Passive immunization against cachectin/tumor necrosis factor protects mice from lethal effect of endotoxin. Science, 1985, 229(4716):869-871. Classical article.

Article (PDF Available)inThe Journal of Immunology 181(1):7-9 · August 2008with2 Reads
Source: PubMed
Abstract
A highly specific polyclonal rabbit antiserum directed against murine cachectinltumor necrosis factor (TNF) was prepared. When BALBIc mice were passively immunized with the antiserum or with purified immune globulin, they were protected against the lethal effect of the endotoxin lipopolysaccharide produced by Escherichia coli. The prophylactic effect was dose-dependent and was most effective when the antiserum was administered prior to the injection of the endotoxin. Antiserum to cachectin/TNF did not mitigate the febrile response of endotoxin-treated animals, and very high doses of endotoxin could overcome the protective effect. The median lethal dose of endotoxin in mice pretreated with 50 microliters of the specific antiserum was approximately 2.5 times greater the median lethal dose for controls given nonimmune serum. The data suggest that cachectinlTNF is one of the principal mediators of the lethal effect of endotoxin.
the
loose
connective
tissue
between
the
macrophage,
which
reacts
to
infectious
and
immunologic
stimuli,
and
the
adipo-
cyte,
whose
energy
stores
must
be
mobi-
lized
at
times
of
physiologic
stress.
That
these
events
are
reversible
with
the
re-
moval
of
cachectin
suggests
approaches
that
may
have
potential
therapeutic
im-
plications
for
humans.
Further
charac-
terization
of
cachectin
and
its
receptor
will
help
to
clarify
the
nature
of
signal
transduction
to
the
adipocyte
nucleus.
References
and
Notes
1.
M.
Kawakami,
P.
H.
Pekala,
M.
D.
Lane,
A.
Cerami,
Proc.
Natl.
Acad.
Sci.
U.S.A.
79,
912
(1982).
2.
P.
H.
Pekala,
M.
Kawakami,
C.
W.
Angus,
M.
D.
Lane,
A.
Cerami,
ibid.
80,
2743
(1983).
3.
A.
B.
Chapman,
D.
M.
Knight,
B.
S.
Dieck-
mann,
G.
M.
Ringold,
J.
Biol.
Chem.
259,
15548
(1984).
4.
A.
B.
Chapman,
D.
M.
Knight,
G.
M.
Ringold,
J.
Cell
Biol.,
in
press;
D.
M.
Knight,
A.
B.
Chapman,
G.
M.
Ringold,
in
preparation.
5.
R.
G.
Ham,
in
Cell
Culture
Methodsfor
Molecu-
lar
and
Cellular
Biology,
D.
W.
Barnes,
D.
A.
Sirbasku,
G.
H.
Sato,
Eds.
(Liss,
New
York,
1984),
vol.
1,
pp.
3-21.
6.
J.
Hirsch
and
J.
L.
Knittle,
Fed.
Proc.
Fed.
Am.
Soc.
E:xp.
Biol.
29,
1516
(1970).
7.
B.
Beutler,
J.
Mahoney,
M.
Le
Trang,
P.
Pe
B.
BEUTLER*
I.
W.
MILSARK
A.
C.
CERAMI
Laboratory
of
Medical
Biochemistry,
Rockefeller
University,
1230
York
Avenue,
New
York
10021
*To
whom
correspondence
should
be
addressed.
Mammals
infected
with
gram-negative
bacteria
often
develop
a
state
of
shock,
which
is
characterized
by
hypotension,
disseminated
intravascular
coagulation,
and
renal,
hepatic,
and
cerebral
injury.
Many
of
these
deleterious
consequences
of
infection
can
be
reproduced
in
animal
models
by
injecting
endotoxin,
a
lipo-
polysaccharide
(LPS)
component
of
the
cell
walls
of
certain
bacteria.
While
the
mechanism
of
action
of
LPS
remains
30
AUGUST
1985
kala,
A.
Cerami,
J.
Exp.
Med.
161,
984
(1983).
8.
B.
Beutler
et
al.,
Nature
(London),
in
press.
9.
F.
M.
Torti,
J.
Larrick,
G.
M.
Ringold,
unpub-
lished
observations.
More
recent
data
indicate
that
recombinant
TNF
affects
adipocyte
gene
expression
in
a
manner
indistinguishable
from
that
reported
here
for
cachectin.
10.
C.
Reznikoff,
D.
Brankow,
C.
Heidelberger,
Cancer
Res.
33,
3231
(1973).
11.
S.
M.
Taylor
and
P.
A.
Jones,
Cell
17,
771
(1979).
12.
J.
M.
Chirgwin,
A.
E.
Przybyla,
R.
J.
MacDon-
ald,
W.
J.
Rutter,
Biochemistry
18,
5294
(1979).
13.
J.
L.
Vannice,
J.
M.
Taylor,
G.
M.
Ringold,
Proc.
Natl.
Acad.
Sci.
U.S.A.
81,
4241
(1984).
14.
D.
J.
Israel
and
J.
P.
Whitlock,
Jr.,
J.
Biol.
Chem.
259,
5400
(1984).
15.
M.
M.
Smith,
A.
E.
Reene,
R.
C.
C.
Huang,
Cell
15,
615
(1978).
16.
R.
L.
Friedman,
S. P.
Manly,
M.
McMahon,
I.
M.
Kerr,
G.
R.
Stark,
ibid.
38,
745
(1984).
17.
T.
Maniatis,
E.
F.
Fritsch,
J.
Sambrook,
Molec-
ular
Cloning:
A
Laboratory
Manual
(Cold
Spring
Harbor
Laboratory,
Cold
Spring
Harbor,
N.Y.,
1982),
p.
202.
18.
We
thank
P.
Gunning
and
L.
Kedes
for
the
,3-
actin
cDNA
clone,
B.
Spiegelman
for
the
GPD
cDNA
clone,
P.
Lomedico
and
Hoffman-La
Roche
for
recombinant
IL-1,
and
L.
Liu,
J.
Larrick,
and
Cetus
Corp.
for
recombinant
tumor
necrosis
factor.
K.
Benight
helped
to
prepare
the
manuscript.
Supported
by
grants
from
NIH
(GM25821
and
AM0131401),
the
March
of
Dimes
(1-877),
Rockefeller
University
(84077)
and
Cetus
Corporation.
F.M.T.
was
supported
by
a
grant
from
the
Veterans
Administration.
G.M.R.
is
an
Established
Investigator
of
the
American
Heart
Association.
30
March
1985;
accepted
16
July
1985
obscure,
it
is
believed
that
the
toxic
effects
are
mediated
by
factors
produced
by
host
cells.
Adoptive
transfer
experi-
ments
with
LPS-resistant
(C3H/HeJ)
and
-sensitive
(C3H/HeN)
congenic
mice
have
implicated
cells
of
hematopoietic
origin
and,
in
particular,
monocytes,
as
the
source
of
these
mediators
(1,
2).
Recently,
we
reported
the
isolation
and
characterization
of
a
monokine,
ca-
chectin,
that
is
made
by
macrophages
stimulated
with
endotoxin
(3-6).
Cachec-
tin
completely
suppresses
the
synthesis
of
lipoprotein
lipase
(LPL)
in
adipocytes
in
vivo
and
in
vitro
(3-6).
Further
struc-
tural
studies
revealed
a
marked
homolo-
gy
between
cachectin
and
human tumor
necrosis
factor
(TNF),
and
subsequent
biological
studies
confirmed
that
purified
cachectin
had
TNF
activity
(7).
Although
most
studies
of
cachec-
tin/TNF
have
centered
on
its
antitumor
activity,
the
protein
is
produced,
in
vivo
and
in
vitro,
in
response
to
LPS
chal-
lenge
(6,
8-11),
and
binds
to
high-affinity
receptors
present
on
a
number
of
normal
host
tissues
(for
example,
liver,
muscle,
and
adipose
tissue)
(3).
We
have
previ-
ously
proposed
that
cachectin
may
func-
tion
as
a
hormone
to
promote
cellular
responses
which,
in
part,
result
in
the
mobilization
of
host
energy
reserves
in
response
to
invasion
(3,
8,
9).
In
the
present
study,
we
reasoned
that
cachectin/TNF
might
also
play
a
role
in
the
lethal
metabolic
effects
of
endotoxin-
mediated
shock.
Accordingly,
we
pas-
sively
immunized
mice
with
antibody
to
cachectin/TNF
and
challenged
them
with
lethal
amounts
of
LPS.
Cachec-
tin/TNF
was
purified
as
previously
de-
scribed
(7).
The
purified
protein
was
prepared
for
use
in
immunization
by
electrophoresis
in
sodium
dodecyl
sul-
fate(SDS)-polyacrylamide
slab
gel.
The
gel
was
sliced
after
completion
of
elec-
trophoresis,
and
approximately
5
jig
of
homogeneous
protein
(still
contained
within
the
gel
slice)
was
emulsified
in
1.0
ml
of
0.05M
ammonium
acetate
solution
and
1.0
ml
of
Freund's
complete
adju-
vant.
A
New
Zealand
White
female
rab-
bit
was
injected
at
multiple
subcutaneous
sites
with
this
material.
Four
additional
injections
were
given
at
monthly
inter-
vals,
with
2
to
5
jig
of
cachectin/TNF,
prepared
as
above
but with
Freund's
incomplete
adjuvant.
Blood
was
with-
drawn
from
the
rabbit
4
days
and
6
days
after
the
final
injection.
Immune
serum
was
tested
for
its
ability
to
precipitate
cachectin/TNF
labeled
with
125I
by
the
iodogen
method
(12)
(Fig.
1).
Approxi-
mately
50
percent
of
the
tracer
was
pre-
cipitated
when
the
serum
dilution
was
1:30,000;
preimmune
serum
was
nonre-
active.
The
specificity
of
immune
and
preimmune
sera
was
analyzed
by
im-
munoblotting
(Fig.
2).
A
single
major
species,
corresponding
to
murine
ca-
chectin/TNF,
was
labeled
when
blot
transfers
of
medium
from
RAW
264.7
cells
previously
incubated
with
LPS
were
exposed
to
the
immune
serum.
Occasionally,
the
presence
of
faint
bands
were
noted
in
the
gel
above
cachec-
tinITNF,
possibly
reflecting
precursor
molecules
or
glycosylation
products.
Preimmune
serum
showed
no
reactivity.
Neither
immune
nor
preimmune
se-
rum
contained
antibodies
reactive
with
LPS.
This
was
assessed
by
the
method
of
Neter
et
al.
(13),
in
which
human
erythrocytes
were
passively
sensitized
with
LPS
and
exposed
to
preimmune
or
immune
serum
over
a
range
of
dilutions
between
1:2
and
1:1000.
No
agglutina-
869
Passive
Immunization
Against
Cachectin/Tumor
Necrosis
Factor
Protects
Mice
from
Lethal
Effect
of
Endotoxin
Abstract.
A
highly
specific
polyclonal
rabbit
antiserum
directed
against
murine
cachectinltumor
necrosis
factor
(TNF)
was
prepared.
When
BALBIc
mice
were
passively
immunized
with
the
antiserum
or
with
purified
immune
globulin,
they
were
protected
against
the
lethal
effect
of
the
endotoxin
lipopolysaccharide
produced
by
Escherichia
coli.
The
prophylactic
effect
was
dose-dependent
and
was
most
effective
when
the
antiserum
was
administered
prior
to
the
injection
of
the
endotoxin.
Antiserum
to
cachectin/TNF
did
not
mitigate
the
febrile
response
of
endotoxin-
treated
animals,
and
very
high
doses
of
endotoxin
could
overcome
the
protective
effect.
The
median
lethal
dose
of
endotoxin
in
mice
pretreated
with
50
microliters
of
the
specific
antiserum
was
approximately
2.5
times
greater
the
median
lethal
dose
for
controls
given
nonimmune
serum.
The
data
suggest
that
cachectinlTNF
is
one
of
the
principal
mediators
of
the
lethal
effect
of
endotoxin.
on January 7, 2010 www.sciencemag.orgDownloaded from
tion
was
observed
at
any
dilution,
sug-
demonstrable
(Table
1)
compared
to
the
the
immune
serum
did
not
abrogate
the
gesting
that
if
LPS
antibodies
were
pre-
mortality
rate
observed
among
control
pyrogenic
effect
of
LPS.
A
similar
level
sent,
they
were
of
extremely
low
titer.
mice
treated
with
preimmune
serum
or
of
protection
was
observed
when
immu-
Similarly,
no
reactivity
was
observed
with
serum
from
other
nonimmune
rab-
noglobulin
G
(IgG)
prepared
from
the
when
LPS
(up
to
50
,ug)
was
subjected
to
bits.
Mice
pretreated
with
serum
from
immune
serum
was
administered
to
mice
electrophoresis,
transferred
to
nitrocel-
rabbits
immunized
against
bovine
serum
prior
to
the
injection
of
LPS.
In
contrast,
lulose,
and
exposed
to
immune
serum.
albumin
according
to
a
schedule
similar
IgG
prepared
from
nonimmune
serum
When
immune
serum
was
adminis-
to
that
used
in
cachectin
immunization
did
not
protect
the
mice.
tered
to
female
BALB/c
mice
by
intra-
fared
no
better
than
the
other
control
The
degree
of
protection
provided
by
peritoneal
injection
1.5
hours
before
the
groups.
After
administration
of
the
LPS,
the
immune
serum
was
assessed
by
ad-
intraperitoneal
injection
of
400
,ug
of
the
mice
in
all
of
the
groups
appeared
ill,
ministering
various
doses
of
LPS
to
mice
LPS,
a
significant
protective
effect
was
and
exhibited
a
febrile
response.
Thus,
that
had
received
either
50
,ul
of
nonim-
A
B
Fig.
1
(left).
Immunoprecipitation
of
125I-la-
25
so
beled
murine
cachectin/TNF.
Immune
serum
(dashed
line)
was
prepared
as
described
in
the
a
I
~~~~~~~~~~~~~~~~~~~~text.
Preimmune
serum
(solid
line)
was
ob-
tamned
from
the
same
animal.
Immunoprecip-
&.
itation
reactions
were
carried
out
in
conical
polypropylene
tubes.
Serum
samples
were
29-
X
0
-
0
-4
S
ffi00e-00
-7
-f
fff-;
0
i
f:
diluted
to
yield
the
final
concentrations
indi-
.g
\
29-
-;.
0.00i
.a
.i
0-.
-070.
i0
;cated,
using
Dulbecco's
phosphate
buffered
,i
o
*
e:-
'
saline
containing
1
percent
RIA-grade
bovine
.-0
i0X;$ri.i-78
5
-
i.-0.
.
serum
albumin
(Sigma)
(PBS/BSA).
Symbol:
7,771
f
ff
-
k
=0,
no
serum
added.
Two
microliters
(approxi-
1
3
5
7
9-
11
0
12
-
mately
0.2
ng;
1.4
x
104
count/min
of
labeled
Serum
dilution
(3-n)
cachectin/TNF
solution,
prepared
as
de-
1
2
3
45S
6
1
2
3
4
5
scribed
(3,
11),
was
mixed
with
20
,ul
of
diluted
serum.
Solutions
were
incubated
at
4°C
for
5
hours.
At
that
time,
10
,ul
of
1:
5
suspension
of
washed
Staphylococcus
aureus
bacterial
adsorbant
(Miles
Laboratories)
in
distilled
water
was
added
to
each
tube.
After
30
minutes,
1.0
ml
of
PBS/BSA
was
added
to
each
tube.
Samples
were
quickly
mixed,
and
the
immunoprecipitates
were
sedimented
with
a
Beckman
microfuge.
Supernatants
were
aspirated,
and
the
pellets
counted
for
radioactivity
with
a
Packard
Autogamma
Scintillation
Spectrometer,
model
578.
Results
are
expressed
as
the
percentage
of
radioactivity
precipitated.
Nonimmune
sera
from
other
rabbits
(data
not
shown)
were
also
incapable
of
precipitating
labeled
cachectin/TNF.
Fig.
2
(right).
Western
blot
of
crude
RAW
264.7
cell
(nmouse
macrophage)
conditioned
medium,
and
of
LPS.
Lane
I
(A
and
B):
50
,g
Escherichia
coli
strain
0127:B8
LPS
(Difco).
Lanes
2
to
6
(A)
and
2
to
5
(B):
25
,ul
of
LPS-stimulated
RAW
264.7
cell
conditioned
medium
(containing
approximately
40
Fig
of
total
protein),
prepared
and
concentrated
50-fold
as
described
(3).
Samples
were
applied
to
a
10
to
15
percent
SDS-polyacrylamide
gradient
gel.
After
completion
of
electrophoresis,
protein
was
transferred
to
nitrocellulose
paper
(Schleicher
and
Schuell)
by
using
an
electroblotting
apparatus
(Bio-Rad).
Immune
(A),
and
preimmune
(B)
sera
were
applied
to
the
nitrocellulose
blot
at
dilutions
of
1:
300,
and
reactive
bands
were
identified
with
125I-labeled
protein
A
(New
England
Nuclear)
according
to
the
protocol
outlined
by
Hotez
et
al.
(14).
The
band
visible
in
(A)
corresponds
precisely
in
molecular
weight
to
murine
cachectin/TNF
when
the
purified
radioiodinated
protein
is
subjected
to
electrophoresis
and
transferred
in
parallel
as
a
marker.
No
reactive
proteins
were
apparent
when
preimmune
serum
was
applied
to
the
blot.
100
_
16
Fig.
3
(left).
Estima-
0
/
_
-
tion
of
the
endotoxin
/
/^
NQ
LD50
in
mice
treated
80
2
12
-
with
immune
and
non-
/'
/
,
| 4 r
4
,immune
sera.
Male
We
0
/
8
.
li
.
BALB/c
mice
(19
to
60
8
/
zT
A
21
g)
were
randomly
.
' /
l * / 1
_
assigned
to
11
groups,
>
4
-(P>0.2)
each
containing
15
to
040
2
I
/
X
17
members.
In
four
C(P<0.02)
groups
(triangles),
20
0
0
E(P<0.001)
D
(P
<0.001)
mice
were
given
intra-
/
2
4
--
peritoneal
injections
0
2
4
6
of
200
,ul
of
immune
O
X
time
after
injection
of
LPS
(days)
serum
diluted
1:4
0
240
360
480
soo
with
sterile
isotonic
Dose
(2g/mouse)
saline;
in
the
remaining
seven
groups
(open
circles),
mice
were
given
the
same
volume
of
1:4
diluted
nonimmune
serum.
After
5
hours,
mice
were
given
intraperitoneal
injections
of
various
quantities
of
E.
coli
strain
0127:B8
LPS,
dissolved
in
sterile
saline.
All
mice
were
observed
for
48
hours
after
LPS
injection.
Curves
were
fitted
to
the
data,
and
computation
of
LD50
was
accomplished
as
described
by
Bliss
(15),
and
by
Litchfield
et
al.
(16).
Horizontal
bars
indicate
the
95
percent
confidence
limits
of
LD50
determinations
(91
jig
for
mice
treated
with
nonimmune
serum,
and
227
,ug
for
mice
treated
with
immune
serum).
The
administration
of
larger
volumes
of
immune
serum
to
mice
(up
to
200
,ul)
did
not
further
improve
their
survival.
Fig.
4
(right).
Kaplan-Meier
plot
of
survival
after
LPS
treatment.
Eighty
male
BALB/c
mice
(19
to
21
g)
were
randomly
divided
into
five
groups
of
16
members
each.
Mice
were
protected
by
an
intraperitoneal
injection
of
200
,ll
of
antiserum
to
cachectin/TNF
given
6
or
3
hours
before,
concurrent
with,
or
3
or
6
hours
after
intraperitoneal
injection
of
400
,ug
of
E.
coli
strain
0127:B8
LPS
(Difco).
Censored
determinations
of
survival
were
made
at
the
in-
dicated
time
points
following
LPS
injection.
(A)
Survival
was
optimal
when
mice
were
treated
with
antiserum
6
hours
before
LPS
administration.
(B-D)
A
decrease
in
survival,
with
respect
to
these
values,
was
noted
when
mice
were
given
antiserum
to
cachectin/TNF
3
hours
before
LPS
administration
(B),
at
the
time
of
LPS
administration
(C),
3
hours
after
LPS
administration
(D),
or
6
hours
after
LPS
administration
(E).
Tests
of
the
null
hypothesis
that
the
time-related
probability
of
survival
in
each
group
was
equal
to
that
in
the
group
protected
6
hours
before
LPS
administration
were
performed
according
to
the
Gehan
method
(17);
P
values
(single-tail
normal
distribution;
with
Bonferroni
correction
for
multiple
comparisons)
are
indicated
for
each
curve.
870
SCIENCE,
VOL.
229
on January 7, 2010 www.sciencemag.orgDownloaded from
mune
serum
or
50
,ul
of
immune
serum
5
hours
previously
(Fig.
3).
The
median
lethal
dose
(LD50)
of
LPS
in
animals
treated
with
immune
serum
was
signifi-
cantly
higher
than
the
LD50
for
control
mice
treated
with
nonimmune
serum.
The
time
at
which
the
antiserum
was
administered
relative
to
the
time
of
LPS
administration
was
found
to
be
of
crucial
importance
in
producing
a
protective
ef-
fect.
Mice
that
were
injected
with
im-
mune
serum
3
or
6
hours
prior
to
admin-
istration
of
LPS
fared
better
than
those
passively
immunized
at
the
time
of
LPS
injection
or
several
hours
after
(Fig.
4).
This
finding
suggests
that
endotoxin
elic-
its
cachectin/TNF
production
soon
after
its
administration,
and
that
cachec-
tin/TNF
mediates
its
lethal
injury
within
a
very
short
time.
In
rabbits,
cachec-
tin/TNF
is
produced
within
minutes
after
the
intravenous
administration
of
LPS,
and
peak
plasma
concentrations
are
ob-
served
after
2
hours,
with
a
rapid
decline
in
concentrations
occurring
thereafter
(11).
Hence,
in
this
model,
the
animal
is
exposed
to
high
concentrations
of
the
hormone
only
briefly;
it
is
within
this
interval
of
time
that
effective
antibody
concentrations
must
be
present
if
protec-
tion
is
to
be
achieved.
Presumably
the
necessity
for
prior
administration
of
the
antiserum
reflects
the
time
required
for
complete
distribution
of
the
antibody
within
the
recipient
animal.
These
data
give
evidence
for
the
role
of
cachectin/TNF
in
mediating
the
lethal
effects
of
LPS.
Cachectin/TNF
is
clearly
only
one
of
the
mediators
responsible
for
the
numerous
pathological
effects
evoked
by
LPS,
since
the
passively
im-
munized
mice
become
febrile,
and
con-
tinue
to
appear
ill
and
distressed.
It
is
possible,
for
example,
that
cachec-
tinfTNF
acts
in
concert
with
other
medi-
ators
(for
example,
interleukin-1,
inter-
ferons,
and
lymphotoxin)
in
order
to
elicit
the
lethal
effect
of
LPS.
It
is
important
to
note
that
mice
are
relatively
resistant
to
the
effects
of
LPS
when
compared
to
most
other
mammals;
rabbits,
for
example,
are
approximately
1000-fold
more
sensitive.
In
LPS-sensi-
tive
species,
TNF
may
play
a
more
prominent
role
as
a
mediator
of
shock.
Immunization
against
TNF
might
then
be
expected
to
yield
a
higher
level
of
protection.
The
potential
utility
of
passive
immu-
nization
with
antisera
to
cachectin/TNF
in
animals
with
shock
induced
by
septi-
cemia
(or
possibly
other
causes)
needs
further
exploration.
An
obvious
corol-
lary
is
the
possibility
that
agents
which
affect
the
synthesis
or
binding
of
cachec-
tin/TNF
to
its
receptor
might
be
of
utility
30
AUGUST
1985
Table
1.
Protective
effect
of
antiserum
to
cachectin/TNF.
Female
BALB/c
mice
(20
to
24
g)
were
randomly
divided
into
six
groups
and
injected
intraperitoneally
with
serum
from
immune
or
nonimmune
rabbits
1.5
hours
before
being
injected
with
400
iag
of
LPS
from
E.
coli
strain
0127:B8.
Serum
samples
were
diluted
with
sterile
isotonic
saline
and
injected
in
a
final
volume
of
0.2
ml
per
mouse.
LPS
was
also
diluted
in
sterile
saline
and
injected
in
a
volume
of
0.2
ml.
Mortality
was
recorded
daily,
and
the
experiment
was
considered
complete
when
no
deaths
were
observed
in
any
group
for
3
days.
The
data
show
the
number
of
survivors
7
days
after
LPS
injec-
tion.
Serum
volume
injected
(,A)
Serum
10
50
200
Immune
3/14
6/14
7/14
Nonimmune
0/14
1/14
0/14
P
>
0.05
<
0.05
<0.01
*Chi-square
test.
in
this
setting
without
compromising
the
host's
immune
system.
From
these
stud-
ies,
a
better
understanding
of
the
mecha-
nisms
by
which
the
immune
system
in-
fluences
other
tissues
may
emerge.
JAMES
HANKEN
Department
of
Environmental,
Population,
and
Organismic
Biology,
University
of
Colorado,
Boulder
80309
The
origin
of
novel
morphological
de-
sign
is
a
primary
focus
of
evolutionary
morphology.
One
trend
that
may
pro-
mote
morphological
novelty
is
phyloge-
netic
decrease
in
body
size,
or
miniatur-
ization.
Unique
morphological
arrange-
ments
are
a
common
feature
of
dwarfed
invertebrates
(1,
2),
many
of
which
rep-
resent
"entirely
new
types
of
organiza-
tion"
(2).
In
vertebrates,
a
frequent
asso-
ciation
between
miniaturization
and
morphological
novelty
has
been
docu-
mented
in
many
taxa,
including
teleost
fishes
(3),
anuran
(4)
and
urodele
(5-7)
amphibians,
and
squamate
(8-10)
and
amphisbaenid
(11)
reptiles.
Size
de-
References
and
Notes
1.
B.
M.
Sultzer,
Nature
(London)
219,
1253
(1968).
2.
J.
P.
Filkins,
Fed.
Proc.
Fed.
Am.
Soc.
Exp.
Biol.
44,
300
(1985).
3.
B.
Beutler,
J.
Mahoney,
N.
Le
Trang,
P.
Pekala,
A.
Cerami,
J.
Exp.
Med.
161,
984
(1985).
4.
,
Fed.
Proc.
Fed.
Am.
Soc.
Exp.
Biol.
44,
1704
(1985).
5.
B.
Beutler,
J.
Mahoney,
P.
Pekala,
N.
Le
Trang,
A.
Cerami,
Blood
64,
65a
(1984).
6.
J.
R.
Mahoney,
Jr.,
B.
A.
Beutler,
N.
Le
Trang,
W.
Vine,
Y.
Ikeda,
M.
Kawakami,
A.
Cerami,
J.
Immunol.
134,
1673
(1985).
7.
B.
Beutler
et
al.,
Nature
(London),
in
press.
8.
M.
Kawakami
and
A.
Cerami,
J.
Exp.
Med.
154,
631
(1981).
9.
M.
Kawakami,
P.
H.
Pekala,
M.
D.
Lane,
A.
Cerami,
Proc.
Natl.
Acad.
Sci.
U.S.A.
79,
912
(1982).
10.
M.
Kawakami,
Y.
Ikeda,
N.
Le
Trang,
W.
Vine,
A.
Cerami,
Proceedings
of
the
IUPHAR
(Mac-
millan,
New
York,
1984),
p.
377.
11.
B.
Beutler
and
A.
Cerami,
in
preparation.
12.
P.
J.
Fraker
and
J.
C.
Speck,
Biochem.
Biophys.
Res.
Commun.
80,
849
(1978).
13.
E.
Neter,
0.
Westphal,
0.
Luderitz,
E.
Gor-
zynski,
E.
Eichenberger,
J.
Immunol.
76,
377
(1956).
14.
P.
J.
Hotez,
N.
Le
Trang,
J.
H.
McKerrow,
A.
Cerami,
J.
Biol.
Chem.
in
press.
15.
C.
I.
Bliss,
Q.
J.
Pharm.
Pharmacol.
11,
192
(1938).
16.
J.
T.
Litchfield,
Jr.,
and
F.
Wilcoxon,
J.
Phar-
macol.
Exp.
Ther.
96,
99
(1949).
17.
E.
A.
Gehan,
Biometrika
52,
203
(1965).
18.
Supported
by
National
Institutes
of
Health
grant
AM01314
and
by
Rockefeller
Foundation
grant
84077.
We
thank
N.
Agabian
for
helpful
discus-
sions.
31
May
1985;
accepted
16
July
1985
crease
also
has
been
implicated
as
a
critical
factor
in
the
evolution
of
higher
taxa
such
as
frogs
and
salamanders
(5),
lizards
(8),
and
snakes
(10).
Lungless
salamanders
(Plethodonti-
dae)
provide
some
of
the
best
examples
of
miniaturization
among
vertebrates.
Decreased
body
size
has
evolved
in
sev-
eral
lineages;
one,
the
Mexican
genus
Thorius,
comprises
the
smallest
extant
tailed
tetrapods
(6,
7,
12).
In
this
report
I
analyze
the
consequences
of
miniatur-
ization
in
Thorius
for
forelimb
skeletal
morphology.
The
analysis
is
based
on
quantitative
estimates
of
natural
varia-
tion
in
limb
osteology,
on
a
comparison
of
skeletal
unit
homology,
and
on
an
electrophoretically
derived
molecular
phylogeny.
It
reveals
the
following:
(i)
miniaturization
of
the
genus
as
a
whole
correlates
with
the
appearance
of
several
871
Morphological
Novelty
in
the
Limb
Skeleton
Accompanies
Miniaturization
in
Salamanders
Abstract.
Salamanders
of
the
genus
Thorius
(Plethodontidae)
are
among
the
smallest
tetrapods.
Hypotheses
of
limb
skeletal
evolution
in
these
vertebrates
were
evaluated
on
the
basis
of
estimates
of
natural
variation,
comparisons
of
skeletal
homology,
and
analysis
of
molecular
phylogeny.
Nine
carpal
arrangements
occur
in
Thorius,
more
than
in
all
twelve
related
genera
of
typically
larger
salamanders;
six
of
these
arrangements
are
unique.
They
represent
a
trend
toward
a
decrease
in
the
number
of
separate
cartilages
that
is
independent
of
locomotor
and
ecological
specialization.
Miniaturization
may
be
an
important
source
of
morphological
novelty,
distinct
from
local
adaptation,
in
vertebrates.
on January 7, 2010 www.sciencemag.orgDownloaded from
    • "But as predicted in the wider literature (above), it generated side effects that mimicked the onset of severe sepsis [48]. Yet while prior anti-TNF antibody prevented the illness caused by injecting LPS to induce experimental sepsis, it was of no help when administered to mice that had already been made sick by LPS [39]. This gave fair warning that such antibody was likely to be impractical for treating acute inflammatory disease once it is underway, and so it proved when etanercept, a major commercial anti-TNF biological agent, was first tested in sepsis patients [49]. "
    [Show abstract] [Hide abstract] ABSTRACT: Tumor necrosis factor (TNF) is an ancient and widespread cytokine required in small amounts for much physiological function. Higher concentrations are central to innate immunity, but if unchecked this cytokine orchestrates much chronic and acute disease, both infectious and noninfectious. While being a major proinflammatory cytokine, it also controls homeostasis and plasticity in physiological circumstances. For the last decade or so these principles have been shown to apply to the central nervous system as well as the rest of the body. Nevertheless, whereas this approach has been a major success in treating noncerebral disease, its investigation and potential widespread adoption in chronic neurological conditions has inexplicably stalled since the first open trial almost a decade ago. While neuroscience is closely involved with this approach, clinical neurology appears to be reticent in engaging with what it offers patients. Unfortunately, the basic biology of TNF and its relevance to disease is largely outside the traditions of neurology. The purpose of this review is to facilitate lowering communication barriers between the traditional anatomically based medical specialties through recognition of shared disease mechanisms and thus advance the prospects of a large group of patients with neurodegenerative conditions for whom at present little can be done.
    Full-text · Article · Jul 2015
    • "The first group is the early proinflammatory mediators, such as TNF-α and IL-1β, which are induced within hours after the induction of sepsis. Initial experimental observations have suggested that the inhibition of these cytokines with neutralizing antibodies is beneficial in the treatment of sepsis [31]. However, clinical trials and more elegant animal studies have shown that the neutralization of early cytokines does not protect against sepsis [32]. "
    [Show abstract] [Hide abstract] ABSTRACT: The nuclear DNA binding protein high mobility group box 1 (HMGB1) has recently been suggested to act as a late mediator of septic shock. The effect of ((S)-6,7-dihydroxy-1-(4-hydroxynaphthylmethyl)-1,2,3,4-tetrahydroisoquinoline alkaloid, also known as THI-56, in an experimental model of sepsis was investigated. THI-56 exhibited potent anti-inflammatory properties in response to LPS in RAW 264.7 cells. In particular, THI-56 significantly inhibited the expression of inducible nitric oxide synthase (iNOS) and the release of HMGB1 in activated macrophages. THI-56 activated NE-F2-regulated factor 2 (Nrf-2)/heme oxygenase 1 (HO-1). The specific knockdown of the HO-1 gene by HO-1 siRNA significantly reversed the inhibitory effects of THI-56 on iNOS expression and HMGB1 release in LPS-stimulated macrophages. Importantly, THI-56 administration protected animals from death induced by either a lethal dose of LPS or cecal ligation and puncture (CLP). Furthermore, the ALT, AST, BUN, creatinine, and HMGB1 levels in the blood were significantly increased in CLP-induced septic mice, and the administration of THI-56 reduced these levels in a concentration-dependent and zinc protoporphyrin IX (ZnPPIX)-sensitive manner. In addition, the administration of THI-56 significantly ameliorated not only lung damage but also macrophage infiltration in the livers of CLP-induced septic mice, and these effects were also abrogated in the presence of ZnPPIX. Thus, we conclude that THI-56 significantly attenuates the proinflammatory response induced by LPS and reduces organ damage in a CLP-induced sepsis model through the upregulation of Nrf-2/HO-1.
    Full-text · Article · Oct 2013
  • Article · Jan 1993
Show more