ArticlePDF Available

A Periodic Diet that Mimics Fasting Promotes Multi-System Regeneration, Enhanced Cognitive Performance, and Healthspan

Authors:

Abstract and Figures

Prolonged fasting (PF) promotes stress resistance, but its effects on longevity are poorly understood. We show that alternating PF and nutrient-rich medium extended yeast lifespan independently of established pro-longevity genes. In mice, 4 days of a diet that mimics fasting (FMD), developed to minimize the burden of PF, decreased the size of multiple organs/systems, an effect followed upon re-feeding by an elevated number of progenitor and stem cells and regeneration. Bi-monthly FMD cycles started at middle age extended longevity, lowered visceral fat, reduced cancer incidence and skin lesions, rejuvenated the immune system, and retarded bone mineral density loss. In old mice, FMD cycles promoted hippocampal neurogenesis, lowered IGF-1 levels and PKA activity, elevated NeuroD1, and improved cognitive performance. In a pilot clinical trial, three FMD cycles decreased risk factors/biomarkers for aging, diabetes, cardiovascular disease, and cancer without major adverse effects, providing support for the use of FMDs to promote healthspan.
Content may be subject to copyright.
Clinical and Translational Report
A Periodic Diet that Mimics Fasting Promotes Multi-
System Regeneration, Enhanced Cognitive
Performance, and Healthspan
Graphical Abstract
Highlights
dFMD rejuvenates the immune system and reduces cancer
incidence in C57BL/6 mice
dFMD promotes hippocampal neurogenesis and improves
cognitive performance in mice
dFMD causes beneficial changes in risk factors of age-related
diseases in humans
Authors
Sebastian Brandhorst, In Young Choi,
Min Wei, ..., Todd E. Morgan,
Tanya B. Dorff, Valter D. Longo
Correspondence
vlongo@usc.edu
In Brief
Brandhorst et al. develop a fasting
mimicking diet (FMD) protocol, which
retains the health benefits of prolonged
fasting. In mice, FMD improved
metabolism and cognitive function,
decreased bone loss and cancer
incidence, and extended longevity. In
humans, three monthly cycles of a 5-day
FMD reduced multiple risk factors of
aging
Brandhorst et al., 2015, Cell Metabolism 22, 1–14
July 7, 2015 ª2015 Elsevier Inc.
http://dx.doi.org/10.1016/j.cmet.2015.05.012
Cell Metabolism
Clinical and Translational Report
A Periodic Diet that Mimics Fasting
Promotes Multi-System Regeneration,
Enhanced Cognitive Performance, and Healthspan
Sebastian Brandhorst,
1,15
In Young Choi,
1,15
Min Wei,
1
Chia Wei Cheng,
1
Sargis Sedrakyan,
2
Gerardo Navarrete,
1
Louis Dubeau,
3
Li Peng Yap,
4
Ryan Park,
4
Manlio Vinciguerra,
5
Stefano Di Biase,
1
Hamed Mirzaei,
1
Mario G. Mirisola,
6
Patra Childress,
7
Lingyun Ji,
8
Susan Groshen,
8
Fabio Penna,
9
Patrizio Odetti,
10
Laura Perin,
2
Peter S. Conti,
4
Yuji Ikeno,
11
Brian K. Kennedy,
12
Pinchas Cohen,
1
Todd E. Morgan,
1
Tanya B. Dorff,
13
and Valter D. Longo
1,14,
*
1
Longevity Institute, School of Gerontology, and Department of Biological Sciences, University of Southern California, Los Angeles,
CA 90089, USA
2
GOFARR Laboratory, Children’s Hospital Los Angeles, Division of Urology, Saban Research Institute, University of Southern California,
Los Angeles, CA 90089, USA
3
Department of Pathology, Keck School of Medicine, University of Southern California, Los Angeles, CA 90089, USA
4
Molecular Imaging Center, Department of Radiology, Keck School of Medicine, University of Southern California, Los Angeles,
CA 90089, USA
5
Institute for Liver and Digestive Health, Division of Medicine, University College London, Royal Free Hospital, London NW3 2PF, UK
6
Department of Pathobiology and Medical Biotechnology, University of Palermo, 90100 Palermo, Italy
7
Global Medicine Program, Keck School of Medicine, University of Southern California, Los Angeles, CA 90089, USA
8
Department of Preventive Medicine, Keck School of Medicine, University of Southern California, Los Angeles, CA 90089, USA
9
Department of Clinical and Biological Sciences, University of Torino, 10100 Torino, Italy
10
Department of Internal Medicine, University of Genova, 16146 Genova, Italy
11
Department of Pathology, Barshop Institute, University of Texas Health Science Center, San Antonio, TX 78229, USA
12
Buck Institute for Research on Aging, Novato, CA 94945, USA
13
Norris Comprehensive Cancer Center, Keck School of Medicine, University of Southern California, Los Angeles, CA 90089, USA
14
IFOM, FIRC Institute of Molecular Oncology, 20139 Milano, Italy
15
Co-first author
*Correspondence: vlongo@usc.edu
http://dx.doi.org/10.1016/j.cmet.2015.05.012
SUMMARY
Prolonged fasting (PF) promotes stress resis-
tance, but its effects on longevity are poorly
understood. We show that alternating PF and
nutrient-rich medium extended yeast lifespan
independently of established pro-longevity genes.
In mice, 4 days of a diet that mimics fasting
(FMD), developed to minimize the burden of PF,
decreased the size of multiple organs/systems, an
effect followed upon re-feeding by an elevated
number of progenitor and stem cells and regenera-
tion. Bi-monthly FMD cycles started at middle age
extended longevity, lowered visceral fat, reduced
cancer incidence and skin lesions, rejuvenated
the immune system, and retarded bone mineral
density loss. In old mice, FMD cycles promoted
hippocampal neurogenesis, lowered IGF-1 levels
and PKA activity, elevated NeuroD1, and improved
cognitive performance. In a pilot clinical trial,
three FMD cycles decreased risk factors/bio-
markers for aging, diabetes, cardiovascular dis-
ease, and cancer without major adverse effects,
providing support for the use of FMDs to promote
healthspan.
INTRODUCTION
Dietary composition and calorie level are key factors affecting
aging and age-related diseases (Antosh et al., 2011; Blagos-
klonny et al., 2009; Fontana et al., 2010; Gems and Partridge,
2013; Lo
´pez-Otı
´n et al., 2013; Tatar et al., 2003). Dietary restric-
tion (DR) promotes metabolic and cellular changes that affect
oxidative damage and inflammation, optimize energy meta-
bolism, and enhance cellular protection (Haigis and Yankner,
2010; Johnson et al., 2000; Lee et al., 2012b; Longo and Finch,
2003; Mair and Dillin, 2008; Narasimhan et al., 2009; Smith
et al., 2008). Fasting, the most extreme form of DR, which entails
the abstinence from all food, but not water, can be applied in a
chronic manner as intermittent fasting (IF) or periodically as cy-
cles of prolonged fasting (PF) lasting 2 or more days (Longo
and Mattson, 2014). In rodents, IF promotes protection against
diabetes, cancer, heart disease, and neuro-degeneration (Longo
and Mattson, 2014). In humans, IF and less-severe regimens
(e.g., consumption of approximately 500 kcal/day for 2 days a
week) have beneficial effects on insulin, glucose, C-reactive pro-
tein, and blood pressure (Harvie et al., 2011).
PF cycles lasting 2 or more days, but separated by at least a
week of a normal diet, are emerging as a highly effective strategy
to protect normal cells and organs from a variety of toxins and
toxic conditions (Raffaghello et al., 2008; Verweij et al., 2011)
while increasing the death of many cancer cell types (Lee et al.,
2012a; Shi et al., 2012). PF causes a decrease in blood glucose,
Cell Metabolism 22, 1–14, July 7, 2015 ª2015 Elsevier Inc. 1
Please cite this article in press as: Brandhorst et al., A Periodic Diet that Mimics Fasting Promotes Multi-System Regeneration, Enhanced Cognitive
Performance, and Healthspan, Cell Metabolism (2015), http://dx.doi.org/10.1016/j.cmet.2015.05.012
insulin, and insulin-like growth factor 1 (IGF-1) (Lee et al., 2010)
and is accompanied by autophagy (Cuervo et al., 2005; Madeo
et al., 2010). Recently, we have shown that PF causes a major
reduction in the levels of white blood cells followed by stem-
cell-based immune system regeneration upon refeeding (Cheng
et al., 2014). Others have reported on the role of PF in causing ma-
jor decreases in liver and body mass in rats (Wasselin et al., 2014).
However, prolonged water-only fasting is difficult for the great
majority of the population, and its extreme nature could cause
adverse effects, which include the exacerbation of previous mal-
nourishments and dysfunctions, particularly in old and frail sub-
jects. These concerns point to the need for dietary interventions
that induce PF-like effects while minimizing the risk of adverse
effects and the burden of complete food restriction.
Here we identified a diet that mimics the effects of fasting (fast-
ing mimicking diet, FMD) on markers associated with the stress
resistance caused by PF, including low levels of glucose and
IGF-1 and high levels of ketone bodies and IGFBP-1 (Longo and
Mattson, 2014). We tested the hypothesis that cycles of the
FMD lasting 4 days followed by a standard ad libitum diet could
promote healthspan in mice. Additionally, we tested the effects
of three cycles of a similar FMD in a pilot randomized clinical study
with 38 subjects, 19 of whom were assigned to the FMD group.
RESULTS AND DISCUSSION
Periodic Fasting in S. cerevisiae Extends Lifespan and
Induces Stress Resistance
To determine whether the benefits of periodic starvation can be
achieved in a simple organism, we tested the effects of cycles
of prolonged fasting (PF) in S. cerevisiae. PF was implemented
by switching wild-type yeast cells back and forth from nutrient-
rich medium to water every 48 hr. This duration was selected
to match the length of fasting shown to be effective in mice,
but also to allow cells to undergo at least 4 cycles of PF within
its lifespan. PF cycles extended both medium and maximum
chronological lifespan (Figures 1A and 1B) and increased the
number of yeast cells that survive hydrogen peroxide treatment
by more than 100-fold (Figure 1C). Surprisingly, the deletion of
the serine threonine kinase Rim15, or of its downstream stress
response transcription factors Msn2/4 and Gis1, well estab-
lished to be important or essential for longevity extension by ge-
netic and dietary interventions (Fabrizio et al., 2001; Wei et al.,
2008), did not prevent the lifespan effects of PF (Figures 1A
and 1B). These results indicate that PF can protect simple organ-
isms from both toxins and aging by mechanisms that are in part
independent of conserved pro-longevity transcription factors, in
agreement with findings in C. elegans that complete deprivation
of food does not require the stress response transcription factor
DAF-16, analogue of yeast Msn2/4 and Gis1 (Greer and Brunet,
2009; Kaeberlein et al., 2006).
Periodic FMD in Aged Mice
Periodic FMD without an Overall Reduction in Calorie
Intake Promotes Visceral Fat Loss
We developed a very low calorie/low protein fasting mimicking
diet (FMD) that causes changes in markers associated with
stress resistance or longevity (IGF-1, IGFBP-1, ketone bodies,
and glucose) that are similar to those caused by fasting (Table
S1). Mice were fed the FMD starting at 16 months of age for
4 days twice a month and were fed an ad libitum diet in the period
between FMD cycles. Mice on the control diet reached maximum
weight (36.6 ± 5.2 g) at 21.5 months of age, whereas those in the
FMD group lost 15% weight during each FMD cycle but re-
gained most of the weight upon re-feeding (Figure S1A). How-
ever, FMD group mice maintained a constant weight between
16 and 22 months and then gradually lost weight (Figure 1D).
Although FMD group mice were severely calorically restricted
during the diet, they compensated for this restriction by over-
eating during the ad libitum period, resulting in a 14-day cumula-
tive calorie intake equivalent to that of the ad libitum groups
(Figures 1E and S1B). The average caloric intake in both cohorts
increased after 25 months of age (Figure 1E).
At the end of the FMD and before re-feeding, blood glucose
levels were 40% lower than those in the control diet group.
Throughout the study, glucose returned to normal levels within
7 days of re-feeding (Figure S1C). Ketone bodies increased
9-fold by the end of the FMD but returned to normal levels after
re-feeding (Figure S1D). Serum insulin levels were reduced
10-fold after 4 days of the FMD and returned to baseline levels
after re-feeding (Figure S1E). Reduced signaling of the growth
hormone/IGF-1 axis extends health- and lifespan in rodents
(Brown-Borg, 2009; Guarente and Kenyon, 2000; Harrison
et al., 2009; Junnila et al., 2013; Wullschleger et al., 2006).
IGF-1 was reduced by 45% by the end of the FMD period
but returned to normal levels, even after multiple FMD cycles
(Figure S1F). IGFBP-1, which inhibits IGF-1, increased 8-fold
by the end of the FMD regimen, but its concentration returned
to levels similar to those for ad libitum mice within 1 week of
re-feeding (Figure S1G).
To investigate diet-induced body composition changes, we
evaluated lean body mass and body fat localization by microCT.
At 28 months, FMD group mice showed a trend (p = 0.06) for
reduced total adipose tissue measured during the ad libitum
diet period between cycles (Figure 1F). Although subcutaneous
adipose tissue volume (Figures 1G, 1J, and 1K; gray area) was
not affected, visceral fat deposits (Figures 1H, 1J, and 1K; red
area) were reduced in the FMD group compared to control group
mice (p < 0.05). Lean body mass remained similar in the two
groups (Figure 1I). These results indicate that FMD cycles can
have profound effects on visceral fat, glucose, and IGF-1 levels,
but in mice the latter changes are reversed by the return to the ad
libitum diet.
Reduced Organ Size and Regeneration
FMD (20.5 months), FMD-RF (7 days after resuming the ad libi-
tum diet post-FMD; 20.5 months), and ad libitum-fed (16 and
20.5 months) mice were euthanized, and organ weights were
measured. At the end of the FMD, we observed a reduction in or-
gan weight in kidneys, heart, and liver (Figures 1L–1N), but not in
the lungs, spleen, and brain (Figures S1L and S1M), and a reduc-
tion in body weight (Figure S1H–S1J). The weights of these or-
gans returned to pre-FMD levels after re-feeding.
The chronic use of bi-weekly FMD cycles caused no differ-
ences in systolic and diastolic left ventricular volume, ejection
fraction, and left ventricular mass, as measures of cardiac func-
tion in 25-month-old mice (Figures S1N–S1Q). Serum alanine
transaminase, a liver atrophy marker, increased at the end
of the FMD but returned to control levels upon re-feeding
2Cell Metabolism 22, 1–14, July 7, 2015 ª2015 Elsevier Inc.
Please cite this article in press as: Brandhorst et al., A Periodic Diet that Mimics Fasting Promotes Multi-System Regeneration, Enhanced Cognitive
Performance, and Healthspan, Cell Metabolism (2015), http://dx.doi.org/10.1016/j.cmet.2015.05.012
(Figure S1R). Following re-feeding, liver cells repopulated in
proximity to the hepatic blood vessels (Figure 1O 4, arrow).
The effect of the FMD on hepatic regeneration 24 hr post re-
feeding was supported by a 10-fold induction of a marker for he-
patic cellular proliferation (Ki67), which is absent in G
0
cells (Fig-
ures 1P and S1S). Ki67 remained elevated for at least 3 days
post-FMD. Renal function, assessed by serum creatinine and
blood urea nitrogen measurements, revealed no alterations (Fig-
ures S1T and S1U) (Schnell et al., 2002). Renal histology, to eval-
uate glomerular and interstitial fibrosis, also showed no change
in the number of sclerotic glomeruli (Figure S1V). These data
are supportive of hepatic regeneration as a consequence of
FMD-re-feeding cycles and with the absence either liver or kid-
ney toxicity even after 4 months on the FMD.
Figure 1. Periodic FMD Promotes a Lean Bodyweight, Improves Healthspan, and Promotes Tissue Regeneration
(A and B) Periodic fasting (PF, alternating cycles of SDC media and water) prolongs lifespan in wild-type (WT) S. cerevisiae (DBY746) and rim15D(A) and msn2D
msn4Dgis1DDBY746 (B) mutants.
(C) PF induces cellular stress resistance against hydrogen peroxide in S. cerevisiae (DBY746).
(D) Mouse body weight profile. Dotted lines represent FMD cycles.
(E) Consumed kcal/g of bodyweight.
(F–I) Total adipose tissue (TAT) (F), subcutaneous adipose tissue (SAT) (G), visceral adipose tissue (VAT) (H), and lean body mass (I) at 28 months of age.
n = 3/group.
(J and K) Representative images of the SAT (gray) (J) and VAT (red) (K) in the lumbar L3 region.
(L–N) Kidney (L), heart (M), and liver (N) weight as percentage change. n = 8–10/group.
(O) Liver H&E staining of control (1, 2) and FMD mouse at the end of the FMD regimen (3) or 24 hr after re-feeding (4). Unorganized cells (arrow) indicate liver
repopulation. 1, 3: 403magnification; 2, 4: 203magnification.
(P) Hepatic proliferative index (Ki67
+
) after 1, 3, and 7 days of refeeding compared to control. n = 3–4/group.
(Q and R) Pax7 (Q) and p62 (R) protein expression level. n = 3–4/group.
(S) Tissue mineral density (mg Hydroxyapatite/cm
3
) of the femur. n = 5/group.
All data are expressed as the mean ± SEM.
Cell Metabolism 22, 1–14, July 7, 2015 ª2015 Elsevier Inc. 3
Please cite this article in press as: Brandhorst et al., A Periodic Diet that Mimics Fasting Promotes Multi-System Regeneration, Enhanced Cognitive
Performance, and Healthspan, Cell Metabolism (2015), http://dx.doi.org/10.1016/j.cmet.2015.05.012
Postnatal growth and regeneration of the skeletal muscle re-
quires myogenic precursors termed satellite cells (Sinha et al.,
2014). Pax7 expression is critical for satellite cell biogenesis,
survival, and self-renewal (Olguin et al., 2007), whereas the
myogenic transcription factors MyoD and MyoG promote mus-
cle development and differentiation (Perry and Rudnick, 2000).
Pax7 upregulation and reduced MyoD expression is observed
in undifferentiated myogenic cells (Olguin et al., 2007). An age-
dependent decline in Pax7 (Figure 1Q) and MyoD (and less pro-
nounced in MyoG) was detected in 20-month-old mice (Figures
S1W and S1X). At the end of the FMD, Pax7 expression was
reduced by 40% compared to that in control animals. A similar
trend was also observed for MyoG (p = 0.074). 1 week after re-
feeding, Pax7 expression in 20-month-old FMD group animals
reached levels similar to those in 12-month-old ad libitum fed
animals (Figure 1Q). By contrast, MyoD expression in old animals
was not altered by the FMD (Figures S1W and S1X). Taken
together, these changes are consistent with muscle regeneration
and rejuvenation upon re-feeding, although further analyses
similar to those performed for the hematopoietic and nervous
systems (see below) are necessary to confirm this hypothesis
and determine the mechanisms responsible for it. Heterochronic
parabiosis has been shown to increase the proliferative index of
aged hepatocytes, as well as the proliferative and regenerative
capacity of aged muscle satellite cells, and to promote adult neu-
rogenesis in an age-dependent fashion in mice (Conboy and
Rando, 2012; Villeda et al., 2011). One of the proteins that has
been implicated in muscle and brain regeneration, and which
may contribute to regenerative effects in multiple systems, is
GDF11 (Katsimpardi et al., 2014; Sinha et al., 2014). It will be
interesting to determine if part of the rejuvenating effect of the
FMD may involve factors including or related to GDF11.
The failure to induce autophagy contributes to cellular dam-
age, carcinogenesis, and aging (Cuervo et al., 2005). Autophagy
can be monitored by indirectly measuring autophagic sequestra-
tion (LC3) and degradation (p62) (Moscat and Diaz-Meco, 2011)
(Figure S1Y and S1Z). p62 is consistently increased in auto-
phagy-deficient cells (Komatsu et al., 2007). An age-dependent
increase in muscle p62 was observed in 20-month-old mice
from the ad libitum, but not FMD, group (Figure 1R), indicating
that the FMD and possibly the associated regeneration protects
muscle cells from age-dependent functional decline, including
the ability to maintain normal expression of autophagy proteins.
Tissue mineral density in both femora decreased in 28-month-
old C57BL/6 mice compared to that in 12-month-old mice (Fig-
ure 1S), in agreement with previously published data (Shen
et al., 2011). At 28 months, femoral bone density was higher in
the FMD group compared to that in the control diet group (Fig-
ure 1S), indicating that FMD cycles either attenuated age-depen-
dent bone mineral density loss or induced bone regeneration.
Cancer and Inflammation
C57BL/6 mice are prone to hematopoietic tumors and mainly
malignant lymphomas (Blackwell et al., 1995). Subcutaneous
and internal masses caused by neoplasia, abscesses, or both
were detected in aging mice (Figures 2A–2H and S2A–S2D).
Necropsies indicated a 45% reduction in neoplasia incidence
in the FMD group compared to that in the control group (Fig-
ure 2I). By the end of life, lymphomas affected 67% of control
mice, but only 40% of mice in the FMD group (Figure 2J),
although the FMD did not cause a shift in the type of neoplasms.
Notably, the FMD also postponed the occurrence of neoplasm-
related deaths by over 3 months, from 25.3 ± 0.66 months in the
controls to 28.8 ± 0.72 months of age in the FMD cohort (p =
0.003) (Figure 2K). Furthermore, necropsies revealed that the
number of animals with multiple (3 or more) abnormal lesions
was more than 3-fold higher in the control than in the FMD group
(p = 0.0067; Fisher’s exact test) (Figure 2L). Therefore, the cycles
of the FMD started at middle age reduced tumor incidence, de-
layed their onset, and caused a major reduction in the number of
lesions, which may reflect a general switch from malignant to
benign tumors.
Inflammation can play a key role in the development of many
age-related diseases including cancer (Bartke et al., 2013; Mor-
gan et al., 2007). Pathological analysis showed a reduced num-
ber of tissues with inflammation (e.g., reactive lymph nodes or
chronic hepatic inflammation, Table S2) in the FMD mice
compared to those in the control group (Figure 2M). One of the
inflammatory conditions observed in C57BL/6 mice is severe ul-
cerating dermatitis (Figure 2N). Control animals had an 20%
incidence of progressing skin lesions that required animal sacri-
fice in contrast to the 10% incidence for mice in the FMD-fed
group. These results indicate that the FMD protects against
inflammation and inflammation-associated skin lesions (Coppe
´
et al., 2010).
Effects of the FMD on Immunosenescence and
Bone-Marrow-Derived Stem and Progenitor Cells
The age-associated decline in hematopoiesis causes a dimin-
ished or altered production of adaptive immune cells, a phenom-
enon known as ‘‘immunosenescence,’’ manifested as a shift in
the lymphoid-to-myeloid ratio and elevated incidence of anemia
and myeloid malignancies (Figures 2O–2S) (Muller-Sieburg et al.,
2004; Shaw et al., 2010). Complete blood counts indicated that
the FMD causes a rejuvenation of the blood profile (Figures
2O–2S; Figures S2E–S2R; Table S3) and a reversal of the age-
dependent decline in the lymphoid-to-myeloid ratio (L/M) (Fig-
ure 2 P), as well as of the age-dependent decline in platelets,
and hemoglobin (Figures 2Q–2S). Also, 4 months of FMD cycles
resulted in an increase in red blood cell count and hemoglobin
levels compared to baseline (Figures 2Q–2S). We also measured
a panel of 23 cytokines but did not detect changes except for
elevated IL-12 and RANTES, as well as reduced GM-CSF in
the FMD group (Figures S2S–S2U). These results indicate that
chronic use of the FMD promotes immune system regeneration
and rejuvenation, in agreement with our previous results on the
effect of fasting on lymphocyte number (Cheng et al., 2014).
Among the bone marrow-derived stem cells, hematopoietic
stem cells and mesenchymal stem cells represent a potential
source for adult tissue and organ regeneration. To investigate
whether the rejuvenating effects of the FMD may involve stem
cells, we measured hematopoietic (HSPC, lin
Scal-1
+
C-kit
+
CD45
+
) and mesenchymal (MSPC, lin
Scal-1
+
CD45
)stemand
progenitor cells in the bone marrow. The number of HSPCs is
known to increase with age, possibly to compensate for a reduc-
tion in function (Geiger and Van Zant, 2002; Morrison et al., 1996).
This age-dependent increase may mask the effects of fasting
or FMD in promoting stem cell self-renewal, which we have
recently shown for younger mice (Figure S2V) (Cheng et al.,
2014). Unlike that of HSPCs, the number of MSPCs declines
4Cell Metabolism 22, 1–14, July 7, 2015 ª2015 Elsevier Inc.
Please cite this article in press as: Brandhorst et al., A Periodic Diet that Mimics Fasting Promotes Multi-System Regeneration, Enhanced Cognitive
Performance, and Healthspan, Cell Metabolism (2015), http://dx.doi.org/10.1016/j.cmet.2015.05.012
with age (Bellantuono et al., 2009; Kasper et al., 2009). We
confirmed this age-dependent decline comparing MSPC number
in mature (8–10 months) and 20.5-month-old mice (Figure 2T), in
agreement with previous reports (Kasper et al., 2009; Ratajczak
et al., 2008). The number of MSPCs increased 5-fold in the FMD
cohort (469.8 ± 179.5 FMD versus 95.5 ± 16.7 CTRL; Figures 2T
Figure 2. Periodic FMD Cycle Reduces and Delays Cancer, Rejuvenates the Hematopoietic System, and Induces Mesenchymal Stem/Pro-
genitor Cells
(A) Hepatic lymphomatous nodules (bar, 400 microns).
(B–D) Lymphoma in the renal medulla (bar, 100 microns) (B), in a mesenteric lymph node (bar, 100 microns) (C), and in the spleen (bar, 100 microns) (D).
(E) Hepatic lymphoma containing atypical cells with abnormal DNA (circle) and mitosis (arrows, bar, 100 microns).
(F and G) Subcutaneous fibrosarcoma in relationship to the epidermis (F) and with invasion into the skeletal muscle tissue (G).
(H) Cytological details (bar, 100 microns).
(I) Autopsy-confirmed neoplasms.
(J) Lymphoma incidence.
(K) Neoplasms in relationship to the onset (arrow) of the FMD diet.
(L) Number of animals with 0 to greater than 5 abnormal lesions determined at autopsy.
(M) Inflammatory incidence.
(N) Dermatitis incidence in percentage. Images show progression of dermatitis.
(O–S) The number of white blood cells (O), the lymphoid:myeloid ratio (P), as well as the number of platelets (Q), red blood cells (R), and hemoglobin (S) are shown.
n = 7–12/group. Other complete blood count parameters are summarized in Figure S2 and Table S3.
(T) lin
Scal-1
+
CD45
mesenchymal stem/progenitor cells (MSPC) in bone marrow cells from control mature (M, 8–10 month), old (O, 20.5 month), and FMD mice
7 days after refeeding (FMD-RF; 20.5 month). n = 4–5/group.
All data are expressed as the mean ± SEM.
Cell Metabolism 22, 1–14, July 7, 2015 ª2015 Elsevier Inc. 5
Please cite this article in press as: Brandhorst et al., A Periodic Diet that Mimics Fasting Promotes Multi-System Regeneration, Enhanced Cognitive
Performance, and Healthspan, Cell Metabolism (2015), http://dx.doi.org/10.1016/j.cmet.2015.05.012
and S2W), and that of BrdU
+
MSPCs increasedby 45-fold in FMD-
treated mice (69.8 ± 34.0 FMD versus 1.5 ± 0.6 CTRL) (Figures 2T
and S2X). Taken together, these data suggest that cycles of FMD
are effective in promoting increases in hematopoietic and mesen-
chymal stem and progenitor cells, which are likely to contribute to
the regeneration of various cell types/systems.
Effects of the FMD on Motor Coordination, Memory,
and Neurogenesis
Aging is associated with the decline in locomotor and cognitive
function (Lynch, 2004). To evaluate motor coordination and bal-
ance, we tested mouse performance on the accelerating rotarod
(Shiotsuki et al., 2010). 23-month-old mice fed the FMD every
2 weeks (FMD-RF, tested 1 week after resuming the normal
diet) were able to stay longer on the rotarod than mice in the con-
trol diet group (Figure 3A). We also assessed motor learning abil-
ity by examining performance improvement during subsequent
trials. The mice from the FMD-RF group performed consistently
better by staying on the accelerating rod longer than mice on
the ad libitum diet, although the rate of learning was similar in
the two groups (sessions 2–5; Figure 3B). Mouse body weight
and best rotarod performance were negatively associated (Pear-
son correlation coefficient r = 0.46; p = 0.005). When corrected
for weight, rotarod performance improvement was no longer sig-
nificant (p = 0.34; data not shown), indicating that the FMD mice
benefit from the fat loss.
To test the effect of the diet on cognitive performance, we
carried out working memory tests (Beninger et al., 1986)at
23 months of age (Figure 3C). Mice in the FMD cohort displayed
enhanced spontaneous alternating behavior compared to con-
trol mice, with no difference in the total number of arm entries
(a measure of activity) (Figure S3A). Short-term cognitive perfor-
mance and context-dependent memory were assessed with the
novel object recognition test (Figures 3D and 3E) (Bernabeu
et al., 1995). FMD mice had a higher recognition index (RI =
0.60) compared to controls (RI = 0.52; p < 0.01) (Figure 3D). An
increase in exploration time was observed for the FMD mice
for the new object, while the total exploration time remained
the same (13.6 ± 0.9 CTRL versus 13.4 ± 0.9 FMD-RF), suggest-
ing enhanced short-term cognitive performance, not general ac-
tivity (Figure 3E; Figure S3B).
Figure 3. Periodic FMD Cycle Improves Motor Coordination, Hippocampal-Dependent Learning, and Short- and Long-Term Memory
(A) Best rotarod performance score at 23 months. n = 18/group.
(B) Rotarod performance as linear regression for each cohort (dashed lines). n = 18/group.
(C) Spontaneous alternation behavior (SAB) at 23 months. n = 11/group.
(D) Recognition index at 23 months in the novel object recognition task.
(E) Exploration time of the old versus novel object (New, dashed bar). n = 8/group.
(F–I) Error number (F), deviation (G), latency (H), and success rate (I) in the Barnes maze at 23 months. n = 7–12/group.
(J and K) Control (J) and FMD-RF (K) strategies used to locate escape box.
All data are expressed as the mean ± SEM.
6Cell Metabolism 22, 1–14, July 7, 2015 ª2015 Elsevier Inc.
Please cite this article in press as: Brandhorst et al., A Periodic Diet that Mimics Fasting Promotes Multi-System Regeneration, Enhanced Cognitive
Performance, and Healthspan, Cell Metabolism (2015), http://dx.doi.org/10.1016/j.cmet.2015.05.012
As a measure of long-term memory, we measured spatial
learning and memory using the Barnes maze: a hippocampus-
dependent cognitive task requiring spatial reference memory
to locate a unique escape box by learning and memorizing visual
clues (Figures 3F–3K) (Barnes, 1988). During the 7-day training
period, FMD mice performed better with regard to errors, devia-
tion, latency, and success rate compared to controls (Figures
3F–3I). In the retention test, the FMD group displayed better
memory indicated by reduced deviation at day 14 (Figure 3G).
Deviation of control diet mice at day 14 was similar to that
at day 1, indicating that these mice did not remember the box
location they had learned by day 7. Improvements in the search
strategy, including the shifting from a random and serial search
strategy to spatial strategies, were observed for the FMD, but
not the control diet group after days 3–4 (Figures 3J and 3K).
Together, the behavioral tests suggests that FMD cycles
improve motor learning and hippocampus-dependent short-
and long-term memory in old animals.
Adult neurogenesis plays an important role in learning and
memory (Clelland et al., 2009; Deng et al., 2010; Mattson,
2012). To determine whether the diet affected neurogenesis,
we measured BrdU incorporation in the subgranular layer of
control mice at the age of 8 weeks, 12 weeks, 6 months, and
24 months (Figure 4B). Similarly to previously reported data,
we observed an age-dependent decline in BrdU incorporation
in the dentate gyrus (Lee et al., 2012c)(Figure 4B). To assess
whether the cognitive improvements in the FMD group are asso-
ciated with neural regeneration, we measured the proliferative in-
Figure 4. Periodic FMD Cycle Promotes
Adult Neurogenesis
(A) Hippocampal immunohistochemistry of control
(top row) and FMD (bottom row, see Experimental
Procedures for details)-fed 23-month-old animals
for BrdU (left, green), DCX (middle, red), and
BrdU
+
DCX
+
(right).
(B) Age-dependent BrdU
+
cell counts in
sub-granular zone of the dentate gyrus (DG)
(n = 4/group).
(C) BrdU
+
cells in the DG at the end of the FMD
(n = 4/group).
(D) DCX
+
staining in the DG in 23-month-old ani-
mals (n = 4/group).
(E) Percentage of double-positive BrdU
+
DCX
+
cells in the DG (n = 4/group).
(F) Hippocampal IGF-1 level after FMD (n =
3/group).
(G) IGF-1R mRNA level in the DG (n = 3/group).
(H) PKA activity level in the DG (n = 5/group).
(I) NeuroD1 mRNA level in the DG (n = 3/group).
All data are expressed as the mean ± SEM.
dex of DCX
+
immature neurons in the
sub-granular cell layer of the dentate gy-
rus. BrdU
+
or BrdU
+
DCX
+
double-label-
ing indicated an increased proliferation
of immature neurons in the FMD group
compared to that in controls (Figures
4C–4E). To investigate mechanisms of
FMD-induced neurogenesis, we fed
6-month-old mice, in which cellular prolif-
eration in the dentate gyrus is reduced by more than 50%
compared to that in 8-week-old mice (Figure 4B), with a single
cycle of the FMD. After 72 hr on the FMD, we observed a reduc-
tion in circulating (Figure S1E) and hippocampal IGF-1 (Figure 4F)
but increased IGF-1 receptor mRNA expression in the dentate
gyrus region of the hippocampal formation (Figure 4G). Micro-
dissected dentate gyrus-enriched samples from FMD mice
displayed a major reduction in PKA activity (Figure 4H) and a
2-fold induction in the expression of NeuroD1 (Figure 4I), a tran-
scription factor important for neuronal protection and differenti-
ation (Gao et al., 2009). Similarly, a single cycle of the FMD
increased radial glia-like cells (type I) and non-radial precursor
(type II) neural stem cells (Figures S4B, S4C, S4F, and S4G),
immature neurons (Figures S4D and S4I–S4Q), and the
dendrite-covered area (Figures S4E and S4H) in CD-1 mice.
These results in two genetic backgrounds indicate that the
FMD promotes neurogenesis in adult mice. Notably, the brain
did not undergo a measurable weight reduction during the
FMD, indicating that regeneration can also occur independently
of the organ size increase after refeeding. Thus, we hypothesize
that alterations in circulating factors, such as the reduction in
IGF-1 levels and PKA signaling, can induce pro-regenerative
changes that are both dependent and independent of the major
cell proliferation that occurs during re-feeding, in agreement with
our previous finding in bone marrow and blood cells (Cheng
et al., 2014). Most likely, the increase in IGF-1 and PKA after re-
feeding also contributes to the proliferative and regenerative
process, raising the possibility that both low and high levels of
Cell Metabolism 22, 1–14, July 7, 2015 ª2015 Elsevier Inc. 7
Please cite this article in press as: Brandhorst et al., A Periodic Diet that Mimics Fasting Promotes Multi-System Regeneration, Enhanced Cognitive
Performance, and Healthspan, Cell Metabolism (2015), http://dx.doi.org/10.1016/j.cmet.2015.05.012
these proteins can promote regeneration depending on the
timing of their expression. Alternatively, the FMD may increase
survival of newly differentiated neurons, as observed in the den-
tate gyrus of alternate day-fed rodents (Lee et al., 2002; Mattson
et al., 2001). The observed improvements in cognitive perfor-
mance in the FMD cohort might be affected by a PKA/CREB-
dependent regulation of NeuroD1 (Cho et al., 2012; Sharma
et al., 1999), which is known to increase neuronal survival and
differentiation of hippocampal progenitors (Roybon et al.,
2009), enhance functional integration of new neurons, and alle-
viate memory deficits in a mouse model of Alzheimer’s disease
(Richetin et al., 2015).
FMD and Lifespan
Control mice had a median lifespan of 25.5 months (Figure 5A),
which was extended to 28.3 months (11% extension) in the
FMD group (p < 0.01). The FMD showed an 18% extension effect
at the 75% survival point, but only a 7.6% extension effect on the
25% survival point and no effect on maximum lifespan (Figures
5A and 5B), indicating that at very advanced ages the 4-day
FMD may be beneficial for certain aspects and detrimental for
others. Further analysis indicated that many deaths at very old
ages occurred during or shortly (within 3 days) after completion
of the FMD cycle (Figure 5E, asterisk). Based on this observation,
at 26.5 months we shortened the FMD diet from 4 to 3 days and
halted the FMD diet completely at 29.5 months. Analyses of the
data indicate that whereas the shortening of the FMD from 4 to
3 days was associated with reduced mortality rates between
26.5 and 29.5 months, the halting of the FMD diet at 29.5 months
did not reduce mortality further (Figure 5D). These results sug-
gest that FMD cycles can have a potent effect on lifespan and
healthspan, but, at least for very old mice, a less-severe (3 versus
Figure 5. Periodic FMD Cycle Increases
Median Lifespan, but Does Not Affect
Maximum Lifespan
(A) Kaplan-Meier survival curve for control and
FMD cohort (n = 46 and 29, respectively).
(B) Overview for onset of death, 75%, median,
25%, and maximum lifespan in months with
percent change.
(C) Cumulative incidence rates of deaths associ-
ated with neoplasia.
(D) Cumulative incidence rates of deaths not
associated with neoplasia.
(E) Overview over the date of death not associated
with neoplasms. The change from the 4-day FMD
to 3-day FMD is indicated by the green shaded
area at 26.6 months. The stop of the 3 day FMD
and switch to the ad libitum control diet after 6
FMD cycles is indicated by the white shaded area.
Numbers over the red squares indicate the num-
ber of animals deceased on the particular date;
asterisk indicates that death during the FMD
regime or within 3 days of refeeding.
All data are expressed as the mean ± SEM.
4 days) low-calorie and low-protein diet
may be preferable to continue to provide
beneficial effects while minimizing
malnourishment, in agreement with our
recent work demonstrating opposite
roles of high protein intake on health/mortality in mice and
humans of middle to old and very old ages (Levine et al., 2014).
Periodic FMD in a Pilot Randomized Clinical Trial
Markers of Aging and Diseases
To evaluate the feasibility and potential impact of a periodic low-
protein and low-calorie FMD in humans, we conducted a pilot
clinical trial in generally healthy adults. The components and
levels of micro- and macro-nutrients in the human FMD were
selected based on their ability to reduce IGF-1, increase
IGFBP-1, reduce glucose, increase ketone bodies, maximize
nourishment, and minimize adverse effects (Figure 6) in agree-
ment with the FMD’s effects in mice (Figure S1). The develop-
ment of the human diet took into account feasibility (e.g., high
adherence to the dietary protocol) and therefore was designed
to last 5 days every month and to provide between 34% and
54% of the normal caloric intake with a composition of at least
9%–10% proteins, 34%–47% carbohydrates, and 44%–56%
fat. Subjects were randomized either to the FMD for 5 days every
month for 3 months (3 cycles) or to a control group in which they
continued to consume their normal diet (Figure 6A). Subjects
were asked to resume their normal diet after the FMD period
and were asked to not implement any changes in their dietary
or exercise habits. 5% of the subjects were disqualified due to
non-compliance to the dietary protocol. 14% of the enrolled sub-
jects withdrew from the study due to non-diet-related reasons
(e.g., work- and travel-related scheduling issues). We present
results of the pilot randomized clinical trial that includes a set
of 19 participants who successfully completed 3 FMD cycles,
as well as data for 19 participants who were randomized to
continue on their normal diet and serve as controls. The control
8Cell Metabolism 22, 1–14, July 7, 2015 ª2015 Elsevier Inc.
Please cite this article in press as: Brandhorst et al., A Periodic Diet that Mimics Fasting Promotes Multi-System Regeneration, Enhanced Cognitive
Performance, and Healthspan, Cell Metabolism (2015), http://dx.doi.org/10.1016/j.cmet.2015.05.012
group included 9 females (47.4%) and 10 males (52.6%) with an
average age of 35.4 ± 5.5 years and 38.0 ± 1.7 years, respec-
tively. The FMD cohort included 7 females (36.8%) and 12 males
(63.2%) with an average age of 41.8 ± 4.9 years and 42.5 ± 3.5
years, respectively (Figures S5A and S5B). The age range was
19.8–67.6 years for the control cohort and 27.6–70 years for
the FMD cohort. The ethnicity was 58% White, 18.5% Hispanic,
18.5% Asian, and 5% Black (Figure S5C). Subjects were evalu-
ated by a baseline examination (Figure 6A). For the FMD group,
the follow-up examinations occurred before resuming normal
food intake at the end of the first FMD cycle (FMD) and after
5–8 days of normal dieting following the third FMD cycle (FMD-
RF, Figure 6A). The average time between the baseline and the
FMD-RF assay/measurement points was 75.2 ± 2.7 days,
whereas the time between baseline and the final examination
was 74.5 ± 6.0 days in the control group. For all three FMD cy-
cles, study participants self-reported adverse effects following
Common Terminology Criteria for Adverse Events (Figure S5D).
Adverse effects were higher after completion of the first FMD cy-
cle compared to those during the second and third FMD cycles.
However, the average reported severity of the side effects was
very low and below ‘‘mild’’ (<1 on a scale of 1–5).
In the FMD subjects, fasting blood glucose levels were
reduced by 11.3% ± 2.3% (p < 0.001; FMD) and remained
5.9% ± 2.1% lower than baseline levels after resuming the
normal diet following the third FMD cycle (p < 0.05; Figure 6B).
Serum ketone bodies increased 3.7-fold at the end of the FMD
regimen (p < 0.001) and returned to baseline levels following
normal food intake (Figure 6C). Circulating IGF-1 was reduced
by 24% by the end of the FMD period (p < 0.001) and remained
15% lower after resuming the normal diet (p < 0.01; Figure 6D).
IGFBP-1 was increased 1.5-fold at the end of the FMD regimen
(p < 0.01) and returned to baseline levels following normal food
intake (Figure 6E). These results indicate that the FMD group
was highly compliant and generally did not consume foods not
included in the FMD box provided to them.
Figure 6. Effects of a Human-Adapted FMD Regimen in a Pilot Clinical Trial
(A) Subjects were randomized to either the fasting mimicking diet (FMD) or a control group. Subjects in the FMD cohort consumed the FMD for 5 consecutive days
every month for 3 months and returned to normal diet in between FMDs. Control subjects continued their normal diet. Measurements were performed prior to the
diet (Baseline), immediately after the first FMD cycle (FMD), and during the recovery period after the third cycle (FMD-RF). Subjects in the control group were
evaluated within the same time frame as the FMD-RF subjects (End).
(B) Glucose (n = 19).
(C) b-hydroxybutyrate (FMD n = 19, Control n = 18).
(D) IGF-1 (FMD n = 19, Control n = 18).
(E) IGFBP-I (FMD n = 19, Control n = 17).
(F) Body weight (n = 19).
(G and H) Trunk fat (FMD n = 18, Control n = 19) (G) and lean body mass (H) evaluated by dual energy X-ray absorptiometry.
(I) C-reactive protein (CRP; FMD n = 19, Control n = 18) levels of all subjects (left) and subjects in the average or high-risk group for heart disease (n = 8; right).
(J) Percentage of lin
CD184
+
CD45
mesenchymal stem/progenitor cells (MSPC) in the peripheral blood mono-nucleated cell population (FMD n = 16, Control
n = 14).
All data are expressed as the mean ± SEM.
Cell Metabolism 22, 1–14, July 7, 2015 ª2015 Elsevier Inc. 9
Please cite this article in press as: Brandhorst et al., A Periodic Diet that Mimics Fasting Promotes Multi-System Regeneration, Enhanced Cognitive
Performance, and Healthspan, Cell Metabolism (2015), http://dx.doi.org/10.1016/j.cmet.2015.05.012
Weight, Abdominal Fat, Lean Body Mass, and
Metabolic Markers
In mice, the FMD caused weight loss and reduced visceral fat.
We studied whether the FMD could have similar effects in hu-
mans by measuring body weight, abdominal fat, and lean body
mass. The FMD resulted in a 3% reduction in body weight
(3.1% ± 0.3%; p < 0.001; Figure 6F) that remained lower at the
completion of the study (p < 0.01; Figure 6F). Trunk fat percent-
age, measured by dual-energy X-ray absorptiometry, showed a
trend (p = 0.1) for reduction after 3 FMD cycles and 1 week of
normal dieting (Figure 6G), while the relative lean body mass
adjusted for body weight was increased after completion of 3 cy-
cles (Figure 6H), indicating that fat loss accounts for most of the
weight loss. Pelvis bone mineral density was not affected by the
FMD (Figure S5D).
A complete metabolic panel (Figures S5E–S5L) indicated no
persistent metabolic changes due to the FMD except for lowered
bilirubin and alkaline phosphatase following the return to the
normal diet. Blood urea nitrogen, bilirubin, creatinine, alanine
transaminase, and aspartate transaminase showed changes
immediately following the FMD, which remained within a safe
physiological range. Together with the self-reported Common
Terminology Criteria for Adverse Events, these results provide
initial evidence that the periodic FMD is generally safe and
causes fat loss without reducing lean body mass.
Cardiovascular Disease Risk Factors
In mice, the FMD caused a reduction in inflammation-associated
diseases (Figure 2). In humans, the serum level of C-reactive pro-
tein (CRP) is a marker of inflammation and risk factor for cardio-
vascular disease. At baseline, the average CRP level for the FMD
subjects was 1.45 ± 0.4 mg/l (Figure 6I) and similar to the control
group (1.29 ± 0.5 mg/l), indicating an average moderate risk for
cardiovascular disease. CRP levels were reduced by the FMD
cycles. 8 of the 19 FMD subjects had CRP levels in the moderate
or high cardiovascular disease risk range (levels above 1.0 and
3 mg/l, respectively) at baseline. For 7 of them, the levels re-
turned to the normal range (levels below 1.0 mg/l) after 3 FMD cy-
cles (Figure 6I). For the 11 participants with CRP levels below
1.0 mg/l at baseline, no changes were observed at the comple-
tion of the trial. These results indicate that periodic FMD cycles
promote anti-inflammatory effects and reduce at least one risk
factor for CVD.
Regenerative Markers
In mice, cycles of the FMD promoted an increase of mesen-
chymal stem and progenitor cells (MSPC; Figure 2). We therefore
analyzed lin
CD184
+
CD45
MSPCs in the peripheral blood of
human FMD subjects (Figure 6J). Although not significant, the
percentage of MSPC in the peripheral blood mono-nucleated
cell population showed a trend (p = 0.1) to increase from
0.15 ± 0.1 at baseline to 1.06 ± 0.6 at the end of FMD, with a sub-
sequent return to baseline levels after re-feeding (0.27 ± 0.2). A
larger randomized trial will be required to determine whether
the number of specific populations of stem cells is in fact
elevated by the FMD in humans.
In summary, this study indicates that FMD cycles induce long-
lasting beneficial and/or rejuvenating effects on many tissues,
including those of the endocrine, immune, and nervous systems
in mice and in markers for diseases and regeneration in humans.
Although the clinical results will require confirmation by a larger
randomized trial, the effects of FMD cycles on biomarkers/risk
factors for aging, cancer, diabetes, and CVD, coupled with the
very high compliance to the diet and its safety, indicate that
this periodic dietary strategy has high potential to be effective
in promoting human healthspan. Because prolonged FMDs
such as the one tested here are potent and broad-spectrum,
they should only be considered for use under medical
supervision.
EXPERIMENTAL PROCEDURES
Subjects
Experimental design and report were prepared following the CONSORT stan-
dards for randomized clinical trials where applicable. Available data from an
ongoing pilot trial are presented. Subjects were recruited under protocols
approved by the IRB (HS-12-00391) of the University of Southern California
based on established inclusion (generally healthy adult volunteers, 18–70
years of age, BMI: 18.5 and up) and exclusion (any major medical condition
and chronic diseases, mental illness, drug dependency, hormone replacement
therapy [DHEA, estrogen, thyroid, testosterone], females who are pregnant or
nursing, special dietary requirements or food allergies, alcohol dependency)
criteria. All participants signed informed consent forms and were not offered
financial compensation for participation. Subjects were allocated (base d on
stratified sampling for age and gender) into a control (n = 19) or experimental
diet group (FMD, n = 19), followed by baseline examination. The control group
continued normal food consumption and returned for a follow-up examination
3 months after enrollment. Subjects in the FMD cohort consumed the provided
experimental diet consisting of 3 cycles of 5 continuous days of FMD followed
by 25 days of normal food intake. During all three FMD cycles, study partici-
pants self-reported adverse effects following Common Terminology Criteria
for Adverse Events. For the FMD group, follow-up examinations occurred
before resuming normal food intake at the end of the first cycle (FMD) and
also after 5–8 days of normal feeding following the end of the third diet cycle
(FMD-RF). Pre-specified outcome measures include adherence to the dietary
protocol and evaluation of physiological markers during and after completion
of the study. Examinations included height, dressed weight, body composition
(including whole-body fat, soft lean tissue, and bone mineral content)
measured by dual-energy X-ray absorptiometry (DEXA), and blood draw
through venipuncture. All data were collected at the USC Diabetes & Obesity
Research Institute. Complete metabolic panels were assayed by the Clinical
Laboratories at the Keck Medical Center of USC immediately following blood
draw. Data analysis was performed independent of study design. Complete
data will be made available elsewhere upon completion of the study.
Human Diet
The human fasting mimicking diet (FMD) program is a plant-based diet
program designed to attain fasting-like effects while providing micronutrie nt
nourishment (vitamins, minerals, etc.) and minimize the burden of fasting. It
comprises proprietary vegetable-based soups, energy bars, energy drinks,
chip snacks, chamomile flower tea, and a vegetable supplement formula tablet
(Table S4). The human FMD diet consists of a 5 day regimen: day 1 of the diet
supplies 1,090 kcal (10% protein, 56% fat, 34% carbohydrate), days 2–5 are
identical in formulation and provide 725 kcal (9% protein, 44% fat, 47%
carbohydrate).
Animals
All animal protocols were approved by the Institutional Animal Care and Use
Committee (IACUC) of the University of Southern California. Experimental
design and report were prepared following the ARRIVE standards for mouse
work. 110 9-month-old female C57Bl/6 (Charles River) retired breeders were
maintained in a pathogen-free environment and housed in clear shoebox ca-
ges in groups of three animals per cage with constant temperature and humid-
ity and 12 hr/12 hr light/dark cycle and unlimited access to water. At 16 months
of age, animals were randomly divided (by cage to avoid fighting) into the ad
libitum-fed control (CTRL) group and the fasting mimicking diet (FMD) group.
Bodyweight of individual animals was measured routinely every 2 weeks, prior
10 Cell Metabolism 22, 1–14, July 7, 2015 ª2015 Elsevier Inc.
Please cite this article in press as: Brandhorst et al., A Periodic Diet that Mimics Fasting Promotes Multi-System Regeneration, Enhanced Cognitive
Performance, and Healthspan, Cell Metabolism (2015), http://dx.doi.org/10.1016/j.cmet.2015.05.012
to starting a new FMD cycle. n = 9 mice were measured daily in the FMD and
control cohort for safety evaluation and to establish a weight profile during the
FMD cycle. Food intake was measured daily. Upon indication of progressing
dermatitis, animals were treated with a triple antibiotic ointment (Fougera
Pharmaceuticals) and were euthanized if the condition progressed. To reduce
subjective bias, mice were randomly assigned (using the online Random Num-
ber Calculator from GraphPad) to any behavioral and physiological assess-
ments shortly before any experiment. Mice that appeared weak and/or
showed signs of illness were not included in any experiment. Upon death/sac-
rifice, autopsies were performed and all abnormal classified lesions submitted
for evaluation by a pathologist. Autopsies were performed on 73 mice; 2 mice
(one from each cohort, respectively) were cannibalized and not available
for autopsy. We also utilized strain-matched younger animals to establish
age-dependent changes in risk factors using identical methods. In addition,
6-month-old female CD-1 mice (Charles River) were used in supplemental ex-
periments to measure adult neurogenesis.
Rodent Diets
Mice were fed ad libitum with irradiated TD.7912 rodent chow (Harlan Teklad)
containing 15.69 kJ/g of digestible energy (3.92 kJ/g animal-based protein,
9.1 kJ/g carbohydrate, 2.67 kJ/g fat).
The FMD is based on a nutritional screen that identified ingredients that
allow nourishment during periods of low calorie consumption (Brandhorst
et al., 2013). The FMD consists of two different components designated as
day 1 diet and day 2–4 diet that were fed in this respective order. The day 1
diet consists of a mix of various low-calorie broth powders, a vegetable medley
powder, extra virgin olive oil, and essential fatty acids; day 2–4 diet consist of
low-calorie broth powders and glycerol. Both formulations were then
substituted with hydrogel (Clear H2O) to achieve binding and to allow the
supply of the food in the cage feeders. Day 1 diet contains 7.67 kJ/g (provided
at 50% of normal daily intake; 0.46 kJ/g protein, 2.2 kJ/g carbohydrate,
5.00 kJ/g fat); the day 2–4 diet is identical on all feeding days and contains
1.48 kJ/g (provided at 10% of normal daily intake; 0.01 kJ/g protein/fat,
1.47 kJ/g carbohydrates). An alternative FMD containing 0.26 kJ/g
(0.01 kJ/g protein/fat, 0.25 kJ/g carbohydrates) was supplied for 3 days for
the evaluation of adult neurogenesis. Mice consumed all the supplied food
on each day of the FMD regimen and showed no signs of food aversion. At
the end of either diet, we supplied TD.7912 chow ad libitum for 10 days before
starting another FMD cycle. Prior to the FMD, animals were transferred into
fresh cages to avoid feeding on residual chow and coprophagy.
Survival Analysis
The endpoint considered was survival defined as the duration in time between
treatment starting date and date of death. Mice showing signs of severe
stress, deteriorating health status, or excess tumor load were designated
as moribund and euthanized. Two mice in the FMD group were sacrificed
due to seizure and head/neck injury, and one mouse died during anesthesia.
A total of 75 mice were included in the survival analysis, 46 in the control
group, and 29 in the FMD group. 12 were sacrificed due to progressing
dermatitis (CTRL n = 9, FMD n = 3) and considered as deaths for the assess-
ment of healthspan. Two cannibalized mice were considered as dead due to
reasons other than neoplasia in the analysis. A secondary analysis that
considered the 2 mice as dead because of neoplasia rendered similar results
(data not shown).
Physiological Biomarkers
Prior to blood collection and glucose measurements, mice were withheld from
food for up to 4 hr to avoid interferences caused by food consumption. For
mice, blood glucose was measured with the Precision Xtra blood glucose
monitoring system (Abbott Laboratories). An overview of all utilized commer-
cial kits is given in the Supplemental Experimental Procedures.
Complete Blood Counts and Cytokines
Complete blood counts were performed using the Mindray BC-2800 VET auto
hematology analyzer following the manufacturer’s protocol. In brief, blood was
collected from the tail vein in heparin-coated micro-hematocrit tubes. 20 mlof
the heparinized blood was added to CDS diluent (Clinical Diagnostics Solu-
tion), and whole-blood parameters were evaluated. Cytokines were measured
using the Bio-Plex Cytokine Assay (Bio-Rad), following the manufacturers
recommendation for serum analysis.
Echocardiography
Animals were anesthetized with 2% isoflurane, and the left hemithorax was
shaved. The mice were placed on a temperature-controlled heating pad,
and heart rate was continuously monitored (400–550 bpm). Ultrasound
trans-mission gel (Parker Laboratories) was used, and the heart was imaged
in the parasternal short-axis view. 2D B-mode images were obtained at the
papillary muscle level using the high-resolution Vevo 770 Ultrasound system
(VisualSonics) and analyzed using Vevo 770 V2.2.3 software (VisualSonics).
X-Ray Computed Tomography Scans
Mice (representing average body weight) were anesthetized using 2% inhalant
isoflurane and placed in a fixed position on their back. Due to prolonged anes-
thesia times, animal number was kept at n = 3 to minimize the risk of accidental
death of old mice. Tissue bone mineral density (mg Hydroxyapatite/cm
3
)of
both femora was measured in vivo for n = 5/group using the Siemens InveonCT
scanner. A detailed description is given in the Supplemental Experimental
Procedures.
Bone Marrow Collection and FACS Analysis
Bone marrow cells were harvested from femurs and tibia of mice in alpha-MEM
media (Corning Cellgro). For mice, freshly collected bone marrow cells were
washed with PBS and stained with lineage-specific, Scal-1, c-Kit, and BrdU
antibodies (BD Biosciences) according to manufacturer’s instructions. Anal-
ysis was performed using BD FACS diva on LSR II. Human Lin
CD184
+
CD45
mesenchymal stem/progenitor cells in the peripheral blood mono-nucleated
cell population were identified using human hematopoietic lineage FITC cock-
tail, anti-human CD45 APC, and anti-human CD184-PE (eBioscience, #22-
7778-72, #17-9459-42, #12-9999-42).
Immunohistochemistry
For the detection of hematopoietic cell genesis, mice were injected intraperi-
toneal with 2% filter-sterilized BrdU (10 mg/ml stock solution, Sigma) at a sin-
gle dose of 200 mg/kg bodyweight in PBS 24 hr prior to the bone marrow
collection. To analyze adult neurogenesis, BrdU was injected at 50 mg/kg
for 3 or 4 consecutive days (Figure S4) prior to FMD feeding. Staining for
BrdU, Ki67, Sox2, GFAP, and doublecortin was performed as described in
the Supplemental Experimental Procedures.
Western Blotting
A detailed description is given in the Supplemental Experimental Procedures.
qPCR
Relative transcript expression levels were measured by quantitative real-time
PCR as described in the Supplemental Experimental Procedures.
Behavior Studies
A detailed description is given in the Supplemental Experimental Procedures.
Y Maze
11 mice per treatment group were tested at 23 months of age. Spontaneous
alternation behavior (SAB) score was calculated as the proportion of alterna-
tions (an arm choice differing from the previous two choices) to the total num-
ber of alternation opportunities.
Accelerating Rotarod
At 23 months of age, 18 mice/group were evaluated using an accelerating ro-
tarod. The speed and time after which the mice fell off were recorded. On two
consecutive days, the mice were given three successive trials, for a total of six
trials. Performance was measured with two variables: the mean of the individ-
ual best performance over the two consecutive trial days and the mean time
the mice of each treatment group remained in balance over the six trial session
as an index of training.
Novel Object Recognition
The testing session comprised two trials of 5 min of each. During the first trial
(T1), the apparatus contained two identical objects. After a 1 hr delay interval,
mice were placed back in the apparatus for the second trial (T2), now with one
familiar and one new object. The time spent exploring each object during T1
Cell Metabolism 22, 1–14, July 7, 2015 ª2015 Elsevier Inc. 11
Please cite this article in press as: Brandhorst et al., A Periodic Diet that Mimics Fasting Promotes Multi-System Regeneration, Enhanced Cognitive
Performance, and Healthspan, Cell Metabolism (2015), http://dx.doi.org/10.1016/j.cmet.2015.05.012
and T2 was recorded manually. Recognition index was calculated as the time
(in seconds) spent between familiar and new object.
Barnes Maze
12 mice/group were tested twice daily for 7 days at 23 months of age. Success
rate (100%, finding the escape box [EB] within 2 min; 0%, not finding the EB
within 2 min), latency (time to enter the EB), number of errors (nose pokes
and head deflections over false holes), deviation (how many holes away
from the EB was the first error), and strategies used to locate the EB were re-
corded and averaged from two tests to obtain daily values. Search strategies
were classified as random (crossings through the maze center), serial
(searches in clockwise or counter-clockwise direction), or spatial (navigating
directly to the EB with both error and deviation scores of no more than 3).
Retention was assessed by testing once on day 14.
Yeast Intermittent Fasting
Yeast cells were streaked out from frozen stock onto YPD plates and incubated
at 30C for 2 days. Next, 3–5 colonies were inoculated in 2 ml of liquid SDC and
incubated overnight. 100 ml of the overnight culture was added to 10 ml of fresh
SDC in a 50 ml flask and incubated at 30C for 3 days . On day 3, a dilution of the
culture was plated on YPD plates to evaluate the number of viable cells. The re-
maining culture was spun down, media was removed, and the pellet was
washed with sterile dH
2
O twice before re-suspending in 10 ml of sterile dH
2
O
in a 50 ml flask followed by incubation for 2 days. On day 5, once again a dilution
of the culture was plated on YPD, and the remainder was pelleted and re-sus-
pended in 10 ml of expired media followed by 48 hr incubation. This process
was repeated by alternating dH
2
O and expired media treatment every 2 days
until the number of viable cells reached below 10% of the original culture.
To prepare expired media, 3–5 colonies were inoculated in 5 ml of SDC over-
night. 500 ml of the overnight culture was added to 200 ml SDC in a 500 ml flask
and incubated in an orbital shaker for 4 days. After the incubation period, the
cultures were filtered using a 0.22 micron filter and used for the duration of the
experiment.
Statistical Analysis
All data are expressed as the mean ± SEM. For mice, all statistical analyses
were two sided, and p values < 0.05 were considered significant (*p < 0.05,
**p < 0.01, ***p < 0.001). Differences among groups were tested by either Stu-
dent’s t test comparison, one-way ANOVA followed by Tukey’s multiple
comparison, or two-way ANOVA (for Barnes maze) using GraphPad Prism
v.5. Kaplan-Meier survival curves were compared using the Gehan-Breslow-
Wilcoxon test. Competing risk analysis was performed to assess statistical
differences in the rate of deaths. For human subjects, statistical analysis
was performed using the Wilcoxon signed-rank test, and p values < 0.05
were considered significant (*p < 0.05, **p < 0.01, ***p < 0.001).
SUPPLEMENTAL INFORMATION
Supplemental Information includes Supplemental Experimental Procedures,
five figures, and four tables and can be found with this article online at
http://dx.doi.org/10.1016/j.cmet.2015.05.012.
AUTHOR CONTRIBUTIONS
Preclinical studies: S.B. and I.Y.C., collected and analyzed the data. H.M. and
M.G.M. performed the yeast experiments. G.N. and C.W.C. collected and pro-
cessed the CBC data. C.W.C. performed FACS analysis. S.S. performed
creatinine, BUN, ALT, and renal histology. L.D. performed autopsies and his-
tology. L.P.Y. and R.P. performed X-ray computed tomography. L.P.Y.,
R.P., and M.V. performed echocardiography. S.D.B. performed cytokine
assay. F.P. performed protein expression for autophagy and myogenesis.
L.J. and S.G. performed bioinformatics analyses. P.O., L.P., P.S.C., Y.I.,
B.K.K., and P.C. were involved in study design. S.B., T.E.M., and V.D.L. de-
signed the mouse study. V.D.L. supervised all yeast and mouse studies. Clin-
ical trial: V.D.L and M.W. designed the clinical trial. V.D.L., M.W., and T.B.D.
supervised the clinical trial. M.W. and T.B.D. performed data collection and
analysis together with S.B., S.G., and H.M. S.B., I.Y.C., and V.D.L. wrote the
paper. S.B. and I.Y.C. contributed equally to this work. All authors discussed
the results and commented on the manuscript.
CONFLICT OF INTEREST
V.D.L. and T.E.M. have equity interest in L-Nutra, a company that develops
medical food. 100% of the L-Nutra equity belonging to V.D.L. will be donated
to non-profit organizations. Neither author had any role in data analysis.
ACKNOWLEDGMENTS
We would like to thank Giusi Taormina, Shawna Chagoury, and Lynn Baufeld
for their assistance in the yeast chronological lifespan experiments. Funding
was provided by the NIH and NIA grants (AG20642, AG025135, AG034906),
The Bakewell Foundation, The V Foundation for Cancer Research, and a
USC Norris Cancer Center pilot grant to V.D.L. The human study was funded
by the USC Edna Jones chair fund. The Molecular Imaging Center at USC is
supported in part by the National Center for Research Resources (NCRR,
S10RR017964-01). The funding sources had no involvement in study design;
in the collection, analysis, and interpretation of data; in the writing of the report;
or in the decision to submit the article for publication. The content is solely the
responsibility of the authors and does not necessarily represent the official
views of the National Institute on Aging or the NIH. The University of Southern
California has licensed intellectual property to L-Nutra that is under study in
this research. As part of this license agreement, the University has the potential
to receive royalty payments from L-Nutra.
Received: February 2, 2015
Revised: April 2, 2015
Accepted: May 8, 2015
Published: June 18, 2015
REFERENCES
Antosh, M., Whitaker, R., Kroll, A., Hosier, S., Chang, C., Bauer, J., Cooper, L.,
Neretti, N., and Helfand, S.L. (2011). Comparative transcriptional pathway bio-
informatic analysis of dietary restriction, Sir2, p53 and resveratrol life span
extension in Drosophila. Cell Cycle 10, 904–911.
Barnes, C.A. (1988). Spatial learning and memory processes: the search for
their neurobiological mechanisms in the rat. Trends Neurosci. 11, 163–169.
Bartke, A., Sun, L.Y., and Longo, V. (2013). Somatotropic signaling: trade-offs
between growth, reproductive development, and longevity. Physiol. Rev. 93,
571–598.
Bellantuono, I., Aldahmash, A., and Kassem, M. (2009). Aging of marrow stro-
mal (skeletal) stem cells and their contribution to age-related bone loss.
Biochim. Biophys. Acta 1792, 364–370.
Beninger, R.J., Jhamandas, K., Boegman, R.J., and el-Defrawy, S.R. (1986).
Effects of scopolamine and unilateral lesions of the basal forebrain on T-
maze spatial discrimination and alternation in rats. Pharmacol. Biochem.
Behav. 24, 1353–1360.
Bernabeu, R., de Stein, M.L., Fin, C., Izquierdo, I., and Medina, J.H. (1995).
Role of hippocampal NO in the acquisition and consolidation of inhibitory
avoidance learning. Neuroreport 6, 1498–1500.
Blackwell, B.N., Bucci, T.J., Hart, R.W., and Turturro, A. (1995). Longevity,
body weight, and neoplasia in ad libitum-fed and diet-restricted C57BL6
mice fed NIH-31 open formula diet. Toxicol. Pathol. 23, 570–582.
Blagosklonny, M.V., Campisi, J., and Sinclair, D.A. (2009). Aging: past, present
and future. Aging (Albany, N.Y. Online) 1, 1–5.
Brandhorst, S., Wei, M., Hwang, S., Morgan, T.E., and Longo, V.D. (2013).
Short-term calorie and protein restriction provide partial protection from che-
motoxicity but do not delay glioma progression. Exp. Gerontol. 48, 1120–11 28.
Brown-Borg, H.M. (2009). Hormonal control of aging in rodents: the somato-
tropic axis. Mol. Cell. Endocrinol. 299, 64–71.
Cheng, C.-W., Adams, G.B., Perin, L., Wei, M., Zhou, X., Lam, B.S., Da Sacco ,
S., Mirisola, M., Quinn, D.I., Dorff, T.B., et al. (2014). Prolonged fasting reduces
IGF-1/PKA to promote hematopoietic-stem-cell-based regeneration and
reverse immunosuppression. Cell Stem Cell 14, 810–823.
Cho, I.S., Jung, M., Kwon, K.S., Moon, E., Cho, J.H., Yoon, K.H., Kim, J.W.,
Lee, Y.D., Kim, S.S., and Suh-Kim, H. (2012). Deregulation of CREB signaling
12 Cell Metabolism 22, 1–14, July 7, 2015 ª2015 Elsevier Inc.
Please cite this article in press as: Brandhorst et al., A Periodic Diet that Mimics Fasting Promotes Multi-System Regeneration, Enhanced Cognitive
Performance, and Healthspan, Cell Metabolism (2015), http://dx.doi.org/10.1016/j.cmet.2015.05.012
pathway induced by chronic hyperglycemia downregulates NeuroD transcrip-
tion. PLoS ONE 7, e34860.
Clelland, C.D., Choi, M., Romberg, C., Clemenson, G.D., Jr., Fragniere, A.,
Tyers, P., Jessberger, S., Saksida, L.M., Barker, R.A., Gage, F.H., and
Bussey, T.J. (2009). A functional role for adult hippocampal neurogenesis in
spatial pattern separation. Science 325, 210–213.
Conboy, I.M., and Rando, T.A. (2012). Heterochronic parabiosis for the study
of the effects of aging on stem cells and their niches. Cell Cycle 11, 2260–2267.
Coppe
´, J.P., Desprez, P.Y., Krtolica, A., and Campisi, J. (2010). The senes-
cence-associated secretory phenotype: the dark side of tumor suppression.
Annu. Rev. Pathol. 5, 99–118.
Cuervo, A.M., Bergamini, E., Brunk, U.T., Dro
¨ge, W., Ffrench, M., and Terman,
A. (2005). Autophagy and aging: the importance of maintaining ‘‘clean’’ cells.
Autophagy 1, 131–140.
Deng, W., Aimone, J.B., and Gage, F.H. (2010). New neurons and new mem-
ories: how does adult hippocampal neurogenesis affect learning and memory?
Nat. Rev. Neurosci. 11, 339–350.
Fabrizio, P., Pozza, F., Pletcher, S.D., Gendron, C.M., and Longo, V.D. (2001).
Regulation of longevity and stress resistance by Sch9 in yeast. Science 292,
288–290.
Fontana, L., Partridge, L., and Longo, V.D. (2010). Extending healthy life
span—from yeast to humans. Science 328, 321–326.
Gao, Z., Ure, K., Ables, J.L., Lagace, D.C., Nave, K.A., Goebbels, S., Eisch,
A.J., and Hsieh, J. (2009). Neurod1 is essential for the survival and maturation
of adult-born neurons. Nat. Neurosci. 12, 1090–1092.
Geiger, H., and Van Zant, G. (2002). The aging of lympho-hematopoietic stem
cells. Nat. Immunol. 3, 329–333.
Gems, D., and Partridge, L. (2013). Genetics of longevity in model organisms:
debates and paradigm shifts. Annu. Rev. Physiol. 75, 621–644.
Greer, E.L., and Brunet, A. (2009). Different dietary restriction regimens extend
lifespan by both independent and overlapping genetic pathways in C. elegans.
Aging Cell 8, 113–127.
Guarente, L., and Kenyon, C. (2000). Genetic pathways that regulate ageing in
model organisms. Nature 408, 255–262.
Haigis, M.C., and Yankner, B.A. (2010). The aging stress response. Mol. Cell
40, 333–344.
Harrison, D.E., Strong, R., Sharp, Z.D., Nelson, J.F., Astle, C.M., Flurkey, K.,
Nadon, N.L., Wilkinson, J.E., Frenkel, K., Carter, C.S., et al. (2009).
Rapamycin fed late in life extends lifespan in genetically heterogeneous
mice. Nature 460, 392–395.
Harvie, M.N., Pegington, M., Mattson, M.P., Frystyk, J., Dillon, B., Evans, G.,
Cuzick, J., Jebb, S.A., Martin, B., Cutler, R.G., et al. (2011). The effects of inter-
mittent or continuous energy restriction on weight loss and metabolic disease
risk markers: a randomized trial in young overweight women. Int J Obes (Lond)
35, 714–727.
Johnson, T.E., Cypser, J., de Castro, E., de Castro, S., Henderson, S.,
Murakami, S., Rikke, B., Tedesco, P., and Link, C. (2000). Gerontogenes
mediate health and longevity in nematodes through increasing resistance to
environmental toxins and stressors. Exp. Gerontol. 35, 687–694.
Junnila, R.K., List, E.O., Berryman, D.E., Murrey, J.W., and Kopchick, J.J.
(2013). The GH/IGF-1 axis in ageing and longevity. Nat. Rev. Endocrinol. 9,
366–376.
Kaeberlein, T.L., Smith, E.D., Tsuchiya, M., Welton, K.L., Thomas, J.H., Fields,
S., Kennedy, B.K., and Kaeberlein, M. (2006). Lifespan extension in
Caenorhabditis elegans by complete removal of food. Aging Cell 5, 487–494.
Kasper, G., Mao, L., Geissler, S., Draycheva, A., Trippens, J., Ku
¨hnisch, J.,
Tschirschmann, M., Kaspar, K., Perka, C., Duda, G.N., and Klose, J. (2009).
Insights into mesenchymal stem cell aging: involvement of antioxidant defense
and actin cytoskeleton. Stem Cells 27, 1288–1297.
Katsimpardi, L., Litterman, N.K., Schein, P.A., Miller, C.M., Loffredo, F.S.,
Wojtkiewicz, G.R., Chen, J.W., Lee, R.T., Wagers, A.J., and Rubin, L.L.
(2014). Vascular and neurogenic rejuvenation of the aging mouse brain by
young systemic factors. Science 344, 630–634.
Komatsu, M., Waguri, S., Koike, M., Sou, Y.S., Ueno, T., Hara, T., Mizushima,
N., Iwata, J., Ezaki, J., Murata, S., et al. (2007). Homeostatic levels of p62 con-
trol cytoplasmic inclusion body formation in autophagy-deficient mice. Cell
131, 1149–1163.
Lee, J., Duan, W., and Mattson, M.P. (2002). Evidence that brain-derived neu-
rotrophic factor is required for basal neurogenesis and mediates, in part, the
enhancement of neurogenesis by dietary restriction in the hippocampus of
adult mice. J. Neurochem. 82, 1367–1375.
Lee, C., Safdie, F.M., Raffaghello, L., Wei, M., Madia, F., Parrella, E., Hwang,
D., Cohen, P., Bianchi, G., and Longo, V.D. (2010). Reduced levels of IGF-I
mediate differential protection of normal and cancer cells in response to fast-
ing and improve chemotherapeutic index. Cancer Res. 70, 1564–1572.
Lee, C., Raffaghello, L., Brandhorst, S., Safdie, F.M., Bianchi, G., Martin-
Montalvo, A., Pistoia, V., Wei, M., Hwang, S., Merlino, A., et al. (2012a).
Fasting cycles retard growth of tumors and sensitize a range of cancer cell
types to chemotherapy. Sci. Transl. Med. 4, 24ra27.
Lee, C., Raffaghello, L., and Longo, V.D. (2012b). Starvation, detoxification,
and multidrug resistance in cancer therapy. Drug Resist. Updat. 15, 114–122.
Lee, S.W., Clemenson, G.D., and Gage, F.H. (2012c). New neurons in an aged
brain. Behav. Brain Res. 227, 497–507.
Levine, M.E., Suarez, J.A., Brandhorst, S., Balasubramanian, P., Cheng, C.W.,
Madia, F., Fontana, L., Mirisola, M.G., Guevara-Aguirre, J., Wan, J., et al.
(2014). Low protein intake is associated with a major reduction in IGF-1, can-
cer, and overall mortality in the 65 and younger but not older population. Cell
Metab. 19, 407–417.
Longo, V.D., and Finch, C.E. (2003). Evolutionary medicine: from dwarf model
systems to healthy centenarians? Science 299, 1342–1346.
Longo, V.D., and Mattson, M.P. (2014). Fasting: molecular mechanisms and
clinical applications. Cell Metab. 19, 181–192.
Lo
´pez-Otı
´n, C., Blasco, M.A., Partridge, L., Serrano, M., and Kroemer, G.
(2013). The hallmarks of aging. Cell 153, 1194–1217.
Lynch, M.A. (2004). Long-term potentiation and memory. Physiol. Rev. 84,
87–136.
Madeo, F., Tavernarakis, N., and Kroemer, G. (2010). Can autophagy promote
longevity? Nat. Cell Biol. 12, 842–846.
Mair, W., and Dillin, A. (2008). Aging and survival: the genetics of life span
extension by dietary restriction. Annu. Rev. Biochem. 77, 727–754.
Mattson, M.P. (2012). Energy intake and exercise as determinants of brain
health and vulnerability to injury and disease. Cell Metab. 16, 706–722.
Mattson, M.P., Duan, W., Lee, J., and Guo, Z. (2001). Suppression of brain ag-
ing and neurodegenerative disorders by dietary restriction and environmental
enrichment: molecular mechanisms. Mech. Ageing Dev. 122, 757–778.
Morgan, T.E., Wong, A.M., and Finch, C.E. (2007). Anti-inflammatory mecha-
nisms of dietary restriction in slowing aging processes. Interdiscip. Top.
Gerontol. 35, 83–97.
Morrison, S.J., Wandycz, A.M., Akashi, K., Globerson, A., and Weissman, I.L.
(1996). The aging of hematopoietic stem cells. Nat. Med. 2, 1011–1016.
Moscat, J., and Diaz-Meco, M.T. (2011). Feedback on fat : p62-mTORC1-auto-
phagy connections. Cell 147, 724–727.
Muller-Sieburg, C.E., Cho, R.H., Karlsson, L., Huang, J.F., and Sieburg, H.B.
(2004). Myeloid-biased hematopoietic stem cells have extensive self-renewal
capacity but generate diminished lymphoid progeny with impaired IL-7
responsiveness. Blood 103, 4111–4118.
Narasimhan, S.D., Yen, K., and Tissenbaum, H.A. (2009). Converging path-
ways in lifespan regulation. Curr. Biol. 19, R657–R666.
Olguin, H.C., Yang, Z., Tapscott, S.J., and Olwin, B.B. (2007). Reciprocal inhi-
bition between Pax7 and muscle regulatory factors modulates myogenic cell
fate determination. J. Cell Biol. 177, 769–779.
Perry, R.L., and Rudnick, M.A. (2000). Molecular mechanisms regulating
myogenic determination and differentiation. Front. Biosci. 5, D750–D767.
Raffaghello, L., Lee, C., Safdie, F.M., Wei, M., Madia, F., Bianchi, G., and
Longo, V.D. (2008). Starvation-dependent differential stress resistance
Cell Metabolism 22, 1–14, July 7, 2015 ª2015 Elsevier Inc. 13
Please cite this article in press as: Brandhorst et al., A Periodic Diet that Mimics Fasting Promotes Multi-System Regeneration, Enhanced Cognitive
Performance, and Healthspan, Cell Metabolism (2015), http://dx.doi.org/10.1016/j.cmet.2015.05.012
protects normal but not cancer cells against high-dose chemotherapy. Proc.
Natl. Acad. Sci. USA 105, 8215–8220.
Ratajczak, M.Z., Zuba-Surma, E.K., Wysoczynski, M., Ratajczak, J., and
Kucia, M. (2008). Very small embryonic-like stem cells: characterization, devel-
opmental origin, and biological significance. Exp. Hematol. 36, 742–751.
Richetin, K., Leclerc, C., Toni, N., Gallopin, T., Pech, S., Roybon, L., and
Rampon, C. (2015). Genetic manipulation of adult-born hippocampal neurons
rescues memory in a mouse model of Alzheimer’s disease. Brain 138,
440–455.
Roybon, L., Hjalt, T., Stott, S., Guillemot, F., Li, J.Y., and Brundin, P. (2009).
Neurogenin2 directs granule neuroblast production and amplification while
NeuroD1 specifies neuronal fate during hippocampal neurogenesis. PLoS
ONE 4, e4779.
Schnell, M.A., Hardy, C., Hawley, M., Propert, K.J., and Wilso n, J.M. (2002).
Effect of blood collection technique in mice on clinical pathology parameters.
Hum. Gene Ther. 13, 155–161.
Sharma, A., Moore, M., Marcora, E., Lee, J.E., Qiu, Y., Samaras, S., and Stein,
R. (1999). The NeuroD1/BETA2 sequences essential for insulin gene transcrip-
tion colocalize with those necessary for neurogenesis and p300/CREB binding
protein binding. Mol. Cell. Biol. 19, 704–713.
Shaw, A.C., Joshi, S., Greenwood, H., Panda, A., and Lord, J.M. (2010). Aging
of the innate immune system. Curr. Opin. Immunol. 22, 507–513.
Shen, J., Tsai, Y.T., Dimarco, N.M., Long, M.A., Sun, X., and Tang, L. (2011).
Transplantation of mesenchymal stem cells from young donors delays aging
in mice. Sci Rep 1,67.
Shi, Y., Felley-Bosco, E., Marti, T.M., Orlowski, K., Pruschy, M., and Stahel,
R.A. (2012). Starvation-induced activation of ATM/Chk2/p53 signaling sensi-
tizes cancer cells to cisplatin. BMC Cancer 12, 571.
Shiotsuki, H., Yoshimi, K., Shimo, Y., Funayama, M., Takamatsu, Y., Ikeda, K.,
Takahashi, R., Kitazawa, S., and Hattori, N. (2010). A rotarod test for evaluation
of motor skill learning. J. Neurosci. Methods 189, 180–185.
Sinha, M., Jang, Y.C., Oh, J., Khong, D., Wu, E.Y., Manohar, R., Miller, C.,
Regalado, S.G., Loffredo, F.S., Pancoast, J.R., et al. (2014). Restoring
systemic GDF11 levels reverses age-related dysfunction in mouse skeletal
muscle. Science 344, 649–652.
Smith, E.D., Kaeberlein, T.L., Lydum, B.T., Sager, J., Welton, K.L., Kennedy,
B.K., and Kaeberlein, M. (2008). Age- and calorie-independent life span exten-
sion from dietary restriction by bacterial deprivation in Caenorhabditis elegans.
BMC Dev. Biol. 8,49.
Tatar, M., Bartke, A., and Antebi, A. (2003). The endocrine regulation of aging
by insulin-like signals. Science 299, 1346–1351.
Verweij, M., van Ginhoven, T.M., Mitchell, J.R., Sluiter, W., van den Engel, S.,
Roest, H.P., Torabi, E., Ijzermans, J.N., Hoeijmakers, J.H., and de Bruin, R.W.
(2011). Preoperative fasting protects mice against hepatic ischemia/reperfu-
sion injury: mechanisms and effects on liver regeneration. Liver Transpl. 17,
695–704.
Villeda, S.A., Luo, J., Mosher, K.I., Zou, B., Britschgi, M., Bieri, G., Stan,
T.M., Fainberg, N., Ding, Z., Eggel, A., et al. (2011). The ageing systemic
milieu negatively regulates neurogenesis and cognitive function. Nature
477, 90–94.
Wasselin, T., Zahn, S., Maho, Y.L., Dorsselaer, A.V., Raclot, T., and Bertile, F.
(2014). Exacerbated oxidative stress in the fasting liver according to fuel
partitioning. Proteomics 14, 1905–1921.
Wei, M., Fabrizio, P., Hu, J., Ge, H., Cheng, C., Li, L., and Longo, V.D. (2008).
Life span extension by calorie restriction depends on Rim15 and transcription
factors downstream of Ras/PKA, Tor, and Sch9. PLoS Genet. 4, e13.
Wullschleger, S., Loewith, R., and Hall, M.N. (2006). TOR signaling in growth
and metabolism. Cell 124, 471–484.
14 Cell Metabolism 22, 1–14, July 7, 2015 ª2015 Elsevier Inc.
Please cite this article in press as: Brandhorst et al., A Periodic Diet that Mimics Fasting Promotes Multi-System Regeneration, Enhanced Cognitive
Performance, and Healthspan, Cell Metabolism (2015), http://dx.doi.org/10.1016/j.cmet.2015.05.012
Cell Metabolism, Volume 22
Supplemental Information
A Periodic Diet that Mimics Fasting Promotes Multi-System Regeneration, Enhanced Cognitive
Performance, and Healthspan
Sebastian Brandhorst, In Young Choi, Min Wei, Chia Wei Cheng, Sargis Sedrakyan, Gerardo Navarrete,
Louis Dubeau, Li Peng Yap, Ryan Park, Manlio Vinciguerra, Stefano Di Biase, Hamed Mirzaei, Mario G.
Mirisola, Patra Childress, Lingyun Ji, Susan Groshen, Fabio Penna, Patrizio Odetti, Laura Perin, Peter S.
Conti, Yuji Ikeno, Brian K. Kennedy, Pinchas Cohen, Todd E. Morgan, Tanya B. Dorff, Valter D. Longo
Supplementary Data
Supplementary Figures
Insulin
0.0
0.1
0.2
0.3
0.4
***
(g/L)
Food Intake
020 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 360 380 400 420
0
5
10
15
20
25
30
35
40 CTRL
FMD
1+3 days 1+2 days
Days on Diet
(kcal/day)
Bodyweight
020 40 60 80 100 120 140 160180 200 220 240 260 280300 320 340 360 380400 420
50
60
70
80
90
100
110
120 CTRL
FMD
1+3 days 1+2 days
Days on Diet
(%)
IGF-1
#9
#16
0
50
100
150
200
250
300
350
FMD Cycle
(ng/ml)
Glucose Profile
040 80 120 160 200 240 280 320 360 400
50
100
150
200
2501+3 days 1+2 days
CTRL
FMD
FMD-RF
Day
Glucose (mg/dl)
Ketone bodies
0.0
0.5
1.0
1.5
2.0
2.5 ***
OH Butyrate (mM)
IGF-1
0
100
200
300
400
500
***
(ng/ml)
Glucose
0
50
100
150
200
250
***
Glucose (mg/dl)
Figure S1
A
BIGFBP-1
0
10
20
30
40
50
60 CTRL
FMD
FMD-RF
*** ***
ng/ml
DC
E
Heart
20 30 40 50
0.00
0.05
0.10
0.15
0.20
0.25
r2=
P-value=
0.2027
0.0075
Body weight (g)
Organ Weight (g)
Lung
20 30 40 50
0.0
0.1
0.2
0.3
0.4
0.5
r2=
P-value=
0.1416
Body weight (g)
Organ Weight (g)
Liver
20 30 40 50
0.0
0.5
1.0
1.5
2.0
2.5
r2=
P-value=
0.5670
< 0.0001
Body weight (g)
Organ Weight (g)
Spleen
20 30 40 50
0.0
0.1
0.2
0.3
0.4
0.5 r2=
P-value=
0.06699
0.1394
Weight
Organ Weight (g)
Kidney (R)
20 30 40 50
0.0
0.1
0.2
0.3
0.4
0.5 r2=
P-value=
0.1956
0.0088
Body weight (g)
Organ Weight (g)
Brain
20 30 40 50
0.0
0.2
0.4
0.6
0.8
r2=
P-value=
0.01647
0.4694
Control (16 months)
Control (20.5 months)
FMD-RF (20.5 months)
FMD (20.5 months)
Body weight (g)
Organ Weight (g)
LV Diastole
0
20
40
60
80
100
(ml)
LV Systole
0
10
20
30
40
50
(ml)
EF
0
20
40
60
80
100
%
LW Mass
0
50
100
150
200 3month CTRL
10month CTRL
25month CTRL
25month FMD
25month FMD-RF
(mg, Anterior Wall)
ALT
16.5
20.5
25
0
10
20
30
40 ** *
(U/L)
Creatinine
16.5
20.5
25
0.00
0.25
0.50
0.75
1.00
(mg/dL)
BUN
16.5
20.5
25
0
5
10
15
20
25
30
35
40
*
Age
(mg/dL)
Fibrosis
0
5
10
15
20 CTRL
FMD
FMD-RF
%
MyoD
0
25
50
75
100
%
MyoG
0
25
50
75
100 CTRL (12 month)
CTRL (20 month)
FMD (20 month)
FMD-RF (20 month)
%
LC3 I
0
50
100
150
200
%
LC3 II
0
50
100
150
200
250
300
350
*
*
%
Control FMD-RF Day1
FMD-RF Day3 FMD-RF Day7
G
H I J K L
N O P
R T U V
Q
M
S W X
Y Z
F
Figure S1 (Related to Figure 1)
A) Body weight profile in % compared to the onset of the FMD diet for the ad lib fed (gray) and
FMD fed (red) cohort. Dashed lines and arrows indicate the onset of the bi-monthly FMD cycle.
Red arrows for the 1+3 day feeding cycles, black arrows indicate the 1+2 day schedule later during
the study. See text for details. n = 9/group. B) Food intake in kcal per day for ad lib (gray) and
FMD (red) fed groups during the study period. Dashed lines and arrows indicate the onset of the
bi-monthly FMD cycle. Red arrows for the 1+3 day feeding cycle, black arrows indicate the 1+2
day schedule later during the study. See text for details. n = 15/group. C) Blood glucose (in mg/dL)
profile for the ad lib fed (gray) and FMD fed (red) cohort. Animals at the end of the FMD cycle in
bright red, animals of the FMD group 7 days after re-feeding with normal rodent chow (FMD-RF)
in light red. Dashed lines and arrows indicate the onset of the bi-monthly FMD cycle. Red arrows
for the 1+3 day feeding cycle, black arrows indicate the 1+2 day schedule later during the study.
See text for details. n = 3-5/group. D) Ketone bodies, E) Serum insulin levels and F) Serum insulin-
like growth factor (IGF)-1 level in ad lib fed control animals, FMD fed and FMD-mice 7 days post
re-feeding (FMD-RF) after the first cycle; as well as cycles 9 and 16 for IGF-1. n = 3-5/group. G)
IGF binding protein-1 (IGFBP-1) measurements for ad lib fed control (n = 5) mice or at the end
of the FMD feeding cycle (n = 5) at 17 month of age. 7 days after re-feeding, FMD animals were
measured again (FMD-RF, n = 5). H) - M) Control fed animals were euthanized at 16 months (n
= 10) prior to the start of the FMD diet to obtain baseline measurements for organ weight.
Comparison of organ weight in relation to body weight for H) right kidney, I) heart, J) liver, K)
lung, L) spleen and M) brain. N)- Q) Ultrasound echocardiogram data of young (3 months),
middle-aged (10 months) and old (25 months) control fed animals as well as 25 months-old
animals from the FMD cohort. FMD data obtained on the last day of the dietary regimen, FMD-
RF data 7 days post re-feeding with standard rodent chow. n = 5/group. N) Left ventricular (LV)
volume in diastole and O) systole, P) ejection fraction (EJ), and Q) the (corrected) relative left
ventricular wall (LW) mass were calculated. R) Serum alanine transaminase (ALT) level at 16.5,
20.5 and 25 months in control fed and FMD fed mice. For reference, ALT levels 7 days post re-
feeding in the FMD cohort are shown (FMD-RF). All data presented as mean ± SEM; * p<0.05,
** p<0.01, ANOVA, Tukey’s multiple comparison. S) Liver immunofluorescent staining with
Ki67 (red) and DAPI (blue) in control and FMD mice at day 1, 3 and 7 post refeeding. T) Serum
creatinine and U) blood urea nitrogen (BUN) level at 16.5, 20.5 and 25 months in control fed and
FMD fed mice. For reference, creatinine and BUN levels 7 days after re-feeding mice in the FMD
cohort with normal food are shown (FMD-RF). V) Quantification of Masson’s trichrome stain to
evaluate renal histopathology and glomerular/interstitial fibrosis in control fed mice and animals
from the FMD cohort both at the end of diet and the end of re-feeding (FMD-RF). To evaluate
myogenesis markers, W) MyoD and X) MyoG protein expression levels in ad lib fed control
animals at 12 and 20 months of age in comparison to 20 months-old animal at the end of the FMD
dietary cycle or one week after re-feeding (FMD-RF) with normal food were measured.
Representative results shown below. n = 4/group. All data presented as mean ± SEM. Protein
expression of the autophagy marker Y) LC3 I and Z) LC3 II was measured in ad lib fed control
animals at 12 and 20 months of age in comparison to 20 months old animal at the end of the FMD
dietary cycle or one week after re-feeding (FMD-RF) with normal food. Representative results
shown below. n = 4/group. All data are expressed as the mean ± SEM.
Figure S2 (Related to Figure 2)
A) Subcutaneous (asterisk) or C) internal masses (arrow) commonly found in aging animals. The
appearance of these masses can be attributed to subcutaneous inflammation with resulting enlarged
lymph nodes (B, arrowhead) or multi-systemic neoplasia (D; a: liver, b: ovaries, c: spleen). E)-
R) Complete blood counts (CBC) in 4 months, 16.5 months and at 20.5 months-old mice in the ad
lib fed control and FMD cohort at the end of the FMD dietary cycle or one week after re-feeding
(FMD-RF) with normal food were measured. n = 6-10/group. S)- U) Significantly changed
cytokines out of a panel of 23 markers at age 16.5 months and 20.5 months for CTRL, FMD and
FMD-refed animals. V) Bone marrow cells were harvested from the femur and tibia of 20.5
months-old control mice and FMD-fed mice 7 days after refeeding (FMD-RF) and stained with
lineage-specific Scal-1 (PE-Cy-A) and c-Kit (APC-A) antibodies to measure hematopoietic stem
and progenitor cells (HSPC). n = 6/group. All data presented as mean ± SEM. W) - X) To detect
Monocytes
0.0
0.2
0.4
0.6
0.8
Counts (109/L)
Granulocytes
0
2
4
6
Counts (109/L)
Hematocrit
0
10
20
30
40
50
% (rel. Volume)
Mean Corpuscular
Volume
0
20
40
60
(fL)
Mean Corpuscular
Hemoglobin
0
5
10
15
20
(pg)
Mean Corpuscular
Hemoglobin Conc.
0
10
20
30
40
(g/L)
Red Cell
Distribution Width
0
5
10
15
20
**
%
Platelet Distribution Width
0
5
10
15
20 *
%
Plateletcrit
0.0
0.1
0.2
0.3
0.4
0.5
*
*** *CTRL 4 month
CTRL 16.5 month
CTRL 20.5 month
FMD 20.5 month
FMD-RF 20.5 month
%
Mean Platelet Volume
0
2
4
6
8** * *
**
(fL)
Lymphocytes
0
20
40
60
80 ****
%
Monocytes
0
5
10
15 ***
%
Granulocytes
0
20
40
60 ** *
%
E F G H
L M N
P
Figure S2
Q R
J K
O
A B
C D
*
a
bc
GM-CSF
0
50
100
150 *
Fluorescence
IL-12 (p40)
0
200
400
600
800 *
Fluorescence
RANTES
0
100
200
300
400 *
*
CTRL 16.5 month
Control 20.5 month
FMD 20.5 month
FMD-RF 20.5 month
Fluorescence
HSPC
CTRL (O)
FMD-RF (O)
0
100
200
300
400
500
600
No. of HSPCs /105 BM cells
BrdU-MSPC
CTRL (M)
CTRL (O)
FMD-RF (O)
0
200
400
600
800
*
*
No. of cells / 106 BM cells
BrdU+ MSPC
CTRL (M)
CTRL (O)
FMD-RF (O)
0
30
60
90
120
**
*
No. of cells / 106 BM cells
V W XS T U
Lymphocytes
0
2
4
6
8****** ***
Counts (109/L)
I
the generation of stem/progenitor cells, mice were injected with bromodeoxyuridine (BrdU) 24
hours before bone marrow collection. Bone marrow cells harvested from control fed mature (M,
8-10 month) and old (O, 20.5 month) as well as 20.5 month old FMD-mice 7 days after refeeding
(FMD-RF) were sorted by flow cytometry to measure pre-existing (BrdU-) and newly-generated
(BrdU+) Lin-Sca1+CD45- mesenchymal stem/progenitor cells (MSPC), also known as very small
embryonic like stem cells (VSEL). All data are expressed as the mean ± SEM.
Figure S3 (Related to Figure 3)
A) Total number of arm entries of 23 months old (n =11) ad lib fed control animals and FMD-
mice 7 days after refeeding (FMD-RF) in the Y-maze behavioral test. B) Exploration time for the
two identical objects placed left and right during the adjustment phase of the novel object
recognition task. n = 8/group. All data presented as mean ± SEM.
Y-maze
CTRL
FMD-RF
0
10
20
30
40
50
# of arm entries
Figure S3
A B Exploration Time
0
5
10
15
20
CTRL FMD-RF
Right Object
Left Object
Time (s)
Figure S4 (Related to Figure 4)
A) Experimental scheme to assess neural stem cell proliferation and differentiation based on the
FMD schedule (see Methods for details) and different BrdU injection time points in 6 months-old
CD1 mice. B) GFAP+ Sox2+ (Type I neural stem cell) in the sub-granular zone (SGZ) in the control
group and at the end of FMD (BrdU schedule A). C) BrdU+ Sox2+ (Type I and Type II neural stem
cell) in SGZ in the control group and at the end of FMD (BrdU schedule B). D) BrdU+ DCX+ in
SGZ in the control group and at the end of FMD (BrdU schedule A). E) Dendrites (DCX+, schedule
C) covered molecular layer. F) Quantification of GFAP+ Sox2+ in SGZ. G) BrdU+ Sox2+ in SGZ,
FMD-RF
GFAP
SOX2
CTRL
GFAP
SOX2
BrdU+ (B)
CTRL
FMD-RF
0
5
10
15
Average Count
DCX+ (B)
CTRL
FMD-RF
0
20
40
60
80
Average Count
Double Positive (%) (B)
CTRL
FMD-RF
0
20
40
60
% BrdU+ DCX+ / BrdU+
BrdU+ (C)
CTRL
FMD-RF
0
5
10
15
20
Average Count
DCX+ (C)
CTRL
FMD-RF
0
10
20
30
40
50
Average Count
Double Positive % (C)
CTRL
FMD-RF
0
20
40
60
% BrdU+ DCX+ / BrdU+
GFAP+ Sox2+SGZ
CTRL
FMD-RF
0
20
40
60
Average Count
BrdU+ (A)
CTRL
FMD-RF
0
5
10
15
20
Average Count
BrdU
0 1 234
FMD0
(A)
72h
BrdU
(B)
7 8 17 18 1920 21
BrdU
(C)
22
Collection
(A) & (B) Collection
(C)
BrdU
SOX2
BrdU
SOX2
A
B
F
C
E
D
HG
I J K
L M N
O P Q
Double Positive (%) (A)
CTRL
FMD-RF
0
20
40
60
80 *
% BrdU+ DCX+ / BrdU+
BrdU+ Sox2+ SGZ
CTRL
FMD-RF
0
2
4
6
8
10 *
Average Count
Dendrites Covered Area
CTRL
FMD-RF
0.0
0.5
1.0
1.5
2.0 *
% of Pixel
DCX+ (A)
CTRL
FMD-RF
0
20
40
60 **
Average Count
Figure S4
DCX DCX
increased BrdU incorporation in Sox2+ cells indicating self-renewal/proliferation of neural stem
cell in the FMD cohort (n = 4/group). H) Quantification of dendritic covered area of the molecular
layer 3 weeks after FMD. I-Q) Quantification of BrdU+ (I- H), DCX+ (L- N) and double positive
for BrdU+ DCX+ over BrdU+ cells (O- Q) at time points A, B and C. All data are expressed as the
mean ± SEM.
Figure S5 (Related to Figure 6)
A) Percentage of female (F) and male (M) participants in the control or FMD groups that were part
of a randomized clinical study to evaluate the human FMD version. B) Gender specific average
age per group. C) Ethnic background. D) Subject self-reported adverse effects based on Common
Terminology Criteria for Adverse Effects (CTCAE; 1 = Mild, 2 = Moderate, 3 = Severe, 4 = Life-
threatening, 5 = Death). E) Bone mineral density (BMD in g/cm2; n = 19) was evaluated by dual-
energy X-ray absorptiometry and compared to control. Safety and feasibility were evaluated based
on complete metabolic panels (F- M; n = 19) prior to (Baseline, BL), during (FMD) and after
completion of 3 FMD cycles (FMD-RF). All data are expressed as the mean ± SEM.
Table S1 Fasting and the Fasting Mimicking Diet (FMD) induce a similar physiological
response in mice. (Related to Figure 1)
Fasting (72 hours)
FMD (96 hours)
Effect
Change
Reference
Effect
Change
Blood glucose
~45%
(Lee et al., 2010; Wang et al., 2006)
~40%
Ketone bodies
~10-fold *
(Shimazu et al., 2013)
~10-fold
IGF-I
~40- 70%
(Brandhorst et al., 2013; Frystyk et al., 1999;
Lee et al., 2010)
~45%
IGFBP-1
~7- 11-fold
(Frystyk et al., 1999; Lee et al., 2010)
~8-fold
* after 24 hours of fasting. Abbreviations: IGF-I, insulin-like growth factor 1; IGFBP-1, insulin-
like growth factor binding protein 1.
Table S2 Overview of autopsy results from the ad lib control and Fasting Mimicking Diet
(FMD) mouse cohorts. (Related to Figure 2)
Total incidence and percentage is shown per organ (gray rows) and individual abnormality.
Lymphoma incidence involves both systemic and localized cases. ^, consistent with chronic
inflammation; *, possibly due to torsion; #, consistent with lymphoma; †, associated with chronic
inflammation with numerous plasma cells and reactive mast cells.
Table S3 Complete blood counts of young and 20.5 months-old mice. (Related to Figure 2)
4 Month
20.5 month
CTRL
CTRL
FMD
FMD-RF
White Blood Cell count (109/L)
9.13 ± 0.85
4.49 ± 0.59
^
5.38 ± 0.92
9.90 ± 2.04
Lymphocyte count (109/L)
6.48 ± 0.49
1.79 ± 0.22
^^^
2.55 ± 0.43
‡ ‡
5.31 ± 0.73
†††
Monocyte count (109/L)
0.31 ± 0.04
0.47 ± 0.11
0.33 ± 0.14
0.54 ± 0.18
Granulocyte count (109/L)
2.32 ± 0.34
2.23 ± 0.50
2.50 ± 0.84
4.04 ± 1.37
Lymphocyte (%)
72.29 ± 1.74
43.09 ± 8.26
^^
51.07 ± 9.27
57.94 ± 6.61
Monocyte (%)
3.34 ± 0.22
10.34 ± 1.70
^^^
6.05 ± 1.18
6.19 ± 2.01
Granulocyte (%)
24.20 ± 1.60
46.57 ± 7.14
^
42.88 ± 8.49
30.76 ± 7.11
Red Blood Cell count (109/L)
8.59 ± 0.19
7.10 ± 0.22
^^
8.64 ± 0.34
**,
7.47 ± 0.32
Hemoglobin (g/L)
12.88 ± 0.27
10.57 ± 0.30
^^
13.42 ± 0.57
**,
11.57 ± 0.54
Hematocrit (rel. volume of erythrocytes)
36.23 ± 1.35
30.70 ± 0.94
36.50 ± 1.62
*
32.86 ± 1.24
Mean Corpuscular Volume (fL)
47.42 ± 3.32
43.36 ± 0.62
42.30 ± 0.38
44.10 ± 0.54
Mean Corpuscular Hemoglobin (pg)
14.94 ± 0.05
14.86 ± 0.23
15.47 ± 0.09
15.44 ± 0.12
Mean Corpuscular Hemoglobin Concentration (g/L)
34.84 ± 1.34
34.41 ± 0.49
36.22 ± 0.46
*
35.10 ± 0.39
Red Cell Distribution Width
12.23 ± 0.13
13.57 ± 0.28
^
13.75 ± 0.31
13.99 ± 0.32
Platelet Count (109/L)
348.40 ± 13.38
192.90 ± 82.75
712.80 ± 53.82
**,
325.00 ± 124.00
Mean Platelet Volume (fL)
5.38 ± 0.0.
6.10 ± 0.16
^^
5.45 ± 0.19
6.16 ± 0.24
Platelet Distribution Width
14.83 ± 0.05
14.97 ± 0.12
14.70 ± 0.19
15.10 ± 0.12
Plateletcrit
0.18 ± 0.01
0.11 ± 0.05
0.39 ± 0.034
**
0.19 ± 0.07
^ p< 0.05, ^^ p<0.01, ^^^ p<0.001 young (4 months) compared to old (20.5 months); * p< 0.05,
** p<0.01, *** p<0.001 young (4 months) compared to old (20.5 months); † p< 0.05, †† p<0.01,
††† p<0.001 young (4 months) compared to old (20.5 months); ‡ p< 0.05, ‡‡ p<0.01, ‡‡‡
p<0.001 young (4 months) compared to old (20.5 months).
Table S4 Caloric content of the human FMD regimen. (Related to Figure 6)
Day 1
Day 2- 5
Calories
~1090
~725
Protein (%)
10
9
Fat (%)
56
44
Carbohydrates (%)
34
47
Supplemental Experimental Procedures:
Physiological Biomarkers β–hydroxybutyrate was measured with a colorimetric assay kit
following the manufacturer’s protocol (#700190, Cayman Chemical). Insulin level were measured
using a mouse/rat specific ELISA (Millipore, #EZRMI-13K) following the manufacturer’s
protocol. Human serum IGF-I and IGFBP-1 was measured with an in-house enzyme-linked
immunosorbent assay (ELISA) based on paired specific antibodies (R&D Systems) and validated
against the commercial kit from Diagnostic Systems Laboratories. CRP levels were measured with
a human specific ELISA kit following the manufacturer’s protocol (R&D Systems, #DCRP00).
Mouse serum IGF-I was measured using a mouse specific ELISA kit (R&D Systems). Mouse
serum IGFBP-1 levels were measured by in-house ELISA assays using recombinant mouse
proteins and antibodies from R&D Systems (MAB 1240 as capture antibody and BAF 1240 as
detection antibody, R&D Systems, Minneapolis, MN, USA) as described previously (Gray et al.,
2011). PKA activity was measured using ENZO PKA kinase activity kit (ENZO Lifesciences ADI-
EKS-390A). Mouse kidney function was evaluated by serum creatinine and blood urea nitrogen
(BUN) at 16.5, 22 and 27 months of age based on a quantitative colorimetric assay
(QuantichromeTM; DICT-500 and DIUR-500, respectively) following the manufacturer’s protocol
(BioAssay Systems). In brief, 30 µl serum (for creatinine) or 5 µl (for BUN) and 200 µl working
solution were added into a 96 well plate. Mouse liver function was analyzed by measuring serum
alanine transaminase (EnzyChromTM; EALT-100) following the manufacturer’s protocol
(BioAssay Systems). In brief, 20 µl serum and 200 µl working solution were added into a 96 well
plate.
X-ray computed tomography (CT)-Scans (continued) CT phantoms for density calibration were
used to fit data and determine the slope and intercept. Images were reconstructed and densities
calculated based on the phantom scan values. The slope and intercept of the phantoms were used
to interpolate/extrapolate the tissue mineral density (TMD) for the femur of all animals. The
reported tissue mineral density of the investigated bone volume contains hydroxyapatite
contribution from both the cortical and trabecular bone.
The body fat composition was measured in vivo for N=3/group. The abdominal region of each
mouse was scanned with a Siemens InveonCT scanner at the following settings: 80 kV, 250 µA,
220° total rotation in 180 rotation steps, binning of 4 and 300 ms exposure time. Two-dimensional
gray-scale image slices were reconstructed into a three-dimensional tomography. Scans were
reconstructed between the proximal end of L1 and the distal end of L5 using COBRA software.
The region of interest for each animal was defined based on skeletal landmarks from gray-scale
images. To analyze total fat volume, a threshold segmenting fat from other tissues and background
was determined by ex vivo microCT imaging of a freshly harvested fat pad, muscle and liver tissue
from a C57BL/6J mouse. The abdominal muscular wall was used as the demarcation line to
separate visceral adipose tissue from subcutaneous adipose tissue. Fat scans by means of microCT
were performed 5 days following refeeding to avoid interferences during the immediate recovery
time from the FMD.
Immunohistochemistry Adult mice were anesthetized with isoflurane and intracardially perfused
with saline followed by 4% paraformaldehyde (PFA). The tissues were removed immediately and
post-fixed in 4% PFA for 24 hours and stored in 0.05% sodium azide. Brain was cut sagittally (40
μm), and stored in 0.05% sodium azide solution. Briefly, the sections were rinsed 3 times in
phosphate buffered saline (PBS) for 5 minutes and denatured in 2N HCl at 37°C for 20 minutes.
Sections were neutralized with 0.1M boric acid for 10 minutes and blocked with 2% Normal
Donkey Serum (NDS; Jackson ImmunoResearch) for 1 hour at room temperature. For liver,
samples were obtained from the right lobe and processed for paraffin embedding and sectioning at
the USC Stem Cell Core. Hepatic proliferation was assessed by Ki67 (Santa Cruz) staining on days
1, 3 and 7 post refeeding following the BrdU protocol. For the evaluation of adult neurogenesis in
the hippocampus, sections were incubated in BrdU (Serotec, 1:200), doublecortin (Santa Cruz,
1:200), GFAP (Cell Signaling, 1:200), Sox2 (Millipore, 1:100), diluted in 2% NDS in 0.3% triton
overnight at 4°C. The sections were rinsed 3 times in PBS for 10 min, incubated in anti-rat IgG
tagged with Alexa Flour488 and anti-goat IgG tagged with Alexa Flour598 (Invitrogen, 1:400)
diluted in 2% NDS. Sections are mounted using Vectashield (Vector). Free-floating hippocampal
(one out of every 6th) were processed for fluorescent immunohistochemistry. Co-expression was
confirmed by fluorescent- and confocal-microscopy. Ki67-positive cells were quantified by
averaging at least 5 consecutive, non-overlapping image frames from at least two tissue sections
30 μm apart. Digital images were collected on a Leica SL confocal microscope located at the
Multiphoton Imaging Core of the University of Southern California. For quantification, serological
counting methods were used.
Western blotting (continued) About 50 mg of muscle (m. gastrocnemius) was homogenized in 80
mM Tris-HCl, pH 6.8, containing 100 mM DTT, 70 mM SDS, and 1 mM glycerol, with freshly
added protease and phosphatase inhibitor cocktails, kept on ice for 30 min, centrifuged at 15000 x
g for 10 min at 4°C, and the supernatant was collected. Protein concentration was assayed using
BSA as working standard. Equal amounts of protein (30 µg) were heat-denaturized in sample-
loading buffer (50 mM Tris-HCl, pH 6.8, 100 mM DTT, 2% SDS, 0.1% bromophenol blue, 10%
glycerol), resolved by SDS-PAGE and transferred to nitrocellulose membranes (Bio-Rad,
Hercules, CA, USA). The filters were blocked with Tris-buffered saline (TBS) containing 0.05%
Tween and 5% non-fat dry milk and then incubated overnight with antibodies directed against
MyoD, and myogenin (Santa Cruz Biotechnology, CA, USA), the monoclonal antibody against
Pax7 (developed by Atsushi Kawakami, obtained from the Developmental Studies Hybridoma
Bank (University of Iowa)) and LC3B (L7583; BD Biosciences, San Jose, CA). Peroxidase-
conjugated IgG (Bio-Rad, Hercules, CA, USA) was used as secondary antibody. Membrane-bound
immune complexes were detected by an enhanced chemiluminescence system (Santa Cruz
Biotechnology, USA) on a photon-sensitive film (Hyperfilm ECL, GE Healthcare, Milano, Italy).
Protein loading was normalized according to tubulin or GAPDH expression. Quantification was
performed by densitometric analysis using TotalLab software (NonLinear Dynamics, Newcastle
upon Tyne, UK).
Quantitative PCR Relative transcript expression levels were measured using a SYBR Green-
based method. IGF1R F-CAAGCTGTGTGTCTCCGAAA/R-CTCCGTTGTTCCTGGTGTTT
and NeuroD1 F-ATTGCGTTGCCTTAGCACTT/R-TGCATTTCGGTTTTCATCCT. Average
fold changes were calculated by differences in threshold cycles (Ct) between pairs of samples.
Behavior Studies (continued) To prevent starvation-induced hyper-activity (e.g. foraging
associated movement (personal observation SB)), FMD animals were exposed to the behavior tests
not earlier than 3 days after re-feeding. Ymaze Shortterm spatial recognition memory was
examined by a spatial novelty preference task in the Y maze. The Y maze was made of black
plexiglas and comprised three identical arms (50 × 9 × 10 cm), radiating from a central triangle (8
cm on each side) and spaced 120° apart from each other. The test started placing the rodent in one
of the arms of the maze. The mouse was allowed to freely explore the environment for 8 minutes
and the total numbers of arm entries and arm choices were recorded. Arm choices are defined as
both fore-paws and hind-paws fully entering the arm. We used an Accelerating rotarod consisting
of a 3 cm diameter rotating rod (suspended 15 cm above the base) and divided by flanges so that
up to 5 mice could be tested simultaneously. Mice were placed on the rotating rod and the speed
gradually increased from 4 rpm to 40 rpm within a 5 min session. The exact speed at which the
mice fell off and time that the mice were able to stay on the bar were recorded. On two consecutive
days, the mice were given three successive trials, for a total of six trials. Novel Object Recognition
The novel object recognition test was introduced to assess the ability of rodents to recognize a
novel object in a familiar environment. The test includes a habituation phase (5 min on day one)
and trial phases (5 min each on the second day) for each mouse. Briefly, in the habituation phase,
the mouse was placed into a rectangular cage (50 x 50 x 40 cm) made of black acryl plexiglas for
5 min on day one without any object. The testing session comprised two trials with the duration of
each trial being 5 min. Mice were always placed in the apparatus facing the wall at the middle of
the front segment. Exploration of the objects is defined as any physical contact with an object
(whisking, sniffing, rearing on or touching the object) as well as positioning its nose toward the
object at a distance of less than 2 cm; however, sitting or standing on top of the object is not
counted toward the exploration time. After the first exploration period, the mice were placed back
in their home cage. To control for odor cues, the open field arena and the objects were thoroughly
cleaned with water, dried, and ventilated for a few minutes between mice. After a 1 hour delay
interval, mice were placed back in the apparatus for the second trial (T2), but now with two
dissimilar objects, a familiar one and a new one. Barnes Maze The maze consists of a platform
with 20 holes (San Diego Instruments) and 20 boxes underneath each hole; with only one hole big
enough to allow the entire mouse to enter/hide (escape box, ”EB”). A unique position for the EB
was randomly assigned to each mouse; this position was always located underneath the same hole
for a specific animal. In order to minimize the inter-maze cues, the platform was rotated after each
trial. All mice were trained once daily on days 0 to 7. During training sessions, mice were allowed
to freely explore the maze until either entering the EB or after 2 min time elapsed. If the mouse
did not enter the EB by itself, it was gently guided to and allowed to stay in the EB for 30 seconds.
After the training session, mice were tested twice daily for 7 days. Testing was similar to training,
but if after 2 min the mouse did not find the EB, it was directly returned to its cage.
... This would be enhanced even more by culturing under constant flow and media exchange to mimic more what is happening in the body with proximity to blood and lymphatic vessels. referred to a spheroid cracking or tearing [139], whereby the spheroid loses its structure. This is a challenge in comparing controls to treated samples, when relying on size. ...
... The consumption of reduced calories, but not below nutritional levels, has been shown to reduce resting metabolic rate [136], and depends on mitochondrial function for its beneficial outcomes [137]. Calorie restriction and associated dietary restriction seem to exert their effects specifically through mTOR, AMPK, and glucose handling (IGF/insulin) pathways [138,139] with outcomes such as lifespan extension, reduced inflammation and cancer. Taking the restriction further to a pure ketogenic diet has been shown in mice to reduce metabolic activity and increase uncoupling protein 2 and the ketone, beta-hydroxybutyrate [140]. ...
... Drug testing would be better in more complex, physiological system embedded in a matrix and under continuous media exchange using flow, but then throughput becomes an issue. With all or just some of these in place, especially a wider range of samples analysed, using the metrics to create in silico models[117,128,139, 154] offers a further way to analyze the relationship between the different factors and make predictions on how changes to the system may affect phenotype and drug responses. Adding different protein markers to the panel and including different protein expression analysis such as IHC or immunofluorescence would improve the understanding of protein differences in different conditions and models. ...
Book
Current drug screening protocols use in vitro cancer cell panels grown in 2D to evaluate drug response and select the most promising candidates for further in vivo testing. Most drug candidates fail at this stage, not showing the same efficacy in vivo as seen in vitro. An improved first screening that is more translatable to the in vivo tumor situation could aid in reducing both time and cost of cancer drug development. 3D cell cultures are an emerging standard for in vitro cancer cell models, being more representative of in vivo tumour conditions. To overcome the translational challenges with 2D cell cultures, 3D systems better model the more complex cell-to-cell contact and nutrient levels present in a tumour, improving our understanding of cancer complexity. Furthermore, cancer cells exhibit altered metabolism, a phenomenon described a century ago by Otto Warburg, and possibly related to changes in nutrient access. However, there are few reports on how 3D cultures differ metabolically from 2D cultures, especially when grown in physiological glucose conditions. Along with this, metabolic drug targeting is considered an underutilized and poorly understood area of cancer therapy. Therefore, the aim of this work was to investigate the effect of culture conditions on response to metabolic drugs and study the metabolism of 3D spheroid cultures in detail. To achieve this, multiple cancer cell lines were studied in high and low glucose concentrations and in 2D and 3D cultures. We found that glucose concentration is important at a basic level for growth properties of cell lines with different metabolic phenotypes and it affects sensitivity to metformin. Furthermore, metformin is able to shift metabolic phenotype away from OXPHOS dependency. There are significant differences in glucose metabolism of 3D cultures compared to 2D cultures, both related to glycolysis and oxidative phosphorylation. Spheroids have higher ATP-linked respiration in standard nutrient conditions and higher non-aerobic ATP production in the absence of supplemented glucose. Multi-round treatment of spheroids is able to show more robust response than standard 2D drug screening, including resistance to therapy. Results from 2D cultures both over and underestimate drug response at different concentrations of 5-fluorouracil (5-FU). A higher maximum effect of 5-FU is seen in models with lower OCR/ECAR ratios, an indication of a more glycolytic metabolic phenotype. In conclusion, both culture method and nutrient conditions are important consideration for in vitro cancer models. There is good reason to not maintain in vitro cultures in artificially high glucose conditions. It can have downstream affects on drug response and likely other important metrics. If possible, assays should also be implemented in 3D. If not in everyday assays, at least as a required increase in complexity to validate 2D results. Finally, metabolism even in the small scope presented here, is complex in terms of phenotypic variation. This shows the importance of metabolic screening in vitro to better understand the effects of these small changes and to model how a specific tumor may behave based on its complex metabolism.
... Despite a wealth of literature on the mechanisms and effects of CR, its clinical applicability remains limited because of challenges with long-term sustainability. Thus, newly designed dietary compositions aimed at inducing fasting-like effects that enable nutrition are beginning to emerge as potential therapies for delaying agerelated diseases, such as cancer [22]. These fasting-like diets, such as intermittent fasting (IF), also promote the regeneration and rejuvenation of multiple systems by promoting stem cell self-renewal and white blood cell formation [22]. ...
... Thus, newly designed dietary compositions aimed at inducing fasting-like effects that enable nutrition are beginning to emerge as potential therapies for delaying agerelated diseases, such as cancer [22]. These fasting-like diets, such as intermittent fasting (IF), also promote the regeneration and rejuvenation of multiple systems by promoting stem cell self-renewal and white blood cell formation [22]. Therefore, novel, and periodic forms of fasting-like diets