Neuron Module

Quantitative Physiology II
Organ Systems
BMEN E4002
Professor Morrison

Dendrite
Soma
Hillock
Axon
Presynaptic Terminal
Bouton

Copyright 2002 Elsevier Science (USA)

All rights reserved

Todays Overview

10.1

Information Flow

General purpose of the nervous system

General structures (at neuron level)
Introduction to myelin
Difference between PNS and CNS
Structure of a peripheral nerve
Simple electrical model of a cell

Excitatory Post Synaptic Potential: EPSP

Action Potential: AP

What is the purpose of the nervous system?

2

Copyright 2002 Elsevier Science (USA)

All rights reserved

Why a Neuron Module?

10.9

Myelin

What are the applications?

Neural engineering

CNS: Oligodendrocytes
PNS: Schwann cells

Restoration of lost function

Cochlear implants (~100,000 in use)
Brain computer interface
Replace cognitive / higher order processing

Treat diseases
Deep brain stimulators
Parkinsons, epilepsy, even severe depression
http://www.neuro.jhmi.edu/DBS/cases.htm
3

Copyright 2002 Elsevier Science (USA)

All rights reserved

10.12

Motivation
Central nervous system (CNS)

Why do we want to model neuron

behavior?

Dense and complex connections

Designed for computation

To gain a deeper understand of data

To better understand normal function
To better understand disease states
To identify underlying causes of pathology
To develop treatments / cures

Peripheral nervous system (PNS)

Transport of information to/from periphery
Mechanically active environments
Specialized structures
Mechanical protection

Epilepsy
9

13

Peripheral Nerve

Epilepsy
Coordinated and repetitive discharges
Of large populations of cells
Bursting

Capture normal behavior with a model

Alter the model to produce pathology
Models help explain experimental data
Models can guide new experiments
Copyright 2002 Elsevier Science (USA)
10
All rights reserved

14

10.12

Motivation

Phospholipids

Response

Stimulus current
Vm

Hyperpolarizing
stimulus

7.2

3-5nm Thick

Vm
Depolarizing
stimulus

11

Copyright 2002 Elsevier Science (USA)

All rights reserved

15

Copyright 2002 Elsevier Science (USA)

All rights reserved

2.1

Capacitor Model

Membrane Resistance
Passive element 2:
Membrane resistance: Rm
Leak current is Ohmic

Ohms
Rm

Obeys V=IR
Alternatively I= g V
g conductance = 1/R (Siemens)

16

Copyright 2002 Elsevier Science (USA)

All rights reserved

19

6.9

Parallel Plate Capacitor

Passive Membrane Model

Passive Element 1

Vm is measured: Inside - Outside

Membrane capacitance: Cm
Q = charge (Coulombs)
Vm is membrane voltage
C2
C: Farads

Q
Vm

Cm

Vm

Intracellular
Cm

Rm

N m

Iin

Extracellular
17

20

Membrane Capacitance
Rearrange

Cm

Q
Vm

Circuit Equations

C mVm

Differentiate wrt time (assuming Cm const.)

Implies Cm dictates speed

dQ
dt
dQ
dt
I

Cm

Kirchoffs Current Law (KCL)

Kirchoffs Voltage Law (KVL)
KCL: Sum of current into a node = 0

dVm
dt

ij

I
Cm

dVm
dt

18

0
21

Kirchoffs Current Law

Passive Model
Vm

Use KCL to sum I in top node

Requires a sign convention

Current into a node is positive

i4

i1

Define:
units?

I in

Cm

dVm
dt

Rm C m

i1

i2

i3

i4

Rm

Extracellular

N m s
Coulomb 2
Coulomb 2
N m

dVm
dt

First order ODE

Cm

Vm
Rm
R

i3
i2

Intracellular

V m I in Rm

22

25

Kirchoffs Voltage Law

Sum of the voltages in a loop = 0

vj

Separate variables and integrate

v1

1
dVm
V m I in Rm

v2

ln V m I in Rm

Voltage drop
across nodes
v4

dt

v3

V m I in Rm

De

Algebraic sum
Assign +/- to each end of voltage drop

23

26

Kirchoffs Voltage Law

v1

v2

Integration constant from IC

At t=0, Vm(0) = 0

v1

v4

v2

v3

V m I in Rm

v3

V m(t )

v4

De

I in Rm 1 e

0
24

27

Initial Condition

I(t)
Iin

I = 0; t < 0
I = Iin; t 0

Find D from IC

At t = 0; Vm(0) = 0

This model is valid from to

What is Vm(to)?

IinRm

V m(t )

Vm(t)

I in Rm 1 e

to

V m(to )

I in Rm 1 e

Vm(to)

63%

Speed
Vm(0)

to

t=

28

31

IC for Discharge

I(t)
I = 0; t < 0
I = Iin; 0 t < to
I = 0; to t

Iin

to

At t = 0; Vm(0) = 0

V m(to )

De

to

V m(to )

Vm(t)

to

I in Rm 1 e

D V m(to )e

Vm(to)
0

Vm (t

t < to

to

to ) V m(to )e

to

29

32

Discharge
Reexamine the system now
No current source: Iin = 0
dVm
dt
dVm
Vm
1
Vm

Vm

Substitute for Vm(to)

Vm

to

Intracellular
Cm

V m(t o )

Rm

I in Rm 1 e

dt
1

dVm
t

ln V m

dt

Collect terms

Extracellular

to

Vm (t

to )

I in Rm 1 e

( t to )

V m De

30

33

Potential Energy Storage

I(t)
I = 0; t < 0
I = Iin; 0 t < to
I = 0; to t

Iin

Ion gradients store energy

t

V m(0 t

Vm(t)

Vm (t

to )

to )

I in Rm 1 e

I in Rm 1 e

to

Concentration potential
Electrical potential

( t to )

Ux

Vm(to)

to

34

39

Typical values

Nernst Potential
Potential of the concentration gradient

Rm and Cm related to membrane area

Unit capacitance

Unit resistance

X
RT
ln i
zF
Xo

Vm

Ex

Rm

Rm

RT
X
ln o
zF
Xi

RT
X
ln i
zF
Xo

Ex

E is measured Ein Eout same as Vm

[ ] = concentration

cm 2

2000
Rm

Nernst potential

1 F
cm 2

R = ideal gas constant = 8.314 J/K/Mole

T = absolute temperature in K (C=273)
z = algebraic charge of the ion
F = Faradays constant 96,500 Coulomb/mole

Rm
35

40

Nernst Potential Calculation

Ion

ATPase

subunit

[Intracellular]

Na+

NaK PUMP

50mM

400mM

400mM

10mM

Cl-

40mM

540mM

ATP

Na : Ein

subunit

[Extracellular]

K+

E out

ADP

RT
400
ln
1 F
50

54mV

+ Pi

Na+

K : Ein

E out

Cl : Ein

E out

Extracellular
space
3.8; 5-8

Cytosol

37

RT
10
ln
1 F
400

RT
540
ln
1 F
40

96mV

68mV
41

Copyright 2002 Elsevier Science (USA)

All rights reserved

11

Driving Potential
Intracellular

Vm

Vm Ex
Vm ~ -70mV
Na+: Vm ENa = -124mV
K+: Vm EK = 26mV
Cl-: Vm ECl = -2mV

Cm

Sign determines direction

gNa

gK

gCl

ENa

EK

ECl

With respect to the voltage drop (Vm ENa)

I =V/R

Extracellular
44

47

Resting Membrane Potential

Vm

Q: How can ions cross the membrane?

Intracellular

At Vm, I = 0
Apply KCL

Membrane core - hydrophobic

Ions polar

Cm

Current out of node is +

A: Embedded channel proteins

Form hydrophilic pores

I Na
V is

Extracellular space

IK

I Cl

dV
constant: dtm

Cm

dVm
dt

gNa

gK

gCl

ENa

EK

ECl

Extracellular

And ICl ~ Ileak (IL)

+

K channel

6.9

Na channel

Ca

2+

channel

Cl channel

Cytosol

45

48

Copyright 2002 Elsevier Science (USA)

All rights reserved

Ohmic Current
Ion movement generates a current

Substitute definitions for each channel

Assume it is an Ohmic current

I=V*g
INa: (Vm ENa ) * gNa
IK: (Vm EK ) * gK
ICl: (Vm ECl ) * gCl

g Na (Vm

E Na )

g K (Vm

EK )

g L (Vm

EL )

Rearrange
Vm

g Na E Na g K E K g L E L
g Na g K g L

Add these to a new cell model

46

49

13

AP Threshold
Vm

Ex

g Na E Na g K E K g L E L
g Na g K g L

X
RT
ln o
zF
Xi

Ion

Na+

K+

Cl-

Vm

0.05

RT 120
ln
F
12
Vm

0.5

[In]

[Out] Conductance

g Na
g

0.05

4mM

gK
g

0.5

4mM 120mM

g Cl
g

0.45

12mM 120mM

120mM

RT
4
ln
F
120

0.45

RT 120
ln
F
4

81mV

50

54

Copyright 2002 Elsevier Science (USA)

All rights reserved

7.1&2

Stimulus Intensity & Duration

Action Potential

Intracellular

Vm

Cm

Combine Ri

Iin

Ra

Em
Extracellular

52

7.1&2

55

Copyright 2002 Elsevier Science (USA)

All rights reserved

Vm

AP Regenerative Conduction

Intracellular
Cm

Iin

Ra

Apply KCL to the top node

0

I in

Cm

Define

53

7.2

I in Ra

dVm
dt

Vm

Em

Em
Ra

Extracellular

Ra C m

dVm
dt

Vm

Em

56

Copyright 2002 Elsevier Science (USA)

All rights reserved

15

Intracellular

Vm

Cm

Iin

Ra

Em

Rearrange and separate

V m(t )

Extracellular

dVm
Vm E m I in Ra
dt
dVm
1
dt
Vm E m I in Ra

Integrate

ln Vm

Em

I in Ra

I in Ra 1 e

Em

For a given Vth

How are stimulus intensity and time related?
Rearrange
Vth E m
I in
t

Ra 1 e

Vm

Em

I in Ra

D e

57

Cm

I in Ra

Iin

Ra

I in Ra

Em

D e

Em
Ra

tj
D e0

Em

Ra 1 e

Ij

I in Ra

Vth

I in

Extracellular

Vth

Em

I in Ra 1 e

Em

t 0
Vm E m
Vm

V m(t )

Apply initial conditions

Vm(0) = Em
Find D
Em

Rheobase

Intracellular

Vm

Vm

60

The minimum level of current capable of generating an AP if applied

for an infinite time

58

61

Adding to the model

Cannot reproduce the Action Potential

Substitute

Hodgkin Huxley neuron model

Invented the Voltage Clamp

Vm Em
Collect terms
V m(t )

I in Ra

Holds the cell voltage constant

Measures the necessary current

I in Ra e
t

I in Ra 1 e

Vm
(mV)

Em

20
80
TOTAL IONIC CURRENT
1

Im
(mA/cm2) 0
1

59

7.5

Outward
current
Inward
current

1963 Nobel Prize in Physiology / Medicine62

Copyright 2002 Elsevier Science (USA)
All rights reserved

17

Hodgkin Huxley Assumptions

Ions are separated by the membrane

Current flow is Ohmic
Ions flow through channels

H&H assumed
Changes in Vm were due to
Time dependent changes in conductances

Their conductances are variable

Function of Vm and time

Em

g Na E Na g K EK g L EL
g Na g K g L

Increasing gi will drive Vm to Ei

Add these to the model

63

66

Hodgkin Huxley Model

Intracellular

Vm

Calculated Nernst potentials earlier

E Na
54mV

Cm
gK

gNa

gL

EK

ENa

EL

EK

96mV

EL

68mV

Leak current is mainly Cl current

INa

IK

IL

IT

IC

Extracellular
64

Vm

Em

Ij

X
RT
ln o
zF
Xi
g Na E Na g K EK g L EL
g Na g K g L

g j (Vm

Intracellular
Cm

Apply KCL
Ej

67

gNa

gK

gL

ENa

EK

EL

INa

IK

IL

Conductance Changes in AP

IC
1)
2)
3)

Vm

IT

Extracellular

Increase gNa to drive Vm to ENa (+54mV)

Increase gK to drive Vm to EK (-96mV)
Reset gK and gNa to drive voltage back to Vm

0
Em

Ej)

g Na E Na g K EK g L EL
g Na g K g L

70

Current out of bottom node as positive

0

IT

Cm

dV m
dt

gNa

I Na

IK

gK

Reset

IL
65

68

Copyright 2002 Elsevier Science (USA)

All rights reserved

7.2

19

Experimental Requirements

Experimental Methods

Needed to measure gNa and gK

Functions of both time and Vm
Measure INa and IK
Calculate g from
Ij

g Na
gK

Measured the membrane current to different

voltage steps

g j (V m
Vm

With this set-up

Ej)

I Na
E Na

t (msec)

IK
Vm

EK
73

69

Experimental Methods

Separate Current

Experiments to understand physiology

Needed to separate IK from INa

Drive innovation and technology development

Constant in time
Constant in space along the axon

Today wed use pharmacology

Tetrodotoxin TTX to block Na channels
Tetraethyl ammonium TEA to block K channels

Devised a method to hold Vm constant

None available - ?
Separate mathematically

Voltage

Space Clamp

Experiments in normal sea water

Experiments in sea water with reduced [Na]
74

71

Experimental Methods

Experimental Paradigm

Needed to measure the current necessary

to hold Vm at a desired level
Feedback Amplifier
Second wire to apply a current
Feedback circuitry to measure current
for a desired Vm

Voltage

Current

Classic experiments
Nobel prize winning work!

Step and hold voltage at V

Measure current through the membrane
Calculate gNa and gK

76

72

21

V (mV
90

K Channel

70
55
40
25
10
5

gK
1/

Examine K channel first

Concept of a gate
Probability it is open = n
Assumed 4 gates in the K channel
All must be open for conduction

t (msec)

Reproduces the sigmoidal shape of the curve

gNa
1/

t (msec)

77

82

Model Equations

K Channel

Hodgkin and Huxley devised a model

Fit the model to the data
Examine K channel first

Assume each gate operates independently

Probability that a channel is open
n*n*n*n = n4

The proportion of open channels in a

population

Formulated the concept of a gate

To explain the data
That is fit the experiment to a model

n4

Probability it is open = n
Assumed 4 gates in the K channel

gK n4

gK
78

83

K Channel Conduction Gates

Time Dependence

Probability of one gate being opened = n

Gates can transition from open to closed

Rate constant associated with each transition

O
+ +

++

+ +

+ +

+ +

++

++

++

C
C

Write a differential equation for n

dn
dt
79

(n )

(1 n )
84

23

dn
dt

n
n

Homogeneous solution

dn
dt
Define

n (t )

1
n

c2

n (t )

85

Apply an initial condition

(1 e

Fit a first order exponential to find

n (t )

Calculated

i.e. the system comes to a steady state

from their definitions

1
n

and

89

for other voltages

V (mV
90

and

(1 e

Captures time dependency of the channel

Estimates an n and n valid for one Vm

Need to find
n

n
86

dn
dt

88

Normalized gK and took the 4th root

Determined n from the data

Solve for a particular solution

t
dn
dt

The model was then fit to the data

n(0) = 0
c1 = -c2

c1

c1e

c1e

c1

n (t )

n
t

70
55
40
25
10
5

gK
1/

n
t (msec)

87

25

90

Na Channel Kinetics

Determined n and n for many Vm

Fit a smooth function through the points

Rate Constant (1/msec)

Empirical functions only

Capture voltage dependence
gate
gate

0.01

e
n

0.125 e

mV

dm
dt
dh
dt

80

= V-Vrest

91

(1 m)

10
10

10

(1 h )

(m)

(h )

94

Na Channel
Kinetics are more complex

Similar procedure as for the K channel

Fit function to the data

Activation followed by inactivation

m,

gNa
1/

m,

h,

h
m

are all functions of Vm

e

Proposed two kinds of gates

Activation: m
Inactivation: h

g Na

+ +

+ +

+ +

Closed

Closed

30
10

95

Passive properties of the membrane

Nernst potentials of each ion
Time / voltage dependent conductances

Open

Complete H&H Model

I


+ +

4 e 18
0.07 e 20

Probability of an open channel = m3h

92

Na Channel Conduction Gates

+ +

= V-Vrest

25
10

g Na m 3h

+ +

25

0.1

t (msec)

Closed

Ej

Cm

dVm
dt

g LL Vmm E
ELL

g Na Vm

E Na

g K Vm

EK

RT
X
ln o
zF
Xi

93

96

27

Complete H&H Model

Numerical Simulation Results

Time / voltage dependent conductances

I

Cm

dVm
dt

g L Vm

g Na m3h Vm

EL

g K n 4 Vm

E Na

Depolarizing
Stimulus

EK

20
10
0
0

10 voltage15
membrane
(mV)

20

25

60

dn
dt
dm
dt
dh
dt

(n)

(1 n )

(1 m)

(1 h)

( m)

10

0.01

0.125 e

10
10

80

25

0.1

25
10

4 e 18

0.07 e 20

HHSim
Tutorial

20
0

30
10

10

15

20

25

-20

( h)
= V-Vrest

-40
-60

197

-80

Example

103

AP Characteristics

Discover a new organism extremophile

90C

40

Ion

[In] (mM)

[Out] (mM)

g (Siemens)

Gd+++

30

520

12

P--

450

12

AP is an all or nothing phenomenon

No half amplitude AP
Magnitude and duration are fixed

What are the Nernst potentials for Gd3+ & P2-?

Communicates by a P channel
What is the theoretical maximum potential when
this channel opens?

Absolute Refractory Period

A second stimulus cannot elicit an AP
If close in time to the first stimulus

Both can be explained by the model

What is the resting membrane potential?

98

104

Example

AP Initiation

Discovered a cuboid cell: 100 m per edge

Em = -40mV
cm2
rm = 2500
cm = 5 F/cm2
Vth = -5mv

AP begins when Vm > Vth

Two opposing currents
INa depolarizing
IK hyperpolarizing

Vth corresponds to INa > IK

What is the minimum current injection to

get the cell to fire in 5ms?

Initiates a positive feedback

Stopped when Na channels shut down (h)

99

105

29

ENa

50
0
-50

-100

Gate
Activation

10

15

20

25

EK

HW 2

m
h
n

0.75

Will explore the H&H model in more detail

Using a numerical simulation
Links are in the homework
Detailed instructions in the homework

0.5

0.25
t

n (t )

(1 e

0
0

10

15

20

10625

109

Absolute Refractory Period

Two stimuli

Two stimuli

Two APs

H&H model reproduces the AP

Only one AP?

Captures experimental data

Leads to testable predictions

Limitations of what weve modeled so far?

Treats the neuron as a single compartment

Vm
x
107

Absolute Refractory Period

No spatial information
Cant represent a realistic neuron

115

Actual Neuron Morphology

h must reset
1

h
n

0.75

0.5

0.25

0
0

10

15

20

10825

Duke-Southampton archive of neuronal morphology

31

116

Passive Conduction

Signal Attenuation

Model the passive spread of signal down a

neuronal process
Processes are not perfectly insulated
conductors

Insulation is not perfect

Current loss through the membrane

Signal gets attenuated as it travels

There is a finite resistance between the
intracellular and extracellular spaces

Injection

117

120

Copyright 2002 Elsevier Science (USA)

All rights reserved

7.22

Why model neuron structure?

Passive Conduction

Important for modeling networks

Constant velocity of propagation

Networks are the basis for computation

Capture network behavior

V
1

Understand higher order functions

Ocular dominance columns (p. 370)
We can design a replacement circuit
Replace defective neuron networks
Repair damaged neuron networks
Berger et al., Restoring lost cognitive function, IEEE Eng
Med Biol Mag. 24: 30-44, 2005
Cohen and Nicolelis, Reduction of single-neuron firing
uncertainty by cortical ensembles during motor skill
learning, J Neurosci 24: 3574-3582, 2004

V 70

x3
x2
x1

80

Slope = conduction velocity

t1 t2 t3

118

Passive Conduction

121

Cable Theory

Signal is attenuated
Signal becomes spread out

First developed by William Thomson,1855

University of Glasgow
Later Lord Kelvin of absolute 0 K fame
Describe conduction in the

V
2

Stimulus current

If we understand network behavior

Trans-Atlantic Telegraph Cable, was knighted for it

Applied to neurons by
Stimulus current
V 70

Hodgkin and Rushton (1946)

Rall (1957-1969)

B&B pp. 207-211

80

119

7.2

122

Copyright 2002 Elsevier Science (USA)

All rights reserved

33

Cable Theory Assumptions

Components
ro

Need a new model for the neuron

For the neuronal structure (process)

Uniform cylindrical core

Length >> diameter
Uniform membrane properties
Uniform core properties

123

Definitions
ro: external resistance per axial length

io

cm

ri: internal resistance per axial length

cm

cm: capacitance of membrane per unit

length

F
cm

Extracellular

rm

cm

ri

Intracellular
126

Currents
ro

io+dio
im

rm: resistance across the membrane times

unit length
cm
124

rm

ii

ri

Vo

Voltages
ro

Cable Model
Simple model of the plasma membrane

cm

ii+dii

127

Vo+dVo

Linked in series with two resistances

Internal and external resistance
ro

ro

ro

ro

ro

ro

Membrane
rm

cm
ri

Cytoplasm

7.22

ri
Extracellular fluid

ri

ri

ri

rm

cm

ri

V(t,x)

Vi

125

Copyright 2002 Elsevier Science (USA)

All rights reserved

35

ri

Vi+dVi

128

Membrane Current
im

Vi

dx
im

dVi Vo
rm

lim

dVo

cm

d Vi

Axial Currents

dVi Vo dVo
dt

V=IR
Extracellular current

Vo+dVo

Vi Vo
rm

cm

im

d Vi Vo
dt

rm

cm

Vo

dVo Vo

dVo
dx

Vo

io

ro

Vo+dVo

io ro dx

io ro
ri

Vi+dVi

129

132

External Current
Use KCL on top and
bottom nodes
Top node

0 io
0
im

io

ro

io+dio

rm

dio im dx
ii

ri

V=IR
Intracellular current

Vi (Vi

im

dio im dx io
dio
dx

Axial Currents

cm

dVi
dx

ro

dVi ) ii ri dx
ii ri
Vi ii

ii+dii
130

ri

Vi+dVi
133

Internal Current
Use KCL on top and
bottom nodes
Bottom node

0 ii im dx (ii
0
im

Subtract the currents

io

io+dio
im

dii )

rm

im dx dii
dii
dx

ro

ii

ri

cm
ii+dii

ii io
dVi
dx

dVo
dx

d Vi Vo
dx

131

ii ri io ro

ii ri io ro
134

37

Membrane Current
d Vi Vo
dx

ii ri io ro

ri

Differentiate with respect to x

d d Vi Vo
dx
dx

d (ii ri )
dx

d d Vi Vo
dx
dx

di
ri i
dx

d (io ro )
dx

im

di
ro o
dx

d d Vi Vo
ro dx
dx

Vi Vo
rm

im

Vo

d Vi Vo
cm
dt

rm

dio
dx

im

d d Vi Vo
dx
dx

dii
dx

dio
dx

di
ri i
dx

d d Vi Vo
dx
dx

dii
dx

1
ri

di
ro o
dx

dii
ri
dx

ro

Vi Vo
x

Vi Vo
rm

cm

Rearrange:
Vi Vo

ro

1 d d Vi Vo
ro dx
dx
im

rm
ri

ro

Define:

Vi Vo
x2

cm rm

Vi Vo
t

Vi Vo

ri

139

Define the space constant (distance units)

dii
dx

rm
ro

dii
dx

d d Vi Vo
ro dx
dx

138

Vi Vo
t

136

ri

cm
Vi

135

im

im

ri

Define the time constant (time units)

cm rm

Cable Equation

im

( x, t )
137

x2

t
140

39

Steady State Solution

Semi infinite cable

must be negative for a finite solution

From the initial condition

0<x<

Inject a constant current for a long time

Injected at x = 0

What is the distribution of

Final solution is

o
x

I(x=0)

As x
t

=0

146

149

I(x=0)

At x=0, membrane has reached

I(x=0)

o x 0

Steady state so

When x =

= 37% of
t

A e
2

We have experimental measures of:

x2

Try a solution of:

Rm Cross membrane resistance

cm 2

A e

and

e
150

Calculate
( x, t

147

Ri Cytoplasm resistivity
2

A e

cm

Cm membrane capacitance

F
cm 2

1
148

152

41

 Need to calculate parameters in cable

equation

cm: capacitance of membrane per unit

length
F

Take into account the specific geometry

cm

a = radius

Cm is capacitance per unit area

And direction of integration

F
cm 2

Multiply by perimeter
Capacitance per unit length

cm

Cm 2 a

153

156

Typical Values
a

rm: resistance across the membrane times

unit length
cm
Rm is resistance times surface area

cm 2

Divide it by the perimeter

To get resistance times unit length

rm

Rm
2 a

25 m

Rm

2000

Ri

60

Cm

F
1 2
cm

cm 2
cm

rm

127,000

ri

3,000,000

cm

0.0157

cm

cm
F
cm

Neglect ro because it is small

154

159

Space Constant
ri: internal resistance per axial length

cm

Ri is resistance times axial length

cm

Divide by cross sectional area

To get resistance per length

ri

Ri
a2

rm

~ 2000 m

ri

ro

Rm / 2 a
Ri / a 2

a Rm
2 Ri

rm
ri

Rm and Ri are constants for the cell

Therefore:
155

a
160

43

Space Constant
a

Length constant scales with

Post-synaptic side: dendrite

The larger the cable radius

The better the conduction

Input structure of the neuron

Flow of information
Synapse

Axon
Soma
Synapse

Dendrites

a
161

Copyright 2002 Elsevier Science (USA)

All rights reserved

171

10.9

Axon

Time Constant
Synapse

rm c m
Rm
2 a Cm
2 a

Presynaptic
Terminal

Neurotransmitter
Vesicles

Rm Cm
Neurotransmitter

Time constant is independent of radius

Depends on electrical properties of the lipid

Extracellular
space

162

Postsynaptic
Terminal

NT

Dendrite
Neurotransmitter
binds
Opens an ion channel
Depolarizes the
membrane

V
3

Stimulus current
V 70

No action potential
No voltage gated channels
Cytosol
Excitatory post-synaptic
potential

80

170

7.2

Ligandgated
channel

Post-synaptic side

Signal transfer in the dendrites is passive

172

12.2; 8.2

Extracellular
space

Passive Propagation

Dendrite

Copyright 2002 Elsevier Science (USA)

All rights reserved

Copyright 2002 Elsevier Science (USA)

All rights reserved

EPSP
Copyright 2002 Elsevier Science (USA)

45

NT

Ion

173

12.15; 13.13

Change 20
in Vm
(mV)
0

Passive propagation of the depolarization

Cable equation
+ + + + +

- - - - -

+ + + + + + + + + + +

- - - - - - - - - - -

Channel opening generates local depolarization

Depolarizes the next segment

Change
20
in Vm
(mV)
0

Soma

- -

Passive transmission

+ +

Dendrites

By passive propagation

Until Vth is reached

174

Copyright 2002 Elsevier Science (USA)

All rights reserved

11.3; 12.3

Then action potential is regenerated

177

7.21

Copyright 2002 Elsevier Science (USA)

V
1

Stimulus current
V 70

80

Neurotransmitters bind to receptors

Channels open

Propagation of action potential in space

Dependent on passive axon properties

Depolarize or hyperpolarize
Depends on the neurotransmitter and the channel

Local change is voltage

Regeneration of action potential

Requires voltage activated channels

Transmitted to the cell body

Attenuated along the way
Cable equation

Integrates the input from many synapses

Determines if an action potential is fired

175

178

Cable Propagation for the Axon

Action potential is initiated in the hillock

Myelination
Myelinated or unmyelinated axons

If Vm > Vth
How does it get propagated?
By depolarizing the next piece of membrane
How does the next piece of membrane get to Vth?

Inactive

Active

Inactive
176

Copyright 2002 Elsevier Science (USA)

All rights reserved

47

179

10.12

Myelin Diseases

Arranged in Series

Myelin plays a critical physiological role in

neuronal activity
Myelin damage causes severe disruption
of nerve function

What are the implications for the model?

Examine the capacitance first
Assume each layer has capacitance C
Capacitors in series:
1
1
1
1
...
C T C1 C2 C3
For n identical C

Multiple sclerosis, amyotrophic lateral

sclerosis (ALS, Lou Gehrigs disease),
progressive multifocal leukoencephalopathy

We can explain the pathology through our

models and make predictions of function

CT

180

C
n

183

Myelin Sheath
Composed of many layers

Each myelin layer adds to the axon size

Plasma membrane (phospholipids)

Wrapped tightly around the axon

Radius a = n * layer thickness

Produced by Schwann Cells in the PNS

Produced by Oligodendrocytes in the CNS

n
a

Hundreds of layers

Acts as an insulator
Decreases leak currents to the outside
ro

ro

ro

ro

ro

ri

ri

ri

ri

C
a

CT

ro

Membrane
rm

cm
ri

Cytoplasm

181r

184

Extracellular fluid

Myelin

C
R

Use our passive model again

Implications for the resistance

One layer of myelin

Arranged in series
C
R

How are the layers arranged?

In series or parallel?

RT

R1 R2

RT

n R

RT
C

C
R

R3 ...

How does the myelin affect the cable equation

182

185

49

Effects of Myelin
Space constant:

Improves conduction by

a Rm
2 Ri

Increasing the space constant

No effect on time constant

Ri does not change due to myelin sheath

Rm is now different R
n R
T

Assuming material properties of myelin

Equal those of regular plasma membranes

RT

Not entirely true

Myelin resistance is greater
Myelin capacitance is lower

Rm
186

189

Myelinated Nerve Structure

a Rm
2 Ri

Myelin is not continuous

Nodes of Ranvier
Separate myelin sheaths

Myelinated

a a Rm
Ri
2

No ion channels beneath myelin sheath

Un-myelinated

RT

Rm
a

187

Rm Cm

Time constant:

Rm

194

AP jumps from node to node

Saltatory conduction
Possible because of the increased insulation

C
a

CT

Concentrated at the Nodes of Ranvier

Myelin
Decreased loss of signal

Cm
188

Copyright 2002 Elsevier Science (USA)

All rights reserved

51

195

7.21

 Implications for diseases and injuries

Myelinated nerve
Channels concentrated at nodes
What happens if it becomes unmyelinated?

Injury
Multiple Sclerosis
Amyotrophic Lateral Sclerosis

> Vth?

What happens to information transfer?

196

200

Normal condition

Assume that

Vth
Vth

Injured or diseased condition

1
o
e
37%

How long can the gap be

For conduction to be maintained?

Will the nerve continue to conduct?

Copyright 2002 Elsevier Science (USA)
All rights reserved

197

201

7.21

Re-state the problem

Use our steady state solution:

x

Will the next Node of Ranvier

o

Reach Vth to re-initiate the action potential?

1
e

How can we calculate the voltage

As a function of distance?

o
x

Assume steady state applies

Cable equation

> Vth

x
199

202

53

Inhibitory Signals
Using some typical values
a

2. 5 m

Rm

2000

Ri

60

cm 2

Neurons transmit information by AP

Vm

Electrical coupling

a Rm
2 Ri

Excitatory signal propagation only

AP in up stream cell

70

Always produces an AP
in down stream cell

cm

640 m

Called gap junction

Unmyelinated gap < 640 m

Exist in certain cells

Cardiac myocytes

For conduction to continue

203

208
Copyright 2002 Elsevier Science (USA) 8.1

Example
Injury has destroyed a section of neurons
myelin will it continue to conduct?
Distance between nodes of Ranvier = 505 m

Model the neuron as infinitely long

Assume the steady state solution applies
Em = -63mV; Vth = -39mV; VAP = +27mV
Rm = 2000 cm2; Ri = 250 cm
Radius = 4 m

Chemical synapse
No physical connection between the cells
Message is transmitted by chemicals
Released from the pre-synaptic terminal
Bind to the post-synaptic cell
Specialized structures two halves
Pre-synaptic machinery
Post-synaptic machinery

Advantage
Signal can be excitatory or inhibitory
Depends on the chemical
Neurotransmitters

205

209

Why learn about it?

Neurotransmission
How one neuron communicates with the next

Current neural prostheses interface with

neurons with electrodes
Limits information sensing to only excitation

Via the synapse

Electrically isolated from one another
One cell does not touch the other

Can only measure action potentials

Potential for new prosthetic design

Transmission is mediated by the release of

Sense the neurotransmitters directly

Neurotransmitter

Seems overly complicated

What is the advantage?
207

55

Both excitatory and inhibitory information

Greater information, more accurate response
Probably necessary for cognitive prostheses
Bi-directional flow of information

210

Pre-synaptic Terminal

Fast Axonal Transport

Axon

Action potential
Vesicle fusion
Transmitter Release
Neurotransmitter
Vesicles
Free diffusion

Proteins, vesicles, and mitochondria

Transported down the axon (~400mm/day)
An ATP dependent process
By motor proteins

Presynaptic
Terminal

Kinesin (anterograde direction)

Dynein (retrograde direction)

30nm gap

Receptor
Binding

Neurotransmitter
Kinesin

Dynein

Extracellular
space

211

Postsynaptic
Terminal

12.2; 8.2

Copyright 2002 Elsevier Science (USA)

Dendrite

214

10.2
Copyright 2002 Elsevier Science (USA)

Post-Synaptic Side

Fast Axonal Transport

Post-synaptic side - dendrite

Neurotransmitter binds
Activates a protein

Extracellular
space

Ion channel
Signaling cascade

NT

Ligandgated
channel

Motor proteins run on a track

Made of a component of the cytoskeleton
Microtubules

NT

ER

Golgi

Vesicles

Mitochondria

Synaptic
terminal

NT

NT

Receptor

Nucleus
G
protein

AC

Axon

Soma

Cytosol

Retrograde

Ion
Copyright 2002 Elsevier Science (USA)

215

10.2

Copyright 2002 Elsevier Science (USA)

Pre-Synaptic Terminal

Pre-synaptic Terminal

Where do the transmitter vesicles come

from?
Axon

Fast axonal transport

Axon

Of vesicles

Vesicles are generated in the soma

To the synapse
Neurotransmitter
Loaded with
Vesicles
neurotransmitters
Docked to the
Neurotransmitter

By the packaging machinery there

Golgi and the endoplasmic reticulum
40-200nm in diameter

Anterograde

212

12.15; 13.13

++

Ca

Axon could be a meter long!

213

Presynaptic
Terminal

Plasma membrane
Active
Zone

12.2, 4; 13.3, 8.2

Copyright 2002 Elsevier Science (USA)

57

Extracellular
space

Postsynaptic
Terminal

216

Dendrite

Vesicle Fusion
For transmitter release

Very complex process

Vesicle must fuse with the plasma membrane

INITIAL STATE
How does it depend on the AP?

Tightly regulated
Prevent unwanted release of neurotransmitter
Ensure rapid release when wanted

Synaptic
vesicle

Multiple components are required

FUSION AND EXOCYTOSIS

Not fully understood

Definitely Ca++ dependent

217
Pre-synaptic membrane

Ca2+

8.16

220

Copyright 2002 Elsevier Science (USA)

AP Coupling to Vesicle Release

AP depolarizes the pre-synaptic terminal
Opens voltage gated Ca++ channels

Docking
Vesicle is docked

TIGHTENING OF TERNARY
SNARE COMPLEX

Close to the membrane

Awaiting Ca++ influx

Similar to the K+ and Na+ channel

Surrounded by channels

But only allow Ca++ to pass

Ca++ is a special ion

FORMATION OF TERNARY
COMPLEX OF SNARES

Huge concentration gradient

0.1 M inside
2mM outside

SNAP-25
Syntaxin
Synaptobrevin

20,000X difference
218

223

Vesicle Fusion
Opening of Ca++ channels

Axon

Action potential depolarizes the membrane

Ca++ influx
FUSION AND EXOCYTOSIS
Sensed by

Allows Ca++ to enter

Increases the local [Ca++]
Located near the vesicles

Synaptotagmin (yellow)
Initiates fusion

Vesicle fusion
Ca++ dependent

Mechanism still unknown

Ca++

TIGHTENING OF TERNARY
SNARE COMPLEX

Ca2+

219

224

59

Toxins

Excitatory Synapses (CNS)

Vesicle fusion will not occur spontaneously

Main neurotransmitter is Glutamate

Also aspartate

Requires this full set of protein machinery

Multiple glutamate receptors

Ionotropic glutamate receptors

Steps can be disrupted by toxins

Prevent the release of NTs
Tetanus toxin
Botulinum toxin

Glutamate receptors linked to ion channels

Open when bound to glutamate

Proteases which degrade:

Synaptobrevin or SNAP-25 or syntaxin

Depolarize the cell

NH 2

HOOC

CH

CH2

CH2

COOH
Glutamate

225

229

Post-synaptic Side
Release of neurotransmitter
Diffuses across the synapse (30nm)
Bind to receptors

Named for specific agonists

AMPA

-amino-3-hydroxy-5-methyl-4-isoxazole proprionic acid

Kainate
NMDA

Proteins embedded in the plasma membrane

Bind specifically to certain compounds
Thereby altering their confirmation (structure)

HN

HO2C

N-methyl-D-aspartate

HO2C

OH

N
H3C

CO2 H

Kainate

AMPA

O
CO2H

NH 2
HOOC

8.3

CH2

CH 2

CH

COOH

CO2H

NMDA

Glutamate

226

CH3

230

Copyright 2002 Elsevier Science (USA)

The combination generates specificity

AMPA / Kainate channels

Pre-synaptic side could release different NTs

Post-synaptic side could express different
receptors

Bind glutamate
Open up a cation channel

Glutamate

Na

AMPA
K

Channel for positive ions

Both K+ and Na+ flow through the channel
Net effect is to depolarize the membrane
Nernst potential between ENa and EK

One reason for a chemical synapse

Examine some of the combinations

In most fast excitatory synapses of the CNS

Design parameters for a neural prosthesis

227

231

12.18; 13.16

61

Copyright 2002 Elsevier Science (USA)

Inhibitory Synapse
NMDA channels

GABAA channels

Bind glutamate
Opens up a cation channel

Site of action for sedatives

Valium
Barbiturates

Channel for positive ions

Both K+ and Na+ flow through the channel
Plus Ca++ too

Influx of Ca++ activates many signaling cascades

232

235

Metabotropic Receptors
Complex behavior

Another class of post-synaptic receptors

At normal Vm, channel is blocked by Mg++

Does not conduct until Vm is depolarized
And glutamate is bound
Glutamate

Bind the same ligands

Glutamate and GABA
Ca2+ Na+

Glutamate

Mg

NMDA

NMDA
Mg

Do not open an ion channel

Activate G-proteins
Initiate signaling cascades

2+

2+

+
2+

Ca

Vm = -80mV

Vm = -40mV
233

12.18; 13.16

Copyright 2002 Elsevier Science (USA)

236

8.3

Copyright 2002 Elsevier Science (USA)

Inhibitory Synapse
GABAA channels

Many other CNS neurotransmitters

-aminobutyric acid
Ionotropic

Amino acids
glutamate

Monoamines

Open ion channels

Specifically Cl- channels
Drives Vm to ECl

Acetylcholine
Serotonin

Hyperpolarizes the membrane by ~10mV

Catecholamines

NH 2
HOOC

CH2

CH 2

C
H

Norepinephrine

Peptides

H
GABA

234

Endorphin

63

237

Summary
Purpose of a neuron
Transmission of information
Takes the form of an action potential

Input structure
Dendrite

Output structure
Axon

240

Because of the chemical synapse

Dendritic input could be either
Excitatory (ionotropic)
Inhibitory (ionotropic)
Modulatory (metabotropic)

Transmitted by the cable equation

Inputs are integrated in the soma
Makes a decision whether to fire an AP
Vm > Vth
241

The timing and magnitude of the AP

Determined by the gate kinetics
Well described by the Hodgkin Huxley model

AP propagates down the axon

Releases neurotransmitter at the synapse
Vesicle fusion is Ca++ dependent

Neurotransmitter activates receptors

On the post-synaptic side
242

65