 0, 500, 1000, 2000, 3000, 4000, 5000, and 6000 s/mm2.

Proton spectroscopy measured water signal. The voxel was adjusted to each animal in order

to cover the brain as much as possible (The average voxel size was 13mm×10mm×5mm).

Following localized manual shimming, a fully relaxed short-echo-time proton MR spectrum was

measured using the PRESS (point resolved spectroscopy) sequence: TR/TE = 10,000/30ms,

number of acquisitions 1, preparation scans 1, vector size 2048 points, bandwidth 2500Hz,

without water suppression. In this way only the unsuppressed water signal was recorded.

1.3. Data processing:

Diffusion analysis was performed using the MATLAB curve fitting toolbox. Both

monoexponential and biexponential approaches were used while performing curve fitting of the

data. For standard mono-exponential fitting we used Eq. VI.1.

Eq. VI.1: M = M0 × exp(−b × ADC)

where M is the signal in the presence of diffusion sensitization, M0 is that in the absence of

diffusion sensitization, b is the b - value describing the diffusion sensitizing gradient used for the

measurement (b - values of only 0, 500, 1000s/mm2 were used for the mono-exponential fit),

ADC is the apparent diffusion coefficient value.

By analyzing the whole measured b - value range (b 0–6000s/mm2), the ADC values of the rat

cortex were broken down into ADCfast and ADCslow. In order to do this, Eq. VI. 2 was used:

38

Eq. VI.2: M/M0= pfast exp(−b × ADCfast) ± pslow exp(−b × ADCslow)

where M is the signal in the presence of diffusion sensitization, M0 is that in the absence of

diffusion sensitization, ADCfast and ADCslow are apparent diffusion coefficient values, and pfast

and pslow are the percentages of contributions to the signal from the fast and slow-diffusing water

compartments (pfast =1−pslow).

Freehand ROIs were drawn on diffusion weighted images covering both sides of the cortex.

Parameters were calculated according to the selected region.

For the water signal evaluation standard spectroscopy software was used on a Siemens Leonardo

workstation. For signal processing only Fourier transformation was applied. The frequency

domain of the water signal was fitted according to a Lorenzian distribution, and the integral

value of the fitted signal was used for the comparison between the different time points.

ADC and water signal changes are referenced to the average of the initial three base line

measurements.

1.4. Statistical analysis:

Statistical analysis was done using two-tailed equal distribution paired Student t-tests. Values are

expressed as means ± standard deviation.

2. Results:

The water loading caused a severe hyponatremia reaching a serum sodium level of

107.93±4.07 mmol/l. This hyponatremia was accompanied with a decrease of serum osmolality

to levels of 264.5±10.05 mOsm/l (Table VI.1).

Table VI. 1:

Blood parameter analysis: Serum data B-1 B-2 B-3 B-4 B-5 B-6 B-7 B-8 B-9 B-10 Average

Standard deviation

Na (mM/l) 103.2 110.9 108.2 108 114.5 108 112.5 105.9 100.9 107.2 107.93 4.0743

K (mM/l ) 6.34 5.3 6.43 6.73 6.38 8.2 6.53 6.02 7.68 5.76 6.537 0.8554 Osmol (mOsm/l) 257 257 252 272 272 277 280 265 258 255 264.5 10.058 Blood parameter analysis: Serum sodium, potassium, and osmolality at the end of the 2 hour MR protocol.

39

Prior to the water loading, an average ADC of 724±43µm2/s was measured in the cerebral

cortex of the animals. Shortly after commencing the hydration protocol, the ADC (ADCmono)

values started to significantly decline (p < 0.05) practically immediately from the reference value

of 724±43 to 706±32µm2/s. These values continued to decline reaching 682±26µm2/s at the

120min mark (p < 0.0001) (Fig.VI.1).

Fig.VI.1:

ADC – monoexponential: Average ADC values, measured every 10 minutes, in the rat cerebral cortex. Immediately following the intraperitoneal injection (IPI) the ADC values start to decline. *(p<0.05); **(p<0.001); ***(p<0.0001)

With further evaluation of the fast and slow components of ADC, we found that the ADCfast

values were declining as well. This decrease became significant at 20min, when the ADCfast

values decreased from 948±122 to 895±67µm2/s (p < 0.05). At the end of the 2 h, ADCfast

reached its lowest point of 840±66µm2/s (p < 0.001) (Fig. VI.2).

ADCslow parameters exhibited a very small yet significant decrease towards the end of the MR

measurement. The ADCslow values reached their lowest point of 191±74µm2/s (p < 0.05) at the

110 min. time point (Fig. VI.3).

40

Fig.VI.2:

ADC fast: Average values of the fast component of ADC, measured every 10 minutes, in the rat cerebral cortex. Immediately following the intraperitoneal injection (IPI) the ADC-fast values start to decline. *(p<0.05); **(p<0.001) Fig.VI.3:

ADC slow: Average values of the slow component of ADC, measured every 10 minutes, in the rat cerebral cortex. Following the intraperitoneal injection (IPI) the ADC-slow exhibits a tendency to decline. *(p<0.05)

41

An evaluation of the slow (Pslow) and fast (Pfast) volume fractions, it is evident that while the Pfast

exhibits a tendency to increase, the Pslow shows the opposite tendency to decline. The changes in

the Pfast and Pslow of 76.56±7.79% to 81.2±7.47% and 23.43±7.79% to 18.79±7.47% respectively

become significant (p < 0.05) at the 110min time point (Fig.VI. 4).

Fig.VI.4:

Percentage fast and slow fractions: Average percentages of the fast (% fast) and slow (% slow) components of ADC, measured every 10 minutes, in the rat cerebral cortex. Following the intraperitoneal injection (IPI) the ADC percentages of the fast and slow components exhibit a tendency to increase and decrease, respectively. *p<0.05

Reliability of the biexponential analysis considerably depends on the noise level, however the

signal to noise ratio was large enough even on the images acquired with a b-value of 6000s/mm2

(Fig. VI. 5).

The results of the spectroscopy measurements showed that the water signal was increasing.

The water signal continued to elevate up until the 110min time point. Reaching an increase of

4.98±3.52% from the averaged base line (p < 0.01) (Fig. VI. 6).

42

Fig.VI.5:

Signal and noise-level vs. b-value

1

10

100

1000

0 1000 2000 3000 4000 5000 6000

b value (s/mm2)

Sig

nal i

nten

sity

Signal

Noise

bi-exponential signal curveR2=0.9998104

13

29

12084

51

1914

In bold italic: signal-to-noise ratio (SNR)

Signal and noise-level vs. b-value: Average diffusion weighted signal decay originating from rat brains before edema induction. The error bars are hardly visible because of the low standard deviations. The noise level is demonstrated by a dotted line. The figure also shows the signal to noise ratio (i.e. average signal intensity over the standard deviation of the noise) at each b-value.

Fig.VI.6:

Water signal: The evolution of the total brain average water signal values, measured every 10 minutes. Following the intraperitonel injection (IPI) the water signal was increasing. *p<0.05; **p<0.01

43

Electron microscope observations in the examined areas of both the temporal cortex and

the caudo-putamen, with negligible exceptions, the excess water was found in astrocytes (i.e.

intracellular edema); within their cell bodies (Fig.VI. 7b) as well as in their perivascular end-feet

(Fig.VI.7a). The bulk of excess water was present around arterioles (Fig.VI 7c), less was found

next to venules or capillaries (Fig. VI. 7a and b). In most cases the water-filled astrocytic profiles

were surrounded by membranes with very thin extracellular space between them (Fig.VI. 7a and

b), indicating that the role of the latter was unimportant in storing the excess water. In cases

where relatively large volumes of excess water were present around a blood vessel, the

membranes between two water-filled end-feet of astrocytic processes were focally torn, and

partially or completely disintegrated (Fig. VI 7c), but the excess water did not invade the opened

extracellular space (Fig.VI. 7d). In several places, remnants of the disintegrated organelles of

swollen astrocytes self-assembled into regular (Fig. VI. 7e) or irregular myelin bodies (Fig. VI.

7f).

3. Discussion: The biexponential approach is considered to be a robust model which gives consistent ADCs and

volume fraction values in the brain tissue. It appears to be insensitive to inversion pulse (T1),

magnetization transfer, and extension of echo time (T2)85,87. Similar values of ADC and volume

fractions were recorded regardless of field strength53,85,88,73,139,140, diffusion time and

sequences73,87,88 ,140. A recent paper, however, suggests that ADCslow increases by extending the

diffusion time while ADCfast and the volume fractions are still insensitive to this94. The

predominance of the fast volume fraction (i.e. Pfast≈70–80%) is so consistent that it does not

change much, regardless of species (human brain73,87,140, mouse brain53,73, rat brain85) or tissue

types (brain53,73,85, muscle141, prostate142,143). These characteristics of the bi-exponential approach

may substantiate that it will have diagnostic value in the daily clinical practice73,139,140 similar to

the mono-exponential approach in the low b-value range. Despite the fact that the cause of ADC

changes in brain pathologies is still unclear and that the monoexponential fit was invented for

“free” diffusion in solutions, it became the gold standard for diagnosis e.g. in stroke144,145.

In our experiment, we induced an intracellular edema that resulted in unexpected volume

fractions. Contrary to stroke53,85, 139 , Pfast increased.

44

Fig. VI.7: Photoelectron micrograph.

Photoelectron micrograph showing water accumulation and myelin body formation: a: L marks the lumen of a venule, asterisks end-feet of astrocytes. A long thin arrow points to a pericyte, a short thin arrow to an endothelial cell, and a short thick arrow to the basement membrane. b: L marks the lumen of a capillary, N the nucleus of a perivascular astrocyte, P a protrusion of the astrocyte, and asterisks end-feet of astrocytes. c: L marks the lumen of an arteriole, an asterisks a large fluid-filled space originated from the fusion of several astrocytic end feet. The arrows mark torn ends of the membranes of neighboring astrocytic processes d: The asterisks mark two fluid-filled adjacent astrocytic end-feet with a torn section of the membranes between them. An arrow points to the extracellular space between the membranes of the two neighboring astrocytic endfeet. e: A regular myelin body. f.: An irregular myelin body.

45

The magnitude of ADCmono decrease from the baseline (−5.71±4.0%) is considerably smaller

than that detected in stroke (−40%146). However, there are substantial differences that can

explain the discrepancy. Based on the electron microscopic images (Fig. VI. 7), in intracellular

edema (i.e. water load) the perivascular astrocytes are swollen, while most of the neurons appear

to be intact. Cells can cope with the intracellular edema since the energy stores are not

compromised due to intact blood supply. There is a lack of tissue necrosis that could have

resulted in degradation of macromolecules and dysfunction of membrane pumps and channels.

Therefore, based on our experiments it can be deduced that the intracellular edema alone (i.e.

water shift from extracellular to intracellular space) results in volume fractions changes that are

different from the ones observed in stroke.

Our results are in apparent contradiction with in-vitro studies describing cell swelling (i.e. extra

to intracellular water shift) due to hypotonic medium or inhibition of Na-K ATPase96,97 or

application of NMDA147. The in-vitro results showed no change in ADCfast and ADCslow, while

Pfast was decreasing. Several limitations about the in-vitro studies should be mentioned: (i) The

measurements were carried out at room temperature. This has the potential to cause

conformational changes in proteins and significantly slow down enzymatic processes. (ii) A

well-documented fact from in-vivo studies is that a radical ADC decrease is almost immediate

(2–3min) if anoxia-ischemia takes place146. Therefore, the time required for the tissue

preparation would have been enough to cause ischemia. This notion is supported by the volume

fractions (Pfast was 53%) in the in-vitro measured rat hippocampus97 which highly resemble

volume fractions detected during in-vivo brain ischemia experiments53,139. Unfortunately, due to

the application of room temperature, the ADC values are not comparable. (iii) The in-vitro

application of hypotonic solutions to elicit intracellular edema does not model the in-vivo

circumstances since it appears that in-vivo mostly astrocytes are swollen while in-vitro all tissue

elements are presumed to swell (i.e. no blood flow, and all cells that come into contact with the

hypotonic solution swell).

Table VI. 2 summarizes results of biexponential diffusion measurements in different pathological

conditions. If we take into account only edematous tissue and in-vivo studies, it seems that the

intracellular and extracellular edema do have MR fingerprints, that would be of interest in the

daily clinical practice. It is clear from Table VI. 2, that ADCfast would have a diagnostic value,

since it decreases in intracellular edema (i.e. stroke, hypotonic conditions) and increases in

46

extracellular (e.g. vasogenic) edema. The latter is obvious also in muscle edema in which the

presumed extracellular water increase causes Pfast and ADCfast elevations141.

Table VI. 2:

Biexponential DWI parameters in different pathologies: Paper Species and Pathology ADC (f) change ADC (s) change Fraction (f) change Our results

Rat, hyponatremia Decrease Very slight decrease

Increase: From 76.56% to 81.2%

Brugières et al.139

Human, stroke. Slight increase 1 day after stroke.

Increase. Decrease: From 74.3% to 49.1%.

Schwacz et al.73

i) Mice, cold injury. (perifocal edema).

ii) Human, pritumoral edema.

i) Increase. ii) Increase.

i) No change.

ii) No change.

i) Increase: From 75% to 79%*

ii) Increase: From 63% to 89%.

Schwacz et al.53

i) Mice, stroke. ii) Mice, core cold injury.

i) Decrease. ii) Increase.

i) Decrease. ii) Decrease.

i) Decrease: From 79% to 57%.

ii) Decrease: From 78% to 67%

Niendorf et al.85

Rat (caudate nucleus), global ischemia.

Decrease. Decrease. Decrease: From 80% to 69%.

Maier et al.140

Human, peritumoral edema.

Increase. No change. Increase: From 69% to 90%.

Ababneh et al.141

Rat, extracellular muscle edema.

Increase. Increase. Increase: from 84% to 89%.

Buckley et al.97

Rat (hippocampal slice), Ouabain intoxication.

No change. No change. Decrease: From 53% to 44%.

Bui et al.147

Rat (hippocampal slice), NMDA intoxication.

No change. No change. Decrease: From 55% to 50%.

Shepherd et al.96

Human (perfused hippocampal slice) i) Hyponatremia. ii) Hypernatremia.

i) No change.

ii) No change.

i) No change. ii) Increase.

i) Decrease: From 71% to 65%.

ii) Increase: From 71% to 78%.

Biexponential DWI parameters in different pathologies. *(no statistical significance)

Another parameter that would have diagnostic value in differentiating the different conditions

and subtypes of edema (i.e. identifying components of free vs. bound water rather than

compartments of intracellular vs. extracellular edema) is the Pfast. It is hypothesized that the fast

component represents free, bulk water that would be the ideal candidate to be eliminated from

edematous tissue. Opposing this theory is the fact that the core region of cold injury shows a

47

different behavior (Table VI. 2): Pfast decreases despite the presence of vasogenic edema.

However, considering that the core of the cold injury contains disintegrated cells53, meaning that

necrosis and vasogenic edema are present at the same location. If necrosis-apoptosis dominate

over the vasogenic edema the net result will cause a marked decrease in Pfast. By expanding our

observation, and evaluating the penumbra region of the cold injury which presumably contains

living tissue with extracellular edema73 we find that in this location, according to expectation

ADCfast and Pfast are increasing.

48

VII. Limitations of our studies:

1. Limitations of the Benzamil studies:

The clinical implications of our observations are not apparent, although they appear to provide

new approaches to prevent or treat brain edema in hyponatremic patients. Selective, reversible

and non-toxic inhibition of brain sodium channels in the early phase of edema formation is likely

to meet the clinical requirements of brain edema treatment. For this therapeutic approach,

however, new sodium channel blockers that freely cross the BBB need to be developed, and the

timing, dosage and route of effective and safe drug administration need to be established. In this

regard it is to be considered that whereas ICV administration of Benzamil abolished the increase

of mean arterial pressure, heart rate, sympathetic outflow and vasopressin release in rat models

with NaCl-induced hypertension, intravenous infusion of Benzamil did not affect either of these

parameters99.

2. Limitations of the MRI studies:

The clinical scanner`s limitations manifest in the relative inflexibility of the software and

gradient strength.

In the biexponential DWI study a long echo and diffusion times were applied due to the

hardware restrictions of a clinical scanner. In our case, the ADC parameters were higher, while

the volume fractions are lower than those obtained by Niendorf et al.85 in the rat brain. However,

these differences may be explained by the following facts: (i) trace weighting was used in our

study vs. only one direction, (ii) in our study, data from the cortex are presented vs. caudate

nucleus data (i.e. more restrictions due to white matter fibers passing through the caudate

nucleus), (iii) the echo and diffusion times are much longer in our study. Regardless of these

limitations, our results are in overall very good agreement especially with human data measured

in brain cortex140,148. Other mathematical models (e.g. diffusional kurtosis imaging, inverse

Laplace transformation etc.) are also available for analysis. However, considering that only

limited number of papers deal with these alternative approaches, the application of alternative

fitting models was beyond the scope of the present edema study.

49

VIII. Future plans:

1. Exploring the effects of Benzamil on the hyponatremic brain edema by way of

biexponential diffusion analysis.

2. Using the accumulated data to test new compounds under development, like the AQP

modulators and drugs like steroids aimed at treating the different kinds of brain edema,

in the hope of finding new, better, and innovative ways to treat patients suffering from

these pathologies.

50

IX. Summary of the thesis:

Our observations appear to suggest that an early cellular sodium uptake through the

Benzamil-sensitive activated sodium channels may be intimately involved in the development of

brain edema in rats exposed to hyponatremic stress. The opposing effects of systemic

hyponatremia on the brain-specific AQP4 and ENaCs, provide an important defense mechanism

to act against excessive water accumulation and to protect brain volume.

ICV pretreatment with Benzamil has an immediate anti-edematous effect that lasts for at

least 2 hours on the cellular brain edema associated with acute hyponatremic stress in rats.

The changes in serial MR spectroscopy (water signal) give us yet another tool to continuously

follow the progress of brain edema evolution in-vivo. The in-vivo results confirm in-vitro

observations that Benzamil prevents water accumulation in hyponatremic rats.

These observations suggest that, in the future, selective non-toxic inhibition of brain

specific sodium channels by agents that can freely cross the BBB may provide a new therapeutic

approach to prevent and treat hyponatremia associated brain edema in the early stages of its

formation. In order to do this, the exact pharmacokinetics and the ideal timing of treatment with

such sodium channel blockers would have to be worked out, especially when considering the

clinical point of view, where the development or exact starting point of a brain edema are harder

to determine.

Our MRI experiments point to the fact that reliable small animal imaging as well as DWI

and MRS can be carried out on 3T clinical scanners.

The unexpected volume fraction changes that were detected in hypotonic edema appear to be

substantially different from those observed in stroke. This may suggest that contrary to general

opinion and radiology textbook data, the ADC decrease in stroke cannot be explained by mere

water shift from extra to intracellular space (i.e. intracellular edema). It is plausible that other

mechanisms such as decreased cytoplasmic streaming, membrane pump dysfunctions, changes in

physical state of water (i.e. sol–gel transitions, free-bound transitions) etc. are involved in the

ADC decrease seen in stroke.

Based on our results and the literature, the biexponential DWI analysis would have a clinical role

in brain edema diagnosis and follow up of treatment.

51

X. References:

1.) Klatzo I. Evolution of brain edema concepts. Acta Neurochir Suppl (Wien). (1994) 60:3-

6.

2.) Kimelberg HK. Current concepts of brain edema. Review of laboratory investigations.

J neurosurg. (1995) 83:1051-1059.

3.) Klatzo I. Presidental address. Neuropathological aspects of brain edema. J Neuropathol

Exp Neurol. (1967) 26(1):1-14.

4.) Fishman RA. Brain edema. N Engl J Med. (1975) 293:706-71.

5.) Papadopoluos MC, Binder DK, Verkman AS. Enhanced macromolecular diffusion in

brain extracellular space in mouse models of vasogenic edema measured by cortical

surface photobleaching. FASEB J. (2005) 19:425-427.

6.) Marmarou A. A review of progress in understanding the pathophysiology and treatment

of brain edema. Neurosurg Focus. (2007) 22(5):E1.

7.) Barron KD, Dentinger MP, Kimelberg HK, Nelson LR, Bourke RS, Keegan S, Mankes

R, Cragoe EJ Jr. Ultrastructural features of a brain injury model in cat. I. Vascular and

neuroglial changes and the prevention of astroglial swelling by a fluorenyl (aryloxy)

alkanoic acid derivative (L-644, 711). Acta Neuropathol. (1988) 75:295–307.

8.) Klatzo I. Pathophysiological aspects of brain edema. Acta Neuropathol. (1987) 72:236–

239.

9.) Rash JE, Yasumura T, Hudson CS, Agre P, Nielsen S. Direct immunogold labeling of

aquaporin-4 in square arrays of astrocyte and ependymocyte plasma membranes in rat

brain and spinal cord. Proc Natl Acad Sci U S A. (1998) 95(20):11981-6.

10.) Rash JE, Yasumura T. Direct immunogold labeling of connexins and aquaporin-4 in

freeze-fracture replicas of liver, brain, and spinal cord: factors limiting quantitative

analysis. Cell Tissue Res. (1999) 296(2):307-21.

11.) Ayata C, Ropper AH. Ischaemic brain oedema. J Clin Neurosci. (2002) 9:113–124.

12.) Kuroiwa T, Miyasaka N, Fengyo Z, Yamada I, Nakane M, Nagaoka T, Tamura A, Ohno

K. Experimental ischemic brain edema: morphological and magnetic resonance imaging

findings. Neurosurg Focus. (2007) 22(5):E11.

52

13.) Liang D, Bhatta S, Gerzanich V, Simard JM. Cytotoxic edema: mechanisms of

pathological cell swelling. Neurosurg Focus. (2007) 22(5):E2.

14.) Upadhyay A, Jaber BL, Madias NE. Epidemiology of hyponatremia. Semin Nephrol.

(2009) 29(3):227-238

15.) Arieff AI, Llach F, Massry SG. Neurological manifestations and morbidity of

hyponatremia: correlation with brain water and electrolytes. Medicine (Baltimore).

(1976) 55(2):121-129.

16.) Ayus JC, Armstrong D, Arieff AI. Hyponatremia with hypoxia: effects on brain

adaptation, perfusion, and histology in rodents. Kidney Int. (2006) 69(8):1319-1325.

17.) Melton JE, Nattie EE. Brain and CSF water and ions during dilutional and isoosmotic

hyponatremia in the rat. Am J Physiol. (1983) 244(5)R724-R732.

18.) Olson JE, Mishler L, Dimilich RV. Brain water content, brain blood volume, blood

chemistry, and pathology in a model of cerebral edema. Ann Emerg Med. (1990)

19(10)1113-1121.

19.) Olson JE, Katz-Stein A, Reo NV, Jolesz FA. Evaluation of acute brain edema using

quantitative magnetic resonance inmaging: effects of pretreatment with dexamethasone.

Magn Reson Med. (1992) 24(1):64-74.

20.) Verbalis JG, Goldsmith SR, Greenberg A, Schrier RW, Sterns RH. Hyponatremia

treatment guidlines 2007: Expert panel recommendations. Am J Med. (2007) 120(11

suppl 1)S1-S21.

21.) Agre P. Nobel Lecture. Aquaporin water channels. Biosci Rep. (2004) 24(3):127-63.

22.) Papadopoulos MC, Manley GT, Krishna S, Verkman AS. Aquaporin-4 facilitates

reabsorption of excess fluid in vasogenic brain edema. FASEB J. (2004) 18(11):1291-3.

23.) Manley GT, Fujimura M, Ma T, Noshita N, Filiz F, Bollen AW, Chan P, Verkman AS.

Aquaporin-4 deletion in mice reduces brain edema after acute water intoxication and

ischemic stroke. Nat Med. (2000) 6(2):159-63.

24.) Papadopoulos MC, Verkman AS. Aquaporin-4 gene disruption in mice reduces brain

swelling and mortality in pneumococcal meningitis. J Biol Chem. (2005) 280(14):13906-

12.

25.) Prough DS, Lang J. Therapy of patients with head injuries: key parameters for

management. J Trauma. (1997) 42 (5 Suppl):S10-18.

53

26.) Roberts I, Schierhout G, Alderson P. Absence of evidence for the effectiveness of five

interventions routinely used in the intensive care management of severe head injury: a

systematic review. J Neurol Neurosurg Psychiatry. (1998) 65(5):729-33.

27.) Amiry-Moghaddam M, Otsuka T, Hurn PD, Traystman RJ, Haug FM, Froehner SC,

Adams ME, Neely JD, Agre P, Ottersen OP, Bhardwaj A. An α-syntrophin dependent

pool of Aqp-4 in astroglial end-feet confers bidirectional water flow between blood and

brain. Proc Natl Acad Sci USA. (2003)100:2106–2111

28.) Amiry-Moghaddam M, Williamson A, Palomba M, Eid T, de Lanerolle NC, Nagelhus

EA, Adams ME, Froehner SC, Agre P, Ottersen OP. Delayed K+ clearance associated

with aquaporin-4 mislocalization: phenotypic defects in brains of α-syntrophin-null mice.

Proc Natl Acad Sci USA. (2003) 100:13615–13620

29.) Hasegawa H, Ma T, Skach W, Matthay MA, Verkman AS. Molecular cloning of a

mercurial-insensitive water channel expressed in selected water-transporting tissues. J

Biol Chem. (1994) 269:5497–5500

30.) Yool AJ, Brown EA, Flynn GA. Roles for novel pharmacological blockers of aquaporins

in the treatment of brain oedema and cancer. Clin Exp Pharmacol Physiol. (2010)

37:403–409

31.) Okada Y. Volume-expansion sensing outward-rectifier Cl- channel: fresh start to the

molecular identity and volume sensor. Am J Physiol. (1997) 273:C755–C789

32.) Pasantes-Morales H, Morales Mulia S. Influence of calcium on regulatory volume

decrease: role of potassium channels. Nephron. (2000) 86:414–427

33.) Pasantes-Morales H, Lezama RA, Ramos-Mandujanu G, Tuz KL. Mechanisms of cell

volume regulation in hypo-osmolality. Am J Med. (2006) 119(7a):S4–S11.

34.) Kellenberger S, Schild L. Epithelial sodium channel/degenerin family of ion channels: a

variety of functions for a shared structure. Physiol Rev. (2002) 82(3):735-67.

35.) Canessa CM, Horisberger JD, Rossier BC. Epithelial sodium channel related to proteins

involved in neurodegeneration. Nature. (1993) 361: 467–470.

36.) Huang M, Chalfie M. Gene interactions affecting mechanosensory transduction in

Caenorhabditis elegans. Nature. (1994) 367: 467–470.

54

37.) Grunder S, Muller A, Ruppersberg JP. Developmental and cellular expression pattern of

epithelial sodium channel alpha beta and gamma subunits in the inner ear of the rat. Eur J

Neurosci. (2001) 13: 641–648.

38.) Couloigner V, Fay M, Djelidi S, Farman N, Escoubet B, Runembert I, Sterkers O,

Friedlander G, Ferrary E. Location and function of the epithelial Na channel in the

cochlea. Am J Physiol Renal Physiol. (2001) 280: F214–F222.

39.) Benson CJ, Xie J, Wemmie JA, Price MP, Henss JM, Welsh MJ, Snyder PM.

Heteromultimers of DEG/ENaC subunits form H+-gated channels in mouse sensory

neurons. Proc Natl Acad Sci USA. (2002) 99: 2338–2343.

40.) Price MP, McIlwrath SL, Xie J, Cheng C, Qiao J, Tarr DE, Sluka KA, Brennan TJ,

Lewin GR, Welsh MJ. The DRASIC cation channel contributes to the detection of

cutaneous touch and acid stimuli in mice. Neuron. (2001) 32: 1071–1083.

41.) Chen CC, England S, Akopian AN, Wood JN. A sensory neuron-specific, proton-gated

ion channel. Proc Natl Acad Sci USA. (1998) 95: 10240–10245.

42.) Bassilana F, Champigny G, Waldmann R, de Weille JR, Heurteaux C, Lazdunski M. The

acid-sensitive ionic channel subunit ASIC and the mammalian degenerin MDEG form a

heteromultimeric H-gated Na channel with novel properties. J Biol Chem. (1997) 272:

28819–28822.

43.) Wang HW, Amin MS, El-Shahat E, Huang BS, Tuana BS, Leenen FH. Effects of central

sodium on epithelial sodium channels in rat brain. Am J Physiol Regul Integr Comp

Physiol. (2010) 299(1) R222-33.

44.) Garty H, Palmer LG. Epithelial sodium channels. Function, structure, and regulation.

Physiol Rev. (1997) 77: 359–396.

45.) Canessa CM, Schild L, Buell G, Thorens B, Gautschi I, Horisberger JD, Rossier BC.

Amiloride-sensitive epithelial Na+ channel is made of three homologous subunits.

Nature. (1994) 367: 463–467.

46.) Marunaka Y. Physiological requirement of aldosterone action on the Na+/K+ pump in

ADH-stimulated Na+ absorption in renal epithelium. Jpn J Physiol. (1996) 46(4):357-61.

47.) Niisato N, Marunaka Y. Activation of the Na+-K+ pump by hyposmolality through

tyrosine kinase-dependent Cl- conductance in Xenopus renal epithelial A6 cells. J

Physiol. (1999) 518:417-32.

55

48.) Marunaka Y, Hagiwara N, Tohda H. Insulin activates single amiloride-blockable Na

channels in a distal nephron cell line (A6). Am J Physiol. (1992) 263(3 Pt 2):F392-400.

49.) De Smet P, Simaels J, Declercq PE, Van Driessche W. Regulatory volume decrease in

cultured kidney cells (A6): role of amino acids. J Gen Physiol. (1995) 106(3):525-42.

50.) Niisato N, Van Driessche W, Liu M, Marunaka Y. Involvement of protein tyrosine kinase

in osmoregulation of Na+ transport and membrane capacitance in renal A6 cells. J Membr

Biol. (2000) 175:63-77

51.) Taruno A, Niisato N, Marunaka Y. Hypotonicity stimulates renal epithelial sodium

transport by activating JNK via receptor tyrosine kinase. Am J Physiol. (2007) 293:F128-

F138

52.) Zhi-Pei Liang, Paul C. Lauterbur. Principles of Magnetic Resonance Imaging: A Signal

Processing Perspective. (2000).

53.) Schwarcz A, Bogner P, Meric P, Correze JL, Berente Z, Pal J, Gallyas F, Doczi T, Gillet

B, Beloeil JC. The existence of biexponential signal decay in magnetic resonance

diffusion-weighted imaging appears to be independent of compartmentalization. Magn

Reson Med. (2004) 51(2):278-285.

54.) Beuf O, Jaillon F, Saint-Jalmes H. Small-animal MRI: signal-to-noise ratio comparison at

7 and 1.5 T with multiple-animal acquisition strategies. Magma. (2006) 19(4):202-208.

55.) Brockmann MA, Ulmer S, Leppert J, Nadrowitz R, Wuestenberg R, Nolte I, Petersen D,

Groden C, Giese A, Gottschalk S. Analysis of mouse brain using a clinical 1.5 T scanner

and a standard small loop surface coil. Brain Res. (2006) 1068(1):138-142.

56.) Chen F, De Keyzer F, Wang H, Vandecaveye V, Landuyt W, Bosmans H, Hermans R,

Marchal G, Ni Y. Diffusion weighted imaging in small rodents using clinical MRI

scanners. Methods. (2007) 43(1):12-20.

57.) Chen F, Suzuki Y, Nagai N, Peeters R, Coenegrachts K, Coudyzer W, Marchal G, Ni Y.

Visualization of stroke with clinical MR imagers in rats: a feasibility study. Radiology.

(2004) 233(3):905-911.

58.) Guzman R, Lovblad KO, Meyer M, Spenger C, Schroth G, Widmer HR. Imaging the rat

brain on a 1.5 T clinical MR-scanner. J Neurosci Methods. (2000) 97(1):77-85.

56

59.) Lee FK FM, Antonio GE, Yeung DK, Chan ET, Zhang LH, Yew DT, Ahuja AT.

Diffusion tensor imaging (DTI) of rodent brains in vivo using a 1.5T clinical MR

scanner. J Magn Reson Imaging. (2006) 23(5):747-751.

60.) Linn J, Schwarz F, Schichor C, Wiesmann M. Cranial MRI of small rodents using a

clinical MR scanner. Methods. (2007) 43(1):2-11.

61.) Thorsen F, Ersland L, Nordli H, Enger PO, Huszthy PC, Lundervold A, Standnes T,

Bjerkvig R, Lund-Johansen M. Imaging of experimental rat gliomas using a clinical MR

scanner. J Neurooncol. (2003) 63(3):225-231.

62.) Ulmer S, Reeh M, Krause J, Herdegen T, Heldt-Feindt J, Jansen O, Rohr A. Dynamic

contrast-enhanced susceptibility-weighted perfusion MRI (DSC-MRI) in a glioma model

of the rat brain using a conventional receive-only surface coil with a inner diameter of 47

mm at a clinical 1.5 T scanner. J Neurosci Methods. (2008) 172(2):168-172.

63.) Yang YM, Feng X, Yao ZW, Tang WJ, Liu HQ, Zhang L. Magnetic resonance

angiography of carotid and cerebral arterial occlusion in rats using a clinical scanner. J

Neurosci Methods. (2008) 167(2):176-183.

64.) Sandner B, Pillai DR, Heidemann RM, Schuierer G, Mueller MF, Bogdahn U,

Schlachetzki F, Weidner N. In vivo high-resolution imaging of the injured rat spinal cord

using a 3.0T clinical MR scanner. J Magn Reson Imaging. (2009) 29(3):725-30.

65.) Zhang S, He H, Lu W, Xu Q, Zhou B, Tang X. Tracking intrahepatically transplanted

islets labeled with Feridex-polyethyleneimine complex using a clinical 3.0-T magnetic

resonance imaging scanner. Pancreas. (2009) 38(3):293-302.

66.) Daire JL, Jacob JP, Hyacinthe JN, Croisille P, Montet-Abou K, Richter S, Botsikas D,

Lepetit-Coiffé M, Morel D, Vallée JP. Cine and tagged cardiovascular magnetic

resonance imaging in normal rat at 1.5 T: a rest and stress study. J Cardiovasc Magn

Reson. (2008) 10(1):48.

67.) Chang NK, Jeong YY, Park JS, Jeong HS, Jang S, Jang MJ, Lee JH, Shin SS, Yoon W,

Chung TW, Kang HK. Tracking of neural stem cells in rats with

intracerebral hemorrhage by the use of 3T MRI. Korean J Radiol. (2008) 9(3):196-204.

68.) Pinkernelle JG, Stelter L, Hamm B, Teichgräber U. Small animal MRI: clinical MRI as

an interface to basic biomedical research. Rofo. (2008)180(6):505-13.

57

69.) Pfefferbaum A, Adalsteinsson E, Sullivan EV. In vivo structural imaging of the rat brain

with a 3-T clinical human scanner. J Magn Reson Imaging. (2004) 20(5):779-785.

70.) Sevick RJ, Kanda F, Mintorovich J, Arieff AI, Kucharczyk J, Tsuruda JS, Norman D,

Moseley ME. Cytotoxic brain edema: Assessment with diffusion weighted MR imaging.

Radiology. (1992) 185:687-690.

71.) Vajda Z, Pedersen M, Dóczi T, Sulyok E, Stødkilde-Jørgensen H, Frøkiaer J, Nielsen S.

Effects of centrally administered arginine vasopressin and atrial natriuretic peptide on the

development of brain edema in hyponatremic rats. Neurosurgery. (2001). 49(3):697-704.

72.) Slivka AP, Murphy EJ. High-dose methylprednisolone treatment in experimental focal

cerebral ischemia. Exp Neurol. (2001) 167(1):166-72.

73.) Schwarcz A, Ursprung Z, Berente Z, Bogner P, Kotek G, Meric P, Gillet B, Beloeil JC,

Dóczi T. In vivo brain edema classification: New insight offered by large b-value

diffusion-weighted MR imaging. J Magn Reson Imaging. (2007) 25(1):26-31.

74.) Le Bihan D, Delannoy J, Levin RL. Temperature mapping with MR imaging of

molecular diffusion: application to hyperthermia. Radiology. (1989). 171:853-857.

75.) Jiang C, Wang J, Li X, Liu C, Chen N, Hao Y. Progesterone exerts neuroprotective

effects by inhibiting inflammatory response after stroke. Inflamm Res. (2009) 58(9):619-

24.

76.) Le Bihan D, Moonen CT, Van Zijl PC, Pekar J, DesPres D. Measuring random

microscopic motion of water in tissue with MR imaging: a cat brain study. J Comput

Assist Tomogr. (1991) 15(1):19-25.

77.) Zhong J, Petroff OA, Prichard JW, Gore JC. Changes in water diffusion and relaxation

properties of rat cerebrum during status epilepticus. Magn Reson Med. (1993) 30:241-

246.

78.) Neumann-Haefelin T, Kastrup A, de Crespigny A, Yenari MA, Ringer T, Sun GH,

Moseley ME. Serial MRI after transient focal cerebral ischemia in rats: Dynamics of

tissue injury, blood brain barrier damage, and edema formation. Stroke. (2000) 31:1965-

1972.

79.) Brazo P, Marmarou A, Fatouros P, Hyasaki K, Corwin F. Contribution of vasogenic and

cellular edema to traumatic brain swelling measured by diffusion-weighted imaging. J

Neurosurg. (1997) 87:900-907.

58

80.) Szafer A, Zhong J, Gore JC. Theoretical model for water diffusion in tissues. Magn

Reson Med. (1995) 33(5):697-712.

81.) Ebisu T, Naruse S, Horikawa Y, Ueda S, Tanaka C, Uto M, Umeda M, Higuchi T.

Discrimination between different types of white matter edema with diffusion-weighted

MR imaging. J Magn Reson Imaging. (1993) 3(6):863-8.

82.) Sykova E, Svoboda J, Polak J, Chvatal A. Extracellular volume fraction and diffusion

characteristics during progressive ischemia and terminal anoxia in the spinal cord of the

rat. J Cereb Blood Flow Metab. (1994) 14(2):301-11.

83.) Moseley ME, Cohen Y, Mintorovitch J, Chileuitt L, Shimizu H, Kucharczyk J, Wendland

MF, Weinstein PR. Early detection of regional cerebral ischemia in cats: Comparison of

diffusion-and T2-weighted MRI and spectroscopy. Magn Reson Med. (1990) 14:330-346.

84.) Mintorovitch J, Moseley ME, Chileuitt L, Shimizu H, Cohen Y, Weinstein PR.

Comparison of diffusion-and T2-weighted MRI for the early detection of cerebral

ischemia and reperfusion in rats. Magn Reson Med. (1991) 18:39-50.

85.) Niendorf T, Dijkhuizen RM, Norris DG, van Lookeren Campagne M, Nicolay K.

Biexponential diffusion attenuation in various states of brain tissue: implications for

diffusion-weighted imaging. Magn Reson Med. (1996) 36(6):847–57.

86.) Mulkern RV, Haker SJ, Maier SE. On high b diffusion imaging in the human brain:

ruminations and experimental insights. Magn Reson Imaging. (2009) 27(8):1151–62.

87.) Clark CA, Le Bihan D. Water diffusion compartmentation and anisotropy at high b

values in the human brain. Magn Reson Med. (2000) 44(6):852–9.

88.) Grant SC, Buckley DL, Gibbs S,Webb AG, Blackband SJ. MR microscopy of multi

component diffusion in single neurons. Magn ResonMed. (2001) 46(6):1107–12.

89.) Branco G. An alternative explanation of the origin of the signal in diffusion weighted

MRI. Neuroradiology. (2000) 42(2):96–8.

90.) Le Bihan D. The wet mind: water and functional neuroimaging. Phys Med Biol. (2007)

52(7):R57–90.

91.) Kiselev VG, Il’yasov KA. Is the “biexponential diffusion” biexponential? Magn Reson

Med. (2007) 57(3):464–9.

59

92.) Ronen I, Moeller S, Ugurbil K, Kim DS. Analysis of the distribution of diffusion

coefficients in cat brain at 9.4 T using the inverse Laplace transformation. Magn Reson

Imaging. (2006) 24(1):61–8.

93.) Minati L, Aquino D, Rampoldi S, Papa S, Grisoli M, Bruzzone MG, Maccagnano E.

Biexponential and diffusional kurtosis imaging, and generalized diffusion-tensor imaging

(GDTI) with rank 4 tensors: a study in a group of healthy subjects. MAGMA. (2007)

20(5–6):241–53.

94.) Minati L, Zucca I, Carcassola G, OcchipintiM, Spreafico R, BruzzoneMG. Effects of

diffusion-sensitizing gradient timings on the exponential, biexponential, and diffusional

kurtosis model parameters: in-vivo measurements in the rat thalamus. MAGMA. (2010)

23(2):115–21.

95.) Bennett KM, Schmainda KM, Bennett RT, Rowe DB, Lu H, Hyde JS. Characterization

of continuously distributed cortical water diffusion rates with a stretched-exponential

model. Magn Reson Med. (2003) 50(4):727–34.

96.) Shepherd TM, Wirth 3rd ED, Thelwall PE, Chen HX, Roper SN, Blackband SJ. Water

diffusion measurements in perfused human hippocampal slices undergoing tonicity

changes. Magn Reson Med. (2003) 49(5):856–63.

97.) Buckley DL, Bui JD, Phillips MI, Zelles T, Inglis BA, Plant HD, Blackband SJ. The

effect of ouabain on water diffusion in the rat hippocampal slice measured by high

resolution NMR imaging. Magn Reson Med. (1999) 41(1):137–42.

98.) Rink PA. Magnetic resonance in medicine. The basic textbook of the European magnetic

resonance forum. Fifth revised edition. (2003).

99.) Nishimura M, Ohtsuka K, Nanbu A, Takahashi H, Yoshimura M. Benzamil blockade of

brain Na+ channels averts Na+-induced hypertension in rats. Am J Physiol. (1998) 274(3

Pt 2):R635-44.

100.) Ecelbarger CA, Kim G-H, Terris J, Masilamani S, Mitchell C, Reyes I, Verbalis JG,

Knepper MA. Vasopressin-mediated regulation of epithelial sodium channel abundance

in rat kidney. Am J Physiol Renal Physiol. (2000) 279:F46–F53.

101.) Chen L, Williams SK, Schaffer JA. Differences in synergistic actions of vasopressin and

deoxycorticosterone in rat and rabbit CCD. Am J Physiol Renal Fluid Electrolyte

Physiol. (1990) 259:F147–F156.

60

102.) Frind G, Burg MB. Effect of vasopressin on sodium transport in renal cortical collecting

duct tubules. Kidney Int. (1972) 1:224–231.

103.) Reif MC, Troutman SL, Schafer JA. Sodium transport by rat cortical collecting tubule.

Effects of vasopressin and desoxycorticosterone. J Clin Invest. (1986) 77:1291–1298.

104.) Tomita K, Pisano JJ, Knepper MA. Control of sodium and potassim transport in the

cortical collecting duct of the rat: Effect of bradykinin, vasopressin, and

deoxycorticosterone. J Clin Invest. (1985) 76:132–136.

105.) Canessa CM, Schaffer JA. AVP stimulates Na+ transport in primary cultures of rabbit

cortical collecting duct cells. Am J Physiol Renal Fluid Electorlyte Physiol. (1992)

262:F454–F461.

106.) Djelidi S, Fay M, Cluzenaud F, Escoubet B, Eugene E, Capurro C, Bovalet JP, Farman

N. Blot-Chabaud M. Transcriptional regulation of sodium transport by vasopressin in

renal cells. J Biol Chem. (1997) 272(52):32919–32924.

107.) Bindels RJM, Schaffer JA, Reif MC. Stimulation of sodium transport by aldosterone and

arginine vasotocin in A6 cells. Biochim Biophys Acta. (1988) 972(3):320–330.

108.) Wong SM, Chase HS. Effect of vasopressin on intracellular Ca++ and Na+ transport in

cultured toad bladder cells. Am J Physiol Fluid Electrolyte Physiol. (1988) 255:F1015–

F1024.

109.) Nakhoul NL, Hering-Smith KS, Gambala CT, Hamm LL. Regulation of sodium transport

in M-1 cells. Am J Physiol Renal Physiol. (1998) 275:F998–F1007.

110.) Kleyman TR, Ernst SA, Coupaye-Gerard B. Arginine vasopressin and forskolin regulate

apical cell surface expression of epithelial Na+ channels in A6 cells. Am J Physiol Renal

Fluid Electrolyte Physiol. (1994) 266:F506–F511.

111.) Prat AG, Ausiello DA, Cantiello HF. Vasopressin and protein kinase A activate G protein

-sensitive epithelial Na+ channels. Am J Physiol Cell Physiol. (1993) 265:C218–C223.

112.) Saito M, Tamara A, Sugimoto T. 1-Desamino-8-D-arginine vasopressin (DDAVP) as an

agonist on V1b vasopressin receptor. Biochem Pharmacol. (1997) 53:1711–1717.

113.) Hernando F, Schoots O, Lolait SJ, Burbach JP. Immunohistochemical localization of the

vasopressin V1b receptor in the rat brain and pituitary gland: Anatomical support for its

involvement in the central effect of vasopressin. Endocrinology. (2001) 142:1659–1668.

61

114.) Lolait SJ, O’Carroll A-M, Mahan LC, Felder CC, Button DC, Young WC, Mezey E,

Brownstein MJ. Extrapituitary expression of rat V1b vasopressin receptor gene. Proc

Natl Acad Sci USA. (1995) 92:6783–6787.

115.) Noda M. Hydromineral neuroendocrinology: mechanism of sensing sodium levels in the

mammalian brain. Exp Physiol. (2007) 92:513–522.

116.) Nishimura M, Ohtsuka K, Iwai N, Takahashi H, Yoshimura M. Regulation of brain renin-

angiotensin system by Benzamil-blockable sodium channels. Am J Physiol. (1999)

276:R1416-24.

117.) DePasquale M, Patlak CS, Cserr HF. Brain ion and volume regulation during acute

hyponatremia in Brattleboro rats. Am J Physiol. (1989) 256: F1059-F1066.

118.) Sarfaraz D, Fraser CL. Effects of arginine vasopressin on cell volume regulation in brain

astrocytes in culture. Am J Physiol Endocrinol Metab. (1999) 276:E596–E601

119.) Kanda F, Arieff AI. Vasopressin inhibits calcium-coupled sodium efflux system in rat

brain. Am J Physiol. (1994) 266: R1169-R1173.

120.) Han Z, Wax MB, Patil RV. Regulation of aquaporin-4 water channels by phorbol ester-

dependent protein phosphorylation. J Biol Chem. (1998) 273:6001–6004.

121.) Vajda Z, Promeneur D, Dóczi T, Sulyok E, Frokiaer J, Ottersen OP, Nielsen S. Increased

aquaporin-4 immunoreactivity in rat brain in response to systemic hyponatraemia.

Biochem Biophys Res Commun. (2000) 270:495–503.

122.) Niisato N, Post M, VanDriessche W, Marunaka Y. Cell swelling activates stress-

activated protein kinases, p38MAPK Kinase and JNK, in renal epithelial A6 cells.

Biochem Biophys Res Common. (1999) 266:547–550.

123.) Mayer D, Zahr NM, Adalsteinsson E, Rutt B, Sullivan EV, Pfefferbaum A. In vivo fiber

tracking in the rat brain on a clinical 3 T MRI system using a high strength insert gradient

coil. Neuroimage. (2007) 35(3):1077—85.

124.) Adalsteinsson E, Hurd RE, Mayer D, Sailasuta N, Sullivan EV, Pfefferbaum A. In vivo

2D J-resolved magnetic resonance spectroscopy of rat brain with a 3 T clinical human

scanner. Neuroimage. (2004) 22(1):381—6.

125.) Mayer D, Zahr NM, Sullivan EV, Pfefferbaum A. In vivo metabolite differences between

the basal ganglia and cerebellum of the rat brain detected with proton MRS at 3 T.

Psychiatry Res. (2007) 154(3):267—73.

62

126.) Natt O, Watanabe T, Boretius S, Radulovic J, Frahm J, Michaelis T. High-resolution 3D

MRI of mouse brain reveals small cerebral structures in vivo. J Neurosci Methods. (2002)

120(2):203—9.

127.) Chen F, Suzuki Y, Nagai N, Jin L, Yu J, Wang H, Marchal G, Ni Y. Rodent stroke

induced by photochemical occlusion of proximal middle cerebral artery: evolution

monitored with MR imaging and histopathology. Eur J Radiol. (2007) 63(1):68—75.

128.) Dong Z, Dreher W, Leibfritz D. Toward quantitative short echo-time in vivo proton MR

spectroscopy without water suppression. Magn Reson Med. (2006) 55(6):1441—6.

129.) Pfeuffer JTI, Provencher SW, Gruetter R. Toward an in vivo neurochemical profile:

quantification of 18 metabolites in short-echo-time H NMR spectra of the rat brain. J

Magn Reson. (1999) 141(1):104—20.

130.) Schwarcz A, Natt O, Watanabe T, Boretius S, Frahm J, Michaelis T. Localized proton

MRS of cerebral metabolite profiles in different mouse strains. Magn Reson Med. (2003)

49(5):822—7.

131.) Bize V, Horisberger JD. Sodium self-inhibition of human epithelial sodium channel:

selectivity and affinity of the extracellular sodium sensing site. Am J Physiol Renal

Physiol. (2007) 293(4): F1137–F1146.

132.) Crowe WE, Ehrenfeld J, Brochiero E, Wills NK. Apical membrane sodium and chloride

entry during osmotic swelling of renal (A6) epithelial cells. J Membr Biol. (1995)

144:81–91.

133.) Wills NK, Millinoff LP, Crowe WE. Na+ channel activity in cultured renal (A6)

epithelium: regulation by solution osmolarity. J Membr Biol. (1991) 121:79–90.

134.) Miyazaki H, Shiozaki A, Niisato N, Marunaka Y. Physiological significance of

hypotonicity–induced regulatory volume decrease: reduction in intracellular Cl-

concentration acting as an intracellular signaling. Am J Physiol Renal Physiol. (2007)

292:F1411–F1417.

135.) Niisato N, Eaton BC, Marunaka Y. Involvement of cytosolic Cl- in osmoregulation of α-

ENaC gene expression. Am J Physiol Renal Physiol. (2004) 287:F932–F939.

136.) Liu X, Nakayama S, Amiry-Moghaddam M, Ottersen OP, Bhardwaj A. Arginine-

vasopressin V1 but not V2 receptor antagonism modulates infarct volume, brain water

63

content and aquaporin-4 expression following experimental stroke. Neurocrit Care.

(2010) 12:124–131.

137.) Niermann H, Amiry-Moghaddam M, Holthoff K, Witte OW, Ottersen OP. A novel role

of vasopressin in the brain: modulation of activity-dependent water flux in the neocortex.

J Neurosci. (2001) 21:3045–3051.

138.) Kovács B, Bukovics P, Gallyas F. Morphological effects of transcardially perfused SDS

on the rat brain. Biol Cell. (2007) 99(8):425-32.

139.) Brugières P, Thomas P, Maraval A, Hosseini H, Combes C, Chafiq A, Ruel L, Breil S,

Peschanski M, Gaston A. Water diffusion compartmentation at high b values in ischemic

human brain. AJNR Am J Neuroradiol. (2004) 25(5):692–8.

140.) Maier SE, Bogner P, Bajzik G, Mamata H, Mamata Y, Repa I, Jolesz FA, Mulkern RV.

Normal brain and brain tumor: multicomponent apparent diffusion coefficient line scan

imaging. Radiology. (2001) 219(3):842–9.

141.) Ababneh Z, Beloeil H, Berde CB, Gambarota G, Maier SE, Mulkern RV. Biexponential

parametrization of diffusion and T2 relaxation decay curves in a rat muscle edema model:

decay curve components and water compartments. Magn Reson Med. (2005) 54(3):524–

31.

142.) Shinmoto H, Oshio K, Tanimoto A, Higuchi N, Okuda S, Kuribayashi S, Mulkern RV.

Biexponential apparent diffusion coefficients in prostate cancer. Magn Reson Imaging.

(2009) 27(3):355–9.

143.) Mulkern RV, Barnes AS, Haker SJ, Hung YP, Rybicki FJ, Maier SE, Tempany CM.

Biexponential characterization of prostate tissue water diffusion decay curves over an

extended b-factor range. Magn Reson Imaging. (2006) 24(5):563–8.

144.) Roberts TP, Rowley HA. Diffusion weighted magnetic resonance imaging in stroke. Eur

J Radiol. (2003) 45(3):185–94.

145.) Albers GW. Diffusion-weighted MRI For evaluation of acute stroke. Neurology. (1998)

51(Suppl. 3):S47–9.

146.) Davis D, Ulatowski J, Eleff S, et al. Rapid monitoring of changes in water diffusion

coefficients during reversible ischemia in cat and rat brain. Magn Reson Med. (1994)

31(4):454–60.

64

147.) Bui JD, Buckley DL, Phillips MI, Blackband SJ. Nuclear magnetic resonance imaging

measurements of water diffusion in the perfused hippocampal slice during N-methyl-d-

aspartate-induced excitotoxicity. Neuroscience. (1999) 93(2):487–90.

148.) Maier SE, Mulkern RV. Biexponential analysis of diffusion-related signal decay in

normal human cortical and deep gray matter. Magn Reson Imaging. (2008) 26(7):897–

904.

65

XI. Bibliography:

1. Articles grounding the thesis:

• Sulyok E, Pál J, Vajda Z, Steier R, Dóczi T. Benzamil prevents brain water accumulation

in hyponatraemic rats. Acta Neurochir (Wien). (2009) 151(9):1121-5.

• Aradi M, Steier R, Bukovics P, Szalay C, Perlaki G, Orsi G, Pál J, Janszky J, Dóczi T,

Schwarcz A. Quantitative proton MRI and MRS of the rat brain with a 3 T clinical

scanner. [published online ahead of print Mar 22 2010]. J Neuroradiol (2010).

• Steier R, Aradi M, Pál J, Bukovics P, Perlaki G, Orsi G, Janszky J, Schwarcz A, Sulyok

E, Dóczi T. The influence of Benzamil hydrochloride on the evolution of hyponatremic

brain edema as assessed by in vivo MRI study in rats. [Epub ahead of print 2011 Mar 29]

Acta Neurochir (Wien). (2011).

• Steier R, Aradi M, Pál J, Perlaki G, Orsi G, Bogner P, Galyas F, Bukovics P, Janszky J,

Dóczi T, Schwarcz A. A biexponential DWI study in rat brain intracellular edema. [Epub

ahead of print 2011 Apr 15] Eur J Radiol. 2011).

2. Presentations and posters grounding the thesis:

• Oral scientific progress reports at semiyearly meetings of the EU FP-6

"Aquaglyceroporins" to the rest of the consortium members. (Steier R, Aradi M, Pál J,

Bukovics P, Perlaki G, Orsi G, Schwarcz A, Sulyok E, Dóczi T.)

• Oral contribution at the Neuroimaging Workshop in Szeged Hungary. March 2010:

Effects of Benzamil in rat brain edema investigated by quantitative MR experiments.

(Steier R, Aradi M, Pál J, Bukovics P, Perlaki G, Orsi G, Janszky J, Schwarcz A, Sulyok

E, Dóczi T.)

• Oral presentation at the 5th Pannonian Symposium on Central Nervous System Injury in

Pecs Hungary. May 2010: Effects of Benzamil in rat brain edema investigated by

quantitative MR experiments. (Steier R, Aradi M, Pál J, Perlaki G, Orsi G, Bogner P,

Galyas F, Bukovics P, Janszky J, Dóczi T, Schwarcz A.)