Citation
The relationship among eternal calcium, internal free calcium and neurite growth in rat hippocampal neurons and neuroblastoma cells

Material Information

Title:
The relationship among eternal calcium, internal free calcium and neurite growth in rat hippocampal neurons and neuroblastoma cells
Creator:
Frank, James Andrew
Publication Date:
Language:
English
Physical Description:
xi, 66 leaves : illustrations (some color) ; 29 cm

Subjects

Subjects / Keywords:
Calcium in animal nutrition ( lcsh )
Neurons -- Growth ( lcsh )
Calcium in animal nutrition ( fast )
Neurons -- Growth ( fast )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Bibliography:
Includes bibliographical references.
General Note:
Submitted in partial fulfillment of the requirements for the degree, Master of Arts, Department of Integrative Biology.
Statement of Responsibility:
by James Andrew Frank.

Record Information

Source Institution:
University of Colorado Denver
Holding Location:
Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
26018917 ( OCLC )
ocm26018917
Classification:
LD1190.L45 1991m .F73 ( lcc )

Downloads

This item has the following downloads:


Full Text
THE RELATIONSHIP AMONG EXTERNAL CALCIUM, INTERNAL FREE
CALCIUM AND NEURTTE GROWTH IN RAT HIPPOCAMPAL NEURONS
AND NEUROBLASTOMA CELLS
by
James Andrew Frank
B.A., University of Colorado, 1989
A thesis submitted to the Faculty of the
Graduate School of the
University of Colorado in partial fulfillment
of the requirements for the degree of
Master of Arts
Department of Biology
1991


This thesis for the Master of Arts
degree by
James Andrew Frank
has been approved for the
Department of
Biology
by
Gerald Audesirk
Teresa Audesirk
V- 2 9/
Date
Alan Brockway


Frank, James Andrew (M.A., Biology)
The Relationship Among External Calcium, Internal Free Calcium and Neurite
Growth in Rat Hippocampal Neurons and Neuroblastoma Cells
Thesis directed by Associate Professor Gerald Audesirk
Intracellular free calcium ([Ca2+]i) plays a critical role in the growth of neurites.
A change in the gradient for calcium across the plasma membrane may result in
altered influx of the cation. To maintain ail optimal [Ca2+]i for neurite growth, cells
may have to overcome the difference in Ca2+ influx. This study was designed to
explore the effects of altered external calcium concentration ([Ca2+]0) on [Ca2+]i and
neurite growth in rat hippocampal neurons, mouse peripheral nervous system
neuroblastoma N1E-115 and rat central nervous system neuroblastoma B50.
Neurite initiation and elongation were quantified in each cell type in six different
extracellular calcium concentrations (OX, 0.25X, 0.5X, IX, 2X, 4X, where IX
was 1.8 mM Ca2+). Using Fura-2 digital imaging analysis, [Ca2+]i was determined
in each [Ca2+]0 for N1E-115 cells and rat hippocampal neurons. The data showed
that initiation of neurites was significantly reduced in rat hippocampal neurons
cultured in media with calcium levels other than IX. The two transformed cell lines
showed no differences from IX. Elongation of neurites was unaffected in all three
cell types by altered [Ca2+]0. Analysis of [Ca2+]i showed that in rat hippocampal
neurons, [Ca2+]i changed monotonicly when [Ca2+]p changed with a plateau near
4X. N1E-115 cells showed no differences in [Ca2+]i from IX values in any of the
altered [Ca2+]0. N1E-115 cells may be able to regulate [Ca2+]i more efficiently than
rat hippocampal neurons when challenged with an alteration in the influx of Ca2+.


Also, neurite elongation seems to be less sensitive than neurite initiation to changes
in [Ca2+]i in rat hippocampal neurons.
The form and content of this absract are approved. I recommend its publication.
Signed
Gerald Audesirk
IV


To Pooka


CONTENTS
Figures....................................................viii
Tables........................................................x
Acknowledgements.............................................xi
CHAPTER
1. INTRODUCTION................................................. 1
Calcium Homeostasis....................................... 3
Regulation of Calcium by the Plasma Membrane..............4
Regulation of Calcium by Organelles...................... 7
Calcium in the Cytosol....................................9
The Role of Calcium in Neurite Growth......................12
Initiation...............................................12
The Role of Calcium in Growth Cone Motility
and Neurite Elongation.................................. 15
The Effects of Abnormally High or Low Calcium
Concentrations on Elongation and Initiation............... 19
The Toxic Effects of Increased
Internal Calcium Concentration...........................19
The Toxic Effects of Decreased
Internal Calcium Concentration.......................... 20
Fura-2 AM Ester as a Tool to Study Internal
Calcium Concentration .....................................22
Summary of Experiments.....................................23
Research Hypothesis......................................23
Experiment...............................................26


2. EXPERIMENTAL PROCEDURES..................................... 27
Culturing Media............................................27
Cell Culture.............................................. 28
Analysis of Survival, Initiation and Elongation............31
Analysis of Intracellular Calcium Levels with Fura-2.......34
Statistical Analysis.......................................36
3. EXPERIMENTAL RESULTS.........................................37
Survival, Neurite Initiation and Elongation................37
Rat Hippocampal Neurons..................................37
N1E-115 Cells............................................43
B50 Neuroblastoma Cells..................................46
The Effects of Changes in Extracellular Calcium
Concentration on Intracellular Calcium Concentration.......49
Rat Hippocampal Neurons..................................49
N1E-115 Neuroblastoma Cells..............................50
4. DISCUSSION...................................................53
Culturing Experiments......................................54
Survival.................................................54
Initiation............................................. 54
Elongation...............................................56
Fura-2 Studies............................................ 58
Causes of Differential Regulation of Internal
Calcium Concentration in Two Cell Types....................58
Further Study..............................................60
LITERATURE CITED...................................................62
vii


FIGURES
Figure
1.1. The involvement of calcium in the
growth of neurites................................................ 2
1.2. The major pathways of calcium
regulation in a nerve cell........................................ 4
1.3. Characteristics of voltage-sensitive
calcium channels.................................................. 5
1.4. Cytosolic and membrane bound proteins
that are affected by calcium......................................11
1.5. Optimal calcium concentrations for initiation,
elongation and membrane vesicle binding...........................16
1.6. Fura-2 calcium chelator.......................................... 22
1.7. Graph of excitation intensities of Fura-2 AM..................... 23
1.8. Quantex calcium imaging system.................................. 24
3.1. Rat hippocampal neurons and
N1E-115 cells in culture......................................... 39
3.2. The effect of calcium on survival and neurite
initiation in rat hippocampal cells at 48 hours...............41
3.3. The effect of calcium on the number neurites/cell and mean
axon length in rat hippocampal neurons at 48 hours............42
3.4. A comparison of neurite initiation and mean
neurite length in rat hippocampal cells....................... 43
3.5. The effect of calcium on survival and neurite
initiation in N1E-115 cells at 48 hours....................... 44
3.6. The effect of calcium on the number neurites/cell and mean
neurite length in NIE-115 at 48 hours......................... 45
3.7. The effect of calcium on survival and neurite
initiation in B50 cells at 48 hours...........................47


3.8. The effect of calcium on the number neurites/cell and mean
neurite length in B50 cells at 48 hours.....................48
3.9. A ratioed image of a rat hippocampal cell with neurites .... 50
3.10. Ratioed images of rat hippocampal cells
cultured in 4X and OX MEM...................................51
3.11. A ratioed image of an N1E-115 cell in
IX DMEM-F12................................................. 52
3.12. Internal calcium concentration versus external calcium
concentration in rat hippocampal neurons and N1E-115 cells ... 52
IX


TABLES
Table
2.1. Calcium Concentrations Used for Culturing......................28
2.2. Summary of Culturing Media by Cell Type........................30
3.1. Rat Hippocampal Culturing Data.................................40
3.2. N1E-115 Culturing Data.........................................40
3.3. B50 Culturing Data.............................................40
3.4. Internal Calcium Concentration Resulting From External
Calcium in Rat Hippocampal Neurons and NIE-115 Cells ... 49
x


ACKNOWLEDGEMENTS
I would like to thank Dr. Gerald Audesirk for his valuable suggestions
regarding this manuscript and for his guidance during the course of this project. I
also thank Drs. Teresa Audesirk and Alan Brockway for critically reading earlier
drafts of this paper and offering helpful comments. Finally, I extend my gratitude
to Leigh Cabel-Kluch, Charles Ferguson and David Shugarts for technical
assistance with this work. This research was supported by grants from the
Environmental Protection Agency and the National Institutes of Health.
xi


CHAPTER 1
INTRODUCTION
Calcium is a ubiquitous second messenger found in every type of cell. In
neurons, cytoplasmic calcium plays an important role in nearly every facet of
nervous system development and regulation. Calcium is particularly important in
neurite growth. The growth of neurites (axons and dendrites) has been described
as having three steps: initiation from the cell body, neuronal growth cone motility
and neurite shaft elongation (Audesirk et al., 1990).
Initiation from the cell body involves the formation of a new growth cone.
Growth cones are typically found on the leading end of neurites and can assume
several different morphologies that correspond to the type of neurite that is
developing. During neurite growth, it is the motility of actih microfilaments in the
filopodia of these cones that pulls a neurite along, while the shaft pushes.
Calcium regulates the motility of the microfilaments in the growth cone and the
polymerization of tubulin for microtubules; these are required for neurite growth. It
is believed that relatively high calcium levels in the growth cone lead to maximum
cone motility, while relatively low calcium levels promote maximum neurite
elongation (Figure 1.1).
Neurite initiation, elongation and growth cone motility are all dependent on the
concentration of free calcium inside the cell ([Ca2+]i). [Ca2+]i is the amount of
unbound calcium within the cytosol of the cell. The total concentration of calcium


includes calcium that is bound to proteins and that which is sequestered in
organelles. The continued study of the role of intracellular calcium will provide
valuable insight into the development and maintenance of the nervous system (Mills
and Kater, 1990).
This introduction will first address the regulation of [Ca2+]i, including
regulation by membrane proteins and enzymes in the cytosol. Next, the role of
calcium in neurite growth will be discussed in detail. Finally, the hypothesis of this
research will be presented and the specific experiments will be outlined.
2


Calcium Homeostasis
Maintenance of a relatively constant [Ca2+]i (homeostasis) is of vital importance
to the functioning of a cell. The regulation of [Ca2+]i involves the integration of
several biochemical pathways within a neuron (Mills and Kater, 1990). Studies
have shown that neurons have a great capacity to regulate [Ca2+]i, but that not all
cell types use the same mechanisms to achieve homeostasis (Cohan, 1987; Tank et
al., 1988; Thayer et al., 1988). The efficiency of regulatory mechanisms within a
cell is demonstrated by the fact that [Ca2+]i is about 100 nM to 200 nM in most
mammalian cells, while the extracellular calcium concentration ([Ca2+]0) is normally
between 1 mM and 5 mM (Rawn, 1990). The [Ca2+]i of a cell tends to oscillate
mildly around a set point (Carafoli, 1987).
Not all cell types maintain the same [Ca2+]i. For example, embryonic rat
hippocampal neurons maintain a calcium concentration of 45 nM to 70 nM (Mattson
et al., 1990) while the mouse neuroblastoma cell N1E-115 keeps calcium levels
between 100 nM and 200 nM (Pang et al., 1990). Additionally, cells of the same
type may not regulate calcium with the same efficiency throughout their lives. Wahl
et al. (1989) demonstrated that 8-day-old cerebral cortex cell cultures were unable to
expel a large influx of calcium, but 2-day-old cells in culture could.
In light of the above findings, it is apparent that the level of calcium within a cell
depends on many factors. The degree and efficiency to which specific regulatory
mechanisms are used is probably the single most important factor in determining
[Ca2+]i and maintaining cell viability. The manner in which [Ca2+]i is controlled by
the plasma membrane, cellular organelles and cytosolic proteins will now be
3


discussed (Figure 1.2). The biochemical role of calcium in normal cell function
will also be addressed.
Figure 1.2. The major pathways of calcium regulation in a nerve celL
Plasma membrane (PM): a, represents the VSCC; b, RMCC influx; c, calcium ATPases that pump
Ca?+ out of the cell; d, the sodium/calcium exchanger e,Ca2+ATPases that pump Ca2+into the
endoplasmic reticulum (ER); f, receptor mediated Ca2+ channels responsible for efflux from the ER;
g, calcium uniporter that only transports calcium into the mitochondria; h, inner mitochondrial
membrane @MM) sodium/calcium exchanger, i, cytosolic molecules that bind calcium (horn
Nicholls, 1986).______________________________________________________________________________
Regulation of Calcium bv the Plasma Membrane
Although calcium in the cytoplasm of a cell can originate from the extracellular
medium or from intracellular stores, the majority of calcium that enters the cell
cytoplasm comes from outside the cell due to the large gradient for influx (Carafoli,
1987). For this reason, the plasma membrane is the most important regulatory
barrier for calcium.
Calcium Channels. Calcium channels are ion channels which, when activated,
predominatly allow calcium to piass through the membrane along its gradient. The
plasma membrane contains two general categories of calcium channels: voltage-
PM
EVEM
4


sensitive calcium channels (VSCCs) and receptor-mediated calcium channels
(RMCCs). Most of the calcium that enters the cell through the plasma membrane
comes through these channels (Nicholls, 1986).
The VSCCs can be subdivided into long-lasting (L), transient (T) and neither L
or T (N) types. The names correspond to the length of time the channel conducts a
calcium current in response to membrane depolarization. L-type channels have a
relatively high threshold and tend to have long-lasting currents during a given
depolarization, while T-type channels have a low threshold and are rapidly
inactivated. N-type channels have an intermediate threshold voltage and have a
current conduction duration that is in between L apd T. Therefore, more calcium
enters a cell through L-type than N- or T-type channels. (Figure 1.3).
T L
mV -35 + 10
100 -40

+250
o < Q.
-500
0 ms 1000 u me 1000 ms
Figure 1.3. Characteristics of voltage-sensitive calcium channels. In the the upper left is a graph of a voltage stimulus used to activate a typical T-type calcium current. On the lower left is a typical image of the T-type channel current as seen dunng a patch clamp experiment On the right is a typical L-type channel current (below) and the stimulus used to activate it (aboveMSilver et al., 1990). Note that the T channel has a lower threshold and is inactivated quickly. The stimulus voltage is in millivolts (mV), the current is irt oicoamDers (dA) and the time is in milliseconds (ms).
VSCCs are found all over the cell body and synaptic endings of neurites.
Channels with the general characteristics of those listed above have been found in
5


the neuroblastoma cell line N1E-115, rat hippocampal cells and other cell types
(Anglister et al., 1982; Bean, 1989; Ozawa et al., 1989; Silver et al., 1990; Toselli
and Taglietd, 1990). The role of specific calcium channels in neurite development
will be discussed later.
When bound by a ligand, receptor-mediated calcium channels (RMCCs)
undergo changes in conformation that lead to the opening of a calcium channel.
RMCCs are found on the endoplasmic reticulum (ER) and calcisomes (small single-
membrane organelles that are produced by the ER), as well as the cell body plasma
membrane (Rasmussen, 1989). One type of RMCC is the N-methyl-D-aspartate
(NMD A) receptor that opens a calcium channel in response to glutamate binding in
vivo. These channels are found on rat hippocampal pyramidal neurons and other
neuron types (Siesjo, 1990).
Calcium ATPases. Ca2+ATPases hydrolyze ATP to provide energy to move
calcium across a membrane. These transporters, which are found on the plasma
membrane and the ER, pump calcium out of the cytosol, resulting in a lower
[Ca2+]i. Calcium ATPases hive been described as the most important mechanism
available to a cell to lower [Ca2+]i (Miller, 1988). Obviously, the action of these
transporters is directly related to the amount of ATP within a cell. If cell
metabolism is decreased, the activity of the pumps will decrease and [Ca2+]i will
increase.
Sodium/Calcium Exchanger. This transporter is found predominantly on the
plasma membrane, but can also be found on the inner mitochondrial membrane.
The stoichiometry of the exchanger is 3 Na+ to one Ca2+, but hydrogen ions and
other monovalent cations can also be exchanged for calcium across a membrane
with this protein. Further, the affinity of the exchanger for certain ions is affected
6


by phosphorylation. The affinity for calcium increases tenfold and the affinity for
sodium increases twofold with the addition of ATP in vivo, probably due to
calmodulin-dependent phosphorylation.
The driving force for the exchanger is the large sodium gradient across the
plasma membrane. Theoretically, the sodium gradient provides enough energy to
maintain a [Ca2+h of about 100 nm, but the system works too slowly to overcome
the large calcium influx that occurs during depolarization in a neuron. Also, the
exchanger is electrogenic and would seem to have limits based on the membrane
potential (Cohan and Kater, 1987). Laboratory experiments have confirmed that
the exchanger could not be solely responsible for maintaining [Ca2+]i (Carafoli,
1987). The most unique feature of the exchanger is that it can move in both
directions, depending on the relative concentrations of calcium and sodium. This
exchanger is probably more important in muscle than in neurons (Nashchen, 1985).
Regulation of Calcium bv Organelles
Cytosolic calcium concentration is influenced by regulatory proteins on the
membranes of intracellular organelles; specifically, the endoplasmic reticulum (ER)
and the inner mitochondrial membrane are involved in the regulation of [Ca2+]i.
Other organelles, including the nucleus, have about the same calcium concentration
as the cytosol (Meldolesi and Volp6,1988).
Endoplasmic Reticulum. As previously mentioned, the ER membrane contains
Ca2+ ATPase and RMCC proteins to regulate the concentration of calcium in the ER
and in the cytosol. The ER and calcisomes, which are single membrane organelles
produced by the ER, usually have veiy high calcium concentrations relative to the
cytoplasm and take up. Ca2+ in an ATP-dependent manner. The ER is perhaps the
7


only organelle that plays a crucial role in maintaining the [Ca2+]i of a healthy cell
because it can temporarily sequester large amounts of calcium in the event of influx
(Meldolesi and Volpd, 1988; Nahorski, 1988). Despite this ability, the importance
of the ER is far outweighed by the regulatory capacity of the plasma membrane
(NichoUs, 1986).
The release of Ca2+ from the ER is initiated with the hydrolysis of
phosphotidylinositol phosphate (PIP) by G proteins or by phospholipase C. PIP, a
phospholipid in the membrane, is hydrolyzed to form inositol trisphosphate (IP3), a
fatty acid, and diacylglycerol (DAG), a diglyceride. IP3 binds to RMCCs on the
ER, resulting in the opening of the calcium channels and the efflux of calcium into
the cytosol along its gradient (Meldolesi and Volpd, 1988) (Figure 1.4). Calcium in
the cytosol is known to activate phospholipase C and G proteins via calcium
dependent protein kinases. Both of these proteins can also be activated by the
binding of hormones to the plasma membrane.
Mitochondria. The inner membrane of the mitochondrion (IMM) holds two
types of calcium transporters: the sodium/calcium exchanger, which has already
been discussed, and the membrane potential-dependent uniporter. This uniporter is
exclusive to the IMM and transports C electrical gradient that exists between the cytosol and the mitochondrial matrix. The
intermembrane space of the mitochondrion is quite positively charged relative to the
matrix, which is due, in part, to the hydrogen ion pumping activity of cytochromes
and reductases on the IMM. This voltage gradient provides energy for positively-
charged calcium ions to flow into the matrix (NichoUs, 1986).
Although mitochondria have a high concentration of calcium relative to the
cytosol, they play only a smaU role in regulating [Ca2+]i because the Na+/Ca2+
8


exchanger is the only IMM protein capable of transporting calcium out of the
mitochondria. Sodium levels iri the cytosol and matrix are usually not very
different; therefore, little energy is available for the extrusion of calcium into the
cytosol. Also, it has been noted that mitochondrial calcium levels increase only
when cytosolic calcium concentrations get very high (Nicholls, 1986).
It should be noted that calcium plays an important regulatory role within the
mitochondrion in that it activates pyruvate dehydrogenase and other citric acid cycle
and oxidative phosphorylation dehydrogenases that are involved in energy release.
If calcium levels in the matrix become abnormal, energy production is decreased
(Rawn, 1990).
Calcium in the Cvtosol
Within the cytosol, the calcium concentration is regulated to a small extent by
soluble proteins; however, the majority of these proteins triggers biochemical
events in the cell when bound to calcium (Kennedy, 1988; Koike et al., 1989). The
next section will address the cytosolic proteins that bind calcium and the
biochemical effects that result from the binding.
Proteins Affected bv Calcium. Calcium can enter the cytoplasm through the
plasma membrane or horn the ER. Cytosolic calcium, regardless of origin, can
bind with several proteins in the cytosol or on the inside of the plasma membrane
and regulate their action (Figure 1.4). The two types of membrane-bound proteins
which are influenced by [Ca2+]i are channels and enzymes.
There are at least two types of ion channels that are regulated by calcium. The
first type consists of voltage-sensitiVe calcium channels (VSCCs). As explained
previously, VSCCs allow calcium to pass through the plasma membrane in
9


response to depolarization of the cell. Some VSCCs are inactivated by calcium.
The other type of ion channels includes calcium channels, as well as channels for
other ions, such as potassium, chloride and hydrogen. Unlike VSCCs, these
channels are opened in response to calcium binding (Bean, 1989).
As well as influencing channels, calcium can also affect membrane-bound and
cytosolic enzymes. On the plasma membrane, calcium activates enzymes such as
phospholipase C. When phospholipase C is activated, it can hydrolyze PIP to form
IP3, which binds to a receptor mediated calcium channel on the ER, causing the
release of calcium into the cytosol. This pathway can result in a positive feedback
loop in which [Ca2+]i increases (Kennedy, 1988; Siesjo, 1990).
In the cytosol, calcium can bind with other enzymes and enzyme regulators,
such as protein kinase C, calmodulin, paravalbumin and calpain, which will now
be discussed. The binding of calcium activates protein kinase C to phosphorylate
other neuronal proteins, including ion channels, resulting in the alteration of
electrical excitability in the cell. Additionally, protein kinase C phosphorylates
proteins involved in the regulation of synaptic transmission.
The second cytosolic protein, calmodulin, is an enzyme regulator and has four
binding sites for calcium that exhibit positive cooperativity with respect to the
binding constants (Kd) for each site. There are at least four important functions of
calcium-bound calmodulin. When calcium binds to calmodulin, a conformational
change occurs that allows calmodulin to bind to and activate adenylate cyclase,
which synthesizes cyclic AMP from ATP. Calmodulin also activates cyclic
nucleotide phosphodiesterase to hydrolyze cyclic AMP. This is significant because
in heart muscle, Ca2+ATPase is directly phosphorylated by cyclic AMP-dependent
10


PM
Phospholipase
/ Calpain
DAG
| Calcineurin
CAM Kinase II
[(^a 2+]Calmodulin-
Adenylate
Cyclase
- Phospho-
diesterase
Figure 1.4. Cytosolic and membrane bound proteins that are affected by [Ca2+]i.
All arrows indicate direct or indirect activation with the exception of diacylglvcerol (DAG) that
inactivates C Kinases. Also, the binding of IP3 to the ER causes an efflux of calcium into the
cytosol (from Kennedy,1988).
protein kinase causing calcium to be pumped out of the cell (Neyes et al., 1985). It
is possible that this may also occur in neurons. Another function of calmodulin is
to activate calmodulin-dependent kinase II, which can then phosphorylate a
Ca2+ATPase, resulting in the decrease of [Ca2+]i. Calmodulin-dependent kinase II
is also important in the modification of ion channels involved in long-term
potentiation within pyramidal cells of the hippocampus. The fourth function of
calcium-bound calmodulin is to activate calcineurin, which inactivates calcium-
dependent L-type VSCCs in pyramidal neurons (Kennedy, 1988; Neyes et al.,
1985).
Another cytosolic enzyme, paravalbumin, is probably more important in muscle
cells than neurons. Paravalbumin binds calcium in the cytoplasm and may go on to
activate proteases. The fourth enzyme is the protease calpain, which is also
activated when Ca2+-bound. Calpain is important in the regulation of cytoskeletal
architecture. Both paravalbumin and calpain are found in very small quantities in
il


the cytosol of neurons and some membrane-bound forms also exist (Kennedy,
1988; Neyes et al., 1985; Rawn, 1990; Scott et al., 1990).
All four of these cytosolic proteins regulate [Ca2+]i to a very small degree, with
the exception of the proteins that allow calcium into or out of the cell when Ca2+-
activated. Regulation may be a misnomer, but the amount of calcium bound to
cytosolic proteins is significant enough to warrant notice. For the most part,
however, it is the regulatory activities of the plasma membrane and organelles that
attempt to maintain a balance in the calcium-mediated reactions listed above. In
summary, the proteins most important in neuronal calcium regulation are the
calcium-dependent VSCCs and RMCCs, calmodulin-dependent kinase n,
phopholipases, calmodulin and protein kinase C (Kennedy, 1988).
The Role of Calcium in Neurite Growth
As previously mentioned, neurite growth has three steps; initiation from the cell
body, neuronal growth cone motility and neurite shaft elongation (Audesirk et al.,
1990). Internal free calcium concentration may affect each of these steps, but is
probably not the only intracellular messenger involved in their regulation.
Initiation
Neurite initiation can be defined as the formation of a new growth cone. Two
ideas dominate current theories of initiation. The first is that an influx of calcium
into the soma of a neuron is necessary for the development of a neurite. According
to this hypothesis, this calcium can then interact with other intracellular messengers,
resulting in a change in cellular morphology. The second idea is that a calcium
influx is not required to trigger neurite outgrowth and that other biochemical events
12


that may involve pre-existing calcium stores produce a new neurite. Although both
of the mechanisms discussed above are calcium-dependent, the dependence on
extracellular calcium for neurite development may be different among cell types and
neurite types. Some cells may incorporate both of the above mechanisms in the
initiation of different neurite types (Kater et al., 1990).
Calcium Influx and Initiation. Many stimuli that regulate neuromorphology
exert their effects by changing [Ca2+]j (Mills and Kater, 1990). It has been
observed that a transient increase in [Ca2+]j that occurs through VSCCs is required
for neurite initiation in Xenopus embryonic cells (Holliday and Spitzer, 1990).
Other studies have indicated that an influx of calcium through L-type VSCCs may
be required for initiation of certain neurites in N1E-115 cells, embryonic chick
neurons and superior cervical ganglion cells (Audesirk et al., 1990; Rogers and
Hendry, 1990). This influx may lead to the formation of a new growth cone and,
thus, result in neurite initiation from the soma.
This type of calcium influx may also be involved in neurite sprouting since
increased calcium could lead to increased microfilament motility, the activation of
other cellular messengers and the formation of a new growth cone (Kruskal et al.,
1986). L-type VSCCs have been found to cluster on growth cone membranes,
creating calcium hotspots in the cones (Silver et al., 1990). Within the growth
cone, calcium promotes the actin microfilament motility that is responsible for the
pulling action of the filopodia on the cone. A particular neuron may have some
growth cone types that require an influx of calcium and others that do not (Kater et
al., 1990).
The pulling action of the actin microfilaments depends on ATP hydrolysis,
calcium and certain kinases for motile force. Calcium-dependent gelsolin proteins
13


act to stabilize actiii filaments by capping the filament ends, thereby preventing
subunit depolymerization or unnecessary polymerization. Many acdn binding
proteins involved in the movement of the filaments and microtubule-associated
proteins are also modulated by calcium dependent kinases (Smith, 1988). Too
much or too little calcium would impede proper growth cone development in at least
some growth cone types (Kater et al., 1990).
A transient increase in [Ca2+]i may not be required to trigger initiation in all cell
types (Tolkovsky et al., 1990). Several researchers have reported that cyclic AMP
promotes neurite initiation in embryonic and transformed rat cells in the absence of
an influx of calcium (Mann et al., 1989; Reboulleau, 1986; Rydel and Greene,
1988; Schubert et al., 1974). Kater and Mattson (1988) reported that cAMP along
with protein kinase C may cause initiation directly or via a pathway involving
calcium in the absence of a large calcium influx in rat hippocampal neurons. These
molecules probably regulate the formation of cytoskeletal components. One
possible explanation of these findings is that dependence on calcium is different
among different neurite types in the same cell. For example, in pryamidal rat
hippocampal neurons, axons develop in areas within the cell with lower calcium
concentrations than those in which dendrites develop (Mattson et al., 1990).
Because several morphological types of growth cones have been observed, it is
reasonable to assume that their dependence on calcium arid other cellular
messengers is also variable (Kater et al., 1990). Calcium is probably involved in
the initiation of all neurites, but an influx or transiently-enhanced gradient may not
be required. (Duprat and Kan, 1981).
Recently, it has been reported that concentration gradients of calcium within a
cell, rather than a Ca2+ influx, are necessary for axon genesis. In this study,
14


Mattson et al. (1990) caused a calcium influx into the soma of rat hippocampal
pyramidal cells by transecting an existing axon, 'they noted that when [Ca2+]i
increased from about 50 nM to 130 nM near the transection site, axon initiation was
inhibited, but in other areas of the cell where the [Ca2+]i was relatively low, axon
initiation was enhanced. This would indicate that a brief, enhanced calcium
gradient in the cell causes axon initiation in rat hippocampal cells. This may be
related to previous findings by the group that demonstrated that axons had a lower
[Ca2+]i than dendrites. It could be that this lower calcium level in axons facilitated
longer, faster elongation (Mattson et al., 1990).
The Role of Calcium in Growth Cone Motility and Neurite Elongation
After initiation of the growth cone occurs, the second and third steps of neurite
growth are neuronal growth cone motility and neurite shaft elongation. These steps
may occur simultaneously; the movement of the growth cone pulls the neurite
along, while the microtubules in the shaft push it
Calcium Hypothesis. The so-called calcium hypothesis for neurite growth
states that there are different optimal calcium levels for both growth cone motility
and neurite elongation (Figure 1.5). High concentrations of calcium relative to the
cytosol promote maximum growth cone motility by stimulating the actin
microfilaments contained within the cone. This type of motility may enhance
neurite growth and cause the formation of new growth cones. If [Ca2+]j becomes
too high or too low, however, growth cone motility will decrease (Kater et al.,
1990). This could lead to a decrease in neurite initiation; this was not tested,
however (Figure 1.5).
15


Concentrations of calcium that are lower than that of the growth cone afford
maximum neurite shaft elongation because calcium inhibits the polymerization of
microtubules that are required to form the cytoskeleton of a developing neurite
(Audesirk et al.t 1990; Cohan and Kater,1987; Kater and Mattson, 1988). If
[Ca2+]i was too high, elongation would be inhibited (Figure 1.5).
Percent
of
Maximum
[Ca2+10
Figure 1.5. Optimal calcium concentrations for initiation, elongation and vesicle binding to the
membrane in a hypothetical neuron.
The mechanism of the inhibition of polymerization may involve the interaction of
calcium, calmodulin and microtubule associated proteins. Similarly, it is possible
that abnormally low [Ca2+]i may also inhibit neurite elongation (Manalan and Klee,
1984).
Finally, the calcium hypothesis also states that there is an optimal calcium level
for the fusion of vesicles to the plasma membrane. If [Ca2+]i is out of the optimal
range, initiation and elongation will decrease (Figure 1.5). These vesicles provide
membrane for the growing neurite (Kater et al., 1990).
16


In accordance with the calcium hypothesis, some neurons may work to
establish these optimal calcium levels in the different parts of the cell. Connor
(1986) showed that actively-growing neurites had relatively high concentrations of
calcium in their growth cones compared to the rest of the cell. This concentration
difference was unaffected by calcium channel blockers. Similarly, calcium is
known to inhibit tubulin polymerization and may not be required at all in the neurite
shaft during elongation. Therefore, cells rriay try to keep [Ca2+]i to a minimum in
this region (Tolkovsky et al., 1990). The findings of these studies and those of
many others support the calcium hypothesis (Mattson and Kater, 1987; Mills et al.,
1990; Silver et al., 1990).
According to the calcium hypothesis, calcium levels within the optimal range for
microfilament motility are probably higher than the [Ca2+]i of the cytosol. When
calcium is within this range, maximum initiation and growth cone motility occur.
Studies that support the calcium hypothesis show that an influx of calcium into
growth cones of N1E-115 cells (caused by depolarization) has been shown to cause
increased growth cone motility (Anglister et al., 1982). A study done by Silver et
al. (1990) showed that growth cones of N1E-115 cells contain hotspots of L-type
VSCCs that allow calcium into the cell when depolarization occurs. This leads to a
selective increase of calcium levels in the growth cone and growth cone expansion.
This may be responsible, in part, for neurite elongation. A similar mechanism may
be involved in neurite initiation with hotspots of L-type VSCCs on the soma,
although this was not tested.
Other studies seem to contradict the calcium hypothesis. Calcium in the neurite
shaft and soma may enhance neurite elongation by affecting calcium-dependent
phosphorylation by protein kinase C or calmodulin. These proteins are involved in
17


the regulation of microtubule associated proteins and tubulin polymerization
(Campenot and Draker, 1989). This is in apparent disagreement with findings that
state that high calcium concentrations within a cell (neurite shaft) inhibit elongation,
but this may be due to differences in optimal calcium levels in the neurite type
observed in that study (Mattson et a/.,1990).
In summary, optimal calcium levels must be achieved for maximum initiation
and growth. These optimal levels vary between cell types and neurite types. Cells
seem to produce calcium maps (areas of different optimal calcium levels for a
specific function). For example, neurons may establish high calcium
concentrations in growth cones and low concentrations in elongating neurite shafts.
If calcium levels were pathologically high, a neurite would not elongate and growth
cone motility might also decrease. If calcium levels were low, the growth cone
might not fully develop, leading to a decrease in initiation even if elongation could
occur in a very low calcium concentration. In this manner, calcium affects the
initiation of neurites (Kater et al., 1990).
The ultimate length of the neurite, however, may also depend on the
concentration of calcium in the elongating neurite shaft because tubule
polymerization is regulated by [Ca2+]j. Since calcium inhibits tubulin
polymerization, the optimal [Ca2+]i in the shaft is most likely lower than that of the
growth cone (Figure 1.5).
The optimal calcium level within an elongating shaft may also vary with neurite
type. Axons tend to have lower [Ca2+]( than dendrites; this lower [Ca2+]i enables
greater elongation. Generally, cells may work to keep calcium levels high in the
growth cone and low in the elongating shaft, but ihe exact model for neurite growth
18


must involve the summation of a variety of cellular messengers, the substrate for
growth and calcium levels (Kater et al., 1990).
The Effects of Abnormally High or Low Calcium Concentrations on Neurite
Elongation arid Initiation
Many toxins that adversely effect the growth and survival of neurons target
[Ca2+]i. This would indicate that the regulation of [Ca2+]i is of the utmost
importance to the cell. If [Ca2+]i does become abnormally high or low, many
cellular functions can be undermined (KomUlainen et al., 1988).
The Toxic Effects of Increased Internal Calcium Concentration
It is possible that an increase in the external calcium concentration increases the
internal calcium concentration beyond the optimal range for the initiation of some
neurites within a cell. Researchers have shown that an increase in [Ca2+]0 causes
an increase in [Ca2+]i in synaptosomes. Nashchen (1985) showed that when
external calcium increased from 0.02 mM to 2 mM, [Ca2+]i increased from 50 nM
to 150 nM. Other researchers have shown that an increase in extracellular calcium
concentration ([Ca2+]0) causes a decrease in the number of neurites in embryonic
chick cells. When [Ca2+]0 was increased from 1.8 mM to about 7.2 mM, the
number of cells with neurites decreased by 30 percent; [Ca2+]i was not tested,
however (Audesirk et al., 1990).
Studies have shown that increased [Ca2+]i affects several metabolic pathways
in the cell. When [Ca2+]i gets too high, calcium-activated reactions become
uncontrolled and cell functioning, particularly cytoskeletal formation, is completely
disrupted (Siesjo, 1990). As previously discussed, calcium inhibits the
polymerization of microtubules (Manalan and Klee, 1984). If [Ca2+]i is extremely
19


high, the microtubule formation required for neurite development may be reduced
and fewer neurites may develop. The motility of microfilaments may also be
stopped when [Ca2+]i is high due to the overacrivatiort of calcium-dependent
capping proteins mentioned earlier (Smith, 1988).
When [Ca2+]i increases, membrane structure can be damaged and membrane ion
channels can be modified. As previously discussed, some cells demonstrate a
calcium-dependent calcium uptake, as well as calcium-dependent calcium release
from the ER, creating a positive feedback loop, potentially leading to toxic calcium
levels in the cytosol (Kennedy, 1988; Rawn, 1990; Siesjo, 1990).
If [Ca2+]i increases to a very high level, the concentration of calcium in the
mitochondria will also increase to an abnormally high level. This could block
oxidative phosphorylation, resulting in a decreased concentration of ATP in the cell.
Reduced ATP stores would cause a decrease in ATPase function, causing a positive
feedback loop that would lead to an even higher [Ca2+]i. Fluctuations in cellular
sodium and potassium could also occur if ATP levels were too low; this would be
detrimental to the cell (Komulainen et al 1987; Nashchen, 1985; Siesjo, 1990).
Increases in [Ca2+]i may also affect nucleic acids in the cell. High calcium
levels have been associated with a decrease in DNA synthesis in neuroblastoma
cells (Reboulleau, 1986). Extremely high calcium concentrations may also activate
endonucleases to fragment DNA (Siesjo, 1990). Any abnormal alteration in DNA
synthesis could cause cell death.
The Toxic Effects of Decreased Internal Calcium Concentration
The sustained lack of calcium in the soma of a rat sympathetic neuron as a result
of low [Ca2+]0 causes neurite degeneration (Campehot arid Draker, 1989). As
20


previously mentioned, Nashchen (1985) showed that a decrease in external calcium
in vitro may cause a decrease in internal calcium concentration in synaptosomes. A
decrease in calcium in culture medium also causes a decrease in neurite initiation in
embryonic chick cells. When external calcium was lowered from 2 mMtoO.l
mM, neurite initiation decreased by 25 percent (Audesirk et al., 1990). This may
be because new growth cone formation was reduced due to the lack of calcium. A
lowered internal calcium concentration in the cone may also result in the
underactivation of calcium-dependent metabolic pathways, particularly actin
filament motility and protective capping resulting in decreased initiation (Kater and
Mattson, 1988; Smith, 1988).
21


Fura-2 AcetoxvmethvK'AM'l Ester as a Tool to Study Internal Calcium
Concentration
CH
COO' Fura-2 AM is a cell-permeable penta-
ester. Once inside a cell, Fura-2 AM
is hydrolyzed and trapped (Figure
1.6). Fura-2 (free acid) then binds
with calcium in the same manner as
EGTA chelation. The excitation of
calcium-bound Fura-2 occurs at the
ultraviolet wavelengths 340 nm and
380 nm (Figure 1.7). Using charge-
coupled device (CCD) digital imaging,
the [Ca2+]i can be measured by
calculating a ratio of emission
intensities at the two excitation
wavelengths. Fura-2 has been called
the best tool currently available to
accurately measure [Ca2+]i (Grynkiewicz et al 1985; Magoroli et al., 1987; Tsien
et al., 1985).
The graph in figure 1.7 demonstrates how emission intensity at the excitation
wavelengths varies with calcium concentration. Maximum excitation occurs at
340 nm and with proper calibration, a unique ratio for each calcium concentration
can be obtained. It is from this relationship that calcium concentration can be
calculated. Images of cells are obtained with an intensified charge coupled device
Figure 1.6. Fura-2 calcium chelator.
22


(ICCD) camera attached to a microscope fitted with special filters that control the
wavelength of light passing through a cell. A series of images from a cell are then
passed to a processor that averages them and calculates the ratio mentioned above.
The processor can subtract background fluorescence and add psuedo-color to the
images that reflects the relative calcium concentration. All of this information is
presented on a video monitor. This processor is controlled by a computer that
stores the images and operates the filter wheel attached to the microscope (Figure
1.8).
Research Hypothesis
Neurite initiation and elongation may be affected differently in rat
hippocampal neurons, transformed rat central nervous system neuroblastoma (B50)
and transformed mouse peripheral nervous system neuroblastoma (N1E-115) by
altered [Ca2+]0. A possible cause of these differences might be differential
23


Figure 1.8. Diagram of the Quantex Calcium Imaging System. .A camera attached to a
microscope feeds into the processor that can then determine the cellular internal calcium
concentration. The ratioed image of a cell is viewed and manually analyzed on the color monitor
with the aid of another computer.
KEYDOAfin (foucil IMousi


efficiency in the regulation of calcium influx or [Ca2+]i when the calcium gradient
across the membrane changes; calcium influx and [Ca2+]i have been shown to be an
integral part of neurite development. Differences in the regulation of calcium could
result in higher or lower [Ca2+]i when the calcium gradient changes. If the amount
of calcium inside a cell is changed, then neurite initiation and elongation could be
affected.
According to the calcium hypothesis bf neurite growth, optimal calcium levels
must be maintained for maximum neurite growth. The optimal range for neurite
initiation and growth cone motility is probably slightly higher than baseline levels of
calcium in the soma. The increases may be caused by influxes of calcium through
VSCCs. The optimal range of [Ca2+]i for elongation is probably somewhat lower
than the level of calcium in the soma because calcium inhibits microtubule
polymerization. When [Ca2+]0 is changed drastically, these optimal calcium
gradients, or calcium maps, within a cell may be disrupted. By this theory, cells
with [Ca2+]i that is too high or too low will initiate neurites less often than cells with
optimal calcium levels. If [Ca2+]i is increased, elongation of initiated neurites may
decrease. If [Ca2+]i were decreased, elongation would probably increase.
When cells are cultured in medium that enhances or reduces the calcium gradient
across the membrane, the regulatory mechanisms of the cell are challenged. Cells
that could not maintain optimal [Ca2+]i or optimal calcium influx during this
challenge would show differences in neurite initiation and elongation from cells of
the same type grown in more physiological medium and from other cell types that
could regulate calcium more efficiently when challenged in the same way.
25


Experiment
Rat hippocampal cells, B50 cells and N1E-115 cells were cultured in media
containing high and low calcium concentrations relative to control concentrations.
Cell survival, the percent of living cells growing neurites and neurite length were
examined in each cell type in each calcium concentration at 48 hours. Several cells
from each culture dish were evaluated and averaged to provide one data point.
In addition to the culturing experiments listed above, tests of intracellular
calcium concentration were conducted for N1E-115 cells and rat hippocampal cells
in each extracellular calcium concentration. These experiments were completed
using Fura-2 fluorescence imaging. Whole cell averages of growing cells were
measured and several cells on each dish were averaged to obtain one data point
Measurements of [Ca2+]i were made at 48 hours.
These experiments were designed to demonstrate whether or not altered
[Ca2+]0 affects survival, elongation and neurite initiation in the cells studied.
Also, tests to determine if [Ca2+]i changes with [Ca2+]0 in N1E-115 and rat
hippocampal neurons were conducted. Information on the regulation of [Ca2+]i
and its apparent effect on neurite development in tumor and non-tumor cells was
obtained. Finally, the usefulness of B50 and N1E-115 neuroblastoma cells as
research models was analyzed.
26


CHAPTER 2
EXPERIMENTAL PROCEDURES
Culturing Media
The three cell types had to be cultured in different media, but the calcium
concentrations were the same in the different media. Also, the two transformed
cell lines had to be differentiated using different techniques, but both techniques
resulted in elevated cyclic AMP levels during part of the process (Table 2.2).
Rat hippocampal neurons were cultured in Eagles minimum essential medium
(MEM) modified from Mattson and Kater (1987) with 2% fetal calf serum (FCS),
0.2% glucose, 20 mM KC1,25 mM HEPES, 10 mM sodium bicarbonate, 1 mM
sodium pyruvate and 1% antibiotic-antimycotic (Sigma). This has 1.8 mM
calcium and will be called pre-made MEM. Potassium chloride was added to
make a final concentration of 20 iiiM to cause depolarization in the cells and
promote neurite growth and survival (Mattson and Kater, 1987). Additional
modified MEM with variable calcium concentrations was prepared from inorganic
salt-free MEM (Sigma, St Louis, MO., special order) to which all salts were
added to the normal concentration except CaCl2*2H20. The calcium concentration
of pre-made MEM was assigned the value IX. Calcium concentrations were
adjusted to range from about 0.1 mM to 7.2 mM, OX through 4X (Table 2.1).
The calcium in the OX medium comes from the FCS and from the other
ingredients as a contaminant


Mouse peripheral nervous system (PNS) neuroblastoma N1E-115 cells were
maintained in Dulbeccos MEM (DMEM) with 10% FCS and cultured in an
inorganic salt-free 1:1 preparation of DMEM and Hams F12 (DMEM-F12)
(Sigma, St. Louis, MO, special order) with 2% FCS to which all salts except
CaCl2*2H20 had been added to their normal Concentrations. Calcium
concentrations were adjusted to match the concentrations in the variable calcium
MEM media (Table 2.1).
Rat central nervous systerii (CNS) neuroblastoma B50 cells were maintained
in DMEM with 10% FCS and cultured in DMEM-F12 2% FCS, prepared as for
the N1E-115 cells (Table 2.1), to which 250 uM dibutyryl cyclic AMP (dbcAMP)
was added to maintain differentiation (Schubert et al., 1974).
Table 2.1. Calcium Concentrations Used for Cell Culturing.
Calcium Concentration (CaCh) Assigned Value
=0.1 mM OX
0.45 mM 0.25X
0.90 mM 0.5X
1.80 mM IX
3.60 mM 2X
7.20 mM 4X
Cell Culture
Hippocampal cells. Hippocampi were removed from 18-day-old rat embryos
and were incubated for 15 minutes at 37C in Hanks balanced salt solution
(HBSS) with 10 mM HEPES and 2 mg/ml trypsin. The hipppocampi were then
rinsed and soaked for 5 minutes in HBSS containing 2 mg/ml trypsin inhibitor.
Cells were dissociated with a pasteur pipette and frozen (-70C) at 4 million cells
28


per milliliter (in vials containing 250 |il) in MEM modified as above except with
2.0% glucose and 8% DMSO (Mattson and Kater, 1987).
Cells were thawed by adding 1 ml of IX medium to the the frozen vial and
gently agitating it in a 37C water bath. Cells were then gently triturated though a
pasteur pipette and a cell count was performed using a hemocytometer. For the
analysis of initiation and growth the cells were plated on 35 mm poly-D-lysine
(Sigma #P-0899, MW 105,800) coated (O.lmg/ml) culture dishes containing 2 ml
of medium ranging from OX to 4X Ca2+. At least one IX pre-made dish and one
dish of each of the concentrations listed in table 2.1 was used per experiment.
Cells were plated at 2x10s cells per dish and incubated at 37C in a humidified 5%
CO2 atmosphere for 48 hours. Four hours after plating, the media were changed
to remove the DMSO and a survival count was preformed (see below).
Afterward, the media were not changed. Cells used for the analysis of [Ca2+]i
were plated on poly-D-lysine coated 35 mm culture dishes modified by removing
the center of the dish bottom and sealing a number 1 glass cover slip to the dish.
This was done to allow the maximum amount of 340 nm and 380 nm UV light
though for the Fura-2 fluorescence analysis. Because the cells did not attach as
well to glass as they did to plastic, the cells were plated at 3.3x10s per dish to
achieve approximately the same cell density on the plastic dishes plated with fewer
cells. The cells were incubated for 48 hours, as described above.
N1E-115 Cells. Mouse PNS neuroblastoma N1E-115 cells were maintained
in DMEM with 10% FCS. To differentiate the cells, they were cultured in
DMEM-F12 with N2 supplements (Bottenstein, 1985), 17.5 Hg/ml
isobutylmethylxanthine (IBMX) and 1.75 |ig/ml prostaglandin El (PGE1) for
three days at 37C, 5% CO2. PGE1 stimulates adenylate cyclase and IBMX
29


inhibits cyclic nucleotide phosphodiesterase resulting in elevated cyclic AMP
levels in the cells.
After three days, cells were plated onto plastic 35 mm culture dishes
containing 2 ml of medium with each of the calcium concentrations listed in table
2.1, at lxlO5 cells per dish. They were incubated, as described, for 48 hours.
The media were not changed (Audesirk et al., 1990).
For experiments on [Ca2+]i, N1E-115 cells were plated on cover slip dishes as
described above at 7,500 cells per dish with the cells being placed in the center of
the dish only. This achieved approximately the same cell density at 48 hours as in
the other N1E-115 cultures.
B50 Cells. Rat CNS neuroblastoma B50 cells were maintained in DMEM with
10% FCS as with N1E-115 cells. B50 cells were plated onto 35 mm plastic culture
dishes containing 2 ml medium at 7.5x10s cells per dish. The cells were incubated
for 48 hours at 37C in a 5% CO2 atmosphere. The medium was not changed. To
achieve differentiation, 0.25 mM dbcAMP was added to each dish upon plating
(Schubert et al., 1974). Analysis of [Ca2+]i was not done on B50 cells.
Table 2.2. Summary of Culturing Media by Cell Type
Cell Type Medium / Treatment
Rat Hippocampal Cell (CNS Primary Cell) Modified (Kater, 1987) MEM + 2% Fetal Calf Serum.
N1E-115 (PNS Transformed Cell) DMEM-F12 + 2% FCS after soaking for 3 days in DMEM-F12, N2, IBMX/PGE1 (cAMP stimulators) for differentiation.
B50 (CNS Transformed Cell) DMEM-F12 + 2% FCS with 0.25 mM dbcAMP for differentiation.
30


Analysis of Survival. Initiation and Elongation
Rat Hippocampal Cells. Four hours after plating, the media on the
hippocampal cells were changed to remove the DMSO left over from the freezing
medium. Also at 4 hours after plating, a survival count was performed using 4
microscope fields that measured approximately 675 (im by 425 Jim on an
Olympus inverted microscope. Cells were judged to be alive if they were round,
phase-bright and attached. This survival count was repeated on the same fields at
48 hours and a percentage was calculated. It was assumed that cells that were not
attached at 4 hours were dead; therefore, this 4 hour count established a baseline
for the count at 48 hours.
Hippocampal cell cultures were examined for neurite initiation and growth at
48 hours. Dishes were letter coded so that the calcium concentration of the dish
was not known until after the count was completed. When the percent of neurons
growing neurites was calculated, all cells growing neurites were counted and this
number was divided by the number of living cells. When elongation was
measured, only cells with a characteristic pyramidal form of one axon at least
twice the length of all other neurites (dendrites) were measured. Hippocampal
neurons with a more symmetrical array of neurites were not measured. These
cells were either stellate neurons of the hippocampus, or had not established
neurite polarity, indicating that they were probably at a more immature stage of
development than the pyramidal ceils. This corresponds to the stages of neuronal
migration seen in the hippocampus in vitro (Fletcher and Banker, 1989).
The count was conducted field by field until at least 30 living neurons and 10
growing neurons were counted. To avoid investigatior bias, the count was
stopped at 50 if 10 growing neurons were not found. Complete fields were
31


always counted, even if the described limits were reached in the middle of a field.
Measurements of the number of living cells, the number of living cells that grew
axons and dendrites, the number of axons and dendrites per field, the length of all
axons and dendrites in a field, the total number of axonal cells, the total number of
neurite branches per field and the mean length of all axons and dendrites were
performed using a computerized digitizing tablet and Sigma-Scan by Jandel
Scientific. A dish was considered one data point Six valid data points were
collected for each concentration of calcium.
The minimum criteria for a valid set of experiments were that, in the controls,
there should have been at least 35% survival at 48 hours and initiation of neurites
in at least 35% of the cells counted. If these criteria were not met by the controls
(pre-made IX and IX from salt-free) of the experiment, then none of the dishes of
any calcium concentration were not used as data points.
N1E-115 Cells. N1E-115 cell cultures were examined for survival, neurite
initiation and elongation at 48 hours. Dishes were letter coded so that the calcium
concentration of the dish was not known until after the count was completed.
Survival, neurite elongation and initiation were compared to that of the IX (of
MEM) calcium concentration. Three randomly chosen fields that measured
approximately 1350 nm by 850 nm on an Olympus inverted microscope were
counted. Measurements of the total number of living cells, the number of living
cells that grew neurites, the number of neurites per field, the length of all neurites
in a field and the mean length of all neurites were performed using a computerized
digitizing tablet and Sigma-Scan by Jandel Scientific. A dish was considered
one data point A minimum of eight valid data points were collected for each
concentration of calcium.
32


The minimum criteria for validity were that, in the controls, there should have
been at least 10 living cells, on average, per field and no more than 25 cells per
field More than 25 cells per field would indicate that the cells were not
differentiated and that excessive cell division occured If 100% of the plated cells
survived, the average number of cells per field would be about 15 to 20. Cells
were judged to be alive if they were round, phase-bright and attached. Also, at
least 40% of the cells that were living had to be growing neurites.
B50 Cells. B50 cell cultures were examined for neurite initiation and growth
at 48 hours. Dishes were letter coded so that the calcium concentration of the dish
was not known until after the count was completed. Growth and initiation were
compared to that of the IX (of MEM) calcium concentration. Three randomly
chosen fields that measured approximately 675 nm by 425 nm on an Olympus
inverted microscope were counted. Measurements of the number of living cells,
the number of living cells that grew neurites, the number of neurites per field the
length of all neurites in a field and the mean length of all neurites were performed
using a computerized digitizing tablet and Sigma-Scan by Jandel Scientific. A
dish was considered one data point A minimum of eight valid data points was
collected for each concentration of calcium.
The minimum criteria for validity were that in the controls, there should have
been at least 10 living cells, on average, per field and no more than 25 cells per
field. More than 25 cells per field would indicate that the cells were not
differentiated and that excessive cell division occured If 100% of the plated cells
survived, the average number of cells per field would be 20. Cells were judged to
be alive if they were round phase-bright and attached. Also, at least 40% of the
cells that were living had to be growing neurites.
33


Analysis of Intracellular Calcium Levels with Fura-2
Obtaining Images. Calcium concentrations were measured from the ratio of
Fura-2 fluorescence intensities at 340 nm and 380 nm light excitation. When the
concentration of calcium increases, the 340 nm fluorescence increases and the 380
nm fluorescence decreases (Figure 1.7). Calcium calibration was performed
using Fura-2 pentapotassium salt, a zero calcium EGTA solution and a saturated
(2 mM) calcium solution on cover-slips. The ratio of fluorescence intensity (R) is
converted to calcium concentration using the formula [Ca2+] = Kd[(R-
Rmin)/(Rmax-R)] (Fq/Fs) (Grynkiewicz et al., 1985). These values are calculated
from the fluorescence intenstities of two solutions. Rmax= the ratio of the
fluorescence of the saturated calcium calibration solution (340/380). Rmin = the
ratio of the fluorescence of the EGTA calibration solution (340/380). Fo/Fs= 380
fluorescence of EGTA solution/380 fluorescence of the saturated calcium solution.
Kd is the binding constant of Fura-2 for calcium (224 mM). Fura-2 analysis was
performed on rat hippocampal cells and N1E-115 cells. Calibration of the
Quantex imaging system was performed daily.
Cells were loaded by adding 2 jil of 3 mM Fura-2 AM in DMSO (Molecular
Probes, Eugene, OR) for a final Fura-2 concentration of 3 |iM. The Fura-2 AM
was added directly onto the culture dishes (2 (0.1/2 ml). The cells were incubated
for 30 minutes at 37C in a 5% carbon dioxide atmosphere. Cells were then
rinsed in the same medium in which they were cultured and were incubated
another 30 minutes. De-esterfication of Fura-2 AM occurred during .this time.
The glass bottom culture dishes were then viewed on a Nikon inverted microscope
with an ICCD camera that was connected to a Quantex digital ratioed image
34


processor (Quantex, Santa Clara, CA, see figure 1.6). Images of cells were taken
and averaged for 16 frames at each wavelength of light The wavelength of light
was controlled by a computer driven filter wheel in the path of the Xenon lamp.
When rat hippocampal cells were studied, the 100X objective was used. The
CCD was set to 100, the camera intensity was Set to 0 and no neutral density filter
was placed in the path of the Xenon lamp. When the'larger N1E-115 cells were
studied the 40X objective was used. The CCD was set to 100, the camera
intensity was set to 900 and a Nikon neutral density filter (ND16) was placed in
the path of the Xenon lamp. At least 10 randomly chosen cells from any one dish
were imaged and were used as one data point Also, one blank field was imaged
in each dish for use as a background standard to be subtracted from all images
taken from that dish. At least one dish from every calcium concentration was used
in every run and cells were not left in Fura-2 containing medium for more than 3
hours. At least 7 dishes at 10 cells each were imaged for each calcium
concentration for rat hippocampal cells. At least 3 dishes at 10 cells each were
imaged for N1E-115 cells.
Image Analysis. With each image, the standard background for the dish was
subtracted. For each image minus background, a halo value was determined and
subtracted from the image. When the halo was subtracted, most neurites were no
longer visible. Boxes used to calculate the average halo intensity for an image were
kept to an approximately constant size. Images were then ratioed and viewed in
pseudo-color. Histograms of whole cells were calculated and the mean calcium
concentration was recorded. Dishes with a calculated calcium concentration less
than 10 nM were given the value of 10 nM due to the insensitivity of the machine at
low calcium levels. All dishes were letter coded so that the extracellular calcium
35


concentration of the dishes was not known during the imaging Or analysis
processes.
Statistical Analysis
All measurements were tested for significant differences from the mean of the
control (IX) cultures. Two-tailed t-tests were performed on all measurements and a
significance level of 0.05 was used. In addition to t-tests, analysis of variance
(ANOVA) tests were performed on the data when significant differences were
found by the t-tests. A significance level of 0.01 was used for the ANOVA tests.
36


CHAPTER 3
EXPERIMENTAL RESULTS
Survival. Neurite Initiation and Elongation
Rat Hippocampal Neurons
There were no significant (p>0.05) differences between hippocampal cells
cultured in pre-made IX MEM and IX MEM prepared from salt-free in any of the
measurements taken.
Survival. At 48 hours all cultures containing calcium levels other than 1.8
mM showed significantly lower survival (p<0.05). The mean survival in 1.8 mM
Ca2+ was 82.51% 2.02% (SEM) (Table 3.1, Figure 3.2 A).
Initiation. Hippocampal cells growing neurites had either a characteristic
pyramidal morphology with one long axon and several shorter dendrites, or a
more symmetrical array of neurites with no distinguishable axons or dendrites.
The percentage of living neurons with neurites at 48 hours was 41.27 3.85% in
the control (IX) MEM. This percentage was calculated by dividing the number of
living cells with neurites by the number of cells alive. External calcium levels
other than IX resulted in significantly lower initiation except 4X. Analysis of
variance (ANOVA) confirmed significant differences among the different
concentrations (Table 3.1, Figure 3.2 B). The control value for the total number
of neurites per living cell was 1.64 0.25. Cells cultured in OX and 4X medium
showed no significant difference from controls, but the other concentrations


resulted in significant differences in t-test analysis. ANOVA analysis confirmed
that significant differences existed (Table 3.1, Figure 3.3 A).
Neurite Elongation. Analysis of mean axon length yielded a value of 331 30
|im for control cultures. There were no significant differences from the control
value in any of the cultures with altered calcium concentrations (Table 3.1, Figure
3.3 B). Figure 3.4 is a comparison of neurite initiation and mean axon length in
rat hippocampal cells at 48 hours. Further, there were no differences from
controls in mean dendrite length, dendrite branch length or axon branch length.
38


Figure 3.1. Rat hippocampal neurons and neuroblastoma N1E-115 cells in culture at 48 hours.
Embryonic rat hippocampal cells (Above). Mouse (PNS) neuroblastoma N1E-115 (Below).
39


Table 3.1. Rat Hippocampal Culturing Data. These data were recorded at 48 hours
in cells plated in media containing z variable calcium evels. All values are SEM.
[Ca2+]o in medium % Survival % Initiation # Neurites per Cell Mean Axon Length (pm)
0.1 68.153.14 29.272.77 1.230.19 32738
0.45 74.233.12 28.743.31 0.950.16 30529
0.90 72.473.11 25.274.21 0.740.15 32537
1.80 82.512.02 41.273.85 1.640.25 33030
3.60 77.271.56 28.203.08 0.920.06 28654
7.20 69.234.89 29.573.64 1.080.14 31033
Table 3.2. N1E-115 Culturing Data. These data were recorded at 48 hours
in cells plated in media containin S variable calcium evels. All values are SEM.
[Ca2+]0 in medium # Alive % Initiation # Neurites per Cell Mean Axon Length (pm)
-0.1 12.71iO.88 60.563.04 4.050.24 996
0.45 13.370.31 56.312.26 4.030.14 1084
0.90 14.040.58 58.741.89 3.640.24 986
1.80 12.790.34 58.331.84 3.420.24 945
3.60 12.930.31 54.481.80 3.770.16 925
7.20 13.420.46 58.751.80 3.550.27 1059
Table 3.3. B50 Culturing Data. These data were recorded at 48 hours
in cells plated in media containing variable calcium evels. All values are SEM.
[Ca2+]0 in medium # Alive % Initiation # Neurites per Cell Mean Axon Length (pm)
0.1 13.961.75 88.763.47 1.630.11 558
0.45 12.631.15 90.892.36 1.700.07 594
0.90 14.371.80 94.530.87 1.700.05 608
1.80 14.071.54 88.863.26 1.740.07 616
3.60 12.830.59 91.031.98 1.980.11 6910
7.20 12.722.08 88.891.62 1.970.14 677
40


0
1
DC
>
'cE
3
w
o
X
X
c
o
CO
s
c
-p
O'
Figure 3.2. The effect of [Ca^+]j on survival and neurite initiation in rat hippocampal cells at
48 hours. A) Changing the external calcium concentration from the control value of 1.8 mM
results in significantly decreased cell survival at 48 hours. The horizontal line highlights the
control value. Statistically significant points are shaded (p<0.05 in a two-tailed t-test). B)
Altered calcium levels also result in significantly fewer neurons producing neurites at 48 hours.
All values are SEM.
41


3
Figure 3.3. The effect of external calcium concentration on the number of neurites per cell and
the mean length of axons in rat hippocampal cells. A) Hippocampal cells cultured in 2X, 0.5X
and 0.2SX calcium show significantly fewer neurites per living cell than controls. The
horizontal line high-lights the control value. Statistically significant points are shaded. B)
Altering external calcium concentration causes no change in mean axon length in hippocampal
cells. All values are SEM.
42


Figure 3.4. A comparison of neurite initiation and mean neurite length in rat hippocampal cells
cultured in varied external calcium concentrations at 48 hours. Mean neurite length shows no
significant difference from the control value, but the percent initiation greatly reduced when
external calcium levels are changed. The graph displays values as a percent difference from the
control (IX) calcium cultures. All values are SEM.
N1E-115 Cells
All values were compared to cells cultured in 1.8 mM (IX) calcium. At 48 hours,
N1E-115 cells cultured in varied external calcium concentrations displayed no
significant differences in survival, number of living cells with neurites or mean
neurite length (p>0.05, two-tailed t-test) (Table 3.2). The only exception was the
cells cultured in 0.25X Ca2+; these cells grew slightly longer neurites and, upon
further analysis, showed more neurites per cell. ANOVA calculation showed that
no significant differences existed among the means of this data set Figure 3.1 B
shows N1E-115 cells in culture at 48 hours. Figure 3.5 A and B shows N1E-115
survival and percent initiation. Figure 3.6 A and B shows the total number of
neurites and mean neurite length as a function of external calcium concentration.
43


20
I
<
w

O
in
in
S
iiiiiiiiii T 1 1 1 1
2 3 4 5 6
[Calcium] 0 (mM)
A
Figure 3.5. The effect of calcium on survival and neurite initiation in N1E-115 cells at 48
hours. A) No significant differences in the number of N1E-115 cells alive per field were found
at 2 days. B) Initiation of neurites in N1E-115 cells is not affected by changing the external
calcium concentration. The horizontal lines high-light the IX values. All values are SEM.
44



o
in
LLJ
Z
i

Q.
W


Z
4fc
Figure 3.6. The effect of calcium on the number of neurites per cell and mean neurite length in
N1E-115 cells at 48 hours. A) A change in external calcium concentration does not effect the
number of neurites per cell in N1E-115 cells. Cells cultured in 0.2SX medium showed a
significant difference with t-test analysis, but ANOVA showed no significant differences exsisted.
B) The mean length of the neurites produced by N1E-1 IS cells is also not effected by a change
in the external calcium concentration with the exception of cells cultured in 0.2SX calcium.
Statistically significant points are shaded. The horizontal lines high-light the IX values. All
values are SEM.
45


B50 Neuroblastoma Cells
Like N1E-115 cells, rat central nervous system neuroblastoma B50 cells
showed no significant differences in survival growth or initiation of neurites at 48
hours (Table 3.3). These results are displayed graphically on Figures 3.7 and
3.8.
46


25
0 1 2 3 4 5 6
[Calcium] o (mM)
f
A
Figure 3.7. B50 neuroblastoma cells cultured in different external calcium concentrations. A)
No significant differences can be seen in the number of cells alive at 48 hours. B) Likewise, no
differences are seen in the percentage of neurons growing neurites. The horizontal lines high-
light the control value. All values are SEM.
47


4

o
O
LO
CQ

Q.
O)
£
*c
3

4b
3 -
3
0
0
T 1 1 I 1 I
2 3 4 5 6 7 8
[Calcium] o (mM)
A
Figure 3.8. The number of neurites produced and the mean length of neurites in B50
neuroblastoma cells. A) Altered external calcium has no effect on the number of neurites per
cell in B50 cells at 48 hours. B) Differences in [Ca^+]0 do not affect the mean length of B50
neurites. Horizontal lines high-light the control value. All values are SEM.
48


The Effects of Changes in Extracellular Calcium Concentration on Intracellular
Calcium Concentration in Rat Hippocampal Cells and N1E-115 Cells
Rat Hippocampal Neurons
Fura-2 analysis of intracellular calcium concentrations at 48 hours in rat
hippocampal cells cultured in varied extracellular calcium concentrations showed
that differences exist between cells cultured in high and low calcium levels (Table
3.4, Figures 3.9,3.10 and 3.12). Both t-test analysis and ANOVA confirmed
this finding. The [Ca2+]i of cells cultured in the control medium proved to be 74
7 nM (mean SEM). The [Ca2+]i of cells cultured in 0.5X (0.9 mM) calcium
was not significantly different from control values, but all other extracellular
calcium concentrations caused significant changes (two-tailed t-test and ANOVA).
[Ca2+]i showed a linear relationship to [Ca2+]0 in media containing from OX to 2X
calcium. At 4X, [Ca2+]i began to plateau. The SEM values for [Ca2+h were less
than 20% of the [Ca2+]j for each extracellular concentration.
Table 3.4. Internal Calcium Concentration ( SEM) Resulting from External Calcium Levels in
Rat Hippocampal Neurons and N1E-115 Cells________________________________________________
[Ca2+]0 (mM) Rat Hippocampal [Ca2+]j (nM) N1E-115 [Ca2+]i (nM)
-0.1 36 7 127 35
0.45 47 5 135 55
0.90 65 6 160 10
1.80 Control) 74 7 132 22
3.60 124 12 138 29
7.20 132 13 157 27
49


N1E-115 Neuroblastoma Cells
Analysis of N1E-115 cells showed that [Ca2+]i is not affected by the external
calcium concentration at 48 hours (Table 3.4 and Figures 3.11 and 3.12). Images
of N1E-115 cells and of rat hippocampal cells were done with different
microscope objectives and, therefore, different calibration scales. The result is
that the colors on the pictures in Figure 3.10 are the same when the [Ca2+]i is
different.
Figure 3.9. A ratioed image of a rat hippocampal cell with neurites. This is a photograph of a
ratioed Quantex image of a rat hippocampal cell that has background and halo subtracted. After
histogram analysis of this cell that was cultured in IX modified MEM, the internal calcium
concentration was found to be 74 nM.
50


A
Figure 3.10. Ratioed images of rat hippocampal cells cultured in 4X and OX MEM. A) This
growing cell was cultured for 48 hours in 4X modified MEM. Histogram analysis showed that
the internal calcium concentration was 128 nM. B) This growing cell was cultured for 48 hours
in OX modified MEM. Histogram analysis showed the internal calcium concentration was 35
nM.
51


Figure 3.11. Ratioed image of an N1E-115 Cell at 48 hours in IX DMEM-F12. Histogram
analysis of this image of a growing N1E-115 cell showed the internal calcium concentration to
be 125 nM. Because different objectives were used (40X for N1E-115 and 100X for rat
hippocampal cells), this image displays the same pseudo-color characteristics as a rat
hippocampal cell with a much lower internal calcium concentration.
+
Figure 3.12. Internal calcium concentration versus external calcium concentration in rat
hippocampal cells and N1E-115 neuroblastoma cells. Calcium concentrations are displayed as
percentage differences from control values. The horizontal line high-lights the control
extracellular calcium concentration (1.8 mM). Values are displayed SEM. Statistically
significant points are denoted by (*).
52


CHAPTER 4
DISCUSSION
Previous studies indicate that [Ca2+]i may influence neurite initiation and
elongation (Anglister et al., 1982; Audesirk et al., 1990; Cohan and Kater, 1987;
Connor, 1986; Holliday and Spitzer, 1990; Kater et al., 1990; Manalan and Klee,
1984; Mattson et al., 1990; Mills and Kater, 1990; Reboulleau, 1986; Silver et al.,
1990; Smith, 1988; Tolkovsky et al., 1990). Changing [Ca2+]0 will change the
calcium gradient across the membrane, altering the amount of calcium entering a
cell. Therefore, the potential exists for changes in [Ca2+]i to occur, depending on
the cells regulatory efficiency for calcium.
From this, altered [Ca2+]0 (higher or lower) may decrease neurite initiation to
the extent that [Ca2+]i is changed. Low [Ca2+]i may result in greater elongation,
while high [Ca2+]i may cause decreased elongation (Kater, 1990) (Figure 1.1).
The present study explored the effects of altered [Ca2+]0 on neurite initiation,
neurite elongation and cell survival and compared [Ca2+]i of cells cultured in
different [Ca2+]0 at 48 hours. This study indicates that the transformed cell N1E-
115 can regulate [Ca2+]i more efficiently than rat hippocampal neurons; therefore, in
N1E-115 cells neurite initiation, growth and cell survival are not altered by changes
in [Ca2+]0. These parameters are affected in rat hippocampal neurons when [Ca2+]0
is altered. Furthermore, the transformed CNS cell B50 is not affected by altered
[Ca2+]0 with respect to initiation, elongation or survival; we predict that [Ca2+]i is
also probably unchanged.


Culturing Experiments
Survival
Incubation in media containing different calcium levels affects non-transformed
rat hippocampal cells, but seems to have no effect on transformed cell lines.
At 48 hours, N1E-115 and B50 cells showed no significant change in survival
in all [Ca2+]0 compared to controls, but hippocampal cells showed a decrease in
survival in media with higher or lower than control [Ca2+]0 (Tables 3.1,3.2, 3.3;
Figures 3.5,3.7). This would indicate that in hippocampal cells, [Ca2+]j changed
when [Ca2+]0 changed. Altered [Ca2+]i would affect cell survival due to decreased
regulation of calcium dependent reactions. This change in [Ca2+]i could affect cell
metabolism, resulting in decreased cell survival (Komulainen et ai, 1987; Rawn,
1990; Siesjo, 1990). It should be noted that in decreased calcium medium,
hippocampal cells were less able to attach to the culture dish bottom resulting in
decreased survival counts. This could mean that true survival might not be affected,
just attachability.
Initiation
When [Ca2+]i changed, neurite initiation was affected. Rat hippocampal cells
showed a decrease in neurite initiation when [Ca2+]0 increased or decreased, but
N1E-115 and B50 cells showed no differences. This was probably due to the
changes in [Ca2+]i in the hippocampal cells.
When the calcium gradient across the plasma membrane was enhanced, the
[Ca2+]i in the hippocampal cells increased (Table 3.4, Figures 3.9, 3.12).
Increased or decreased calcium may have inhibited neurite initiation by interfering
54


with the polymerization and motility of actin filaments (Smith, 1988). In the
absence of proper filament motility, tiew growth cones could not develop. Because
initiation was not completely stopped, not all types of emerging growth cones may
have been affected by these changes in [Ca2+]0. Different types of growth cones
with different dependences on calcium have been found within the same cell (Kater
et al., 1990). It may be that optimal calcium ranges for initiation are somewhat
wide, but when the limits are approached, a decrease in initiation occurs.
In N1E-115 and B50 cells, neurite initiation was unaffected by altered calcium
influx across the membrane. [Ca2+]i was also unchanged in N1E-115 cells by
altered [Ca2+]0. This indicates that in N1E-115 cells, calcium regulatory
mechanisms must exist that are more efficient than those found in the rat
hippocampal cells. Perhaps because [Ca2+]i was unchanged, neurite initiation
occurred with the same frequency in cells cultured in all of the [Ca2+]0 media.
[Ca2+]i was not tested in B50 cells, but because initiation was not affected, it could
be that [Ca2+]i was unchanged in the B50 cultures. Another possibility is that the
dbcAMP used to differentiate the B50 cells affected neurite initiation to a greater
extent than a change in [Ca2+]i may have. Other data have shown that cAMP can
affect initiation (Mann etal., 1989; Reboulleau, 1986; Rydel and Greene, 1988).
Based on these data, it is also possible that a flux of calcium across the
membrane is required for initiation to occur. The exact [Ca2+]i may not be directly
related to the initiation of new neurites. Other studies have reported similar
findings. For example, a flux of calcium caused by VSCCs in growth cones
appears to be necessary for cone motility in N1E-115 cells (Silver et al., 1990).
Furthermore, Mattson et al. (1990) showed that a brief enhanced calcium gradient
within a cell seemed to stimulate axon initiation in rat hippocampal cells. The
55


[Ca2+]i of a cell may affect the size of the flux by altering membrane channels.
[Ca2+]i is known to activate and inhibit calcium channels directly and via the
activation of other kinases (Kennedy, 1988; Rawn, 1990).
Audesirk et al. (1990) showed that the blockage of calcium channels caused a
decrease in initiation in NIE-115 cells. Therefore, it is possible that a flux of
calcium and not [Ca2+]i directly that is responsible for stimulating initiation. In this
study, a gradient of calcium always existed, even in the 0.1 mM medium. If [Ca2+]i
was not changed, then the flux may not have been disrupted. However, it can not
be ruled out from the Audesirk et al. study that a decrease in [Ca2+]i caused the
decrease in initiation because [Ca2+]i was not tested.
Elongation
Elongation was not dramatically changed in any of the cells with altered
[Ca2+]0. No significant differences were seen in any of the cell types. Elongation
may not be dependent on calcium once initiation has occurred (Tolkovsky et al.,
1990). When [Ca2+]i was reduced in rat hippocampal cells, longer neurites were
not seen as expected. Likewise, when [Ca2+]i was increased in the rat hippocampal
cells, elongation was not decreased (Table 3.1; Figure 3.2).
Because calcium inhibits microtubule polymerization, one would expect higher
[Ca2+]i to inhibit growth. It could be that the optimal [Ca2+]i range for elongation is
quite wide or that a different form of calcium regulation exists in neurite shafts than
in cell bodies that helps to maintain low to moderate calcium levels. For example,
neurite shafts have few VSCCs that would allow calcium into the shaft in response
to depolarization (Silver et al., 1990). This alone may have prevented a decrease in
elongation when [Ca2+]i in the soma increased.
56


The [Ca2+]i of neurites was not explored in thisexperiment; however, it was
noted that neurites consistently had a lower [Ca2+]i than cell bodies because during
the image analysis process, neurites were subtracted away from the soma images
when halos were subtracted. During the analysis of images, it is necessary to
subtract the halo, or glow, from around a cell body to get an accurate
measurement of [Ca2+]i in the soma. The imaging system subtracts the calculated
value of the halo intensity from the entire image. Regions of the image with low
calcium levels are also subtracted during this process if their [Ca2+]i levels are lower
than the value calculated for the halo. This would indicate neurites had lower
[Ca2+]i than cell bodies.
Because lower [Ca2+]i did not increase elongation (Tables 3.1,3.2,3.3; Figures
3.3,3.4,3.6,3.8), there may be a threshold of calcium inhibition of microtubule
polymerization. [Ca2+]i was below the level that causes inhibition in all of these
cultures; microtubule assembly was therefore unaffected by calcium and occurred at
an uninhibited rate. Similar results with elongation have been reported in
embryonic chick cells (Audesirk et al., 1990). [Ca2+]i was not tested in this study,
however.
Mattson and Kater (1987) showed that in the snail Helisoma, neurite elongation
was reduced by increased [Ca2+]i. Helisoma may have a lower threshold of calcium
for neurite elongation than rat hippocampal cells, chick cells, or N1E-115 cells.
Therefore, in Helisoma, increased [Ca2+]i may have inhibited elongation by
interfering with microtubule polymerization, as previously discussed.
57


Fura-2 Studies
Fura-2 digital analysis demonstrated that rat hippocampal neurons and N1E-115
cells regulated [Ca2+]i differently in altered [Ca2+]0. The [Ca2+]i in rat hippocampal
cells changed linearly with [Ca2+]0, but N1E-115 cells showed no changes (Table
3.4; Figures 3.11, 3.12). Fura-2 analysis of B50 cells was not conducted.
Causes of Differential Regulation of Internal
Calcium Concentration in Two Cell Types
There are two likely explanations of the differences in [Ca2+]i in high [Ca2+]0.
The first is the differential regulation of plasma membrane and endoplasmic
reticulum calcium ATPases. Calcium ATPases have been called the most important
component in the maintenance of [Ca2+]i. The second is the up and/or down
regulation of VSCCs on the plasma membrane (Carafoli, 1987; Nicholls, 1986).
N1E-115 cells may have a greater ability to increase the activity of calcium ATPases
in times of rising [Ca2+]i. One possibility would be that calcium, calcium-bound
protein kinases, or phosphatases act as positive effectors on calcium ATPase. For
example, calcium ATPase is phosphorylated directly by cAMP-dependent protein
kinase in heart cells (Neyes et al., 1985). As [Ca2+]j increases, the calcium ATPase
activity would increase. This ability may be enhanced in the transformed N1E-115
cells, affording better regulation of calcium ATPases when [Ca2+]i is increasing. In
a similar manner, a decrease in [Ca2+]i resulting from lower [Ca2+]0 would cause
reduced calcium ATPase activity because the kinases or phosphatases that are
activated by calcium would be less active. This would slow down the extrusion of
calcium from the cell when [Ca2+]i was too low.
58


A second possibility for enhanced regulation of [Ca2+]i in decreasing [Ca2+]0 in
N1E-115 cells involves plasma membrane calcium channels. Calcium channel
blockers have been shown to cause a decrease in initiation in N1E-115 cells
(Audesirk et al., 1990a). It could be possible that this channel blockage lead to a
decreased [Ca2+]i; this was not tested, however. There may be a calcium-activated
protein that is a negative effector of calcium channels on the membrane that, in
times of reduced [Ca2+]i, would enhance calcium influx, because the inhibition of
the channels would be removed. Voltage sensitive calcium channels that are
inhibited by calcium-bound calmodulin have been found in other nerve cell types
(Kennedy, 1988; Neyes et al., 1985). In a similar fashion, N1E-115 cells may also
be able to release calcium from intracellular stores more effectively than rat
hippocampal cells using a pathway involving inositol triphosphate. Rat
hippocampal cells may have less extensive calcium stores or less efficient
sodium/calcium exchange proteins.
It is possible that the increased transcription of DNA seen in many transformed
cells results in an increased number of regulatory proteins in the N1E-115 cells,
affording them greater protection from calcium as a by-product of their
transformation. Differences may also be due to differences in PNS versus CNS,
but because neurite initiation in B50 (CNS) cells and N1E-115 (PNS) cells were
not affected by [Ca2+]0 changes, whereas initiation in rat hippocampal neurons and
chick brain cells is affected by [Ca2+]0, the differences probably result from
transformation. It should be noted, however, that B50 cells were cultured with
increased cAMP levels that may have affected initiation and growth. Also, Fura-2
studies were not done on B50s or chick brain neurons.
59


Further Study
[Ca2+]i at 48 hours may not reflect the mean [Ca2+]i during the entire 2-day
incubation period, since calcium may be affected drastically during the first few
hours after plating. These potentially bigger changes in [Ca2+]i in the short term
may have been responsible for the decrease in initiation. [Ca2+]i at shorter-term
time points should be explored to study the pattern in [Ca2+]i over the 2-day period.
Particular attention could be paid to the few hours after plating where much
initiation occurs.
More Fura-2 experiments should be completed on N1E-115 cells to decrease the
variability obtained in [Ca2+]j (SEM was up to 39 percent). The general finding that
[Ca2+]i in NIE-115 cells is not affected by [Ca2+]0 would probably not change,
however.
An assay of the activity of calcium ATPase would be a very valuable addition to
these experiments. If the activity of these calcium pumps could be analyzed in the
two cell types, information on the mechanisms involved in the enhanced regulation
of [Ca2+]i in NIE-115 cells could be obtained. These pumps may be regulated
directly by [Ca2+]i, calmodulin, or another calcium-dependent kinase or
phosphatase. One would expect N1E-115 cells to show greater calcium ATPase
activity than rat hippocampal cells when [Ca2+]0 increased, and possibly lower
calcium ATPase activity when [Ca2+]0 decreased.
There may be differences in DNA transcription between the two cell types.
Calcium is known to affect DNA-binding proteins in many cells (Kennedy, 1988).
Slight increases or decreases in calcium flux across the membrane may result in the
increased or decreased transcription or translation of calcium ATPase proteins or
60


VSCCs. This would result in increased or decreased numbers of these proteins on
the membrane to alleviate an abnormal flux of calcium. Studies of DNA
transcription and RNA translation of calcium channels and calcium ATPases would
provide information about the regulation of [Ca2+]i when [Ca2+]0 changes.
The flux of calcium across the membrane rather than the [Ca2+]i may be
responsible for initiation in some neurons, because calcium channel blockers, but
not reduced [Ca2+]0 cause a decrease in initiation in N1E-115 cells (Audesirk et al.,
1990). A study with the calcium channel blocker dihydropyridine similar to the one
conducted by Audesirk et al. (1990) should be conducted on N1E-115 cells. Such
a study should include an analysis of [Ca2+]i. This would provide information on
the specific mechanism of calcium in the initiation of new neurites. If initiation is
found to be decreased in the presence of the channel blocker, but [Ca2+]i is
unchanged, this would support the hypothesis that an influx of calcium is the
required prerequisite for initiation.
Completion of the above studies will provide valuable information on the
regulation of [Ca2+]i and its role in neurite initiation and elongation.
61


LITERATURE CITED
Anglister, L., Farber, I., Shahar, A., Grinvald, A. (1982). Localization of voltage-
sensitive calcium channels along developing neurites: their possible role in
regulating neurite elongation. Developmental Biology. 94:351-365.
Audesirk, G., Audesirk, T., Ferguson, C., Lomme, M., Shugarts, D Rosack, J.,
Caracciolo, P., Gisi, T., Nichols, P. (1990a). L-Type calcium channels may
regulate neurite initiation in cultured chick embryo brain neurons and N1E-115
neuroblastoma cells. Developmental Brain Research. 494:421-34.
Bean, B. (1989). Classes of calcium channels in vertebrate cells. Annual Review of
Physiology. 51:367-384.
Bottenstein, J. (1985). Culture methods for growth of neuronal cell lines in defined
media. In D.W. Barnes, G.H. Sato (Eds). Methods for Serum Free Culture of
Neuronal and Lymphiodal Cells, Liss, New York, pp3-13.
Campenot, R. and Draker, D. (1989). Growth of sympathetic nerve fibers in
culture does not require extracellular calcium. Neuron. 3:733-743.
Carafoli, E. (1987). Intracellular calcium homeostasis. Annual Review of
Biochemistry. 56:345-443.
Choi, D. (1988). Calcium mediated neurotoxicity: relationship to specific channel
types and role in ishemic damage. TINS. 11:465-468.
Cohan, C. and Kater, S. (1987). Electrically and chemically mediated increases in
intracellular calcium in neuronal growth cones. Journal of Neuroscience.
7:3588-3599.
Connor, J. (1986). Digital imaging of free calcium changes and of spatial gradients
in growing processes in single, mammalian central nervous system cells. Proc.
National Acad, of Science. 83:6179-6183.
Connor, J., Kater, S., Cohan, C., Fink, L. (1990). Calcium dynamics in neuronal
growth cones: regulation and changing patterns of calcium entry. Cell Calcium.
11:233-239.
Duprat.A. and Kan, P. (1981). Stimulating effect of the divalent cation ionophore
A 23187 on in vitro neuroblast differentiation; comparative studies with
myoblasts. Experimentia. 37:154-158.


Fletcher, T. and Banker, G. (1989). The establishment of polarity in hippocampal
neurons: the relationship btween the stage of a cells development in situ and
its subsequent development in culture. Developmental Biology. 136:446-454.
Grynkiewicz, G., Poenie, M., Tsien, R. (1985). A new generation of calcium
indicators with greatly improved fluorescence properties. Journal of Biological
Chemistry. 260:3440-3450.
Haugland, R. (1989). Fura-2 excitation intensity at various calcium concentrations.
Molecular Probes Handbook of Fluorescence Probes and Research Chemicals.
p.78.
Holliday, J. and Spitzer, N. (1990). Spontaneous calcium influx and its roles in
differentiation of spinal neurons in culture. Developmental Biology. 141:13-
23.
Kater, S. and Mattson, M. (1988). Calcium regulation of the neuronal growth
cone. TINS. 12:315-320.
Kater, S., Mills, L., Guthrie, P. (1990). Intracellular calcium and the control of
neuronal growth and form. Trophic Factors and the Nervous System.L.A.
Honrocks (Ed). Raven Press Ltd. New York, p.231-245.
Kennedy, K. (1988). Regulation of neuronal function by calcium. TINS. 12:417-
421.
Koike, T., Martin, D., Johnson, E. (1989). Role of calcium channels in the ability
of membrane depolarization to prevent neuronal death induced by trophic factor
deprivation: evidence that levels of internal calcium determine nerve growth
factor dependence of ganglion cells. Proc. National Acad. Science. 86:6421-
6425.
Komulainen, H. (1988). Increased free cellular calcium by toxic agents: an index of
potential neurotoxicity? TIPS. 5:231-234.
Kruskal, B., Shak, S., Maxfield, F. (1986). Spreading of human neutrophils is
immediately preceded by a large increase in cyotoplasmic free calcium. Proc.
National Acad. Science. 83:2919-2923.
Malgoroli, A., Milani, D., Meldolesi, J., Pozzan, T. (1987). Fura-2 measurement
of cytosolic free calcium in monolayers and suspensions of various types of
animal cells. Journal of Cell Biology. 105:2145-2155.
Manalan, A. and Klee, B. (1984). Calmodulin and cyclic nucleotide protein
phosphorylation. Research. 18:227-278.
Mann, D., Doherty, P., Walsh, F. (1989). Increased intracellular cyclic AMP
differently modulates nerve growth factor induction of three neuronal
recognition molecules involved in neurite outgrowth. Journal of
Neurochemistry. 53:581-588.
63


Matthews,E. (1986). Calcium and membrane permiability. British Medical Bulletin.
42:391-397.
Mattson, M. and Kater, S. (1987). Calcium regulation of neurite elongation and
growth cone motility. Journal of Neuroscience. 7:4034-4043.
Mattson, M., Murain, M., Guthrie, P. (1990). Localized calcium influx orients
axon formation in embryonic hippocampal pyramidal neurons. Developmental
Brain Research. 52:201-209.
Meldolesi, J. and Volpe, P. (1988). The intracellular distribution of calcium. TINS.
11:449-452.
Miller, R. (1988). Calcium signaling in neurons. TINS. 11:415-419.
Mills, L.and Kater, S. (1990). Neuron specific and state specific differences in
calcium homeostasis regulate the generation and degeneration of neuronal
architecture. Neuron. 2:149-163.
Nahorski, S. (1988). Inositol phosphates and neuronal calcium homeostasis.
TINS. 11:444-448.
Nashchen, D. (1985). Regulation of cytosolic calcium concentration in presynaptic
nerve endings isolated from rat brain. Journal of Physiology. 363:87-101.
Neyes, L., Reinlib, L., Carafoli, E. (1985). Phosphorylation of the calcium
pumping ATP ase of heart sarcolemriiU and erythrocyte plasma membrane by
the cyclic AMP dependent protein kinase. Journal of Biological Chemistry.
18:10283-10287.
Nicholls, D. (1986). Intracellular calcium homeostasis. British Medical Bulletin.
42:353-358.
Ozawa, S., Tsuzuki, M-, Ogura, A., Kudo, Y. (1989). Three types of of voltage
dependent calcium current in cultured rat hippocampal neurons. Brain
Research. 495:329-336.
Pang, P Wang, R., Karpinski, J., Shan, J., Benishin, A.(1990). Control of
calcium channels in neuroblastoma cells N1E-115. Experimental Gerentology.
25:247-253.
Rasmussen, H. (1989). The cycling of calcium as an intracellular messenger.
Scientific American. 42:66-73.
Rawn, J. (1990). Biochemistry. Patterson Publishers. Burlington, North
Carolina.pp:253-257; 352-353.
64


Reboulleau, C. (1986). Extracellular calcium induced neuroblastoma cell
differentiation:involvement of phosphatidylinosittol turnover. Journal of
Neurochemistry. 46:920-930.
Rogers, M and Hendry, I. (1990). Involvment of dihydropyrodine-sensitive
calcium channels in nerrve growth factor-dependent neurite outgrowth by
sympathetic neurons. Journal of Neuroscience. 26:447-454.
Rydel, R.and Greene, L. (1988). Cyclic AMP analogs promote survival and neurite
outgrowth in cultures of rat sympathetic and sensory neurons independently of
nerve growth factor. Proc. National Acad. Science. 85:1257-1261.
Schubert, D., Heinemann, W., Carlisle, H., Kimes.B., Patrick, J., Steinbach, J.,
Culp, W., Brant, L. (1974). Clonal cell lines from the rat central nervous
system. Nature. 249:224-227.
Scott, R. (1990). Voltage-dependent modulation of rat sensory neurone calcium
channel currents by G protein activation: effect of a dihydropyridine
antagonist British Journal of Pharmacology. 99:629-630.
Siesjo, B. (1990). Calcium in the brain under physiological and pathalogical
conditions. European Neurology. 30(supp2):3-9.
Silver, A., Lamb, A., Bolsover, S. (1989). Elevated cytosolic calcium in the
growth cone inhibit neurite elongation in neuroblastoma cells: correlation of
behavioral states with cytosolic calcium concentration. Journal of
Neuroscience. 9:4007-4020.
Silver, A., Lamb, A., Bolsover, S. (1990). Calcium hotspots caused by L-channel
clustering promote morphological changes in neuronal growth cones. Nature.
343:751-754.
Smith, S. (1988). Neuronal cytomechanics: the actin-based motility of growth
cones. Science. 242:708-715.
Tank, D.,Sugimori, S., Connor, J. (1988). Spatially resolved calcium dynamics of
mammalian cells in cerebral slice. Science. 242:773-776.
Thayer, S., Pemey, P., Miller, R. (1988). Regulation of calcium homeostasis in
sensory neurons by bradykinin. Journal of Neuroscience. 8:4089-4097.
Tolkovsky, A., Walker, A., Murrell, R. (1990). Calcium transients are not
required as signals for long term neurite outgrowth from cultured sympathetic
neurons. Journal of Cell Biology. 110:1295-1306.
Toselli, M., Taglietti, V. (1990). Pharmacalogical characterization of voltage
dependent calcium currents in rat hippocampal cells. Neuroscience Letters.
112:70-75.
65


Tsien, R., Rink, T., Poenie,M. (1985). Measurement of cytosolic free calcium in
individual small cells using flourescence microscopy with dual excitation
wavelengths. Cell Calcium. 6:145-157.
Wahl, P., Schousboc, A., Honore, T., Dreijer, J. (1989). Glutimate induced
increase in intracellular calcium in cerebral cortex neurons is transient in
immature cells but permanent in mature cells. Journal of Neurochemistry.
53:1316-1319.
66


Full Text

PAGE 1

THE RELATIONSHIP AMONG EXTERNAL CALCIUM, INTERNAL FREE CALCIUM AND NEURITE GROWTH IN RAT HIPPOCAMPAL NEURONS AND NEt*OBLASTOMA CELLS by James Andrew Frank B.A., University of Colorado, 1989 A thesis submitted to the Faculty of the Graduate School of the University of Colorado in partial fulfillment of the requirements for the degree of Master of Arts Departnient of Biology 1991

PAGE 2

This thesis for the Master of Arts degree by James Andrew Frank has been approved for the Department of Biology by Gerald Audesirk Teresa Audesirk Alan Brockway

PAGE 3

Frank, James Andrew (M.A., Biology) The Relationship Among External Calcitlrtl, Intemal Free Calcium and Neurite Growth in Rat Hippocampal Neurons and Neuroblastoma Cells Thesis directed by Associate Professor Gerald.Audesirk Intracellular free calcium ([Cal+];) plays a critical role in the growth of neurites. A change in the gradient for calcium across the plaslila membrane may result in altered influx of the cation. To maintain ail optimal [Ca2+]i for neurite growth, cells may have to overcome the difference in Ca2+ influx. This study was designed to explore the effects of altered external calcium concentration ([Ca2+]0 ) on [Ca2+]i and neurite growth in rat hippocarilpal neurons, mouse peripheral nervous system neuroblastoma NIE-II5 and rat central nervous system neuroblastoma B50. Neurite initiation and elongation were quantified in each cell type in six different extracellular calcium concentrations (OX, 0.25X, 0.5X, IX, 2X, 4X, where IX was 1.8 mM Ca2+). Using Fura-2 digital imaging analysis, [Ca2+]i was determined in each [Ca2+]0 for NIE-II5 cells and rat hippocampal neurons. The data showed that initiation of neurites was sigirlficantly reduced in rat hippocampal neurons cultured in media with calcium levels other than IX. The two transformed cell lines showed no differences from IX. Elongation of neurites was unaffected in all three cell types by altered [Ca2+]0 Analysis of [Ca2+]i showed that in rat hippocampal neurons, [Ca2+]i changed monotonicly when [Ca2+]9 changed with a plateau near 4X. NIE-II5 cells showed no differe11:ces in [Ca2+]i from IX values in any of the altered [Ca2+]0 NIE-II5 cells may be able to regulate [Ca2+]i more efficiently than rat hippocampal neurons when challenged with an alteration in the influx of Ca2+.

PAGE 4

Also, neurite elongation seems to be less sensitive than neurite initiation to changes in [Ca2+]i in rat hippocampal neurons. The form and content of this absract are approved. I recommend its publication. Gerald Audesirk iv

PAGE 5

ToPooka

PAGE 6

Figures . Tables .. Acknowledgements CONTENTS viii .... X xi CHAPTER 1. INTRODUCTION Calcium Homeostasis . . . . . . Regulation of Calcium by the Plasma Membrane Regulation of Calcium by Organelles 1 3 4 7 Calcium in the Cytosol . . 9 The Role of Calcium in Neurite Growth 12 Initiation . . . . . 12 The Role of Calcium in Growth Cone Motility and Neurite Elongation . . . . . . . 15 The Effects of Abnormally Fligh or Low Calcium Concentrations on Elongation and Initiation The Toxic Effects ofincreased Internal Calcium Concentration The Toxic Effects of Decreased Internal Calcium Concentration Fura-2 AM Ester a,s a Tool to Study Internal Calcium Concentration . Summary of Experiments 19 19 20 22 23 Research Hypothesis . Experiment . . . . . . 23 26

PAGE 7

2. EXPERIMENTALPROCEDURES. 27 Culturing Media . . . 27 Cell Culture. . . . . . . . . . 28 Analysis of Survival, Initiation and Elongation . . 31 Analysis of Intracellular Calcium Levels with Fura-2 . 34 Statistical Analysis . . . . . . . . 36 3. EXPERIMENTAL RESULTS . . . Survival, Neurite Initiation and Elongation Rat Hippocampal Neurons. N1E-115 Cells . . B50 Neuroblastoma Cells The Effects of Changes in Extracellular Calcium ....... 37 . 37 37 . 43 ... 46 Concentration on IntracellQlar Calcium Concentration . . 49 RatHippocampalNeurons. . . . . . . 49 N1E-115 Neuroblastoma Cells. 50 4. DISCUSSION. . . . . . 53 Culturing Experiments. . . . . . . . . 54 Survival Initiation . . . 54 54 Elongation . . . . . . . . 56 Ftira-2 Studies . . . . . Causes of Differential Regulation of Internal Calcium Concentration in Two Cell Types Further Study . . LITERATURE CITED . . . vii 58 58 60 62

PAGE 8

Figure 1.1. 1.2. 1.3. 1.4. FIGURES The involvement of calcium in the growth of neurites . . The major pathways of calcium regulation in a nerve cell . . Characteristics of voltage-sensitive calcium channels . . . Cytosolic and membrane bound proteins that are affected by calcium . . 1. 5. Optimal calcium concentrations for initiation, elongation and membrane vesicle binding 1.6. Fura-2 calcium chelator . . 1. 7. Graph of excitation intensities of Fura-2 AM 1. 8. Quantex calcium imaging system 3.1. 3.2. Rat hippocampal neurons and N1E-115 cells in culture ... The effect of calcium on survival and neurite initiation in rat hippocampal cells at 48 hours 3. 3. The effect of calcium on the number neurites/cell and mean 2 4 5 11 16 22 23 24 39 41 axon length in rat hippocampal neurons at 48 hours 42 3.4. A comparison of neurite initiation and mean neurite length in rat hippocampal cells . 43 3.5. The effect of calcium on survival and neurite initiation in NlE-115 cells at 48 hours . 44 3.6. The effect of calcium on the number neurites/cell and mean neurite length in NlE-115 at 48 hours . 45 3.7. The effect of calcium on survival and neurite initiation in B50 cells at 48 hours . . 47

PAGE 9

3. 8. The effect of calcium on the number neurites/cell and mean neurite length in B50 cells at 48 hours . . . 48 3. 9. A ratioed image of a tat hippocampal cell with neurites 50 3.1 0. Ratioed images of rat hippocampal cells cultured in 4X and OX MEM . . . . . . . 51 3.11. A ratioed image of an NlE-115 cell in lX DMEM-Fl2 . . . . 3.12. Internal calcium concentration versus external calcium 52 concentration in rat hippocampal neurons and NlE-115 cells . 52 ix

PAGE 10

TABLES Table 2.1. Calcium Concentrations Used for Culturing 2.2. Summary of Culturing Media by Cell Type 3.1. Rat Hippocampal Culturing Data 3.2. NlE-115 Culturing Data 28 30 40 40 3.3. B50 Culturing Data . 40 3.4. Internal Calcium Concentration Resulting From External Calcium in Rat Hippocampal Neurons and NlE-115 Cells 49

PAGE 11

ACKNOWLEDGEMENTS I would like to thank Dr. Gerald Al!desirk for his valuable suggestions regarding this manuscript and for his gUidance during the course of this project. I also thank Drs. Teresa Audesirk and Alan Brockway for critically reading earlier drafts of this paper and offering helpful comments. Finally, I extend my gratitude to Leigh Cabel-Kluch, Charles Ferguson and David Shugarts for technical assistance with this work. This research was supported by grailts from the Environmental Protection Agency and the National Institutes of Health. xi

PAGE 12

CHAPTER! Calcium is a ubiquitous second messenger found in every type of cell. In neurons, cytoplasmic calcium plays an role in nearly every facet of nervous system development and regulation. Calcium is particularly important in neurite growth. The growth of neurites (axons and dendrites) has been described as having three steps: initiation from the cell body, neuronal growth cone motility and neurite shaft elongation (Audesirk et al.. 1990). Initiation from the cell body involves the formation of a new growth cone. Growth cones are typically found ori the leading end of neurites and can assume several different morphologies that correspond to the type of neurite that is developing.. During neurite growth, it is the motility of actin microfilaments in the filopodia of these cones that "pulls" a neurite along, while the shaft "pushes." Calcium regulates the motility of microfilaments in the growth cone and the polymerization of tubulin for microtubules; these are required for neurite growth. It is believed that relatively high calcium levels in the growth cone lead to maximum cone motility, while relatively low calcium levels promote maximum neurite elongation (Figure 1.1 ). Neurite initiation, elongation and growth cone motility are all dependent on the concentration of free calcium inside the cell ([Ca2+]i.). [Ca2+]i is the amount of unbound calcium within the cytosol of the cell. The total concentration of calcium

PAGE 13

Neurite Shaft Growth Cone High (ca2+Jin actin-based motility and exocytosis Low (ca2+Jin enhances Microtubule polymerrzation __ -:::;:;;,.r Filopodia (pull) Figure 1.1. The involvement of calcium in the growth of neurites. Calcium enhances growth cone motili but inhibits microtubule I merization. includes calcium that is bound to proteins and that which is sequestered in organelles. The continued study of the role of intracellular calcium will provide valuable insight into the development and maintenance of the nervous system (Mills and Kater, 1990). Tills introduction will first address the regulation of [Ca2+]i, including regulation by membrane proteins and enzymes in the cytosol. Next, the role of calcium in neurite growth will be discussed in detail. Finally, the hypothesis of this research will be presented and the specific experiments will be outlined. 2

PAGE 14

Calcium Homeostasis Maintenance of a relatively cortstailt [Ca2+]i (homeostasis) is of vital importance to the functioning of a cell. The regulation of [Ca2+]i involves the integration of several biochemical pathways within a neuron (Mills and Kater, 1990). Studies have shown that neurons have a great capacity to regulate [Ca2+]io but that not all cell types use the same mechanisms to achieve homeostasis (Cohan, 1987; Tank et al . 1988; Thayer et al., 1988). The efficiency of regulatory mechanisms within a cell is demonstrated by the fact that [Ca2+]i is about 100 nM to 200 nM in most mammalian cells, while the extracellular calcium concentration ([Ca2+]0 ) is nonnally between 1 mM and 5 mM (Rawn, 1990). The [Ca2+]i of a cell tends to oscillate mildly around a set point (Carafoli, 1987). Not all cell types maintain the same [Ca2+]i. For example, embryonic rat hippocampal neurons maintain caicium concentration of 45 nM to 70 nM (Mattson et al., 1990) while the mouse neuroblastoma cell N1E-115 keeps calcium levels between 100 nM and 200 nM (Pang et al.. 1990). Additionally, cells of the same type may not regulate calcium with the same efficiency throughout their lives. Wahl et al. (1989) demonstrated that 8-day-old cerebral cortex cell cultures were unable to expel a large influx of calcium, but 2-day-old cells in culture could. In light of the above findings, it is apparent that the level of calcium within a cell depends on many factors. The degree and efficiency to which specific regulatory mechanisms are used is probably the single most important factor in determining [Ca2+]i and maintaining cell viability. The manner in which [Ca2+]i is controlled by the plasma membrane, cellular organelles and cytosolic proteins will now be 3

PAGE 15

discussed (Figure 1.2). The biochemical role of calcium in normal cell function will also be addressed. PM a ea2+ .... + b c Protein + Ca t I [Ca2+]. 1 Figure 1.2. The maE! pathways of calciUlll in a nerve cell. Plasma membrane {PM): a, represents the VSCC; b, RMCC influx; c, calcium A TPases that pump oil+ out of the cell; d., the sodium/calcilliil exchanger: e, Ca2+ A TPases that pump ea2 + into the endoplasmic reticulum (ER); f, receptor mediated ea2+ channels for efflux from the ER; g, calcium lll!ii!.O!:fer that only calcium into the mitochondria; h, inner mitochondrial membrane aMM) sodiwn/caicium exchanger. i, cytosolic molecules that bind calcium (from Nicholls 1986 Regulation of Calcium by the Plasma Membrane Although calcium in the cytoplasm of a cell can originate from the extracellular medium or from intracellular stores, the majority of calcium that enters the cell cytoplasm comes from outside the cell due to the large gradient for influx (Carafoli, 1987). For this reason, the plasn1a membrane is the most important regulatory barrier for calcium. Calcium Channels. Calcium channels are ion channels which, when activated, predominatly allow calcium to pass through the membrane along its gradient. The plasma membrane contains two general categories of calcium channels: voltage-4

PAGE 16

sensitive calcium channels (VSCCs) and receptor-mediated calcium channels (RMCCs). Most of the calcium that enters the cell through the plasma membrane comes through these channels (Nic;holls, 1986). The VSCCs can be subdivided iflto long-lasting (L), transient (T) and neither L or T (N) types. The names correspond to the length of time the channel conducts a calcium current in response to membrane depolarization. L-type channels have a relatively high threshold and tend to have lang-lasting currents during a given depolarization, while T -type channels have a law threshold and are rapidly inactivated. N-type channels have an intermediate threshold voltage and have a current conduction duration that is in between L apd T. Therefore, more calcium enters a cell through L-type than N-or T-type channels. (Figtire 1.3). pA mV -1 oo I +250 0 T L -35 +10 L -40 L n ms 1 ooo o ms 1 ooo FiguJe 1.3. Characteristics of voltage-sensitive calcium channels. In the the left is a graJ>h of a voltage stimulus used to activate a typical T -(Y.Pe calcium current On the lower left is a typical image of the T-type channel cmrent as seen dunng a patch clamp experiment On the right ts a typiCal L-type channel current (below) and the stimulus used to activate it (above) (Silver et al., 1990). Note that the T channel has a lower threshold and is inactivated gwckly. The stimulus volta e is in millivolts m V the ctirrent is in icoain rs A and the time is m milliseconds ms VSCCs are found all over the cell body and synaptic endings of neurites. Channels with the general characteristics of those listed above have bee.n found in 5

PAGE 17

the neuroblastoma cell line N1E-115, rat hippocampal cells and other cell types (Anglister et al., 1982; Bean, 19S9; Ozawa et al., 1989; Silver et al., 1990; Toselli and Taglietti, 1990). The role of specific calcium channels in netnite development will be discussed later. When bound by a ligand, receptor-mediated calcium channels (RMCCs) undergo changes in conformation that lead to the opening of a calcium channel. RMCCs are found on the endoplaSmic reticulum (ER) and calcisomes (small single membrane organelles that are produced by the ER), as well as the cell body plasma membrane (Rasmussen, 1989). One J;ype of RMCC is the N-methyl-D-aspartate (NMDA) receptor that opens a calcium channel in response to glutamate binding in vivo. These channels are found on rat hippocampal pyramidal neurons and other neuron types (Siesjo, 1990). Calcium A TPases. Ca2+ ATPases hydrolyze A TP to provide energy to move calcium across a membrane. These transporters, which are found on the plasma membrane and the ER, pump calcium out of the cytosol, resulting in a lower [Ca2+]i. Calcium ATPases have been described as the most important mechanism available to a cell to lower [Ca2+] (Miller, 1988). Obviously, the action of these transporters is directly related to the amount of A TP within a cell. If cell metabolism is decreased, the activity of the pumps will decrease and [Ca2+]i will increase. Sodium/Calcium Exchancer. This transporter is found predominantly on the plasma membrane, but can also be found on the inner mitochondrial membrane. The stoichiometry of the exchanger is 3 Na+ to one Ca2+, but hydrogen ions and other monovalent cations can also be exchanged for calcium across a membrane with this protein. Further, the affinity of the exchanger for certain ions is affected 6

PAGE 18

by phosphorylation. The affmity for calcium increases tenfold and the affmity for sodium increases twofold with the addition of ATP irt vivo, probably due to calmodulin-dependent phosphorylation. The driving force for the exchanger is the large sodium gradient across the plasma membrane. Theoretically, the sodium gradient provides enough energy to maintain a [Ca2+]i of about 100 nm, but the system works too slowly to overcome the large calcium influx that occurs dUring depolarization in a neuron. Also, the exchanger is electrogenic and would seem to have limits based on the membrane potential (Cohan and Kater, 1987). Laboratory experiments have confl.IlD.ed that the exchanger could not be solely responsible for maintaining [Ca2+]i (Carafoli, 1987). The most unique feature of the exchanger is that it can move in both directions, depending on the relative concentrations of calcium and sodium. This exchanger is probably more important in muscle than in neurons (Nashchen, 1985). Regulation of Calcium by Organelles Cytosolic calcium concentration is influenced by regulatory proteins on the membranes of intracellular organelles; specifically, the endoplasmic reticulum (ER) and the inner mitochondrial membrane are involved in the regulation of [Ca2+]i. Other organelles, including the nucleus, have about the same calcium concentration as the cytosol (Meldolesi and Volpe, 1988). Endoplasmic Reticulum. As previously mentioned, the ER membrane contains Ca2+ A TPase and RMCC proteins to the concentration of calcium in the ER and in the cytosol. The ER and calcisomes, which are single membrane organelles produced by the ER, usually have very high calcium concentrations relative to the cytoplasm and take up. Ca2+ in an A TP-dependent manner. The ER is perhaps the 7

PAGE 19

only organelle that plays a crucial role in maintaining the [Ca2+]i of a healthy cell because it can temporaztly sequester large amounts of ca}cium in the event of influx (Meldolesi and Volpe, 1988; Nahorski, 1988). Despite this ability, the importance of the ER is far outweighed by the regulatory capacity of the plasma membrane (Nicholls, 1986). The release of Ca2+ from the ER is initiated with the hydrolysis of phosphotidylinositol phosphate (PIP) by G proteins or by phospholipase C. PIP, a phospholipid in the membrane, is hydrolyzed to form inositol trisphosphate (IP3), a fatty acid, and diacylglycerol (DAG), a diglyceride. IP3 binds to RMCCs on the ER, resulting in the opening of the calcium channels and the effiux of calcium into the cytosol along its gradient (Meldolesi and Volpe, 1988) (Figure 1.4). Calcium in the cytosol is known to activate phospholipase C and G proteins via calcium dependent protein kinases. Both of these proteins can also be activated by the binding of hormones to the plasma membrane. Mitochondria. The inner membrane of the mitochondrion (IMM) holds two types of calcium transporters: the sodium/calcium exchanger, which has already been discussed, and the membrane potential-dependent uniporter. This uniporter is exclusive to the IM:M and transports ca2+ into the mitochondrial matrix along the electrical gradient that exists between the cytosol and the mitochondrial matrix. The intermembrane space of the mitochondrion is quite positively charged relative to the matrix, which is due, in part, to the hydrogen ion pumping activity of cytochromes and reductases on the IMM. This voltage gradient provides energy for positively charged calcium ions to flow into the (Nicholls, 1986). Although mitochondria have a high concentration of calcium relative to the cytosol, they play only a smallrole in regulating [Ca2+ li because the Na+/Ca2+ 8

PAGE 20

exchanger is. the only IM:M protein capable of transporting calcium out of the mitochondria. Sodium levels hi the cytosol and are usually not very different; therefore, little energy is available for the extrusion of calcium into the cytosol. Also, it has been noted that mitochondrial calcium levels increase only when cytosolic calcium concentrations get very high (Nicholls, 1986). It should be noted that calcium plays an important regulatory role within the mitochondrion in that it activates pyruvate dehydrogenase and other citric acid cycle and oxidative phosphorylation dehydrogeriases that are involved in energy release. If calcium levels in the matrix become abiion:hal, energy production is decreased (Rawn, 1990). Calcium in the Cytosol Within the cytosol, the calcium concentration is regulated to a small extent by soluble proteins; however, the majority of these proteins triggers biochemical events in the cell when bound to calcium (Kennedy, 1988; Koike et al., 1989). The next section will address the cytos9lic proteins that bind calcium and the biochemical effects that result from the binding. Proteins Affected by Calcium. Calcium can enter the cytoplasm through the plasma membrane or from the ER. Cytosolic calcium, regardless of origin, can bind with several proteins in the Cytosol or on the inside of the plasma membrane and regulate their action (Figure 1.4). The two types of membrane-bound proteins which are influenced by [Ca2+]i are channels and enzymes. There are at least two types of ion channels tluJ.t are regulated by calcium. The first type consists of voltage-sensitive calcium channels (VSCCs). As explained previously, VSCCs allow calcium to pass through the plasma membrane in 9

PAGE 21

response to depolarization of the cell. Some VSCCs are inactivated by calcium. The other type of ion channels includes calcium channels, as well as channels for other ions, such as potassium, chloride and hydrogen. Unlike VSCCs, these channels are opened in response to calcium binding (Bean, 1989). As well as influencing channels, calcium can also affect membrane-bound and cytosolic enzymes. On the plasma membrane, calcium activates enzymes such as phospholipase C. When phospholipase Cis activated, it can hydrolyze PIP to form IP3, which binds to a receptor mediated calcium channel on the ER, causing the release of calcium into the cytosol. This pathway can result in a positive feedback loop in which [Ca2+]i increases (Kennedy, 1988; Siesjo, 1990). In the cytosol, calcium can bind with other enzymes and enzyme regulators, such as protein kinase C, calmodulin, paravalbumin and cal pain, which will now be discussed. The binding of calcium activates protein kinase C to phosphorylate other neuronal proteins, including ion channels, resulting in the alteration of electrical excitability in the cell. Additionally, protein kinase C phosphorylates proteins involved in the regulation of synaptic transmission. The second cytosolic protein, calmodulin, is an enzyme regulator and has four binding sites for calcium that exhibit positive cooperativity with respect to the binding constants (Kci) for each site. There are at least four important functions of calcium-bound calmodulin. When calcium binds to calmodulin, a conformational change occurs that allows calmodulin to bind to and activate adenylate cyclase, which synthesizes cyclic AMP from ATP. Calmodulin also activates cyclic nucleotide phosphodiesterase to hydrolyze cyclic AMP. This is significant because in heart muscle, Ca2+A TPase is directly phosphorylated by cyclic AMP-dependent 10

PAGE 22

PM Cal pain Calmodulin Calcine min CAM Kinase IT Adenylate Cyclase Phospho diesterase Figure 1.4. Cytosolic and membrane bound proteins that are affected by [Ca2+]i. All arrows indicate direct or indirect activation with the exception of diacylglycerol (DAG) that inactivates C Kinases. Also, the binding of IP3 to the ER causes an efflux of calcium into the c osol from Kenned 1988 protein kinase causing to .be pumped out of the cell (Neyes et al., 1985). It is possible that this may also occur in neurons. Another function of calmodulin is to activate calmodulin-dependent kinase II, which can then phosphorylate a Ca2+ATPase, resulting in the decrease of [Ca2+]i. Calmodulin-dependent kinase IT is also important in the modification of ion channels involved in long-term potentiation within pyramidal cells bf the hippocampus. The fourth function of calcium-bound calmodulin is to activate calcineurin, which inactivates calciumdependent L-type VSCCs in pyramidal neurons (Kennedy, 1988; Neyes et al., 1985). Another cytosolic enzyme, paravalbumin, is probably more important in muscle cells than neurons. Paravalbumin binds calcium in the cytoplasm and may go on to activate proteases. The fourth enzyme is the protease calpain, which is also activated when Ca2+-bound. Calpain is important in the regulation of cytoskeletal architecture. Both paravalbumin and calpain are found in very small quantities in

PAGE 23

the cytosol of neurons and some membrane-bound forms also exist (Kennedy, 1988; Neyes et al., 1985; Rawn, 1990; Scott et al., 1990). All four of these cytosolic proteins regUlate [Ca2+ Ji to a very small degree, with the exception of the proteins ailow calcium into or out of the cell when Ca2+. activated. "Regulation" may be a misnomer, but the amount of calcium bound to cytosolic proteins is significant enough to warrant notice. For the most pan, however, it is the regulatory activities of the plasma membrane and organelles that attempt to maintain a balance in the calcium-mediated reactions listed above. In summary, the proteins most important in neuronal calcium regulation are the calcium-dependent VSCCs and RMCCs, calmodulin-dependent kinase II, phopholipases, calmodulin and protein kinase C (Kemtedy, 1988). The Role of Calcium in Neurite Growth As previously mentioned, neurite growth has three steps; initiation from the cell body, neuronal growth cone motility and neurite shaft elongation (Audesirk et al., 1990). Internal free calcium concentration may affect each of these steps, but is probably not the only intracellular messenger involved in their regulation. Initiation Neurite initiation can be denned as the formation of a new growth cone. Two ideas dominate current theories of initiation. The first is that an influx of calcium into the soma of a neuron is necessary for the development of a neurite. According to this hypothesis, this calcium can then interact with other intracellular messengers, resulting in a change in cellular morphology. The second idea is that a calcium influx is not required t9 trigger neurite outgrowth and that other biochemical events 12

PAGE 24

that may involve pre-existing calcium stores produce a new neurite. Although both of the mechanisms discussed above ate calcium-dependent, the dependence on extracellular calcium for neurite development may be different among cell types and neurite types. Some cells tna,y incorporate both of the above mechanisms in the initiation of different neurite types (Kater et al., 1990). Calcium Influx and Initiation. Many stimuli that regulate neutomorphology exert their effects by changing [Ca2+]i (Mills and Kater, 1990). It has been observed that a transient increase in [Ca2+]i that occurs through VSCCs is required for neurite initiation in Xenopus embryonic cells (Holliday and Spitzer, 1990). Other studies have indicated that an influx of calcium through L-type VSCCs may be required for initiation of certairi neurites in N1E-115 cells, embryonic chick neurons and superior cervical ganglion cells (Aridesirk et al., 1990; Rogers and Hendry, 1990). This influx may lead to the formation of a new growth cone and, thus, result in neurite initiation from the soma. This type of calcium influx may also be involved in neurite sprouting since increased calcium could lead to increased microfllameni motility, the activation of other cellular messengers and the formation of a new growth cone (Kruskal et al., 1986). L-type VSCCs have been found to cluster on growth cone membranes, creating calcium hotspots in the cones (Silver et al., 1990). Within the growth cone, calcium promotes the actin microfilament motility that is responsible for the pulling action of the filopodia on the cone. A particular neuron may have some growth cone types that require an influx of calcium and others th.at do not (Kater et al., 1990). The pulling action of the actin microfllaments depends on A TP hydrolysis, calcium and certain kiDases for motile force. Calcium-dependent gelsolin proteins 13

PAGE 25

act to stabilize actin filaments by capping the filament ends, thereby preventing subunit depolymerization or unnecessary polymerization. Many actin binding proteins involved in the movement of the filaments and microtubule-associated proteins are also modulated by calcium dependent kinases (Smith, 1988). Too much or too little calcium would impede proper growth cone development in at least some growth cone types (Kater et a/.,). A transient increase in may not be required to trigger initiation in all cell types (Tolkovsky et al., 1990). Several researchers have reported that cyclic AMP promotes neurite initiation in embryonic and transformed rat cells in the absence of an influx of calcium (Mann et al., 1989; Reboulleau, 1986; Rydel and Greene, 1988; Schubert er al., 1974). Kater and Mattson (1988) reported that cAMP along with protein kinase C may cause ihltiation direcdy or via a pathway involving calcium in the absence of a large calcium influx in rat hippocampal nemons. These molecules probably regulate the formation of cytoskeletal components. One possible explanation of these findings is that dependence on calcium is different among different neurite types in the same cell. For example, in pryamidal rat hippocampal neurons, axons develop in areas within the cell with lower calcium concentrations than those in whiclt dendrites develop (Mattson et al., 1990). Because several morphological types of growth cones have been observed, it is reasonable to assume that their dependence on calcium and other cellular messengers is also variable (Kater et al., 1990). Calcium is probably involved in the initiation of all neurites, but an influx or transiently-enhanced gradient may not be required. (Duprat and Kan, 1981). Recently, it has been reported that concentration gradients of calcium within a cell, rather than a Ca2+ influx, are necessary for axon genesis. In this study, 14

PAGE 26

Mattson et al. (1990) caused a calcium influx into the soma of rat hippocampal pyramidal cells by transecting an existing axon. They noted that when [Ca2+]i increased from about 50 nM to 130 nM near the transection site, axon initiation was inhibited, but in other areas of the cell where the [Ca2+]i was relatively low, axon initiation was enhanced. This would indicate that a brief, enhanced calcium gradient in the cell causes axon initiation in rat hippocampal cells. This may be related to previous findings by the gro\lp that demonstrated that axons had a lower [Ca2+]i than dendrites .. It could be that this lower calcium level in axons facilitated longer, faster elongation (Mattson et al., 1990). The Role of Calcium in Growth Cone Motility and Neurite Elongation After initiation of the growth cone occqrs, the second and third steps of neurite growth are neuronal growth cone motility neurite shaft elongation. These steps may occur simultaneously; the movement of the growth cone pulis the neurite along, while the microtubules in the shaft push it. Calcium Hypothesis. The so-called calcium hypothesis for neurite growth states that there are different optimal calcium levels for both growth cone motility and neurite elongation (Figure 1.5). High concentrations of calcium relative to the cytosol promote maximum growth cone motility by stimulating the actin microfllaments contained within the cone. This type of motility may enhance neurite growth and cause the formation of new growth cones. If [Ca2+]i becomes too high or too low, however, growth cone motility will decrease (Kater et al., 1990). This could lead to a decrease in neurite initiation; this was not tested, however (Figure 1.5). 15

PAGE 27

Concentrations of calcium that are lower than that of the growth cone afford maximum neurite shaft elongation because calcium inhibits the polymerization of microtubules that are required to form the cytoskeleton of a developing neurite (Audesirk et al., 1990; Cohan and Kater,1987; Kater and Mattson, 1988). If [Ca2+]i was too high, elongation would be inhibited (Figure 1.5). Percent of Maximum elongation \ Initiation and cone motility \ [Ca2+ 1 0 membrane binding l Figw:e 1.5. Optimal calciwn concentrations for initiation. elongation and vesicle binding to the membrane in a h thetical neuron. The mechanism of the inhibition of polymerization may involve the interaction of calcium, calmodulin and microtubule associated proteins. Similarly, it is possible that abnormally low [Ca2+]i may also inhibit neurite elongation (Manalan and Klee, 1984). Finally, the calcium hypothesis also states that there is an optimal calcium level for the fusion of vesicles to the plas111a membrane. If [Ca2+ li is out of the optimal range, initiation and elongation will decrease (Figure 1.5). These vesicles provide membrane for the neurite (Kater et al., 1990). 16

PAGE 28

In accordance with the calcium hypothesis, some neurons may work to establish these optimal calcium levels 'in the different parts of the cell. Connor (1986) showed that actively-growing neurites had relatively high concentrations of calcium in their growth cones compared to the rest of the cell. this concentration difference was unaffected by calcium channel blockers. Similarly, calcium is known to inhibit tubulin polymerization and may not be required at all in the neurite shaft during elongation. Therefore, cells may tty to keep [Ca2+]i to a minimum iD this region (Tolkovsky et al., 1990). The findings of these studies and those of many others support the calcium hypothesis (Mattson and Kater, 1987; Mills et al., 1990; Silver et al., 1990). According to the calcium hypothesis, levels within the optimal range for microfllament motility are probably higher than.the [Ca2+]i of the cytosol. When calcium is within this range, maximum initiation and growth cone motility occur. Studies that support the calcium hypothesis show that an influx of calcium into growth cones of N1E-115 cells (caused by depolarization) has been shown to cause increased growth cone motility (Anglister et al., 1982). A study done by Silver et al. (1990) showed that growth cones ofN1E-115 cells contain hotspots ofL-type VSCCs that allow calcium into the cell when depolarization occurs. This leads to a selective increase of calcium levels in the growth cone and growth cone expansion. This may be responsible, in part, for neurite elongation. A similar mechanism may be involved in neurite initiation with hotspots of L-type VSCCs on the soma, although this was not tested. Other studies seem to contradict the calcium hypothesis. Calcium in the neurite shaft and soma may enhance neurite elongation by affecting calcium-dependent phosphorylation by protein kinase C or calmodulin. These proteins are involved in 17

PAGE 29

the regulation of microtubule associated proteins and tubulin polymerization (Campenot and Draker, 1989). This is in apparent disagreement with fmdings that state that high calcium concentrations within a ceij (neurite shaft) inhibit elongation, but this may be due to differences in optimal calcium levels in the neurite type observed in that study (Mattson et a/.,1990). In summary, optimal levels must be achieved for maximum initiation and growth. These optimal levels vary between cell types and neurite types. Cells seem to produce "calcium maps" (areas of different optimal calcium levels for a specific function). For example, neurons may establish high calcium concentrations in growth cones and low concentrations in elongating neurite shafts .. If calcium levels were pathologically high, a neurite would not elongate and growth cone motility might also decrease. If calcium levels were low, the growth cone might not fully develop, leading to a decrease in initiation even if elongation could occur in a very low calcium concentration. In this manner, calcium affects the initiation of neurites (Kater et al., 1990). The ultimate length of the neurite, however, may also depend on the concentration of calcium in the elongating nelirite shaft because tubule polymerization is regulated by [Ca2+]i. Since calcium inhibits tubulin polymerization, the optimal [Ca2+]i in the shaft is most likely lower than that of the growth cone (Figure 1.5). The optimal calcium level within an elongating shaft may also vary with neurite type. Axons tend to have lower [Ca2+]j" than dendrites; this lower [Ca2+]i enables greater elongation. Generally, cells may work to keep calcium levels high in the growth cone and low in the elongating shaft, but the exact tnodel !or neurite growth 18

PAGE 30

must involve the summation of a variety of cellular messengers, the substrate for and calcium levels (Kater et al., 1990). The Effects of Abnormally High or Low Calcium Concentrations on Neurite Elongation arid Initiation Many toxins that adversely effect the growth and stnvival of neurons target [Ca2+]i. This would indicate that. the regulation of [Ca2+]i is of the unnost importance to the cell. If [Ca2+]i does become high or low, many cellular functions can be undemlined (KomWainen et al., 1988). The Toxic Effects oflncreased Internal Calcium Concentration It is possible that an increase in the external calcium concentration increases the internal calcium concentration beyond the optimal range for the initiation of some neurites within a cell. Researchers have shown that art increase in (Ca2+]0 causes an increase in [Ca2+]i in synaptosomes. Nashchen (1985) showed that when external calcium increased from 0.02 mM to 2 mM, [Ca2+]i increased from 50 nM to 150 nM. Other researchers have shown that an increase in extracellular calcium concentration ([Ca2+]0 ) causes a decrease in the number of neurites in embryonic chick cells. When [Ca2+]0 was increased from 1.8 mM to about 7.2 mM, the number of cells with neurites decreased by 30 percent; [Ca2+]i was not tested, however (Audesirk et al., 1990). Studies have shown that increased [Ca2+]i affects several metabolic pathways in the cell. When [Ca2+]i gets too high, calcium-activated reactions become uncontrolled and cell functioning, particularly cytoskeletal fonnation, is completely disrupted (Siesjo, 1990). As previously discussed, calcium inhibits the polymerization ofmicrotubules (Manalan and Klee, 1984). If [Ca2+]i is extremely 19

PAGE 31

high, the microtubule formation required for neurite development may be reduced / and fewer neurites may develop. Tpe motility of microfilaments may also be stopped when [Ca2+]i is high due to the overactivatiort of calcium-dependent capping proteins mentioned earlier (Smith, 1988). When [Ca2+]i increases, membrane structure can be damaged and membrane ion channels can be modified. As previously discussed, some cells demonstrate a calcium-dependent calcium uptake, as well as calcium-dependent calcium release from the ER, creating a positive feedback loop, potentially leading to toxic calcium levels in the cytosol (Kennedy, 1988; Rawn, 1990; Siesjo, 1990). If [Ca2+]i increases to a very high level, the concentration of calcium in the mitochondria will also increase to an abnormally high level. This could block oxidative phosphorylation, resulting in a decreased concentration of A TP in the cell. Reduced A TP stores would cause a decrease in A TPase function, causing a positive feedback loop that would lead to an even higher [Ca2+]i. Fluctuations in cellular sodium and potassium could also occur if A TP levels were too low; this would be detrimental .to the cell (Komi.Ilainen eta/., 1987; Nashchen, 1985; Siesjo, 1990). Increases in [Ca2+]i may also affect nucleic acids in the cell. High calcium levels have been associated with a decrease in DNA synthesis in neuroblastoma cells (Reboulleau, 1986). Extremely calcium concentrations may also activate endonucleases to fragment DNA (Siesjo, 1990). Any abnormal alteration in DNA synthesis could cause cell death. The Toxic Effects of Decreased Internal Calcium Concentration The sustained lack of calcium in the soma of a rat sympathetic neuron as a result oflow [Ca2+]0 causes degeneration (Campehot arid Draker, 1989). As 20

PAGE 32

previously mentioned, Nashchen (1985) showed that a decrease in external calcium in vitro may cause a decrease in internal calcium concentration in synaptosomes. A decrease in calcium in culture medium also causes a decrease in neurite initiation in embryonic chick cells. When external calcium was lowered from 2 mM to 0.1 mM, neurite initiation decreased by 25 percent (Audesirk et al., 1990). This may be because new growth cone formation was reduced due to the lack of calcium. A lowered internal calcium concentration m the cone may also reslilt in the underactivation of calcium-dependent metabolic pathways, particularly actin filament motility and protective capping resulting in decreased initiation (Kater and Mattson, 1988; Smith, 1988). 21

PAGE 33

Fura-2 AcetoxymethylCAMl Ester as a Tool to Study Internal Calcium Concentration coo coo-l '--N coo '--Nl(\ ) 0 coo-Fura-2 AM is a cell-permeable penta ester. Once inside a cell, Fura-2 AM is hydrolyzed and trapped (Figure 1.6). Fura-2 (free acid) then binds with calcium in the same manner as EGTA chelation. The excitation of calcium-bound Fura-2 occurs at the ultraviolet wavelengths run and 380 nm (Figure 1.7). Using charge coupled device (CCD) digital imaging, the [Ca2+]i can be measured by calculating a ratio of emission intensities at the two excitation Figure 1.6. Fura-2 calcium chelator. wavelengths. Fura-2 has been called the best tool currently available to accurately measure [Ca2+]i (Grynkiewicz et al., 1985; Magoroli et al., 1987; Tsien et al . 1985). The graph in figure 1. 7 demonstrates how emission intensity at the excitation wavelengths varies with calcium concentration. Maximum excitation occurs at 340 run and with proper calibration, a unique ratio for each calcium concentration can be obtained. It is from this reiationship that calcium concentration can be calculated. Images of are obtained with an intensified charge coupled deVice 22

PAGE 34

(ICCD) camera attached to a microscope qtted with special filters that control the wavelength of light passing through a cell. A series of images from a cell are then passed to a processor that averages them and calculates the ratio mentioned above. The processor can subtract background fluorescence and add psuedo-color to the images that reflects the relative calcium concentration. All of this information is presented on a video monitor. This processor is controlled by a computer that stores the images and operates the filter wheel attached to the microscope (Figure 1.8). Research Hypothesis > 1-"' z w 1-z z 0 ;: < !:: lJ X w 250 Fura-2 + ca2+ EM 510 NM 43.5 JJM 0.756 0.441 0.284 0.189 0. 1 2 6 .... ___ o .o e 1 -rtlf+.'-,.-.,,. 0.047 0. 0 2 1-HH/1'-H.-/ 300 350 400 WAVELENGTH (NM) Summary of Experiments 450 Neurite initiation and elongation may be affected differently in rat hippocampal neurons, transformed rat central nervous system neuroblastoma (B50) and transformed mouse peripheral nervous system neuroblastoma (N1E-115) by altered [Ca2+]0 A possible cause of these differences might be differential 23

PAGE 35

.!.J-Ir-------.11 ,...-lo.L.i...---..... 0 "' en w E 0 :1: w :1: Ps I 1 ::s IW u IIIII mi ..JC: oo C::l--1-zZ oo u:: ... .. I .n CL:J u Figure 1.8. DiaW
PAGE 36

efficiency in the regulation of influx or [Ca2+]i when the calcium gradient across the membraD.e changes; calcium influx and [Ca2+]i have been shown to be an integral part of neurite developinertt. Differences iri the regulation of calcium could result in higher or lower [Ca2+]i when the calcium gradient changes. If the amount of calcium inside a cell is changed, then neurite initiation and elongation could be affected. According to the calcium hypothesis of neurite growth, optimal calcium levels must be maintained for maximum peurite growth. The optimal range for neurite initiation and growth cone motility is probably slightly higher than baseline levels of calcium in the soma The increases may be caused by influxes of calcium through VSCCs. The optimal range of [Ca2+]i for elongation is probably somewhat lower than the level of calcium in the soma because calcium inhibits microtubule polymerization. When [Ca2+]0 is changed drastically, these optimal calcium gradients, or calcium maps, within a cell may be disrupted. By this theory, cells with [Ca2+]i that is too high or too low will initiate neurites less often than cells with optimal calcium levels. If [Ca2+]i is increased, elongation of initiated neurites may decrease. If [Ca2+ ]i were decreased, elongation would probably increase. When cells are cultured in mepium that enhances or reduces the calcium gradient across the membrane, the regulatory mechanisms of the cell are challenged. Cells that could not maintain optimal [Ca2+]i or optimal calcium influx during this challenge would show differences in neurite initiation and elongation from cells of the same type grown in more physiological medium and from other cell types that could regulate calcium more efficiently when challenged in the same way. 25

PAGE 37

Experiment Rat hippocampal cells, B50 cells and NlE-115 cells were cultured in media containing high and low calcium concentrations relative to control concentrations. Cell survival,. the percent of living cells growing rtelirites and neurite length were examined in each cell type in each calcium concentration at 48 hours. Several cells from each culture dish were evaluated and averaged to provide one data point. In addition to the culnirlng experiments listed above, tests of intracellular calcium concentration were conducted for NlE-115 cells and rat hippocampal cells in each extracellular calcium concentration. These experiments were completed using Fura-2 fluorescence imaging. Whole cell averages of growing cells were measured and several cells on each dish were averaged to obtain one data point Measurements of [Ca2+]i were made at 48 hours. These experiments were designed demonstrate whether or not altered [Ca2+]0 affects survival, elongation and neurite initiation in the cells studied Also, tests to determine if [Ca2+]i changes with [Ca2+]0 in N1E-115 and rat hippocampal neurons were conducted. Information on the regulation of [Ca2+]i and its apparent effect on neurite development in tumor and non-tumor cells was obtained. Finally, the usefulness of B50 and N1E-115 neuroblastoma cells as research models was analyzed. 26

PAGE 38

Culturing Media CHAPTER2 The three cell types had to be cultured in different media, but the calcium concentrations were the same in the different medla. Also, the two transformed cell lines had to be differentiated using different techniques, but both techniques resulted in elevated cyclic AMP levels during part of the process (Table 2.2). Rat hippocampal neurons were cultured in Eagle's minimum essential medium (MEM) modified from Mattson and Kater (1987) with 2% fetal calf serum (FCS), 0.2% glucose, 20 mM KCl, 25 mM HEPES, 10 mM sodium bicarbonate, 1 mM sodium pyruvate and 1% antibiotic-antimycotic (Sigma). This has 1.8 mM calcium and will be called pre-made MEM. Potassium chloride was added to make a final concentration of 20 IilM to cause depolarization in the cells and promote neurite growth and survival (Mattson and Kater, 1987). Additional modified MEM with variable calcium concentrations was prepared from inorganic salt-free MEM (Sigma, SL Louis, MO., special order) to which all salts were added to the normal concentration except CaCl2H20. The calcium concentration of pre-made MEM was assigned the value IX. Calcium concentrations were adjusted to range from about 0.1 mM to 7.2 mM, OX through 4X (Table 2.1). The calcium in the OX medium comes from the FCS and from the other ingredients as a contaminant

PAGE 39

Mouse peripheral nervous system (PNS) rteuroblastoma N1E-115 cells were maintained in Dulbecco's MEM (DMEM) with 10% FCS and cultured in an inorganic salt-free 1:1 preparation of OMEM and Ham's F12 (DMEM-F12) (Sigma, St. Louis, MO, special order) with 2% FCS to which all salts except Ca02H20 had been added to their nonilal concentrations. Calcium concentrations were adjusted to match the concentrations in the variable calcium MEM media (Table 2.1). Rat central nervous system (CNS) neuroblastoma B50 cells were maintained in DMEM with 10% FCS and cultured in DMEM-F12 2% FCS, prepared as for the N1E-115 cells (Table 2.1), to which 250 uM dibutyryl cyclic AMP (dbcAMP) was added to maintain differentiation (Schubert et al., 1974). T bl 21 Cal C a e .. ctum oncentrations Uedfl CllCI s or e u tunnl!. Calcium Concentration (CaCl2) Assigned Value =0.1 mM ox 0.45mM 0.25X 0.90mM 0.5X 1.80mM 1X. 3.60mM 2X 7.20mM 4X Cell Culture Hi:g:gocampal cells. Hippocampi were removed from 18-day-old rat embryos and were incubated for 15 minutes at 37C in Hank's balanced salt solution (HBSS) with 10 mM HEPES and 2 mglml trypsin. The hipppocampi were then rinsed and soaked for 5 minutes in HBSS containing 2 mglml trypsin inhibitor. Cells were dissociated with a pasteur pipette and frozen ( -70C) at 4 million cells 28

PAGE 40

per milliliter (in vials containing 250 J.ll) in MEM modified as above except with 2.0% glucose and 8% DMSO (Mattson and Kater, 1987). Cells were thawed by adding 1 ml of lX medium to the the frozen vial and gently agitating it in a 37C water bath. Cells were then gently triturated though a pasteur pipette and a cell count was performed using a hemocytometer. For the analysis of initiation and growth the cells were plated on 35 mm poly-D-lysine (Sigma #P-0899, MW 105,800) coa.ted (O.lmg/ml) culture dishes containing 2 ml of medium ranging from OX to 4X Ca2+. At least one 1X pre-made dish and one dish of each of the concentrations listed in table 2.1 was used per experiment. Cells were plated at 2x10S cells per dish and incubated at 37C in a humidified 5% atmosphere for 48 hours. Four hours after plating, the media were changed to remove the DMSO and a survival count was preformed (see below). Afterward, the media were not changed. Cells used for the analysis of [Ca2+]i were plated on poly-D-lysine coated 35 mm culture dishes modified by removing the center of the dish bottom and sealing a number 1 glass cover slip to the dish. This was done to allow the maximuQl amount of 340 run and 380 nm UV light though for the Fura-2 fluorescence analysis. Because the cells did not attach as well to glass as they did to plastic, the cells were plated at 3.3x1Q5 per dish to achieve approximately the same cell density on the plastic dishes plated with fewer cells. The cells were incubated for 48 hours, as described above. Nl E-115 Cells. Mouse PNS neuroblastoma N1E-115 cells were maintained in O:MEM with 10% FCS. To differeQ.tiate the cells, they were cultured in DMEM-F12 with N2 supplements (Hottenstein, 1985), 17.5 J.lg/ml isobutylmethylxanthine (ffiMX) and 1.75 J.lg/ml prostaglandin E1 (PGE1) for three days at 37C, 5% C02. PGEl stimulates adenylate cyclase and ffiMX 29

PAGE 41

inhibits cyclic nucleotide phosphodiesterase resulting in elevated cyclic AMP levels in the cells. After three days, cells were plated onto plastic 35 mm culture dishes containing 2 ml of medium with each of the calcium concentrations listed in table 2.1, at 1x10S cells per dish. They were incubated, as described, for 48 hours. The media were not changed (Audesirk et al., 1990). For experiments on [Ca2+]io N1E-115 cells were plated on cover slip dishes as described above at 7,500 cells per dish with the cells being placed in the center of the dish only. This achieved approxitnately the same cell density at 48 hours as in the other N1E-115 cultures. B50 Cells. Rat CNS neuroblastoma B50 cells were maintained in DMEM with 10% FCS as with N1E-115 cells. B50 cells were plated onto 35 Iilm phistic culture dishes containing 2 ml medium at 7.5x10S cells per dish. The cells were incubated for 48 hours at 37C in a 5% C02 atmosphere. The medium was not changed. To achieve differentiation, 0.25 mM dbcAMP was added to each dish upon plating (Schubert et al., 1974). Analysis of [Ca2+]i was not done on B50 cells. T bl 22 S a e .. ummaryo rc 1 M d. b c n T u tunng e 1a >y e lype Cell Type Medium I Treatment Rat Hippocampal Cell (CNS Primary Cell) Modified (Kater, 1987) MEM + 2% Fetal Calf Serum. N1E-115 (PNS Transformed Cell) DMEM-Fl2 + 2% FCS after soaking for 3 days in DMEM-F12, N2, ffiMX/PGE1 (cAMP stimulators) for differentiation. B50 (CNS Cell) DMEM-Fl2 + 2% FCS with 0.25 mM dbcAMP for differentiation. 30

PAGE 42

Analysis of Survival. Initiation and Elongation Rat Cells. Four hours after plating, the media on the hippocampal cells were changed to remove the DMSO left over from the freezing medium. Also at 4 hours after plating, a, survival count was performed using 4 microscope fields that measured approximately 675 J.Lm by 425 J.Lili on an olympus inverted microscope. Cells were judged to be alive if they were round, phase-bright and attached. This survival count was repeated on the same fields at 48 hours and a percentage was calculated. It was assumed that cells that were not attached at 4 hours were dead; therefore, this 4 hour count established a baseline for the count at 48 hours. Hippocampal cell cultures were examined for neurite initiation and growth at 48 hours. Dishes were letter coded so that the calcium concentration of the dish was not known until after the count was completed. When the percent of neurons growing neurites was cells growing neurites were counted and this number was divided by the number of living cells. When elongation was measured, only cells with a characteristic pyramidal form of one axon at least twice the length of all other neurites (dendrites) were measured. Hippocampal neurons with a more symmetrical array of neurites were not measured. These cells were either stellate neurons of the hippocampus, or had not established neurite polarity, indicating that they were probably at a more immature stage of development than the pyramidal ce1Is. This corresponds to the stages of neuronal migration seen in the hippocampus in vitro (Fletcher and Banker, 1989). The count was conducted field by field until at least 30 living neurons and 10 growing neurons were counted. to avoid investigatior bias, the count was stopped at 50 if 10 growing neurons were not found. Complete fields were 31

PAGE 43

always even if the described limits were reached in the middle of a field. Measurements of the number of living cells, the number of living cells that grew axons and dendrites, the number of axons and dendrites per field, the length of all axons and dendrites in a field, the total number of axonal cells, the total number of neurite branches per field and the mean length of all axons and dendrites were performed using a computerized digitizing tablet and Sigma-Scan by J andel Scientific. A dish was considered one data point Six valid data points were collected for each concentration of calcium. The minimum criteria for a valid set of experiments were that, in the controls, there should have been at least 35% survival at 48 hours and initiation of neurites in at least 35% of the cells counted. If these criteria were not met by the controls (pre-made lX and lX from salt-free) of the experiment, then none of the dishes of any calcil.im concentration were not used as data points. N1E-115 Cells. N1E-115 cell cultures were examined for survival, neurite initiation and elongation at 48 hours. Dishes were letter coded so that the calcium concentration of the dish was not known until after the count was completed. Survival, neurite elongation and initiation were compared to that of the lX (of :MEM) calcium concentration. Thtee raridoinly chosen fields that measured approximately 1350 nm by 850 nm on an Olympus inverted microscope were counted. Measurements of the total number of living cells, the number of living cells that grew neurites, the number of neurites per field, the length of all neurites in a field and the mean length of all neurites were performed using a computerized digitizing tablet and Sigma-Scan by Jande! Scientific. A dish was considered one data point A minimum of eight valid data points were collected for each concentration of calcium. 32

PAGE 44

The minimum criteria for validity were that, in the controls, there should have been at least 10 living cells, on average, per field and no more than 25 cells per field. More than 25 cells per field would indicate that the cells were not differentiated and that excessive cell division occured. If 100% of the plated cells survived, the average number of cells per field would be about 15 to 20. Cells were judged to be alive if they were round, phase-bright and attached. Also, at least 40% of the cells that were living had to be growing neurites. B50 Cells .. B50 cell cultures were examined for neurite initiation and growth at 48 hours. Dishes were letter coded so that the calcium concentration of the dish was not known until after the count was completed. Growth and initiation were compared to that of the 1X (of MEM) calcium concentration. Three randomly chosen fields that measured approximately 675 nm by 425 nm on an Olympus inverted microscope were counted. Measurements of the number of living cells, the number of living cells that grew neurites, the number of neurites per field, the length of all neurites in a field and the mean length of all neurites were performed using a computerized digitizing tablet and Sigma-Scan by Jandel Scientific. A dish was considered one data point A minimum of eight valid data points was collected for each concentration of calcium. The minimum criteria for validity were that, in the controls, there should have been at least 10 living cells, ort average, per field and no more than 25 cells per field. More than 25 cells per field would indicate that the cells were not differentiated and that excessive cell division occured. If 100% of the plated cells survived, the average number of cells per field would be 20. Cells were judged to be alive if they were round, phase-bright and attached. Also, at least 40% of the cells that were living had to be growing neurites. 33

PAGE 45

Analysis oflntracellular Calcium Levels with Fura-2 Obtaining Images. Calcium concentrations were measured from the ratio of Fura-2 fluorescence intensities at 340 nm and 380 nm light excitation. When the concentration of calcium increases, the 340 nm fluorescence increases and the 380 nm fluorescence decreases (Figure L7). Calcium calibration was performed using Fura-2 pentapotassium salt, a zero calcium EGTA solution and a saturated (2 mM) calcium solution on cover-slips. The ratio of fluOrescence intensity (R) is converted to calcium concentration using the formula [Ca2+] = Rmm)/CRmaxR)] (Fo/F5 ) (Grynkiewicz et al., 1985). These values are calculated from the fluorescence intenstities of two solutions. Rma.x= the ratio of the fluorescence of the saturated calcium calibration solution (340/380). Rmm = the ratio of the fluorescence of the EGTA calibration solution (340/380). FofFs= 380 fluorescence of EGTA solution/380 fluorescence of the saturated calcium solution. the binding constant of Fura-2 for calcium (224 mM). Fura-2 analysis was performed on rat hippocampal cells and N IE-115 cells. Calibration of the Quantex imaging system was performed daily. Cells were loaded by adding 2 J.Ll of 3 mM Fura-2 AM in DMSO (Molecular Probes, Eugene, OR) for a fmal Fura-2 concentration of 3 J.LM. The Fura-2 AM was added directly onto the culture dishes (2 J.LV2 ml). The cells were incubated for 30 minutes at 37C in a 5% carbon dioxide atmosphere. Cells were then rinsed in the same medium in which they were cultured and were incubated another 30 minutes. De-esterfication ofFura-2 AM occurred during .this time. The glass bottom culture dishes were then viewed on a Nikon inverted microscope with an ICCD camera was connected to a Quantex digital ratioed image 34

PAGE 46

proces.sor (Quantex, Santa Clara, CA, see figure 1.6). Images of cells were taken and averaged for 16 frames at each wavelength of light. The wavelength of light was controlled by a computer driven filter wheel in the path of the Xenon lamp. When rat hippocampal cells were studied, the lOOX objective was used. The CCD was set to 100, the camera intensity was set to 0 and no neutral density filter was placed in the path of the Xenon lamp. When the larger N1E-115 cells were studied the 40X objective was used. The CCD was set to 100, the camera intensity was set to 900 and a Nikon neutral density fllter (ND16) was placed in the path of the Xenon lamp. At least io randomly chosen c.ells from any one dish were imaged and were used as one data point. Also, one blank field was imaged in each dish for use as a background standard to be subtracted from all images taken from that dish. At least one dish from every calcium concentration was used in every run and cells were not left in Fura-2 containing medium for more than 3 hours. At least 7 dishes at 10 cells each were imaged for each calcium concentration for rat hippocampal cells. At least 3 dishes at 10 cells each were imaged for N1E-115 cells. Analysis. With each image, the standard background for the dish was subtracted. For each image minus background, a halo value was determined and subtracted from the image. When the halo was subtracted, most neurites were no longer visible. Boxes used to calculate the halo for an image were kept to an approximately constant size. linages were then ratioed and viewed in pseudo-color. Histograms of whole cells were calculated and the tnean calcium concentration was recorded. Dishes with a calculated calcium concentration less than 10 nM were the value of 10 nM due to the insensitivity ofthe machine at low calcium levels. Ail dishes were letter coded so that the extracellular calcium 35

PAGE 47

concentration of the dishes was not known during the itnaging or analysis processes. Statistical Analysis All measurements were tested for significant differences from the mean of the control (IX) cultures. Two-tailed t-tests were performed on all measurements and a significance level of 0.05 was used. In addition to t-tests, analysis of variance (ANOV A) tests were performed on the data when significant differences were found by the t-tests. A significance level of 0.01 was used for the ANOV A tests. 36

PAGE 48

CHAPTER3 EXPERIMENTAL RESULTS Survival. Neurite Initiation and Elongation Rat Hippocampal Neurons There were no significant (p>0.05) differences between hippocampal cells cultured in pre-made 1X MEM and 1X MEM prepared from salt-free in any of the measurements taken. Survival. At 48 hours all cultures containing calcium levels other than 1.8 mM showed significantly lower survival (p<0.05). The mean survival in 1.8 mM Ca2+ was 82.51% 2.02% (SEM) (Table 3.1, Figure 3.2 A). Initiation. Hippocampal cells growing neurites had either a characteristic pyramidal morphology with one long axon and severa1 shorter dendrites, or a more symmetrical array of neurites with no distingilishable axorts or dendrites. The percentage of living neurons with neurites at 48 hours was 41.27 3.85% in the control (1X) MEM. This percentage was calculated by dividing the number of living cells with neurites by the number of cells alive. External calcium levels other than IX resulted in significantly lower initiation except 4X. Analysis of variance (ANOV A) confirmed significant differences among the different concentrations (Table 3.1, Figure 3.2 B). The control value for the total number of neurites per living cell was 1.64 0.25. Cells cultured in OX and 4X medium showed no significant .difference from controls, but the other concentrations

PAGE 49

resulted in significant differences in t-test analysis. ANOV A analysis confinned that-significant differences existed (Table 3.1, Figure 3.3 A); Neurite Elonc-ation. Analysis of mean axon length yielded a value of331 30 Jl.Dl for control cultures. There were no significant differences from the control value in any of the cultures with altered calcium concentrations (Table 3.1, Figure 3.3 B). Figure 3.4 is a comparison of neurite initiation and mean axon length in rat hippocampal cells at 48 hours. Further, there were no differences from controls in mean dendrite length, dendrite branch length or axon branch length. 38

PAGE 50

Figure 3.1. Rat hippocampal neurons and neuroblastoma NlE-115 cells in culture at 48 hours. Embryonic rat hippocampal cells (Above). Mouse (PNS) neuroblastoma N1E-115 (Below). 39

PAGE 51

Table 3.1. Rat Hippocampal Culturing Data. These data were recorded at 48 hours U lated eel" tainin bl al 1 1 All al SEM mce spt mm 1acon vana e c ciUm eve s. v uesare [Ca2+]0 in %Survival % Initiation #Neurites Mean Axon medium per Cell Length (J.Lm) 10:10.1 68.15.14 29.27.77 1.23.19 327 0.45 74.23.12 28.74.31 0.95.16 305 0.90 72.47.11 25.27.21 0.74.15 325 1.80 82.51.02 41.27.85 1.64.25 330 3.60 77.27.56 28.20.08 0.92.06 286 7.20 69.23.89 29.57.64 1.08.14 310 Table 3.2. NlE-115 Culturing Data. These data were recorded at 48 hours Us Ia ed edi . bl al 1 I All al SEM mce :pi t mm a contamio vana e c Cium eve s. v ues are [Ca2+]o in #Alive % Initiation #Neurites Mean Axon medium per Cell Length (Jlm) ,..0,1 12.71.88 60.56.04 4.05.24 99 0.45 13.37.31 56.31.26 4.03.14 108 0.90 14.04.58 58.74.89 3.64.24 98 1.80 12.79.34 58.33.84 3.42.24 94 3.60 12.93.31 54.48.80 3.77.16 92 7.20 13.42.46 58.75.80 3.55.27 105 Table 3.3. BSO Culturing Data. These data were recorded at 48 hours U I ed. ed" . "abl cal. 1 1 All al SEM mce splat m m 1a contamin:! van e c1um eves. v uesare [Ca2+]o in #Alive % Initiation #Neurites Mean Axon medium per Cell Length (Jlm) 10:10.1 13.96.75 88.76.47 1.63.11 55 0.45 12.63.15 90.89.36 1.70.07 59 0.90 14.37.80 94.53.87 1.70.05 60 1.80 14.07.54 88.86.26 1.74.07 61 3.60 12.83.59 91.03.98 1.98.11 69 7.20 12.72.08 88.89.62 1.97.14 67 40

PAGE 52

90 0 J: 80 a: (ij > :::::1 70 (/) 0 60 0 2 3 4 5 6 7 8 [Calcium] o (mM) A 50 40 0 J: a: 30 c 0 20 :e .s 0 10 0 0 2 3 4 5 6 7 8 [Calcium] 0 (mM) 8 Figure 3.2. The effect of [Ca2+]i on survival and neurite initiation in rat hippocampal cells at 48 hours. A) Changing the external calcium concentration from the control value of 1.8 mM results in significantly decreased cell survival at 48 hours. The horizontal line highlights the control value. Statistically significant points are shaded (p<0.05 in a two-tailed t-test). B) Altered calcium levels also result in significantly fewer neurons producing neurites at 48 hours. All values are SEM. 41

PAGE 53

3 a; 0 0 c. c. 2 J: ... CD c. en Q) 1 .:!::: ... :::J Q) z =#:: 0 0 2 3 4 5 6 7 8 [Calcium] 0 (mM) A 400 E :::1. .s:: 360 CD c::: CD ..J 320 c::: 280 c::: ca CD ::E 240 0 :I: a: 200 0 1 2 3 4 5 6 7 8 [Calcium] 0 (mM) B Figure 3.3. The effect of external calcium concentration on the number of neurites per cell and the mean length of axons in rat hippocampal cells. A) Hippocampal cells cultured in 2X. O.SX and 0.2SX calcium show significantly fewer nemites per living cell than controls. The horizontal line the control value. Statistically significant points are shaded. B) Altering external calcium concentration causes no change in mean axon length in hippocampal cells. All values are SEM. 42

PAGE 54

+ C\1 cu 25 .. percent Initiation 0 mean axon length E CX? 0 .... E g -25 G) u c:: Q) .... -50 i5 :::f Q 75 ......... -1 0 1 2 3 4 56 7 8 [Calcium] 0 (mM) Figure 3.4. A comparison of neurite initiation and mean neurite length in rat hippocampal cells cultured in varied external calcium concentrations at 48 hours. Mean neurite length shows no significant difference &om the control value, but the percent initiation greatly reduced when external calcium levels are changed. The graph displays values as a percent diffet:ence from the control (lX) calcium cultures. AU values are SEM. N1E-115 Cells All values were compared to cells cultured.in 1.8 mM ( 1X) calcium. At 48 hours, N1E-115 cells cultured in varied external calcium concentrations displayed no significant differences in survival, number of living cells with neurites or mean neurite length (p>0.05, two-tailed t-test) (Table 32). The only exception was the cells cultured in 0.25X Ca2+; these cells grew slightly longer neurites and, upon further analysis, showed more neurites per cell. ANOV A calculation showed that no significant differences existed among the means of this data set. Figure 3.1 B shows N1E-115 cells in culture at 48 hours. Figure 3.5 A and B shows N1E-115 survival and percent initiation. Figure 3.6 A and B shows the total number of neurites and mean neurite length as a function of external calcium concentration. 43

PAGE 55

20 < 15 J!2 a; (.) an 10 ,... ,... I w ,... z 5 0 0 100 80 w ,... z 60 c .Q 1U E 40c 0 20 0 0 1 2 3 4 5 6 (Calcium]0 (mM) A I I 2 3 4 5 6 (Calcium] 0 (mM) B 7 8 -7 8 Figure 3.5. The effect of calcium on survival and neurite initiation in NlE-115 cells at 48 hours. A) No significant differences in the number ofNlE-115 cells alive per field were found at 2 days. B) Initiation of neurites in NlE-115 cells is not affected by changing the external calcium concentration. The horizontal lines high-light the 1X values. All values are SEM. 44

PAGE 56

Q; 6 0 ll) ,... ,... I w 4 ,... z ... Q) c. en (I) 2 ;:: ;:::::, (I) z ::tt:: 0 0 1 2 3 4 5 6 7 8 [Calcium] 0 (mM} A 175 -E ::I. .c::: C) 125 c:: Q) ....1 .! c ;:::::, Q) 75 z c:: ca (I) 25 0 1 2 3 4 5 6 7 8 [Calcium] 0 (mM) B Figure 3.6. The effect of calcium on the number of neurites per cell and mean neurite length in NlE-115 cells at 48 hours. A) A change in external calcium concentration does not effect the number of neurites per cell in N1E-115 cells. Cells cultured in 0.25X medium showed a significant difference with t-test analysis, but ANOV A showed no significant differences exsisted. B) The mean length of the neurites produced by NlE-115 cells is also not effected by a change in the external calcium concentration with the exception of cells cultured in 0.25X calcium. Statistically significant points are shaded. The horizontal lines high-light the IX values. All values are SEM. 45

PAGE 57

B50 Neuroblastoma Cells Like N1E-115 cells, rat central nervous system neuroblastoma B50 cells showed no significant differences in survival growth or initiation of neurites at 48 hours (Table 3.3). These results are displayed graphically on Figures 3.7 and 3.8. 46

PAGE 58

25 20
PAGE 59

4 a; 0 0 3 LO co '-G) c. 2 ;I;_ X (/) -------:1: .! --c ::::s CD 1 -z =tl:: 0 I I 0 2 3 4 5 6 7 8 [Calcium] 0 (mM) A 100 -E =.. .c eo C) c: 60 ::::s CD z c: 40 ca G) ::! 20 0 2 3 4 5 6 7 8 [Calcium) 0 (mM) B Figure 3.8. The number of oeuriies produced and the mean length of oeurites in BSO neuroblastoma cells. A) Altered external calcium has no effect on the number of oeurites per cell in BSO cells at 48 hours. B) Differences in [Ca2+]0 do not affect the mean length of BSO neurites. Horizontal lines high-light the control value. All values are SEM. 48

PAGE 60

The Effects of Changes in Extracellular Calcium Concentration on Intracellular Calcium Concentration in Rat Hip_pocampal Cells and NlE-115 Cells Rat Hi:gpocampal Neurons Fura-2 analysis of intracellular calcium concentrations at 48 hours in rat hippocampal cells cultured in varied extracellular calcium concentrations showed that differences exist between cells cultured in high and low calcium levels (Table 3.4, Figures 3.9, 3.10 and 3.12). Both t-test analysis and ANOVA confinned this finding. The [Ca2+]i of cells cultured in the control medium proved to be 74 7 nM (mean SEM). The [Ca2+]i of cells cultured in 0.5X (0.9 mM) calcium was not significantly different from control values, but all other extracellular calcium concentrations caused significant changes (two-tailed t-test and ANOV A). [Ca2+]i showed a linear relationship to [Ca2+]0 in media containing from OX to 2X calcium. At 4X, [Ca2+]i began to plateau. The SEM values for [Ca2+]i were less than 20% of the [Ca2+]i for each extracellular concentration. Table 3.4. Internal Calcium Concentration ( SEM) Resulting from External Calcium Levels in Rat Hi al Neurons and NIE-115 Cells [Ca2+]0 Rat Hippocampal NlE-115 (mM) [Ca2+]i (nM) [Ca2+]i (nM) 36 127 35 0.45 47 135 0.90 65 160 10 1.80 Control) 74 132 3.60 124 12 138 7.20 132 13 157 49

PAGE 61

NlE-115 Neuroblastoma Cells Analysis ofN1E-115 cells showed that [Ca2+]i is not affected by the external calcium concentration at 48 hours (Table 3.4 and Figures 3.11 and 3.12). Images ofNlE-115 cells and of rat hippocampal cells were done with different microscope objectives and, therefore, different calibration scales. The result is that the colors on the pictures in Figure 3.10 are the same when the [Ca2+]i is different. Figure 3.9. A ratioed image of a rat hippocampal cell with neurites. Th..is is a photograph of a ratioed Quantex image of a rat hippocampal cell that has background and halo subtracted. After histogram analysis of this cell that was cultured in IX modified MEM, the internal calcium concentration was found to be 74 oM. 50

PAGE 62

A B Figure 3.10. Ratioed images of rat hippocampal cells cultured in 4X and OX MEM. A) This growing cell was cultured for 48 hours in 4X modified MEM. Histogram analysis showed that the internal calcium concentration was 128 nM. 8) This growing cell was cultured for 48 hours in OX m
PAGE 63

Figte 3.11. Ratioed image of an N1E-115 Cell at 48 hours in IX DMEM-F12. Histogram analysis of this image of a growing N 1 E-115 cell showed the internal calcium concentration to be 125 nM. Because different objectives were used (40X for N1E-l15 and lOOX for rat hippocampal cells), this image displays the same pseudo-color characteristics as a rat hipJX>Campal cell with a much lower internal calcium concentration. + (\1
PAGE 64

CHAPTER4 DISCUSSION Previous studies indicate that [Ca2+]i may influence neurite initiation and elongation ( Anglister et al., 1982; Audesirk et al., 1990; Cohan and Kater, 1987; Connor, 1986; Holliday and Spitzer, 1990; Kater et al., 1990; Manalan and Klee, 1984; Mattson et al., 1990; Mills and Kater, 1990; Reboulleau, 1986; Silver et al., 1990; Smith, 1988; Tolkovsky et al., 1990). Changing [Ca2+]0 will change the calcium gradient across the altering the amount of calcium entering a cell. Therefore, the potential exists for changes in [Ca2+]i to occur, depending on the cell's regulatory efficiency for calcium. From this, altered [Ca2+]0 (higher or lower) may decrease neurite initiation to the extent that [Ca2+]i is changed. Low [Ca2+]i may result in greater elongation, while high [Ca2+]i may cause decreased elongation (Kater, 1990) (Figure 1.1). The present stUdy explored the effects of altered [Ca2+]0 on neurite initiation, neurite elongation and cell survival and compared [Ca2+]i of cells cultured in different [Ca2+]0 at 48 hours. This study indicates that the transformed cell N1E-115 can regulate [Ca2+]i more efficiently than rat hippocampal neurons; therefore, in N1E-115 cells neurite initiation, growth and cell survival are not altered by changes in [Ca2+]0 These parameters are affected in rat hippocampal neurons when [Ca2+]0 is altered. Furthermore, the transformed CNS cell B50 is not affected by altered [Ca.2+]0 with respect to initiation, elo11gation or survival; we predict that [Ca2+]i is also probably unchanged.

PAGE 65

Culturin& Experiments Survival Incubation in media containing different calcium levels affects non-transformed rat hippocampal cells, but seems to have no effect on transformed cell lines. At 48 hours, N lE-115 and B50 cells showed no significant change in survival in all [Ca2+]0 compared to controls, but hippocampal cells showed a decrease in survival in media with higher or lower than control [Ca2+]0 (Tables 3.1, 3.2, 3.3; Figures 3.5, 3.7). This would indicate that in hippocampal cells, [Ca2+]i changed when [Ca2+]0 changed Altered [Ca2+]i would affect cell survival due to decreased regulation of calcium dependent reactions. This change in [Ca2+]i could affect cell metabolism, resulting in decreased cell survival (Komulainen et al., 1987; Rawn, 1990; Siesjo, 1990). It should be noted that in decreased calcium medium, hippocampal cells were less able to attach to the culture dish bottom resulting in decreased survival counts. This could mean that true survival might not be affected, just "attachability." Initiation When [Ca2+]i changed, neurite initiation was affected. Rat hippocampal cells showed a decrease in neurite initiation when [Ca2+]0 increased or decreased, but N1E-115 and B50 cells showed no differences. This was probably due to the changes in [Ca2+]i in the hippocampal cells. When the calcium gradient across the plasma membrane was enhanced, the [Ca2+]i in the hippocampal cells increased (Table 3.4, Figures 3.9, 3.12). Increased or decreased calcium may have inhibited neurite initiation by interfering 54

PAGE 66

with the polymerization and motility of actin fl.laments (Smith, 1988). In the absence of proper filament motill.tr, riew growth could not develop. Because initiation was not completely stopped, not all types of emerging growth cones may have been affected by these changes in [Ca2+]0 Different types of growth cones with different dependences on calcium have been found within the same cell (Kater et al., 1990). It may be that optimal calcium ranges for initiation are somewhat wide, but when the limits are approached, a decrease in initiation occurs. In N1E-115 and B50 cells, neurite initiation was unaffected by altered calcium influx across the membrane. [Ca2+]i was also unchanged in NlE-115 cells by altered [Ca2+]0 This indicates that in NlE-115 cells, calcium regulatory mechanisms must exist that are more efficient than those found in the rat hippocampal cells. Perhaps because [Ca2+]i was uilchanged, neurite initiation occurred with the same frequency in cells cultured in all of the [Ca2+]0 media. [Ca2+]i was not tested in B50 cells, but because initiation was not affected, it could be that [Ca2+]i was unchanged in the B50 cultures. Another possibility is that the dbcAMP used to differentiate the B50 cells affected neurite initiation to a greater extent than a change in [Ca2+]i may have. Other data have shown that cAMP can affect initiation (Mann et al., 1989; Reboulleau, 1986; Rydel and Greene, 1988). Based on these data, it is also possible that a flux of calcium across the membrane is required for initiation to occur. The exact [Ca2+]i may not be directly related to the initiation of new neurites. Other studies have reported similar findings. For example, a flux of calcium caused by VSCCs in growth cones appears to be necessary for cone motility in N1E-115 cells (Silver et al., 1990). Furthermore, Mattson et al. ( 1990) showed lbat a brief enhanced calcium gradient within a cell seemed to stimulate axon inithition in rat hippocampal cells. The 55

PAGE 67

[Ca2+]i of a cell may affect the size of the flux by altering membrane channels. [Ca2+]i is known to activate and inhibit calcium channels directly and via the activation of other kinases (Kennedy, 1988; Rawn, 1990). Audesirk eta/. (1990) showed that the blockage of calcium channels caused a decrease in initiation in N1E-115 cells. Therefore, it is possible that a flux of calcium and not [Ca2+]i directly that is responsible for stimulating initiation. In this study, a gradient of calcium always existed, even in the 0.1 mM medium. If [Ca2+]i was not changed, then the flux may not have been disrupted. However, it can not be ruled out from the Audesirk eta/. study that a decrease in [Ca2+]i caused the decrease in initiation because [Ca2+]i was not tested. Elongation Elongation was not dramatically changed in any of the cells with altered [Ca2+]0 No significant differences were seen in any of the cell types. Elongation may not be dependent on calcium once initiation has occurred (Tolkovsky et al., 1990). When [Ca2+]i was reduced in rat hippocampal cells, longer neurites were not seen as expected. Likewise, when [Ca2+]i was increased in the rat hippocampal cells, elongation was not decreased (Table 3.1; Figure 3.2). Because calcium inhibits microtubule polymerization, one would expect higher [Ca2+]i to inhibit growth. It could be that the optimal [Ca2+]i range for elongation is quite wide or that a different form of calcium regulation exists in neurite shafts than in cell bodies that helps to maintain low to moderate calcium levels. For example, neurite shafts have few VSCCs that would allow calcium into the shaft in response to depolarization (Silver et al., 1990). This alone may have prevented a decrease in elongation when [Ca2+]i in the soma increased. 56

PAGE 68

The [Ca2+]i of neurites was not explored in thisexperiment; however, it was noted that neurites consistently had a lower [Ca2+]i than cell bodies because during the image analysis process, neurites were subtracted away from the soma images when halos were subtracted. During the analysis of images, it is necessary to subtract the "halo," or glow, from around a cell body to get an accurate measurement of [Ca2+]i in the soma. The imaging system "subtracts" the calculated value of the halo intensity from the entire image. Regions of the image with low calcium levels are also subtracted dining this process if their [Ca2+]i levels are lower than the value calculated for the ha,lo. This would indicate neurites had lower [Ca2+]i than cell bodies. Because lower [Ca2+]i did not increase elongation (Tables 3.1, 3.2, 3.3; Figures 3.3, 3.4, 3.6, 3.8), there may be a threshold of calcium inhibition of microtubule polymerization. [Ca2+]i was below the level that causes inhibition in all of these cultures; microtubule assembly was therefore unaffected by calcium and occurred at an uninhibited rate. Similar results with elongation have been reponed in embryonic chick cells (Audesirk et al., 1990). [Ca2+]i was not tested in this study, however. Mattson and Kater (1987) showed that in the snail Helisoma, neurite elongation was reduced by increased [Ca2+]i. Helisoma may have a lower threshold of calcium for neurite elongation than rat hippocampal cells, chick cells, or N1 E-115 cells. Therefore, in Helisoma, increased [Ca2+]i may have inhibited elongation by interfering with microtubule polymerization, as previously discussed. 57

PAGE 69

Fura-2 Studies Fura-2 digital analysis demonstrated that rat hippocampal neurons and N1E-115 cells regulated [Ca2+]i differently in altered [Cai+]0 The [Ca2+]i in rat hippocampal cells changed linearly with [Ca2+]0 but N1E-115 cells showed no changes (Table 3.4; Figures 3.11, 3.12). Fura-2 analysis of BSO cells was not conducted. Causes of Differential Regulation of Internal Calcium Concentration in Two Cell TXPeS There are two likely explanations of the differences in [Ca2+ li in high [Ca2+]0 The first is the differential regulation of plasma membrane and endoplasmic reticulum calcium A TPases. Calcium A TPases have been called the most important component in the maintenance of [Ca2+]i. The second is the up and/or down regulation ofVSCCs on the plasma membrane (Carafoli, 1987; Nicholls, 1986). N1E-115 cells may have a greater ability to increase the activity of calcium A TPases in times of rising [Ca2+]i. One possibility would be that calcium, calcium-bound protein kinases, or phosphatases act as positive effectors on calcium A TPase. For example, calcium A TPase is phosphorylated directly by cAMP-dependent protein kinase in heart cells (Neyes et al., 1985). As [Ca2+]i increases, the calcium ATPase activity would increase. This ability may be enhanced in the transformed N1E-115 cells, affording better regulation of calcium A TPases when [Ca2+]i is increasing. In a similar manner, a decrease in [Ca2+]i resulting from lower [Ca2+]0 would cause reduced calcium A TPase activity because the kinases or phosphatases that are activated by calcium would be less active. This would slow down the extrusion of calcium from the cell when [Ca2+]i was too low. 58

PAGE 70

A second possibility enhanced regulation of [Ca2+]i in decreasing [Ca2+]0 in N1E-115 cells involves plasma membrane calcium channels. Calcium channel blockers have been shown to cause a decrease in initiation in N1E-115 cells (Audesirk et al., 1990a). It could be pos&ible that this channel blockage lead to a decreased [Ca2+]i; this was not tested, however. There may be a calcium-activated protein that is a negative effector of calcium channels on the membrane that, in times of reduced [Ca2+]i. would enhance calcium influx, because the inhibition of the channels would be removed. Voltage sensitive calcium channels that are inhibited by calcium-boWld calmodulin have been foWld in other nerve cell (Kennedy, 1988; Neyes et al., 1985). In a similar fashion, N1E-115 cells may also be able to release calcium from intracellular stores more effectively than rat hippocampal cells using a pathway involving inositol triphosphate. Rat hippocampal cells may have less extensive calcium stores or less efficient sodium/calcium exchange proteins. It is possible that the increased transcription of DNA seen in many transformed cells results in an increased number of regulatory proteins in the N1E-115 cells, affording them greater protection from calcium as a by-product of their transformation. Differences may also be due to differences in PNS versus CNS, but because neurite initiation in BSO (CNS) cells and NlE-115 (PNS) cells were not affected by [Ca2+]0 changes, whereas initiation in rat hippocampal neurons and chick brain cells is affected by [Ca2+]0 the differences probably result from transformation. It should be noted, however, that BSO cells were cultured with increased cAMP levels that may have affected initiation and growth. Also, Fura-2 studies were not done on BSOs or chick brain neurons. 59

PAGE 71

Further Study [Ca2+]i at 48 hours may not reflect the mean [Ca2+]i during the entire 2-day incubation period, since calcium may be affected drastically during the first few hours after plating. These potentially bigger changes in [Ca2+]i in the short term may have been responsible for the decrease in initiation. [Ca2+]i at shorter-term time points should be explored to study the pattern in [Ca2+]i over the 2-day period. Particular attention could be paid to the few hours after plating where much initiation occurs. More Fura-2 experiments should be completed on N1E-115 cells to decrease the variability obtained in [Ca2+]i (SEM was up to 39 percent). The general fmding that [Ca2+]i in N1E-115 cells is not affected by [Ca2+]0 would probably not change, however. An assay of the activity of calcium A TPase would be a very valuable addition to these experiments. If the activity of these calcium pumps could be analyzed in the two cell types, information on the mechanisms involved in the enhanced regulation of [Ca2+]i in N1E-115 cells could be obtained. These pumps may be regulated directly by [Ca2+]i, calmodulin, or another calcium-dependent kinase or phosphatase. One would expect N 1 E-115 cells to show greater calcium A TPase activity than rat hippocampal cells when [Ca2+]0 increased, and possibly lower calcium A TPase activity when [Ca2+]0 decreased. There may be differences in DNA transcription between the two cell types. Calcium is known to affect DNA-binding proteins in many cells (Kennedy, 1988). Slight increases or decreases in calcium flux across the membrane may result in the increased or decreased transcription or translation of calcium A TPase proteins or 60

PAGE 72

VSCCs. This would result in increased or decreased numbers of these proteins on the membrane to alleviate an abnormal fltpt of calciU:Iil. Studies of DNA transcription and RNA translation of calcium channels and calcium A TPases would provide information about the regulation of [Ca2+]i when [Ca2+]0 changes. The flux of calcium across the membrane rather than the [Ca2+]i may be responsible for initiation in some neurons, because calcium channel blockers, but not reduced [Ca2+]0 cause a decrease in initiation in N1E-115 cells (Audesirk et al., 1990). A study with the calcium channel blocker dihydropyridine similar to the one conducted by Audesirk et al. (1990) should be conducted on N1 E-115 cells. Such a study should include an analysis of [Ca2+]i. This would provide information on the specific mechanism of calcium in the initiation of new neurites. If initiation is found to be decreased in the presence of the channel blocker, but [Ca2+]i is unchanged, this would support the hypothesis that an influx of calcium is the required prerequisite for initiation. Completion of the above studies will provide valuable information on the regulation of [Ca2+]i and its role in neurite initiation and elongation. 61

PAGE 73

LITERATURE CITED Anglister, L., Farber, 1., Shahar, A., Grinvald, A. (1982). Localization of voltage sensitive calcium channels along developing neurites: their possible role in regulating neurite elongation. Developmental Biology. 94:351-365. Audesirk, G., Audesirk, T., Ferguson, C., Lomme, M., Shugarts, D., Rosack, J., Caracciolo, P., Gisi, T., Nichols, P. ( 1990a). LType calcium channels may regulate neurite initiation in cultured chick embryo brain neurons and N1E-115 neuroblastoma cells. Developmental Brain Research. 494:421-34. Bean, B. (1989). Classes of calcium channels in vertebrate cells. Annual Review of Physiology. 51:367-384. Hottenstein, J. (1985). Culture methods for growth of neuronal cell lines in defined media. In D.W. Barnes, G.H. Sato (Eds). Methods for Serum Free Culture of Neuronal and Lymphiodal Cells, Liss, New York, pp3-13. Campenot, R. and Draker, D. (1989). Growth of sympathetic nerve fibers in culture does not require extracellular calcium. Neuron. 3:733-743. Carafoli, E. (1987). Intracellular calcium homeostasis. Annual Review of Biochemistry. 56:345-443. Choi, D. (1988). Calcium mediated neurotoxicity: relationship to specific channel types and role in ishemic damage. TINS. 11:465-468. Cohan, C. and Kater, S. (1987). Electrically and chemically mediated increases in intracellular calcium in neuronal growth cones. Journal of Neuroscience. 7:3588-3599. Connor, J. (1986). Digital imaging of free calcium changes and of spatial gradients in growing processes in single, mammalian central nervous system cells. Proc. National Acad. of Science. 83:6179-6183. Connor, J., Kater, S., Cohan, C., Fink, L. (1990). Calcium dynamics in neuronal growth cones: regulation and changing patterns of calcium entry. Cell Calcium. 11:233-239. Duprat,A. and Kan, P. (1981 ). Stimulating effect of the divalent cation ionophore A 23187 on in vitro neuroblast differentiation; comparative studies with myoblasts. Experimentia. 37:154-158.

PAGE 74

Fletcher, T. and Banker, G. (1989). The establishment of polarity in hippocampal neurons: the relationship btween the stage of a cell's development in situ and its subsequent development in culture. Developmental Biology. 136:446-454. Grynkiewicz, G., Poenie, M., Tsien, R. (1985). A new generation of calcium indicators with greatly improved fluorescence properties. Journal of Biological Chemistry. 260:3440-3450. Haugland, R. (1989). Fura-2 excitation intensity at various calcium concentrations. Molecular Probes Handbook of Fluorescence Probes and Research Chemicals. p.78. Holliday, J. and Spitzer, N. (1990). Spontaneous calcium influx and its roles in differentiation of spinal neurons in culture. Developmental Biology. 141:1323. Kater, S. and Mattson, M. (1988). Calcium regulation of the neuronal growth cone. TINS. 12:315-320. Kater, S., Mills, L., Guthrie, P. (1990). Intracellular calcium and the control of neuronal growth and form. Trophic Factors and the Nervous System.L.A. Horrocks (Ed). Raven Press Ltd. New York. p.231-245. Kennedy, K. (1988). Regulation of neuronal function by calcium. TINS. 12:417-421. Koike, T., Martin, D., Johnson, E. (1989). Role of calcium channels in the ability of membrane depolarization to prevent neuronal death induced by trophic factor deprivation: evidence that levels of internal calcium determine nerve growth factor dependence of ganglion cells. Proc. National Acad. Science. 86:64216425. Komulainen, H. (1988). Increased free cellular calcium by toxic agents: an index of potential neurotoxicity? TIPS. 5:231-234. Kruskal, B., Shak, S., Maxfield, F. (1986). Spreading of human neutrophils is immediately preceded by a large increase in cyotoplasmic free calcium. Proc. National Acad. Science. 83:2919-2923. Malgoroli, A., Milani, D., Meldolesi, J., Pozzan, T. (1987). Fura-2 measurement of cytosolic free calcium in monolayers and suspensions of various types of animal cells. Journal of Cell Biology. 105:2145-2155. Manalan, A. and Klee, B. (1984). Calmodulin and cyclic nucleotide protein phosphorylation. Research. 18:227-278. Mann, D., Doherty, P., Walsh, F. (1989). Increased intracellular cyclic AMP differently modulates nerve growth factor induction of three neuronal recognition molecules involved in neurite outgrowth. Journal of Neurochemistry. 53:581-588. 63

PAGE 75

Matthews,E. (1986). Calcium ancl membrane permiability. British Medical Bulletin. 42:391-397. Mattson, M and Kater, S. (1987). Calci\nn regulation of neurite elongation and growth cone motility. Journal of Neuroscience. 7:4034-4043. Mattson, M., Murain, M., Guthrie, P. (1990). Localized calcium influx orients axon formation in embryonic hippocampal pyramidal neurons. Developmental Brain Research. 52:201-209. Meldolesi, J. and Volpe, P. (1988). The intracellular distribution of calcium. TINS. 11 :449-452. Miller, R. (1988). Calcium signaling in neurons. TINS. 11:415-419. Mills, L.and Kater, S. (1990). Neuron specific and state specific differences in calcium homeostasis regulate the generation and degeneration of neuronal architecture. Neuron. 2:149-163. Nahorski, S. (1988). Inositol phosphates and neuronal calcium homeostasis. TINS. 11:444-448. D. (1985). Regulation of cytosolic calcium concentration in presynaptic nerve endings isolated from rat brain. Journal of Physiology. 363:87-101. Neyes, L., Reinlib, L., Carafoli, E. (1985). Phosphorylation of the calcium pumping A TP ase of heart sarcoleiiiiiut and erythrocyte plasma membrane by the cyclic AMP dependent protein kinase. Journal of Biological Chemistry. 18:10283-10287. Nicholls, D. (1986). Intracellular calcium homeostasis. British Medical Bulletin. 42:353-358. Ozawa, S., Tsuzuki, M., Ogura, A., Kudo, Y. (1989). Three types of of voltage dependent calcium current in cUltured rilt hippocampal neurons. Brain Research. 495:329-336. Pang, P., Wang, R., Karpinski, J., Shan, J., Benishin, A.(1990). Control of calcium channels in neuroblastoma cells N1E-115. Experimental Gerentology. 25:247-253. Rasmussen, H. (1989). The cycling of calcium as an intracellular messenger. Scientific American. 42:66-73. Rawn, J. (1990). Biochemistry. Patterson Publishers. Burlington, North Carolina.pp:253-257; 352-353. 64

PAGE 76

Reboulleau, C. ( 1986). Extracellular calcium induced neuroblastoma cell differentiation:involvement of phosphatidylinosittol turnover. Journal of Neurochemistry. 46:920-930. Rogers, M and Hendry, I. (1990). Involvment of dihydropyrodine-sensitive calcium channels in nerrve growth factor-dependent neurite outgrowth by sympathetic neurons. Journal of Neuroscience. 26:447-454. Rydel, R.and Greene, L. (1988). Cyclic AMP analogs promote survival and neurite outgrowth in cultures of rat sympathetic and sensory neurons independently of nerve growth factor. Proc. National Acad. Science. 85:1257-1261. Schubert, D., Heinemann, W., Carlisle, H., Kimes,B., Patrick, J., Steinbach, J., Culp, W., Brant, L. (1974). Clonal cell lines from the rat central nervous system. Nature. 249:224-227. Scott, R. (1990). Voltage-dependent modulation ofrat sensory neurone calcium channel currents by G protein activation: effect of a dihydropyridine antagonist British Journal of Pharmacology. 99:629-630. Siesjo, B. (1990). Calcium in the brain under physiological and pathalogical conditions. European Neurology. 30(supp2):3-9. Silver, A., Lamb, A., Bolsover, S. (1989). Elevated cytosolic calcium in the growth cone inhibit neurite elongation in neuroblastoma cells: correlation of behavioral states with cytosolic calcium concentration. Journal of Neuroscience. 9:4007-4020. Silver, A., Lamb, A., Bolsover, S. (1990). Calcium hotspots caused by L-channel clustering promote morphological changes in neuronal growth cones. Nature. 343:751-754. Smith, S. (1988). Neuronal cytomechanics: the actin-based motility of growth cones. Science. 242:708-715. Tank, D.,Sugimori, S., Connor, J. (1988). Spatially resolved calcium dynamics of mammalian cells in cerebral slice. Science. 242:773-776. Thayer, S., Perney, P., Miller, R. (1988). Regulation of calcium homeostasis in sensory neurons by bradykinin. Journal of Neuroscience. 8:4089-4097. Tolkovsky, A., Walker, A., Murrell, R. (1990). Calcium transients are not required as signals for long term neurite outgrowth from cultured sympathetic neurons. Journal of Cell Biology. 110:1295-1306. Toselli, M., Taglietti, V. (1990). Pharmacalogical characterization of voltage dependent calcium currents in rat hippocampal cells. Neuroscience Letters. 112:70-75. 65

PAGE 77

Tsien, R., Rink, T., Poenie,M. (1985). Measurement of cytosolic free calcium in individual small cells using flourescence microscopy with dual excitation wavelengths. Cell Calcium. 6:145-157. Wahl, P., Schousboe, A., Honore, T., Dreijer, J. (1989). Glutimate induced increase in intracellular calcium in cerebral cortex neurons is transient in immature cells but permanent in mature cells. Journal of Neurochemistry. 53:1316-1319. :. 66