Aging and N-methyl-D-aspartate receptors

MISSING IMAGE

Material Information

Title:
Aging and N-methyl-D-aspartate receptors
Physical Description:
x, 136 leaves : ill. ; 29 cm.
Language:
English
Creator:
Mitchell, Josephine Jean, 1964-
Publication Date:

Subjects

Subjects / Keywords:
Research   ( mesh )
Receptors, N-Methyl-D-Aspartate -- ultrastructure   ( mesh )
Receptors, N-Methyl-D-Aspartate -- physiology   ( mesh )
Aging   ( mesh )
Cell Aging   ( mesh )
Neuronal Plasticity   ( mesh )
Brain -- physiology   ( mesh )
Brain Chemistry   ( mesh )
Glutamic Acid -- physiology   ( mesh )
Age Factors   ( mesh )
Rats, Inbred F344   ( mesh )
Department of Neuroscience thesis Ph.D   ( mesh )
Dissertations, Academic -- College of Medicine -- Department of Neuroscience -- UF   ( mesh )
Genre:
bibliography   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph.D.)--University of Florida, 1995.
Bibliography:
Bibliography: leaves 124-135.
Statement of Responsibility:
by Josephine Jean Mitchell.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 002298782
oclc - 49818746
notis - ALQ2044
System ID:
AA00009034:00001


This item is only available as the following downloads:


Full Text












AGING AND N-METHYL-D-ASPARTATE RECEPTORS


By

JOSEPHINE JEAN MITCHELL











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


UNIVERSITY OF FLORIDA


1995






























To Dad










ACKNOWLEDGMENTS


I would like to thank my advisor, Dr. Kevin J. Anderson,

for sharing his extensive knowledge and for his support

throughout my doctoral studies. My research was supported by

a National Institute of Aging grant (NIA #AG08843). I would

also like to recognize my other committee members: Dr. A.

John MacLennan, Dr. Joanna Peris, Dr. Tom Vickroy and Dr. Don

Walker for their constructive criticisms and suggestions. I

am also very grateful to Dr. Janet Zengel's for her support

and friendship through the years.

Michael S. Sapper deserves a special acknowledgement for

his technical assistance as well as his friendship from the

very beginning. Special thanks go out to Tanya McGraw for

her friendship and support as well as to many of my other

fellow graduate students.

I wish to especially mention my father, Ignaceous J.

Maddalena, who unfortunately passed away just before I began

my doctoral studies. He was and always will be the person

who fueled my love for academics and taught me the importance

of hard work and perseverance in the pursuance of knowledge.

My mother, Helen, has been a loving and supportive figure

throughout my life and I wish to thank her with all of my

heart as well as the rest of my family and friends.


iii









Finally, and most importantly, I wish to acknowledge my

husband, Thomas Mitchell (Mitch), who has given me invaluable

support throughout my academic pursuits as the 'one and only

love of my life' and father to our two beautiful sons Kyle

Ryan and Ty Joseph. I could never have achieved all I have

worked for without the constant love and support of my family

and friends.















TABLE OF CONTENTS


ACKNOWLEDGMENTS ........................................iii-iv

ABSTRACT ...................... ...................... ... vii-ix

CHAPTERS

1 INTRODUCTION.................................... ..... 1

Background ............................................. 1
Previous Studies Examining Age-Related Changes in
NMDA Receptors ...................................... 4
Molecular Cloning and Characterization of the NMDA
Receptor/Channel Complex .......................... 8
The Use of NMDAR1 and NMDAR2A/B Antisera in the
Analysis of NMDA Receptor/Channel Protein
Density and Distribution in the Rat CNS .......... 17
Cooperative Modulation of [3H]MK-801 Binding to the
NMDA Receptor-ion/Channel Complex by L-
Glutamate, Glycine and Polyamines .................. 19
Specific Aims ........... ........ ........ ....... .......22

2 GENERAL METHODS ......... ...... .... ...................... 23

Animal Model ................. .................... ... 23
[3H]MK-801 Autoradiography ............................24
Tissue Preparation ...............................24
[3H]MK-801 Binding Assay ............................25
Image Analysis .....................................26
Statistical Analysis .............................. 27
Stereological Determination of Neuronal Density .......27
Tissue Preparation ..................................27
Data Analysis ...................................... 28
Determination of Neuronal Density in the Lateral
Striatum, Entorhinal Cortex and Inner Frontal
Cortex ... ... ... ............................ .....30
L-Glutamate Stimulation of [3H]MK-801 Binding to
NMDA Receptors ..................................... 31
Tissue Preparation .................................31
[3H]MK-801 Binding Assay ............................31
Image Analysis .................................. .... 32
Data Analysis ......... ..................... ....... 33
In situ Hybridization ................................ 33
Probe Labeling ... ... ........................ ...... 33
Tissue Preparation ................................. 34









Fixation ...........................................35
Hybridization ...................................... 35
Autoradiography .................................. 36
Data Analysis ......................................37
Immunocytochemistry .................................. 37
Tissue Preparation .................................37
Antibody Specificity .............................. 37
Immunocytochemical Procedure .......................38

3 AGE-RELATED CHANGES IN [3H]MK-801 BINDING IN F-344
RATS .............. ............................ .......... 40

Introduction ............... ............... .......... 40
Methods .................. oo .. ... ..... .... 42
Statistical Analysis for [3H]MK-801 Binding ........42
Results .............................................. 44
[3H]MK-801 Binding ................................44
Discussion ........................................... 51

4 EFFECT OF AGE ON L-GLUTAMATE STIMULATION OF [3H]MK-
801 BINDING IN F-344 RAT BRAIN..........................59

Introduction ........................... ....... ...... 59
Methods ............................................... 61
Tissue Preparation .................................61
[3H]MK-801 Binding Assay ..........................61
Image Analysis ............. ........................ 62
Data Analysis ...... .............................. ..62
Binding Isotherm Plots ............................. 63
Statistical Analysis ..............................63
Results ............................................... 64
Discussion ................................ ....... ....70

5 AGE-RELATED CHANGES IN THE LEVELS OF mRNA CODING
FOR SPECIFIC NMDA RECEPTOR SUBUNITS IN THE CNS OF F-
344 RATS ................................................. 77

Introduction o ......................................77
Methods ............................................... 78
In Situ Hybridization ............................. 78
Data Analysis .................................... 79
Statistical Analysis ............................. 79
Results .............................................. 80
Discussion ............................................97

6 AGE-RELATED EFFECTS ON NMDAR1 AND NMDAR2A/B PROTEIN
LEVELS IN F-344 RAT BRAIN UTILIZING
IMMUNOHISTOCHEMISTRY................... ............. 102

Introduction ................................. ...... 102
Methods ............................................. 105
Tissue Preparation ................................ 105









Antibody Specificity ............................ 105
Immunocytochemical Procedure ......................106
Data Analysis .................................... 106
Statistical Analysis .............................. 107
Results .............................................. 108
Discussion ................................... .......113

7 SUMMARY AND DISCUSSION............................... 117
Research Summary .................... ............. 117
Discussion .......................................... 121

LIST OF REFERENCES ...................................... 124

BIOGRAPHICAL SKETCH ....................................... 136


vii















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


AGING AND THE N-METHYL-D-ASPARTATE RECEPTOR

By

Josephine Jean Mitchell

December, 1995




Chairman: Kevin J. Anderson
Major Department: Neuroscience


Aging is associated with a reduction in cognition and

memory in humans and other animals. Excitatory amino acids

(EAA) play pivotal roles in learning and memory. L-Glutamate

is thought to be the major EAA transmitter in mammalian

brain. L-Glutamate transmitter systems may therefore be

involved in age-related deficits. The N-methyl-D-aspartate

(NMDA) ionotropic glutamate receptor subtype appears

important in learning and memory. It is critical to long-

term potentiation and NMDA antagonists impair performance of

rats in spatial and reference memory tasks. Antagonism of

NMDA receptor-mediated neurotransmission produces behavioral

deficits strikingly similar to those detected in aged

animals.


viii









Thus deficits in NMDA receptor neurotransmission may

underlie age-related changes in neuronal plasticity. One

possible way NMDA neurotransmission could be reduced in aged

animals is by alterations in NMDA receptor/channel complexes.

The central hypothesis tested in this study was that

there are measurable, anatomically specific age-related

changes in the NMDA receptor and its individual subunits. An

initial examination of age-related differences in NMDA

receptor density in brains of 6-, 12- and 24-month-old F344

rats was performed with [3H]MK-801 in vitro autoradiography.

An age-related decrease was found in the entorhinal and inner

frontal cortices and the lateral striatum.

An analysis of L-glutamate's ability to enhance [3H]MK-

801 binding to NMDA receptor channels was performed by

varying L-glutamate concentration. Emax and EC50 values were

obtained for several brain regions showing an age-dependent

decrease in Emax values without changes in EC50.

The possibility that receptor binding changes were due

to decreases in specific NMDA receptor subunits was analyzed

using in situ hybridization. A significant difference was

seen in NMDAR1 subunit mRNA in all brain regions. Splice

variants NRlOxx, NR1ixx and NRlxlx changed in fewer regions

while NMDAR2 subunits did not change with age.

No age-related differences in NMDAR1 and NMDAR2A/B

protein density levels were found using immunocytochemistry.

These results indicate the NMDA receptor undergoes

regionally specific changes during aging and these changes






x


may account for some of the cognitive deficits in the aging

population.















CHAPTER 1
INTRODUCTION


Background


Ramon y Cajal, as early as the turn of the century,

proposed the connective foundation of neural memory (Cajal,

1911). Many years later, Hebb (1949) formulated his

principles of memory formation which was based upon

facilitation of contacts between neurons. Both Cajal and

Hebb focused on the synapse as the location of plastic change

in the formation of memory. They also theorized that the

processes of learning and memory result from changes in the

strength of synaptic transmission (Cajal, 1911; Hebb, 1949).

Bliss and Lomo (1973) found that in the hippocampus, a

brain region known to be important for learning (Nicoll et

al., 1988), brief repetitive activation of excitatory

pathways resulted in a substantial increase in synaptic

strength that lasted for many hours and, in vivo, even for

weeks. Since its initial discovery, this synaptic

enhancement, referred to as long-term potentiation (LTP), has

provided a model in vertebrate brain for a cellular mechanism

of learning and memory (Bliss and Lomo, 1973). LTP is

defined as an electrophysiological phenomenon of persistent

change in synaptic strength or efficacy as a result of









impulse transmission across synapses. These activity-induced

changes in the efficacy of existing synapses are thought to

be mediated by excitatory synaptic transmission throughout

the central nervous system (Bliss and Lomo, 1973).

Excitatory glutamatergic synapses are abundant

throughout the central nervous system (CNS), especially in

the hippocampus and cerebral cortex (Westbrook and Jahr,

1989). There are at least three types of ionotropic

glutamate receptors classified by the potent, selective

agonists N-methyl-D-aspartate (NMDA), kainate (KA) and a-

amino-3-hydroxy-5-methylisoxazole-4-propionic acid (AMPA)

(Foster and Fagg, 1984). During low-frequency synaptic

transmission, glutamate is released from the presynaptic

terminal and acts on both the NMDA receptor subtypes as well

as the non-NMDA receptor subtypes (KA and AMPA). AMPA and KA

receptors are permeable to K+ and Na+ and provide a voltage-

independent means of depolarizing the postsynaptic neuron.

However, the NMDA receptor has unique biophysical

characteristics making it highly relevant to neural

plasticity and memory formation (Collingridge, 1987; Cotman

and Iversen, 1987; Harris et al., 1984). During low

frequency stimulation, ion flux through the NMDA channel is

blocked by Mg+2 in a voltage-dependent manner. Mg+2 ions are

expelled from membrane channels after the postsynaptic

membrane has reached a certain level of depolarization.

Following relief of the Mg+2 block, the receptor channel is

permeable to Na+ and Ca+2. Furthermore, NMDA receptor-









mediated current increases as a function of the degree of

depolarization are longer in duration than those from non-

NMDA receptors. The necessity for both presynaptic release

of transmitter to bind to the NMDA receptor as well as

sufficient postsynaptic depolarization to relieve the

voltage-dependent channel block by Mg+2 give this receptor

Hebb-like properties.

It is now thought that the activation of NMDA receptors

is responsible for the induction of LTP in the hippocampus

(see Nicoll et al., 1988 for review). NMDA antagonists have

been shown to block the development of LTP in the CA1 region

of the hippocampus during high-frequency tetanic stimulation

of the Schaffer commissural pathway (Collingridge et al.,

1983) as well as LTP in the dentate gyrus after perforant

path stimulation (Errington et al., 1987). In addition, a

variety of NMDA antagonists, both competitive (e.g., 2-amino-

5-phosphonovalerate [AP5], CGS 19755) and noncompetitive

(e.g., MK-801, phencyclidine [PCP]), have been shown to

impair performance of rats in the acquisition of spatial

working and reference memory tasks (Shapiro and Caramanos,

1990). NMDA receptor channel antagonism by MK-801 also

impairs performance of rats in aversively motivated complex

maze tasks (Spangler et al., 1991). These results provide

evidence for the NMDA receptor playing a role in memory and

spatial learning and LTP (see also Danysz et al., 1988;

Morris et al., 1986).









Aging is associated with a reduction in cognition and

memory in humans as well as other animals (Barnes, 1979;

Barnes and McNaughton, 1985). Senescent rats are slower than

young rats in learning various spatial tasks (Barnes, 1979;

DeToledo-Morrell et al., 1984) which may be due to age-

related changes in NMDA receptor-mediated neurotransmission.

In support of this, it has been shown that antagonism of NMDA

receptor-mediated neurotransmission produces behavioral

deficits strikingly similar to those detected in aged animals

(Bonhaus et al., 1990).

Taken together, the above evidence suggests that a

reduction in NMDA receptor-mediated neurotransmission may

underlie age-related changes in neuronal plasticity. One

means by which NMDA neurotransmission could be reduced in

aged animals is by an alteration in the NMDA receptor/channel

complex itself.


Previous Studies Examining Age-Related Changes in NMDA
Receptors


Previous studies have examined the effects of aging on

the density of central excitatory amino acid (EAA) receptors

in brain membrane preparations. For example, Tamaru et al.

(1991) reported a reduction in the number of NMDA receptors

in the cerebral cortex and hippocampus of aged rats.

Specifically, they found a significant decrease in

[3H]glutamate binding displaceable by NMDA, in strychnine-

insensitive [3H]glycine binding and in [3H]MK-801 binding in









both brain regions. Scatchard analysis revealed that the

reduction was due to a decrease in the number of binding

sites, not to an alteration in affinity. The reduction in

the number of receptors was apparently not due to a relative

increase of insoluble protein in membrane fractions since the

protein concentrations and ratio of protein concentration to

tissue wet weight were not significantly different between

young and aged animals. Similar results were obtained by

Ingram et al. (1992), who observed a marked (>50%) age-

related reduction in NMDA receptor binding in rat hippocampal

brain homogenates, and by Peterson and Cotman (1989) who

found reductions in NMDA-displaceable [3H]L-glutamate binding

density in two strains of aged mice.

The effect of increasing age on the binding of [3H]L-

glutamate to NMDA receptors in the brain of the BALB/c and

C57B1 mouse strains was also determined using in vitro

quantitative autoradiographic analysis (Magnusson and Cotman,

1993). In this study, a significant decrease in binding to

NMDA receptors occurred with increasing age (ranging from 3

through 30 months). NMDA receptors, as opposed to non-NMDA

receptors, were selectively affected by age-related changes

in the majority (17 of 20) of the cortical, subcortical and

hippocampal regions assayed in both strains of mice.

[3H]Kainate and [3H]AMPA (non-NMDA) binding, on the other

hand, was decreased in only 7 of 21 and 4 of 21 regions,

respectively. They concluded that the NMDA receptor is









selectively vulnerable to the aging process throughout most

cerebral cortical, subcortical and hippocampal regions.

A similar autoradiographic study has also demonstrated

that specific brain regions in aged animals exhibit a decline

in NMDA receptor density (Anderson et al., 1989). The

density of NMDA receptors labeled with [3H]L-glutamate in

young-adult (4-month-old) Fischer 344 rats was compared to

aged (24- to 26-month-old) rats. The areas that showed a

greater than 30% decrease in aged rats included the lateral

striatum, inner layers of the entorhinal cortex and the

lateral septal nucleus. The receptor density in inner

parietal cortex and anterior cingulate cortex was decreased

by 10% in aged rats when compared to young-adults. Within a

given brain region there was no significant difference in the

affinity (KD) of the NMDA receptor for [3H]L-glutamate.

However, a significant decrease in the Bmax was seen within

the brain regions most affected by increasing age which

indicated that there was a decrease in the total number of

receptors (Anderson et al., 1989). This group also utilized

[3H]glycine binding to probe the associated allosteric

activating site on the NMDA receptor. Areas that showed the

greatest degree of loss of [3H]glycine binding included the

lateral striatum, parietal cortex and the entorhinal cortex.

There was no significant decrease in the lateral septal

nucleus and the loss of glycine binding sites in aged rats

was more profound in the parietal cortex and outer portions









of the entorhinal cortex when compared to [3H]L-glutamate

binding.
Age-related changes in the glutamatergic neuro-

transmitter systems in rats and rhesus monkeys has been

examined utilizing NMDA-displaceable [3H]L-glutamate binding

to brain homogenates (Wenk et al., 1991). [3H]L-Glutamate

binding density was decreased in many brain regions in aged

rats (24 month-old) when compared to young (5 month-old)

rats. Specifically, the sensory-motor cortex, the parietal-

occipital cortex, the hippocampus and the caudate nucleus all

showed a significant decrease in binding density. In monkey

brains, NMDA-displaceable [3H]L-glutamate binding was

decreased in most brain regions analyzed and particularly

noticeable in the frontal and temporal lobes of the aged

monkeys (29-34 years) when compared to young (4-9 years)

monkeys.

Senescence-accelerated mice (SAM-P/8) have been used as

a murine model of aging and memory dysfunction (Kitamura et

al., 1992). This strain of mice shows an age-related

deterioration of learning and memory at an earlier age when

compared with control mice. In brain homogenates of

hippocampus and cerebral cortex from SAM-P/8 mice there was a

significant increase in the content of glutamate and

glutamine when compared to controls. Potassium-evoked

endogenous glutamate release from the brain slices of SAM-P/8

mice was increased in comparison with the control strains at

9 and 11 months. Additionally, the Bmax of [3H]dizocilpine








(MK-801) binding in the cerebral cortex was decreased in SAM-

P/8 but not in controls. This suggests that synaptic

dysfunctions in the glutamatergic system occur in the CNS of

the SAM-P/8 mouse strain (Kitamura et al., 1992).


Molecular Cloning and Characterization of the NMDA
Receptor/Channel Complex


A functional cDNA clone for the rat NMDA receptor

(NMDAR1) was first isolated in 1991 (Moriyoshi et al., 1991).

The single protein encoded by the cDNA was shown to form a

functional receptor/ion-channel complex with

electrophysiological and pharmacological properties

characteristic of the NMDA receptor; i.e. agonist and

antagonist selectivity, modulation by glycine, Ca2+

permeability, a voltage-dependent channel block by Mg2+, and

Zn2+ inhibition (Moriyoshi et al., 1991).

Three cDNAs encoding different NMDA receptor subunits

were isolated by polymerase chain reaction (PCR) (Monyer et

al., 1992). Two degenerate oligonucleotide primers were

designed after largely conserved peptide sequences in

ionotropic EAA receptor subunits, with which NMDAR1 shares

several small sequence islands around putative transmembrane

(TM) segments. These primers were used to PCR amplify

homologous sequences from rat brain cDNA. Three full-length

cDNAs, having sequences identified from the PCR products,

were named NMDAR2A (NR2A), NR2B and NR2C. The predicted

proteins were between 55% (NR2A and NR2C) and 70% (NR2A and









NR2B) identical to each other, but were only about 20%

identical to homologous AMPA-selective glutamate receptor

subunits (GluRs) and NMDAR1 (Monyer et al., 1992).

The functional properties of the expressed NR2 subunit

were examined using a Xenopus oocyte expression system. No

detectable calcium currents were recorded after bath

application of glutamate or NMDA to oocytes expressing one or

two NR2 subunits, which indicated that NR2 subunits may not

form functional homomeric or heteromeric channels. However,

large currents were measured in oocytes coexpressing NR1 and

any one of the NR2 subunits. On average, the NMDA-induced

currents in oocytes expressing NR1 and NR2A, NR2B or NR2C

were 100 times larger than they were in oocytes expressing

homomeric NR1 channels. These currents also more closely

resemble native NMDA receptors. This indicated that

heteromeric configurations are likely to form from NR1

subunits and members of the NR2 subunit family (Monyer et

al., 1992).

NMDAR1 and NMDAR2 subunits carry an asparagine residue

in the putative channel forming region TMII, whereas a

glutamine or arginine residue resides in the homologous

position of the AMPA-selective glutamate receptor subunits

(Nakanishi et al., 1992). The importance of the asparagine

residue in the regulation of Ca+2 permeability and channel

blockade was shown by electrophysiological characterization

of receptors in which the asparagine was replaced with either

glutamine or arginine (Nakanishi et al., 1992). These









substitutions reduced or abolished Ca+2 permeability and

inhibition by Mg+2, Zn+2, and MK-801. Thus, this particular

asparagine residue may constitute a distinctive functional

determinant in subunits belonging to the NMDA receptor

(Monyer et al., 1992).
Previous in situ hybridization studies have revealed

that NMDARl messenger RNA (mRNA) in adult rat brain is

expressed in almost all neuronal cells throughout the brain

(Moriyoshi et al., 1991). This group observed prominent

expression of NMDARl mRNA in the cerebellum, hippocampus,

cerebral cortex and olfactory bulb. High expression was also

seen in the granular layer of the cerebellum, in granule

cells of the hippocampal dentate gyrus and in pyramidal cells

throughout hippocampal areas CAl-CA4.

The anatomical distribution of mRNA for NMDAR2 subunits

has been examined with in situ hybridization (Buller et al.,

1993; Kutsuwada et al., 1992; Meguro et al., 1992; Monaghan

et al., 1993; Monyer et al., 1992; Moriyoshi et al., 1991;

Nakanishi, 1992). NMDAR2A mRNA is widely expressed in many

brain regions, and this expression is prominent in the

cerebral cortex, hippocampus, olfactory bulb, some thalamic

nuclei, pontine nuclei, inferior olivary nuclei and

cerebellar cortex. The NMDAR2B mRNA expression is prominent

in most of the telencephalic and thalamic regions but

relatively low in the hypothalamus, cerebellum and lower

brain stem. The distribution of the NMDAR2C mRNA is more

discrete; its expression is extremely high only in the









granular layer of the cerebellum. NMDAR2D mRNA is mainly

expressed in the diencephalic and lower brain stem regions.

Recently, the distribution of NMDA receptor subtypes, as

identified with autoradiography, have been shown to

correspond to specific receptor subunits (Monaghan et al.,

1993). Early studies had indicated that there were two

populations of NMDA receptors; agonist-preferring (striatal-

type) and antagonist-preferring (thalamic-type) (Monaghan et

al., 1988). These populations were classified by whether

they displayed a higher affinity for agonist or for

antagonist. Monaghan (1991) showed that low concentrations

of L-glutamate selectively promoted the binding of [3H]MK-801

to striatal-type NMDA sites. He concluded that low

concentrations of L-glutamate appeared to preferentially

activate striatal NMDA receptors without activating thalamic

and cortical NMDA receptors. Conversely, with higher

concentrations of L-glutamate, [3H]MK-801 will also label

NMDA receptors of the lateral thalamic nuclei and cerebral

cortex (Monaghan and Anderson, 1991).

More recently, two other pharmacologically-distinct

populations of NMDA receptors have been defined by

autoradiography. These receptors were identified in the

midline thalamic nuclei and in the cerebellum (Monaghan et

al., 1993). It has been shown that the anatomical

distribution of each of these four native receptor

populations corresponds to a specific NMDAR2 subunit

(Monaghan et al., 1993). The NMDAR2A transcript has a









distribution very similar to the "antagonist-preferring" NMDA

receptor subtype. The NMDAR2B subunit mRNA is the only

NMDAR2 species found in regions enriched in "agonist-

preferring" sites. NMDAR2C subunits are largely restricted

to the cerebellum which contains a pharmacologically-distinct

receptor subtype and the NMDAR2D subunits are restricted to

the midline-thalamic nuclei NMDA receptor subtype (Monaghan

et al., 1993). These data suggest that NMDAR2 subunits may

contribute to the pharmacological diversity of native NMDA

receptors.

Alternative splicing has been shown to generate several

functionally distinct NMDAR1 subunits (Anatharam et al.,

1992; Durand et al., 1992; Nakanishi et al., 1992; Sugihara

et al., 1992). Structures and properties of seven isoforms

of the NMDAR1 receptor are differentiated from each other by

an insertion at the extracellular amino-terminal regions or

deletions at two different carboxy-terminal regions, or by

combinations of the insertion and deletions. All of these

isoforms have been shown in the Xenopus oocyte expression

system to induce electrophysiological responses to NMDA and

respond to various antagonists selective to the NMDA receptor

(Sugihara et al., 1992). The nomenclature of Durand et al.

(1993) denotes each NMDAR1 splice variant by the presence or

absence of the three alternatively spliced exons in the 5' to

3' direction. A subscripted o denotes exclusion of an exon

while a subscripted 1 denotes its inclusion. For example,









NRl111 has all three exons, NRI000 has none, and NR1100 has

only the N-terminal insert (Figure 1-1).
insert 1 insert2 insert
#of aa: 190 21 673 / 37 38 22


5' 3'


L--_ N terminus-- C terminus I .L-_-_J


l coding non-coding 6 coding when insert is absent




Figure 1-1. Proposed gene structure of the NMDAR1 receptor
(Durand et al., 1993). Three putative inserts can be spliced
in or out to form the mature mRNA. The NMDAR1 splice
variants are denoted by subscripts indicating the presence or
absence (1 or 0) of the three inserts in the 5' to 3'
direction.

A recent study by Buller et al. (1994) described the

anatomical distributions of NMDAR1ixx (the NMDAR1 splice

variant containing insert 1) and NMDAR1oxx (the NMDAR1 splice

variant lacking insert 1). They found NMDAR11xx mRNA density

varied across cortical regions with the parietal, temporal,

and superficial entorhinal cortices displaying threefold

higher levels of NMDAR1Ixx mRNA than the anterior cingulate,

perirhinal, and insular cortices. In contrast, higher levels

of NMDAR1oxx mRNA were found in the anterior cingulate,

perirhinal, and insular cortices. They concluded that the

localization of NMDAR1ixx and NMDAR1Oxx mRNA between cortical

regions paralleled the distribution of antagonist-preferring

and agonist-preferring NMDA receptors, respectively.









NMDAR11xx mRNA displayed a lateral-to-medial gradient pattern

within the striatum whereas NMDAR1oxx was shown to be

moderately higher (15%) in the medial striatum than the

lateral striatum. NMDAR1ixx was present in low levels

throughout the septum while high levels of NMDARloxx mRNA were

found in this region. NMDARlixx mRNA was present at high

levels throughout the thalamus with low levels of NMDAR10xx.

Overall, this study showed that agonist-preferring NMDA

receptors are found predominately in the subset of brain

regions that contain both NMDAR2B and NMDAR10xx mRNA whereas

the antagonist-preferring NMDA receptors are found

predominately in brain regions containing both NMDAR2A and

NMDAR11xx mRNA. Monaghan and Buller (1994) also found that

NMDAR1 mRNAs that are alternatively spliced at the second and

third inserts have distribution patterns that are dissimilar

to that of previously described NMDA receptor subtypes. No

differences in NMDA receptor pharmacological properties have

been found after studying homomeric receptors of NMDAR1 mRNA

that contain alternatively spliced exons at the second and

third insert sites (Durand et al., 1993; Nakanishi et al.,

1992). Because of these findings, alternative splicing at

these C-terminal sites is thought to have less of an effect

upon NMDA receptor pharmacology than N-terminal alternative

splicing events.

Expression cloning in Xenopus oocytes isolated two

different cDNAs encoding functional NMDA receptor subunits.

These receptor subunits were termed NMDA-R1A (NR1011) and -RIB









(NR1111) (Durand et al., 1992). The two subunits displayed

different pharmacologic properties as a consequence of

alternative exon addition within the putative ligand-binding

domain. The splicing choice is regulated such that NR1111 is

the predominate form of the receptor in the cerebellum,

whereas NR1011 predominates in the cerebral cortex,

hippocampus and olfactory bulb. Durand et al. (1992) showed

that the functional differences between NR1111 and NR1011 are

marked and include differences in agonist affinity and

potentiation by spermine. Clearly, alternative splicing

contributes to NMDA receptor diversity. The expression of

distinct NMDA receptors with different electrophysiological

properties and anatomical distribution may be responsible for

the different forms of activity-dependent synaptic plasticity

in the mammalian brain such as the generation of different

forms of LTP (e.g., hippocampal associative LTP and

motorcortical LTP) (Nakanishi et al., 1992).

Recent studies examining the transmembrane topology of a

glutamate receptor GluRI (an AMPA subunit) showed that the N-

terminus is extracellular, whereas the C-terminus is

intracellular (Hollmann et al., 1994). In addition, three

transmembrane domains (TMD), (designated TMD A, TMD B, and

TMD C) corresponded to the previously proposed TMDs I, III,

and IV, respectively. It was found by N-glycosylation

tagging that, contrary to earlier models (Barnard et al.,

1987; Hollmann et al., 1989), the putative channel-lining

hydrophobic domain TMD II does not span the membrane.









Instead, this domain was suggested to either lie in close

proximity to the intracellular face of the plasma membrane or

loop into the membrane without traversing it. Furthermore,

the region between TMDs III and IV, in previous models

thought to be intracellular, is an entirely extracellular

domain (Hollmann et al., 1994).

Previously, Durand et al. (1993) showed that the 21-

amino acid insert in the N-terminal domain reduced the

apparent affinity of homomeric NMDAR1 receptors for NMDA and

nearly abolished potentiation by spermine and glycine. For

both of these properties, the N-terminal insert was the

determining structural feature; the C-terminal domain

produced only a very minor effect. These observations

correlate with a three transmembrane spanning topology where

the N-terminal domain is extracellular and therefore should

affect the ligand binding affinity characteristics of the

NMDA receptor. Therefore, it is not surprising that the C-

terminus would have minimal, if any, effect on ligand binding

characteristics due to its intracellular location. This was

demonstrated by Durand et al. (1993) when they examined the

electrophysiological characteristics of six splice variants

of NMDAR1 receptors and concluded that variants differing

only in their C-terminal domain showed little change in

agonist affinity or spermine potentiation.








The Use of NMDAR1 and NMDAR2A/B Antisera in the Analysis of
NMDA Receptor/Channel Protein Density and Distribution in the
Rat CNS


One of the first studies examining the NMDAR1 protein in

the rat CNS was provided by Hennegriff et al. (1992).

Polyclonal antibodies were raised against synthetic peptides

corresponding to the carboxyterminal region of the putative

NMDA receptor cloned by Nakanishi (1991). The affinity

purified antibodies to the NMDA receptor subunit labeled

antigens of 75 and 81 kDa. In order to determine the

regional distribution of NMDAR1, the 75/81 kDa doublet

protein was quantitated by laser densitometry from Western

blots in homogenate samples prepared from seven brain

regions. They found the doublet to be most abundant in

neocortex, followed by hippocampus > striatum >> thalamus >

olfactory bulb > cerebellum >> brain stem. The same regional

distribution was found in synaptosomes using antibodies to

NMDAR1 and resembled earlier autoradiographic studies

measuring NMDA-sensitive [3H]glutamate binding sites

(Monaghan and Cotman, 1985). A more thorough light and

electron microscopic analysis of the distribution of the NMDA

receptor subunit NMDAR1 in the rat CNS was performed by

Petralia et al. (1994a). This group made a polyclonal

antiserum that recognized four of the seven splice variants

of NMDAR1 and subsequently utilized this antiserum to perform

a comprehensive immunohistochemical survey of the

distribution of this antigen. They showed that the NMDAR1









subunit is widespread throughout the rat CNS. The most

densely stained cells included the pyramidal and hilar

neurons of the CA3 region of the hippocampus, Purkinje cells

of the cerebellum and paraventricular neurons of the

hypothalamus. Ultrastructural localization of NMDAR1 antigen

showed labeling present in postsynaptic densities in a

pattern consistent with the synthesis, processing and

transport of this protein. Staining was seen in the

cytoplasm of dendrites and concentrated in patches associated

with groups of microtubules and/or the surface of one pole of

a mitochondrion. In addition, patches of staining in the

cell bodies were found to form similar associations with

microtubules and mitochondria, as well as an association with

rough endoplasmic reticulum, Golgi apparatus, and the nuclear

envelope. No staining was found in the synaptic cleft. The

antiserum did not cross-react with extracts from transfected

cells expressing other glutamate subunits, nor did it label

non-neuronal tissues. The pattern of staining correlated

closely with previous in situ hybridization studies but

differed somewhat from binding studies (Petralia et al.,

1994a).

Another study by this laboratory (Petralia et al.,

1994b) examined the histological and ultrastructural

localization patterns of the NMDA receptor subunits NMDAR2A

and NMDAR2B. They made a polyclonal antiserum to a C-

terminus peptide of NMDAR2A. In analysis of membranes from

transfected cells, this antiserum recognized NMDAR2A and








NMDAR2B, and to a slight extent, NMDAR2C and NMDAR2D.

Immunostained sections of rat brain showed significant

labeling throughout the CNS that was similar to that seen

previously with their antiserum to NMDAR1. Dense staining

was present in postsynaptic densities in the cerebral cortex

and hippocampus. Since there is physiological evidence that

both NMDAR1 and NMDAR2 subunits coexist in the native NMDA

receptor, their findings are consistent with this idea

(Petralia et al., 1994b).


Cooperative Modulation of r3H]MK-801 Binding to the NMDA
Receptor-ion/Channel Complex by L-Glutamate, Glycine and
Polyamines


L-Glutamate binding to the NMDA receptor, in the

presence of glycine, produces channel opening. MK-801 is an

NMDA receptor antagonist that binds with high affinity (in

the low nanomolar range) to a site located within the NMDA

channel. L-Glutamate markedly stimulates [3H]MK-801 binding

by increasing the affinity for the ligand. Other NMDA

receptor agonists (e.g. NMDA, D-aspartate) are capable of

enhancing this binding as well, but the non-NMDA receptor

agonists (AMPA and kainate) have no effect on [3H]MK-801

binding (Foster and Wong, 1987). Because MK-801 binding has

been shown to take place only if the NMDA receptor channel is

in the transmitter-activated state (Huettner and Bean, 1988),

the efficacy with which L-glutamate produces receptor

activation and channel opening can be measured by determining









the amount of [3H]MK-801 binding as a function of glutamate

concentration.

Glycine appears to be a constitutive co-agonist for NMDA

receptor activation. Glycine alone is ineffective in opening

NMDA-linked cation channels. However, glycine greatly

potentiates the frequency of channel opening in response to

NMDA receptor agonists (Johnson and Ascher, 1987; Reynolds et

al., 1987). Glycine also potentiates NMDA-stimulated Ca2+

influx in cultured striatal neurons (Reynolds et al., 1987).

An allosteric role for glycine in NMDA receptor function is

supported by autoradiographic studies. These studies showed

that the distribution of strychnine-insensitive [3H]glycine

binding sites in supraspinal brain regions mirrored the

regional distribution of NMDA-sensitive L-glutamate sites

(Monaghan and Cotman, 1985). Since glycine has been shown to

promote channel activation in response to agonist, it can be

predicted to enhance [3H]MK-801 binding since MK-801 prefers

to bind to the activated state of the channel.

Along with the allosteric co-agonist glycine site on the

NMDA receptor/complex, there is also a polyamine recognition

site. The polyamines spermine and spermidine have been shown

to increase both the maximum NMDA response amplitude (Lerma,

1992; McGurk et al., 1990; Ransom and Deschenes, 1990;

Sprosen and Woodruff, 1990) as well as the binding of [3H]MK-

801 to the NMDA receptor channel (Ransom and Stec, 1988).

Because spermine does not activate NMDA receptors in the

absence of glutamate and glycine, it has been suggested to








act at an allosteric site independent from the glycine site

(Ransom and Stec, 1988).

The mechanism by which L-glutamate and these other

agonists affect [3H]MK-801 binding affinity has not been

critically addressed. In response to varying agonist

concentrations, [3H]MK-801 binding sites may undergo a

continuum of conformational changes expressing a smooth range

of different affinities for [3H]MK-801. However, it seems

more probable that [3H]MK-801 binding is an all (open

channel) or nothing (closed channel) phenomenon since most

ion channels have not been shown to exhibit broad conductance

ranges (i.e. they are either open or closed [Ransom and Stec,

1988]). This would be more consistent with the

electrophysiological characterization of MK-801 as a use-

dependent, open channel blocker (Foster and Wong, 1987). A

more likely explanation for the observed changes in KD for

[3H]MK-801 is that they result from changes in the length of

time a channel spends in the activated state (Ransom and

Stec, 1988). At high agonist concentrations, channels are

opened for a proportionately greater amount of time and the

affinity for [3H]MK-801 is high. At low agonist

concentrations, the frequency with which the channel opens is

comparatively low. The result is an apparent reduction in

[3H]MK-801 binding affinity since there is a much lower

probability, due to decreased access, that an infrequently

opened channel will bind radioligand (Ransom and Stec, 1988).









Specific Aims

The central hypothesis tested in this dissertation

project was that there are measurable, anatomically specific

aging-related changes in the NMDA receptor/channel complex

and its individual subunits. Four specific hypotheses were

addressed within the framework of the central hypothesis.

The first hypothesis was that there are age-related

differences in the density of the NMDA receptor in F-344 rat

brain as a function of age. This hypothesis was addressed

using in vitro quantitative [3H]MK-801 binding analyses

(Chapter 3). The second hypothesis was that there are age-

related differences in the ability of L-glutamate to enhance

[3H]MK-801 binding to NMDA receptors. [3H]MK-801 binding as

a function of L-glutamate concentration was performed to test

this hypothesis (Chapter 4). The third hypothesis was that

there are selective changes in mRNA coding for subunits of

the NMDA receptor as a function of age. In situ

hybridization analyses were performed with cDNA probes

specific to NMDA receptor subunits (Chapter 5). The fourth

hypothesis was that there are specific age-related changes in

NMDAR1 and NMDAR2 protein density in the rat CNS.

Immunocytochemistry was performed with antisera specific to

the NMDA receptor subunit protein (Chapter 6).















CHAPTER 2
GENERAL METHODS



A description of the general methods that were used

throughout this dissertation will follow and will be referred

to when describing individual experiments.


Animal Model


Fischer 344 (F-344) male rats 6-, 12-, and 24-months-of-

age were used for these experiments. These animals were

obtained from the National Institute on Aging (NIA) breeding

colony (Harlan). In 1981, the Committee on Animal Models for

Research on Aging described the life span characteristics of

two rat strains, the F-344 and the Brown Norway (BN) rat, and

recommended their use as models for aging research. Among

the reasons for choosing F-344 and BN rat strains over other

rat strains was that these rats displayed delayed onset of

kidney problems nephropathyy) as well as certain tumors.

Currently, the NIA recommends and maintains a total of three

rat strains for aging studies, the F-344, the BN, and the F1

F-344/BN hybrid. The F-344 was chosen for these studies

because it has long been a standard model for aging research

and thus has been extremely well characterized. Survival

curves for the F-344 rat show that 24-month-old males have a









40% probability of survival and the probability of survival

curve drops rapidly to only 10% by 27-months-of-age.

Therefore, 24-month-old animals represent advanced stages of

aging in this rat strain and 12-month-old animals can be

classified as middle-aged. 6-month-old animals represent

young-adults and are used throughout these studies for

comparisons to middle-aged and aged rats.


13H1MK-801 Autoradiography


Tissue Preparation


Male F-344 rats 6-, 12- and 24-months-of-age (n=6 per

age group) were decapitated, their brains rapidly removed and

frozen with powdered dry ice. Brains were stored at -80 OC

until used. Sections (6pm thick) were cut on a cryostat and

representative sections from each age group were thaw-mounted

onto chromic acid washed and gelatin-subbed slides. A rat

brain atlas (Paxinos and Watson, 1982) was referenced in

order to delineate the brain regions for analyses. The

brains were sectioned in the horizontal plane. The bregma

coordinate corresponded to the dorsoventral distance of the

sections from the horizontal plane passing through bregma and

lambda on the surface of the skull. The start and stop

bregma coordinates were -4.1 mm and -6.1 mm, respectively.

This anatomical range was used to thoroughly analyze brain

regions of interest that were present on each sectional

profile. Every fifth serial section was taken and a total of









fifty sections were cut from each rat brain. One section

from each of the three age groups was mounted onto each slide

which represented one block of animals. Ten slides were used

per block of animals for each [3H]MK-801 binding assay. Two

slides per assay were used to determine non-specific binding

and these values were subtracted from each total binding

value in order to obtain specific binding. These tissue

sections were used immediately or stored for no longer than

24 hours at -20 oC prior to the [3H]MK-801 binding assay.


13H1MK-801 Binding Assay


Slides were thawed and preincubated at room temperature

for 10 minutes in 50 mM Tris acetate with 1.0 mM EDTA and

0.1% saponin (pH 7.7). Sections were then rinsed at 30 OC

for 60 minutes in 50 mM Tris-acetate buffer (pH 7.7). This

treatment removes endogenous glutamate, glycine and various

ions. Sections were then incubated for 60 minutes at room

temperature in lOnM [3H]MK-801 (30 Ci/mmol;New England

Nuclear, Boston,MA,U.S.A.) in 50 mM Tris-acetate buffer

containing 20 pM D-2-amino-5-phosphonopentanoic acid (D-

AP5), 250 pM spermine, 25 pM glycine and 20 pM L-glutamate.

D-AP5, a competitive NMDA antagonist, was added to all

[3H]MK-801 incubations to give lower and more consistent

basal binding levels (Monaghan 1991). The addition of

spermine, glycine and L-glutamate ensures a maximal degree of

stimulation of the NMDA receptor/ channel complex and

therefore optimal [3H]MK-801 binding. Sections were then









washed for 60 minutes in ice-cold Tris-acetate buffer

containing 20 pM D-AP5. Nonspecific binding was defined in

sections treated identically in the presence of 50 pM MK-801.

Following the rinse, the sections were dried under an air

stream. The sections were then placed in x-ray cassettes and

apposed to tritium-sensitive film (Hyperfilm, Amersham).

Tritium standards calibrated against brain paste were

included in the cassettes (Microscales, Amersham). The film

was exposed for six weeks and then developed using Kodak D-19

developer.

Receptor densities (expressed in pmol/mg protein) from

the 12- and 24-month rat brains were normalized against

values derived from 6-month-old animals. Data from the 12-

and 24-month-old rats were presented as a percentage of the

receptor density values from the 6-month-old animals.


Image Analysis


Each slide contained a representative horizontal section

from each age group. Ten slides were analyzed per block of

animals, with each block consisting of an animal from each

age group. Autoradiograms were analyzed, with the

investigator blind to age, by computer assisted densitometry

with a Microcomputer Imaging Device (MCID, Imaging Research,

Inc., St. Catherines, Ont.). Densitometric measurements were

converted on line to pmol/mg protein binding.









Statistical Analysis


Six blocks of animals, each block consisting of an

animal from each age group, were assessed. The experimental

design consisted of a within-subject factor with eight levels

(brain region), a between-subject fixed factor with three

levels (age), and each of six levels (experimental day).

Repeated measures analysis of variance (ANOVA) with between-

subject randomized blocks and age considered as a linear

covariate were used to model the amount of receptor density

as a function of age within each brain region and to

determine if this pattern differed significantly among brain

regions.

The brain regions that were analyzed included the outer

and inner frontal cortices (OFCTX and IFCTX); the entorhinal

cortex (ERC); the molecular layer of the dentate gyrus (DG);

hippocampal area CA1 stratum radiatum (CA1); the lateral

septum (LSEP); the lateral striatum (LSTR); and the lateral

thalamus (LTHAL).


Stereological Determination of Neuronal Density



Tissue Preparation


Thirty micron sections of brains from 6-, 12-, and 24-

month-old F-344 rats (n=4 per age) were cut on a cryostat.

One section from each age group was thaw-mounted on each

slide with a total of fifty slides for each block of animals.









The brain regions of interest in these studies were those

that showed an age-related decrease in [3H]MK-801 binding as

determined by the first experiment in Chapter 3. These

regions were the lateral striatum (LSTR), the inner frontal

cortex (IFCTX) and the entorhinal cortex (ERC). Sections

were stained with Cresyl violet in order to view neuronal

cell bodies.

A rat brain atlas (Paxinos and Watson, 1982) was

referenced in order to delineate the stereotaxic coordinates

of brain regions for analyses. Brains were sectioned in the

horizontal plane. Bregma coordinates correspond to the

dorsoventral distance of the sections from the horizontal

plane passing through bregma and lambda on the surface of the

skull. The start and stop bregma coordinates were -3.1 mm

and -7.9 mm, respectively. This broader anatomical range was

chosen to thoroughly analyze the dorsal to ventral extent of

the lateral striatum as well as to view the entorhinal and

inner frontal cortices. Two serial sections were taken every

120pm throughout the stereotaxic range.


Data Analysis


Five blocks of animals were used for these analyses with

each block containing one 6-month-, one 12-month- and one 24-

month-old rat. An initial analysis was performed to

ascertain whether the lateral striatum (LSTR) underwent

volumetric changes by sampling this brain region through its

dorsoventral extent. The volume of this structure could be









readily determined because its entire anatomical distribution

could be viewed throughout the series of sections. The LSTR

was sampled at five separate anatomical levels (2 serial

slides per level) with computer assisted image analysis

(Microcomputer Imaging Device, Imaging Research, Inc.). A

determination was then made as to whether there was a

significant age-related change in the overall volume (mm3) of

this structure. In addition, a quantitative analysis of the

laminar thickness of the inner frontal cortex (IFCTX) and the

entorhinal cortex (ERC) in 6-, 12- and 24-month-old animals

was performed. This study was performed to determine whether

there were changes in laminar thickness in these cortical

structures in the aged brain when compared to the young

animals. Calculations of laminar thickness were used because

both the entorhinal cortex and inner frontal cortex have

highly organized laminar structures. There were no

significant age-related differences in the volume of the LSTR

or the laminar thickness of the ERC or the IFCTX.

Neuronal packing density units are in neurons/mm2. No

significant age-related differences in the volume of the LSTR

or in the laminar thickness of the ERC or IFCTX were found,

therefore packing density should not vary as a function of

age. Therefore, any change seen in neuronal packing density

would be attributable to differences in the number of neurons

per mm2. In support of this, Coleman et al. (1987) examined

the volumes of the components of the hippocampus in the aging

F-344 rat and found that there was no change in volume









between 12 and 37 months. They suggested that neuronal

packing density (expressed as the number of neurons per mm2)

could be compared in aged animals independent of a volumetric

effect.


Determination of Neuronal Density in the Lateral Striatum,
Entorhinal Cortex and Inner Frontal Cortex


An MCID impage analysis system was used to determine

neuronal packing density in three brain regions. A fully

automated sampling scheme was utilized where the general

target characteristics were defined. The two target

definitions that were used were target density and target

size. For the first criteria of target density, MCID was

given upper and lower density thresholds (segmentation

ranges) and all pixels darker or lighter than the threshold

values were ignored. Pixels that were lying within the

segmentation range were valid targets. The second criteria

for target acceptance was defined as a range of target

lengths where the target neurons smaller or larger than the

defined size range ( >5pm and <20pm) would be ignored with

those targets fulfilling the length criteria being counted.

After the system found the defined targets, post-scan editing

of the digitized image was performed using editing tools.

This enabled the investigator to separate any neuronal

targets that may have been overlapping.

The area searched for targets (scan area) was also

quantified, thus neuronal packing density (number of









neurons/mm2) was determined for each brain region. Ten

serial sections were analyzed from each age group with the

investigator blind to age. Neuronal density from 6-month-old

animals was used to normalize counts from 12-, and 24-month-

old animals. Results were then statistically compared

between age groups (one-way repeated measures ANOVA, paired

t-tests).


L-Glutamate Stimulation of f3H1MK-801 Binding to NMDA
Receptors



Tissue Preparation


Male F-344 rats 6-, 12- and 24-months-of-age (n=6 per

age) were decapitated and their brains were rapidly removed

and frozen with powdered dry ice. The brains were stored at

-80 OC until used. Fifty 6pm sections were cut on a cryostat

and representative sections from each age group were thaw-

mounted onto acid washed and gelatin-subbed slides. These

tissue sections were used immediately or stored for no longer

than 24 hours at -20 oC prior to the [3H]MK-801 binding

assay.


13H1MK-801 Binding Assay


The [3H]MK-801 binding assay was the same as described

previously with the exception that varying concentrations of

L-glutamate were added to the incubation step in order to

examine glutamate's ability to stimulate [3H]MK-801 binding.









The following concentrations (in micromolar) of L-glutamate

were used: 0.0, 0.05, 0.1, 0.25, 0.5, 1.0, 2.5, 5, 10 and 20.

Non-specific binding was defined by the addition of 50 pM

unlabeled MK-801 with 20pM L-glutamate. Each slide contained

one section from each of the three age groups. A total of

four slides, from each block of animals, were analyzed per

assay (n=6) for each L-glutamate concentration.

Receptor density (expressed in pmol/mg protein) from 12-

and 24-month rat brains was normalized against 6-month-old

rats. Data from 12- and 24-month-old animals were

statistically compared to receptor density values from 6-

month-old animals.


Image Analysis


Each slide consisted of a horizontal section from each

age group taken from between bregma coordinates -4.1 mm and

-6.1 mm (the dorsoventral distance of the sections from the

horizontal plane passing through bregma and lambda on the rat

skull). This stereotaxic range was used for a thorough

analysis of the brain regions of interest that were present

on each sectional profile. A total of fifty sections were

cut for each block of animals, with every fifth section being

taken throughout the stereotaxic range. Autoradiograms were

analyzed, blind to age, by computer assisted densitometry

with an MCID system. Four sections from each block of

animals, for each concentration of L-glutamate, were

analyzed. Two slides per assay were analyzed for non-









specific binding and these values were subtracted from each

total binding value in order to obtain specific binding.

Densitometric measurements obtained were converted on line to

pmol/mg protein binding.


Data Analysis


The brain regions that were analyzed included the OFCTX

and IFCTX; the ERC; the molecular layer of the DG;

hippocampal area CAl (stratum radiatum); the LSEP; the LSTR;

and the LTHAL.

INPLOT (GraphPad Software, San Diego, CA) was used to

generate rectangular hyperbola plots or binding isotherms.

These binding isotherm plots yielded EC50 (effective

concentration at 50%) and Emax (maximal effect) values. EC50

values and Emax values were obtained for each age group and

each brain region. Statistical analyses were then performed

to ascertain significant age-related differences in these

parameters (repeated measures ANOVA; F tests).


In situ Hybridization


Probe Labeling


Oligonucleotide probes for this study were graciously

supplied by Dr. Daniel T. Monaghan, University of Nebraska

Medical Center, Omaha NE. These probes are 45 nucleotides in

length (45 mer) and are specific to NMDAR1, four

alternatively spliced versions of NMDAR1 (NR1oxx, NRlixx,









NRlxlx NRlxxl) and the members of NMDAR2 subunit family

(NMDAR2A, NMDAR2B, NMDAR2C, NMDAR2D). The specificity of

these probes was previously confirmed by Buller and Monaghan

et al. (1994) by incubation of the radiolabeled

oligonucleotides in the presence of excess (100 nM )

unlabeled probe. In this study, only oligonucleotides with

the same sequence inhibited probe hybridization.

The oligonucleotide probes were labeled using a New

England Nuclear (NEN) labeling kit and following the methods

described by Wisden et al. (1991). The probes were labeled

at the 3'- end using [35S]dATP (1000-1500 Ci/mmol, NEN) and 3'

terminal deoxynucleotidyl transferase (NEN). After labeling,

the probes were used immediately or stored for a limited time

at -70 oC.


Tissue Preparation


Frozen brains from 6-, 12- and 24-month-old rats (n=6

per age) were cut at 12 pm, in the horizontal plane, on a

cryostat and thaw-mounted on Fisher Superfrost Plus glass

slides. The stereotaxic bregma coordinates were from -4.1 mm

to -6.1 mm. Sections were refrozen in the cryostat. One

section from each age group was mounted per slide for a total

of three sections per slide. A total of fifty slides was

prepared per block of animals. Four slides per probe, each

containing a section from each age group, were used. Gloves

were worn at all times during sectioning of the brains and









handling of the slides in order to avoid contamination with

ribonucleases.


Fixation


The sections were fixed immediately after cutting by

placing the slides in 4% paraformaldehyde at 40C for 5-15

minutes. Slides were washed in 0.1M phosphate-buffered

saline (PBS) for 1 minute and 70% EtOH (RNase-free) for

several minutes. Slides were then stored in 95% EtOH in a

cold room (40C) until needed. All solutions were made with

diethylprocarbonate (DEPC)-treated water.


Hybridization


Slides were removed from storage and air dried at room

temperature. Each labeled probe was dissolved in appropriate

amounts of hybridization buffer (NEN, Boston, MA), containing

0.2 M dithiothreitol to achieve a final concentration of 2000

cpm/pl. 100 pl of this solution was applied to the slide and

covered with a glass coverslip. Slides were incubated at 42

OC overnight with a parafilm wrap. The following day,

parafilm and coverslips were removed and the slides were

placed in ix SSC (saline sodium citrate buffer) at 60 oC for

20 minutes. Finally, the sections were rinsed again in lx

SSC at 60 oC for 5 seconds and then rapidly dried under an

air stream.

In situ hybridization is an experimental method that can

be used to determine the relative density of individual mRNA









species. Each 45 mer probe was constructed to be highly

specific and hybridizes to its particular NMDA subunit mRNA.

One potential drawback is that only the relative density of

mRNA can be obtained from this method. Therefore, only semi-

quantitative analyses were performed. Another drawback is

that each probe corresponding to specific NMDA receptor

subunits could only be compared across each age group. No

quantitative comparisons could be made between various

probes. In other words, the relative density of the NMDAR1

mRNA could only be compared between 6-, 12- and 24-month-old

animals from the same assay. This information could not be

compared to the data obtained for any of the other mRNAs

(i.e. NMDAR2A-D and the four alternatively spliced versions

of NMDAR1). Another drawback is that the analyses gave an

assessment of the relative density of mRNA while not all mRNA

is necessarily translated into protein.


Autoradiography


Dried slides were placed in an x-ray cassette and

apposed to B-max film (Hyperfilm, Amersham) for two weeks,

developed in D-19 (Kodak) developer and fixed with Rapid Fix

(Kodak). Autoradiograms were analyzed by computer assisted

densitometry with a Microcomputer Imaging Device, Imaging

Research, Inc.(MCID) image processing system.









Data Analysis


The brain regions analyzed included the OFCTX, IFCTX,

LSEP, LSTR, LTHAL, ERC, DG and hippocampal areas CAl and CA3.

Brain regions were compared between the three age groups and

the relative density of each subunit mRNA was normalized

against the values obtained for young animals. Probes were

labeled simultaneously and labeling experiments were grouped

such that variability was kept to a minimum. A semi-

quantitative measurement of mRNA levels and the distribution

of the different subunits was compared between ages.


Immunocytochemistry



Tissue Preparation


Brains from 6-, 12- and 24-month rats (n=5) were

serially sectioned at 30 pm on a cryostat and thaw-mounted

onto subbed microscope slides and then refrozen in the

cryostat. Every fifth section was taken for a total of fifty

slides. One brain section from each block of animals was

mounted per slide n=5.


Antibody Specificity


AB1516 (Chemicon Intl. Inc., Temecula, CA), an antiserum

raised against a synthetic peptide corresponding to the C-

terminus of rat NMDA receptor subunit (NMDAR1), is selective

for splice variants NRl011, NR1111, NR1001, NR1101. These









appear to be the major splice variants expressed in rat brain

(Hollmann et al., 1993). It has been shown previously that

there is no cross reactivity of this antiserum with other

glutamate receptor subunits (Petralia et al., 1994a). AB1548

(Chemicon Intl. Inc., Temecula, CA), an antiserum raised

against a synthetic peptide corresponding to the C-terminus

of rat NMDAR2A receptor subunit, recognizes both NMDAR2A and

NMDAR2B subunits equally. This antiserum shows no cross

reactivity with NMDAR1 or other glutamate receptor subunits

(Petralia et al., 1994b).


Immunocytochemical Procedure


An optimal antibody concentration was determined

empirically. Previous studies had used an AB1516 antibody

concentration between 2 and 4 pg/ml (Petralia et al., 1994a)

and AB1548 antibody concentration of 0.5-1.5 pg/ml (Petralia

et al., 1994b). In the present experiments, a concentration

of 2.5 pg/ml was found to produce optimal results with both

NMDAR1 and NMDAR2A/B antibodies. Sections were incubated in

10% normal goat serum in PBS (pH 7.4) for 1 hour and then

kept overnight in primary antisera (AB1516 or AB1548) in PBS.

Sections were then washed and incubated in biotinylated goat

anti-rabbit antisera (1:250 dilution) for 1 hour, washed,

incubated in avidin-horse radish-peroxidase (1:200 dilution)

for 1 hour, washed, treated for 15 minutes with 3',3-

diaminobenzidine tetrahydrochloride (0.5 mg/ml PBS + 5 pl/ml

of 0.6% hydrogen peroxide), and washed. All washes were









performed with PBS (2 x 15 min.). Sections with PBS

substituted for the primary antibody (PBS controls) were also

used in order to correct for nonspecific staining.

Staining density was semi-quantitated by an MCID image

analysis system. Data were presented as raw optical

densities. Raw optical density of NMDAR1 and NMDAR2A/B

immunoreactivity from 6-month-old animals was used to

normalize data from 12- and 24-month-old animals. Data from

middle-aged and aged animals were presented as a percentage

of the six-month-old animals.















CHAPTER 3
AGE-RELATED CHANGES IN [3H]MK-801 BINDING IN F-344 RATS





Introduction


Aging is associated with a reduction in many physiologic

forms of neuronal plasticity. Recent findings indicate a

direct correlation in age-dependent deficits in spatial

memory and in hippocampal kindling (a measure of neuronal

plasticity) as measured in 26-month-old Fischer 344 rats

(DeToledo-Morrell et al., 1984). N-Methyl-D-aspartate (NMDA)

receptor-mediated glutamatergic neurotransmission has been

shown to be important for learning and memory in animals and

man (Collingridge, 1987; Cotman and Iversen, 1987; Harris,

1984). Profound learning and memory impairments are seen in

a variety of animal species after NMDA-receptor activation is

blocked by either competitive or non-competitive antagonists

(Morris et al., 1986). There are also some striking

similarities between NMDA antagonist induced memory

impairments and age-related cognitive deficits, suggesting

that age-related alterations in NMDA receptors may be

involved in cognitive decline with increasing age (Bonhaus et

al., 1990; Ingram et al., 1992).









To test the hypothesis that NMDA receptors decline

during aging, an examination of the density of NMDA receptors

was performed in aged Fischer 344 rats. In this study,

[3H](+)-5-methyl-10,11-dihydro-5H-dibenzo(a,d)-cycloheptan-

5,10-iminehydrogen maleate (MK-801 or dizocilpine), a highly

specific non-competitive NMDA receptor antagonist was used to

determine the density of NMDA receptors. This antagonist

binds with high affinity (low nanomolar range KD) to a site

located within the NMDA receptor ion channel. Thus, MK-801

binding occurs only if the NMDA receptor is in the

transmitter-activated (open) state (Huettner and Bean, 1988).

Channel activation by an NMDA receptor agonist like L-

glutamate, in the presence of glycine, promotes high-affinity

MK-801 binding. Glycine alone cannot open NMDA receptor

channels, but is a constitutive co-agonist. Glycine, in the

presence of L-glutamate or other NMDA receptor agonists

(i.e., NMDA or D-aspartate) has been shown to greatly

potentiate the frequency of channel opening (Johnson and

Ascher, 1987; Reynolds et al., 1987). In addition, the

polyamines spermine and spermidine increase both the maximum

NMDA response amplitude as well as binding of MK-801 to the

NMDA receptor. Alone, polyamines do not activate the NMDA

receptor, suggesting the presence of an allosteric binding

site independent from the other co-agonist sites (Lerma,

1992; McGurk et al., 1990; Ransom and Deschenes, 1990;

Sprosen and Woodruff, 1990).









Because of the high degree of specificity and affinity
of MK-801 for the NMDA receptor channel complex, [3H]MK-801

is an ideal ligand to examine age-related changes in NMDA

receptors in the CNS. Also, because MK-801 only binds to the

activated state of the NMDA receptor, the total number of

available NMDA receptors can be determined by using optimal

agonist and co-agonist concentrations.


Methods

For a detailed description of the experimental tissue

used in these studies, the [3H]MK-801 binding assay

procedure, and image analyses see Chapter 2.


Statistical Analysis forf3HlMK-801 Binding


The data acquired in this study consisted of

quantitative analyses of the average density of [3H]MK-801

binding (pmol/mg protein) obtained from bilateral brain

regions for each of the 6-,12-,and 24-month-old rats (n=6 for

each age). One "group" consisted of one 6-month-, one 12-

month-, and one 24-month-old animal mounted per series of

slides. Ten slides, from each of these groups, were analyzed

per assay. The regions that were analyzed included the outer

frontal cortex (OFCTX), inner frontal cortex (IFCTX),

entorhinal cortex (ERC), dentate gyrus of the hippocampus

(DG), hippocampal area CA1, lateral septum (LSEP), lateral

striatum (LSTR) and the lateral thalamus (LTHAL).








Repeated measures analysis of variance (ANOVA) with

between-animal randomized blocks and age considered as a

linear continuous covariate was used to model the pattern of

MK-801 binding as a linear function of age within each brain

region. F-tests were also performed for the presence of a

significant interaction between age and brain region receptor

density effects. In assessing the validity of ANOVA

assumptions, it was noted that experimental day age profiles

of receptor binding were parallel within each brain region

and that between-animal variability as estimated by the

standard deviation was similar among age groups within each

brain region. This indicated that the patterns of [3H]MK-801

binding obtained for each group of 6-, 12-, and 24-month-old

rats were the same, independent of which experimental day

each group was assayed. Within age groups however, brain

region between-animal standard deviations tended to increase

as the mean level of receptor density in those regions

increased. ANOVA models assume that standard deviations do

not vary in this manner. Therefore, receptor binding density

values were transformed logarithmically, prior to analysis,

in order to appropriately model this inherent pattern of

variability within the ANOVA framework. Plots of residuals

versus predicted values, residual histograms, and residual

normal probability plots were examined to assess goodness of

fit in the ANOVA model. The statistical analyses determined

which mean age profiles of MK-801 binding density appeared to








be linearly decreasing in specific brain regions and which

profiles had slopes which were significantly less than zero.


Results



13H1MK-801 Binding


The IFCTX, ERC and LSTR were the only regions analyzed

that underwent a significant change (decrease) in [3H]MK-801

binding when 24-month-old animals were compared to 6-month-

old animals. The mean age profiles for the IFCTX, ERC and

the IFCTX all linearly decreased and had slopes significantly

less than 0 with p values of .0026, .0151 and .0018

respectively (Figure 3-1). The other brain regions analyzed,

the OFCTX, the LSEP, the LTHAL, the molecular layer of the DG

and hippocampal area CA1, did not change as a function of

increasing age (Figures 3-2 and 3-3). A representative

autoradiograph shows the distribution of [3H]MK-801 binding

in the rat brain (Figure 3-4).









1.0



0.8

6mo.
o 0.6
--" U 12 mo.











IFCTX ERC LSTR

Brain Region
Figure 3-1. Brain regions that underwent a significant age-
related decrease in [3H]MK-801 binding. (IFCTX=inner frontal
cortex; ERC=entorhinal cortex; LSTR=lateral striatum).
Repeated measures ANOVA with between-animal randomized blocks
and age as a linear covariate presented as a function of
increasing age. The p values represent mean age profiles with
slopes significantly less than 0.












2-




M 6mo.
'E
1 2 l2mo.

E 24mo.
0







0-
OFCTX LSEP LTHAL
Brain Region
Figure 3-2. Brain regions that did not undergo a significant
age-related change in [3H]MK-801 binding. (OFCTX= outer
frontal cortex; LSEP= lateral septum; LTHAL= lateral
thalamus). Repeated measures ANOVA with between subject
randomized blocks and age as a linear covariate presented as
a function of increasing age.














liT


TIT


* 12 mo.

* 24 mo.


DG CA1


Brain Region
Figure 3-3. Brain regions that did not undergo a significant
age-related change in [3H]MK-801 binding. (DG= molecular
layer of the hippocampal dentate gyrus; CA1= hippocampal area
CA1 stratum radiatum). Repeated measures ANOVA with between
subject randomized blocks and age as a linear covariate
presented as a function of increasing age.


6 mo.


co~

~00>
SE
E^
CLa


0Q






























Figure 3-4. NMDA receptor distribution as determined by
[3H]MK-801 binding in : (A) 6-month-old, (B) 12-month-
old, and (C) 24-month-old rat brain.






49

















A B









TABLE 3-1. Laminar thickness (mm), volume (mm3), and
neuronal density (neurons/mm2) values from Fischer 344 rat
brain.
r6-month-old 12-month-old 24-month-old
Laminar


Thickness
(mm)
Frontal Cortex


Entorhinal
Cortex

Volume (B=m31
Lateral
Striatum

Neuronal
Density
(per mm2)

Entorhinal
Cortex


Inner Frontal
Cortex


Lateral
Striatum


n=5 animals
Der aae arounD


Mean= 0.1952
SEM = 0.0013

Mean= 0.2988
SEM = 0.0031



Mean=41.8491
SEM = 1.2064





Mean=1343.86
SEM = 36.60


Mean=1285.50
SEM = 35.17


Mean=1284.23
SEM = 24.01


0.19561
0.00211

0.2940
0.0037



41.8147
1.0065





1300.38
127.03


1279.89
10.86


1284.71
51.28


0.1945
0.0017

0.2980
0.0026



41.3812
0.5626





1319.50
41.33


1285.31
33.81


1297.88
21.01


No statistically significant differences found with ANOVA
(multi-comparison significance level at 95%, repeated
measures).









Discussion


These studies showed that the number of NMDA receptors,

as determined by [3H]MK-801 binding, is reduced in the ERC,

the LSTR and the IFCTX in aged F-344 rats. However, receptor

density in the OFCTX, the LSEP, the LTHAL, the molecular

layer of the DG as well as hippocampal area CA1 did not

change as a function of increasing age.

It is interesting that the decrease in the density of

[3H]MK-801 binding was limited to only three of the eight CNS

structures analyzed. The ERC was one structure that did

undergo an age-related decrease in MK-801 binding. The ERC

projects to the hippocampal formation, a component of the

limbic system associated with brain mechanisms for memory.

More specifically, the ERC is the major source of extrinsic

afferents to the dentate gyrus granule cells and pyramidal

cells of the hippocampus proper (Amaral and Witter, 1989).

There are decreased numbers of NMDA receptors in the ERC of

aged rats and this may affect the connectivity from this

structure to the hippocampus. This may result in an

alteration in NMDA receptor-mediated neurotransmission and

subsequently alter some aspects of memory (e.g., spatial

information processing) (Barnes and McNaughton, 1985; Morris

et al., 1978). It has been shown that excitotoxic lesions of

the rat ERC impair the retention of reference memory tasks

(Levisohn and Isacson, 1991). Age-dependent deficits in

spatial memory have been shown to be related to impairments









in hippocampal "kindling" (used as a measure of neuronal

plasticity) in F-344 rats (DeToledo-Morrell et al., 1984).

The kindling phenomenon is dependent upon sequential synaptic

modification in a cascading neural system (DeToledo-Morrell

et al., 1984), and would be expected to be especially

vulnerable to age-related alterations in postsynaptic

mechanisms such as a decrease in NMDA receptors in the

entorhinal cortex. The ERC, in addition to having

connections with the hippocampus, has reciprocal connections

with a wide range of cortical association areas (Amaral and

Witter, 1989). The decrease in MK-801 binding in the ERC

could therefore also affect cortical circuitry by decreasing

the overall amount of NMDA receptor-mediated

neurotransmission.

The decrease in MK-801 binding observed in the IFCTX of

the aged animals may contribute to some alteration in the

circuitry necessary for learning and memory as well.

Cortical-striatal projections have been shown to utilize NMDA

receptors (Cherubini et al., 1988), and a decreased density

of NMDA receptors in both IFCTX and the LSTR may affect the

NMDA receptor-mediated neurotransmission between these

projections. Piggott et al. (1992) found that [3H]MK-801

binding declined with age in human frontal cortical

membranes. More recently, Serra et al. (1994) found that the

total number of binding sites for [3H]MK-801 was decreased in

the hippocampus, cerebral cortex and striatum of 18- and 24-

month-old rats, relative to 3-month-old animals. Other









groups have also shown a significant age-related decrease in

[3H]MK-801 binding in cerebral cortical and hippocampal brain

homogenates (Kitamura et al., 1992; Tamaru et al., 1991). In

another study, the ERC, IFCTX and LSTR were among various

brain regions that were shown to undergo the greatest percent

decline in NMDA-displaceable [3H]L-glutamate binding when 30-

month-old mice were compared to 3-month-olds (Magnusson and

Cotman, 1993). These regions may be some of the most

vulnerable to the effects of aging. Since these regions are

implicated in normal memory function and plasticity of the

CNS (DeToledo-Morrell et al., 1984; Gonzales et al., 1991;

Petit, 1988), an age-related decrease in NMDA receptors in

these regions suggests that some of the cellular mechanisms

encoding memory are subsequently impaired.

It was determined that the age-related decrease seen in

[3H]MK-801 binding was not a result of a decrease in neuronal

density in the ERC, IFCTX or the LSTR. Age comparisons were

made in neuronal packing densities for each of these brain

regions. As West et al. (1993) stated, the volume and/or

potential shrinkage laminarr thickness) of specific brain

regions needs to be determined in order to obtain accurate

measurements of changes that may be occurring in specific

neuronal populations. He stated that only when these

parameters are known can neuron packing density be used,

instead of total neuronal number, in a meaningful discussion

of the functional state of a neural structure (West et al,

1993). In the present study, neuronal packing density was









measured as the number of neurons per square millimeter.

Therefore, this measurement would only be valid after

determining that these brain regions did not undergo an age-

related change in overall volume (LSTR) or laminar thickness

(ERC and IFCTX) (i.e., the mm2 denominator value was not

affected by age). No significant age-dependent differences

were found in the volume of the LSTR or in the laminar

thickness of the ERC and IFCTX (Table 3-1). Neuronal packing

densities were then determined utilizing an MCID image

analysis system for the ERC, the IFCTX and the LSTR. There

was no statistically significant difference in neuronal

density measurements in any brain region examined. Therefore

the decrease seen in [3H]MK-801 binding represents a loss of

NMDA receptors within these brain regions rather than a loss

of neurons.

In this study, [3H]MK-801 binding density was expressed

in pmol/mg protein. Therefore, changes in protein

concentration in specific brain regions as a function of age

may effect these values. Burnett and Zahniser (1989)

performed a quantitative autoradiographic study that examined

age-related changes in a-1 adrenergic receptors in F-344 rat

brain. Protein levels were measured in the same tissue

sections by using a staining procedure described by Miller et

al. (1988). They found no significant differences in protein

concentration between age groups (i.e., 3- to 4-month-old,

16- to 18-month-old and 24- to 28-month-old) in the thalamus,

cerebral cortex, hippocampus, striatum, cerebellum,








brainstem, and olfactory tubercle. Since the ages and strain

of rats used in the present study were identical to Burnett

and Zahniser (1989) it is expected that total protein would

also remain unchanged as a function of age. Therefore, it is

probable that age-related differences in [3H]MK-801 binding

density in specific brain regions is due to alterations in

receptor number and not in protein concentration.

It should be noted that finding an age-related effect on

[3H]MK-801 binding in only three of eight brain regions

analyzed may reflect the inability of MK-801 to label all of

the NMDA receptors, since ligand binding can only assay

surface-presented receptors. Studies investigating mRNA

coding for the NMDA receptor is presented in Chapter 5. The

density of NMDA protein is also presented in Chapter 6.

These further studies address the hypotheses that the density

of mRNA coding for the NMDA receptor and the density of NMDA

protein change as a function of age.

These [3H]MK-801 binding experiments have determined

some of the age-dependent changes taking place in the NMDA

receptor/channel complex. The decrease in NMDA receptor

density seen in specific brain regions may account for some

of the deleterious effects of aging on learning and memory

(Barnes, 1979; Barnes and McNaughton, 1985). Gonzales et al.

(1991) studied NMDA receptor-mediated responses in the

hippocampus, cortex, and striatum of F-344 rats of various

ages (3- to 5-, 12- to 14-, and 24- to 28-months-of-age) to

determine whether aging alters some of the functional









properties of this receptor complex. They examined NMDA-

stimulated release of norepinephrine and dopamine as indices

of NMDA receptor function and found that NMDA-mediated

responses were attenuated with increasing age. In Chapter 4,

the effects of increasing age on L-glutamate's ability to

enhance [3H]MK-801 binding is studied. This approach was

taken to see if there were differences in the Emax (maximal

response or density of [3H]MK-801 bound) or EC50 (the

concentration of L-glutamate producing a half-maximal

response) as a function of age.

Many of the processes underlying learning and memory are

known to utilize, to a great extent, NMDA receptors (Danysz

et al., 1988; Morris et al., 1986). Therefore, elucidation

of age-dependent changes in NMDA receptors may help target

potential pharmaceutical interventions aimed at alleviating

age-related cognitive decline. It has been suggested that

pharmacological manipulation of glutamatergic

neurotransmission may prove beneficial for cognitive

enhancement (Ingram et al., 1994). However, the utilization

of NMDA receptor agonists may be an unsafe strategy due to

the potential neurotoxic effects associated with

overstimulation of the NMDA receptor (Cotman and Monaghan,

1988). Therefore, indirect activation through other

modulatory sites identified on the NMDA receptor complex

could be investigated, such as the co-agonist glycine site

and/or the polyamine modulatory sites.









Milacemide and D-cycloserine are two strong candidates

for agonists at the glycine site and have both been shown to

enhance memory performance in rats (Flood et al., 1992;

Quartermain et al., 1991). A relatively new class of

chemical substances, nootropics, were developed specifically

to alleviate age-related cognitive deficits and have been

shown to enhance NMDA receptor density (10 to 25%) in aged

brain (Cohen and Muller, 1992; Davis et al., 1993; Fiore and

Rampello, 1989). Phosphatidylserine is one example of a

nootropic drug. Cohen and Muller (1992) showed that chronic

treatment with phosphatidylserine ameliorated age-associated

deficits of the NMDA receptor in the forebrain of aged mice

by restoring the density of NMDA receptors to levels similar

to young animals and also by normalizing the sensitivity of

aged animals for the stimulating effects of L-glutamate and

glycine. Cohen and Muller (1993) also showed significant

enhancement (20%) of NMDA receptor density in aged mice

treated with another nootropic drug, piracetam. The kinetic

constants for L-glutamate stimulation of [3H]MK-801 binding

in the aged animals were not significantly different from

untreated young mice (Cohen and Muller, 1993). In addition,

the nootropic agent L-acetylcarnitine was shown to attenuate

the age-dependent decrease of NMDA-sensitive glutamate

receptors in the rat hippocampus (Davis et al., 1993; Fiore

and Rampello, 1989). Taken together, these studies imply

that specific pharmacological targeting of the NMDA receptor






58

may indeed prove to be a viable treatment strategy for

restoration of cognitive deficits in the aged population.














CHAPTER 4
EFFECT OF AGE ON L-GLUTAMATE STIMULATION OF [3H]MK-801
BINDING IN F-344 RAT BRAIN



Introduction


In Chapter 3, a decrease in the number of NMDA receptors

in specific brain regions in the F-344 rat was shown. Other

studies have described a similar reduction in NMDA receptors

in various species (Anderson et al., 1989; Ingram et al.,

1992; Magnusson and Cotman, 1993; Peterson and Cotman, 1989;

Serra et al., 1994; Tamaru et al., 1991; Wenk et al., 1991)

including humans (Piggott et al., 1992). In Chapter 3,

[3H]MK-801 was utilized to study the NMDA receptor/channel

complex. This antagonist binds with high affinity to a site

within the channel of the NMDA receptor complex, thus MK-801

binding will only take place if the NMDA receptor is in the

transmitter-activated (open) state (Huettner and Bean, 1988).

Channel activation by an NMDA agonist such as L-glutamate, in

the presence of glycine allows MK-801 binding to take place.

Glycine alone cannot activate the NMDA receptor. In addition

to the glycine site on the NMDA receptor/complex, there is an

allosteric polyamine recognition site. The polyamines

spermine and spermidine have been shown to increase both

maximum NMDA response amplitude (Lerma, 1992; McGurk et al.,









1990; Ransom and Deschenes, 1990; Sprosen and Woodruff, 1990)

as well as binding of MK-801 to the NMDA receptor channel

(Ransom and Stec, 1988).

L-glutamate is the endogenous transmitter for the NMDA

receptor channel complex. In this study, an evaluation of

the efficacy of L-glutamate to stimulate MK-801 binding

within the NMDA receptor/channel complex was performed by

using various concentrations of L-glutamate. These studies

examined the efficacy and potency with which L-glutamate

produces receptor activation and channel opening as a

function of increasing age by comparing dose-response curves

from 6-, 12-, and 24-month-old animals.

The hypothesis tested in these experiments was that

there would be an age-related decrease in the maximal

response (Emax) elicited by L-glutamate. It was also

hypothesized that there would not be a concomitant change in

the concentration of L-glutamate producing a half-maximal

response (EC50) as a function of age. In these experiments,

the response of the NMDA receptor was defined as the density

of [3H]MK-801 binding determined as a function of L-glutamate

concentration.

Many studies have shown that there are decreased numbers

of NMDA receptors in aged animals (Anderson et al., 1989;

Ingram et al., 1992; Magnusson and Cotman, 1993; Peterson and

Cotman, 1989; Piggott et al., 1992; Serra et al., 1994;

Tamaru et al.,1991; Wenk et al., 1991). Consequently, there

are fewer NMDA receptors available for occupation by ligand









in aged rats. Because there are fewer NMDA receptors

available for occupation by ligand in aged rats it would be

expected that there would be a decreased maximal response

([3H]MK-801 bound) in aged rats due to fewer NMDA receptor

channels available to be activated (opened) by agonist. In

addition, previous studies have shown an age-related decrease

in the functional capacity of the NMDA receptor/channel

complex as defined by a reduction in various NMDA-mediated

responses (Gonzales et al., 1991; Pittaluga et al., 1993;

Serra et al., 1994). The ability of L-glutamate to stimulate

[3H]MK-801 binding was therefore compared across varying age

groups in the F-344 rat.


Methods


Tissue Preparation


The male F-344 rats, 6-, 12- and 24-months-of-age, that

were used in these experiments were the same as those

described in Chapter 3. For a more detailed description of

the tissue preparation see Chapter 2.


r3H1MK-801 Binding Assay


For a detailed description of this protocol, see Chapter

2. Varying concentrations of L-glutamate (i.e., 0.0 pM, 0.05

pM, 0.1 pM, 0.25 pM, 0.5 pM, 1.0 pM, 2.5 pM, 5.0 pM, 10 yM,

and 20 pM) were added to the incubation step in order to

examine glutamate's ability to stimulate [3H]MK-801 binding.









A total of four slides, from each block of animals, were

analyzed per assay (n=6) for each L-glutamate concentration.


Image analysis


See Chapter 2 for a detailed description of image

analysis.


Data Analysis


The data consist of the EC50 and Emax values obtained

from binding isotherm plots. These plots were generated from

average density measurements of [3H]MK-801 binding (pmol/mg

protein) in the presence of increasing concentrations of L-

glutamate. Binding densities were obtained from bilateral

brain regions for each of the 6-,12-, and 24-month-old rats

(n=6 for each age). One "group" consisted of one 6-month-,

one 12-month-, and one 24-month-old animal mounted per series

of slides. Each [3H]MK-801 binding assay was run with the

three ages of animals mounted per slide. Four slides, from

each of these groups, were analyzed per assay per L-glutamate

concentration, with each [3H]MK-801 binding assay being

performed on one group of animals (n=6). Two slides per

assay were analyzed for non-specific binding and these values

were subtracted from each specific binding value in order to

obtain total binding. Brain regions that were analyzed

included the outer and inner frontal cortices (OFCTX, IFCTX);

the entorhinal cortex (ERC); the molecular layer of the

dentate gyrus (DG); hippocampal area CA1 stratum radiatum








(CA1); the lateral septum (LSEP); the lateral striatum

(LSTR); and the lateral thalamus (LTHAL).


Binding Isotherm Plots


INPLOT (GraphPad Software, San Diego, CA) was used to

generate linear regression plots from the density values

obtained in these experiments. Plots gave estimated EC50

(effective concentration at 50%) and Emax (maximal [3H]MK-801

binding elicited by L-glutamate) values for L-glutamate

stimulation of [3H]MK-801 binding. Estimates were then used

to generate rectangular hyperbola plots or binding isotherms.

Binding isotherm plots then yielded the true EC50 and Emax

values for each age group and brain region analyzed. As the

concentration of L-glutamate increased, there was a

concominant increase in the density of [3H]MK-801 binding

(Figure 4-1). EC50 and Emax values obtained from [3H]MK-801

binding isotherm plots for the 6-month-old animals were used

to normalize the 12- and 24-month-old animal values.


Statistical Analysis


Six blocks of animals, each block consisting of an

animal from each age group, were assessed. A different block

of 3 animals, 1 per age group, was assessed on each of 6

experimental days. The experimental design consisted of a

within-animal factor with 8 levels (brain region), a

quantitative between-animal fixed factor with 3 levels (age),

and a between-animal random blocking factor with 6 levels








(experimental day). Repeated measures analysis of variance

(ANOVA) with between-animal randomized blocks and age

considered as a linear covariate were used to model the EC50

and Emax data as a linear function of age within each brain

region and to determine if this pattern differed

significantly among brain regions. The change in EC50 and

Emax per month-of-age (i.e. the linear slope) was estimated

and assessed for statistical significance within each brain

region while blocking on experimental day. A general

comparison of these slopes among brain regions was then

performed by testing for the presence of a significant

interaction between age and brain region EC50 and Emax

effects. Mauchly's sphericity test was used to determine if

a multivariate F test (Wilks' Lambda) or a univariate F test

should be used for the age x brain region interaction test.

In assessing the validity of ANOVA assumptions, it was noted

that experimental day age profiles were relatively parallel

within each brain region and that between-subject variability

as estimated by the standard deviation was similar among age

groups within each brain region. Plots of residuals versus

predicted values, residual histograms, and residual normal

probability plots were examined to assess goodness of fit in

the ANOVA models.


Results


[3H]MK-801 binding increased as a function of L-

glutamate concentration in all the age groups and a








representative autoradiogram is presented in Figure 4-1.

Significant differences between the Emax values in nearly all

brain regions were found when middle-aged and aged animals

were compared to the young rats (Table 4-1). The only region

where Emax did not differ in middle-aged animals, when

compared to young-adults, was the OFCTX. The areas showing

the greatest percent decline in aged animals when compared to

young were the LSTR (26.4%), ERC (24.7%), and IFCTX (21.5%).

Areas showing the greatest percent decline in middle-aged

animals when compared to young were the LSTR (12.9%), LSEP

(10.3%), and IFCTX (9.3%).

No significant differences were found in the EC50 values

when comparing middle-aged and aged rats to young (Table 4-

2).










TABLE 4-1. Emax values (pmol/mg protein) from [3H]MK-801
binding isotherms.

BRAIN REGION 6 month 12 month 24 month


OFCTX 1.588 + 0.29 1.468 + 0.29 1.441 + 0.24*

IFCTX 0.959 + 0.16 0.870 + 0.17* 0.753 + 0.14*

ERC 0.830 + 0.12 0.758 + 0.14* 0.625 + 0.12*

DG 1.874 + 0.37 1.737 + 0.40* 1.669 + 0.39*

CA1 1.712 + 0.36 1.618 + 0.41* 1.537 + 0.40*

LSEP 0.951 + 0.20 0.853 + 0.19* 0.850 + 0.21*

LSTR 0.780 + 0.15 0.679 + 0.14* 0.574 + 0.12*

LTHAL
0.793 + 0.15 0.747 + 0.15* 0.665 + 0.14*


MEAN + SEM MEAN + SEM MEAN + SEM

_n=6 n=6 n=6
Asterisks (*) indicates significance when compared to 6-
month (p < 0.05 ANOVA; F-test). Abbreviations: OFCTX=outer
frontal cortex; IFCTX=inner frontal cortex; ERC=entorhinal
cortex; DG=molecular layer of the dentate gyrus;
CAl=hippocampal area CAl; LSEP=lateral septum; LSTR=lateral
striatum; LTHAL=lateral thalamus









TABLE 4-2. EC50 values (yM)
isotherms.


from [3H]MK-801 binding


BRAIN REGION 6 month 12 month 24 month


OFCTX 0.764 + 0.14 0.758 + 0.18 0.798 + 0.22

IFCTX 1.311 + 0.23 1.085 + 0.26 1.148 + 0.39

ERC 1.108 + 0.17 1.197 + 0.32 0.921 + 0.25

DG 0.951 + 0.21 0.936 + 0.25 1.040 + 0.20

CA1 0.782 + 0.17 0.660 + 0.17 0.725 + 0.20

LSEP 0.432 + 0.11 0.444 + 0.14 0.410 + 0.14

LSTR 0.578 + 0.13 0.586 + 0.15 0.599 + 0.23

0.993 + 0.20 0.861 + 0.18 0.892 + 0.23


MEAN + SEM MEAN + SEM MEAN + SEM

n___=6 n=6 n=6
No significant differences were found when 12-, and 24-
month-old rats were compared to 6-month-old rats (p > 0.05
ANOVA; F-test). Abbreviations: OFCTX=outer frontal cortex;
IFCTX=inner frontal cortex; ERC=entorhinal cortex;
DG=molecular layer of the dentate gyrus; CAl=hippocampal area
CA1; LSEP=lateral septum; LSTR=lateral striatum;
LTHAL=lateral thalamus

























Figure 4-1. Total [3H]MK-801 binding in a range of
concentrations of L-glutamate and a constant
concentration of [3H]MK-801 (10 nM). Concentration of L-
glutamate was as follows: (A), 0.0 pM; (B), 0.25 pM;
(C), 2.5 pM; (D), 10 pM. (A) Note that only background
levels of binding were present in sections incubated
without L-glutamate. (B-D) As the concentration of L-
glutamate increased, [3H]MK-801 binding increased in a
dose-dependent manner. Autoradiographs are shown in
inverse exposure so that white indicates high density of
binding and black indicates low levels of binding.






`69









Discussion


The lateral striatum (LSTR), entorhinal cortex (ERC) and

inner frontal cortex (IFCTX) were the three brain regions

that were shown in Chapter 3 to undergo an age-related

decrease in [3H]MK-801 binding when maximal levels of

glutamate, glycine and spermine were present. These regions

also showed the greatest percentage decrease in Emax values

without concomitant changes in EC50 values as a function of

age. Because both middle-aged and aged animals displayed a

significant decrease in Emax values without differences in

their EC50 values, it appears that there is an age-related

effect on the maximum density of [3H]MK-801 bound as a

function of L-glutamate concentration.

The maximal effect that L-glutamate has on stimulating
[3H]MK-801 binding is decreased in aged rats. This age-

dependent decrease in Emax may be the result of a down-

regulation of NMDA receptors in specific brain regions.

These brain regions may undergo down-regulation to protect

against excitotoxic damage. In a developmental study, Oster

and Schramm (1993) demonstrated that the process of down-

regulation of NMDA receptor activity occurs in rat cerebellar

granule cells. In these experiments, NMDA was added to cell

cultures derived from postnatal day 8 rats. This resulted in

suppression of NMDA receptor-mediated 45Ca uptake without

affecting the viability or total cell protein of the cultured

neurons. The down-regulation also rendered the neurons









resistant to NMDA toxicity. They proposed that a similar

form of down-regulation may play a role in adjusting the

activity of postsynaptic NMDA receptors following

synaptogenesis.

Jakoi et al. (1992) showed that activation of EAA

receptors in cultured hippocampal neurons caused a down-

regulation of the protein ligatin at both physiologic and

excitotoxic levels of glutamate stimulation. This down-

regulation was shown to be mediated by the NMDA receptor and

it was hypothesized that EAA receptor activation may alter

expression of NMDA receptors. This effect on NMDA receptor

expression may be a central mechanism that underlies some of

the long-lasting functional and pathophysiological effects of

EAA receptor activation on cell function (Jakoi, et al.,

1992). Therefore, it is possible that down-regulation of

NMDA receptors in the aging CNS may affect the efficacy of L-

glutamate to maximally enhance the receptor's binding

capacity in the aged brain.

Neurons bearing NMDA receptors are vulnerable to

excitotoxic injury associated with an excessive concentration

of extracellular glutamate or related agonists (Choi, 1987).

An age-related increase in basal glutamate release in mouse

striatal and hippocampal slices has been reported (Freeman

and Gibson, 1987). Low affinity glutamate uptake into rat

cerebral cortical slices (Matsumoto et al., 1982) and brain

mitochondria (Victorica et al., 1985) has been shown to be

reduced as a function of age. Furthermore, an age-related









loss in the number of high affinity glutamate transport

(uptake) sites of rat striatal (Price et al., 1981) and

cortical synaptosomes (Wheeler and Ondo, 1986) has been

reported. Taken together, these reports suggest that

extracellular glutamate levels are elevated with aging.

Therefore, the age-related down-regulation in NMDA receptors

may essentially be a functional trade-off in order to protect

cells bearing NMDA receptors from excitotoxic insult.

A recent study reported data that contradict some of the

findings here (Serra et al., 1994). Using hippocampal,

striatal and cerebral cortical brain homogenates from 3-, 18-

and 24-month-old male Wistar Kyoto rats a decrease (20 to

25%) was seen in the total number of NMDA receptors in 18-

and 24-month-old rats. No significant differences were found

in KD between young and aged rats. In addition, no

difference was seen in the sensitivity of [3H]MK-801 binding

as a function of glutamate concentration in the cerebral

cortex, striatum or hippocampus of the aged rats. However,

in the hippocampus of 18-month-old rats, glycine and

glutamate stimulated [3H]MK-801 with a higher efficacy than

in 3- or 24-month-old rats. This group speculated that the

loss of NMDA receptors in the hippocampus of 18-month-old

rats is counteracted by a physiological compensatory

mechanism. This mechanism somehow increases NMDA receptor

activity and thereby prevents a decline in cognitive function

in the "early phase of aging" (e.g., 18-months-of-age). The

compensation ceases to function adequately during, what this









group termed, the "late phase of aging" (i.e. 24 months of

age) (Serra et al., 1994). One possible explanation for the

differences in these findings is that 18-month-old animals

were not examined in the present study. Although Serra et

al. (1994) used thorough washing techniques on the rat

membrane homogenates they speculated that the sensitivity of

[3H]MK-801 binding to endogenous concentrations of agonists

and allosteric modulators of the NMDA receptor may have

affected their results. Therefore, another possible

explanation for this discrepancy is the aging-induced changes

in the brain concentrations of glycine and glutamate and/or

other modulators which could affect [3H]MK-801 binding

(Freeman and Gibson, 1987).

Gonzales et al. (1991) found an age-related decrease in

NMDA receptor function in aged animals by analyzing NMDA-

stimulated neurotransmitter release in rat cortex,

hippocampus and striatum. They showed that NMDA-stimulated

[3H]NE (norepinephrine) and [3H]DA (dopamine) release were

decreased as a function of age. In a similar study,

Pittaluga et al. (1993) showed an age-related decrease in

NMDA receptor-mediated noradrenaline release in rat

hippocampus.

It should be noted that there are advantages to using

quantitative autoradiographic analysis over brain tissue

homogenates in that it allows for a more detailed analysis of

any anatomical distributional changes occurring in the rat

CNS. However, although there are other methods that can also









test the NMDA receptor's functional capacity (e.g.

electrophysiological patch clamping techniques, molecular

biological oocyte expression systems etc.), the experiments

performed in this study tested the ability of glutamate to

enhance [3H]MK-801 binding. It should be noted that a

confounding variable possibly influencing these results is

the phenomenon of receptor desensitization, which is defined

as the lack or decline of a response as a result of previous

activation. Since brain slices in these experiments were

exposed to glutamate for long periods of time, the receptors

may have desensitized. Therefore, responses of NMDA

receptors may have been examined in the desensitized state.

The sections used in this study were rinsed for a total

of 60 minutes at 300C prior to incubation with [3H]MK-801.

Previous studies performing [3H]MK-801 binding assays have

shown that prewashing sections for as little as thirty

minutes removed substantial amounts of endogenous amino acids

since [3H]MK-801 binding following incubation was stimulated

significantly with the addition of exogenous glutamate

(Sakurai et al., 1990). Therefore, any potential effects of

desensitization were probably not seen prior to the

incubation with radiolabeled MK-801. It should also be noted

that the time-course for the amount of [3H]MK-801 bound to

reach equilibrium in the binding assay occurred prior to the

one hour incubation period utilized in these experiments (the

incubation step is when the brain sections are exposed to

[3H]MK-801 in the presence of glutamate, glycine and








spermine). Since, at equilibrium, the receptors have reached

their greatest maximal response, comparisons can be made

between the various age groups with any observed differences

in the density of [3H]MK-801 binding being attributed to the

age factor. Monaghan (1991) examined the differential

stimulation of [3H]MK-801 binding to subpopulations of NMDA

receptors and stated that since excess concentrations of

glycine and spermine were present in all incubations and both

sites analyzed could be 'activated' neither site appeared to

be in a desensitized form. However, if desensitization

occurs within minutes or seconds of glutamate application,

then age-related differences in [3H]MK-801 binding may not be

able to be interpreted as clearly due to potential age-

related differences in densensitization mechanisms. In order

to alleviate some of these potential problems in these

experiments, desensitization will be defined as a decreased

density of [3H]MK-801 (response) as a function of increased

glutamate concentration. It is important to note that the

density of [3H]MK-801 bound did not decrease after reaching

maximal levels in any of the age groups, suggesting that

desensitization may not be a significant factor in these

studies.

Consistent with L-glutamate's ability to open the ion

channel associated with the NMDA receptor, it was shown in

this study that [3H]MK-801 binding increased as a function of

L-glutamate concentration in all age groups. However, these

findings suggest that there is an effect of age on the









maximal increase in [3H]MK-801 binding induced by L-

glutamate. These age-dependent changes in the maximal

response elicited by L-glutamate may be due to some

functional alteration in the NMDA receptor/channel complex.

Hollenberg (1985) proposed that receptors can be modified by

reactions leading to either covalent or noncovalent bond

formation. In covalent modifications, reactions involving

receptor phosphorylation, disulfide-sulfhydryl exchange and

receptor proteolysis are all possible mechanisms involved in

receptor function (Hollenberg, 1985). Non-covalent

interactions would include changes in membrane potential,

receptor distribution (patching, capping), allosteric

interactions involving either protein-protein (e.g. mobile

receptor model) or small ligand cationss, anions,

nucleotides, etc.) receptor interactions, and alterations in

the membrane lipid environment (e.g. lipid methylation or

hydrolysis) (Hollenberg, 1985). All of these processes could

result in changes in the functional modification of the NMDA

receptor. In conclusion, there is probably a combination of

both an age-dependent decrease in the total number of NMDA

receptors available for binding as well as some age-related

change in the functional capacity of the remaining NMDA

receptors. This may account for some of the differences

found in the maximal increase in [3H]MK-801 induced by L-

glutamate as a function of age.














CHAPTER 5
AGE-RELATED CHANGES IN THE LEVELS OF mRNA CODING FOR SPECIFIC
NMDA RECEPTOR SUBUNITS IN THE CNS OF F-344 RATS.


Introduction


Two NMDA receptor subunit families have been cloned and

are named NMDAR1 (NR1) and NMDAR2 (NR2)(Monyer et al., 1992;

Moriyoshi et al., 1991). NMDAR1 has at least eight

alternatively spliced forms and these are differentiated from

each other by an insertion at the extracellular amino-

terminal region, deletion at two carboxy-terminal regions, or

by combinations of both (Moriyoshi et al., 1991). NMDAR1 and

its isoforms have been shown in Xenopus oocyte expression

systems to exhibit electrophysiological and pharmacological

responses characteristic of the NMDA receptor. These include

agonist and antagonist selectivity, glycine modulation,

permeability to calcium, voltage-dependent channel block by

magnesium as well as inhibition by zinc (Moriyoshi et al.,

1991).

NMDAR2 subunit family members have been shown to

potentiate the electrophysiological responses of NMDAR1 but

are not functional in homomeric configurations with each

other. NMDA-induced currents in oocytes expressing NR1 and

NR2A, NR2B or NR2C are approximately 100 times larger than

they are in oocytes expressing homomeric NR1 channels. It









should be noted that these currents more closely resemble

native NMDA receptors and thus native NMDA receptors probably

represent heteromeric configurations formed from NR1 subunits

and members of the NR2 subunit family (Monyer et al., 1992).

The NMDA receptor has been implicated in age-related

learning and memory deficits in humans as well as other

animals (Barnes, 1979; Barnes and McNaughton, 1985).

Detailed in situ hybridization analyses can now be performed

due to the recent isolation of functional cDNA clones for the

rat NMDA receptor NR1 (NMDAR1), splice variants of NR1 as

well as for the members of the NR2 subunit family (i.e.,

NR2A, NR2B, NR2C and NR2D) (Buller et al., 1993; Kutsuwada et

al., 1992; Meguro et al., 1992; Monaghan et al., 1993; Monyer

et al., 1992; Moriyoshi et al., 1991; Nakanishi, 1992). In

this study, this technique was utilized to examine mRNA

coding for the NMDA receptor subunits in the aged brain.


Methods


In situ Hybridization


Oligonucleotide probes for this study (45 mer) were

constructed from published sequences and specific to NMDAR1,

four alternatively spliced versions of NMDAR1 (NRl-not insert

1 or NRl1xx; NR1-insert 1 or NR11xx; NR1-insert 2 or NRlxlx;

and NR1-insert 3 or NRlxxl), and the members of the NMDAR2

subunit family (NR2A, NR2B, NR2C, and NR2D). The probes for

this study were graciously supplied by Dr. Daniel T.








Monaghan, University of Nebraska Medical Center, Omaha NE.

See Chapter 2 for details on the probe labeling procedure,

tissue sectioning, fixation and hybridization.


Data Analysis


Computer assisted semi-quantitative densitometric

measurements were performed as described in detail in Chapter

2. Six-month (young), twelve-month (middle-aged) and twenty-

four month (aged) F-344 rat brains were used (n=6 for each

age group). The outer frontal cortex (OFCTX), inner frontal

cortex (IFCTX), lateral septum (LSEP), lateral striatum

(LSTR), lateral thalamus (LTHAL), entorhinal cortex (ERC),

dentate gyrus (DG), and hippocampal areas CA1 and CA3 were

the brain regions analyzed in this study. Relative mRNA

density levels were obtained for each of the oligonucleotide

NMDA probes. Density values from young adults were compared

to the middle-aged and aged animals.

Statistical Analysis


Six blocks of animals, each block consisting of an

animal from each age group, were assessed. The experimental

design consisted of a within-subject factor with nine levels

(brain region), a quantitative between-subject fixed factor

with three levels (age), and each of six levels (experimental

day). Repeated measures analysis of variance (ANOVA) with

between-subject randomized blocks and age considered as a

linear covariate were used to model the density of mRNA per








oligonucleotide probe as a function of age within each brain

region and to determine if this pattern differed

significantly among brain regions.


Results

Significant age-related changes were found in NMDAR1, as
well as three of the splice variants of NMDAR1: NRl0xx,

NR11xx, and NRlxlx (Figures 5-1,5-2,5-4,5-5 and 5-7). More

specifically, NMDAR1 mRNA measured in 12-month-old rats

showed a significant decrease in the outer frontal cortex

(17.8%), inner frontal cortex (15.0%), hippocampal area CA3

(9.0%) and the lateral striatum (14.1%). In 24-month-old rat

brains, there was a significant decrease seen in all brain

regions analyzed, with the exception of the molecular layer

of the dentate gyrus. Areas showing the greatest percent

decline included the entorhinal cortex (28.1%) > lateral

thalamus (24.2%) > lateral striatum (20.7%) > outer frontal

cortex (18.6%) > inner frontal cortex (17.7%) > hippocampal

areas CA3 (14.5%) and CA1 (14.4%) > lateral septum (10.4%).

Relative mRNA density for NMDAR1 splice variant NR1Oxx

(not containing the 21 amino acid N-terminal insert) showed a

statistically significant decrease in the entorhinal cortex

(5.6%) as well as the CA3 region of the hippocampus (9.2%) in

24-month-old animals. This splice variant also showed a

significant increase in mRNA density in hippocampal area CA1

(5.5%) from 12-month-old animals. There was a significant

age-related change in relative mRNA density for two other









splice variants of NMDAR1. The NRl1xx (those NMDAR1 isoforms

containing the N-terminal insert) showed an increase in the

lateral septum (9.6%) from 12-month-old animals when compared

to 6-month-old animals (Figure 5-5). NRlxlx isoforms (those

splice variants containing the first of the two C-terminal

inserts) also showed an increase in the lateral septum (5.3%)

along with the lateral striatum (3.4%) and lateral thalamus

(4.7%) from 12-month-old animals (Figure 5-7). In contrast,

no significant differences were observed in either NR11xx or

NRlxlx mRNA density in any of the brain regions analyzed from

24-month-old animals (Figures 5-5, 5-6, 5-7, and 5-8). The

density of NMDAR1 splice variant NRlxxl (containing the

second C-terminal insert) did not show any differences in any

of the brain regions analyzed at any age (Figures 5-9 and 5-

10).

The NMDAR2 subunit family members, NMDAR2A and NMDAR2B,

did not show any age-related differences in relative density

of mRNA (Figures 5-11, 5-12, 5-13 and 5-14). NMDAR2C subunit

mRNA was not present in any of the brain regions examined.

Since this subunit is found only in the cerebellum it was

excluded from these analyses. There was no detectable

NMDAR2D mRNA present in any of the age groups. This may be

due to the fact that this subunit appears to be

developmentally regulated such that there is an apparent

shift in relative expression of NMDAR2D to NMDAR2B expression

at about six months-of-age in the rat CNS. Representative

autoradiograms are seen in Figures 5-15, 5-16 and 5-17.











NMDAR1
100-

80-
U middle-aged
60 [] aged

40

20 significant
at 95%
0
OFCtx IFCtx LSep LStr LThal
Brain Region
Figure 5-1. In situ hybridization analyses of NMDAR1 mRNA in
cortical and subcortical brain regions. All data are the mean
+ S.E.M. Asterisk indicates densitometric values
significantly different than young animals.(OFCtx=outer
frontal cortex; IFCtx=inner frontal cortex; LSep= lateral
septum; LStr= lateral striatum; LThal= lateral thalamus).

NMDAR1
100- *

80- middle-aged

_60 [] aged

40

*- significant
20 at 95%


ERC DG CAl CA3
Brain Region
Figure 5-2. In situ hybridization analyses of NMDAR1 mRNA in
hippocampus and entorhinal cortex. All data are the mean +
S.E.M. Asterisk indicates densitometric values significantly
different than young animals. (ERC= entorhinal cortex; DG=
molecular layer of the dentate gyrus; CA1 & CA3= hippocampal
areas).









NR1 Oxx


* middle-aged

D aged





*- significant
at 95%


OFCtx IFCtx LSep LStr LThal
Brain Region

Figure 5-3. In situ hybridization analyses of NMDAR1 splice
variant NR10xx mRNA. All data are the mean + S.E.M. Asterisk
indicates densitometric values significantly different than
young animals. (OFCtx=outer frontal cortex;IFCtx=inner
frontal cortex;LSep=lateral septum;LStr=lateral striatum;
LThal=lateral thalamus).


NR1 Oxx


* middle-aged

[] aged



*- significant
at 95%


ERC DG CA1 CA3
Brain Region
Figure 5-4. In situ hybridization analyses of the NMDAR1
splice variant NRl0xx mRNA. All data are the mean + S.E.M.
Asterisk indicates densitometric values significantly
different than young animals. (ERC=entorhinal cortex;DG=
molecular layer of the dentate gyrus;CA1 & CA3=hippocampal
areas).


120

100









NR1 1xx


* middle-aged

[] aged





*- significant
at 95%


OFCtx IFCtx LSep LStr LThal
Brain Region
Figure 5-5. In situ hybridization analyses of NMDAR1 splice
variant NRl1xx mRNA. All data are the mean + S.E.M. Asterisk
indicates densitometric values significantly different than
young animals. (OFCtx=outer frontal cortex;IFCtx=inner
frontal cortex;LSep= lateral septum;LStr= lateral striatum;
LThal=lateral thalamus).

NR1 1xx


* middle-aged

[ aged




*- significant
at 95%


ERC DG CA1 CA3
Brain Region
Figure 5-6. In situ hybridization analyses of the NMDAR1
splice variant NRl1xx mRNA. All data are the mean + S.E.M.
Asterisk indicates densitometric values significantly
different than young animals. (ERC=entorhinal cortex; DG=
molecular layer of the dentate gyrus;CA1 & CA3= hippocampal
areas).


120

100

80









NR1 xlx


* middle-aged

] aged.





*- significant
at 95%


OFCtx IFCtx LSep LStr LThal
Brain Region
Figure 5-7. In situ hybridization analyses of NMDAR1 splice
variant NRlxlx mRNA. All data are the mean + S.E.M. Asterisk
indicates densitometric values significantly different than
young animals. (OFCtx=outer frontal cortex;IFCtx=inner
frontal cortex;LSep=lateral septum;LStr=lateral striatum;
LThal=lateral thalamus).


NR1 xlx


120

100


M middle-aged

[ aged




*- significant
at 95%


ERC DG CAl CA3
Brain Region
Figure 5-8. In situ hybridization analyses of the NMDAR1
splice variant NRlxlx mRNA. All data are the mean + S.E.M.
Asterisk indicates densitometric values significantly
different than young animals. (ERC=entorhinal cortex;
DG=molecular layer of the dentate gyrus;CA1 & CA3=hippocampal
areas).


120

100

80










NR1 xxz


* middle-aged

] aged



*- significant
at 95%


OFCtx IFCtx LSep LStr LThal
Brain Region
Figure 5-9. In situ hybridization analyses of NMDAR1 splice
variant NRlxxl mRNA. All data are the mean + S.E.M. Asterisk
indicates densitometric values significantly different than
young animals. (OFCtx=outer frontal cortex;IFCtx=inner
frontal cortex;LSep=lateral septum;LStr=lateral striatum;
LThal=lateral thalamus).


NR1 xxi


* middle-aged

[] aged



*- significant
at 95%


ERC DG CAl CA3
Brain Region
Figure 5-10. In situ hybridization analyses of the NMDAR1
splice variant NRlxxl mRNA. All data are the mean + S.E.M.
Asterisk indicates densitometric values significantly
different than young animals. (ERC=entorhinal cortex;DG=
molecular layer of the dentate gyrus;CA1 & CA3=hippocampal
areas).


120

100

80












120

100


NMDAR2A


* middle-aged
] aged





*- significant
at 95%


OFCtx IFCtx LSep LStr LThal
Brain Region
Figure 5-11. In situ hybridization analyses of NMDAR2A mRNA.
All data -are the mean + S.E.M. Asterisk indicates
densitometric values significantly different than young
animals. (OFCtx=outer frontal cortex;IFCtx=inner frontal
cortex;LSep= lateral septum;LStr= lateral striatum;LThal=
lateral thalamus).
NMDAR2A


* middle-aged

] aged



*- significant
at 95%


ERC DG CA1 CA3


Brain Region

Figure 5-12. In situ hybridization analyses of NMDAR2A mRNA.
(ERC=entorhinal cortex; DG=molecular layer of the dentate
gyrus;CA1 & CA3=hippocampal areas). All data are the mean +
S.E.M. Asterisk indicates densitometric values significantly
different than young animals.












* middle-aged

[] aged





*- significant
at 95%


OFCtx IFCtx LSep LStr LThal
Brain Region
Figure 5-13. In situ hybridization analyses of NMDAR2B mRNA.
All data are the mean + S.E.M. Asterisk indicates
densitometric values significantly different than young
animals. (OFCtx=outer frontal cortex;IFCtx=inner frontal
cortex;LSep= lateral septum;LStr=lateral
striatum;LThal=lateral thalamus).


NMDAR2B


* middle-aged

] aged



*- significant
at 95%


ERC DG CA1 CA3


Brain Region
Figure 5-14. In situ hybridization analyses of NMDAR2B mRNA.
All data are the mean + S.E.M. Asterisk indicates
densitometric values significantly different than young
animals. (ERC=entorhinal cortex;DG= molecular layer of the
dentate gyrus;CAl & CA3=hippocampal areas).


120

100

80































Figure 5-15. The distribution of mRNA specific to NMDAR1
in A) 24-month-old; B) 12-month-old; and C) 6-month-old
rat brain.







Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EJI9P8GIX_Z7R4Q2 INGEST_TIME 2012-02-20T21:13:01Z PACKAGE AA00009034_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES