The Effects of Diet on Testosterone Part 1: Calories and Protein
by Thomas Incledon and Lori Gross
This article will be divided into two parts. Part 1 presents an overview
of how testosterone is stimulated in the body, shows how calorie balance
affects T production, and discusses how dietary protein intake affects
circulating T levels. Part 2 explains how carbohydrates and fats impact
testosterone synthesis and circulation, and then puts it all together for
you to make informed decisions. Keep in mind that this is a very complicated
and dynamic process. References will be limited primarily to studies on men.
However, animal research will be cited when it becomes necessary to discuss
proposed mechanisms, or how the actual changes in the body take place. While
the information may get technical at times, read on because you will learn a
great deal that you may wish to apply to your own diet.
The HPT Axis
An article on the effects of diet on hormones would be incomplete without
a basic overview of the relationships between the organs and hormones of the
axis. The term axis simply refers to the pathway in question. The glands of
this pathway include the hypothalamus, pituitary, and testes. The sequence
of events culminating with the production and/or release of T begins at the
hypothalamus. Here specialized nerve cells release a hormone called
gonadotropin-releasing hormone (GnRH). GnRH is a decapeptide (chain of ten
amino acids) that travels by direct blood vessel connections to the anterior
pituitary where it stimulates the release of luteinizing hormone (LH) (1).
LH is then secreted into the blood where it attaches to receptors on the
Leydig cells of the testes. This induces activity of an enzyme, P-450scc,
referred to as the cholesterol-side-chain-cleavage enzyme (1). Through a
series of five enzymatic steps, cholesterol is converted into T.
The body regulates the circulating blood levels of T via several
mechanisms. Once in the blood, about 44% of T is bound to a protein called
either sex-hormone-binding-globulin (SHBG) or testosterone-binding globulin
(TeBG), to indicate the greater affinity for T over estradiol (E2, an
estrogen). About 54% of T is bound by albumin and other proteins, leaving 2%
to circulate unbound to any protein. This unbound T is termed free
testosterone (fT) (1). It is currently believed that only the fT or albumin
bound T are truly available to interact with the tissues of the body. The
significance of this point will be elaborated upon later in reviews of the
data from different studies. Of the T that is available to interact with
tissues, some of it binds to steroid receptors. In most tissues, like
skeletal muscle, it will directly stimulate protein synthesis. In some
tissues, like the brain and fat cells, it can be converted into E2 via the
aromatase enzyme. In other tissues, like the prostate gland, it can be
converted into dihydrotestosterone (DHT) via the 5-alpha-reductase enzyme. T
either directly or through conversion to E2 or DHT can inhibit its own
future production. The conversion to E2 or DHT can take place both in the
brain and various other tissues. E2 and T exert stronger inhibitory effects
than DHT on T production. This process is called negative-feedback
inhibition. This is the reason why the use of steroids, enzyme inhibitors,
and prohormones are far from perfect in their effects on increasing T
levels. Because it is a dynamic process, as T levels elevate in the blood, a
corresponding increase in inhibitory signals occurs. This results in the
body making less T. The opposite occurs when T levels decrease. This is a
basic overview and presented in a simplistic static fashion. The body is a
highly dynamic organism and many factors come into play to help regulate
this process. This point is made to illustrate the confounding problem that
occurs when trying to increase circulating levels of androgens.
Effects of Calorie Intake on Testosterone
Every minute of the day, someone makes a decision to lose weight. Dieting
by means of restricting calories, while not always successful, is practiced
frequently. There are some people who believe that fasting (or what we call
planned starvation) is a necessary method for cleaning the body of wastes.
What effects does depriving the body of calories have on endocrine responses
within the HPT axis? As you may have already guessed, it screws things up.
Fasting for 5 days can lower LH, T, and fT by 30-50% (2). What appears to
happen is that as the body becomes deprived of energy, less GnRH is released
from the hypothalamus. This, in turn, leads to a weaker signal to release LH.
While the pattern of LH release remains the same, the amount of LH released
at each interval decreases, meaning your body is giving weaker signals to
stimulate T. In addition, research on fasting in rats indicates that
testicular enzymes involved in synthesizing T decrease in function (3). This
means that even if enough LH reaches the testes, they still cannot produce
normal amounts of T. The decrease in T can be a contributing factor to the
loss in lean body mass that occurs with fasting. Of course, this is contrary
to what most of us want to do in the quest to get bigger and stronger.
However, many elite athletes have learned how to apply fasting to their
contest preparation. Fasting before a drug test is a common practice when on
anabolic-androgenic steroids because it helps prevent testing positive. But
before you run out and load up on some "juice" and think you’ll beat a drug
test just by fasting, keep in mind that this method is not always reliable,
nor does it work when you have foreign metabolites in the body.
One of the common problems when dieting is holding onto all that hard
earned muscle. Severe calorie restriction, whether from reduced food intake
or imposed by excessive exercise, lowers testosterone (4). While there are
no numbers written in stone, a decrease in calories by 15% does not lower T
levels (5). This may serve as one factor to consider when planning out a
diet strategy. If you cut back too much on your calories, then you risk
lowering your T, which can cause you to say goodbye to some of your muscle.
The good news is that when refeeding resumes and calorie intake equals
calorie expenditure, in most cases, T levels will rise back to normal. The
bad news is that if you are engaging in chronic high volume endurance
exercise, even extra calories won’t help raise your T levels back to normal.
When male subjects are overfed in an attempt to induce weight gain, there
tends to be a decrease in T levels as upper body fat increases (6). It may
be wise, therefore, to limit calorie intakes to less than 1000 Calories
(kilocalories) above energy requirements. From reviewing the literature, it
seems that with large short-term increases in body fat and small chronic
increases, T levels go down. Perhaps this is due to an inverse relationship
between T and insulin and/or the aromatase enzyme. It is clear that with
excessive body fat, aromatase activity in fat cells increases, thus more of
T is converted into an estrogen called estradiol (E2). The issue with
insulin is far more complicated and not really clear. Some research has
shown insulin to regulate T in a positive fashion (7), while carbohydrate
and protein liquid meals, which elevate insulin, have been shown to decrease
T in resistance- trained males (8,9). This may be due to an increased uptake
by tissues, like skeletal muscle, increased excretion of T in the urine, or
decreased responsiveness of the testes to produce T.
While not related to caloric intake, hydration and sleep status are also
important. A reduction of 3.8% in body weight due to dehydration did not
affect T levels during mild exercise (10). But, don’t take any chances with
hydration. Drink plenty of water every day at the rate of 30 cc per kilogram
of body weight (or roughly one ounce for every two pounds). Get plenty of
sleep, as disturbances in sleep and light/dark cycles can decrease T by
almost 50% (11). Of course, no one ever gets enough sleep!
Dietary Protein Intake & Testosterone
The direct impact of protein by itself on T levels has not been well
studied in humans. Some research on high protein diets deals with the
effects on very obese people and weight loss. While this may not seem
applicable to you, read on and we will put it together for you. In obese
men, feeding 600 calories a day with 400 calories from protein (50 grams of
beef protein and 50 grams of casein) induces lower levels of T than fasting
does (12). Normally, when the kidneys filter T out of the blood, some T gets
reabsorbed back out of the kidneys into the blood. The researchers stated
that the additional protein in the diet generated more ketones. They
concluded that the ketones were filtered out of the blood by the kidneys and
were reabsorbed back into circulation preferentially over T. While most
people reading this may not be obese, higher protein diets are definitely in
vogue, more so for bodybuilders and powerlifters than other groups of
athletes. The potential exists that if a ketogenic diet like the Atkins Diet
or a cyclical ketogenic diet like the Anabolic Diet or Bodyopus is followed,
than urinary excretion of T will be greater during the ketogenic phase of
It is known that protein in the diet can influence the metabolism of a
variety of chemicals. Through a series of experiments, it was demonstrated
that various foods could influence the metabolism of drugs in the body (13).
Vegetables like cabbage and brussel sprouts were found to alter the function
of specific liver enzymes. This, in turn, could change the half-life of a
drug in the blood. Given the variety of diets that people follow and the
variety of prescription medications and over-the-counter drugs people take,
the logical progression was to look at how altering the macronutrient
composition of the diet affected drug metabolism. It turns out that a higher
ratio protein diet, a diet with more calories from protein than
carbohydrates or fat, metabolizes some drugs faster, thus decreasing the
clearance time of the drug. Since diet can affect drug metabolism, perhaps
it could affect liver enzymes involved in the metabolism of endogenous
steroids. Sure enough, it was found that a high ratio protein diet decreased
the reduction of T (14). Reducing the reduction of T could mean a potential
decrease in DHT and/or androsterone in the blood, which is good by most
accounts. However, DHT levels were not measured and, more importantly,
urinary T excretion increased, although it was not statistically
significant. These subjects were not in ketosis, so perhaps ketones do not
increase T excretion rates. Regardless of the exact mechanism, there is
sufficient evidence in the literature that when protein intake exceeds
carbohydrate intake, T clearance increases by excretion in the urine.
A cross over design study used seven normal men from 23-43 years of age
and compared a high protein diet to a high carbohydrate diet (15). This
study has been referenced many times and cited as proof that high protein
diets lower total T levels in the blood. The high carbohydrate diet from
this study will be covered in Part II. The high protein diet consisted of
44% protein, 35% carbohydrate, and 21% fat and supplied between 2400 and
2500 kilocalories per day (kcals/d). Let’s assume it was an even 2450 kcals/d.
The men also had bodyweights that ranged from 64-72 kg. If we assume the
mean was 68 kg, then this would give us an average body weight of about 150
pounds. This means these guys were eating [(2450 kcals/d times .44) (divide
by 4)] 270 grams (g) of protein, [(2450 x.350 /4] 215 g of carbohydrates (CHO)
and [(2450 x .21) /9] 58 g of fat per day.
However, total T is not that big of a deal. The more important measure is
the bioactive fraction of T. (Earlier in the overview of the HPT Axis, it
was mentioned that SHBG-bound T is not considered bioactive, while the other
fractions of T are). While subjects followed the high protein diet, their
total T levels were 28% lower than on the higher CHO diet (15). This is
important because T decreased in all seven subjects, although the magnitudes
of the decrease ranged from 10 to 93%. For the same seven subjects, their
SHBG levels decreased about 39% with a range from 19 to 64%. Looking at this
data gives the impression that the actual bioactivity of T was higher while
the subjects were on a high protein diet. SHBG-bound T and fT were not
measured, so it is not known for sure. On the surface it appears that a mean
decrease of 39% in the SHBG values and only a 28% in the T would leave more
T available for binding to tissues. However, if we calculate out the actual
changes in the hormones using the data from the study, we see something
different. The mean and standard error (M±SE) for T was 371 ± 23 ng/dL. The
currently used units in clinical chemistry are nmol/L. Multiplying the mean
T by the conversion factor of 0.0347 gives us about 12.9 ± .8 nmol/L. The
M±SE SHBG was 23.4 ± 1.6 nmol/L. If we assume that the amount of T bound to
SHBG averages 44%, then .44 x 12.9 ± .8 nmol/L gives us 5.7 ± .4 nmol/L of T
bound to SHBG. That leaves 7.2 ± .4 nmol/L of T to interact with tissues in
the body. However, we don’t know from the data if the amount of SHBG bound T
decreased below or increased above the normal 44%, in which case there would
be more or less T available to interact with tissues.
From work by the same group of researchers using the exact same diet (but
different subjects) we see that the ratio of 5a - reduction to 5b -
reduction (5a /5b ) of T is reduced by about 50%, with the decrease being
attributed to lower rates of 5a - reduction (14). The T values that have
been used thus far (15) already reflect any changes in altered T metabolism,
so the conversion to a 5a - reduced hormone (ie androsterone) is accounted
for at this point. Note that even though there is a decrease in 5a - reduced
hormone production, it does not show up as increased T levels. The decrease
in androsterone probably shows up in small, but statistically insignificant
increases in other metabolites of T (they were statistically insignificant
perhaps due to the small sample size). Another interesting aspect is that
there is an increase in the oxidation of estradiol on the higher protein
diet by about 14-15% (14). Unfortunately estradiol levels were not measured
in this paper. This could have given us clues as to the mechanism by which
higher protein diets lower T (ie increased negative feedback on T levels via
estradiol). At this point, this is only one study and it is still difficult
to come to any final conclusions. However, if this is what really happens,
then a high-protein diet may actually lower the anabolic actions of T in the
body. Unfortunately, this has not been verified through laboratory research
and is just a theory at this point. Perhaps the decrease in T is a result of
increased excretion in the urine either as T or a sulfated metabolite, or
increased conversion to estradiol and oxidation by the liver.
Prelude to the Effects of Diet on Testosterone Part II: Carbohydrates
We hope so far that you have learned something about testosterone
production and the effects of calorie intake and protein intake on
testosterone levels in the blood. Please feel free to contact us if you have
any questions or comments at
[email protected] . In the next article, the effects of carbohydrates
and fat and total T levels and its components are explained. We will then
review the key points and see how the information can be integrated into a
diet strategy. At this point we would like to thank Albert Jenab for his
technical assistance and insight.
1) Griffin JE. & Ojeda SR, editors of: Textbook of
Endocrine Physiology, 3rd edition. New York, Oxford University Press,
2) Aloi JA. Bergendahl M. Iranmanesh A.
Veldhuis JD. Pulsatile intravenous gonadotropin-releasing hormone
administration averts fasting-induced hypogonadotropism and
hypoandrogenemia in healthy, normal weight men. Journal of Clinical
Endocrinology & Metabolism. 82(5):1543-8, 1997 May.
3) Fanjul LF. Ruiz de Galarreta CM. Effects of
starvation in rats on serum levels of testosterone, dihydrotestosterone
and testicular 3 beta-hydroxysteroid dehydrogenase activity. Hormone &
Metabolic Research. 13(6):356-8, 1981 Jun.
4) Marniemi J. Vuori I. Kinnunen V. Rahkila P. Vainikka
M. Peltonen P. Metabolic changes induced by combined prolonged exercise
and low-calorie intake in man. European Journal of Applied Physiology &
Occupational Physiology. 53(2):121-7, 1984.
5) Garrel DR. Todd KS. Pugeat MM. Calloway DH. Hormonal
changes in normal men under marginally negative energy balance. American
Journal of Clinical Nutrition. 39(6):930-6, 1984 Jun.
6) Pritchard J. Despres JP. Gagnon J. Tchernof A. Nadeau
A. Tremblay A. Bouchard C. Plasma adrenal, gonadal, and conjugated
steroids before and after long-term overfeeding in identical twins.
Journal of Clinical Endocrinology & Metabolism. 83(9):3277-84, 1998 Sep.
7) Pasquali R. Macor C. Vicennati V. Novo F. De lasio R.
Mesini P. Boschi S. Casimirri F. Vettor R. Effects of acute
hyperinsulinemia on testosterone serum concentrations in adult obese and
normal-weight men. Metabolism: Clinical & Experimental. 46(5):526-9, 1997
8) Kraemer WJ. Volek JS. Bush JA. Putukian M.
Sebastianelli WJ. Hormonal responses to consecutive days of
heavy-resistance exercise with or without nutritional supplementation.
Journal of Applied Physiology. 85(4):1544-55, 1998 Oct.
9) Chandler RM. Byrne HK. Patterson JG. and Ivy JL.
Dietary supplements affect the anabolic hormones after weight-training
exercise. Journal of Applied Physiology. 76(2): 839-845, 1994 Feb.
10) Hoffman JR. Maresh CM. Armstrong LE. Gabaree CL.
Bergeron MF. Kenefick RW. Castellani JW. Ahlquist LE. Ward A. Effects of
hydration state on plasma testosterone, cortisol and catecholamine
concentrations before and during mild exercise at elevated temperature.
European Journal of Applied Physiology & Occupational Physiology.
11) Cortes-Gallegos V. Sojo Aranda I. Gio Pelaez RM.
Disturbing the light-darkness pattern reduces circulating testosterone in
healthy men. Archives of Andrology. 40(2):129-32, 1998 Mar-Apr.
12) Hoffer LJ. Beitins IZ. Kyung NH. Bistrian BR.
Effects of severe dietary restriction on male reproductive hormones.
Journal of Clinical Endocrinology & Metabolism. 62(2):288-92, 1986 Feb.
13) Anderson KE. Conney AH. Kappas A. Nutrition as an
environmental influence on chemical metabolism in man. Progress in
Clinical & Biological Research. 214:39-54, 1986.
14) Kappas A. Anderson KE. Conney AH. Pantuck EJ.
Fishman J. Bradlow HL. Nutrition-endocrine interactions: induction of
reciprocal changes in the delta 4-5 alpha-reduction of testosterone and
the cytochrome P-450-dependent oxidation of estradiol by dietary
macronutrients in man. Proceedings of the National Academy of Sciences of
the United States of America. 80(24):7646-9, 1983 Dec.
15) Anderson KE. Rosner W. Khan MS. New MI. Pang SY.
Wissel PS. Kappas A. Diet-hormone interactions: protein/carbohydrate ratio
alters reciprocally the plasma levels of testosterone and cortisol and
their respective binding globulins in man. Life Sciences. 40(18):1761-8,
1987 May 4.