2. 2
Table of Contents
06 A compound from beer may help fat loss
Bitter, hop-derived compounds found in beer may actually reduce body fat
levels. Previously only shown in mice, this study tested the theory in humans.
13 Sugar is the ultimate antioxidant and insulin will make you
younger: Appreciating a few poorly recognized but critical
contributions of carbohydrate
By Chris Masterjohn, PhD
Sugar is widely demonized in the media and medical establishment.
Professor Masterjohn provides an eye-opening and detailed view on some
potential protective roles of glucose.
18 Milk gone bad: A1 beta-casein and GI distress
Casein isn’t just the slowly digesting protein that helps prevent muscle
breakdown. This study looked at possible negative effects of the most
common type of casein in milk.
25 Arsenic in rice: big trouble for little infants?
Depending on where it’s grown, rice can have rather high levels of arsenic.
Especially brown rice. This may be important for developing infants.
32 How much protein does grandpa really need?
One of the many downsides to aging is altered protein mechanics. Based on
the theory that protein requirements for seniors may be pegged too low, this
study quantified protein needs in older males.
40 Training hot for performance gains
Athletes know all too well that sudden exposure to heat or altitude can
severely impact performance, so acclimation is a good idea. And it turns out
that exposure to one of these stressors may actually help the other one.
3. 3
Table of Contents
48 The art & science of evidence-based practice and elite performance
By Craig Pickering
As one of the rare athletes to participate in both the Summer and Winter
Olympic Games, Craig has a unique perspective on the intersection of
optimal performance and evidence-based practice.
53 Relaxing arteries with magnesium
To stave off cardiovascular disease, we want our arteries to be more pliable
than stiff. This trial tested six months of magnesium supplementation for the
purpose of reducing arterial stiffness.
60 Beating “the burn” with baking soda
Can you believe that something as simple as baking soda may boost
performance? While this fact has been known for a while, researchers didn’t
know that people’s responses to different doses can vary quite a bit.
68 Is resistance exercise the next frontier for nitrates?
Nitrate use for athletics has exploded in the past few years, but research
typically focuses on aerobic activities like longer-distance cycling or
swimming. Could nitrates also show benefit for weightlifting?
4. 4
From the Editor
What if you could see what’s going on inside your body?
What if you could see what food and supplements did
to your cells, see what stress did to your brain?
That would be the biggest breakthrough since combin-
ing peanut butter and chocolate. Although now I’m
wondering what that peanut butter and chocolate is
doing to my cells.
Speaking of this sweet and salty combination, let’s take
salt as an example of what’s going on inside our bodies
that we may be clueless about. Starting with the 1979
Surgeon General’s report, which clearly labeled salt as
a cause of high blood pressure, we became a salt-wary
country. I remember watching The Cosby Show as a kid,
salivating over the enormous cold cut sandwiches that
Cliff Huxtable tried to hide from Claire Huxtable. This
is natural, as humans have a strong innate salt hunger.
Fast forward thirty years, and many researchers have
now switched stances. Studies suggest that low-salt
guidelines may have been misguided, and very low salt
intakes are actually harmful. But aside from random-
ized trial results, what is the sodium from salt actually
DOING in our bodies?
As always, the annoying answer is “it depends”. Usually
our kidneys do a bang-up job of eliminating sodium we
don’t need. Many people reading this exercise a lot, and
those people will sweat out a decent amount of sodium.
So those people might want to avoid low intakes.
But we know surprisingly little about the possible
health effects of salt. Only in the past couple years did
evidence emerge linking higher salt intakes to head-
aches. But (relatively) higher salt intakes could also
theoretically be protective against bacterial infection,
although the evidence is limited to animals right now.
And it turns out that salt might not impact blood pres-
sure just through fluid balance, it may actually increase
adrenaline levels.
That’s a mishmash of seemingly unrelated and occasion-
ally theoretical health effects. And that’s for just one,
lonesome ingredient. Aside from salt, there’s debate
about much more complex things we eat, like red meat
or low-carb diets. Given the complexity of effects from
a lone element like sodium, nobody should pretend to
know with certainty what the health effects of foods
and diets are, unless they can shrink down to the size of
a molecule and zip around inside a human body.
On the flip side, it might be good to imagine that sce-
nario on occasion. If you consistently eat junk food,
imagine zipping through your body and seeing fat slow-
ly accumulate in your liver, neurochemicals shooting
off and desensitizing your brain, things generally going
awry over time. If you mostly eat healthy and enjoy the
occasional indulgence, these are likely to be blips rather
than sustained physiological changes.
So imagining what goes on inside the body isn’t just a
cool thought experiment, it’s also a potentially helpful
heuristic, as well as a reminder that we generally don’t
know what the hell is going on after we chew and swal-
low. But it sure is fun to follow along as researchers try
to piece things together.
Kamal Patel, Editor-in-Chief
5. 5
Contributors
Researchers
Margaret Wertheim
M.S., RD
Alex Leaf
M.S(c)
Courtney Silverthorn
Ph.D.
Zach Bohannan
M.S.
Anders Nedergaard
Ph.D.
Jeff Rothschild
M.Sc., RD
Greg Palcziewski
Ph.D. (c)
James Graham
Ph.D.
Gregory Lopez
Pharm.D.
Pablo Sanchez Soria
Ph.D.
Kamal Patel
M.B.A., M.P.H.,
Ph.D(c)
Editors
Arya Sharma
Ph.D., M.D.
Natalie Muth
M.D., M.P.H., RD
Stephan Guyenet
Ph.D.
Sarah Ballantyne
Ph.D.
Katherine Rizzone
M.D.
Spencer Nadolsky
D.O.
Mark Kern
Ph.D., RD
Gillian Mandich
Ph.D(c)
Adel Moussa
Ph.D(c)
Reviewers
6. 6
A compound from
beer may help fat loss
Matured hop extract reduces body
fat in healthy overweight humans: a
randomized double-blind, placebo-
controlled parallel group study
7. 7
Introduction
Obesity is an increasingly global problem that is asso-
ciated with a greater risk of developing disorders like
hypertension and diabetes. Although dieting is an effec-
tive strategy, many people find it difficult to maintain
and look for easier alternatives.
A popular alternative approach to effective weight
management is supplementing therapeutic products
that offer ‘fat-burning’ properties. These includes nat-
ural products and ‘functional foods’ that are claimed
to suppress energy intake or actively increase energy
expenditure. There are many commercial products
that supposedly assist in effective weight management,
including compounds like conjugated linoleic acid
and pyruvate, as well as natural food products such as
Irvingia gabonensis and chia seed. However, most stud-
ies on these products have been inconclusive (such as
for Irvingia gabonensis) or shown that these dietary sup-
plements do not assist weight loss (such as chia seed).
On a positive note, there is promising data on the
anti-obesity effects of compounds called isohumu-
lones, or iso-α-acids. These compounds are the major
bitter components in beer and come from the female
hop plant (Humulus lupulus L.). As shown in Figure 1,
iso-α-acids are converted from α-acids during brewing,
and impart flavour and bitterness to beer. These iso-α-
acids have been shown to help obese individuals with
pre-diabetes by reducing hyperglycemia and body fat
content. In addition, iso-α-acids have also been shown
to prevent diet-induced obesity in two different strains
of mice. However, one drawback of using iso-α-acids is
their very strong bitter profile, which makes them quite
unpalatable at the concentrations required to be effec-
tive. Although an isohumulone pill would bypass these
palatability issues, for reasons unknown, it has not been
widely considered.
When beer is stored for long periods of time, there is
a progressive breakdown of the iso-α-acids into more
complex bitter compounds—known as matured hop
bitter acids (MHBA). The MHBA compounds consist
of oxidised derivatives that have similar structures
to iso-α-acids but are less bitter and therefore offer a
more palatable therapeutic agent. Recently, it has been
shown that MHBA reduces body fat in rodents at least
in part by increasing thermogenesis in brown adipose
tissue. Brown adipose tissue is abundant in rodents and
is important for their adaptation to cold environments.
Adult humans have also been shown to have meta-
bolically active brown adipose tissue, so this may be a
possible target for anti-obesity therapies in humans.
Figure 1: Degradation of a prevalent α-acid in hops
into nutritionally promising compounds
Mildly bitter α-acid
Humulone
Highly bitter iso-α-acids
Isohumulones Tricyclooxyisohumulones A
Less bitter MHBA
Further oxidation
during storage
Isomerization
during brewing
Figure 1: Degradation of a prevalent α-acid in hops into
nutritionally promising compounds
8. 8
The predominant source of bitterness in beer is from
the α-acid compounds present in hops. These com-
pounds break down into iso-α-acids during brewing
and in isolation may provide benefits that reduce
body fat in animals and humans.
Who and what was studied?
This study investigated the potential for MHBA to
reduce body fat in healthy overweight humans. Over
the course of 12 weeks, participants were given a
matured hop extract (MHE) that contains 18.3%
MHBA with no detectable amounts of α-acids or iso-α-
acids. The MHE was in the form of a test beverage that
was consumed once per day and then compared against
a placebo beverage of similar taste and appearance.
Researchers recruited Japanese males and females age
20-65 with a BMI of 25-30. This is classified as obesity
level 1 in Japan but is classified as overweight by the
World Health Organization (WHO). The participants
were randomly divided into two groups: the active
group and the placebo group. The active group was
to consume MHE once per day for a 12-week testing
period while the placebo group consumed a placebo
beverage that was of similar taste and appearance.
All participants had to adhere to strict criteria that
included the exclusion of any diets or dietary supple-
ments, medication that affected fat or lipid metabolism,
and excessive alcohol consumption or foods enriched
in hops. They were excluded if they had any current
metabolic disorders or other serious diseases, such as
diabetes or heart disease. Participants were interviewed
regarding their lifestyle to ascertain eligibility and
compliance. They recorded daily calorie intake, physi-
cal activity and subjective symptoms. By the end of the
test, evaluation of non-compliance resulted in the final
numbers of 91 participants in the active group and 87
participants in the placebo group.
Hop pellets were heated to 60 degrees for 120 hours
to oxidize the α-acids into iso-α-acids. Oxidized hop
pellets were soaked in water at 50 degrees for one hour
to extract these iso-α-acids before concentration. The
liquid was then heated to 90 degrees for four hours
to degrade the iso-α-acids into MHBA. The result-
ing matured hop extract contains 18.3% MHBA with
no detectable α-acids or iso-α-acids, as judged by
chromatographic analysis. Test beverages were 350 mil-
lilitres in volume and contained 35 milligrams MHBA.
Anthropometric parameters such as height, weight,
waist, and hip circumferences were measured, as well
as circulatory parameters such as blood pressure
and pulse rate. Body fat ratio was measured through
bioelectrical impedance analysis, while visceral, sub-
cutaneous and total fat were measured through CT
scanning. Blood chemistry and urinalysis were done to
evaluate possible adverse effects of consuming MHE.
This study was a randomized double-blind place-
bo-controlled analysis of MHE consumption in
healthy individuals who were classed as Japanese
obese level one (or overweight according to the
WHO). The hypothesis of this study was that MHE
ingestion would reduce abdominal fat, BMI, waist
circumferences and hip circumferences. Toxicology
analysis included parameters such as blood chemis-
try, haematology and urinalysis to determine if MHE
consumption had any adverse effects.
What were the findings?
The main study results are shown in Figure 2. Healthy
obese individuals who consumed MHE had a signifi-
cant reduction of five square centimetres of visceral fat
and nine square centimetres of total fat area in abdom-
inal areas after 12 weeks when compared to the placebo
group. The reduction in visceral and total fat areas was
approximately twofold lower in the active group than in
the placebo group.
9. 9
BMI and bodyweight was also significantly lower in
the active group than in the placebo group, though the
change from baseline was small. There was approxi-
mately a 0.5 kilogram bodyweight change in the active
group compared to no bodyweight change in the pla-
cebo group after 12 weeks of MHE consumption. Waist
and hip circumferences were also significantly lower
in the active group when compared to baseline, with
approximately one centimetre and 0.7 centimetres
lost, respectively. However, the placebo group lost
approximately 0.5 centimetres from the waist and hip
circumferences and so there was no significant differ-
ences between the two groups.
Safety endpoint analyses showed that there was no
significant variation in blood pressure throughout the
entire study and, except for the four-week pulse rate
measurement in the active group being higher, cir-
culatory parameters did not change from baseline
throughout the study. Blood chemistry and urinalysis
showed that there were no continuously significant
differences between the two groups or consistently
abnormal variations from baseline. All values recorded
were within normal physiological reference ranges.
The researchers also assessed subjective and adverse
effects during the study to determine the safety of MHE
consumption. The active group had 25 cases of cold-
like symptoms during the trial, but the placebo group
also reported 20 cases of cold-like symptoms, so it is
unlikely that MHE causes cold-like symptoms. Other
reports such as stomach ache, diarrhea, heartburn,
nausea, and vomiting were also reported to study inves-
tigators. In total, there were 14 cases in the active group
and 17 cases in the placebo group, suggesting that
there were probably no adverse side effects to digestive
parameters that could be attributed to continuously
ingesting MHE.
Figure 2: Change in measured body indices at 12 weeks
-2
-4
-8
-12
-16
-20
-6
-10
-14
-18
0
Subcutaneous fat area (cm2
)
-8cm2
-6cm2
-2
-4
-8
-12
-16
-20
-6
-10
-14
-18
0
Active Group
Total fat area (cm2
)
Placebo Group
-18cm2
-9cm2
0
-2
-4
-8
-12
-16
-20
Visceral fat area (cm2
)
-6
-10
-14
-18
-9cm2
-4cm2
0
-0.10
-0.20
-0.40
-0.60
-0.80
-0.100
Hip circumference (cm)
-0.30
-0.50
-0.70
-0.90
-0.69cm
-0.46cm
0
-0.10
-0.20
-0.40
-0.60
-0.80
-0.100
Waist circumference (cm)
-0.30
-0.50
-0.70
-0.90 -0.97cm
-0.58cm
0
-0.05
-0.1
-0.2
-0.3
-0.4
-0.5
Body weight (kg)
-0.15
-0.25
-0.35
-0.45 -0.47kg
-0.02kg
Figure 2: Change in measured body indices at 12 weeks
10. 10
This study investigated the effect of consuming
matured hop extract for 12 weeks and wheth-
er it would reduce body fat in otherwise healthy
overweight participants. After 12 weeks of MHE
treatment, visceral fat area was significantly reduced
when compared to the placebo group.
What does the study really
tell us?
This study shows that continual consumption of MHE
causes a significant reduction in body fat in healthy
overweight individuals without lifestyle changes like
increased physical activity or a reduction in calories
consumed. The effect of the MHE on body fat reduction
is likely due to the MHBA content of these beverages.
However, it is noteworthy that the placebo group also
lost significant body fat—although not as much as
the active group. It is possible that the placebo effect
has unconsciously affected the energy balance of the
participants, as has been observed in other studies on
anti-obesity agents, like lactoferrin and Pueraria flower.
Alternatively, seasonal variation may also affect weight,
which have contributed to weight changes observed in
the placebo group.
Although the test beverage contained bitter compounds,
the MHBA concentration was low enough to not give
the test beverage a bitter taste. However, the body has
nutrient-sensing receptors in the gut, which have the
ability to sense the luminal content of the stomach. This
gastric sense allows the gut to initiate an appropriate
response depending on the nutrients or toxins present.
The gut’s bitter-sensing receptors modulate the secretion
of the hunger hormone, ghrelin, when bitter compounds
are ingested. It was shown that the release of ghrelin
stimulates the appetite in the short-term but actually
causes a long-term reduction in food intake, thus effec-
tively reducing overall energy intake. Bitter compounds
were also shown to modulate satiety by altering intestinal
motility—delaying gastric emptying—thus prolonging
satiety and reducing further energy intake.
However, in this study, reduced energy intake was not
observed because participants were instructed to record
all calories consumed, as well as maintaining their cur-
rent lifestyle. Therefore, MHE-induced reduced energy
intake—either by reduced appetite or increased satiety—
cannot be used to explain the observed decreases in
body fat in the active group. Although food intake mea-
surement is notoriously inaccurate, the active group did
consistently consume more calories than the placebo
group throughout the test period.
Instead, it is suggested that MHE may accelerate ener-
gy expenditure, rather than inhibiting energy intake
through the control of appetite or satiety. MHBA has
been reported to enhance thermogenesis in brown
adipose tissue in rodents. It does this by binding to
bitter taste receptors in the GI tract. This receptor bind-
ing was shown to cause the downstream activation of
sympathetic nerve activity that regulates energy and
glucose homeostasis in brown adipose tissue.
Finally, this study was carried out by employees of
the Kirin Company, a global beer company that also
has business interests in pharmaceuticals and other
health-related products.
The consumption of MHE over the course of the
study resulted in a significant loss of body fat, which
may have been due to the acceleration of ener-
gy expenditure rather than reduced appetite or
increased satiety.
The big picture
Obesity is characterised by the excessive and patholog-
11. 11
ical accumulation of body fat known as white adipose
tissue: fat cells specialized in the acquiring and storing
of energy. The increase in visceral fat around intra-ab-
dominal organs is a key driver in insulin resistance and
the increased risk of developing type 2 diabetes. Higher
visceral fat content is therefore also associated with
increased risk of cardiovascular diseases and is a key
target for therapeutic approaches against obesity.
An alternative approach to combating obesity is to
activate brown adipose tissue. Brown adipose tissue
is thermogenic—meaning it releases energy in the
form of heat—and is critical for maintaining core body
temperature in small mammals, as well as in newborn
human infants. New research would appear to suggest
that human brown adipose tissue, when activated, has a
significant impact on energy balance and body weight.
In this particular study, it was shown that there is an
inverse correlation between BMI and amount of brown
adipose tissue in an individual. However, this obser-
vation could be explained by obese individuals being
less thermally challenged due to higher mass and better
insulation. This knowledge could potentially be used to
develop novel therapeutic interventions that effectively
reduce obesity through activation of functional brown
adipose tissue.
Over the years there have been several studies on the
health benefits of consuming hop-derived iso-α-acids,
including regulation of glucose metabolism, anti-in-
flammatory activity and amelioration of blood lipid
profiles. More recently, matured hop bittering compo-
nents were shown to influence brown adipose tissue in
rodents through activation of the sympathetic nerves
that innervate it. By activating brown adipose tissue,
these researchers were able to demonstrate that a func-
tional food could be used to prevent pathological body
fat accumulation.
As shown in Figure 3, hops are not the only natural
product that has shown promising results as a function-
al food to reduce body fat. Thai Ginseng (Kaempferia
parviflora) is a herbaceous member of the ginger family,
commonly found in Thailand. This plant is reported to
have similar properties to MHE. In one study, it was
reported to increase energy expenditure through acti-
vation of brown adipose tissue in mice.
Figure 3: Possible thermogenic agents for the prevention of obesity
Thai ginseng
extract
MHBA extract
UCP1-mediated
thermogenesis in
brown adipocytes
Figure 3: Possible thermogenic agents for the prevention
of obesity
12. 12
The relationship between excess fat distribution and
metabolic disorders is of critical importance in the
fight against obesity. Hop extracts are a potential
new treatment option, as they may activate brown
adipose tissue and accelerate fat loss in overweight
individuals.
Frequently asked questions
What are the long-term effects of MHE consumption?
This study is limited by the short duration of MHE
consumption. The authors postulated that continued
consumption of MHE would continue to provide these
benefits because no plateau in the results were observed.
However, further studies over longer periods are
needed to ensure continued safety and efficacy in the
long-term.
What are the adverse effects of consuming MHE on a
regularly basis?
During this study, participants were asked to commu-
nicate any adverse effects during the study. Although
a number of individuals reported digestive issues,
this was the case for both active and placebo groups.
Therefore, no adverse effects could be attributed to
MHE consumption at the doses tested in this study.
Can I just drink beer and get the same benefits?
The simple answer is, unfortunately, no. Beer—par-
ticularly strong hoppy IPAs—may have an iso-α-acid
content of 30-40 ppm (30-40 milligrams per liter).
However, consumption of large volumes of beer is obvi-
ously not recommended as a strategy for weight loss
due to the excessive calorie content of these beverages.
What should I know?
This randomized double-blind placebo-controlled study
showed that continual consumption of a bitter extract
from the hop plant reduced body fat in healthy over-
weight individuals. Although the reductions in BMI,
body weight and waist circumferences in the active
group were only significant when compared to the
baseline, the loss of visceral and subcutaneous fat from
the abdominal area was significant when compared to
the placebo group. With no obvious lifestyle changes
other than the daily consumption of 35 milligrams of
MHBA, this research suggests that a hop extract safely
reduces abdominal fat in overweight individuals. ◆
Convinced by this study? Unconvinced? Either way,
hop over to the ERD Facebook forum to discuss it.
Can I just drink beer and
get the same benefits?
The simple answer is,
unfortunately, no.
13. 13
Sugar is the ultimate
antioxidant and insulin
will make you younger:
Appreciating a few
poorly recognized but
critical contributions of
carbohydrate
By Chris Masterjohn
14. 14
To be clear, when I use the term “sugar” in the title of
this editorial I am not advocating a diet rich in refined
carbohydrate and washed down with sugar-sweetened
soft drinks. I am, instead, using the word in a less col-
loquial sense to refer to glucose, the primary sugar in
biological energy metabolism. And now that I am 34, I
don’t expect insulin or any other hormone to make me
33 again, but I do expect insulin to make me age more
gracefully and to protect me against many facets of
degenerative disease that tend to accumulate with age
in Westernized populations.
As the stance against saturated fat begins to soften, we
need to exercise caution that we do not replace the demo-
nization of saturated fat with the demonization of sugar.
Toward that end, I would like to use this space to high-
light some of the positive roles of glucose and insulin.
Glucose is the ultimate antioxidant. We tend to think of
antioxidants as compounds that will directly neutralize
reactive oxygen species in a test tube, but we actually
have a complex endogenous antioxidant system that
is ultimately fueled by glucose. Through the pentose
phosphate pathway, glucose supplies hydrogen ions and
electrons – which we could call “reducing power” – to
NADPH, which is derived from niacin, also known
as vitamin B3
. The enzyme glutathione reductase uses
riboflavin, also known as vitamin B2
, to pass this reduc-
ing power on to glutathione. Glutathione, the master
endogenous antioxidant, then uses this reducing power
to neutralize hydrogen peroxide to water, to neutralize
lipid peroxides to less harmful hydroxy-fatty acids, and
to recycle vitamin C. Vitamin C recycles vitamin E, the
principal bulwark against lipid peroxidation in cellu-
lar membranes. Thus, the multiple roles of glutathione
within the antioxidant defense system – mitigating the
accumulation of reactive oxygen species, protecting
vulnerable fatty acids within cellular membranes, and
cleaning up any damage that has slipped through the
system – are all ultimately supported by the reducing
power derived from glucose.
Insulin signaling is also important to the antioxidant
defense system because it increases the synthesis of
glutathione. One small but fascinating study published
in the journal Metabolism in 2006 demonstrated the
relevance of this point in type 2 diabetes. Compared to
healthy controls, diabetic patients had poor glutathione
status. In fact, their ratio of reduced to oxidized gluta-
thione was cut in half. The investigators then exposed
the diabetic patients to a euglycemic hyperinsulinemic
clamp. In this approach, both sugar and insulin are
infused into the patients’ blood at concentrations that
keep blood sugar normal and insulin elevated. After
As the stance against saturated fat
begins to soften, we need to exercise
caution that we do not replace the
demonization of saturated fat with
the demonization of sugar.
15. 15
being exposed to this condition for two hours, the glu-
tathione status of the diabetics normalized to the level
of healthy controls.
Pumping patients’ veins full of glucose and insu-
lin is wildly impractical as a treatment for diabetes.
Nevertheless, the study
provides proof of principle
that – at least with respect
to the antioxidant defense
system – type 2 diabetes is
a condition of inadequate
insulin signaling, not excess
insulin signaling.
Poor glutathione status is
most likely a major contrib-
utor to the accumulation
of advanced glycation
endproducts (AGEs) in
diabetes, which are, in turn,
major contributors to the
development of diabetes
itself as well as its cardio-
vascular complications. But
the contribution of insulin
to glutathione synthesis is
only one of several ways
that insulin protects against
AGEs.
Quantitatively, the two main contributors to AGEs
within human plasma are methylglyoxal and 3-deox-
yglucosone. Let us first consider how insulin would
affect the accumulation of methylglyoxal, and then we
will move on to 3-deoxyglucosone.
Methylglyoxal can be derived from glycolysis when
intermediates within the pathway known as triose phos-
phates accumulate. Insulin clears these intermediates
by stimulating the enzyme glyceraldehyde 3-phosphate
dehydrogenase (GAPDH). It can also be derived from
ketogenesis: acetone, one of the major ketone bodies, is
converted in a two-step process first to acetol and then
to methylglyoxal by the enzyme CYP2E1. Insulin sup-
presses ketogenesis and also suppresses CYP2E1. Once
formed, methylglyoxal is detoxified to pyruvate, which
can then enter energy
metabolism in a variety
of ways. This pathway
requires glutathione and
two enzymes known
as glyoxalase-1 and -2.
Insulin stimulates the
synthesis of glutathione
as well as the expression
of glyoxylase-1. Thus,
insulin suppresses the
generation of methy-
glyoxal from all sources
and stimulates its detox-
ification. This makes
insulin central to the
defense against AGEs.
The only study I know
of that has looked at
how this plays out in
living humans found
that methylglyox-
al levels rise on the
Atkins diet. The study
was small and had no control group, so it should not
be taken as the final word. But when we consider the
biochemistry involved, the findings are strongly consis-
tent with what we would expect from the falling insulin
levels that occur during carbohydrate restriction.
AGEs derived from 3-deoxyglucosone are greater in
concentration than those derived from methylglyoxal
in the plasma and blood cells of healthy controls. Both
classes of AGEs increase in diabetes, but the increase in
[...] at least
with respect to
the antioxidant
defense system
– type 2 diabetes
is a condition
of inadequate
insulin signaling,
not excess insulin
signaling.
16. 16
methylglyoxal-derived AGEs is greater, making them
more numerous than those derived from 3-deoxyglu-
cosone in diabetic plasma. If the findings from diabetic
rats can be generalized to humans, AGEs in tissues
besides blood are likely to be overwhelmingly derived
from methylglyoxal rather than 3-deoxylglucosone.
Nevertheless, for a more complete picture, I will briefly
discuss the metabolism of 3-deoxyglucosone.
3-deoxyglucosone is primarily derived from enzy-
matic metabolism of fructosamines, which form from
the direct interaction of sugar with protein. The most
notable fructosamine is Hba1c, the glycated form of
hemoglobin that is used in diagnosing and monitoring
diabetes. Although the metabolism of 3-deoxyglu-
cosone is poorly understood compared to that of
methylglyoxal, it appears to be primarily metabolized
to 3-deoxyfructose by a group of enzymes known as
aldoketo reductases. These enzymes are under the con-
trol of Nrf2, a transcription factor that regulates a suite
of genes involved in xenobiotic metabolism and antiox-
idant defense.
From a nutritional perspective, the polyphenol com-
pounds found abundantly in unrefined plant foods are
thought to be the principle dietary strategy to stimulate
Nrf2. Insulin would protect against 3-deoxyglucosone
accumulation by clearing glucose from the blood and
stimulating its downstream intracellular metabolism.
The reduction of 3-deoxyglucosone, moreover, requires
NADPH, which derives its reducing power from glu-
cose in the pentose phosphate pathway. Thus, although
high blood glucose concentrations drive the formation
of 3-deoxyglucosone, cytosolic glucose is critical to its
detoxification.
In trying to make sense of why our bodies would coor-
dinate protection against oxidative stress and glycation
to be dependent on glucose and insulin, I use the fol-
lowing paradigm. Our ability to store carbohydrate
is very limited because, compared to fat, glycogen
is voluminous, wet, and heavy. A person of healthy
bodyweight stores about 30 times as much fat as car-
bohydrate. Unlike the virtually unlimited supply of
fat within the body, glycogen stores are easily depleted
or repleted over the course of days. Thus, I believe our
bodies are hardwired to regard leptin (influenced most
strongly by total body fat) as a metric of long-term
energy status and insulin (influenced most strongly by
acute intake of carbohydrate) as a metric of short-term
energy status. Protecting against oxidative stress and
glycation requires energy-intensive processes that are
critical over the long-term but can be sacrificed over
the short-term with relative impunity if the body per-
ceives its short-term energy supply as limited.
In principle, glucose is the ultimate antioxidant and
insulin is central to the defense against oxidative stress
and glycation. Nothing I have written here, however,
implies that more glucose and more insulin is always
better. It would be foolish to think there is no point of
diminishing returns and no possibility of a U-shaped
curve where excesses could pose problems as severe
as inadequacies. If for no other reason, diminishing
returns will be seen when carbohydrate-rich foods
begin to displace protein- and fat-rich foods to the point
where proteins, fats, or the micronutrient profiles that
accompany them become the limiting factors for health.
What I am advocating here is a recognition of the
positive contributions of carbohydrate itself to these
systems. In popular writings, antioxidant defense is
often reduced to vitamin E, vitamin C, and plant poly-
phenols, while glycation is misleadingly attributed to
sugar. This could easily lead us to a diet rich in meat,
vegetables, and fat, without considering positive roles
for whole foods rich in natural sugars and starches.
Recognizing positive benefits of glucose and insulin
within these systems should cause us to open up our
menu to whole foods whose central place in the diet is
to provide carbohydrate.
17. 17
Defining exactly how much carbohydrate is needed to
optimize these systems would be difficult. Randomized
trials testing the long-term effect of isocaloric substi-
tutions of carbohydrate for other macronutrients on
glutathione status and AGE accumulation would be
useful, but to my knowledge have not been done. Even
if we had such studies, the carbohydrate requirement
would heavily depend on contextual factors such as
physical activity. Additionally, in the context of a diet
made from whole foods, shifting macronutrient profiles
will lead to inadvertent shifts in micronutrient profiles,
and the micronutrient profile of the diet could influ-
ence whether increases in carbohydrate supply show a
clear benefit to antioxidant defense and glycation status.
In clinical use, I think titration of the carbohydrate
supply should be one of the tools used to improve an
oxidized or deficient glutathione pool. The European
Laboratory of Nutrients Health Diagnostics and
Research Institute offers a methylation panel that
includes measurements of glutathione in its reduced
and oxidized forms. This kind of test could be used to
determine the need for such a titration and to assess the
efficacy of such a titration.
As we move forward, we need to frame our discussions
of glucose as a nutrient and insulin as a protective
hormone whose protective functions are being lost in
obesity and diabetes. With this framework, we may
be able to shake off the old rhetoric about fats without
redirecting it toward carbohydrate as the new nutrition-
al boogeyman. Then we can look freely at the buffet of
dietary tools at our disposal and study with a clearer
collective mind how to maximally reap their benefits in
a way that is tailored to each of us as an individual. ◆
Chris Masterjohn earned his PhD in Nutritional Science in 2012 from the
University of Connecticut at Storrs, where he studied the role of glutathi-
one and dietary antioxidants in regulating the accumulation of
methylglyoxal. He served as a postdoctoral research associate from
2012 to 2014 at the University of Illinois at Urbana-Champaign, where he
studied interactions between vitamins A, D and K. He is now Assistant
Professor of Health and Nutrition Sciences at Brooklyn College in
Brooklyn, NY, where he is continuing his research on fat-soluble vita-
mins. He has authored or co-authored ten peer-reviewed publications.
His writes a blog, The Daily Lipid, and produces a podcast by the same name. You can also follow
his professional work on Facebook, Twitter, Instagram, YouTube, and Snapchat.
18. 18
Milk gone
bad: A1 beta-
casein and GI
distress
Effects of milk
containing only A2
beta casein versus milk
containing both A1 and
A2 beta casein proteins
on gastrointestinal
physiology, symptoms
of discomfort, and
cognitive behavior
of people with self-
reported intolerance to
traditional cows’ milk
19. 19
Introduction
Milk is an important food for young infants and a com-
mon source of nutrition among adults. However, many
humans stop producing the lactase enzyme responsible
for digesting the milk sugar lactose after weaning, a con-
dition called lactose intolerance. When individuals with
lactose intolerance consume lactose through milk or
other forms of dairy, they may experience varying forms
of gastrointestinal (GI) distress, including abdominal
pain, bloating, gas, nausea, and diarrhea. These symp-
toms are caused by the fermentation of lactose in the
colon, since it was not absorbed in the small intestine.
Roughly 65% of the human population is considered
to have a reduced ability to digest lactose after infancy.
However, the prevalence of true lactose intolerance is
difficult to discern because studies have varied in their
interpretation of what constitutes this condition. Many
surveys rely on self-reported lactose intolerance, but
many individuals who self-report lactose intolerance
show no evidence of lactose malabsorption.
An alternative explanation for the high levels of self-re-
ported lactose intolerance may be the type of protein in
milk. The two major protein groups in milk are whey and
casein, with the latter accounting for about 80% of total
protein. The most common genetic variants of casein pro-
tein in milk are A1 beta-casein and A2 beta-casein.
A2 beta-casein is recognized as the original form of
beta-casein and is the only beta-casein found in the
milk of purebred Asian and African cattle. The A1
beta-casein variant is found among cattle of European
origin and is believed to have arisen more than 5,000
years ago. Accordingly, most milk sold commercial-
ly is a combination of A1 and A2 beta-caseins, as it is
sourced from European cattle or other cattle that have
been crossbred with European cattle. Examples include
Guernsey cows, Holsteins, and Ayrshires. Human
milk and milk from goats and sheep contains only A2
beta-casein.
The beta-casein proteins are degraded into beta-caso-
morphins (BCMs) during the digestive process. The
main difference between A1 and A2 beta-casein is that
A1 beta-casein produces BCM-7 upon digestion while
A2 beta-casein does not. There is a growing body of
evidence suggesting that BCM-7 is bioactive and is
associated with inflammation and several disease states,
such as diabetes and coronary heart disease. However,
these associations are not without criticism.
Up until now, nearly all the evidence investigating
health effects of BCM-7 and the beta-casein variants
has been observational or conducted in test tubes and
animals. The current study was designed to compare
the human health effects of consuming milk containing
only A2 beta-casein with milk containing A1 beta-ca-
sein type in terms of GI function, symptoms, and
inflammation.
The two common forms of casein present in milk are
A1 beta-casein and A2 beta-casein, which differ as a
result of a genetic mutation in cattle over 5,000 years
ago. There is observational, test tube, and animal
evidence to suggest that A1 beta-casein may promote
inflammation and be linked to inflammatory disease
states. The study under review put this to the test in
humans.
Who and what was studied?
This was a double-blind, randomized crossover tri-
al in Shanghai, China in which 45 middle-aged men
and women consumed 250 milliliters of milk after two
meals per day for 14 days. All participants were of Han
Chinese ancestry. They had a self-reported intolerance
to commercial milk (moderate digestive discomfort)
and did not regularly consume dairy, but none had irri-
table bowel syndrome or inflammatory bowel disease. A
urinary galactose test confirmed that 23 of the 45 par-
ticipants were lactose intolerant.
20. 20
Figure 1 summarizes the study design. Over the course of
eight weeks, each participant went through two two-week
milk phases and two two-week washout phases. During
the milk phases, the participants consumed either a
milk containing only the A2 beta-casein (from cows
confirmed to be A2-only producers) or a milk containing
a combination of A1 and A2 beta-casein (milk contain-
ing only the A1 beta-casein is not commercially available
and the A1/A2 combination is standard in consumer
milk). All participants completed both milk phases,
with half beginning with the A2-only intervention and
half beginning with the A1/A2 intervention. Aside
from the differences in casein type, the milk was iden-
tical. During the entire eight weeks, the consumption of
dairy products other than those provided was prohibited.
Participants used daily diaries to record milk intake, GI
symptoms using the Bristol Stool Chart (a medical aid
that classifies faeces into seven groups), and adverse
events. At the beginning and end of each two-week milk
phase, the participants underwent a computer-based
test that measured the speed and effectiveness of infor-
mation processing (Subtle Cognitive Impairment Test;
SCIT) and laboratory testing that included the use of a
smart pill (depicted in Figure 2) to record stomach and
intestinal inflammation and physiology. In addition to
the self-reported milk intake diary, the counting of milk
cartons was used to assess compliance.
Figure 1: Study designFigure 1: Study design
Cross-over Study
Day -13
2-week washout 2-week washout
Arm 1: A1/A2 milk
Participants maintain daily recording of adverse events, gastrointestinal symptoms, trial milk intake, dietary adherence
Phase 1
Arm 2: A2/A2 milk
Phase 2
Timeline
Trial Phase
and Diet
Day 0 Day 1 Day 14 Day 15 Day 28 Day 29 Day 42
Dairy Exclusion Diet Throughout All Washout and Trial Phases
A1 A1
Arm 1: A2/A2 milk
Arm 2: A1/A2 milk
A1 A1
Reference: Timm et al. Br J Nutr. 2011 May.
Figure 2: Functions of a Smart Pill
Uses pH, pressure, and
temperature to measure
transit time.
Reference: Timm et al. Br J Nutr. 2011 May.
Smart Pill
(OMOM Capsule)
Replaced more cumbersome
methods that involved
collecting and X-raying
fecal matter (gross).
Certain models can
wirelessly transmit information.
Some can record pictures
or videos that can be used
for diagnosing GI diseases.
Figure 2: Functions of a Smart Pill
21. 21
This double-blind, randomized crossover trial had
45 middle-aged, dairy-intolerant Chinese men and
women consume 250 milliliters of milk twice daily
after meals for two weeks. The milk contained either
only A2 beta-casein or both A1 and A2 beta-casein.
Measurements of GI function and inflammation, as
well as cognitive function, were assessed before and
after each intervention.
What were the findings?
Consumption of milk containing A1/A2 beta-casein led
to significantly greater increases in interleukin-4 (IL-4),
immunoglobulin (Ig) G, IgE, and IgG1 compared to the
consumption of milk containing A2 beta-casein only.
Additionally, A1/A2 milk significantly reduced fecal levels
of total short-chained fatty acids (SCFAs), acetic acid, and
butyric acid. The latter two are specific types of SCFAs.
GI symptoms significantly worsened with A1/A2 milk
only compared to baseline. Specifically, consuming
A1/A2 milk resulted in more bloating, flatulence, and
borborygmus (the rumbling or gurgling noise made by
the movement of fluid and gas in the intestines). Stool
frequency and stool consistency were also significantly
increased compared to baseline with the consumption
of A1/A2 milk, but not with the consumption of A2
only milk.
Using data from the smart pill, consumption of A1/A2
milk was associated with significantly longer GI transit
time than consumption of A2 only milk, by about six
hours (40 vs. 34 hours, respectively, as seen in Figure 3).
This was due to a significantly longer transit time in the
colon. Intestinal inflammation improved in 36% of par-
ticipants and stomach inflammation improved in 23% of
participants after switching from A1/A2 milk to A2 only
milk. By contrast, intestinal and stomach inflammation
both improved in 11% of participants when switching
from A2 only milk to A1/A2 milk. Almost all other par-
ticipants showed no difference between milk types.
Consuming A2 only milk was associated with signifi-
cantly quicker response time and lower error rate on
the SCIT than consuming A1/A2 milk.
Data was re-analyzed comparing individuals with con-
firmed lactose intolerance to those without. Consuming
A1/A2 milk was associated with significant worsen-
ing of GI symptoms in both groups, with the lactose
intolerant group exhibiting worse symptoms than
the tolerant group. This was not observed with A2
only milk, as GI symptoms were comparable to those
observed after the dairy-free washout period in both
lactose tolerant and intolerant individuals. Moreover,
the GI symptom scores between lactose tolerant and
intolerant groups were not significantly different from
one another when consuming A2 only milk.
Figure 3: A1/A2 gastrointestinal transit timeFigure 3: A1/A2 gastrointestinal transit time
A1-A2 A2-A1
5.0
Small Bowel Transit Time
Phase 1 Phase 2
Hours
4.5
4.0
3.5
3.0
2.5
2.0
2.0
1.0
0.5
0.0
40.0
Colon Transit Time
Phase 1 Phase 2
Hours
35.0
30.0
25.0
20.0
15.0
10.0
5.0
0.0
45.0
40.0
Whole Gastrointestinal Transit Time
Phase 1 Phase 2
Hours
35.0
30.0
25.0
20.0
15.0
10.0
5.0
0.0
22. 22
Both lactose tolerant and intolerant individuals showed
similar increases in whole GI and colon-specific transit
times with A1/A2 milk compared to A2 only milk. Both
groups also showed significant increases in IL-4, IgE,
and IgG1 and significant decreases in total SCFAs with
A1/A2 milk vs A2 only milk.
The primary adverse event was diarrhea and was report-
ed by 10 of 45 participants (22%). Of these, eight were
owed to the consumption of A1/A2 milk, three related
to A2 only milk, and three unrelated to either milk.
Consumption of milk containing only A2 beta-casein
was associated with significantly less serum inflam-
mation and GI symptoms than milk containing both
A1 and A2 beta-caseins. A2 milk was also associat-
ed with significantly greater SCFA production and
cognitive ability on the SCIT test. Individuals with
lactose intolerance reported similar GI symptoms as
those without lactose intolerance when consuming
A2 only milk but reported worse symptoms with A1/
A2 milk.
What does the study really
tell us?
The study under review shows that consuming milk
with only A2 beta-casein is associated with reduced GI
symptoms, lower concentrations of inflammatory bio-
markers, greater SCFA production in the colon, shorter
GI transit time, and shorter response time and lower
error rates on the SCIT compared with milk contain-
ing both A2 and A1 beta-caseins. The increased SCFA
production, primarily butyrate and acetate, is especially
notable considering that SCFAs play a prominent role
in human health and mediate the beneficial health
effects of fiber consumption. The shorter GI transit
time is difficult to interpret, as whether this may aid
with constipation or lead to diarrhea would depend on
the baseline transit time of the individual.
This study also suggested that some GI symptoms
ascribed to lactose intolerance were present only with
the consumption of milk containing A1 beta-casein.
Both milk products contained equal amounts of lactose,
which reinforces the concept that the differences in
outcomes were driven by the presence or absence of A1
beta-casein.
This study has notable limitations, such as the purely
Han Chinese study sample group. It’s also unclear how
a longer time frame would have impacted the results,
which future research will need to investigate, consid-
ering that milk consumption is often consistent and
prolonged in real life. In addition, this study focused
solely on GI symptoms, so any non-GI effects of A1
and A2 beta-caseins were not tested.
Finally, this study was funded by The a2 Milk Company
Limited. One of the six authors was also an employee
of this company, but he was not involved in perform-
ing the study or data analysis. Rather, he conceived and
designed the study, selected variables of interest, and
contributed to the manuscript.
This study tells us that consuming milk containing
only A2 beta-casein may result in less GI distress and
inflammation than consuming milk containing both
A1 and A2 beta-caseins. This applies to individuals
with and without lactose intolerance. More research
is needed to determine if these effects are observed in
populations other than the Han Chinese and if dura-
tion of consumption plays a mediating role.
The big picture
Studies using rats and mice have demonstrated that A1
beta-casein exhibits inflammatory properties mediated
by opioid receptors in the gut. This supports the find-
ings of the current study, but its implications remain
unknown. It is notable that the rat study also found A1
23. 23
beta-casein to increase the production of the enzyme
dipeptidyl peptidase 4 (DPP4) in the small intestine.
DPP4 degrades hormones that help regulate insulin
secretion and blood glucose levels, and DPP4 inhibitors
are widely used in the management of type 2 diabe-
tes. It is therefore possible that long-term exposure to
A1 beta-casein may have an effect on blood glucose
management, although future research will need to
investigate this.
The inflammatory opioid effects of the A1 beta-casein
derivative, BCM-7, have also been postulated to affect
the brain. This has been offered as one potential expla-
nation for the delayed psychomotor development in
cow milk formula-fed infants compared to breastfed
infants, as human breast milk contains only A2 beta-ca-
sein. Additionally, some studies have associated A1
beta-casein and BCM-7 with neurological diseases like
autism, schizophrenia, and psychosis.
The finding of increased response times and error rates
on the SCIT in the current study support the above,
as elevated levels of inflammatory markers have been
shown to play a role in Alzheimer’s disease and an
impairment of executive function and processing speed
in the elderly, even after controlling for age and other
health-related factors. Cognitive impairment has also
been observed in newly-diagnosed patients with celiac
disease, which improves with adherence to a gluten-free
diet. It is well-known that gluten elicits a powerful
immune and inflammatory response in these individuals.
The reduction in SCFAs in the colon could be owed to
an excessive production of mucus that normally pro-
vides a protective barrier and home to the microbiome.
This is because the microbiome organisms are respon-
sible for producing SCFAs as a byproduct of eating the
fiber we cannot digest. The A1 beta-casein derivative,
BCM-7, has been shown to increase mucus production
in test tubes and in rats. Whether these changes occur
in humans and whether they have a physiological effect
remain unknown. However, SCFAs play a prominent
The inflammatory opioid effects of
the A1 beta-casein derivative, BCM-7,
have also been postulated to affect
the brain. This has been offered as
one potential explanation for the
delayed psychomotor development
in cow milk formula-fed infants
compared to breastfed infants
24. 24
role in human health, such as through improved blood
glucose control and insulin sensitivity, and their reduc-
tion would be unfavorable.
Much of the evidence supporting the inflammatory
role of A1 beta-casein is observational or stems from
studies performed in test tubes and animal models. Any
human evidence is largely anecdotal and unreliable.
The only other human trial to investigate differenc-
es between A1 and A2 beta-casein was conducted in
41 men and women from Western Australia. Using
an eight-week crossover design similar to the current
study, this previous work showed that A2 milk was
associated with less bloating, abdominal pain, flatu-
lence, and voiding difficulty than A1 milk. The current
study confirms and extends these findings. Notably, this
previous trial used 750 milliliters of milk daily and an
A1 beta-casein only milk as the comparator to the A2
milk. The current study used less milk (500 milliliters
per day) and a commercially available combination
milk (A1 plus A2), both of which make these results
more applicable to the general population.
It is possible that the milk sugar lactose interacts with
BCM-7 to mediate the observed effects. This is support-
ed by the current study findings that individuals with
lactose intolerance did not report a worsening of GI
symptoms with the consumption of milk containing A2
beta-casein only. Perhaps the inflammatory properties
of BCM-7 affect the production of the lactose-de-
grading enzyme, lactase, leading to malabsorption in
normally lactose-tolerant individuals. Perhaps BCM-
7 changes the microbiome in a manner that makes it
more susceptible to lactose fermentation. These are
plausible theories, but require further testing.
A1 beta-casein and its derivative, BCM-7, have shown
a range of effects both in test tubes and in animal
models. This includes promoting inflammation that
has the potential to disrupt blood glucose manage-
ment and cognitive function over the long-term.
Data from human trials is limited but does support
the inflammatory findings. However, more research
is needed to investigate the long-term impact of A1
beta-casein consumption.
Frequently asked questions
Where can I buy milk that contains only A2 beta-casein?
A2 beta-casein is the only beta-casein found in human,
goat, and sheep milk, making these forms of dairy a
safe bet for reducing exposure to A1 beta-casein. The
a2 Milk Company also produces a milk from selectively
bred cattle that contains only A2 beta-casein, but this
product is not commonplace in the U.S. It is sold as a
premium product in Australia and New Zealand.
What should I know?
Although only two human trials have been published
comparing milk that contains A1 beta-casein to A2
beta-casein, both have shown that A1 beta-casein results
in greater GI distress. The current study adds to this by
showing that it is also associated with increased intes-
tinal inflammation and reduced cognitive functioning.
These findings support test tube and animal research
that has found similar effects. The current study also
suggests that some individuals with self-reported lactose
intolerance may be reacting to A1 beta-casein rather
than lactose, as they do not show symptoms when con-
suming milk containing only A2 beta-casein. ◆
Given the massive amount of dairy consumed in the world, you’d think A1 and A2 milk would be more widely stud-
ied. Hopefully this trial spurs further research. Discuss it at the ERD private Facebook forum.
25. 25
Arsenic in rice: big
trouble for little
infants?
Association of rice and rice-
product consumption with
arsenic exposure early in life
26. 26
Introduction
Arsenic is a naturally occurring element found in soil
and water, as it is present in the Earth’s crust as a constit-
uent of over 200 different minerals. Natural processes
like dust storms and volcanic eruptions, as well as
human application, such as through pesticides, her-
bicides, wood preservation, production of electronics,
paints, and other industry, contribute to the accumula-
tion of arsenic in the environment. Arsenic is found both
in inorganic and organic forms, which are compared in
Figure 1. The inorganic kind is generally recognized as
the more toxic form. Organic arsenic is found primarily
in seafood like fish, shellfish, and seaweed.
Inorganic arsenic is a known carcinogen that can cause
cancers of the lung, bladder, skin, kidney, and liver.
Infants are particularly vulnerable to toxicants, and data
from both observational studies and animal models
suggest that arsenic exposure during early life increases
the risk of respiratory diseases, impaired lung function,
cancer, and cardiovascular disease. By contrast, organic
arsenic exposure has little to no association with toxici-
ty in humans.
Infant rice cereal is a common first food during the tran-
sition away from breast milk or formula, but intake of
rice during infancy is not well characterized in the U.S.
There has been a growing concern over levels of inor-
ganic arsenic in rice and rice-based products because
some rice crops may be cultivated with contaminated
groundwater. This is amplified by the use flooded fields
for growing rice crops, as increased exposure of the
soil to water increases the amount of arsenic released
into water for absorption by the rice. This explains why
arsenic contamination in rice is particularly concerning
when compared, for example, to wheat and barley.
On April 1, 2016, the U.S. Food and Drug
Administration (FDA) proposed setting an upper limit
of 100 parts per billion for arsenic concentrations in
infant rice cereal. This parallels current regulations in
the European Union. Given the vulnerability of infants
to arsenic exposure, the study under review investigat-
ed rice-based food sources of arsenic exposure among
infants during their first year of life.
Figure 1: Inorganic versus organic arsenic
More toxic, both acutely
and chronically. It is this
form that has been
deemed a carcinogen.
More rapidly absorbed
once ingested, which
contributes to its higher
toxicity.
More likely to be found
in apple juice, rice, kelp,
some grains, and
vegetables.
Generally considered
fairly non-toxic,
although certain forms
may still be of concern.
Tend to be more slowly
absorbed, allowing the
liver and kidneys more
time to process them
out of the body.
More likely to be found
in marine organisms
such as fish.
Inorganic Arsenic Organic Arsenic
TOXIC
Figure 1: Inorganic versus organic arsenic
27. 27
Inorganic arsenic is a contaminant that accumulates
in rice. Since rice cereal is a common transitional
food for infants, the current study examined food
sources for arsenic exposure among infants during
their first year of life.
Who and what was studied?
This study included 759 infants born to mothers liv-
ing in New Hampshire with a private water system
(e.g., well or spring). Phone interviews with the moth-
ers were conducted four, eight, and 12 months after
delivery to collect information about the infant’s gen-
eral dietary patterns, introduction of solid foods, and
changes in water supply. Additionally, a home tap water
sample was collected for analysis.
The mothers completed a three-day food log for their
infant 12 months into the study. On the last day of this
food log, urine samples were collected from the infants
for analysis of total urinary arsenic and several specific
types of inorganic arsenic. Only 129 infants had three-day
food logs and information on total urinary arsenic con-
centrations, while 48 had data on specific types of urinary
arsenic. This was partly because researchers excluded
infants that had seafood consumption in their food log,
as it contains organic arsenic, which is not viewed as a
health concern and would have inflated the results.
Rice consumption in the three-day food logs was cat-
egorized into three general categories: no rice, food
mixed with rice (adult food mixed with rice, infant
food mixed with rice, and snacks that have rice as an
ingredient), and rice (pure rice, infant rice cereal, and
adult rice cereal). The difference between the infant and
adult categories is only that the former is specifically
marketed to parents of toddlers. Urinary arsenic con-
centrations were then compared to the consumption of
these various rice products.
Finally, rice snacks frequently reported in the infant
food logs were purchased from online sources and a
local supermarket in Hanover, New Hampshire for
analysis of arsenic content.
Infants had their exposure to arsenic during the first
year of life analyzed and associated with rice-based
food items in their diet.
What were the findings?
The main findings are summarized in Figure 2. Eighty
Figure 2: Rice product consumption and arsenic urinary concentrations
No Rice Adult Food
Mixed With
Rice
Rice
Snacks
RiceBaby Food
Mixed With
Rice
Mixed With Rice Rice
Nonbaby
Rice
Cereal
Baby
Rice
Cereal
% of infants consuming rice products prior to urine collection
ConsumptionofRice-ContainingFoods,%
50
45
35
30
25
15
5
0
10
20
40
No Rice Adult Food
Mixed With
Rice
Baby Food
Mixed With
Rice
Mixed With Rice Rice
Rice
Snacks*
* - P < .001
** - P < .01
*** - P < .01
Rice** Nonbaby
Rice
Cereal***
Baby
Rice
Cereal***
Urinary arsenic concetrations in infants who consumed rice products
TotalUrinaryArsenic,µg/L
5
10
15
20
25
0
Figure 2: Rice product consumption and arsenic urinary concentrations
28. 28
percent of infants were exposed to rice cereal in their
first year of life, with 64% starting between four to six
months. Food logs prior to urine sampling revealed that
55% of infants consumed some type of rice product in
the previous three days, with 33% eating a rice snack,
10% eating infant food mixed with rice, 8% eating pure
white or brown rice, 6% eating infant rice cereal, 6%
eating adult food with rice, and 5% eating adult rice
cereal. These numbers don’t add up to 55% because
some infants consumed more than one category of rice
product. While evaluating one dietary item in an adult
diet may not provide much evidence for determin-
ing total arsenic intake because of the different foods
present in the diet, an infant’s diet is much simpler and
composed of fewer items, making it easier to evaluate
total dietary arsenic contribution by determining the
arsenic content of rice.
Median total urinary arsenic concentrations were 4.11
micrograms per liter, and 75% of the participants had
concentrations between 2.06 and 7.27 micrograms
per liter. Urinary arsenic concentrations were signifi-
cantly greater in infants who consumed rice (4.2-8.0
micrograms per liter) or foods mixed with rice (3.3-5.2
micrograms per liter) compared with infants who did
not eat rice (2.4-3.3 micrograms per liter).
The highest total urinary arsenic concentrations were
observed among infants who consumed infant rice
cereal (9.5 micrograms per liter), followed by adult
rice cereal (5.5 micrograms per liter), rice snacks (5.0
micrograms per liter), and rice (4.5 micrograms per
liter). These values were significantly greater than that
of infants who consumed no rice (2.9 micrograms per
liter). Additionally, frequency of consumption of rice or
foods mixed with rice was significantly associated with
higher urinary arsenic concentrations. Controlling for
infant sex and home tap water arsenic concentrations
did not change the results.
Total inorganic arsenic concentrations of the different
rice-based infant snacks ranged from 4.6 to 201 nano-
grams per gram. As seen in Figure 3, there was notable
variation not only among different products, but also
Figure 3: Arsenic concentrations in infant snacks (ng/g)
Strawberry grain snack
Carrot/Blueberry grain snack
Banana rice biscuits
Apple rice biscuits
Original rice rusks
Banana grain snacks
Blueberry whole grain snack
Vanilla cereal snack
Green vegetable grain snack
0 50 100 150 200
Figure 3: Arsenic concentrations in infant snacks (ng/g)
European Union limit of 100 ng/g
29. 29
among different flavors of the same product. For exam-
ple, the 4.6 nanogram per gram low was found in a
“green vegetable” flavor puffed grain snack, while the
201 nanogram per gram high was found in the “straw-
berry” flavor of the same puffed grain snack.
The majority of infants consuming rice cereal had
significantly higher levels of urinary arsenic concen-
trations than did infants who did not consume rice
cereal. Analysis of the most frequently consumed
rice snacks revealed highly variable levels of inorgan-
ic arsenic.
What does the study really
tell us?
The results of this study indicate that urinary arsenic
concentrations are significantly higher among infants
who consume rice and foods mixed with rice compared
to non-consumers, and that frequency of rice con-
sumption is significantly associated with higher urinary
arsenic concentrations. Additionally, some commonly
consumed infant rice snacks contained arsenic levels
greater than standards put forth by the European Union
and, more recently, the FDA.
What this study does not tell us is whether the differ-
ence in urinary arsenic concentrations between rice
consumers and non-consumers has an impact on
health. Also, the cross-sectional nature of the analysis
prevents establishing causality. Still, considering that
rice products have arsenic in them and that the diets of
infants are relatively monotonous, it stands to reason
that rice is a primary route of arsenic exposure.
People are most likely to be exposed to inorganic arse-
nic through drinking water and, to a lesser extent,
through diet. Considering that only 12.5% of the wom-
en in this study had a private water system with arsenic
concentrations above the 10 parts per billion maximum
contaminant level set by the Environmental Protection
Agency, and also that controlling for the water arsenic
content did not change the results, it stands to rea-
son that diet was a significant route of exposure for
these infants. Even so, other dietary sources of arsenic,
like apple juice, cannot be ruled out as contributors
to infant arsenic exposure. However, the association
between rice consumption and urinary arsenic has also
been observed in both adults and children.
The use of food logs presents another limitation, as it
presents an opportunity for misreporting of intake or
amounts. Also, since not everyone has access to a pri-
vate water system, this does present a limitation when
attempting to generalize these results to other popula-
tions. Water sources in some parts of the United States
have higher naturally occurring levels of inorganic
arsenic than other areas.
Infants who consume rice and foods mixed with rice
have higher urinary arsenic concentrations than
infants who do not consume rice, but the health
implications of this difference cannot be determined
by this study. Additionally, some popular infant rice
snacks contain arsenic concentrations above stan-
dards set by the European Union and FDA.
The big picture
The average total inorganic urinary arsenic concen-
trations in U.S. adults is about 5.6 micrograms per
liter. This value is lower than that of infants consuming
infant rice cereal and around the levels observed in
other rice-eating infants of the current study. However,
due to the small size of infants, this presents a notable
difference in arsenic concentrations per unit of body-
weight. Additionally, glomerular filtration rate is lower
in infants than adults, meaning that they may clear
arsenic into the urine more slowly than adults.
30. 30
Exposure to arsenic during critical windows of vul-
nerability, like infancy, may result in greater health
risks at similar exposure levels. Maternal exposure to
contaminated drinking water has been associated with
increased risk of spontaneous abortion and infant mor-
tality. Other research has shown that the relative risk of
developing lung and bladder cancers were several times
higher when arsenic exposure occurred in early life
compared to exposure in adulthood.
Total urinary arsenic is not considered abnormal until
it is in excess of 100 micrograms per liter, despite this
level being greater than that of 95% of U.S. adults.
However, evidence linking disease states to urinary
arsenic concentrations at the lower levels observed in
the current study is mixed. Higher levels of urinary
arsenic may be associated with increased prevalence of
type 2 diabetes but do not appear to be associated with
hypertension. Another study from Mexico suggests
that urinary arsenic concentrations are associated with
an increased risk of cardiometabolic diseases only after
exceeding 27 micrograms per liter. The limitation with
this research is that it is conducted in adults, not infants.
While the authors suggest that the consumption of rice
was the primary source of arsenic for the infants in this
study, it is important to consider the fact that if chil-
dren consumed approximately 11.25 grams of infant
rice cereal per day (the average infant rice consumption
assumed by the FDA for a 10 kilogram infant), their
rice intake would only account for about 20-30% of
the total urinary arsenic excreted by the infants con-
suming rice, with other important sources likely being
water intake or ingestion of dirt and dust from hand-to-
mouth activities.
Exposure to arsenic during infancy may result in
exaggerated health consequences compared to expo-
sure later in life due to the vulnerability of the infant
life stage. However, although rice may be a primary
source of arsenic exposure for infants, average rice
cereal consumption would lead to rice accounting
for only 20-30% of total arsenic exposure observed in
this study. This means that there are other important
sources of exposure, such as water or ingestion of dirt
and dust.
Frequently asked questions
Is there a difference in the arsenic content of brown and
white rice?
Arsenic accumulates primarily in the outer layer (bran)
of the grain. This portion is removed during the pol-
ishing of rice to produce white rice, but is retained in
brown rice. Accordingly, it stands to reason that brown
rice would contain more arsenic than white rice. This
has been confirmed via an analysis of 697 samples of
rice and rice-containing products. On average, brown
rice had about 80% more inorganic arsenic than white.
Although brown rice of a particular type always had
more inorganic arsenic than white rice, California
brown rice or brown rice imported from India or
Pakistan had notably lower levels of arsenic than other
brown rice.
On average,
brown rice had
about 80% more
inorganic arsenic
than white.
31. 31
When looking solely at various types of white rice,
basmati/texmati had the lowest inorganic arsenic con-
centrations, followed by sushi, jasmine, short-grain,
medium-grain, and long-grain. When further broken
up by place of origin, basmati rice imported from India
or Pakistan or grown in California had the lowest
average levels of inorganic arsenic. Additionally, all
U.S.-grown sushi rice had similar total inorganic arse-
nic levels as imported and California-grown basmati
rice. Rice with origins in the south-central U.S. tend to
have higher levels of arsenic than those from elsewhere,
and rice simply noted to be from the U.S. without spe-
cifics noted have the highest levels.
Can I reduce the arsenic content of rice?
Yes, it has been shown that cooking rice in a greater vol-
ume of water leads to lower levels of inorganic arsenic.
Specifically, cooking rice in 12 parts water to one part
rice removed 53% and 61% of inorganic arsenic from
white and brown rice, respectively. These values were
30% and 40% for cooking in six parts water to one part
rice and 30% and 20% for cooking in three parts water.
Contrary to popular belief, rinsing rice before cook-
ing has little impact on arsenic content, though it does
remove about 10%.
What should I know?
Exposure to arsenic during infancy may be linked to
numerous adverse health outcomes later in life. This
study showed that infants consuming rice and foods
mixed with rice had significantly higher levels of uri-
nary arsenic than did non-consumers. Therefore, while
causality cannot be established, this study does support
the notion that consuming rice may increase arsenic
exposure in infants. ◆
Eating habits may be important when you’re an adult,
but the diets of infants are especially important for their
later health. Talk about arsenic exposure over at the
ERD Facebook forum.
[...] while
causality cannot
be established,
this study does
support the
notion that
consuming rice
may increase
arsenic exposure
in infants.
32. 32
How much protein does
grandpa really need?
Dietary Protein Requirement of Men >65
Years Old Determined by the Indicator
Amino Acid Oxidation Technique Is Higher
than the Current Estimated Average
Requirement
33. 33
Introduction
Large public health organizations like the World Health
Organization (WHO) and the Food and Agriculture
Organization of the United Nations (FAO) make rec-
ommendations for protein intake as part of their efforts
to improve global health. Two numbers are often used
in this context: Estimated Average Requirement (EAR)
and Recommended Daily Allowance (RDA). The EAR
refers to an intake value that is estimated to meet the
requirement of half the healthy individuals in a group,
and by definition implies a high risk of inadequacy (i.e.
only 50% of a healthy population will have a higher
intake). The RDA is set at an intake level that should
meet the needs of about 97% of healthy individuals.
The current EAR for protein intake in healthy adults
according to the WHO and Institute of Medicine is 0.66
grams per kilogram of bodyweight per day, and the
RDA is 0.8 grams per kilogram of bodyweight per day.
Recommendations for healthy adults are generally
derived from studies conducted mainly in younger
adults. This is potentially problematic for a number of
reasons. For example, compared with younger adults,
the muscles of older adults are less sensitive to smaller
doses of protein and appear to require a greater amount
of protein to fully stimulate muscle protein synthesis.
Inadequate protein intake results in loss of lean body
mass, decreased muscle function, and even an impaired
immune response to stress and infection. Previous
research has shown protein intakes at the current RDA
can lead to negative nitrogen balance and loss of muscle
mass in elderly individuals. Due to this, some recom-
mendations now suggest an intake of 1.2-1.5 grams of
protein per kilogram of bodyweight per day for certain
older populations. Observational research has found
associations in elderly people between a greater dietary
protein intake and improved health.
Current recommendations are based on data from
nitrogen balance studies, which are a type of study that
attempts to determine protein requirements by measur-
ing the amount of nitrogen (a key component of amino
acids, the building blocks of protein) that is ingested
and the amount that is excreted (via sweat, urine, feces,
hair, and skin).
Unlike carbon, oxygen, and hydrogen (other building
blocks of amino acids), nitrogen doesn’t turn into gases
and water during metabolism, so it can be measured as
[...] compared with younger adults,
the muscles of older adults are less
sensitive to smaller doses of protein
and appear to require a greater
amount of protein to fully stimulate
muscle protein synthesis.
34. 34
a specific marker of protein balance. Being in negative
nitrogen balance means that the amount of nitrogen
excreted from the body is greater than the amount of
nitrogen ingested, and is associated with wasting dis-
eases and muscle catabolism, while positive nitrogen
balance means that the amount of nitrogen excreted
from the body is less than the amount of nitrogen
ingested, and is associated with growth and muscle
buildup. The use of nitrogen balance to determine opti-
mal protein intake dates back over 100 years.
However, the nitrogen balance method has several
limitations. For instance, variations in overall body
nitrogen and amino acid metabolism can lead to a
change in nitrogen balance, with or without a change
in intake. If excretion is changing without a change in
intake, it becomes difficult to determine whether or not
the effect is due to the dietary protein intake, because
this method is basing conclusions on specific dietary
protein intakes. Also, testing requires five to ten days of
adaptation to each level of amino acid by a participant,
and complete collection and quantification of all sourc-
es of nitrogen excretion (mostly in urine and feces, but
also sweat) is difficult.
An alternative method for studying protein needs is
the indicator amino acid oxidation (IAAO) method,
which is shown in Figure 1. This method is based on
the concept that when one essential amino acid (EAA)
is deficient for protein synthesis, then all other EAAs,
including the “indicator” amino acid, will be oxidized.
This is because the deficient amino acid becomes the
limiting step in protein synthesis. The indicator amino
acid (often phenylalanine, though lysine or leucine can
also be used) is “labeled” with a stable isotope (like a
tracking device). The appearance of the label in exhaled
carbon dioxide is used as an indicator of a protein or
amino acid requirement. This method was originally
developed to determine requirements for individual
amino acids. Researchers would vary the levels of one
Figure 1: IAAO in a nutshell
Essential
amino acids
Essential amino acid
with radiolabelled
carbon = indicator
amino acid
Anabolism
Protein
Catabolism
Nonessential
amino acids
c
Amino acids can either
be used to build protein
or to be oxidized for fuel.
When there aren’t
enought essental amino
acids to build protein ...
...and radiolabelled
carbon dioxide can be
measured from the
indicator amino acid.
...they get burned for fuel...
Figure 1: IAAO in a nutshell
co2
35. 35
EAA and measure oxidation of the tracer. However, in
the current study, the researchers varied the levels of all
EAAs (except phenylalanine and tyrosine) in parallel,
then observed how much of the other EAAs it takes
for phenylalanine oxidation to drop, as measured by
the tracer. Since there is no storage of amino acids, the
indicator is either incorporated into protein or oxidized.
This method is non-invasive, doesn’t require a week of
adaptation at a set protein level, and can be measured
through breath and urine samples.
Recent studies using the IAAO method in older women
(older than 65 years and older than 80 years) and young
men have found the protein requirement to be higher
than the current RDA. The primary goal of the current
study was to use the IAAO method to determine protein
requirements for healthy men older than 65, and to com-
pare that with the needs of older women and younger
men, as previously determined by the same method.
Current recommendations for protein intake may be
underestimated, particularly for older adults. This
study used newer and potentially more accurate test-
ing methodology to determine protein requirements
for men over the age of 65.
Who and what was studied?
Six men aged 65 or older participated in this study.
Potential participants were excluded if they had a recent
history of weight loss, disease, or acute illness that
could affect protein and amino acid metabolism (e.g.,
diabetes, cancer, liver or kidney disease, or HIV). High
blood pressure was not a reason for exclusion, if it was
well controlled by medication.
The study design is depicted in Figure 2. Participants
were randomly assigned to receive test protein intakes
ranging from 0.2 to 2.0 grams per kilogram of body-
weight per day. Each participant was included in seven
studies, with each three-day study period separated by
one to two weeks. During each three-day study, two
adaptation days were followed by the IAAO study on
the third day.
On the two adaptation days, participants received a
lactose-free milkshake maintenance diet that supplied
1.0 gram of protein per kilogram of bodyweight per day
and total calorie intake of 1.7 times each participant’s
measure resting energy expenditure (REE; this is stan-
dard for weight maintenance, with the activity factor
of 1.7 calculated to consider general daily activity). The
daily diet was consumed as four equal meals. The set
protein intake of one gram per kilogram was chosen
because previous research showed that varying the
protein intake had a significant effect on amino acid
kinetics on the study day. This implies that the body
can adapt over time. Using a high-protein run-in with
a one-day test period could artificially inflate protein
requirements because there is no time for adaptation to
Which amino acids are
essential, nonessential,
and conditionally
essential?
Essential amino acids are those that cannot be
made by the body. These are histidine, isoleu-
cine, leucine, lysine, methionine, phenylalanine,
threonine, tryptophan, and valine. In contrast,
non-essential amino acids can be made by the
body, even if we do not consume them. These
are alanine, asparagine, aspartic acid, and
glutamic acid. Finally, conditionally essential
amino acids are only essential during times
of increased stress or illness, or in people that
do not synthesize them in adequate amounts.
These include arginine, cysteine, glycine, gluta-
mine, proline, serine and tyrosine.
36. 36
lower protein levels. Lactose-free milkshakes were cho-
sen because five of the six participants reported being
lactose intolerant.
On the third study day, the IAAO testing was per-
formed. Participants were provided protein intakes
ranging from 0.2 to 2.0 grams per kilogram of body-
weight, after a 12 hour fast. The food was consumed as
eight separate meals given hourly, and was not exactly
appetizing. The experimental diet consisted of Tang
and Kool-Aid, grape seed oil, an amino acid mixture
similar to egg protein, and protein-free cookies. Caloric
intake was provided at 1.5x REE, with the carbohy-
drate content of the diets being adjusted according to
the protein intake in order to keep the calorie intake
the same. The study diet consisted of 40% fat, 37-57%
carbohydrate, and 3-37% protein. Phenylalanine was
used as the “indicator” amino acid, and thus intake was
kept constant at 30 milligrams per kilogram of body-
weight per day across all studies. Additionally, tyrosine
was also kept constant at 40 milligrams per kilogram of
bodyweight per day to ensure adequate tyrosine intake,
which can be formed from phenylalanine. If there was
inadequate tyrosine intake, some of the phenylalanine
would go to making tyrosine and change the results of a
given test intake.
Six men aged 65 years or older were tested seven
times with protein intakes ranging from 0.2 to 2.0
grams per kilogram per day. The non-invasive indica-
tor amino acid oxidation (IAAO) method was used to
determine protein balance.
What were the findings?
Figure 3 summarizes the study findings. The rate of
the labeled CO2
released from the oxidation of phe-
nylalanine (the indicator amino acid) declined with
increasing protein intakes up to 0.94 grams per kilo-
gram of bodyweight per day. Additional increases in
protein intakes did not result in changes in the labeled
CO2
values, meaning there were no additional increas-
es in the incorporation of phenylalanine for protein
Figure 2: Experimental protocol
6 male
subjects over
65 years of age
Repeat for a
total of 7 times,
each with a
different protein
intake on the
IAAO day
between
0.2-2 g/kg
bodyweight
Two days of
adaption on 1g/kg
protein and caloric
intake 1.7 x resting
energy expenditure
IAAO day on
0.2-2g/kg protein
and caloric intake
1.5x resting energy
expenditure
Two days of
adaption on 1g/kg
protein and caloric
intake 1.7 x resting
energy expenditure
IAAO day on
0.2-2g/kg protein
and caloric intake
1.5x resting energy
expenditure
One to two week
washout
Figure 2: Experimental protocol
37. 37
synthesis. This would imply an estimated average
requirement (EAR) of 0.94 grams per kilogram of
bodyweight, with an RDA of 1.24 grams of protein per
kilogram per day required to ensure the needs of 97%
of the population.
The researchers also compared data from this study to
previous studies they had conducted in older women
(older than 65 years) and young men using the same
IAAO method. When expressed in grams per kilogram
of body weight, there were no differences in protein
requirements between the three groups (0.94 grams per
kilogram). However, when comparing per kilogram of
fat-free mass (FFM) the protein needs were higher in
older men and women (1.6 g/kg FFM) compared with
young men (1.14 g/kg FFM).
Using the indicator amino acid oxidation method,
the estimated average protein requirement and RDA
for older men were 0.94 and 1.24 grams per kilogram
per day, respectively.
What does the study really
tell us?
This study suggests that current recommendations for
dietary protein intakes (0.66 g/kg/d for the EAR and
0.8 g/kg/d for the RDA) may be an underestimation for
older adults.
The IAAO method, which has been compared to the
nitrogen balance method, suggests that an intake
between 0.94 g/kg/d and 1.24 g/kg/d may be more suit-
able for a general recommendation. Limitations of the
the study include the small sample size of six men. But
these results concord with previous studies by the same
group, where similar results were observed, implying
that protein requirements are not affected by age or sex
when compared on a bodyweight basis.
However, there do appear to be increases in protein
requirements on the basis of FFM in older, compared
with younger, adults. This could be due to muscle pro-
tein synthesis being impaired in older adults compared
Figure 3: How the IAAO experiment determines adequate protein intake
Data from the 6
different subjects.
...until adequate protein
intake is reached.
As protein intake
increases, less is
burned and released
as carbon dioxide...
2.00
1.75
1.50
1.25
1.00
0.75
0.50
0.25
0.00
0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00
Breakpoint = 0.94 g/kg per day
Protein Intake (g/kg per day)
Amountofradioactivecarbondioxidefromoxidized
indicatoraminoacid
Figure 3: How the IAAO experiment determines adequate
protein intake
38. 38
with younger adults, with greater protein intakes being
required to lead to comparable protein synthesis rates.
Additionally, the study design, and more specifically
this method, does not allow time for the body to adapt
to lower protein intakes, which can occur in response
to consistently low protein intakes.
The big picture
This paper is the third in a series of studies looking at
the protein needs of older adults. Previous data in older
women (older than 65 years) showed similar protein
requirements compared to this study (an EAR of 0.96 g/
kg compared to 0.94 g/kg found in this study). On the
basis of fat-free mass, however, the estimated protein
needs are higher in older adults (about 1.6 g/kg FFM)
compared with younger adults (1.14 g/kg FFM). This
would be expected, because FFM is typically a smaller
percentage of total bodyweight on older adults.
There are a few potential reasons for the increased
dietary protein requirements in older people on the
basis of FFM. Compared with younger adults, the
muscles of older adults are less sensitive to protein and
require a greater amount of protein to fully stimulate
muscle protein synthesis. Also, daily protein turnover is
lower in older adults than it is in younger adults. This
is due to a decreased contribution of skeletal muscle
to whole-body protein breakdown. Skeletal muscle
supplies a significant percentage of the amino acids
required to promote whole-body protein synthesis, and
this decreased contribution could be what’s causing an
increase in required dietary protein intake.
Another study by the same group of researchers in
octogenarian women also found the current EAR and
RDA to be underestimated, with the average protein
requirement (EAR) being 0.85 grams per kilogram and
an adequate allowance (RDA) value to be 1.15 grams
per kilogram. Beyond the IAAO studies, a number
of observational studies provide support for optimal
protein intakes being greater than current recommen-
dations. For further reading, two recent reviews on the
subject are suggested.
Much of the existing body of research on protein
requirements has used nitrogen balance studies.
However, the IAAO method has become widely accept-
This method also allows research
in populations less commonly
studied, such as pregnant women
and children, for whom exposure
to dietary protein deficiency for up
to ten days in a row during testing
would be clearly unethical.
39. 39
ed as a valid model and allows for a less invasive
experimental design for protein requirement studies.
When an essential amino acid is missing or limited in
the diet, protein synthesis is then limited by that par-
ticular amino acid. IAAO uses the oxidation of carbon
as an endpoint, in contrast to nitrogen balance studies
that measure nitrogen intake and excretion. This meth-
od also allows research in populations less commonly
studied, such as pregnant women and children, for
whom exposure to dietary protein deficiency for up
to ten days in a row during testing would be clearly
unethical. An important limitation to this study design
is that it does not allow for a period of adaptation to a
low-protein diet. Evidence from both humans and ani-
mals suggests the body can adjust metabolic processes
in order to maintain nitrogen balance.
Increasing lines of evidence are suggesting that the
current dietary recommendations for protein intake
may be underestimated. This study is supported by
previous research by the same group using the IAAO
method, as well as observational evidence suggesting
an increased minimum dietary protein requirement.
Frequently asked questions
Does it matter what type of protein I consume when
trying to meet the RDA?
Yes. Dietary protein recommendations are made with
the assumption that it consists of “high quality” pro-
tein. Protein quality refers to the balance of amino
acids, digestibility of the protein, and the availabili-
ty of the absorbed amino acids for protein synthesis.
Lower quality proteins will require a greater dietary
intake in order to meet the body’s needs for essential
amino acids. People may be familiar with the Protein
Digestibility Corrected Amino Acid Score (PDCAAS),
but a newer and more accurate estimation of protein
quality is the Digestible Indispensable Amino Acid
Score (DIAAS), recently proposed by the Food and
Agriculture Organization. DIAAS scores for animal
proteins (e.g. milk, eggs, and beef) are well above 100%,
while vegetable proteins are usually less than 80%, with
the exception of soy. Combining incomplete vegetable
proteins such as rice and beans can be used effectively
to provide the body with its complete needs.
How does recommended protein intake compare with
actual protein intake?
This is difficult to answer because large-scale observa-
tional studies are notoriously inaccurate, but there is
likely a fairly large variation of intake. One study in old-
er women (aged 60-90 years) found the average protein
intake to be 1.1 grams per kilogram of bodyweight per
day, though 25% of the cohort consumed less than the
RDA of 0.8 grams per kilogram of bodyweight. Another
study in older men and women reported an average
intake of around 70 grams per day, with a standard
deviation of nearly 25 grams, indicating a very wide
range of individual intakes.
What should I know?
Use of a newer (and possibly, but not definitely, better)
method for determining dietary protein requirements
suggests that the current intake recommendations for
older adults of 0.66 grams per kilogram of bodyweight
per day for the estimated average requirement (EAR)
and 0.8 grams per kilogram of bodyweight per day for
the recommended daily allowance (RDA) may be too
low. Although this study used a very small sample size,
when combined with previous research by the same
group, a more accurate recommendation for this pop-
ulation would be 0.94 g/kg as an EAR and 1.24 g/kg as
an RDA. More research is needed, however, before it
would be prudent to increase the EAR and RDA. ◆
Protein intake is arguably more important for elderly
people than for younger people. Discuss the specifics of
protein recommendations at the ERD private Facebook
forum.
41. 41
Introduction
All exercise causes stress, which is followed by adapta-
tions that allow the body to better handle that stress in
the future. Environmental conditions, such as extreme
temperatures or changes in altitude, or even humidity,
can also cause stress and adaptation. Recent research
has shown that adaptation to one type of stressor can
induce protective responses upon exposure to a differ-
ent stressor, presuming they share common adaptive
responses. When improved cellular protection or
reduced physiological strain is observed in an organism
in response to a different type of stressor, it is referred to
as cross-tolerance. Some examples of possible cross-tol-
erance from rat experiments are shown in Figure 1.
Changes associated with heat acclimation include an
increased cardiac output and plasma volume, an earli-
er onset of sweating with an increased sweat rate (and
more diluted sweat), lower core temperature at rest and
during exercise (reduced thermoregulatory strain) and
a lower heart rate for a given workload during exercise
(reduced cardiovascular strain). Hypoxic exposure
refers to a state of reduced oxygen availability, experi-
enced when traveling to high altitudes or simulated in
a lab. Responses to hypoxia include an increase in red
blood cells and oxygen-carrying capacity, as well as
adaptations for improving carbohydrate metabolism.
At the cellular level, both heat and hypoxic stress stim-
ulate the heat shock response. Heat shock proteins
(HSPs) refer to a large family of proteins that aid in
a cell’s response to physical stress. The importance
of these proteins is evident by the fact that they are
observed in species across the animal kingdom. The
Adapted from : Fregly MJ.Compr Physiol. 2011.
Figure 1: Some cross-adaptions found in rats
Whenexposedto...
...adapts to...
Low
pressure
Low
pressure
Cancer-
like
stressors
Cancer-
like
stressors
Exercise
Exercise
Morphine
Morphine
X-rays
X-rays
Cold
Cold
Restraint
Restraint
Adapted from : Fregly MJ.Compr Physiol. 2011.
= Positive adaption
(fares better)
= Negative adaption
(fares worse)
= Data unclear
Blank = no data
Figure 1: Some cross-adaptions found in rats
42. 42
most commonly studied HSPs expressed in response
to heat or hypoxia are the HSP70 and HSP90 families,
which are named after their molecular weight. In addi-
tion to HSPs, increases in hypoxia-inducible factor 1-α
(HIF1-α), the master regulator of oxygen-regulated
genes, have also been observed in animals after heat
stress and heat acclimation. This suggests an associa-
tion, and potential interaction, between HIF-1α and
HSPs during heat acclimation.
Previous research using exposure to heat and hypox-
ia (alone and in combination) has shown reductions
in physiological strain (measured by the heat shock
response) during subsequent hypoxic exercise. Human
studies have yet to examine the HIF-1α response to heat
acclimation, which may play a role in the improved
performance seen during hypoxic exercise after heat
acclimation. Accordingly, the goal of this new study
was to examine the effects of heat or hypoxic accli-
mation on physiological, cellular, and performance
responses to hypoxic exercise in humans.
Overlapping physiological responses have been
observed after exposure to both heat and altitude,
either real or simulated. However, the magnitude and
relevance of these interactions are not fully under-
stood. The aim of this study was to determine the
effects of heat or normobaric hypoxic acclimation on
physiological, cellular, and performance responses to
hypoxic exercise.
Who and what was studied?
Twenty-one males (average age 22, with a BMI of about
23.5) participated in the study. The baseline fitness level
of the participants was good, with an average VO2
max
of 51 ml/kg/min. This means the participants were
above average compared to the general population, but
were definitely a few rungs below an elite athlete. The
participants were split into three experimental groups
(control, heat acclimation, and hypoxic acclimation)
after being matched for aerobic fitness (VO2
max) and
training experience.
The 10-day study protocol consisted of daily cycling
exercise at 50% VO2
max for 60 minutes in one of three
environments: hot conditions (40° C/ 104° F), hypoxic
conditions (equivalent to an altitude of 3000 meters/
about 9800 feet), or a room temperature control group
(18° C). Pre and post-testing sessions were performed
under hypoxic conditions and consisted of 40 minutes
of moderate intensity cycling (50% of VO max), fol-
lowed by a five minute recovery and a 16.1 kilometer
cycling time trial, which took about 40 minutes to com-
plete. Heat strain was calculated using the physiological
strain index (PSI), which uses heart rate and body
temperature to quantify physiological strain on a scale
of zero to 10. Participants arrived after an overnight
fast for all laboratory visits and were provided a stan-
dard breakfast two hours prior to exercise (386 kcal, 16
grams of protein, 44 grams of carbohydrate, 16 grams
of fat, and 400 milliliters of water).
Common terms related to altitude and exercise
Hypoxia:
deficiency in the
amount of oxygen
reaching the tissues
(which can be experi-
enced at high altitude)
Normoxia:
having normal levels of
oxygen (as experienced
at sea level)
Hypobaric:
lower than normal air
pressure (as experi-
enced at high altitude)
Normobaric:
normal barometric
pressure (as experi-
enced at sea level)
43. 43
Four different categories of outcomes were tested: (1)
Resting physiological responses to heat or hypoxic
acclimation such as heart rate, body temperature, and
SpO2
(which refers to peripheral capillary oxygen sat-
uration and gives an estimate of the amount of oxygen
in the blood); (2) Physiological responses to the perfor-
mance tests conducted under hypoxic conditions; (3)
Cellular stress responses such as monocyte HSP72 and
extracellular HIF-1α; and (4) Time trial performance
responses.
Twenty-one males completed a cycling test before and
after undergoing ten daily 60-min training sessions
in hot (40° C), hypoxic (equivalent to an altitude of
3000 meters), or control conditions. Performance
tests were conducted under hypoxic conditions, with
each participant cycling for 40 minutes at 50% of
VO₂max, followed by a five minute recovery and then
a 16.1 kilometer cycling time trial.
What were the findings?
During the pre-intervention testing, average exer-
cise heart rate was greater in both hot and hypoxic
conditions compared with the control condition.
Additionally, the hot environment induced a greater
mean exercise body temperature and corresponding PSI
compared to hypoxic and control environments. After
ten days of acclimation, an increase in resting plas-
ma volume was observed in the hot condition, while
a decrease was observed in the hypoxic group and no
changes were seen in the control group. Resting body
temperature was lower following acclimation in both
hypoxic and hot conditions.
Some of the main study findings are shown in Figure
2. Improvements in time-trial performance were seen
in both the hypoxic (-6.9%) and heat (-4.8%) accli-
mation groups, while no differences were observed in
the control group. Average heart rate was reduced by
nine beats per minute during the post-intervention test
in the heat acclimated group, but unchanged in both
the control and hypoxic groups when compared with
the pre-intervention tests. Body temperature during
exercise was also reduced after acclimation in the heat
group but not the other groups. The combination of
reduced heart rate and body temperature in the heat
acclimated group resulted in a decreased PSI during
the post-intervention test, while no changes in PSI were
Figure 2: Select results from the study
46
45
44
43
42
41
40
39
38
Pre Post
Control Hypoxia Heat
Timefor16.1km(min)
150
145
140
135
130
125
120
Pre Post
HeartRate(bpm)
Time trial performance times before and after
10-day acclimation to heat or hypoxia
Heart rate during 40 minutes of cycling at 50%
VO2
max under hypoxic conditions
Figure 2: Select results from the study
