Content uploaded by Martha Reed Herbert
Author content
All content in this area was uploaded by Martha Reed Herbert on Jul 19, 2022
Content may be subject to copyright.
Please
cite
this
article
in
press
as:
M.R.
Herbert,
C.
Sage,
Autism
and
EMF?
Plausibility
of
a
pathophysiological
link
part
II,
Pathophysiology
(2013),
http://dx.doi.org/10.1016/j.pathophys.2013.08.002
ARTICLE IN PRESS
PATPHY-777;
No.
of
Pages
24
Pathophysiology
xxx
(2013)
xxx–xxx
Autism
and
EMF?
Plausibility
of
a
pathophysiological
link
part
II
Martha
R.
Herbert a,∗,
Cindy
Sage b
aMassachussetts
General
Hospital
Harvard
Medical
School
Boston,
TRANSCEND
Research
Program
Neurology,
Boston,
MA,
USA
bSage
Associates,
Santa
Barbara,
CA,
USA
Abstract
Autism
spectrum
conditions
(ASCs)
are
defined
behaviorally,
but
they
also
involve
multileveled
disturbances
of
underlying
biology
that
find
striking
parallels
in
the
physiological
impacts
of
electromagnetic
frequency
and
radiofrequency
radiation
exposures
(EMF/RFR).
Part
I
(Vol
776)
of
this
paper
reviewed
the
critical
contributions
pathophysiology
may
make
to
the
etiology,
pathogenesis
and
ongoing
generation
of
behaviors
currently
defined
as
being
core
features
of
ASCs.
We
reviewed
pathophysiological
damage
to
core
cellular
processes
that
are
associated
both
with
ASCs
and
with
biological
effects
of
EMF/RFR
exposures
that
contribute
to
chronically
disrupted
homeostasis.
Many
studies
of
people
with
ASCs
have
identified
oxidative
stress
and
evidence
of
free
radical
damage,
cellular
stress
proteins,
and
deficiencies
of
antioxidants
such
as
glutathione.
Elevated
intracellular
calcium
in
ASCs
may
be
due
to
genetics
or
may
be
downstream
of
inflammation
or
environmental
exposures.
Cell
membrane
lipids
may
be
peroxidized,
mitochondria
may
be
dysfunctional,
and
various
kinds
of
immune
system
disturbances
are
common.
Brain
oxidative
stress
and
inflammation
as
well
as
measures
consistent
with
blood–brain
barrier
and
brain
perfusion
compromise
have
been
documented.
Part
II
of
this
paper
documents
how
behaviors
in
ASCs
may
emerge
from
alterations
of
electrophysiological
oscillatory
synchronization,
how
EMF/RFR
could
contribute
to
these
by
de-tuning
the
organism,
and
policy
implications
of
these
vulnerabilities.
It
details
evidence
for
mitochondrial
dysfunction,
immune
system
dysregulation,
neuroinflammation
and
brain
blood
flow
alterations,
altered
electrophysiology,
disruption
of
electromagnetic
signaling,
synchrony,
and
sensory
processing,
de-tuning
of
the
brain
and
organism,
with
autistic
behaviors
as
emergent
properties
emanating
from
this
pathophysiology.
Changes
in
brain
and
autonomic
nervous
system
electrophysiological
function
and
sensory
processing
predominate,
seizures
are
common,
and
sleep
disruption
is
close
to
universal.
All
of
these
phenomena
also
occur
with
EMF/RFR
exposure
that
can
add
to
system
overload
(‘allostatic
load’)
in
ASCs
by
increasing
risk,
and
can
worsen
challenging
biological
problems
and
symptoms;
conversely,
reducing
exposure
might
ameliorate
symptoms
of
ASCs
by
reducing
obstruction
of
physiological
repair.
Various
vital
but
vulnerable
mechanisms
such
as
calcium
channels
may
be
disrupted
by
environmental
agents,
various
genes
associated
with
autism
or
the
interaction
of
both.
With
dramatic
increases
in
reported
ASCs
that
are
coincident
in
time
with
the
deployment
of
wireless
technologies,
we
need
aggressive
investigation
of
potential
ASC—EMF/RFR
links.
The
evidence
is
sufficient
to
warrant
new
public
exposure
standards
benchmarked
to
low-intensity
(non-thermal)
exposure
levels
now
known
to
be
biologically
disruptive,
and
strong,
interim
precautionary
practices
are
advocated.
©
2013
Elsevier
Ireland
Ltd.
All
rights
reserved.
Keywords:
Autism;
EMF/RFR;
Cellular
stress;
Oxidative
stress;
Mitochondrial
dysfunction;
Oscillatory
synchronization;
Environment;
Radiofrequency;
Wireless;
Children;
Fetus;
Microwave
1.
Recap
of
part
I
and
summary
of
part
II
Part
I
of
this
two-part
article
previously
documented
a
series
of
parallels
between
the
pathophysiological
and
genotoxic
impacts
of
EMF/RFR
and
the
pathophysiologi-
cal,
genetic
and
environmental
underpinnings
of
ASCs.
DNA
∗Corresponding
author.
E-mail
address:
drmarthaherbert@gmail.com
(M.R.
Herbert).
damage,
immune
and
blood–brain
barrier
disruption,
cellular
and
oxidative
stress,
calcium
channel
dysfunction,
disturbed
circadian
rhythms,
hormone
dysregulation,
and
degraded
cognition,
sleep,
autonomic
regulation
and
brainwave
activity—all
are
associated
with
both
ASCs
and
EMF/RFR;
and
the
disruption
of
fertility
and
reproduction
associated
with
EMF/RFR
may
also
be
related
to
the
increasing
inci-
dence
of
ASCs.
All
of
this
argues
for
reduction
of
exposures
now,
and
better
coordinated
research
in
these
areas.
These
0928-4680/$
–
see
front
matter
©
2013
Elsevier
Ireland
Ltd.
All
rights
reserved.
http://dx.doi.org/10.1016/j.pathophys.2013.08.002
Please
cite
this
article
in
press
as:
M.R.
Herbert,
C.
Sage,
Autism
and
EMF?
Plausibility
of
a
pathophysiological
link
part
II,
Pathophysiology
(2013),
http://dx.doi.org/10.1016/j.pathophys.2013.08.002
ARTICLE IN PRESS
PATPHY-777;
No.
of
Pages
24
2
M.R.
Herbert,
C.
Sage
/
Pathophysiology
xxx
(2013)
xxx–xxx
pathophysiological
parallels
are
laid
out
after
identifying
the
dynamic
features
of
ASCs
that
could
plausibly
arise
out
of
such
pathophysiological
dysregulation.
The
importance
of
transduction
between
levels
was
also
discussed
in
Part
I.
Part
II
elucidates
in
much
more
detail
the
possible
interfaces
between
the
cellular
and
molecular
pathophys-
iology
reviewed
above
and
the
higher-level
disruption
of
physiological
systems,
brain
tissue
and
nervous
system
electrophysiology.
It
addresses
mitochondrial
dysfunction,
immune
system
disregulation,
neuroinflammation
and
brain
blood
flow
alterations,
altered
electrophysiology,
disruption
of
electromagnetic
signaling,
synchrony,
and
sensory
pro-
cessing,
de-tuning
of
the
brain
and
organism,
and
behavior
as
an
emergent
property.
The
emergence
of
ever
larger
amounts
of
data
is
transforming
our
understanding
of
ASCs
from
static
encephalopathies
based
on
genetically
caused
brain
damage
to
dynamic
encephalopathies
where
challenging
behaviors
emanate
from
physiologically
disrupted
systems.
In
parallel,
the
emergence
of
ever
larger
bodies
of
evidence
supporting
a
large
array
of
non-thermal
but
profound
pathophysiological
impacts
of
EMF/RFR
is
transforming
our
understanding
of
the
nature
of
EMF/RFR
impacts
on
the
organism.
At
present
our
policies
toward
ASCs
are
based
on
outdated
assumptions
about
autism
being
a
genetic,
behavioral
condition,
whereas
our
medical,
educational
and
public
health
policies
related
to
treatment
and
prevention
could
be
much
more
effective
if
we
took
whole-body,
gene-environment
considerations
into
account,
because
there
are
many
lifestyle
and
environmen-
tal
modifications
that
could
reduce
morbidity
and
probably
incidence
of
ASCs
as
well.
Our
EMF/RFR
standards
are
also
based
on
an
outdated
assumption
that
it
is
only
heating
(ther-
mal
injury)
which
can
do
harm.
These
thermal
safety
limits
do
not
address
low-intensity
(non-thermal)
effects.
The
evi-
dence
is
now
overwhelming
that
limiting
exposures
to
those
causing
thermal
injury
alone
does
not
address
the
much
broader
array
of
risks
and
harm
now
clearly
evident
with
chronic
exposure
to
low-intensity
(non-thermal)
EMF/RFR.
In
particular,
the
now
well-documented
genotoxic
impacts
of
EMF/RFR,
placed
in
parallel
with
the
huge
rise
in
reported
cases
of
ASCs
as
well
as
with
the
de
novo
mutations
associ-
ated
with
some
cases
of
ASCs
(as
well
as
other
conditions),
make
it
urgent
for
us
to
place
the
issue
of
acquired
as
well
as
inherited
genetic
damage
on
the
front
burner
for
scien-
tific
investigation
and
policy
remediation.
With
the
rising
numbers
people
with
ASCs
and
other
childhood
health
and
developmental
disorders,
and
with
the
challenges
to
our
prior
assumptions
posed
ever
more
strongly
by
emerging
evi-
dence,
we
need
to
look
for
and
act
upon
risk
factors
that
are
largely
avoidable
or
preventable.
We
argue
that
the
evi-
dence
is
sufficient
to
warrant
new
public
exposure
standards
benchmarked
to
low-intensity
(non-thermal)
exposure
levels
causing
biological
disruption
and
strong,
interim
precaution-
ary
practices
are
advocated.
The
combined
evidence
in
Parts
I
and
II
of
this
article
provide
substantial
pathophysiological
support
for
parallels
between
ASCs
and
EMF/RFR
health
impacts.
2.
Parallels
in
pathophysiology
2.1.
Degradation
of
the
integrity
of
functional
systems
EMF/RFR
exposures
can
yield
both
psychological
and
physiological
stress
leading
to
chronically
interrupted
homeostasis.
In
the
setting
of
molecular,
cellular
and
tis-
sue
damage,
one
would
predict
that
the
organization
and
efficiency
of
a
variety
of
organelles,
organs
and
functional
systems
would
also
be
degraded.
In
this
section
we
will
review
disturbances
from
EMF/RFR
in
systems
(including
include
oxidative
and
bioenergetics
metabolism,
immune
function
and
electrophysiological
oscillations)
that
include
molecular
and
cellular
components
subject
to
the
kinds
of
damage
discussed
in
the
previous
section.
We
will
review
disturbances
that
have
been
associated
with
EMF/RFR,
and
consider
the
parallel
disturbances
that
have
been
documented
in
ASCs.
2.1.1.
Mitochondrial
dysfunction
Mitochondria
are
broadly
vulnerable,
in
part
because
the
integrity
of
their
membranes
is
vital
to
their
optimal
functioning—including
channels
and
electrical
gradients,
and
their
membranes
can
be
damaged
by
free
radicals
which
can
be
generated
in
myriad
ways.
Moreover,
just
about
every
step
in
their
metabolic
pathways
can
be
targeted
by
envi-
ronmental
agents,
including
toxicants
and
drugs,
as
well
as
mutations
[1].
This
supports
a
cumulative
‘allostatic
load’
model
for
conditions
in
which
mitochondrial
dysfunction
is
an
issue,
which
includes
ASCs
as
well
as
myriad
other
chronic
conditions.
Mitochondria
are
commonly
discussed
in
terms
of
the
bio-
chemical
pathways
and
cascades
of
events
by
which
they
metabolize
glucose
and
generate
energy.
But
in
parallel
with
this
level
of
function
there
also
appears
to
be
a
dimension
of
electromagnetic
radiation
that
is
part
of
the
activity
of
these
organelles.
For
example,
electromagnetic
radiation
can
be
propagated
through
the
mitochondrial
reticulum,
which
along
with
the
mitochondria
has
a
higher
refractive
index
than
the
surrounding
cell
and
can
serve
to
propagate
electro-
magnetic
radiation
within
the
network
[2].
It
is
also
the
case
that
“The
physiological
domain
is
characterized
by
small-
amplitude
oscillations
in
mitochondrial
membrane
potential
(delta
psi(m))
showing
correlated
behavior
over
a
wide
range
of
frequencies.
.
..
Under
metabolic
stress,
when
the
bal-
ance
between
ROS
[reactive
oxygen
species,
or
free
radicals]
generation
and
ROS
scavenging
[as
by
antioxidants]
is
per-
turbed,
the
mitochondrial
network
throughout
the
cell
locks
to
one
main
low-frequency,
high-amplitude
oscillatory
mode.
This
behavior
has
major
pathological
implications
because
the
energy
dissipation
and
cellular
redox
changes
that
occur
during
delta
psi(m)
depolarization
result
in
suppression
of
electrical
excitability
and
Ca2
+
handling.
.
.”
[3].
These
electromagnetic
aspects
of
mitochondrial
physiol-
ogy
and
pathophysiology
could
very
well
be
impacted
by
EMF/RFR.
Please
cite
this
article
in
press
as:
M.R.
Herbert,
C.
Sage,
Autism
and
EMF?
Plausibility
of
a
pathophysiological
link
part
II,
Pathophysiology
(2013),
http://dx.doi.org/10.1016/j.pathophys.2013.08.002
ARTICLE IN PRESS
PATPHY-777;
No.
of
Pages
24
M.R.
Herbert,
C.
Sage
/
Pathophysiology
xxx
(2013)
xxx–xxx
3
Other
types
of
mitochondrial
damage
have
been
docu-
mented
in
at
least
some
of
the
studies
that
have
examined
the
impacts
of
EMF/RFR
upon
mitochondria.
These
include
reduced
or
absent
mitochondrial
cristae
[4–6],
mitochondrial
DNA
damage
[7],
swelling
and
crystallization
[5],
alterations
and
decreases
in
various
lipids
suggesting
an
increase
in
their
use
in
cellular
energetics
[8],
damage
to
mitochondrial
DNA
[7],
and
altered
mobility
and
lipid
peroxidation
after
exposures
[9].
Also
noted
has
been
enhancement
of
brain
mitochondrial
function
in
Alzheimer’s
transgenic
mice
and
normal
mice
[10].
The
existent
of
positive
as
well
as
negative
effects
gives
an
indication
of
the
high
context
dependence
of
exposure
impacts,
including
physical
factors
such
as
fre-
quency,
duration,
and
tissue
characteristics
[11].
By
now
there
is
a
large
amount
of
evidence
for
biochemi-
cal
and
other
abnormalities
in
a
large
portion
of
children
with
autism
that
are
consistent
with
mitochondrial
dysfunction
[12–14].
Recently
published
postmortem
brain
tissue
studies
that
have
added
a
new
dimension
of
evidence
for
mitochon-
drial
abnormalities
in
ASCs
will
be
reviewed
in
the
section
on
alteration
of
brain
cells
below.
Secondary
mitochondrial
dysfunction
(i.e.
environmen-
tally
triggered
rather
than
rooted
directly
in
genetic
mutations)
[15–18]
could
result
among
other
things
from
the
already
discussed
potential
for
EMF/RFR
to
damage
chan-
nels,
membranes
and
mitochondria
themselves
as
well
as
from
toxicant
exposures
and
immune
challenges.
In
a
meta-
analysis
of
studies
of
children
with
ASC
and
mitochondrial
disorder,
the
spectrum
of
severity
varied,
and
79%
of
the
cases
were
identified
by
laboratory
findings
without
associ-
ated
genetic
abnormalities
[16].
2.1.2.
Melatonin
dysregulation
2.1.2.1.
Melatonin,
mitochondria,
glutathione,
oxidative
stress.
Melatonin
is
well-known
for
its
role
in
regulation
of
circadian
rhythms,
but
it
also
plays
important
metabolic
and
regulatory
roles
in
relation
to
cellular
protection,
mitochon-
drial
malfunction
and
glutathione
synthesis
[19–21].
It
also
helps
prevent
the
breakdown
of
the
mitochondrial
membrane
potential,
decrease
electron
leakage,
and
thereby
reduce
the
formation
of
superoxide
anions
[22].
Pharmacological
doses
of
melatonin
not
only
scavenge
reactive
oxygen
and
nitrogen
species,
but
enhance
levels
of
glutathione
and
the
expression
and
activities
of
some
glutathione-related
enzymes
[21,23].
2.1.2.2.
Melatonin
can
attenuate
or
prevent
some
EMF/RFR
effects.
Melatonin
may
have
a
protective
effect
in
the
setting
of
some
EMF/RFR
exposures,
apparently
in
relation
to
these
functions
just
described.
EMF/RFR
can
impact
melatonin;
one
example
is
exposure
to
900
MHz
microwave
radiation
promoted
oxidation,
which
reduced
levels
of
melatonin
and
increased
creatine
kinase
and
caspase-3
in
exposed
as
com-
pared
to
sham
exposed
rats
[24].
Melatonin
can
attenuate
or
prevent
oxidative
damage
from
EMF/RFR
exposure.
In
an
experiment
exposing
rats
to
microwave
radiation
(MW)
from
a
GSM-900
mobile
phone
with
and
without
melatonin
treatment
to
study
renal
impacts
[25],
the
untreated
exposed
rats
showed
increases
of
lipid
peroxidation
markers
as
reduction
of
the
activities
of
superoxide
dismutase,
catalase
and
glutathione
peroxidase
indicating
decrement
in
antioxidant
status.
However
these
negative
effects
were
inhibited
in
the
exposed
rats
treated
with
melatonin.
Melatonin
also
inhibited
the
emergence
of
preneoplastic
liver
lesions
in
rats
exposed
to
EMFs
[26].
The
development
of
DNA
strand
breaks
was
observed
in
RFR
exposed
rats;
this
DNA
damage
was
blocked
by
melatonin
[27].
Exposure
of
cultured
cortical
neurons
to
EMF
led
to
an
increase
in
8-hydroxyguanine
in
neuronal
mitochondria,
a
common
biomarker
of
DNA
oxidative
damage,
along
with
a
reduction
in
the
copy
number
of
mitochondrial
DNA
and
the
levels
of
mitochondrial
RNA
transcripts;
but
these
effects
could
all
be
prevented
by
pretreatment
with
melatonin
[7].
In
a
study
of
skin
lesion
induced
by
exposure
to
cell
phone
radia-
tion,
the
skin
changes
in
the
irradiated
group
(which
included
thicker
stratum
corneum,
epidermal
atrophy,
papillamato-
sis,
basil
cell
proliferation,
increased
epidermal
granular
cell
layer
and
capillary
proliferation,
impaired
collagen
tissue
dis-
tribution
and
separation
of
collagen
bundles
in
dermis)
were
prevented
(except
for
hypergranulosis)
by
melatonin
treat-
ment
[28].
Melatonin
as
well
as
caffeic
acid
phenyethyl
ester
(an
antioxidant)
both
protected
against
retinal
oxidative
stress
in
rates
exposed
long-term
to
mobile
phone
irradiation
[29].
Nitric
oxide
(NO)
was
increased
in
nasal
and
sinus
mucosa
in
rats
after
EMF
exposure,
with
this
NO
possibly
acting
as
a
defense
mechanism
suggesting
tissue
damage;
but
this
was
prevented
by
pretreatment
with
melatonin
[30].
Melatonin
treatment
significantly
prevented
the
increase
in
the
MDA
(malondyaldehyde,
a
marker
of
lipid
peroxidation)
content
and
XO
(xanthine
oxidase)
activity
in
rat
brain
tissue
after
40
days
of
exposure,
but
it
was
unable
to
prevent
the
decrease
of
CAT
activity
and
increase
of
carbonyl
group
contents
[31].
Of
note,
the
melatonin
production
of
infants
in
isolettes
in
neonatal
intensive
care
units
appears
to
be
impacted
by
the
high
ELF-EMF
environment,
in
that
when
infants
were
removed
from
those
exposures
they
showed
an
increase
in
melatonin
levels
[32].
There
is
an
increased
prevalence
of
ASCs
in
children
who
were
born
prematurely
[33–43].
There
are
many
potential
prematurity-associated
factors
that
could
contribute
to
increased
risk
for
ASCs,
but
proper
melatonin
regulation
warrants
EMF/RFR
controls
in
the
newborns’
environment.
2.1.2.3.
Melatonin
and
autism.
Regarding
melatonin
status
in
people
with
ASCs,
a
recent
meta-analysis
summarized
the
current
findings
as
indicating
that
“(1)
Physiological
lev-
els
of
melatonin
and/or
melatonin
derivatives
are
commonly
below
average
in
ASC
and
correlate
with
autistic
behavior,
(2)
Abnormalities
in
melatonin-related
genes
may
be
a
cause
of
low
melatonin
levels
in
ASD,
and
(3)
.
.
.
treatment
with
melatonin
significantly
improves
sleep
duration
and
sleep
onset
latency
in
ASD.”
[44].
Please
cite
this
article
in
press
as:
M.R.
Herbert,
C.
Sage,
Autism
and
EMF?
Plausibility
of
a
pathophysiological
link
part
II,
Pathophysiology
(2013),
http://dx.doi.org/10.1016/j.pathophys.2013.08.002
ARTICLE IN PRESS
PATPHY-777;
No.
of
Pages
24
4
M.R.
Herbert,
C.
Sage
/
Pathophysiology
xxx
(2013)
xxx–xxx
The
meta-analysis
also
showed
that
polymorphisms
in
melatonin-related
genes
in
ASC
could
contribute
to
lower
melatonin
concentrations
or
an
altered
response
to
melatonin,
but
only
in
a
small
percentage
of
individuals,
since
pertinent
genes
were
found
in
only
a
small
minority
of
those
screened.
Based
on
the
common
presence
of
both
sleep
disorders
and
low
melatonin
levels,
Bourgeron
[45]
proposed
that
synap-
tic
and
clock
genes
are
important
in
ASCs,
and
that
future
studies
should
investigate
the
circadian
modulation
of
synap-
tic
function
[45].
A
number
of
melatonin-related
genetic
variants
have
been
identified
as
associated
with
ASCs.
Poly-
morphisms
and
deletions
in
the
ASMT
gene,
which
encodes
the
last
enzyme
of
melatonin
synthesis,
have
been
found
[46–48],
and
variations
have
been
found
as
well
for
mela-
tonin
receptor
genes
[46,47,49].
CYP1A2
polymorphisms
have
been
found
in
slow
melatonin
metabolisers,
in
whom
melatonin
levels
are
aberrant
and
initial
response
to
melatonin
for
sleep
disappeared
in
a
few
weeks
[50].
2.1.2.4.
Autism
AND
melatonin
AND
glutathione.
Whereas
PubMed
searches
for
“autism
AND
melatonin”
and
“autism
AND
glutathione”
each
coincidentally
yielded
72
citations,
and
“melatonin
AND
glutathione”
yielded
803
citations,
the
search
for
“autism
AND
melatonin
AND
glutathione”
yielded
zero
citations.
This
is
interesting
given
the
strong
connection
of
melatonin
and
glutathione
metabolically,
as
discussed
above,
alongside
of
the
strongly
established
inter-
est
in
both
glutathione
and
melatonin
in
ASC
research
and
increasingly
in
clinical
practice.
Hopefully
one
contribution
of
an
investigation
of
EMF/RFR
links
to
ASCs
will
be
to
help
bring
attention
to
this
relationship,
which
may
help
identify
potential
environmental
and
physiological
causes
for
low
melatonin
in
those
without
pertinent
mutations.
Of
perti-
nence,
tryptophan
hydroxylase
(TPH2)
–
the
rate
limiting
enzyme
in
the
synthesis
of
serotonin,
from
which
mela-
tonin
is
derived
–
is
extremely
vulnerable
to
oxidation,
and
tends
to
misfold
when
its
cysteine
residues
are
oxidized,
with
the
enzyme
being
converted
to
a
redox-cycling
quinoprotein
[51–54].
2.1.3.
Disturbed
immune
function
There
is
by
now
a
broad
appreciation
of
the
presence
of
immune
disturbances
in
ASCs,
to
the
point
where
there
is
an
emerging
discussion
of
ASCs
as
neuroimmune
disorders
[55,56].
Research
identifying
immune
features
in
ASCs
spans
from
genetics
where
risk
genes
have
been
identified
to
epi-
genetics
where
altered
expression
of
immune
genes
is
being
reported
as
prominent
in
ASC
epigenetics
[57–59],
and
also
includes
prenatal
infectious
and
immune
disturbances
as
risk
factors
for
autism
as
well
as
other
neurodevelopmental
and
neuropsychiatric
diseases
as
well
as
other
conditions
such
as
asthma
[60–62].
Immune
disturbances
in
infants
and
children
with
ASC
are
heterogeneous,
with
some
but
not
all
manifest-
ing
autoimmunity
[63,64].
Anecdotally,
recurrent
infection
is
common
while
on
the
other
hand
some
get
sick
less
often
than
their
peers.
It
is
common
for
people
with
autism
to
have
family
members
with
immune
or
autoimmune
diseases
[65].
The
immune
system
is
turning
out
to
have
an
impor-
tant
role
in
brain
development
[66–68].
As
mentioned,
glial
activation
associated
with
brain
immune
response
has
been
identified
in
a
growing
number
of
studies.
Whether
or
not
EMF/RFR
contributes
to
these
features
of
ASCs
causally,
based
on
the
evidence
below
regarding
immune
impacts
of
EMF/RFR
exposure
[69],
it
is
certainly
plausible
that
such
exposures
could
serve
as
aggravating
factors.
2.1.3.1.
Low-intensity
exposures.
The
body’s
immune
defense
system
is
now
known
to
respond
to
very
low-
intensity
exposures
[70].
Chronic
exposure
to
factors
that
increase
allergic
and
inflammatory
responses
on
a
continuing
basis
is
likely
to
be
harmful
to
health,
since
the
resultant
chronic
inflammatory
responses
can
lead
to
cellular,
tissue
and
organ
damage
over
time.
Many
chronic
diseases
are
related
to
chronic
immune
system
dysfunction.
Disturbance
of
the
immune
system
by
very
low-intensity
electromagnetic
field
exposure
is
discussed
as
a
potential
underlying
cause
for
cellular
damage
and
impaired
healing
(tissue
repair),
which
could
lead
to
disease
and
physiological
impairment
[71,72].
Both
human
and
animal
studies
report
that
exposures
to
EMF
and
RFR
at
environmental
levels
associated
with
new
technologies
can
be
associated
with
large
immunohistological
changes
in
mast
cells
as
well
as
other
measures
of
immune
dysfunction
and
dysregulation.
Mast
cells
not
only
can
degranulate
and
release
irritating
chemicals
leading
to
allergic
symptoms;
they
are
also
widely
distributed
in
the
body,
including
in
the
brain
and
the
heart,
which
might
relate
to
some
of
the
symptoms
commonly
reported
in
relation
to
EMF/RFR
exposure
(such
as
headache,
painful
light
sensitivity,
and
cardiac
rhythm
and
palpitation
problems).
2.1.3.2.
Consequences
of
immune
challenges
during
preg-
nancy.
As
mentioned,
infection
in
pregnancy
can
also
increase
the
risk
of
autism
and
other
neurodevelopmental
and
neuropsychiatric
disorders
via
maternal
immune
acti-
vation
(MIA).
Viral,
bacterial
and
parasitic
infections
during
pregnancy
are
thought
to
contribute
to
at
least
30%
of
cases
of
schizophrenia
[73].
The
connection
of
maternal
infec-
tion
to
autism
is
supported
epidemiologically,
including
in
a
Kaiser
study
where
risk
was
associated
with
psoriasis
and
with
asthma
and
allergy
in
the
second
trimester
[65],
and
in
a
large
study
of
autism
cases
in
the
Danish
Medical
registry
[74]
with
infection
at
any
point
in
pregnancy
yielding
an
adjusted
hazard
ration
of
1.14
(CI:
0.96
−
1.34)
and
when
infection
occurred
during
second
trimester
the
odds
ratio
was
2.98
(CI:
1.29
−
7.15).
In
animal
models,
while
there
is
much
variation
in
study
design,
mediators
of
the
immune
impact
include
oxidative
stress,
interleukin-6
and
increased
placental
cytokines
[61,68,75].
Garbett
et
al.
[76]
commented
on
several
mouse
models
of
the
effects
of
MIA
on
the
fetal
brain
that
“The
overall
gene
expression
changes
suggest
that
the
response
to
MIA
is
a
neuroprotective
attempt
by
the
Please
cite
this
article
in
press
as:
M.R.
Herbert,
C.
Sage,
Autism
and
EMF?
Plausibility
of
a
pathophysiological
link
part
II,
Pathophysiology
(2013),
http://dx.doi.org/10.1016/j.pathophys.2013.08.002
ARTICLE IN PRESS
PATPHY-777;
No.
of
Pages
24
M.R.
Herbert,
C.
Sage
/
Pathophysiology
xxx
(2013)
xxx–xxx
5
developing
brain
to
counteract
environmental
stress,
but
at
a
cost
of
disrupting
typical
neuronal
differentiation
and
axonal
growth.”
[76].
Maternal
fetal
brain-reactive
autoantibodies
have
also
been
identified
in
some
cases
[62,77–82].
Although
we
have
evidence
of
immune
impacts
of
EMF/RFR,
the
impact
of
repeated
or
chronic
exposure
to
EMF
and
RFR
during
pregnancy
is
poorly
studied;
could
this
trigger
similar
immune
responses
(cytokine
production)
and
stress
protein
responses,
which
in
turn
would
have
effects
on
the
fetus?
Although
this
has
been
poorly
studied,
we
do
have
data
that
very
low
cell
phone
radiation
exposures
during
both
human
and
mouse
pregnancies
have
resulted
in
altered
fetal
brain
development
leading
to
memory,
learning,
and
attention
problems
and
behavioral
problems
[83].
2.1.3.3.
Potential
immune
contributions
to
reactivity
and
variability
in
ASCs.
Immune
changes
in
ASCs
appear
to
be
associated
with
behavioral
change
[84–88],
but
the
mecha-
nisms
are
complex
and
to
date
poorly
understood
[89]
and
likely
will
need
to
be
elucidated
through
systems
biology
methods
that
capture
multisystem
influences
on
the
inter-
actions
across
behavior,
brain
and
immune
regulation
[90],
including
electrophysiology.
Two
of
the
particularly
difficult
parts
of
ASCs
are
the
intense
reactivity
and
the
variability
in
assorted
symptoms
such
as
tantrums
and
other
difficult
behaviors.
Children
with
ASCs
who
also
have
gastrointestinal
symptoms
and
marked
fluctuation
of
behavioral
symptoms
have
been
shown
to
exhibit
distinct
innate
immune
abnormalities
and
tran-
scriptional
profiles
of
peripheral
blood
monocytes
[91].
It
is
worth
considering
EMF/RFR
exposures
could
be
operat-
ing
through
related
mechanisms
so
as
to
add
to
‘allostatic
loading’
in
ways
that
exacerbate
behavior.
In
Johansson
2006
and
2007
a
foundation
is
provided
for
understanding
how
chronic
EMF/RFR
exposure
can
compromise
immune
function
and
sensitize
a
person
to
even
small
exposures
in
the
future
[72,92].
Johansson
discusses
alterations
of
immune
function
at
environmental
levels
resulting
in
loss
of
memory
and
concentration,
skin
redness
and
inflammation,
eczema,
headache,
and
fatigue.
Mast
cells
that
degranulate
under
EMF
and
RFR
exposures
and
substances
secreted
by
them
(histamine,
heparin
and
serotonin)
may
contribute
to
features
of
this
sensitivity
to
electromagnetic
fields
[92].
Theoharides
and
colleagues
have
argued
that
environmental
and
stress
related
triggers
might
activate
mast
cells,
causing
inflammatory
compromise
and
leading
to
gut–blood–brain
barrier
compromise,
seizures
and
other
ASC
ASC
symptoms
[93,94],
and
that
this
cascade
of
immune
response
and
its
consequences
might
also
be
triggered
in
the
absence
of
infec-
tion
by
mitochondrial
fragments
that
can
be
released
from
cells
in
response
to
stimulation
by
IgE/anti-IgE
or
by
the
proinflammatory
peptide
substance
P
[95].
Seitz
et
al.
[96]
reviewed
an
extensive
literature
on
elec-
tromagnetic
hypersensitivity
conditions
reported
to
include
sleep
quality,
dizziness,
headache,
skin
rashes,
memory
and
concentration
impairments
related
to
EMF
and
RFR
[96].
Some
of
these
symptoms
are
common
in
ASCs,
whether
or
not
they
are
due
to
EMF/RFR
exposure,
and
the
experience
of
discomfort
may
be
hard
to
document
due
to
difficulties
with
self-reporting
in
many
people
with
ASCs.
Johansson
[72]
also
reports
that
benchmark
indicators
of
immune
system
allergic
and
inflammatory
reactions
occur
under
exposure
conditions
of
low-intensity
non-ionizing
radiation
(immune
cell
alterations,
mast
cell
degranulation
histamine-positive
mast
cells
in
biopsies
and
immunoreac-
tive
dendritic
immune
cells)
[71,72].
In
facial
skin
samples
of
electro-hypersensitive
persons,
the
most
common
finding
is
a
profound
increase
in
mast
cells
as
monitored
by
various
mast
cell
markers,
such
as
histamine,
chymase
and
tryptase
[97].
In
ASCs,
infant
and
childhood
rashes,
eczema
and
pso-
riasis
are
common,
and
they
are
common
in
family
members
as
well
[98].
2.1.4.
Alteration
of
and
damage
to
cells
in
the
brain
Brain
cells
have
a
variety
of
ways
of
reacting
to
envi-
ronmental
stressors,
such
as
shape
changes,
metabolic
alterations,
upregulation
or
downregulation
of
neurotrans-
mitters
and
receptors,
other
altered
functionality,
structural
damage,
production
of
un-metabolizable
misfolded
proteins
and
other
cellular
debris,
and
apoptosis;
these
range
along
a
spectrum
from
adaptation
to
damage
and
cell
death.
These
types
of
alterations
can
be
looked
at
in
animals
under
controlled
conditions,
but
in
human
beings
direct
cellular
examination
can
only
be
done
on
surgical
biopsy
tissue
–
which
is
hardly
ever
available
in
people
with
ASCs
–
or
after
death,
at
which
point
there
has
been
a
whole
lifetime
of
exposures
that
are
generally
impossible
to
tease
apart
if
there
were
even
motivation
to
do
so.
This
complicates
the
comparison
of
brain
cell
and
tissue-related
pathophysiology
between
what
is
seen
in
ASCs
and
what
is
associated
with
EMF/RFR
exposures.
2.1.4.1.
Brain
cells.
Impact
of
EMF/RFR
on
cells
in
the
brain
has
been
documented
by
some
of
the
studies
that
have
examined
brain
tissue
after
exposure,
although
the
interpre-
tation
of
inconsistencies
across
studies
is
complicated
by
sometimes
major
differences
in
impact
attributable
to
differ-
ences
in
frequencies
and
duration
of
exposure,
as
well
as
to
differences
in
resonance
properties
of
tissues
and
other
poorly
understood
constraints
on
cellular
response.
These
studies
and
methodological
considerations
have
been
reviewed
in
depth
in
several
sections
of
the
2012
BioInitiative
Report
[11,99].
A
few
examples
of
observations
after
exposure
have
included
dark
neurons
(an
indicator
of
neuronal
damage),
as
well
as
alteration
of
neuronal
firing
rate
[100],
and
upregula-
tion
of
genes
related
to
cell
death
pathways
in
both
neurons
and
astrocytes
[101].
Astrocytic
changes
included
increased
GFAP
and
increased
glial
reactivity
[102–105],
as
well
as
astrocyte-pertinent
protein
expression
changes
detected
by
Fragopoulou
et
al.
[322]
as
mentioned
above.
Also
observed
has
been
a
marked
protein
downregulation
of
the
nerve
growth
factor
glial
maturation
factor
beta
(GMF)
which
is
Please
cite
this
article
in
press
as:
M.R.
Herbert,
C.
Sage,
Autism
and
EMF?
Plausibility
of
a
pathophysiological
link
part
II,
Pathophysiology
(2013),
http://dx.doi.org/10.1016/j.pathophys.2013.08.002
ARTICLE IN PRESS
PATPHY-777;
No.
of
Pages
24
6
M.R.
Herbert,
C.
Sage
/
Pathophysiology
xxx
(2013)
xxx–xxx
considered
as
an
intracellular
signal
transduction
regulator
in
astrocytes,
which
could
have
significant
impact
on
neuronal-
glial
interactions
as
well
as
brain
cell
differentiation
and
tumor
development.
Diminution
of
Purkinje
cell
number
and
density
has
also
been
observed,
[106]
including
in
two
stud-
ies
of
the
impacts
of
perinatal
exposure
[107,108].
Promotion
of
pro-inflammatory
responses
in
EMF-stimulated
microglial
cells
has
also
been
documented
[109].
Neuropathology
findings
in
ASCs
have
been
varied
and
have
been
interpreted
according
to
various
frameworks
ranging
from
a
regionalized
approach
oriented
to
iden-
tifying
potential
brain
relationships
to
ASC’s
behavioral
features
[110]
to
identifying
receptor,
neurotransmitter
and
interneuron
abnormalities
that
could
account
for
an
increased
excitation/inhibition
ratio
[111–115].
Studies
have
docu-
mented
a
range
of
abnormalities
in
neurons,
including
altered
cellular
packing
in
the
limbic
system,
reduced
den-
dritic
arborization,
and
reductions
in
limbic
GABAergic
systems
[116].
Over
the
past
decade
a
shift
has
occurred
from
presuming
that
all
pertinent
brain
changes
occurred
prior
to
birth,
to
an
acknowledgement
that
ongoing
cellu-
lar
processes
appear
to
be
occurring
not
only
after
birth
but
well
into
adulthood
[117].
One
of
the
reasons
for
this
shift
was
the
observation
that
head
size
(as
well
as
brain
weight
and
size)
was
on
average
larger
in
children
with
autism,
and
the
head
sizes
of
children
who
became
diagnosed
with
autism
increased
in
percentile
after
birth
[118].
2.1.4.2.
Neuroinflammation,
glial
activation
and
excitotox-
icity.
Although
much
attention
has
been
paid
in
ASC
brain
literature
to
specific
regions
manifesting
differences
in
size
and
activity
in
comparison
to
those
without
ASCs,
there
are
other
observations
that
are
not
strictly
regional
in
nature,
such
as
more
widely
distributed
scaling
differences
(e.g.
larger
brains,
wider
brains,
increased
white
matter
volume,
along
with
altered
functional
connectivity
and
coherence
to
be
discussed
below).
Recently
more
studies
have
appeared
identifying
pathophysiological
abnormalities
such
as
neu-
roinflammation,
mitochondrial
dysfunction
and
glutathione
depletion
in
brain
tissue.
Neuroinflammation
was
first
identi-
fied
in
a
study
of
postmortem
samples
from
eleven
individuals
aged
5–44
who
had
died
carrying
an
ASC
diagnosis,
in
which
activated
astrocytes
and
microglial
cells
as
well
as
abnormal
cytokines
and
chemokines
were
found.
Other
research
has
identified
further
astrocyte
abnormalities
such
as
altered
expression
of
astrocyte
markers
GFAP
abnormali-
ties,
with
elevation,
antibodies,
and
altered
signaling
having
been
documented
[119–121].
Increased
microglia
activation
and
density
as
well
as
increased
myeloid
dendritic
cell
fre-
quencies
have
also
been
documented
[87,122,123],
as
has
abnormal
microglial-neuronal
interactions
[124].
Recently,
through
use
of
the
PET
ligand
PK11105,
microglial
activa-
tion
was
found
to
be
significantly
higher
in
multiple
brain
regions
in
young
adults
with
ASCs
[125].
Genes
associated
with
glial
activation
have
been
documented
as
upregulated.
Garbett
et
al
measured
increased
transcript
levels
of
many
immune
genes,
as
well
as
changes
in
transcripts
related
to
cell
communication,
differentiation,
cell
cycle
regulation
and
chaperone
systems
[126].
Voineaugu
and
colleagues
per-
formed
transcriptomic
analysis
of
autistic
brain
and
found
a
neuronal
module
of
co-expressed
genes
which
was
enriched
with
genetically
associated
variants;
an
immune-glial
module
which
showed
no
such
enrichment
for
autism
GWAS
signals
was
interpreted
as
secondary
[127],
but
this
seems
to
involve
circular
thinking,
since
it
implies
that
the
primary
cause
must
be
genetic,
which
is
an
assumption
deriving
from
a
dominant
model,
but
is
not
a
proven
fact.
Neuroinflammation
also
does
not
appear
to
be
strictly
localized
in
a
function-specific
fashion,
and
it
may
contribute
both
to
more
broadly
distributed
and
more
focal
features
for
tissue-based
reasons.
It
may
be
that
brain
regions
with
particular
prominence
in
ASCs
may
have
distinctive
cellu-
lar
characteristics—e.g.
the
amygdala
[128–138],
which
may
have
a
larger
or
more
reactive
population
of
astrocytes
[139]
or
the
basal
ganglia
which
may
have
greater
sensitivity
to
even
subtle
hypoxia
or
perfusion
abnormalities.
In
this
case
it
may
be
the
histology
of
these
areas
that
makes
them
vul-
nerable
to
environmental
irritants,
and
this
may
contribute
to
how
environmental
factors
such
as
EMF/RFR
might
trigger
or
aggravate
some
of
ASC’s
features.
More
widely
distributed
brain
tissue
pathology
be
part
of
what
leads
to
differences
in
ASCs
in
brain
connectivity.
However
these
types
of
tissue-
function
relationships
have
been
poorly
investigated.
Belyaev
has
intensively
reviewed
physical
considerations
including
the
contribution
of
tissue
differences
to
variability
in
mea-
sured
EMF/RFR
impacts
[11].
Various
signs
of
mitochondrial
dysfunction
and
oxida-
tive
stress
have
also
been
identified
in
the
brain.
Findings
include
downregulation
of
expression
of
mitochondrial
elec-
tron
transport
genes
[140]
or
deficit
of
mitochondrial
electron
transport
chain
complexes
[141],
brain
region
specific
glu-
tathione
redox
imbalance
[142],
and
evidence
of
oxidative
damage
and
inflammation
associated
with
low
glutathione
redox
status
[143].
Oxidative
stress
markers
were
measured
as
increased
in
cerebellum
[144].
Additional
support
for
the
presence
of
tissue
pathophysiology-based
changes
in
brains
of
people
with
ASCs
comes
from
the
various
studies
documenting
reduction
in
Purkinje
cell
numbers
[117,145–150],
possibly
due
to
oxidative
stress
and
an
increased
excitation/inhibition
ratio
that
could
potentially
be
acquired
[150].
Also
of
note
are
changes
in
the
glutamatergic
and
GABAergic
systems,
which
when
imbalanced
can
disturb
the
excitation/inhibition
ratio
and
contribute
to
seizure
disorders;
reductions
in
GABA
receptors
as
well
as
in
GAD
65
and
67
proteins
that
catalyse
the
conversion
of
glutamate
into
GABA
have
been
measured
[151–153].
A
consensus
statement
on
the
cerebellum
in
ASCs
stated
that,
“Points
of
consensus
include
presence
of
abnor-
mal
cerebellar
anatomy,
abnormal
neurotransmitter
systems,
oxidative
stress,
cerebellar
motor
and
cognitive
deficits,
and
neuroinflammation
in
subjects
with
autism”
[150].
Please
cite
this
article
in
press
as:
M.R.
Herbert,
C.
Sage,
Autism
and
EMF?
Plausibility
of
a
pathophysiological
link
part
II,
Pathophysiology
(2013),
http://dx.doi.org/10.1016/j.pathophys.2013.08.002
ARTICLE IN PRESS
PATPHY-777;
No.
of
Pages
24
M.R.
Herbert,
C.
Sage
/
Pathophysiology
xxx
(2013)
xxx–xxx
7
Some
indirect
corroboration
for
these
findings
has
come
from
neuroimaging,
where
the
initial
hypothesis
regarding
the
tissue
basis
of
the
larger
size
of
brains
in
so
many
people
with
autism
–
that
it
was
due
to
a
higher
density
of
neurons
and
more
tightly
packed
axons
–
came
under
question
with
the
emergence
of
contradictory
findings,
well
reviewed
a
few
years
ago
by
Dager
and
colleagues
[154].
These
include
reduced
rather
than
increased
density
of
NAA
(n-acetylaspartate,
a
marker
of
neuronal
integrity
and
den-
sity
that
is
produced
in
the
mitochondria),
reduced
rather
than
increased
fractional
anisotropy
suggesting
less
tightly
packed
axonal
bundles
[155–161]
and
greater
rather
than
lower
dif-
fusivity,
all
of
which
may
be
more
consistent
with
lower
density
of
tissue
and
tissue
metabolites
and
more
fluid,
which
could
be
consistent
with
neuroinflammation
and/or
oxidative
stress.
The
early
postnatal
development
of
such
lower
frac-
tional
anisotropy
and
increased
diffusivity
was
measured
in
the
process
of
occurring
recently,
in
the
first
large
prospec-
tive
longitudinal
imaging
study
of
infants,
who
trended
from
6
months
to
2
years
in
the
direction
of
these
findings
becom-
ing
more
pronounced—but
still
with
substantial
overlap
with
those
infants
who
did
not
develop
autism
[160].
This
trend
was
consistent
with
prior
studies
showing
increase
in
head
size
after
birth,
and
added
some
information
about
what
was
happening
in
the
brain
to
drive
this
size
increase,
although
due
to
its
methods
it
could
only
indirectly
address
the
pos-
sibility
that
emergence
during
the
first
few
years
of
life
of
tissue
pathophysiology
disturbances
such
as
neuroinflamma-
tion
might
be
contributing
to
these
trends
[162].
There
is
also
substantial
variability
across
many
differ-
ent
types
of
brain
findings.
Of
interest
is
that
a
number
of
functional
brain
imaging
and
electrophysiology
studies
have
identified
greater
heterogeneity
in
response
to
stimuli
between
individuals
in
the
ASC
group
than
individuals
in
the
neurotypical
control
group
[163,164].
This
may
make
more
sense
from
the
point
of
view
of
non-linear
response—i.e.
a
disproportionality
between
output
and
input
(as
well
as
state
and
context
sensitivity),
in
a
pathophysiologically
per-
turbed
brain
system.
Nonlinearity
has
also
been
a
significant
methodological
issue
in
EMF/RFR
research
because
linear
methods
of
study
design
and
data
analysis
have
often
been
insensitive
to
effects,
whereas
nonlinear
methods
have
been
argued
to
show
greater
sensitivity
[165–175].
It
is
important
to
entertain
how
environmental
agents
could
contribute
individually
and
synergistically
to
brain
changes
in
ASCs,
how
different
exposures
may
disturb
phys-
iology
similarly
or
differently,
and
how
these
changes
may
develop
over
progress
over
time
after
the
earliest
periods
in
brain
development.
EMF/RFR
exposures
could
be
pre-
conceptional,
prenatal
or
postnatal—or
all
of
the
above;
it
is
conceivable
that
this
could
be
the
case
in
ASCs
as
well.
2.1.4.3.
Altered
development.
There
is
some
evidence
for
altered
brain
and
organism
development
in
relation
to
EMF/RFR
exposure.
Aldad
et
al.
[83]
exposed
mice
in-utero
to
cellular
telephones,
with
resultant
aberrant
miniature
exci-
tatory
postsynaptic
currents,
and
dose-responsive
impaired
glutamatergic
synaptic
transmission
onto
layer
V
pyrami-
dal
neurons
of
the
prefrontal
cortex
[83].
Lahijani
exposed
preincubated
chicken
embryos
to
50
Hz
EMFs,
and
made
the
following
morphological
observations:
“exencephalic
embryos,
embryos
with
asymmetrical
faces,
crossed
beak,
shorter
upper
beak,
deformed
hind
limbs,
gastroschesis,
anophthalmia,
and
microphthalmia.
H&E
and
reticulin
stain-
ings,
TEMS,
and
SEMs
studies
indicated
EMFs
would
create
hepatocytes
with
fibrotic
bands,
severe
steatohepatitis,
vacuolizations,
swollen
and
extremely
electron-dense
mito-
chondria,
reduced
invisible
cristae,
crystalized
mitochondria
with
degenerated
cristae,
myelin-like
figures,
macrophages
engulfing
adjacent
cells,
dentated
nuclei,
nuclei
with
irreg-
ular
envelopes,
degenerated
hepatocytes,
abnormal
lipid
accumulations,
lipid
droplets
pushing
hepatocytes’
nuclei
to
the
corner
of
the
cells,
abundant
cellular
infiltrations
cel-
lular
infiltrations
inside
sinusoid
and
around
central
veins,
disrupted
reticulin
plexus,
and
release
of
chromatin
into
cyto-
sol,
with
partially
regular
water
layers,
and
attributed
cell
damage
to
elevated
free
radical
induced
cell
membrane
dis-
ruptions”
[5].
Although
it
is
of
great
interest
to
characterize
the
changes
in
development
associated
with
ASCs,
it
is
also
difficult
to
do
in
human
beings
because
at
present
diagnosis
is
not
possible
until
at
least
2–3
years
after
birth.
By
now
there
have
been
a
lot
of
prospective
studies
of
infants
at
high
risk
for
autism,
but
the
in
vivo
brain
imaging
and
electrophysiology
data
from
these
studies
is
only
starting
to
be