| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Istituto di Ricerche Farmacologiche [Mario Negri], Consorzio Mario Negri Sud, Department of Pharmacology and Molecular Pathology, Chieti, Italy
1Correspondence: Istituto di Ricerche Farmacologiche [Mario Negri], CMNS, 66030 Santa Maria Imbaro, Chieti, Italy. E-mail: Pessia{at}cmns.mnegri.it
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: potassium channel gating • hKv1.1 • hKv1.2· myokymia • cerebellum
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2
3)
. Interictal nystagmus and migraine are not usually
observed in EA-1 patients. In contrast, they are distinctive symptoms
of individuals affected by episodic ataxia type-2, which has been
associated with mutations in the P/Q-type calcium channel 
1 subunit
gene CACNL1A4 (4)
.
Several heterozygous point mutations have been found in the coding
sequence of the voltage-gated potassium channel gene KCNA1
(hKv 1.1) of various EA-1-affected families. These mutations occur at
highly conserved positions in the transmembrane domain 1 and 2 (S1 and
S2), in the loops linking S2 to S3 and S4 to S5, and in S6
(5
6
7
8
9
10
11)
. The characterization of human Kv1.1 channels
bearing a number of EA-1 mutations in Xenopus oocytes has
revealed that these amino acid substitutions reduce their
delayed-rectifier function by altering several potassium channel
biophysical and biochemical properties (12
13
14
15)
.
Potassium channel diversity is greatly enhanced by the ability of
different types of subunits to heteropolymerize and form channels
with properties different from the parental homomeric channels
(16
17
18
19)
.
Biochemical and biophysical studies have shown that mammalian Kv1.1 and
Kv1.2 subunits are colocalized in several subcellular brain regions
important for the control of movement and heteropolymerize to form
channels (16
, 18
, 20
21
22
23
24
25)
. Particularly intriguing is the
presence of both subunits in the Ranvier's nodes of myelinated axons
and at the level of the cerebellar Pinceau, a structure composed of
several basket cell terminals that embrace the Purkinje cell axon
hillock and proximal axon segment (22
, 23
, 25)
. Purkinje
cell axons represent the only output system of the cerebellar cortex
and the Pinceau appear to passively hyperpolarize these axons by
generating fast inhibitory electrical fields (26)
.
Moreover, patch-clamp recordings from Purkinje cells have revealed that
-DTX selectively blocks the Kv1.1 and Kv1.2 potassium channels from
basket cell presynaptic terminals and increases both the amplitude and
frequency of spontaneous IPSCs mediated by GABAA
receptor activation (27)
. These findings suggest that
Kv1.1 and Kv1.2 heteromeric channels may contribute to the excitability
and the rapid repolarization phase of action potentials in myelinated
axons and basket cell terminals, where they modulate the release of the
neurotransmitter 
-aminobutyric acid (GABA) onto Purkinje cells.
To ascertain whether heteropolymerization occurs between the human potassium channel members hKv1.2 and hKv1.1 and whether EA-1 mutations affect heteromeric channel function, we determined the biophysical properties of channels obtained from the coexpression of both mRNAs in Xenopus oocytes, as well as of channels comprised of the concatenated dimeric subunits, hKv1.2 linked to mutant or wild-type hKv1.1. The results demonstrate that human Kv1.2 and Kv1.1 subunits coassemble to form a distinct channel and that EA-1 mutations profoundly impair several of the novel heteromeric K+ channel properties, thus uncovering new physiopathogenetic mechanisms of episodic ataxia type-1 myokymia syndrome.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 28)
. The nucleotide sequences of all
linked subunits were determined throughout the joined segments by
automated sequencing. Oligonucleotides were obtained from EUROBIO,
Milan, Italy.
Electrophysiology
Xenopus oocytes dissection and injections were
performed as described (12)
. The amount of mRNA injected
was determined by spectrophotometer and ethidium bromide stain.
Two-electrode voltage-clamp recording
The oocyte currents were recorded 1–8 days later with a
GeneClamp 500 amplifier (Axon Instruments, Foster City, Calif.)
interfaced to a Power Macintosh 7200/90 computer with an ITC-16
computer interface (Instrutech Corp., Great Neck, N.Y.). The recording
solution contained (mM): NaCl 96, KCl 2, MgCl2 1,
CaCl2 1.8, HEPES 5, pH=7.4. Data acquisition and
analysis were performed using Pulse+PulseFit (HEKA elektronik GmbH,
Germany), KaleidaGraph (Synergy Software), and IGOR (Wavemetrics)
software. Leak and capacitative currents were subtracted by using a P/4
protocol.
Patch-clamp recording
Inside-out patch recordings were performed with an Axopatch 200B
amplifier (Axon Instruments). The pipette solution contained (mM): NaCl
120, KCl 2, CaCl2 0.1, HEPES 5, pH=7.4; the
cytoplasmic solution contained (mM): KCl 120, EGTA 1, HEPES 5, pH=7.4.
Pipettes were pulled from 7052 glass type (Garner Glass Co., Claremont,
Calif.) and had a resistance of 5–10 M
. The recordings were leak
subtracted by using averaged sweeps with no opening fitted with
exponential functions, filtered at 0.5–2 kHz with a four pole low-pass
Bessel filter and acquired at 5–10 kHz with a Pulse+PulseFit program
(HEKA elektronik GmbH). Channel activity was analyzed (as described by
Hoshi et al., 29
) with a TAC-TACfit program (Bruxton Co.,
Seattle, Wash.) by visual inspection and measurement of channel
openings (event-by-event mode). Duration histograms were corrected for
the corner frequency of the low-pass filter used.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
), suggesting that these subunits may assemble to form heteromeric
channels with different relative stoichiometries.
|
To control the stoichiometry of the coexpressed subunits to determine
the biophysical parameters of a presumed homogeneous population of
heterotetrameric channels composed of two subunits of each type, we
concatenated the hKv1.2 with the hKv1.1 subunit (hKv1.2–1.1wt) with a
flexible glutamine linker. The expression of the hKv1.2–1.1wt
concatemer in Xenopus oocytes gave rise to functional
delayed rectifier currents displaying voltage-dependent parameters
intermediate between the respective homomers, confirming the results
obtained form the coexpression experiments (Fig. 1
A, B; Table 1
).
|
To compare the kinetics of activation and deactivation for hKv1.1,
hKv1.2 and hKv1.2–1.1wt channels, activating and deactivating current
traces were fitted with double and single exponential functions,
respectively (12
, 15)
. The plot of time constants as a
function of membrane potential (Fig. 1C
) revealed that
hKv1.2–1.1wt channels had kinetics of activation distinct and
intermediate between the respective homomeric channels.
A similar sensitivity to tetraethyl-ammonium (TEA) ion blockade of
Drosophila Shaker channels was shown by hKv1.1
homotetrameric channels (Ki= 0.368 ± 0.02 mM;
Fig. 1D
; 13, 15). The tyrosine residue 379, located in the
outer mouth of the hKv1.1 pore endows the channel with high sensitivity
to TEA ions (Fig. 2
A; 13). In contrast, hKv1.2 channels possess a valine residue
at this position and showed a much lower affinity to TEA
(Ki= 232 ± 7.5 mM; Fig. 1D
). On the
other hand, heteromeric hKv1.2–1.1wt channels displayed a TEA
sensitivity intermediate between the homomeric hKv1.2 and hKv1.1
channels (Ki= 85.6 ± 14.8 mM; Fig. 1D
). To investigate the role of the tyrosine residue
endowing concatenated heteromeric channels with intermediate
sensitivity to TEA, we constructed a dimer in which a hKv1.2 subunit
was linked to a hKv1.1 subunit with the Tyr 379 residue mutated into a
valine (hKv1.2–1.1Y379V). The TEA concentration-response curve for
hKv1.2–1.1Y379V channels was shifted toward the hKv1.2 values
(Ki= 200 ± 15 mM; Fig. 1D
). Taken
together, these results demonstrate that human heteromeric
K+ channels composed of hKv1.2 and hKv1.1
subunits may be formed. Moreover, the data suggest that the expression
of hKv1.2–1.1wt linked subunits in Xenopus oocytes gives
rise to heterotetrameric channels containing two subunits of each type,
which are expected to be mostly located tandemly (30)
.
|
E325D and V408A mutations alter the voltage dependence, kinetics,
and inactivation of heteropolymeric channels
To determine whether EA-1 mutations affect the functions of
heteromeric channels, hKv1.2 subunits were linked as dimers with
hKv1.1E325D and hKv1.1V408A mutant subunits (hKv1.2–1.1E325D;
hKv1.2–1.1V408A; Fig. 2A
) because they markedly alter
several gating properties of the channel (12
, 13)
. The
expression of hKv1.2–1.1V408A dimeric construct in Xenopus
oocytes produced functional channels with voltage-dependent parameters
not significantly different from 1.2–1.1wt channels (Fig. 2B, C
; Table 1
). In contrast, the half-maximal activation voltage
V1/2 for hKv1.2–1.1E325D channels was shifted
21.8 mV to more positive potentials and the slope factor k was
increased 1.5-fold compared with 1.2–1.1wt channels (Fig. 2C
; Table 1
). These results demonstrate that the voltage
dependence of heteromeric channels is altered by the assembling of
hKv1.1E325D subunits.
To determine whether E325D and V408A mutations affect the kinetics of
activation and deactivation of heteromeric channels, hKv1.2–1.1E325D
and hKv1.2–1.1V408A current traces were fitted with exponential
functions and the time constants were plotted as a function of test
potential (Fig. 2D
). Such analysis revealed that
hKv1.2–1.1E325D and hKv1.2–1.1V408A channels activate at
V1/2 with ~twofold and ~fivefold faster
kinetics than hKv1.2–1.1wt channels, respectively (Fig. 2D
;
Table 1
). Similarly, the time constants of deactivation at
V1/2 for hKv1.2–1.1E325D and hKv1.2–1.1V408A
channels were ~eightfold and ~sixfold faster than hKv1.2–1.1wt
channels. The data demonstrate that both E325D and V408A mutations
alter the kinetics of activation and deactivation of heteromeric
channels and may destabilize their open state.
In previous studies it was reported that E325D and V408A mutations
increased the rate of C-type inactivation (12
, 13
, 15)
. To
assess whether such mutations affect also the inactivation properties
of heteromeric channels, the membrane potential of oocytes expressing
hKv1.2–1.1E325D and hKv1.2–1.1V408A channels was held at -80 mV and
current traces were evoked by depolarizations at +20 mV for 3.5 min.
The superimposed current traces presented in Fig. 3
A clearly show that hKv1.2–1.1E325D and hKv1.2–1.1V408A
channels undergo a faster C-type inactivation process than
hKv1.2–1.1wild-type channels. The decaying phase of the currents was
best fitted with double exponential functions and the fast time
constants for hKv1.2–1.1E325D and hKv1.2–1.1V408A channels were
~1.6-fold and ~3.4-fold faster than hKv1.2–1.1wild-type channels,
respectively (Table 2
).
|
|
The recovery from C-type inactivation of the heteromeric mutant
channels was also faster (Fig. 3B
). In fact, the slow time
constants for hKv1.2–1.1E325D and hKv1.2–1.1V408A channels were
~3.7-fold and ~2.7-fold faster than hKv1.2–1.1wild-type channels,
respectively (Table 2)
. These results demonstrate that heteromeric
channels comprising either hKv1.1E325D or hKv1.1V408A subunits undergo
a faster C-type inactivation process and possess a less stable
inactivated state.
V408A mutation alters the single-channel kinetics of homomeric and
heteromeric channels
Valine 408 is a highly conserved residue located in the
carboxyl-terminal portion of S6 that lies in the ion-conducting pore
(Fig. 2A
; 31, 32). Moreover, this region appears to comprise
part of an intracellular gate that hypothetically opens and closes
Shaker K+ channels (33)
.
To investigate the effects of V408A mutations on homomeric and
heteromeric single-channel kinetics, inside-out patch-clamp recordings
were performed with Xenopus oocytes expressing
hKv1.2–1.1wt, hKv1.2–1.1V408A, and hKv1.1V408A channels (Fig. 4
). Heteromeric hKv1.2–1.1wt channels showed a single-channel slope
conductance of 9.6 ± 1.2 pS (n=3) that was not
significantly different from the hKv1.2–1.1V408A and hKv1.1V408A
conductances of 9.7 ± 0.6 pS (n=3) and 10.0 ±
1.0 pS (n=5), respectively. In contrast, marked differences
were observed when the open dwell life times of the three channel types
were analyzed (Fig. 5
). The frequency distributions of the open durations for
hKv1.2–1.1wt, hKv1.2–1.1V408A, and hKv1.1V408A channels were
well fitted with a single exponential probability density function and
the time constants resulted to be 14.3 ± 0.8 ms, 8.5 ± 0.6
ms and 4.8 ± 0.6 ms (n=3), respectively (Fig. 5)
.
These results demonstrate that V408A mutation reduces the mean open
duration of homomeric channels ~threefold, probably destabilizing the
conformational changes this region of the channel protein undergoes
after a depolarization (33)
. In addition, the data
demonstrate that this alteration is proportionally transferred to the
heteromeric complex and that the Val 408 residue plays a pivotal role
in the stability of the open state of the channel.
|
|
EA-1 mutations affect cell resting potential
Previous studies have shown that E325D mutations markedly
reduce channel expression, possibly because most of the protein remains
in the endoplasmic reticulum where it is subsequently degraded
(12
, 13)
. To determine the effects of such mutation on the
expression pattern of heteromeric channels, we coinjected in
Xenopus oocytes 2.5 ng of hKv1.2 mRNA with 2.5 ng of
hKv1.1wt or hKv1.1E325D mRNA. The relevant average whole-cell currents
were plotted in comparison with the currents obtained from oocytes
injected with 5 ng of hKv1.1wt, hKv1.1E325D, or hKv1.2 mRNAs alone
(Fig. 6
As expected, E325D mutation reduced the expression level of homomeric
channels by severalfold and, the currents obtained from oocytes
coinjected with equal amounts of hKv1.2 and hKv1.1E325D mRNA were
~0.5 of the control currents.
|
Homomeric hKv1.1E325D channels showed a half-maximal activation
voltage V1/2 and a slope factor k of 29.4 ±
1.9 mV and 24.0 ± 1.0 mV, respectively (12
, 13)
.
These values were ~2.2-fold and ~3.7-fold greater than hKv1.2
homomers (Table 1)
. On the other hand, channels resulting
from the expression of hKv1.2–1.1E325D concatemer, which are
expected to be composed of two subunits of each type, possess
voltage-dependent parameters intermediate between hKv1.2 and
hKv1.1E325D homomers (Table 1)
. Therefore, measurement of the
voltage-dependent parameters of currents resulting from the
coexpression of equal amounts of hKv1.2 or hKv1.1E325D mRNA gives an
estimate of the relative proportion of the channel species that would
be mostly formed. The V1/2 and slope factor k
determined from these cells were -3.6 ± 0.7 mV and 10.5 ±
0.2 mV (n=8), respectively. These values were slightly more
similar to those of hKv1.2 (Table 1)
, suggesting that this channel type
is relatively more abundant on the plasma membrane.
We observed that the expression of hKv1.1 wild-type channels in
Xenopus oocytes shifted the resting membrane potential of
the cell ~20 mV to more hyperpolarized potentials compared with
uninjected oocytes (Fig. 6B
). A similar effect was observed
when hKv1.2 and hKv1.1wt mRNAs were coinjected or by the expression of
the hKv1.2–1.1wt concatemer. In contrast, the expression of
hKv1.1E325D channels was no longer able to significantly modify the
resting potential of the cell. Moreover, when equal amounts of hKv1.2
and hKv1.1E325D mRNAs were coinjected, the membrane potential was
shifted by only ~7 mV (Fig. 6B
).
To determine the effects of hKv1.1E325D or hKv1.2 expression on the ability of the hKv1.2–1.1wt concatemer to shift the cell resting potential, equal amounts of both mRNAs were coinjected. The coexpression of the hKv1.2–1.1wt concatemer with hKv1.1E325D or hKv1.2 shifted the resting potential of the cell ~7 mV and ~15 mV to more hyperpolarized potentials, respectively.
To further confirm the role of hKv1.1wt and hKv1.1E325D channels to
setting the cell resting potential, different concentrations of TEA
were superfused on oocytes expressing these channels and the membrane
potentials were recorded. By reversibly blocking hKv1.1wt channels in a
concentration-dependent fashion, TEA ions shifted the cell resting
potentials toward values that were normally recorded in uninjected
oocytes (Fig. 6C
). The superfusion of cells expressing
hKv1.1E325D channels with a solution containing a maximal concentration
of TEA (96 mM) had no effect on their resting potential (not shown).
Taken together, these results demonstrate that both homomeric and heteromeric channels comprised of hKv1.2 and hKv1.1 subunits may contribute significantly to setting the cell resting potential, and that episodic ataxia mutations markedly impair this property by reducing channel expression and shifting the voltage range of channel activation to more depolarized potentials.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The single-channel conductance of all these channel types was not
significantly different. However, a more detailed analysis of V408A
channels revealed new pathogenetic mechanisms caused by this mutation,
namely, a marked reduction of the open duration of both homomeric and
heteromeric channels. All these observations suggest that the highly
conserved residue Val 408 resides in a pivotal region of the protein
that controls the correct opening and inactivation of the channel. This
is consistent with the postulated presence of an intracellular gate
located in this region, whose function may be affected even by minor
amino acid changes such as the V408A mutation (33)
. Also,
the increased rates of deactivation and recovery from C-type
inactivation caused by E325D and V408A mutations, in addition to the
single-channel results, suggest that these residues influence the
stability of the open and inactivated state of both homomeric and
heteromeric channels.
Evidence strongly suggests that Kv1.1 and Kv1.2 subunits form
heteromeric channels in hippocampal granule cells and CA3 pyramidal
cell terminals, in myelinated and unmyelinated axons, and in cerebellar
basket cell terminals (21
, 22
23
24
25)
. This study
demonstrated that heteromeric channels composed of hKv1.1 and hKv1.2
subunits possess peculiar biophysical properties that may confer
distinctive electrophysiological features to the neurons expressing
this channel type. E325D mutations nearly nullify the expression of the
channel, suggesting that the availability of hKv1.1 subunits for
heteropolymerization with the hKv1.2 in specific subsets of neurons may
be reduced. This effect by itself may provoke the expression of the
homomeric hKv1.2 channel type and modify neuronal excitability
accordingly. On the other hand, when wild-type and mutant subunits are
coexpressed (13)
, the voltage-dependent parameters of the
currents obtained are similar to the relevant concatamers
(12)
, revealing that the above mentioned phenomenon does
not predominate.
The ability of these channels to set the cell resting potential
suggests that in axons where they are the main class of channels
expressed (34)
, they may play a major role in determining
such potential. EA-1 mutations markedly impair this important property
of both homomeric and heteromeric channels, suggesting that in affected
patients axons and terminals may be hyperexcitable because their
resting potentials are shifted to more depolarized values.
The blockade of hKv1.2 and hKv1.1 channels by -DTX causes a dramatic
increase of GABA release from the basket cell terminals
(27)
. EA-1 mutations, by affecting the level of expression
and several biophysical channel properties, markedly reduce both
homomeric hKv1.1 and heteromeric hKv1.2–1.1 delayed-rectifier channel
function, which may be qualitatively comparable to the -DTX blocking
effects. In addition, heteromeric channels bearing EA-1 mutations enter
the C-type inactivation state with a faster rate constant, suggesting
that during high frequency spiking, the accumulation of this
inactivation process may further reduce the number of active channels.
Therefore, it is conceivable that in EA-1-affected patients,
prolongation of the action potential duration and the increased
excitability of the presynaptic basket cell membranes may markedly
increase the release of -aminobutyric acid from such terminals onto
Purkinje cells (Fig. 7
). Moreover, a basket cell makes contacts with a number of Purkinje
cells. Consequently, a single EA affected basket cell may at the same
time alter the output of several Purkinje cells. As a result, the
output of the entire cerebellum to the rest of the brain may be
markedly altered, leading to the attacks of generalized spastic muscle
contractions characteristic of EA-1 syndrome.
|
The heteropolymerization of mammalian voltage-dependent potassium
channel subunits appears to be confined among members of the same
subfamily (35)
. This study suggests that whenever EA-1
subunits form an heteromeric complex with any of the other subunits
from the Kv1 subfamily, they alter the delayed rectifier function of
the resulting channel. Therefore, such phenomena may markedly broaden
the electrophysiological alterations caused by the EA-1 mutations in
the central and peripheral nervous systems of affected patients.
The reported effects of EA-1 mutations on both homomeric and
heteromeric channels suggest that the action potential duration of
central and peripheral myelinated axons are broadened in EA-1-affected
patients. These conclusions are consistent with the prolonged compound
action potential and altered refractory period observed in myelinated
axons of mice lacking the entire Kv1.1 gene (36)
. Even
though the Kv1.1 null mouse is behaviorally distinct from the EA-1
phenotype, the stress-induced tremors in these animals are reminiscent
of attacks of episodic ataxia (37)
.
In conclusion, these findings show that human Kv1.2 and Kv1.1 potassium channel subunits heteropolymerize to form a unique channel and uncover novel pathogenetic mechanisms of episodic ataxia whereby mutations in a single gene disrupt the functions of other closely related proteins.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
-aminobutyric acid; TEA, tetraethyl-ammonium. 
Received for publication January 4, 1999.
Revision received March 24, 1999.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
subunit Kv1.2 in the basket cell nerve terminals of rat cerebellum. Neuroscience 57,1039-1045[Medline]
This article has been cited by other articles:
|
|
 |
|
  P. Imbrici, M. C. D'Adamo, A. Cusimano, and M. Pessia Episodic ataxia type 1 mutation F184C alters Zn2+-induced modulation of the human K+ channel Kv1.4-Kv1.1/Kvbeta1.1 Am J Physiol Cell Physiol, February 1, 2007; 292(2): C778 - C787. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. J. G. Cardnell, D. E. Dalle Nogare, B. Ganetzky, and M. Stern In Vivo Analysis of a Gain-of-Function Mutation in the Drosophila eag-Encoded K+ Channel Genetics, April 1, 2006; 172(4): 2351 - 2358. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C.-C. Shieh, M. Coghlan, J. P. Sullivan, and M. Gopalakrishnan Potassium Channels: Molecular Defects, Diseases, and Therapeutic Opportunities Pharmacol. Rev., December 1, 2000; 52(4): 557 - 594. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. Liu and S. Linn Proteolysis of the human DNA polymerase {varepsilon} catalytic subunit by caspase-3 and calpain specifically during apoptosis Nucleic Acids Res., November 1, 2000; 28(21): 4180 - 4188. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. N. Manganas, S. Akhtar, D. E. Antonucci, C. R. Campomanes, J. O. Dolly, and J. S. Trimmer Episodic Ataxia Type-1 Mutations in the Kv1.1 Potassium Channel Display Distinct Folding and Intracellular Trafficking Properties J. Biol. Chem., December 21, 2001; 276(52): 49427 - 49434. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
=
