J.
Phyiol.
(1966),
183,
pp.
407-417
407
With
1
plate
and
4
text-figure8
Printed
in
Great
Britain
THE
RELATION
BETWEEN
SARCOMERE
LENGTH
AND
ACTIVE
TENSION
IN
ISOLATED
SEMITENDINOSUS
FIBRES
OF
THE
FROG
BY
K.
A.
P.
EDMAN
From
the
Department
of
Pharmacology,
University
of
Lund,
Lund,
Sweden
(Received
16
July
1965)
SUMM>ARY
1.
The
relation
between
sarcomere
length
and
tetanic
tension
at
various
states
of
shortening
was
investigated
in
single
frog
semitendinosus
fibres
that
were
subjected
to
different
degrees
of
prestretch
(2*45-3.0
,).
2.
The
capacity
to
produce
tension
changed
in
a
characteristic
way
during
shortening,
the
tension
output
at
each
length
being
determined
by
the
actual
sarcomere
spacing
without
reference
to
the
striation
spacing
at
the
onset
of
contraction.
3.
The
capacity
to
shorten
against
a
given
load
was
independent
of
the
initial
striation
spacing,
provided
the
load
was
not
great
enough
to
cause
fatigue
of
the
fibre.
4.
The
findings
strongly
suggest
that
the
functionally
relevant
structure
of
the
contractile
system
of the
intact
muscle
cell
is
always
in
the
same
state
at
a
given
sarcomere
length
independent
of
how
the
previous
length
change
has
been
achieved,
by
passive
extension
at
rest
or
by
active
shortening
from
a
prestretched
position.
This
probably
means
that
con-
traction
involves
a
structural
change
of
the
contractile
system,
which,
at
least
in
so
far
as
it
is
of
relevance
to
function,
is
a
true
reversal
of
the
change
produced
by
passive
extension
of
the
resting
fibre.
These
aspects
of
the
contractile
behaviour
of
the
intact
muscle
fibre
are
in
full
accord
with
the
concepts
of
the
sliding-filament
hypothesis
of
muscular
contraction.
INTRODUCTION
According
to
the
sliding-filament
hypothesis
of
muscular
contraction
(A.
F.
Huxley
&
Niedergerke,
1954;
H.
E.
Huxley
&
Hanson,
1954;
A.
F.
Huxley,
1957;
H.
E.
Huxley,
1957,
1960)
the
capacity
of
the
con-
tractile
system
to
produce
tension
is
proportional
to
the
number
of
inter-
acting
sites,
and
hence
to
the
amount
of
overlap
between
the
two
sets
of
interdigitating
filaments
of
the
myofibril.
This
postulate
is
supported
by
the
fact
that
in
the
intact
skeletal
muscle
fibre
stretch
beyond
the
'slack'
length
causes
a
graded
decrease
in
the
ability
to
develop
isometric
tension,
in
parallel
with
diminution
of
the
overlap
region.
There
is
experimental
evidence
that
the
complete
loss
of
contractile
strength
coincides
with
the
disappearance
of
overlap
between
the
A
and
I
filaments
caused
by
the
stretch
(A.
F.
Huxley
&
Peachey,
1961;
Gordon,
A.
F.
Huxley
&
Julian,
1964;
Podolsky,
1964.
However,
cf.
Carlsen,
Knappeis
&
Buchthal,
1961).
If
contraction
is
based
on
a
sliding
of
the
A
and
I
filaments
relative
to
each
other
the
capacity
to
produce
tension
would
be
expected
to
in-
crease
in
a
characteristic
way
when
the
contractile
system
is
allowed
to
shorten
actively
from
a
prestretched
length.
The
tension
output
at
each
length
would
be
determined
by
the
actual
degree
of
overlap
without
reference
to
the
state
of
overlap
when
the
contraction
is
initiated.
Further-
more,
the
total
amount
of
shortening
against
a
given
load
would
be
the
same
irrespective
of
the
initial
degree
of
overlap.
The
filaments,
once
they
are
able
to
slide
at
all
against
the
load,
would
be
expected
to
slide
to
the
same
equilibrium
state
without
regard
to
their
position
at
the
start.
It
has
been
the
aim
of
the
present
experiments
to
investigate
these
points
by
studying
the
length-tension
relation
in
single
frog
semiten-
dinosus
fibres
that
are
subjected
to
various
degrees
of
prestretch.
As
will
be
demonstrated,
the
functional
behaviour
of
the
living
skeletal
muscle
cell
is
in
full
accord
with
the
idea
of
a
sliding-filament
mechanism
of
muscular
contraction.
METHODS
Single
twitch
fibres
of
the
dorsal
portion
of
the
semitendinosus
muscle
of
Rana
temporaria
were
used.
The
isolated
fibre
was
mounted
between
a
tension
transducer
(RCA
5734)
and
a
vertical
isotonic
lever
in
a
Perspex
trough.
For
the
mounting
a
small
loop
of
stainless-steel
wire
(0.
10
mm
thickness)
was
fixed
to
each
tendon
close
to
the
fibre
end.
There
was
no
support
to
the
fibre
at
any
place
in
between
the
end-points.
The
resting
length
and
also
the
degree
of
shortening
of
the
fibre
could
be
varied
by
means
of
two
adjustable
stops
fitted
in
front
of
and
behind
the
lever.
The
fibre
was
stimulated
to
tetanic
contraction
at
3
min
intervals
by
applying
a
square
pulse
train
of
1-2-1-5
sec
duration
(50
c/s,
pulse
width
0
9
msec)
through
a
pair
of
platinum-wire
electrodes
mounted
in
the
floor
of
the
Perspex
trough.
The
cathode
was
placed
midway
between
the
tendons
and
the
anode
near
to
the
tendon
attached
to
the
tension
transducer.
The
fibres
were
usually
mounted
4-5
hr
before
the
experiment
was
started
and
were
tested
occasionally.
Measurements
of
the
sarcomere
length
were
made
by
direct
microscopy
at
each
resting
length
by
means
of
a
Zeiss
D
water
immersion
objective
(40
x
magnification,
focal
length
4-6
mm,
N.A.
0
75,
working
distance
1-6
mm)
and
an
ocular
micrometer.
A
total
magnifica-
tion
of
800
x
was
used
for
these
measurements.
In
certain
cases
photographic
recordings
of
the
resting
band
pattern
were
made
as
well
with
a
Zeiss
35
mm
camera
fitted
on
the
micro-
scope
through
an
extension
tube
provided
with
a
side
viewer.
The
sarcomere
spacing
was
determined
by
measurements
of
sequences
of
sarcomeres
of
about
fifteen
striations.
Mean
values,
as
plotted
in
the
length-tension
diagrams,
were
derived
from
measurements
of
408
K.
A.
P.
EDMAN
SARCOMERE
LENGTH
AND
CONTRACTILITY
four
to
five
different
sequences.
Kodak
Tri-X
film
was
used
and
the
exposure
time
was
sec.
The
photographs
were
taken
at
a
magnification
of
320
x
.
The
sarcomere
spacing,
as
will
be
demonstrated,
was
very
nearly
uniform
at
rest
in
the
entire
fibre
except
for
a
small
section
(about
5
%
of
the
total
fibre
length)
close
to
the
tendons,
where
the
striations
were
shorter
(cf.
A.
F.
Huxley
&
Peachey,
1961
and
Carlsen
et
al.
1961).
The
tension
output
of
the
fibre
was
correlated
with
the
sarcomere
spacing
measured
in
the
'middle
segment'
here
defined
as
that
part
of
the
fibre
(approximately
90%
of
the
total
length)
in
which
the
striation
spacing
at
rest
varied
by
less
than
3
%
of
the
mean.
The
sarcomere
spacing
within
this
segment
during
activity
was
determined
by
calculation
based
on
recording
of
the
striation
spacing
at
rest
and
the
over-all
shortening
between
the
ends
of
the
fibre
or
between
two
markers
(nylon
filaments)
on
the
fibre
during
full
tetanic
activity.
In
certain
experiments
photographic
recordings
of
the
band
pattern
at
various
sites
of
the
fibre
during
the
plateau
of
isometric
tetani
were
also
carried
out.
The
experiments
were
performed
at
a
temperature
of
+
5-6'
C
in
a
Ringer
solution
of
the
following
composition
(m-mole/l.):
NaCl
115-5,
KCI
2-0,
CaCl2
1-8,
MgCl2
1-0,
NaH2PO4-Na2HPO4
buffer
2-0,
pH
7-0.
RESULTS
Mechanical
recording8
The
relation
between
sarcomere
length
and
tetanic
tension
in
frog
semitendinosus
fibres
is
illustrated
in
Text-fig.
1.
Single
muscle
fibres
were
stretched
to
various
degrees
and
the
tetanic
output
at
each
length
was
recorded.
The
ability
to
produce
tension
at
given
shortened
lengths
was
determined
by
allowing
the
fibre
to
shorten
against
a
small
load
(10
mg)
from
slack
length
(about
2-45
,u).
The
sarcomere
lengths
plotted
in
Fig.
1
refer
to
the
'middle
segment'
of
the
fibre
(definition,
see
Methods).
Shortening
from
3-0,
t
caused
a
steady
increase
in
the
capacity
to
develop
tension,
maximum
being
attained
at
2-2
,t
sarcomere
length.
The
tetanic
tension
was
virtually
constant
within
the
range
2-2-2-0
,u.
Shortening
below
2-0
It
was
associated
with
reduction
of
the
active
tension.
There
was
only
a
slight
(about
10
%)
decrease
in
tension
by
reduction
of
the
sarcomere
length
from
2-0
to
1-70
It.
Shortening
below
1-70
,u,
how-
ever,
caused
a
steeper
decline
in
tension.
At
1-40
/t,
for
instance,
the
tension
output
was
only
40
%
of
the
maximrum.
The
phases
of
the
length-
tension
curve
below
2-0
It
may
not
be
strictly
linear.
It
is
clear,
however,
that
there
is
a
distinct
change
of
the
slope
of
the
curve
within
a
relatively
narrow
interval
at
about
1-70
,u.
The
results
are
in
good
agreement
with
the
observations
made
by
Gordon
et
al.
(1964)
in
a
similar
preparation.
According
to
these
authors
the
active
tension
is
constant
between
2-2
and
2-0
It,
and
it
was
found
by
them
that
there
is
an
angle
of
the
length-tension
curve
at
1-65
It.
The
capacity
to
produce
tension
at
a
given
length
was
independent
of
the
sarcomere
length
at
the
start
of
the
contraction.
This
was
found
to
hold
true
over
a
wide
range
of
sarcomere
lengths,
as
is
evident
from
Text-
fig.
2.
For
instance,
the
tetanic
tension
produced
at
2-45
It
was
the
same
409
whether
the
contraction
was
initiated
at
that
length
or
the
fibre
had
shortened
to
2*45
,u
from
a
stretched
position
(2*45-3*0
It).
Similarly,
the
tension
output
produced
at
sarcomere
lengths
only
attainable
by
means
of
active
shortening,
i.e.
below
2-45
It,
was
virtually
unaffected
by
the
amount
of
preceding
unloaded
shortening.
The
results
thus
indicate
that
the
ability
to
produce
tension
is
determined
by
the
actual
sarcomere
spacing
during
activity
of
the
fibre,
without
reference
to
the
sarcomere
100
0~~~~~~~
_
50
C~~~~~~~~~C
0-
so_
81
11
I
I I , I I
1.5
20
25
3.0
Sarcomere
length
(is)
Text-fig.
1.
Relation
between
tetanic
tension
and
sarcomere
length
of
isolated
frog
semitendinosus
fibres.
The
different
symbols
denote
different
preparations.
Values
below
'slack'
length,
about
2
45
gu,
refer
to
contractions
induced
at
2-45-
2*48
/s,
the
load
during
the
initial
isotonic
phase
being
10
mg.
Values
at
and
beyond
2-45
Its
represent
isometric
contractions
without
initial
shortening.
length
existing
at
the
moment
when
the
contraction
is
initiated.
This
finding
is
of
relevance
for
the
interpretation
of
the
molecular
events
under-
lying
the
mechanical
activity
of
the
living
muscle
cell.
As
will
be
further
discussed,
the
results
support
the
view
that
contraction
produces
a
change
of
the
functionally
relevant
structure
of
the
contractile
system,
which
is
a
true
reversal
of
the
change
caused
by
passive
extension
of
the
resting
fibre.
If
the
capacity
to
develop
tension
is
determined
by
the
actual
sarcomere
spacing
at
any
moment
during
activity,
the
fibre
might
be
expected
'to
shorten
to
the
same
degree
against
a
given
load
irrespective
of
the
sar-
comere
length
at
the
start.
Once
the
load
can
be
lifted
at
all,
each
new
410
K.
A.
P.
EDMAN
SARCOMERE
LENGTH
AND
CONTRACTILITY
increment
in
the
capacity
to
produce
tension
would
make
it
easier
for
the
contractile
system
to
lift
the
load
further.
This
of
course
will
only
apply
if
the
fibre
does
not
fatigue.
In
order
to
investigate
this
aspect
of
the
length-tension
relation
experiments
were
performed
where
the
fibres
were
allowed
to
shorten
against
different
loads
from
different
resting
100
0
50
Eo-
1
5
20
25
3*0
Sarcomere
length
(,u)
Text-fig.
2.
Relation
between
tetanic
tension
and
sarcomere
length
of
an
isolated
frog
semitendinosus
fibre.
Lack
of
influence
of
sarcomere
spacing
at
the
start
of
contraction.
Small
circles:
measurements
as
in
Text-fig.
1,
i.e.
values
below
2-45
,u
refer
to
contractions
initiated
at
2
45
,u.
Large
circles:
all
contractions
initiated
at
2-85
,.
Fibre
loaded
by
10
mg
during
initial
isotonic
phase.
lengths.
The
shortening
was
arrested
at
chosen
lengths
by
means
of
a
stop,
and
the
active
tension
that
developed
was
recorded.
Text-figure
3
illu-
strates
typical
oscilloscope
recordings,
and
Text-fig.
4
shows
the
results
of
one
experiment
plotted
in
a
length-tension
diagram.
Text-figure
4
also
includes
the
length-tension
curve
of
Text-fig.
1
representing
the
tension
produced
at
various
lengths
after
shortening
from
'slack
length',
2-45
,u,
against
a
small
load.
As
can
be
seen
virtually
the
same
active
tension
was
produced
at
a
given
shortening
irrespective
of
whether
the
fibre
started
to
contract
from
slack
length
or
from
a
stretched
state
or
whether
the
shortening
occurred
against
a
very
small
load
or
a
heavy
load.
With
still
bigger
loads
than
those
used
in
the
experiment
presented
in
Text-fig.
4
411
412
K.A.P.
EDMAN
the
tension
measurements
were
complicated
by
fatigue
of
the
fibre.
Under
such
conditions
less
tension
was
produced
at
a
given
length,
and
the
tension
deficit,
as
expected,
was
larger
the
more
extensive
the
preceding
shorten-
ing.
The
results
demonstrated
in
Text-fig.
4,
obtained
under
conditions
where
there
was
no
substantial
interference
by
fatigue,
provide
clear
evidence,
however,
that
the
capacity
of
the
contractile
system
to
shorten
against
a
load
is
not
critically
dependent
on
the
sarcomere
spacing
at
the
onset
of
contraction
within
the
wide
range
of
sarcomere
lengths
investigated.
A
B
1
200
mg
100
Jo~~~
,
I
I
I
*
I
*
I
0
1
2
sec
0
1
2
sec
C
D
1400
mg
1200
0
1
2
sec
0
1
2
sec
Text-fig.
3.
Isometric
tetanic
contractions
produced
by
an
isolated
frog
semi-
tendinosus
fibre
at
two
different
sarcomere
lengths.
Lack
of
influence
by
load
and
sarcomere
spacing
at
the
start
of
contraction.
Isotonic
phase
from
zero
time
to
angle
indicated
by
arrow.
From
there
on
isometric
tension
development.
Sarcomere
Sarcomere
spacing
Approx.
spacing
at
start
load
during
during
(zero
time)
isotonic
isometric
(,u)
phase
(mg)
phase
(g)
A
2-91
10
1-55
B
2-91
90
1-55
C
2-72
130
2-00
D
2-91
130
2-00
Note
different
tension
scales
between
A,
B
and
C,
D.
Prolongation
of
isotonic
phase
in
B
(as
compared
with
A)
is
due
to
slowness
of
shortening
by
increased
load
in
D
(as
compared
with
C),
due
to
longer
distance of
shortening.
SARCOMERE
LENGTH
AND
CONTRACTILITY
100
_
'I
0
l,~
l
III
Iul
I
|
1.5
20
25
30
Sarcomere
length
(/u)
Text-fig.
4.
Tetanic
output
of
an
isolated
frog
semitendinosus
fibre
at
two
dif-
ferent
lengths,
2-0
and
1-55,u.
Contractions
initiated
at
2-72
and
2-92,u
with
various
loads
during
shortening
phase.
Length-tension
curve
of
Text-fig.
1
superimposed.
The
paths
indicated
between
the
initial
and
final
states
are
approximate.
Microphotographic
recordings
Photographic
recordings
of
the
sarcomere
pattern
at
rest
and
during
activity
were
performed
to
investigate
the
uniformity
of
contraction
throughout
the
fibre.
Plate
1
shows
the
band
pattern
in
a
prestretched
fibre
at
rest
(A-B),
during
full
tension
development
in
the
stretched
position
(C-D)
and
on
the
height
of
an
isometric
tetanus
at
a
shortened
length
(E-F).
In
the
middle
segment
of
the
fibre,
constituting
approxi-
mately
90
%
of
the
total
fibre
length,
the
resting
sarcomere
length
was
2-76
It.
In
accordance
with
previous
observations
(A.
F.
Huxley
&
Peachey,
1961;
Carlsen
et
al.
1961)
there
was
only
a
small
(<
3
%)
variability
in
the
striation
spacing
within
this
section
of
the
fibre.
Shorter
sarcomeres
existed
at
the
ends
of
the
fibre.
Essentially
the
same
uniformity
of
the
band
pattern
was
found
during
activity
of
the
fibre.
This
held
true
whether
the
fibre
was
contracted
isometrically
at
a
stretched
length
or
was
allowed
to
produce
tetanic
tension
at
various
degrees
of
shortening.
In
the
preparation
illustrated
in
413
P1.
1
the
sarcomeres
in
the
middle
segment
changed
from
an
average
length
of
2*76
,
at
rest
to
2-69
,t
during
isometric
tetanic
contraction
in
the
stretched
state;
the
sarcomere
spacing
varied
between
2-66
and
2-71
#
in
the
different
parts
of
the
middle
segment
during
the
contraction.
The
same
degree
of
uniformity
of
the
band
pattern
existed
during
tetanic
activity
at
shorter
fibre
lengths
as
is
illustrated
by
P1.
1
(E-F).
There
was
no
substantial
yielding
of
the
sarcomeres
in
the
end
regions
whether
the
fibre
contracted
isometrically
in
a
stretched
position
or
was
allowed
to
shorten.
The
almost
complete
reproducibility
of
the
sarcomere
pattern
in
repeated
tetanic
contractions
is
also
illustrated
by
P1.
1.
The
two
pictures
of
E
and
F
covering
the
same
area
of
the
fibre
refer
to
two
different
con-
tractions
separated
by
a
time
interval
of
3
min.
As
can
be
seen,
the
two
recordings
exhibit
nearly
identical
patterns.
It
should
be
pointed
out
that
these
findings
only
apply
to
fibres
which
are
in
a
perfect
functional
con-
dition.
Any
damage
to
the
fibre's
function
caused,
for
example,
by
too
frequent
activation
was
manifested
by
the
occurrence
of
local
yielding
of
the
sarcomeres
during
contraction.
The
fact
that
there
was
a
regular
band
pattern-without
yielding
of
the
sarcomeres-at
steady-state
activity
(plateau
of
tetanic
contraction)
indicates
that
all
parts
of
the
middle
segment
of
the
fibre
were
able
to
generate
the
tetanic
tension
recorded
for
the
fibre
as
a
whole.
It
is
there-
fore
concluded
that
the
results
described
in
the
previous
sections,
corre-
lating
active
tension
with
length,
are
representative
for
the
mechanical
behaviour
of
the
muscle
cell
at
the
sarcomere
level.
It
has
been
demonstrated
by
Deleze
(1961),
studying
the
whole
semi-
tendinosus
muscle
of
the
frog,
that
the
total
amount
of
shortening
against
a
given
load
is
reduced
if
contraction
is
initiated
beyond
a
certain
length
(about
130
%
of
the
length
for
maximum
tension).
It
is
reasonable
to
assume
on
the
basis
of
the
present
results
that
the
observed
effects
of
stretching
the
whole
muscle
may
be
due
to
non-uniform
contraction
of
the
individual
fibres
of
the
preparation
and,
at
least
with
the
heavier
loads
used,
to
fatigue.
At
any
rate,
the
complicated
situation
of
the
whole
muscle,
with
differences
in
functional
state
between
the
various
fibres,
makes
it
impracticable
to
interpret
quantitatively
the
length-tension
relation
of
this
preparation
at
sarcomere
level.
DISCUSSION
There
is
convincing
experimental
evidence
that
the
contractile
system
of
the
skeletal
muscle
cell
is
organiized
in
two
discrete
sets
of
inter-digi-
tating
filaments,
the
A
and
I
filaments,
which
are
freely
movable
relative
to
each
other
when
the
muscle
fibre
is
relaxed
and
its
length
changed
passively
(e.g.
Hanson
&
H.
E.
Huxley,
1955;
H.
E.
Huxley,
1957;
A.
F.
414
K.
A.
P.
EDMAN
SARCOMERE
LENGTH
AND
CONTRACTILITY
Huixley
&
Peachey,
1961;
Carlsen
et
al.
1961).
Several
hypotheses
have
been
advanced
to
explain,
in
mechanical
terms,
the
mode
of
interaction
between
the
filaments
during
activity
of
the
muscle
fibre
leading
to
shortening
and
development
of
tension
(A.
F.
Huxley,
1957;
H.
E.
Huxley,
1957,
1960;
Weber,
1958;
Podolsky,
1959,
1960;
Szent-Gyorgyi
&
Johnson,
1964;
Alexander
&
Johnson,
1965).
The
present
results
may
provide
a
relevant
clue
to
the
elucidation
of
this
process.
The
experiments
have
shown
that
as
the
contractile
system
is
shortening,
its
capacity
to
develop
tension
is
changed
in
a
given
manner
according
to
the
actual
sarcomere
length
at
any
instant,
without
relation
to
the
sarcomere
length
at
the
start.
This
fact
strongly
suggests
that
the
func-
tionally
relevant
structure
of
the
contractile
system
of
the
intact
muscle
cell
is
always
in
the
same
state
at
a
given
sarcomere
length
irrespective
of
the
amount
of
preceding
shortening.
This
is
an
important
finding,
since
it
implies
that
contraction
probably
involves
a
structural
change
of
the
contractile
system,
which,
at
least
in
so
far
as
it
is
of
relevance
to
function,
is
a
true
reversal
of
the
change
produced
by
passive
extension
of
the
resting
fibre.
If,
therefore,
contraction
is
due
to
an
interaction
between
the
A
and
I
filaments,
which
all
the
evidence
would
suggest
are
dissociated
and
freely
movable
relative
to
each
other
in
the
resting
fibre,
it
would
seem
necessary
to
assume
a
pure
sliding
movement
of
the
two
sets
of
filaments
during
activity
without
change
in
their
length.
The
re-
sults
thus
support
the
sliding-filament
hypothesis
of
muscular
contraction
(A.
F.
Huxley,
1957;
H.
E.
Huxley,
1957,
1960),
which
predicts
that
the
tension
development
at
a
given
length
is
a
direct
function
of
the
actual
degree
of
overlap
between
the
A
and
I
filaments
without
reference
to
the
state
of
overlap
when
the
contraction
is
initiated.
The
experimental
findings
on
the
other
hand
provide
evidence
against
the
kinematics
proposed
in
the
contraction
models
advanced
by
Podolsky
(1959,
1960)
and
Szent-Gy6rgyi
&
Johnson
(1964).
Both
models
postulate
that
the
thin
filaments
become
fixed
relative
to
the
thick
filaments
when
activation
occurs
followed
by
shortening
of
the
complex
so
formed,
either
of
the
I
filament
proper
as
proposed
by
Podolsky
or
of
the
myosin
com-
ponent
of
the
A
filament
as
in
the
hypothesis
advanced
by
Szent-Gy6rgyi
&
Johnson.
It
is
clear
that
neither
of
these
models
can
account
for
the
length-tension
relation
exhibited
by
the
living
muscle
cell
as
demon-
strated
by
the
present
experiments.
In
both
cases
the
capacity
to
develop
tension
at
a
given
length
would
be
rendered
critically
dependent
on
the
sarcomere
spacing
existing
at
the
moment
of
activation.
As
pointed
out
by
Gordon
et
al.
(1964)
the
shape
of
the
length-tension
curve
is
easily
explained
in
terms
of
a
sliding-filament
mechanism.
The
active
tension
is
virtually
absent
when
the
sarcomere
spacing
is
3-5-3*6
,u
415
i.e.
when
the
A
and
I
filaments
(total
length
1-60
and
2-05
,u,
respectively)
are
at
approximately
end-to-end
position
relative
to
each
other
(A.
F.
Huxley
&
Peachey,
1961;
Gordon
et
al.
1964;
Podolsky,
1964).
Maximum
tension
is
produced
at
2-2
jt
(Text-figs.
I
and
2
of
the
present
article
and
also
Gordon
et
al.
1964),
at
which
sarcomere
length
the
ends
of
the
I
filaments
are
at
the
'inert'
region
(0
15-0
20
,u
long,
lacking
cross-bridge
structures
(H.
E.
Huxley,
1963))
at
the
centre
of
the
A
filament.
This
degree
of
overlap
would
consequently
provide
the
maximum
number
of
interacting
sites
for
propulsion
of
the
thin
filaments
towards
the
centre
of
the
thick
filaments.
The
decrease
in
tension
output
associated
with
shortening
below
2-0
,u
may
be
attributed
to
an
increase
in
the
passive
resistance
to
the
sliding
movement
caused
by
tighter
packing
of
the
filaments
(double
overlap
of
the
I
filaments
(H.
E.
Huxley,
1960))
and,
at
more
extreme
degrees
of
shortening,
by
compression
of
the
A
filaments
when
coming
up
to
the
Z
disks.
There
may
also
be
a
decrease
of
the
driving
force
for
the
sliding,
owing
to
interference
from
the
set
of
I
filaments
in-
truding
from
the
other
half
of
the
A
segment.
Furthermore,
if
the
I
filaments
are
able
to
interact
with
the
active
sites
on
the
opposite
half
of
the
A
filament,
this
would
even
produce
an
active
braking
of
the
con-
traction
by
generating
a
force
directed
against
the
sliding
movement.
While,
thus,
the
contraction
pattern
of
the
intact
muscle
cell
appears
to
be
in
full
accordance
with
the
sliding-filament
hypothesis,
the
functional
behaviour
of
glycerinated
muscle
(Edman,
1964)
does
not
fit
easily
into
the
concepts
of
this
hypothesis.
The
capacity
of
the
glycerinated
muscle
fibre
to
produce
tension
in
response
to
ATP
is
progressively
diminished
by
shortening.
The
active
tension
at
a
given
length
is
therefore,
by
con-
trast
to
the
situation
in
the
living
cell,
critically
dependent
on
the
sar-
comere
spacing
when
activation
is
initiated.
The
contraction
pattern
holds
true
irrespective
of
whether
the
fibre
is
stretched
before
glycerination
or
after
the
glycerol-treatment
in
a
'relaxing
medium',
and
would
seem
to
reflect
a
genuine
property
of
the
contractile
system
of
the
glycerinated
cell.
The
nature
of
the
aberrant
functional
behaviour
of
the
glycerol-
extracted
fibre
is
unclear.
Evidence
from
electron-microscope
studies
provide
strong
support
for
the
view
that
contraction
of
glycerinated
skeletal
muscle
is
associated
with
a
sliding
movement
of
the
two
sets
of
filaments
without
change
in
length
of
the
individual
elements
(H.
E.
Huxley
&
Hanson,
1954;
Hanson
&
H.
E.
Huxley,
1955).
The
discrepancy
in
contractile
behaviour
between
the
intact
cell
and
the
glycerinated
pre-
paration
is,
therefore,
most
likely
to
be
attributed
to
some
change
of
the
physical
properties
of
the
interacting
sites
of
the
myofilaments
caused
by
the
glycerol-treatment
rather
than
to
a
basic
change
in
the
mode
of
inter-
action
between
the
two
sets
of
filaments.
K.
A.
P.
EDMAN
416
The
Journal
of
Phy8iology,
Vol.
183,
No.
2
9._
.._s
B
C_...
E
K.
A.
P.
EDMAN
Plate
1
(Facing
p.
417)
r.1.
r
SARCOMERE
LENGTH
AND
CONTRACTILITY
417
This
study
was
aided
by
grants
from
Muscular
Dystrophy
Associations
of
America,
Inc.
For
skilful
technical
assistance
I
am
indebted
to
Miss
Kathe
Koch.
REFERENCES
ALEXANDER,
R.
S.
&
JOHNSON,
P.
D.,
Jr.
(1965).
Muscle
stretch
and
theories
of
contraction.
Am.
J.
Physiol.
208,
412-416.
CARLSEN,
F.,
KNAPPEIS,
G.
G.
&
BUCHTHAL,
F.
(1961).
Ultrastructure
of
the
resting
and
contracted
striated
muscle
fibre
at
different
degrees
of
stretch.
J.
biophys.
biochem.
Cytol.
11,
95-117.
D1iL3ZE,
J.
B.
(1961).
The
mechanical
properties
of
the
semitendinosus
muscle
at
lengths
greater
than
its
length
in
the
body.
J.
Physiol.
158,
154-164.
EDMAN,
K.
A.
P.
(1964).
Contractility
of
glycerol-extracted
muscle
fibres
after
stretch.
In
Biochemistry
of
Muscle
Contraction,
ed.
J.
GERGELY.
Boston:
Little,
Brown
and
Co.,
Inc.
GORDON,
A.
M
.,
HUXLEY,
A.
F.
&
JULIAN,
F.
J.
(1964).
The
length-tension
diagram
of
single
vertebrate
striated
muscle
fibres.
J.
Physiol.
171,
28P.
HANSON,
J.
&
HUXLEY,
H.
E.
(1955).
The
structural
basis
of
contraction
in
striated
muscle.
Symp.
Soc.
exp.
Biol.
9,
228-264.
HUXLEY,
A.
F.
(1957).
Muscle
structure
and
theories
of
contraction.
Progr.
Biophys.
biophys.
Chem.
7,
255-318.
HUXLEY,
A.
F.
&
NIEDERGERKE,
R.
(1954).
Structural
changes
in
muscle
during
contrac-
tion:
Interference
microscopy
of
living
muscle
fibres.
Nature,
Lond.,
173,
971-973.
HUXLEY,
A.
F.
&
PEACHEY,
L.
D.
(1961).
The
maximum
length
for
contraction
in
verte-
brate
striated
muscle.
J.
Physiol.
156,
150-165.
HUXLEY,
H.
E.
(1957).
The
double
array
of
filaments
in
cross-striated
muscle.
J.
biophys.
biochem.
Cytol.
3,
631-648.
HuXLEY,
H.
E.
(1960).
Muscle
cells.
In
The
Cell,
vol.
4,
eds.
BRACHET
&
MIRSKY.
New
York:
Academic
Press.
HuXLEY,
H.
E.
(1963).
Electron
microscope
studies
on
the
structure
of
natural
and
syn-
thetic
protein
filaments
from
striated
muscle.
J.
mol.
Biol.
7,
281-308.
HUXLEY,
H.
E.
&
HANSON,
J.
(1954).
Changes
in
the
cross-striations
of
muscle
during
contraction
and
stretch
and
their
structural
interpretation.
Nature,
Lond.,
173,
973-977.
PODOLSKY,
R.
J.
(1959).
The
chemical
thermodynamics
and
molecular
mechanism
of
mus-
cular
contraction.
Ann.
N.Y.
Acad.
Sci.
72,
522-537.
PODOLSKY,
R.
J.
(1960).
Thermodynamics
of
muscle.
In
Structure
and
Function
of
Muscle,
vol.
2,
ed.
G.
H.
BOURNE.
New
York:
Academic
Press.
PODOLSKY,
R.
J.
(1964).
The
maximum
sarcomere
length
for
contraction
of
isolated
myo-
fibrils.
J.
Physiol.
170,
110-123.
SZENT-GYORGYI,
A.
G.
&
JOHNSON,
W.
H.
(1964).
An
alternative
theory
for
contraction
of
striated
muscles.
In
Biochemistry
of
Muscle
Contraction,
ed.
J.
GERGELY.
Boston:
Little,
Brown
and
Co.,
Inc.
WEBER,
H.
H.
(1958).
The
Motility
of
Muscles
and
Cell8.
Cambridge,
Massachusetts:
Harvard
University
Press.
EXPLANATION
OF
PLATE
Sarcomere
pattern
from
'middle
segment'
of
isolated
fibre
at
rest
and
during
constant
tetanic
tension
at
two
different
lengths.
Still
pictures,
-1k
sec
exposure
time.
A-B.
Fibre
at
rest.
Striation
spacing
2-76
,u.
C-D.
Fibre
during
plateau
of
isometric
tetanic
contractions,
0
7
sec
after
the
onset
of
contraction.
Total
length,
including
tendons,
between
attachments
of
the
fibre
the
same
as
in
A-B.
Striation
spacing
2-69
,u.
E-F.
Fibre
during
plateau
of
tetanic
contractions
after
initial
shortening
(load
10
mg)
to
length
for
maximal
tension
development.
Striation
spacing
2-01
g.
Both
pictures
cover
the
same
area
of
the
fibre
and
refer
to
two
different
contractions
induced
in
sequence
with
a
time
interval
of
3
min.
Note
the
uniformity
of
the
band
pattern
and
the
almost
complete
repro-
ducibility
between
the
two
contractions.
Micrometer
scale,
10
,u
divisions.
Physiol.
183
27
