ArticlePDF Available

Abstract and Figures

Hypobaric hypoxia, and dietary protein and fat intakes have been independently associated with an altered gastrointestinal (GI) environment and gut microbiota, but little is known regarding host-gut microbiota interactions at high altitude (HA) and the impact of diet macronutrient composition. This study aimed to determine the effect dietary protein:fat ratio manipulation on the gut microbiota and GI barrier function during weight loss at high altitude (HA), and to identify associations between the gut microbiota and host responses to HA. Following sea level (SL) testing, 17 healthy males were transported to HA (4300m) and randomly assigned to consume provided standard-protein (SP; 1.1g/kg/d, 39% fat) or higher-protein (HP; 2.1g/kg/d, 23% fat) carbohydrate-matched hypocaloric diets for 22d. Fecal microbiota composition and metabolites, GI barrier function, GI symptoms, and acute mountain sickness (AMS) severity were measured. Macronutrient intake did not impact fecal microbiota composition, had only transient effects on microbiota metabolites, and had no effect on increases in small intestinal permeability, GI symptoms, and inflammation observed at HA. AMS severity was also unaffected by diet, but in exploratory analyses was associated with higher SL relative abundance of Prevotella, a known driver of inter-individual variability in human gut microbiota composition, and greater microbiota diversity after AMS onset. Findings suggest that the gut microbiota may contribute to variability in host responses to HA independent of the dietary protein:fat ratio, but should be considered preliminary and hypothesis-generating due to the small sample size and exploratory nature of analyses associating the fecal microbiota and host responses to HA.
Content may be subject to copyright.
RESEARCH ARTICLE Microbiome and Host Interactions
Associations between the gut microbiota and host responses to high altitude
J. Philip Karl,
Claire E. Berryman,
Andrew J. Young,
Patrick N. Radcliffe,
Tobyn A. Branck,
Ida G. Pantoja-Feliciano,
Jennifer C. Rood,
and Stefan M. Pasiakos
Military Nutrition Division, United States Army Research Institute of Environmental Medicine, Natick, Massachusetts;
Oak Ridge Institute for Science and Education, Belcamp, Maryland;
Soldier Performance Optimization Directorate, Natick
Soldier Research, Development and Engineering Center, Natick, Massachusetts; and
Pennington Biomedical Research
Center, Baton Rouge, Louisiana
Submitted 20 July 2018; accepted in final form 8 September 2018
Karl JP, Berryman CE, Young AJ, Radcliffe PN, Branck TA,
Pantoja-Feliciano IG, Rood JC, Pasiakos SM. Associations be-
tween the gut microbiota and host responses to high altitude. Am
J Physiol Gastrointest Liver Physiol 315: G1003–G1015, 2018. First
published September 13, 2018; doi:10.1152/ajpgi.00253.2018.—Hy-
pobaric hypoxia and dietary protein and fat intakes have been inde-
pendently associated with an altered gastrointestinal (GI) environment
and gut microbiota, but little is known regarding host-gut microbiota
interactions at high altitude (HA) and the impact of diet macronutrient
composition. This study aimed to determine the effect of dietary
protein:fat ratio manipulation on the gut microbiota and GI barrier
function during weight loss at high altitude (HA) and to identify
associations between the gut microbiota and host responses to HA.
Following sea-level (SL) testing, 17 healthy males were transported to
HA (4,300 m) and randomly assigned to consume provided standard
protein (SP; 1.1 g·kg
, 39% fat) or higher protein (HP; 2.1
, 23% fat) carbohydrate-matched hypocaloric diets for
22 days. Fecal microbiota composition and metabolites, GI barrier
function, GI symptoms, and acute mountain sickness (AMS) severity
were measured. Macronutrient intake did not impact fecal microbiota
composition, had only transient effects on microbiota metabolites, and
had no effect on increases in small intestinal permeability, GI symp-
toms, and inflammation observed at HA. AMS severity was also
unaffected by diet but in exploratory analyses was associated with
higher SL-relative abundance of Prevotella, a known driver of inter-
individual variability in human gut microbiota composition, and
greater microbiota diversity after AMS onset. Findings suggest that
the gut microbiota may contribute to variability in host responses to
HA independent of the dietary protein:fat ratio but should be consid-
ered preliminary and hypothesis generating due to the small sample
size and exploratory nature of analyses associating the fecal microbi-
ota and host responses to HA.
NEW & NOTEWORTHY This study is the first to examine inter-
actions among diet, the gut microbiota, and host responses to weight
loss at high altitude (HA). Observed associations among the gut
microbiota, weight loss at HA, and acute mountain sickness provide
evidence that the microbiota may contribute to variability in host
responses to HA. In contrast, dietary protein:fat ratio had only mini-
mal, transient effects on gut microbiota composition and bacterial
metabolites which were likely not of clinical consequence.
gut barrier; hypoxia; macronutrient; microbiome; weight loss
The human host and its gut microbiota coexist in a dynamic
mutualistic relationship, with the host providing a favorable
environment for microbes that, in turn, modulate gastrointes-
tinal (GI) health, GI barrier integrity, immunity, and inflam-
mation (13, 35, 36). However, this relationship can be per-
turbed by exposure to environmental and physiologic stressors
that alter the GI environment, the gut microbiota, or both (38,
49). Consequences can include degradation of GI barrier in-
tegrity leading to GI distress, translocation of bacterial antigens
(e.g., LPS) from the gut lumen into circulation, systemic
inflammation, and increased susceptibility to illness and infec-
tion (73, 74).
Hypobaric hypoxia is a physiologic stressor which charac-
terizes high-altitude (HA; 2,500-m elevation) environments.
Exposure to hypobaric hypoxia is associated with increased
inflammation (34), increased risk of illness and infection (41,
54), and increased acute mountain sickness (AMS) (7, 55), a
constellation of symptoms that includes several GI maladies
(4). Rodent studies suggest that host-gut microbiota dynamics
could contribute to these responses, demonstrating that expo-
sure to hypobaric hypoxia increases GI inflammation, oxida-
tive stress, and GI permeability concomitant to changes in gut
microbiota composition and activity (1, 82, 86, 88). Although
few human studies have attempted to corroborate those find-
ings, transient increases in GI permeability (21) and increases
in the abundance of proinflammatory gut bacteria (2, 42) in
association with inflammation (42) have been reported during
mountaineering expeditions. These observations collectively
suggest that the gut microbiota may both be affected by and
contribute to host responses at HA, but greater characterization
of these relationships is needed.
Gut microbiota composition and activity are also modulated
by dietary macronutrient intake (19, 28, 57, 80). It is well
established that carbohydrate fermentation by the gut microbi-
ota promotes the growth of beneficial bacteria and production
of the short-chain fatty acids (SCFAs) acetate, propionate, and
butyrate (48, 75), whereas amino acid fermentation produces
multiple byproducts including SCFAs, branched short-chain
fatty acids (BCFAs; e.g., isovalerate and isobutyrate), and
ammonia (48, 75). While SCFAs, and butyrate in particular,
are beneficial to GI health, several amino acid fermentation
metabolites (e.g., ammonia) may harm the GI barrier (75),
suggesting that protein fermentation could be deleterious to GI
health (83). At sea level (SL), consuming higher protein diets
Address for reprint requests and other correspondence: J. P. Karl, 10 General
Greene Ave., Bldg. 42. Natick, MA 01760 (e-mail:
Am J Physiol Gastrointest Liver Physiol 315: G1003–G1015, 2018.
First published September 13, 2018; doi:10.1152/ajpgi.00253.2018.
Licensed under Creative Commons Attribution CC-BY 4.0: ©the American Physiological Society. ISSN 0193-1857. G1003
Downloaded from at US Army Soldier Systems CMD ( on July 15, 2019.
that are also high in fat (40% total energy), low in carbohy-
drate (20% total energy), and low in fiber has been associated
with reductions in beneficial gut bacteria and fecal SCFAs and
increased fecal concentrations of amino acid fermentation
metabolites (12, 19, 24, 25, 57, 65). When macronutrient
intakes have been more consistent with dietary recommenda-
tions and fiber intakes have been matched between groups,
higher protein diets have been shown to increase fecal concen-
trations of amino acid metabolites without impacting fecal
microbiota composition (5, 79) or markers of GI health (5, 8,
79). Thus protein fermentation does not appear to acutely
degrade GI barrier function when fiber intake is controlled.
However, none of these studies were conducted in environ-
ments such as HA that may render the GI barrier more sensitive
to gut microbes and their metabolites.
Recent interest in studying higher protein, moderate carbo-
hydrate diets at HA (59) has been driven by the knowledge that
unintentional fat-free mass loss is common during HA sojourn
(33, 78), and that higher protein diets preserve fat-free mass
during weight loss at SL (60, 81). However, the effects of these
diets on host-gut microbiota dynamics at HA are unknown.
This study aimed to both address that gap and the need for
greater characterization of host-gut microbiota dynamics at HA
by 1) determining the effect of altering the dietary protein:fat
ratio on GI barrier function, GI symptoms, and gut microbiota
composition and metabolites during weight loss at HA; and 2)
identifying associations among the gut microbiota, weight loss
at HA, and the host response to HA as measured by AMS
Participants and study design. The analyses reported herein were
included as secondary objectives in a randomized, controlled feeding
study designed to examine the efficacy of a higher protein diet for
preserving fat-free mass during HA sojourn (9). Participants were 17
healthy, unacclimatized, physically active men. Although study en-
rollment was open to women, none volunteered. Participants had no
GI abnormalities or disorders, had not taken oral antibiotics or had a
colonoscopy in the previous 3 mo, and did not regularly use laxatives,
stool softeners, or antidiarrheal medications. The study was approved
by the Institutional Review Board at the U.S. Army Research Institute
of Environmental Medicine (Natick, MA) and conducted May–Aug
2016. Investigators adhered to the policies for the protection of human
participants as prescribed by Army Regulation 70-25, and the research
was conducted in adherence with the provisions of 32 CFR Part 219.
(The trial is registered on as NCT02731066.)
Study methods have been previously reported in detail (9). Briefly,
the study was a randomized, controlled trial consisting of two phases
conducted over 43 consecutive days. During the 21-day first phase
(SL), participants resided at SL, consumed a self-selected weight-
maintaining diet, maintained habitual exercise routines, and were free
living but visited the laboratory daily. On SL day 21, participants were
flown from Boston, MA to Denver, CO, where they were placed on
supplemental oxygen until being driven to the summit of Pike’s Peak,
CO (4,300 m) the following morning (HA day 0) where they resided
for the next 22 days at the U.S. Army Research Institute of Environ-
mental Medicine Maher Memorial Laboratory (phase 2; HA). During
HA, participants were under constant supervision, consumed a con-
trolled and measured diet, and engaged in prescribed physical activity.
Diets contained either a standard or higher amount of protein, and
were designed to induce weight loss, which is common during HA
sojourn (33). The estimated energy deficit at HA was 70% or 1,849
kcal/day (SD 511) (9).
Study diets. Beginning on HA1, the first full day of residence at
4,300 m, and continuing until they completed the HA phase of the
study, participants consumed a controlled diet. Participants were
randomized by study staff using computer-generated randomization to
consume either a standard protein [SP; 1.1 g·kg
(SD 0.2)] or
higher protein [HP; 2.1 g·kg
(SD 0.2)] diet during HA. Both
diets were designed to provide 45% of energy as carbohydrate, while
fat intake was reduced in HP to accommodate the higher protein
intake (Table 1). Diets were primarily comprised of entrées, sides, and
snack items included in U.S. military Meals Ready-to-Eat rations and
were supplemented with fresh fruits and vegetables, fruit snacks, olive
oil, and ranch-flavored salad dressing. A whey-protein beverage
(Isopure Zero Carb; Isopure, Hauppauge, NY) was also provided as
appropriate to manipulate protein intake. Water and noncaffeinated
sodas were allowed ad libitum.
Questionnaires. Modified versions of the Irritable Bowel Syn-
drome-Symptom Severity Score (IBS-SSS) Questionnaire (30) and
the Gastrointestinal Quality of Life Index (GIQLI) Questionnaire (27)
Table 1. Baseline participant characteristics, weight loss, and dietary intakes at sea level and over 22 days at high altitude
Age, yr 23 3247
Body mass index, kg/m
27.0 4.0 25.5 3.1
Body fat, % 22.8 7.0 22.6 5.5
, ml·kg
49.2 7.0 53.8 7.2
Weight at HA, ‡ kg 8.0 2.6 7.8 1.2
FFM at HA, ‡ kg 4.0 3.3 3.2 1.5
Energy intake, † kcal/day 2,366 277 1,950 186 2,418 542 1,885 269
Protein intake, % 16 3182142332*
Protein intake, g/day 94 20 88 14 83 15 154 25*
Carbohydrate intake, % 50 6461545471
Carbohydrate intake, g/day 300 62 223 24 327 80 221 35
Fat intake, % 34 4392324233*
Fat intake, g/day 89 9846887477*
Saturated fat intake, g/day 31 4262267152*
Fiber intake, g/day 18 72131910 18 2*
Values are means SD; n8 from standard-protein diet group (SP) and 9 from higher-protein diet group (HP). SL, sea level (weight maintenance); FFM,
fat-free mass; HA, high altitude (4,300 m; energy deficit). Adapted from Berryman et al. (9). *Different from SP during HA, P0.05. †For all dietary variables
independent samples t-test were used to compare groups at SL (ad libitum diet) and HA (provided diet); no significant differences during SL. ‡Linear mixed
model, main effect of time, P0.001.
AJP-Gastrointest Liver Physiol doi:10.1152/ajpgi.00253.2018
Downloaded from at US Army Soldier Systems CMD ( on July 15, 2019.
were administered weekly throughout the study. The GIQLI Ques-
tionnaire asked participants to subjectively rate the frequency of
several GI-related symptoms (e.g., flatulence, constipation, loose
stool, cramping, etc.), which were then used to compute an overall
GIQLI score (27). AMS was assessed by Lake Louise score calculated
from the shortened version of the Environmental Symptoms Ques-
tionnaire (6) and categorized as mild (0.7 and 1.53), moderate
(1.53 and 2.63), and severe (2.63) (7).
Blood biochemistries. Blood samples were collected by venipunc-
ture and following a 10 h fast to measure markers of gut barrier
function, stress, and inflammation. All samples were separated into
serum or plasma and stored at 20 to 80°C until analysis. Plasma
LPS-binding protein (LBP) was measured on SL day 7 and HA days
0,7, and 21 by ELISA according to manufacturer’s instructions
(Abonva, Taipei, Taiwan). Serum glucagon-like peptide-2 (GLP-2), a
gastrointestinal hormone shown to modulate effects of diet-gut mi-
crobe interactions on GI permeability (14), was measured on SL day
7and HA days 0,7, and 21 by ELISA according to manufacturer’s
instructions (EMD Millipore, St. Charles, MO). Serum IL-6 concen-
trations were measured on SL day 0 and HA days 0 and 19 by
multiplex assay (Milliplex MAP; EMD Millipore). Serum cortisol
concentrations were measured on SL day 12 and HA days 2,7,13, and
19 by autoimmunoassay (Immulite 2000; Siemens Healthcare Diag-
nostics, Tarrytown, NY).
Small intestinal permeability. Small intestinal permeability was
assessed by quantifying the urinary excretion of orally ingested sugar
substitutes (45, 53) on SL day 12 and HA days 1 and 18. Measure-
ments began in the morning with fasted participants consuming 5 g
lactulose and 4 g mannitol dissolved in 180 ml of water. All urine
produced over the next 5 h was collected. During the 5-h period,
participants consumed the same individualized diet (breakfast and
morning snack) on all 3 test days, with water allowed ad libitum.
Urine aliquots were immediately frozen and stored at 20 to 80°C
until analysis. Urine lactulose and mannitol concentrations were
measured by HPLC with refractive index detection (Agilent 1100
HPLC, Santa Clara, CA) as previously described (51). Fractional
excretion was calculated by multiplying the measured concentration
of each probe by the total volume of urine collected and dividing by
the dose administered. Small intestinal permeability was then calcu-
lated as the ratio of the fractional excretions of lactulose and mannitol
(LM ratio) (45, 53).
Fecal sample collection and analysis. Participants provided a
single fecal sample during five separate time periods of the study; SL
days 0 – 4 (SL1) and 16 –20 (SL2) and HA days 1–2 (HA1), 8 –11
(HA2), and 18 –21 (HA3), to assess fecal microbiota composition and
fecal concentrations of SCFAs, BCFAs, and ammonia. All samples
were collected into plastic collection containers, immediately refrig-
erated, and processed within 3 h (SD 4; range: 5 min – 21 h) of
production. Aliquots were immediately frozen and stored at 20 to
80°C until analysis.
Fecal metabolites. Fecal SCFA and BCFA concentrations were
measured as previously described (61, 87) with minor modifications.
Fecal aliquots were thawed immediately before extraction, homoge-
nized in distilled water (1:4 wt/vol), and centrifuged. Samples were
then acidified using 50% H
(1:2 wt/vol), and fatty acids were
extracted using diethyl ether (2:5 wt/vol). After incubating on ice for
2 min, samples were centrifuged, the organic layer was removed, and
ethyl butyric acid was added as an internal standard. Samples were then
stored at 80°C until analysis. Acetic acid, propionic acid, butyric acid,
isobutyric acid, and isovaleric acid were quantified using an Agilent
7890A GC system with Flame Ionization Detection (60 m 250 m
0.25 m; DB-FFAP, Agilent J&W). Samples (1 l) were injected by
autosampler in triplicate using a split ratio of 10:1. The temperature
program started with an initial temperature of 110°C for 2 min, increased
10°C/min up to 180°C, and was then maintained at 180°C for 5 min. The
carrier gas was nitrogen with a constant flow of 1 ml/min. Calibration
standards were included for each fatty acid, and used for peak identifi-
cation and quantification.
Fecal ammonia concentrations were measured using a colori-
metric assay according to manufacturer instructions (Abcam, Cam-
bridge, MA).
Fecal microbiota composition. DNA was extracted from fecal
samples using the MoBio PowerFecal DNA isolation kit (Qiagen,
Germantown, MD). Primers designed to amplify the V3-V4 region of
the 16S rRNA gene were used for PCR amplification, and all samples
were sequenced in triplicate on the Illumina MiSeq platform (Illu-
mina, San Diego, CA). Sequencing data were processed using Quan-
titative Insights Into Microbial Ecology (QIIME) v.1.9.1 (16). Read
quality assessment, filtering, barcode trimming, and chimera detection
were performed on demultiplexed sequences using Trimmomatic (11).
Reads were joined in QIIME using a minimum overlap of 32 bp and
a maximum percent difference within the overlap of 20%. Operational
taxonomic units (OTUs) were assigned by clustering sequence reads
at 97% similarity and aligned against the Greengenes database core
set v.13_8 (52) using PyNAST (15). Taxonomic assignment was
completed using the RDP classifier v.2.2 (77).
Read counts averaged 80,024 reads/sample (SD 61,030; range:
31,267– 417,621), and were grouped into 12,966 unique OTUs, which
could be assigned to 134 unique genera and 13 unique phyla. For
genus-level analyses, any OTUs that could not be assigned to a genus
were grouped at the next lowest level of classification possible (e.g.,
family or order).
Diversity metrics were calculated after rarefaction at 31,267
reads/sample. Within-sample diversity (-diversity) was calculated
in QIIME using the Shannon and observed OTUs diversity metrics.
Between-sample diversity (-diversity) was measured using Bray-
Curtis distances calculated using the R packages stats v.3.4.3 and
phyloseq v.1.16.2. Ordinations of -diversity metrics were then
completed using principal coordinates analyses (PCoA) and aver-
age hierarchical clustering within the R package ape.
Statistical analysis. Data were checked to verify adherence to
model assumptions and transformed when necessary to meet model
assumptions. Unless otherwise noted, statistical analyses were com-
pleted using SPSS v.21, data are presented as mean (SD), statistical
significance was set at P0.05, and Pvalues between 0.05 and 0.10
were considered evidence of a trend for an effect.
All study outcomes except for genus-level read counts were ana-
lyzed by linear mixed models, generalized linear mixed models, or
marginal models as appropriate (see table and figure legends for
model specifications). All models accounted for the within-subject
correlation and included diet group, time, and their interaction as fixed
factors and age and baseline body mass index as covariates. The
baseline value of the dependent variable was also included as a
covariate in models where the dependent variable was measured at
multiple SL time points (questionnaires and fecal outcomes) or if the
first measurement at HA was completed immediately after ascent
(plasma LBP and serum GLP-2). In all models, if a significant
interaction or main effect was observed, post hoc comparisons were
conducted using t-tests, and P-values were adjusted using Bonferroni
corrections. Correlations were assessed using Pearson’s or Spear-
man’s correlation as appropriate or by including time-varying cova-
riates in linear models.
Bray-Curtis dissimilarities in gut microbiota composition were
analyzed using distance-based redundancy analysis in the R package
vegan. Unrarefied genus-level read counts were analyzed using
DESeq2 v.1.16.1 (47) and R v.3.4.2 to test for effects of diet, time,
and their interaction on changes in gut microbiota composition while
controlling for individual effects. For these analyses, likelihood ratio
tests were used to test for diet-by-time interactions, and, if no
interactions were observed, main effects of time. False discovery rate
(FDR) was controlled by adjusting Pvalues obtained for diet-by-time
interactions and main effects of time using the Benjamini-Hochberg
correction. If a significant interaction or main effect of time was
AJP-Gastrointest Liver Physiol doi:10.1152/ajpgi.00253.2018
Downloaded from at US Army Soldier Systems CMD ( on July 15, 2019.
observed (FDR 0.20), pairwise contrasts were examined for those
taxa to identify between group differences or differences over time
within the full cohort. P-values for pairwise contrasts were adjusted
for the number of contrasts using Bonferroni corrections.
To address exploratory aims, participants were dichotomized as
“responders” and “nonresponders” to the physiologic stress of HA
using Lake Louise scores measured within the first 48 h of ascent
(peak scores for all participants occurred within 48 h of ascent as
expected; Ref. 7). Moderate or severe AMS (responder) was defined
as a peak Lake Louise score of 1.53, and no or mild AMS
(nonresponder) was defined as a peak Lake Louise score of 1.53 (6).
The AMS-responder group was included in separate marginal or
mixed models as an independent variable to examine changes in
physiologic outcomes. All models accounted for the within-subject
correlation and included diet group, time, and responder group as
fixed factors, the responder group-by-time interaction, and age, base-
line body mass index, and baseline value of the dependent variable as
covariates. For analysis of changes in genus-level read counts over
time, the AMS-responder group was substituted for diet group in the
DESeq2 models.
Finally, linear discriminant analysis of effect size (LEfSe) (67) was
used to identify taxa associated with AMS severity. For these analyses
the AMS-responder group was used as the grouping factor (i.e.,
“class”), time point was used as a subgroup factor (i.e., “subclass”),
and relative abundances of taxa measured at SL were included in the
analysis. Default analysis parameters were used with the exceptions of
reducing the significance threshold to P0.01 due to the small
sample size and conducting pairwise between-group comparisons
only within the same time points. As such, discriminant taxa were
defined as those with an effect size 2.0 between groups (P0.01)
at both SL time points.
Primary and secondary results have been previously reported
(9, 37, 50, 58, 84). Total body mass and fat-free mass losses
during HA did not differ between diet groups (Table 1). During
HA, protein intake was higher in HP relative to SP while fat
intake was lower, and total fiber intake was marginally lower
(Table 1).
Effects of diet and time on GI symptoms and small intestinal
permeability. GIQLI scores were lower (indicating worse
symptomology) throughout HA relative to SL, and did not
differ by diet group (Fig. 1A). IBS symptom severity scores
did not differ over time (main effect of time, P0.69) or
by diet group (diet-by-week interaction, P0.15) (data not
Lactulose excretion was lower on HA day 18 relative to SL
and HA day 1 while mannitol excretion was higher at SL
relative to HA days 1 and 18 (main effect of time, P0.01,
Table 2). The resulting LM ratio was 71% (SD 73) higher on
HA day 1 and 67% (SD 77) higher on HA day 18 relative to
SL (main effect of time, P0.001) indicating increased
small intestinal permeability. Diet had no impact on the LM
ratio (Table 2). When urine volume was included as a
covariate in the model, the main effect of time for lactulose
was no longer significant whereas results for mannitol
excretion and the LM ratio were unchanged, the latter
observation remaining consistent with increased small in-
testinal permeability at HA.
Diet had no significant effect on serum IL-6 concentrations;
however, an upwards trend was observed over time at HA (Fig.
1B). Changes in plasma LBP concentrations over time differed
by diet with concentrations initially increasing at HA in HP but
then dropping to a concentration 24% lower than that measured
in SP on HA day 21 (P0.02; Fig. 1C). Diet had no effect on
serum GLP-2 concentrations (Fig. 1D). However, in the full
cohort, GLP-2 concentrations were lower on HA day 7 relative
P-diet = 0.38
P-week < 0.001
P-diet*week = 0.68
P-diet = 0.22
P-day < 0.001
P-diet*day = 0.03
P-diet = 0.51
P-day < 0.001
P-diet*day = 0.78
P-diet = 0.80
P-day = 0.09
P-diet*day = 0.48
GIQLI score
Plasma LBP (µg/mL)Serum GLP-2 (ng/mL)
Log10 serum IL-6 (pg/mL)
Day Day
Fig. 1. Changes in gastrointestinal (GI) sympto-
mology, GI function, and inflammation at high
altitude (HA; 4,300 m) are largely independent
of dietary protein:fat ratio. GI quality of life
(GIQLI; lower scores indicate worse symptoms;
A) and fasting serum IL-6 (B), plasma LPS
binding protein (LBP; C), and serum glucagon-
like peptide (GLP)-2 (D) concentrations. Data
are analyzed by marginal models controlling for
age, body mass index, and baseline value. Band
C: data were log
-transformed for analysis. HP,
higher protein diet group; SL, sea level; SP,
standard-protein diet group. HP, n9; SP, n
8. Values are means SE. Time points not shar-
ing a superscript letter are significantly different
within a diet group (P0.05). #Significantly
different from SL (P0.05). *Significantly
different from SP (P0.02). †Significantly
different from HA days 0 and 21 (P0.01).
AJP-Gastrointest Liver Physiol doi:10.1152/ajpgi.00253.2018
Downloaded from at US Army Soldier Systems CMD ( on July 15, 2019.
to HA days 0 and 21 and were inversely correlated with LBP
concentrations over time [:2.3 g/ml (SE 0.9), P0.01].
Effects of diet and time on fecal metabolite concentrations.
Total fecal SCFA concentrations were 39% lower during the
second week of HA in HP relative to SP due to lower acetate,
propionate and butyrate concentrations in HP but did not differ
before or after (Table 3). Total BCFA concentrations trended
toward being higher in SP relative to HP independent of diet
(Table 3) due to higher mean concentrations of isovalerate
(main effect of diet, P0.07) in SP throughout the study (data
not shown). Fecal ammonia concentrations did not differ by
diet but were higher during the second and third weeks at HA
relative to SL (Table 3).
Effects of diet and time on fecal microbiota composition. At
the community level, no metric of -diversity changed over
time or differed by diet (Fig. 2A). PCoA (Fig. 2B) and hierar-
chical clustering (data not shown) of Bray-Curtis dissimilari-
ties did not reveal any clustering by diet (diet-by-time interac-
tion, P0.91). The effects of diet on genus-level fecal
microbiota relative abundances were observed for only two
taxa. Both changes in the relative abundances of Holdemania
(phyla: Firmicutes) and unclassified taxa within the Chris-
tensenellaceae family (phyla: Firmicutes) at HA were greater
in HP relative to SP (diet-by-time interaction, P0.003;
FDR 0.20; Fig. 2C). However, both taxa demonstrated a
trend toward being less abundant in HP relative to SP at SL
(LEfSe effect size 2.6, P0.02), which implicated regres-
sion to the mean rather than an effect of diet. Finally, no
between group difference was noted when LEfSe was used to
examine differences in fecal microbiota composition measured
during the final week of HA (P0.01; data not shown). Taken
together these findings indicated that diet had little impact on
fecal microbiota community composition.
PCoA of Bray-Curtis dissimilarities did show evidence of
clustering over time (Fig. 2B, main effect of time, P0.02).
In support, several genera demonstrated significant changes
over time independent of diet (main effect of time, FDR
0.20). Specifically, relative to SL, Lachnospira relative abun-
dance was lower on HA days 1–2 [log
fold change ⫽⫺1.5
(SE 0.5); P0.005] but not thereafter, Turicibacter relative
abundance was lower during only the second week at HA (Fig.
2D), and Bacteroides relative abundance trended toward being
higher during only the second week at HA [log
change 0.9 (SE 0.4); P0.06]. Lactococcus, Streptococcus,
and Lactobacillus (Fig. 2D) relative abundances were all lower
relative to SL during the second and third weeks of HA (range
Table 3. Fecal short-chain fatty acid, branched-chain fatty
acid, and ammonia concentrations measured at sea level and
high altitude
Diet Time Diet time
Acetate, mol/g
wet wt 0.42 0.18 0.08
SL1 35.0 12.1 29.7 14.9
SL2 37.1 26.5 24.5 8.6
HA1 26.2 6.9 23.7 6.7
HA2 28.2 8.9 18.6 10.0
HA3 20.2 8.7 22.3 2.3
Propionate, mol/g
wet wt 0.79 0.14 0.03
SL1 15.4 8.1 16.9 9.5
SL2 14.7 12.8 13.6 5.2
HA1 12.5 4.9 10.1 4.1
HA2 14.1 7.9 8.3 4.5
HA3 7.2 3.7
10.6 3.5
Butyrate, mol/g
wet wt 0.29 0.07 0.02
SL1 13.6 6.2 13.6 9.2
SL2 14.1 12.2 9.7 4.2
HA1 13.0 3.9 11.7 2.3
HA2 12.8 4.2 6.7 4.8
HA3 10.4 4.7 10.9 3.4
Total SCFA, mol/g
wet wt 0.55 0.19 0.04
SL1 64.1 23.8 60.2 30.3
SL2 66.0 51.0 47.8 16.2
HA1 51.6 13.4 45.6 9.5
HA2 55.1 19.3 33.6 18.7
HA3 37.9 15.2 43.8 6.2
BCFA, mol/g
wet wt 0.07 0.32 0.29
SL1 3.2 2.2 2.6 1.4
SL2 3.1 2.7 1.6 1.6
HA1 3.7 2.9 2.2 1.9
HA2 3.8 1.7 1.9 1.0
HA3 1.8 1.2 2.1 1.8
Ammonia, mol/g
wet wt 0.38 0.001 0.42
SL1 38.0 23.9 24.2 17.9
SL2 16.1 7.4 18.6 12.3
HA1 34.2 21.1 23.9 15.1
59.1 27.3 29.5 15.5
28.9 14.1 28.4 17.0
Values are means SD; n8 from standard-protein diet group (SP) and 9
from higher protein diet group (HP). SL, sea level (weight maintenance); HA,
high altitude (4,300 m; energy deficit); BCFA, branched chain fatty acid
(isobutyrate and isovalerate); SCFA, short-chain fatty acid (acetate, propi-
onate, and butyrate). Samples collected on study days 0– 4 (SL1) and 16 –20
(SL2) at sea level and days 1–2 (HA1), 8 –11 (HA2), and 18 –21 (HA3) at
high altitude.
Linear mixed model with Bonferroni corrections controlling
for age, body mass index, and concentration on SL day 0 4. All concen-
trations were log
- transformed for analysis.
Significantly different from
SP (P0.04).
Significantly different from HA2 within SP (P0.05).
Significantly different from SL2 (P0.001).
Trend for difference from SL2
Table 2. Small intestinal permeability measured at sea level
and high altitude
Diet Time Diet time
Lactulose,§ % 0.96 0.01 0.35
SL day 12 0.15 0.07 0.13 0.04
HA day 1 0.16 0.10 0.15 0.05
HA day 18* 0.11 0.10 0.12 0.05
Mannitol, % 0.41 0.001 0.76
SL day 12* 16.2 4.4 17.6 3.0
HA day 1 10.5 2.6 12.8 5.1
HA day 18 8.6 5.4 9.2 2.6
LM ratio§ 0.47 0.001 0.55
SL day 12* 0.010 0.005 0.007 0.002
HA day 1 0.014 0.007 0.013 0.005
HA day 18 0.013 0.007 0.013 0.005
Values are means SD; n8 from standard-protein diet group (SP)
and 9 from higher-protein diet group (HP). SL, sea level (weight mainte-
nance); HA, high altitude (4,300 m; energy deficit). LM, lactulose:manni-
tol. *Significantly different from the other days (P0.05). †Trend for
significant difference relative to HA day 1 (P0.08). ‡Linear mixed
model controlling for age and body mass index with Bonferroni correc-
tions. §Log
transformed for analysis.
AJP-Gastrointest Liver Physiol doi:10.1152/ajpgi.00253.2018
Downloaded from at US Army Soldier Systems CMD ( on July 15, 2019.
of log
fold changes 2.2– 4.9; P0.005). Of note, all three
of these taxa include strains that are commonly used as starter
cultures in dairy products (Streptococcus thermophilus and
Lactobacillus spp. in yogurt and Lactococcus spp. in cheese).
Therefore, SL food records were used to estimate yogurt
consumption during SL (consumption of Lactococcus-contain-
ing cheeses during SL could not be accurately estimated).
Yogurt consumption during SL was correlated with mean
Streptococcus relative abundance during SL (Spearman’s
␳⫽0.68, P0.003) and the change in Streptococcus relative
abundance from SL to the third week at HA (Spearman’s
␳⫽⫺0.71, P0.002). In contrast, SL yogurt consumption
was not correlated with mean Lactobacillus relative abundance
during SL (Spearman’s ␳⫽0.08, P0.75) or the change in
Lactobacillus relative abundance from SL to the third week at
HA (Spearman’s ␳⫽⫺0.03, P0.89). These observations
suggest that the decrease in Streptococcus, but not Lactobacil-
lus, relative abundance at HA was attributable to reduced
yogurt consumption. Of note, neither mean dietary fiber intake
at SL nor the change in mean fiber intake from SL to HA was
correlated with changes in the relative abundance of any of
these six taxa (P0.24 for all).
Responders and nonresponders: AMS severity. AMS sever-
ity as measured by the Lake Louise score represents an overall
cognitive and physical response phenotype resulting from the
physiologic stress induced by hypobaric hypoxia. Therefore, in
additional exploratory analyses, AMS severity was considered
an indicator of the cumulative stress response to HA, and
associations between gut microbiota-related factors and both
the development of and response to AMS were explored.
Eleven participants reported moderate or severe AMS (re-
sponders) at 1 time points during the first 48 h at HA (Table
4). Lake Louise scores were consistently higher during the first
6 days of sojourn in AMS responders relative to those who
experienced no or mild AMS (nonresponders) (Fig. 3A). In
support of AMS severity reflecting the cumulative stress re-
P-diet = 0.66
P-time = 0.56
P-diet*time = 0.42
P-diet = 0.83
P-time = 0.93
P-diet*time = 0.79
Log2fold change from SL
(HP v. SP)
Log2fold change from SL
-0.5 -0.4 -0.3 -0.2 -0.1 0.0 0.1 0.2 0.3
PCo1 (13.3%)
PCo1 (13.3%)
PCo2 (9.5%)
Time point Time point
P-time = 0.02
P-diet*time = 0.91
Time pointTime point
Shannon diversity
Observed OTUs
Fig. 2. Changes in fecal microbiota composi-
tion during weight loss at high altitude (HA
4,300 m) are largely independent of dietary
protein:fat ratio. A:-diversity [Shannon di-
versity and observed operational taxonomic
units (OTUs)] analyzed by marginal models
controlling for age, body mass index, and base-
line diversity (i.e., SL1). B: principle coordi-
nates (PCo) analysis of Bray-Curtis dissimilar-
ities. Individual data points represent the entire
fecal microbiota community of a single individ-
ual at one point in time. Samples closer together
are more similar than samples farther apart. C:
-fold change in relative abundance from SL
in HP relative to SP for taxa demonstrating
significant diet-by-time interactions (false dis-
covery rate 0.20). *Change from SL signifi-
cantly different in HP vs. SP (P0.02). D:
-fold change in relative abundance from SL
in genera demonstrating a significant main ef-
fect of time (false discovery rate 0.20). #Sig-
nificant change from SL (P0.005). HA, high
altitude; HP, higher protein diet group; SL, sea
level; SP, standard protein diet group; Uncl,
unassigned genus level taxonomy. HP, n9;
SP, n8.
AJP-Gastrointest Liver Physiol doi:10.1152/ajpgi.00253.2018
Downloaded from at US Army Soldier Systems CMD ( on July 15, 2019.
sponse, mean serum cortisol concentrations were slightly
higher throughout HA in AMS responders relative to nonre-
sponders (Fig. 3B). Finally, GIQLI scores were lower during
the first 2 wk of sojourn in AMS responders relative to
nonresponders (Fig. 3C), consistent with GI issues contributing
to AMS symptomology. No differences in baseline character-
istics between responder groups were noted (Table 4) nor were
any relationships observed with changes in the LM ratio (Fig.
3D), plasma LBP concentrations (data not shown), or fecal
metabolite concentrations (data not shown). Lake Louise
scores were higher in SP relative to HP on HA day 0 [before
the prescribed diets were administered; mean difference 0.6
(95% confidence interval: 0.2, 0.9), P0.004] and on HA day
2[mean difference 0.7 (95% confidence interval: 0.3, 1.1),
P0.001] but did not differ other days (diet-by-day interac-
tion, P0.02). In contrast, AMS severity did not differ
between diet groups on any day (diet-by-day interaction, P
1.0), and frequency of AMS responder/nonresponder classifi-
cation did not differ by diet group (Table 4), indicating that the
incidence of moderate-to-severe AMS was not affected by
dietary protein:fat ratio.
Differences in fecal microbiota composition between AMS
responder groups were observed before ascent to HA. Specif-
ically, LEfSe analyses of SL gut microbiota composition indi-
cated that Prevotella (phyla: Bacteroidetes) relative abundance
at SL was higher in AMS responders relative to nonresponders
(effect size 4.2, P0.006). Furthermore, the Bacteroides:
Prevotella ratio, which is known to drive interindividual vari-
ation in human gut microbiota composition (18), trended to-
ward being lower at SL in the AMS responders (Table 4). We
previously reported that the proportion of total body mass loss
at HA attributable to fat-free mass also demonstrated substan-
tial interindividual variability in this cohort with 7 participants
losing predominantly fat-free mass and 10 losing predomi-
nantly fat mass (9). Interestingly, the mean Bacteroides:Pre-
votella ratio at SL was lower in individuals whose total body
mass loss at HA was 50% fat-free mass relative to those
whose total body mass loss at HA was 50% fat mass [median
(interquartile range): 1 (1) vs. 203 (1,317), P0.01]. Pre-
votella was also the most discriminant taxa between these two
groups, being enriched in the SL microbiota of individuals
whose weight loss at HA comprised 50% fat-free mass
(LEfSe effect size 4.4, P0.0001).
Differences in fecal microbiota composition between AMS
responder groups were also observed after ascent to HA. A
trend for an AMS responder group-by-time interaction (P
0.08) was observed for both the Shannon and observed OTUs
-diversity metrics (Fig. 3E). Post hoc testing indicated differ-
ences were present only during the second week at HA during
which Shannon diversity was 15% higher and observed OTUs
were 45% higher in AMS responders relative to nonresponders
(P0.03 for both). These results were unchanged after
excluding two individuals from the moderate/severe AMS
group who reported constipation during HA2, and bowel
movement frequency during HA2 did not differ between AMS
responder groups [median: nonresponder 5 per week (inter-
quartile range 8) vs. responder 5 per week (interquartile
range 5), P0.56]. These observations indicated that in-
creased GI transit time, which has been associated with greater
-diversity (63), likely did not explain between group differ-
ences. Furthermore, Bray-Curtis dissimilarities demon-
strated clustering over time as a function of AMS responder
group (Fig. 3F; group-by-time interaction, P0.01). Re-
sponder group-by-time interactions (FDR 0.20) were also
observed for six taxa, which were all within the top third of
the most abundant taxa, and all decreased in relative abun-
dance from SL to HA2 in AMS responders relative to
nonresponders (Fig. 3G).
The first aim of this study was to determine the effect of
increasing the dietary protein:fat ratio on gut microbiota com-
position, gut microbiota-derived metabolites, GI barrier func-
tion, and GI symptoms during weight loss at HA. Results
demonstrated that protein:fat ratio had little impact on fecal
microbiota composition and only transient effects on fecal
SCFA and plasma LBP concentrations. These effects did not
appear clinically meaningful as no between-group differences
in intestinal permeability, GI symptoms, incidence of moder-
ate-to-severe AMS, or inflammation were observed. Although
previous studies have reported unfavorable effects of higher
Fig. 3. Acute mountain sickness (AMS) severity is associated with gastrointestinal (GI) symptoms, and changes in fecal microbiota composition. Lake Louise
scores (A), serum cortisol (B), GI symptomology (C), lactulose:mannitol ratio (L:M; D), and -diversity (E) by AMS severity group (no/mild vs.
moderate/severe). AD: data are means SE and analyzed by marginal models or mixed models controlling for diet group, age, body mass index, and baseline
value. *Different from no/mild AMS (P0.05). F: principle coordinates analysis of Bray-Curtis dissimilarities. Individual data points represent the entire fecal
microbiota community of a single individual at one point in time. Samples closer together are more similar than samples farther apart. G: log
-fold change in
relative abundance from SL to HA2 in taxa demonstrating significant AMS group-by-time interactions (false discovery rate 0.20). #Significant within group
difference from SL. HA, high altitude; SL, sea level; Uncl, unassigned genus level taxonomy. No/mild AMS, n6; moderate/severe AMS, n11.
Table 4. Participant characteristics categorized by peak
acute mountain sickness severity measured within 48 h of
ascent to high altitude
AMS Severity
None or mild Moderate or severe Pvalue
SP/HP, n3/3 5/6 0.86
Age, yr 23 4246 1.00
Body mass index, kg/m
24.7 3.7 27.0 3.3 0.20
Body fat, % 20.6 6.7 23.8 5.7 0.30
, ml·kg
51.8 6.3 51.5 8.1 0.96
LM ratio 0.009 [0.005] 0.008 [0.004] 0.72
Shannon 6.2 0.5 6.0 0.7 0.32
Observed OTU 1,313 269 1,154 364 0.15
Bacteroides:Prevotella ratio 234 [2,219] 1 [37] 0.07
Data were measured at sea level and are means SD or median [interquar-
tile range]. Frequencies were measured using
-tests, and means by indepen-
dent samples t-test. Marginal models were used to analyze gut microbiota
metrics, which include 2 measurements at sea level. Acute mountain sickness
was measured by Lake Louise score. HP, higher protein diet group; LM ratio,
lactulose:mannitol ratio measurement of small intestinal permeability; OTU,
operational taxonomic unit; SP, standard-protein diet group; SL, sea level
(weight maintenance); HA, high altitude (4,300 m; energy deficit).
AJP-Gastrointest Liver Physiol doi:10.1152/ajpgi.00253.2018
Downloaded from at US Army Soldier Systems CMD ( on July 15, 2019.
-0.5 -0.4 -0.3 -0.2 -0.1 0.0 0.1 0.2 0.3
PCo2 (9.5%)
PCo1 (13.3%)
P-AMS = 0.002
P-time < 0.001
P-AMS*time = 0.001
moderate/severe AMS no/mild AMS
P-AMS = 0.05
P-time < 0.001
P-AMS*time = 0.39
P-AMS = 0.06
P-time < 0.001
P-AMS*time = 0.03 *
-4 -2 0 2 4
Log2-fold change from SL to HA2
moderate/severe AMS no/mild AMS
P-AMS = 0.62
P-time = 0.84
P-AMS*time = 0.08
P-AMS = 0.56
P-time = 0.99
P-AMS*time = 0.08
Eno/mild AMS
mod/seve re AMS
PCo1 (13.3%)
Time point
P-AMS*time = 0.01
Time pointTime point
Shannon diversity
Observed OTUs
GIQLI score
Lake Louise score
Serum cortisol (µg/dL)
Day at high altitudeDay at high altitude
Mean serum
cortisol (µg/dL)
L:M ratio (% change from SL)
AJP-Gastrointest Liver Physiol doi:10.1152/ajpgi.00253.2018
Downloaded from at US Army Soldier Systems CMD ( on July 15, 2019.
protein diets on fecal microbiota composition and metabolites,
those results were more likely attributable to reductions in fiber
intake because the taxa most consistently affected were sac-
charolytic (12, 19, 24, 25, 65) and decreased in proportion to
reductions in fiber intake (24, 25, 65). Our findings are more
consistent with experiments conducted under normal environ-
mental conditions in which dietary macronutrient composition
was manipulated within recommended intake ranges and fiber
intake was matched between groups (5, 79). Those studies also
did not demonstrate detrimental effects of higher protein diets
on markers of GI health despite evidence of increased protein
fermentation and/or decreased carbohydrate fermentation (5, 8,
79). As such, any impact of gut microbes and their metabolites
on GI barrier function at HA is likely to be independent of
dietary protein:fat ratio within the macronutrient intake ranges
and time period studied.
The second aim of this study was to identify associations
among the gut microbiota, weight loss at HA, and the host
response to HA as measured by AMS severity. Notably, AMS
severity and weight loss at HA were not affected by diet but
demonstrated associations with fecal microbiota composition
and, in the case of AMS severity, GI symptoms, providing
preliminary evidence of a possible role for the gut microbiota
in individual responses to HA.
Recent evidence indicates that the gut microbiota warrants
consideration in studies aiming to explain interindividual vari-
ance in human phenotypes in response to different exposures
(43, 85), and it has been established that the Bacteroides:
Prevotella ratio is a predominant driver of interindividual
variability within the human gut microbiota (18). Intriguingly,
AMS severity, a phenotype known to demonstrate large unex-
plained interindividual variability (7), and body composition
changes during weight loss at HA, which also demonstrated
large interindividual variability (9), were both associated with
an enrichment of Prevotella at SL and a correspondingly lower
Bacteroides:Prevotella ratio. These results suggest that the
ratio of Bacteroides to Prevotella in the gut microbiota, a
well-established driver of interindividual variability in human
gut microbiota composition, may contribute to the interindi-
vidual variability in host responses at HA.
That higher Prevotella abundance was associated with worse
AMS symptomology and greater fat-free mass losses at HA
was somewhat unexpected. A higher ratio of Prevotella to
Bacteroides is generally considered a marker of a healthy
high-fiber, plant-rich diet (18, 32, 80), Prevotella are known
SCFA producers (48), and certain Prevotella species have been
associated with improved glucose homeostasis (43), which
could be beneficial for mitigating impairments in glucose
tolerance common during HA sojourn (84). However, Pre-
votella can also be detrimental depending on the environment
(44). For example, recent studies suggest that Prevotella may
thrive during oxidative stress, promoting intestinal mucus bar-
rier dysfunction and inflammation (26, 66), and some species
can act as opportunistic pathogens (29). Prevotella-dominated
microbiota are also specialized in the degradation of plant
fibers and have decreased lipolytic and proteolytic fermenta-
tion potential (76). This could have altered the availability of
substrates from the low plant-fiber diets used in the present
study thereby impacting body composition. Of note, the Bac-
teroides:Prevotella ratio has been correlated with long-term
habitual dietary patterns but appears to be relatively stable over
short time frames, even in response to substantial dietary
changes (18, 80). As such, to what extent the Bacteroides:
Prevotella ratio provides a useful intervention target for im-
proving host responses to HA is unclear, especially as current
knowledge suggests that diets designed to increase the Bacte-
roides:Prevotella ratio would likely be high in animal fat and
protein and low in fiber (80). Rather, these preliminary findings
suggest that the Bacteroides:Prevotella ratio could serve as a
possible marker for identifying individuals who may be more
susceptible to the effects of HA.
AMS severity was also associated with higher fecal micro-
biota diversity a week after AMS onset. Of note, our group
recently documented an increase in gut microbiota diversity in
association with larger increases in GI permeability in soldiers
engaged in a multiple-stressor military training exercise con-
ducted at low altitude (39). These observations, which correlate
increased diversity with worse responses to stress, contrast
with animal studies reporting decreased microbiota diversity
following exposure to various psychological and physiologic
stressors (10, 71, 72). The discrepancy could reflect differential
impacts of separate stressors on the gut microbiota (38) and/or
the recognized difficulty in extrapolating animal microbiota
studies to humans (56). Interpreting our findings is also com-
plicated by an inability to separate cause and effect. For
example, as higher diversity is generally considered a marker
of a healthy and resilient microbiota (62), the increase in
diversity in individuals with a more severe stress response
observed in our human studies could reflect a beneficial adap-
tation to stress. This hypothesis would be consistent with
murine studies of cold stress in which changes in the gut
microbiota facilitate host acclimatization to the cold (17, 89).
Alternately, our observations may simply reflect greater, tran-
sient changes in the colonic environment in individuals with a
more severe stress response. Future studies are needed to
separate cause and effect in the relationship between the human
gut microbiota and host stress response, while also determining
the extent to which any stress-induced restructuring of the gut
microbiota is sustained.
Independent of AMS symptomology, weight loss at HA was
associated with reduced relative abundance of both Lactoba-
cillus and Turicibacter. The observed decrease in Lactobacil-
lus relative abundance at HA was of particular interest as this
SCFA-producing genus promotes GI barrier integrity, immune
function, and resistance against enteric pathogens (31). The
transient decrease in Turicibacter relative abundance during
HA was also notable as Turicibacter depletion has been re-
ported in rodent models of immunodeficiency (20, 40) and
inflammatory bowel disease (3, 64) in association with in-
creased inflammation (46). Our findings therefore raise the
possibility that the physiologic effects of weight loss at HA
may include Lactobacillus and Turicibacter depletion, which
could in turn contribute to reduced GI barrier integrity, inflam-
mation, altered immune function, and heightened susceptibility
to bacterial pathogens, which have been reported at HA (41,
54). On the other hand, previous studies have not reported a
reduction in Lactobacillus during HA sojourn (Turicibacter
was not measured) (2, 42) or reductions in the relative abun-
dances of these taxa in individuals exposed to normobaric
hypoxia simulating ~4,000-m altitude (68 –70). The latter ob-
servation implies that sustained hypoxia per se is likely not the
mechanism underpinning observed changes in relative abun-
AJP-Gastrointest Liver Physiol doi:10.1152/ajpgi.00253.2018
Downloaded from at US Army Soldier Systems CMD ( on July 15, 2019.
dance and that some combination of hypobaria, hypoxia,
weight loss, increased physical activity, and dietary change is
responsible. Of note, all of these factors are inevitable or
common during HA sojourn (33).
This combination of factors was also associated with a
physiologic stress response and increased inflammation, which
is consistent with previous reports (34, 41), and could drive
changes in gut microbiota composition. In addition, a rapid and
sustained increase in small intestinal permeability was ob-
served. To our knowledge, only one previous study has directly
measured intestinal permeability during HA sojourn. In that
observational study of mountaineers, Dinmore et al. (21) re-
ported increased small intestinal permeability at 5,570 m but
attributed the response to residual effects of GI infection.
However, rodent studies have shown that intestinal permeabil-
ity increases upon exposure to hypobaric hypoxia (82, 88),
attributing the response to inflammation, oxidative stress, at-
rophy, and villous collapse within the GI epithelium (1, 82, 86,
88). Although the present study did not directly address those
mechanisms, the progressive decrease in mannitol excretion is
consistent with the findings of Dinmore et al. (21) and suggests
that small intestinal absorptive capacity and/or surface area
was reduced during weight loss at HA, which could reflect
villous atrophy.
Our data suggest that the pleiotropic hormone GLP-2 could
be one factor impacting GI permeability at HA. GLP-2 is
secreted from intestinal enteroendocrine cells and is thought to
benefit GI function in part by stimulating blood flow, reducing
inflammation, and enhancing GI barrier integrity (14, 22, 23).
Furthermore, elevated GLP-2 concentrations following intesti-
nal injury and in response to fasting have been shown to
facilitate GI growth and repair (22). We hypothesize that a
reduction in enteroendocrine cell density resulting from dimin-
ished intestinal surface area, as suggested by the decrease in
mannitol excretion, could underpin the observed transient de-
crease in GLP-2 concentrations, which in turn could contribute
to an increase in intestinal permeability (14). The subsequent
GLP-2 rebound may reflect an adaptive response to intestinal
injury and underfeeding to promote GI growth and repair. The
inverse correlation with plasma LBP further implicates GLP-2
as having influenced GI barrier function, although the sus-
tained increase in intestinal permeability indicates that other
unmeasured factors also contributed.
Study findings should be interpreted within the context of
several limitations. First, the sample size was small, and some
between-group differences were likely not detected. As such,
the absence of effects of diet composition on many study
outcomes warrants cautious interpretation and requires repli-
cation. Second, the absence of a weight-maintenance control
group at HA precludes determining whether temporal changes
in study outcomes were attributable to hypobaric hypoxia and
acclimatization, weight loss, changes in diet, other factors, or
their combination. Third, study outcomes were not all assessed
at the same time points which prevented examination of or
complicated interpretation of associations between outcomes.
Fourth, functional assessment of GI permeability was limited
to the small intestine, and it cannot be assumed that small
intestinal permeability mirrors that of the large intestine where
gut microbes and their metabolites are more abundant. Addi-
tionally, intestinal permeability was measured only once at SL
which prevented assessing whether any stress related to re-
search study participation was associated with an increase in
intestinal permeability at SL. However, even if such an effect
were present, it did not mask the increases in permeability
observed at HA. Finally, the findings related to weight loss at
HA, AMS severity, and the gut microbiota must be considered
preliminary and hypothesis generating due to the small sample
size and exploratory and correlative nature of the analysis.
In summary, this tightly controlled study integrated stress-
induced changes in physiology with temporal changes in the
gut microbiota and its metabolites, and objective and subjec-
tive measures of GI health and function thereby allowing for an
analysis of multiple steps along the pathway linking environ-
ment, stress, the gut microbiota, and host health. Findings
demonstrated several novel associations between the gut mi-
crobiota and host responses to HA that were independent of
dietary protein:fat ratio. Intriguingly, the ratio of Bacteroides
to Prevotella in the gut microbiota, a well-established driver of
interindividual variability in human gut microbiota composi-
tion, was associated with interindividual variability in host
responses at HA, and increased gut microbiota diversity, con-
sidered a marker of a healthy and resilient microbiota, was
elevated after HA exposure in individuals who experienced a
more severe stress response at HA as measured by AMS
symptomology. Although the study design precluded establish-
ing a causal role for the gut microbiota in host responses to
HA, findings provide preliminary evidence for a potential role,
and support the need for additional research designed to de-
termine if the gut microbiota can be leveraged to improve
physiologic responses to HA.
We thank the study volunteers, our medical oversight team, and Dr. Stephen
Muza for support and the Pennington Biomedical Research Center Clinical
Laboratory staff, Robert Player, Jason Soares, Katherine Kensil, Karen Conca,
Nancy Murphy, Marques Wilson, Christopher Carrigan, Adrienne Hatch,
Laura Lutz, Matthew Kominsky, Adam Luippold, Bradley Anderson, Grant
Holmes, Anthony Karis, Karleigh Bradbury, Alyssa Kelley, Katakyie Sarpong,
Alfonzo Patino, Dr. Renee Cole, Dr. John Carbone, Dr. Lee Margolis, Dr.
Stephen Hennigar, Dr. Robert Kenefick, and Dr. Scott Montain for significant
This work was supported by the U.S. Army Medical Research and Materiel
Command and the U.S. Department of Defense, Defense Health Program.
The opinions or assertions contained herein are the private views of the
authors and are not to be construed as official or reflecting the views of the
Army or the Department of Defense. Any citations of commercial organiza-
tions and trade names in this report do not constitute an official Department of
the Army endorsement or approval of the products or services of these
organizations. Approved for public release; distribution is unlimited.
No conflicts of interest, financial or otherwise, are declared by the authors.
J.P.K., C.E.B., A.J.Y., J.C.R., and S.M.P. conceived and designed research;
J.P.K., C.E.B., A.J.Y., P.N.R., and S.M.P. performed experiments; J.P.K.,
C.E.B., T.A.B., and I.G.P.-F. analyzed data; J.P.K., C.E.B., A.J.Y., I.G.P.-F.,
and S.M.P. interpreted results of experiments; J.P.K. prepared figures; J.P.K.
drafted manuscript; J.P.K., C.E.B., A.J.Y., P.N.R., T.A.B., I.G.P.-F., J.C.R.,
and S.M.P. edited and revised manuscript; J.P.K., C.E.B., A.J.Y., P.N.R.,
T.A.B., I.G.P.-F., J.C.R., and S.M.P. approved final version of manuscript.
AJP-Gastrointest Liver Physiol doi:10.1152/ajpgi.00253.2018
Downloaded from at US Army Soldier Systems CMD ( on July 15, 2019.
1. Adak A, Maity C, Ghosh K, Mondal KC. Alteration of predominant
gastrointestinal flora and oxidative damage of large intestine under sim-
ulated hypobaric hypoxia. Z Gastroenterol 52: 180 –186, 2014. doi:10.
2. Adak A, Maity C, Ghosh K, Pati BR, Mondal KC. Dynamics of
predominant microbiota in the human gastrointestinal tract and change in
luminal enzymes and immunoglobulin profile during high-altitude adap-
tation. Folia Microbiol (Praha) 58: 523–528, 2013. doi:10.1007/s12223-
3. AlShawaqfeh MK, Wajid B, Minamoto Y, Markel M, Lidbury JA,
Steiner JM, Serpedin E, Suchodolski JS. A dysbiosis index to assess
microbial changes in fecal samples of dogs with chronic inflammatory
enteropathy. FEMS Microbiol Ecol 93: 93, 2017. doi:10.1093/femsec/
4. Anand AC, Sashindran VK, Mohan L. Gastrointestinal problems at high
altitude. Trop Gastroenterol 27: 147–153, 2006.
5. Beaumont M, Portune KJ, Steuer N, Lan A, Cerrudo V, Audebert M,
Dumont F, Mancano G, Khodorova N, Andriamihaja M, Airinei G,
Tomé D, Benamouzig R, Davila AM, Claus SP, Sanz Y, Blachier F.
Quantity and source of dietary protein influence metabolite production by
gut microbiota and rectal mucosa gene expression: a randomized, parallel,
double-blind trial in overweight humans. Am J Clin Nutr 106: 1005–1019,
2017. doi:10.3945/ajcn.117.158816.
6. Beidleman BA, Muza SR, Fulco CS, Rock PB, Cymerman A. Valida-
tion of a shortened electronic version of the environmental symptoms
questionnaire. High Alt Med Biol 8: 192–199, 2007. doi:10.1089/ham.
7. Beidleman BA, Tighiouart H, Schmid CH, Fulco CS, Muza SR.
Predictive models of acute mountain sickness after rapid ascent to various
altitudes. Med Sci Sports Exerc 45: 792–800, 2013. doi:10.1249/MSS.
8. Benassi-Evans B, Clifton P, Noakes M, Fenech M. High-protein/high
red meat and high-carbohydrate weight-loss diets do not differ in their
effect on faecal water genotoxicity tested by use of the WIL2-NS cell line
and with other biomarkers of bowel health. Mutat Res 703: 130 –136,
2010. doi:10.1016/j.mrgentox.2010.08.009.
9. Berryman CE, Young AJ, Karl JP, Kenefick RW, Margolis LM, Cole
RE, Carbone JW, Lieberman HR, Kim IY, Ferrando AA, Pasiakos
SM. Severe negative energy balance during 21 d at high altitude decreases
fat-free mass regardless of dietary protein intake: a randomized controlled
trial. FASEB J 32: 894 –905, 2018. doi:10.1096/fj.201700915R.
10. Bharwani A, Mian MF, Foster JA, Surette MG, Bienenstock J,
Forsythe P. Structural & functional consequences of chronic psychosocial
stress on the microbiome & host. Psychoneuroendocrinology 63: 217–
227, 2016. doi:10.1016/j.psyneuen.2015.10.001.
11. Bolger AM, Lohse M, Usadel B. Trimmomatic: a flexible trimmer for
Illumina sequence data. Bioinformatics 30: 2114 –2120, 2014. doi:10.
12. Brinkworth GD, Noakes M, Clifton PM, Bird AR. Comparative
effects of very low-carbohydrate, high-fat and high-carbohydrate, low-
fat weight-loss diets on bowel habit and faecal short-chain fatty acids
and bacterial populations. Br J Nutr 101: 1493–1502, 2009. doi:10.
13. Cani PD. Crosstalk between the gut microbiota and the endocannabinoid
system: impact on the gut barrier function and the adipose tissue. Clin
Microbiol Infect 18, Suppl 4: 50 –53, 2012. doi:10.1111/j.1469-0691.2012.
14. Cani PD, Possemiers S, Van de Wiele T, Guiot Y, Everard A, Rottier
O, Geurts L, Naslain D, Neyrinck A, Lambert DM, Muccioli GG,
Delzenne NM. Changes in gut microbiota control inflammation in obese
mice through a mechanism involving GLP-2-driven improvement of gut
permeability. Gut 58: 1091–1103, 2009. doi:10.1136/gut.2008.165886.
15. Caporaso JG, Bittinger K, Bushman FD, DeSantis TZ, Andersen
GL, Knight R. PyNAST: a flexible tool for aligning sequences to a
template alignment. Bioinformatics 26: 266 –267, 2010. doi:10.1093/
16. Caporaso JG, Kuczynski J, Stombaugh J, Bittinger K, Bushman FD,
Costello EK, Fierer N, Peña AG, Goodrich JK, Gordon JI, Huttley
GA, Kelley ST, Knights D, Koenig JE, Ley RE, Lozupone CA,
McDonald D, Muegge BD, Pirrung M, Reeder J, Sevinsky JR, Turn-
baugh PJ, Walters WA, Widmann J, Yatsunenko T, Zaneveld J,
Knight R. QIIME allows analysis of high-throughput community se-
quencing data. Nat Methods 7: 335–336, 2010. doi:10.1038/nmeth.f.303.
17. Chevalier C, Stojanovic´ O, Colin DJ, Suarez-Zamorano N, Tarallo V,
Veyrat-Durebex C, Rigo D, Fabbiano S, Stevanovic´ A, Hagemann S,
Montet X, Seimbille Y, Zamboni N, Hapfelmeier S, Trajkovski M. Gut
microbiota orchestrates energy homeostasis during cold. Cell 163: 1360 –
1374, 2015. doi:10.1016/j.cell.2015.11.004.
18. Costea PI, Hildebrand F, Arumugam M, Bäckhed F, Blaser MJ, Bush-
man FD, de Vos WM, Ehrlich SD, Fraser CM, Hattori M, Huttenhower
C, Jeffery IB, Knights D, Lewis JD, Ley RE, Ochman H, O’Toole PW,
Quince C, Relman DA, Shanahan F, Sunagawa S, Wang J, Weinstock
GM, Wu GD, Zeller G, Zhao L, Raes J, Knight R, Bork P. Enterotypes in
the landscape of gut microbial community composition. Nat Microbiol 3:
8 –16, 2018. [Erratum in: Nat Microbiol 3: 388, 2018. 10.1038/s41564-018-
0114-x. 29440750.] doi:10.1038/s41564-017-0072-8.
19. David LA, Maurice CF, Carmody RN, Gootenberg DB, Button JE,
Wolfe BE, Ling AV, Devlin AS, Varma Y, Fischbach MA, Biddinger
SB, Dutton RJ, Turnbaugh PJ. Diet rapidly and reproducibly alters the
human gut microbiome. Nature 505: 559 –563, 2014. doi:10.1038/
20. Dimitriu PA, Boyce G, Samarakoon A, Hartmann M, Johnson P,
Mohn WW. Temporal stability of the mouse gut microbiota in relation to
innate and adaptive immunity. Environ Microbiol Rep 5: 200 –210, 2013.
21. Dinmore AJ, Edwards JS, Menzies IS, Travis SP. Intestinal carbohy-
drate absorption and permeability at high altitude (5,730 m). J Appl
Physiol (1985) 76: 1903–1907, 1994. doi:10.1152/jappl.1994.76.5.1903.
22. Drucker DJ, Yusta B. Physiology and pharmacology of the enteroendo-
crine hormone glucagon-like peptide-2. Annu Rev Physiol 76: 561–583,
2014. doi:10.1146/annurev-physiol-021113-170317.
23. Dubé PE, Brubaker PL. Frontiers in glucagon-like peptide-2: multiple
actions, multiple mediators. Am J Physiol Endocrinol Metab 293: E460 –
E465, 2007. doi:10.1152/ajpendo.00149.2007.
24. Duncan SH, Belenguer A, Holtrop G, Johnstone AM, Flint HJ, Lobley
GE. Reduced dietary intake of carbohydrates by obese subjects results in
decreased concentrations of butyrate and butyrate-producing bacteria in
feces. Appl Environ Microbiol 73: 1073–1078, 2007. doi:10.1128/AEM.
25. Duncan SH, Lobley GE, Holtrop G, Ince J, Johnstone AM, Louis P,
Flint HJ. Human colonic microbiota associated with diet, obesity and
weight loss. Int J Obes 32: 1720 –1724, 2008. doi:10.1038/ijo.2008.155.
26. Elinav E, Strowig T, Kau AL, Henao-Mejia J, Thaiss CA, Booth CJ,
Peaper DR, Bertin J, Eisenbarth SC, Gordon JI, Flavell RA. NLRP6
inflammasome regulates colonic microbial ecology and risk for colitis.
Cell 145: 745–757, 2011. doi:10.1016/j.cell.2011.04.022.
27. Eypasch E, Williams JI, Wood-Dauphinee S, Ure BM, Schmülling C,
Neugebauer E, Troidl H. Gastrointestinal Quality of Life Index: devel-
opment, validation and application of a new instrument. Br J Surg 82:
216 –222, 1995. doi:10.1002/bjs.1800820229.
28. Falony G, Joossens M, Vieira-Silva S, Wang J, Darzi Y, Faust K,
Kurilshikov A, Bonder MJ, Valles-Colomer M, Vandeputte D, Tito
RY, Chaffron S, Rymenans L, Verspecht C, De Sutter L, Lima-
Mendez G, D’hoe K, Jonckheere K, Homola D, Garcia R, Tigchelaar
EF, Eeckhaudt L, Fu J, Henckaerts L, Zhernakova A, Wijmenga C,
Raes J. Population-level analysis of gut microbiome variation. Science
352: 560 –564, 2016. doi:10.1126/science.aad3503.
29. Finegold SM. Anaerobic gram-negative Bacilli (4th ed.). In: Medical
Microbiology, edited by Baron S. Galveston, TX: The University of Texas
Medical Branch at Galveston, 1996.
30. Francis CY, Morris J, Whorwell PJ. The irritable bowel severity scoring
system: a simple method of monitoring irritable bowel syndrome and its
progress. Aliment Pharmacol Ther 11: 395–402, 1997. doi:10.1046/j.
31. Gibson GR, Hutkins R, Sanders ME, Prescott SL, Reimer RA, Salmi-
nen SJ, Scott K, Stanton C, Swanson KS, Cani PD, Verbeke K, Reid
G. Expert consensus document: The International Scientific Association
for Probiotics and Prebiotics (ISAPP) consensus statement on the defini-
tion and scope of prebiotics. Nat Rev Gastroenterol Hepatol 14: 491–502,
2017. doi:10.1038/nrgastro.2017.75.
32. Gorvitovskaia A, Holmes SP, Huse SM. Interpreting Prevotella and
Bacteroides as biomarkers of diet and lifestyle. Microbiome 4: 15, 2016.
AJP-Gastrointest Liver Physiol doi:10.1152/ajpgi.00253.2018
Downloaded from at US Army Soldier Systems CMD ( on July 15, 2019.
33. Hamad N, Travis SP. Weight loss at high altitude: pathophysiology and
practical implications. Eur J Gastroenterol Hepatol 18: 5–10, 2006.
34. Hartmann G, Tschöp M, Fischer R, Bidlingmaier C, Riepl R, Tschöp
K, Hautmann H, Endres S, Toepfer M. High altitude increases circu-
lating interleukin-6, interleukin-1 receptor antagonist and C-reactive pro-
tein. Cytokine 12: 246 –252, 2000. doi:10.1006/cyto.1999.0533.
35. Holmes E, Li JV, Marchesi JR, Nicholson JK. Gut microbiota compo-
sition and activity in relation to host metabolic phenotype and disease risk.
Cell Metab 16: 559 –564, 2012. doi:10.1016/j.cmet.2012.10.007.
36. Hooper LV, Littman DR, Macpherson AJ. Interactions between the
microbiota and the immune system. Science 336: 1268 –1273, 2012.
37. Karl JP, Cole RE, Berryman CE, Finlayson G, Radcliffe PN, Komin-
sky MT, Murphy NE, Carbone JW, Rood JC, Young AJ, Pasiakos
SM. Appetite suppression and altered food preferences coincide with
changes in appetite-mediating hormones during energy deficit at high
altitude, but are not affected by protein intake. High Alt Med Biol 19:
156 –169, 2018. doi:10.1089/ham.2017.0155.
38. Karl JP, Hatch AM, Arcidiacono SM, Pearce SC, Pantoja-Feliciano
IG, Doherty LA, Soares JW. Effects of psychological, environmental
and physical stressors on the gut microbiota. Front Microbiol 9: 2013,
2018. doi:10.3389/fmicb.2018.02013.
39. Karl JP, Margolis LM, Madslien EH, Murphy NE, Castellani JW,
Gundersen Y, Hoke AV, Levangie MW, Kumar R, Chakraborty N,
Gautam A, Hammamieh R, Martini S, Montain SJ, Pasiakos SM.
Changes in intestinal microbiota composition and metabolism coincide
with increased intestinal permeability in young adults under prolonged
physiological stress. Am J Physiol Gastrointest Liver Physiol 312: G559 –
G571, 2017. doi:10.1152/ajpgi.00066.2017.
40. Kellermayer R, Dowd SE, Harris RA, Balasa A, Schaible TD, Wolcott
RD, Tatevian N, Szigeti R, Li Z, Versalovic J, Smith CW. Colonic
mucosal DNA methylation, immune response, and microbiome patterns in
Toll-like receptor 2-knockout mice. FASEB J 25: 1449 –1460, 2011.
41. Khanna K, Mishra KP, Ganju L, Kumar B, Singh SB. High-altitude-
induced alterations in gut-immune axis: a review. Int Rev Immunol 37:
119 –126, 2018. doi:10.1080/08830185.2017.1407763.
42. Kleessen B, Schroedl W, Stueck M, Richter A, Rieck O, Krueger M.
Microbial and immunological responses relative to high-altitude exposure
in mountaineers. Med Sci Sports Exerc 37: 1313–1318, 2005. doi:10.1249/
43. Kovatcheva-Datchary P, Nilsson A, Akrami R, Lee YS, De Vadder F,
Arora T, Hallen A, Martens E, Björck I, Bäckhed F. Dietary fiber-
induced improvement in glucose metabolism is associated with increased
abundance of Prevotella. Cell Metab 22: 971–982, 2015. doi:10.1016/j.
44. Ley RE. Gut microbiota in 2015: Prevotella in the gut: choose carefully.
Nat Rev Gastroenterol Hepatol 13: 69 –70, 2016. doi:10.1038/nrgastro.
45. Li X, Kan EM, Lu J, Cao Y, Wong RK, Keshavarzian A, Wilder-
Smith CH. Combat-training increases intestinal permeability, immune
activation and gastrointestinal symptoms in soldiers. Aliment Pharmacol
Ther 37: 799 –809, 2013. doi:10.1111/apt.12269.
46. Liu W, Crott JW, Lyu L, Pfalzer AC, Li J, Choi SW, Yang Y,
Mason JB, Liu Z. Diet- and genetically-induced obesity produces
alterations in the microbiome, inflammation and Wnt pathway in the
intestine of Apc
mice: comparisons and contrasts. J Cancer 7:
1780 –1790, 2016. doi:10.7150/jca.15792.
47. Love MI, Huber W, Anders S. Moderated estimation of fold change and
dispersion for RNA-seq data with DESeq2. Genome Biol 15: 550, 2014.
48. Macfarlane GT, Macfarlane S. Bacteria, colonic fermentation, and
gastrointestinal health. J AOAC Int 95: 50 –60, 2012. doi:10.5740/jaoacint.
49. Mackos AR, Maltz R, Bailey MT. The role of the commensal microbiota
in adaptive and maladaptive stressor-induced immunomodulation. Horm
Behav 88: 70 –78, 2017. doi:10.1016/j.yhbeh.2016.10.006.
50. Margolis LM, Carbone JW, Berryman CE, Carrigan CT, Murphy
NE, Ferrando AA, Young AJ, Pasiakos SM. Severe energy deficit at
high altitude inhibits skeletal muscle mTORC1-mediated anabolic
signaling without increased ubiquitin proteasome activity. FASEB J
fj.201800163RR, 2018. doi:10.1096/fj.201800163RR.
51. Marsilio R, D’Antiga L, Zancan L, Dussini N, Zacchello F. Simulta-
neous HPLC determination with light-scattering detection of lactulose and
mannitol in studies of intestinal permeability in pediatrics. Clin Chem 44:
1685–1691, 1998.
52. McDonald D, Price MN, Goodrich J, Nawrocki EP, DeSantis TZ,
Probst A, Andersen GL, Knight R, Hugenholtz P. An improved
Greengenes taxonomy with explicit ranks for ecological and evolutionary
analyses of bacteria and archaea. ISME J 6: 610 –618, 2012. doi:10.1038/
53. McOmber ME, Ou CN, Shulman RJ. Effects of timing, sex, and age on
site-specific gastrointestinal permeability testing in children and adults. J
Pediatr Gastroenterol Nutr 50: 269 –275, 2010. doi:10.1097/MPG.
54. Meehan RT. Immune suppression at high altitude. Ann Emerg Med 16:
974 –979, 1987. doi:10.1016/S0196-0644(87)80743-6.
55. Netzer N, Strohl K, Faulhaber M, Gatterer H, Burtscher M. Hypoxia-
related altitude illnesses. J Travel Med 20: 247–255, 2013. doi:10.1111/
56. Nguyen TL, Vieira-Silva S, Liston A, Raes J. How informative is the
mouse for human gut microbiota research? Dis Model Mech 8: 1–16, 2015.
57. O’Keefe SJ, Li JV, Lahti L, Ou J, Carbonero F, Mohammed K,
Posma JM, Kinross J, Wahl E, Ruder E, Vipperla K, Naidoo V,
Mtshali L, Tims S, Puylaert PG, DeLany J, Krasinskas A, Benefiel
AC, Kaseb HO, Newton K, Nicholson JK, de Vos WM, Gaskins HR,
Zoetendal EG. Fat, fibre and cancer risk in African Americans and rural
Africans. Nat Commun 6: 6342, 2015. doi:10.1038/ncomms7342.
58. Pasiakos SM, Berryman CE, Carbone JW, Murphy NE, Carrigan CT,
Bamman MM, Ferrando AA, Young AJ, Margolis LM. Muscle Fn14
gene expression is associated with fat-free mass retention during energy
deficit at high altitude. Physiol Rep 6: e13801, 2018. doi:10.14814/phy2.
59. Pasiakos SM, Berryman CE, Carrigan CT, Young AJ, Carbone JW.
Muscle protein turnover and the molecular regulation of muscle mass
during hypoxia. Med Sci Sports Exerc 49: 1340 –1350, 2017. doi:10.1249/
60. Pasiakos SM, Cao JJ, Margolis LM, Sauter ER, Whigham LD,
McClung JP, Rood JC, Carbone JW, Combs GF Jr, Young AJ. Effects
of high-protein diets on fat-free mass and muscle protein synthesis
following weight loss: a randomized controlled trial. FASEB J 27: 3837–
3847, 2013. doi:10.1096/fj.13-230227.
61. Patrone V, Vajana E, Minuti A, Callegari ML, Federico A, Loguercio
C, Dallio M, Tolone S, Docimo L, Morelli L. Postoperative changes in
fecal bacterial communities and fermentation products in obese patients
undergoing bilio-intestinal bypass. Front Microbiol 7: 200, 2016. doi:10.
62. Petersen C, Round JL. Defining dysbiosis and its influence on host
immunity and disease. Cell Microbiol 16: 1024 –1033, 2014. doi:10.1111/
63. Roager HM, Hansen LB, Bahl MI, Frandsen HL, Carvalho V, Gøbel
RJ, Dalgaard MD, Plichta DR, Sparholt MH, Vestergaard H, Hansen
T, Sicheritz-Pontén T, Nielsen HB, Pedersen O, Lauritzen L, Kris-
tensen M, Gupta R, Licht TR. Colonic transit time is related to bacterial
metabolism and mucosal turnover in the gut. Nat Microbiol 1: 16093,
2016. doi:10.1038/nmicrobiol.2016.93.
64. Rossi G, Pengo G, Caldin M, Palumbo Piccionello A, Steiner JM,
Cohen ND, Jergens AE, Suchodolski JS. Comparison of microbiologi-
cal, histological, and immunomodulatory parameters in response to treat-
ment with either combination therapy with prednisone and metronidazole
or probiotic VSL#3 strains in dogs with idiopathic inflammatory bowel
disease. PLoS One 9: e94699, 2014. doi:10.1371/journal.pone.0094699.
65. Russell WR, Gratz SW, Duncan SH, Holtrop G, Ince J, Scobbie L,
Duncan G, Johnstone AM, Lobley GE, Wallace RJ, Duthie GG, Flint
HJ. High-protein, reduced-carbohydrate weight-loss diets promote metab-
olite profiles likely to be detrimental to colonic health. Am J Clin Nutr 93:
1062–1072, 2011. doi:10.3945/ajcn.110.002188.
66. Scher JU, Sczesnak A, Longman RS, Segata N, Ubeda C, Bielski C,
Rostron T, Cerundolo V, Pamer EG, Abramson SB, Huttenhower C,
Littman DR. Expansion of intestinal Prevotella copri correlates with
enhanced susceptibility to arthritis. eLife 2: e01202, 2013. doi:10.7554/
67. Segata N, Izard J, Waldron L, Gevers D, Miropolsky L, Garrett WS,
Huttenhower C. Metagenomic biomarker discovery and explanation.
Genome Biol 12: R60, 2011. doi:10.1186/gb-2011-12-6-r60.
AJP-Gastrointest Liver Physiol doi:10.1152/ajpgi.00253.2018
Downloaded from at US Army Soldier Systems CMD ( on July 15, 2019.
68. Šket R, Debevec T, Kublik S, Schloter M, Schoeller A, Murovec B,
Vogel Mikuš K, Makuc D, Pecˇnik K, Plavec J, Mekjavic´ IB, Eiken O,
Prevoršek Z, Stres B. Intestinal metagenomes and metabolomes in
healthy young males: inactivity and hypoxia generated negative physio-
logical symptoms precede microbial dysbiosis. Front Physiol 9: 198,
2018. doi:10.3389/fphys.2018.00198.
69. Šket R, Treichel N, Debevec T, Eiken O, Mekjavic I, Schloter M, Vital
M, Chandler J, Tiedje JM, Murovec B, Prevoršek Z, Stres B. Hypoxia
and inactivity related physiological changes (constipation, inflammation)
are not reflected at the level of gut metabolites and butyrate producing
microbial community: The PlanHab Study. Front Physiol 8: 250, 2017.
70. Šket R, Treichel N, Kublik S, Debevec T, Eiken O, Mekjavic´I,
Schloter M, Vital M, Chandler J, Tiedje JM, Murovec B, Prevoršek Z,
Likar M, Stres B. Hypoxia and inactivity related physiological changes
precede or take place in absence of significant rearrangements in bacterial
community structure: The PlanHab randomized trial pilot study. PLoS One
12: e0188556, 2017. doi:10.1371/journal.pone.0188556.
71. Sohail MU, Hume ME, Byrd JA, Nisbet DJ, Shabbir MZ, Ijaz A,
Rehman H. Molecular analysis of the caecal and tracheal microbiome of
heat-stressed broilers supplemented with prebiotic and probiotic. Avian
Pathol 44: 67–74, 2015. doi:10.1080/03079457.2015.1004622.
72. Thaiss CA, Zeevi D, Levy M, Zilberman-Schapira G, Suez J, Tengeler
AC, Abramson L, Katz MN, Korem T, Zmora N, Kuperman Y, Biton
I, Gilad S, Harmelin A, Shapiro H, Halpern Z, Segal E, Elinav E.
Transkingdom control of microbiota diurnal oscillations promotes meta-
bolic homeostasis. Cell 159: 514 –529, 2014. doi:10.1016/j.cell.2014.09.
73. Turner JR. Intestinal mucosal barrier function in health and disease. Nat
Rev Immunol 9: 799 –809, 2009. doi:10.1038/nri2653.
74. van Wijck K, Lenaerts K, Grootjans J, Wijnands KA, Poeze M, van
Loon LJ, Dejong CH, Buurman WA. Physiology and pathophysiology
of splanchnic hypoperfusion and intestinal injury during exercise: strate-
gies for evaluation and prevention. Am J Physiol Gastrointest Liver
Physiol 303: G155–G168, 2012. doi:10.1152/ajpgi.00066.2012.
75. Verbeke KA, Boobis AR, Chiodini A, Edwards CA, Franck A, Kleere-
bezem M, Nauta A, Raes J, van Tol EA, Tuohy KM. Towards microbial
fermentation metabolites as markers for health benefits of prebiotics. Nutr
Res Rev 28: 42–66, 2015. doi:10.1017/S0954422415000037.
76. Vieira-Silva S, Falony G, Darzi Y, Lima-Mendez G, Garcia Yunta R,
Okuda S, Vandeputte D, Valles-Colomer M, Hildebrand F, Chaffron
S, Raes J. Species-function relationships shape ecological properties of
the human gut microbiome. Nat Microbiol 1: 16088, 2016. doi:10.1038/
77. Wang Q, Garrity GM, Tiedje JM, Cole JR. Naive Bayesian classifier
for rapid assignment of rRNA sequences into the new bacterial taxon-
omy. Appl Environ Microbiol 73: 5261–5267, 2007. doi:10.1128/AEM.
78. Westerterp KR, Kayser B. Body mass regulation at altitude. Eur J
Gastroenterol Hepatol 18: 1–3, 2006. doi:10.1097/00042737-200601000-
79. Windey K, De Preter V, Louat T, Schuit F, Herman J, Vansant G,
Verbeke K. Modulation of protein fermentation does not affect fecal
water toxicity: a randomized cross-over study in healthy subjects. PLoS
One 7: e52387, 2012. doi:10.1371/journal.pone.0052387.
80. Wu GD, Chen J, Hoffmann C, Bittinger K, Chen YY, Keilbaugh SA,
Bewtra M, Knights D, Walters WA, Knight R, Sinha R, Gilroy E,
Gupta K, Baldassano R, Nessel L, Li H, Bushman FD, Lewis JD.
Linking long-term dietary patterns with gut microbial enterotypes. Science
334: 105–108, 2011. doi:10.1126/science.1208344.
81. Wycherley TP, Moran LJ, Clifton PM, Noakes M, Brinkworth GD.
Effects of energy-restricted high-protein, low-fat compared with standard-
protein, low-fat diets: a meta-analysis of randomized controlled trials. Am
J Clin Nutr 96: 1281–1298, 2012. doi:10.3945/ajcn.112.044321.
82. Xu CL, Sun R, Qiao XJ, Xu CC, Shang XY, Niu WN. Protective effect
of glutamine on intestinal injury and bacterial community in rats exposed
to hypobaric hypoxia environment. World J Gastroenterol 20: 4662–4674,
2014. doi:10.3748/wjg.v20.i16.4662.
83. Yao CK, Muir JG, Gibson PR. Review article: insights into colonic
protein fermentation, its modulation and potential health implications.
Aliment Pharmacol Ther 43: 181–196, 2016. doi:10.1111/apt.13456.
84. Young AJ, Berryman CE, Kenefick RW, Derosier AN, Margolis LM,
Wilson MA, Carrigan CT, Murphy NE, Carbone JW, Rood JC,
Pasiakos SM. Altitude acclimatization alleviates the hypoxia-induced
suppression of exogenous glucose oxidation during steady-state aerobic
exercise. Front Physiol 9: 830, 2018. doi:10.3389/fphys.2018.00830.
85. Zeevi D, Korem T, Zmora N, Israeli D, Rothschild D, Weinberger A,
Ben-Yacov O, Lador D, Avnit-Sagi T, Lotan-Pompan M, Suez J,
Mahdi JA, Matot E, Malka G, Kosower N, Rein M, Zilberman-
Schapira G, Dohnalová L, Pevsner-Fischer M, Bikovsky R, Halpern
Z, Elinav E, Segal E. Personalized nutrition by prediction of glycemic
responses. Cell 163: 1079 –1094, 2015. doi:10.1016/j.cell.2015.11.001.
86. Zhang F, Wu W, Deng Z, Zheng X, Zhang J, Deng S, Chen J, Ma Q,
Wang Y, Yu X, Kang S, Wang X. High altitude increases the expression
of hypoxia-inducible factor 1and inducible nitric oxide synthase with
intest-inal mucosal barrier failure in rats. Int J Clin Exp Pathol 8:
5189 –5195, 2015.
87. Zhang X, Zhao Y, Zhang M, Pang X, Xu J, Kang C, Li M, Zhang C,
Zhang Z, Zhang Y, Li X, Ning G, Zhao L. Structural changes of gut
microbiota during berberine-mediated prevention of obesity and insulin
resistance in high-fat diet-fed rats. PLoS One 7: e42529, 2012. doi:10.
88. Zhou QQ, Yang DZ, Luo YJ, Li SZ, Liu FY, Wang GS. Over-starvation
aggravates intestinal injury and promotes bacterial and endotoxin translo-
cation under high-altitude hypoxic environment. World J Gastroenterol
17: 1584 –1593, 2011. doi:10.3748/wjg.v17.i12.1584.
89. Zie
˛tak M, Kovatcheva-Datchary P, Markiewicz LH, Ståhlman M,
Kozak LP, Bäckhed F. Altered microbiota contributes to reduced diet-
induced obesity upon cold exposure. Cell Metab 23: 1216 –1223, 2016.
AJP-Gastrointest Liver Physiol doi:10.1152/ajpgi.00253.2018
Downloaded from at US Army Soldier Systems CMD ( on July 15, 2019.
... The analyses reported herein used archived samples and data from a study designed to assess the efficacy of a higher protein diet for preserving fat-free mass during high altitude (HA; 4,300 m) sojourn . The present analyses were conceived after trial completion to explore whether measuring urine metabolites could provide insight into observed inter-individual variability in AMS severity (Karl et al., 2018a). ...
... Peak scores were recorded from all participants during the first 48 h at HA, and were used to categorize AMS severity as mild (≥0.7 and <1.53), moderate (≥1.53 and <2.63), and severe (≥2.63; Beidleman et al., 2013;Karl et al., 2018a). The analyses described below used two group identifiers: AMS and NoAMS. ...
... The mean peak AMS-weighted cerebral factor score for AMS individuals (2.25 ± 0.18; n = 11) was significantly elevated (p < 0.05) compared to in NoAMS subjects (0.78 ± 0.18; n = 6). AMS severity (i.e., NoAMS vs. AMS) was unrelated to diet group (Karl et al., 2018a). PCA analysis indicated that the urinary metabolite profiles for both groups changed over the time course of the study with the AMS group displaying greater variation in data at HA1 compared to NoAMS (Figure 1). ...
Full-text available
Human Molecular and Physiological Responses to Hypoxia Towards the end of the 19th century, the French physician Denis Jourdan was the first to understand and state the critical role of the reduction of oxygen at altitude, which he defined as anoxemia. This term indicated the diminished quantity of oxygen contained in the blood of people living at high altitude, where the tension of the oxygen in the surrounding air is considerably decreased (West and Richalet, 2013). In the following 150 years, studies on hypoxia took off, ranging from purely clinical and functional aspects to cellular and biomolecular ones, from acute to chronic hypoxia and analyzing not only the altitude-hypoxia but also the hypoxia related to underlying diseases. Currently, the study of pathophysiological responses at altitude is a model to investigate the mechanisms of response to hypoxia in any condition, also in critical illnesses (Grocott et al., 2007). In this special issue, a series of ten articles with different approaches applied to the study of molecular and physiological responses to hypoxia were collected.
... In recent years, the effects of hypoxia on the gut microbiota attracted much attention (Mazel, 2019;Han et al., 2021). Short-term or chronic exposure to hypoxia can influence the composition and diversity of the gut microbiota (Li and Zhao, 2015;Karl et al., 2018;Suzuki et al., 2019;Jia et al., 2020), thus helping the hosts to adapt to the environment (Li et al., 2016(Li et al., , 2018(Li et al., , 2019Sun et al., 2019;Ma et al., 2021). Current findings indicated that hypoxia exposure leads to intestinal hypoxia, thus promoting the growth of anaerobic bacteria (Suzuki et al., 2019;Jia et al., 2020). ...
... However, the interventions considerably varied with regards to the hypoxia model such as exposure time and oxygen concentration, and to the confounders such as exercise or diet. Thus, the influences of hypoxia exposure on the gut microbiota are still controversial (Li and Zhao, 2015;Karl et al., 2018;Suzuki et al., 2019;Han et al., 2020Han et al., , 2021Jia et al., 2020). ...
... It was widely accepted that hypoxia exposure significantly decreased the aerobic bacteria and increased the anaerobic bacteria (Mazel, 2019;Han et al., 2021). However, most studies have not excluded the interference of confounding factors such as genes, diet, exercise, and hypoxia exposure model (Li and Zhao, 2015;Karl et al., 2018;Suzuki et al., 2019;Jia et al., 2020;Montoya-Ciriaco et al., 2020). Kleessen et al. (2005) reported that mountaineers exposed to high altitudes above 5,000 m have decreased beneficial Bifidobacteria and increased potentially pathogenic gram-negative bacteria such as Enterobacteriaceae that bring health risks. ...
Full-text available
Hypoxia environment has been widely used to promote exercise capacity. However, the underlying mechanisms still need to be further elucidated. In this study, mice were exposed to the normoxia environment (21% O 2 ) or hypoxia environment (16.4% O 2 ) for 4 weeks. Hypoxia-induced gut microbiota remodeling characterized by the increased abundance of Akkermansia and Bacteroidetes genera, and their related short-chain fatty acids (SCFAs) production. It was observed that hypoxia markedly improved endurance by significantly prolonging the exhaustive running time, promoting mitochondrial biogenesis, and ameliorating exercise fatigue biochemical parameters, including urea nitrogen, creatine kinase, and lactic acid, which were correlated with the concentrations of SCFAs. Additionally, the antibiotics experiment partially inhibited hypoxia-induced mitochondrial synthesis. The microbiota transplantation experiment demonstrated that the enhancement of endurance capacity induced by hypoxia was transferable, indicating that the beneficial effects of hypoxia on exercise performance were partly dependent on the gut microbiota. We further identified that acetate and butyrate, but not propionate, stimulated mitochondrial biogenesis and promoted endurance performance. Our results suggested that hypoxia exposure promoted endurance capacity partially by the increased production of SCFAs derived from gut microbiota remodeling.
... Dysbiosis of gut microbiota has been found to be associated with the pathogenesis of PCOS. [90] Rheumatoid arthritis (RA) Change in gut microbiota has been associated with the pathogenesis of RA. [91] High-altitude (HA) sickness Gut microbiota may contribute to variability in host responses to HA. [92] Hypertension Variation in gut microbial parameters was likely associated with Chinese patients with hypertension. ...
Full-text available
A balanced microbiota composition is requisite for normal physiological functions of the human body. However, several environmental factors such as air pollutants may perturb the human microbiota composition. It is noticeable that currently around 99% of the world's population is breathing polluted air. Air pollution's debilitating health impacts have been studied scrupulously, including in the human gut microbiota. Nevertheless, air pollution's impact on other microbiotas of the human body is less understood so far. In the present review, the authors have summarized and discussed recent studies' outcomes related to air pollution-driven microbiotas' dysbiosis (including oral, nasal, respiratory, gut, skin, and thyroid microbiotas) and its potential multi-organ health risks.
... The increase in community diversity, such as Shannon and Simpson, in mice under acute hypoxia exposure observed in our study could reflect a beneficial response to environmental stress. This finding is consistent with individuals who experienced acute mountain sickness (Karl et al. 2018). Moreover, Jiang et al. (2019) reported an elevated microbiome alpha diversity in mice during spaceflight. ...
Full-text available
Gut microbiota bears adaptive potential to different environments, but little is known regarding its responses to acute high-altitude exposure. This study aimed to evaluate the microbial changes after acute exposure to simulated high-altitude hypoxia. C57BL/6 J mice were divided into hypoxia and normoxia groups. The hypoxia group was exposed to a simulated altitude of 5500 m for 24 h above sea level. The normoxia group was maintained in low altitude of 10 m above sea level. Colonic microbiota was analyzed using 16S rRNA V4 gene sequencing. Compared with the normoxia group, Shannon, Simpson and Akkermansia were significantly increased, while Firmicutes-to-Bacteroidetes ratio and Bifidobacterium were significantly decreased in the hypoxia group. The hypoxia group exhibited lower mobile element containing and higher potentially pathogenic and stress-tolerant phenotypes than those in the normoxia group. Functional analysis indicated that environmental information processing was significantly lower, metabolism, cellular processes and organismal systems were significantly higher in the hypoxia group than those in the normoxia group. In conclusion, acute exposure to simulated high-altitude hypoxia alters gut microbiota diversity and composition, which may provide a potential target to alleviate acute high-altitude diseases.
... We acknowledge that the acute exercise protocol used in the present study likely lacks some generalizability. Karl et al. (2018) A potential limitation of the present study was the difference in the relative exercise intensity between the NORM and HYP trials. While we attempted to match for the exercise intensity via workload (watts) and absolute oxygen consumption (l min −1 ), the lowerV O 2 max which has been reported at high altitudes (Calbet et al., 2003) (Swenson et al., 2020). ...
New finding: What is the central question of this study? What is the effect of hypobaric hypoxia on markers of exercise-induced intestinal injury and symptoms of GI distress? What is the main finding and its importance? Exercise performed at 4300 m of simulated altitude increased I-FABP, CLDN-3, and LBP which together suggest that exercise-induced intestinal injury may be aggravated by concurrent hypoxic exposure. Increases in I-FABP, LBP, CLDN-3 were correlated to exercise-induced GI symptoms, providing some evidence of a link between intestinal barrier injury and symptoms of GI distress. Abstract: We sought to determine the effect of exercise in hypobaric hypoxia on markers of intestinal injury and gastrointestinal (GI) symptoms. Using a randomized and counterbalanced design, 9 males completed two experimental trials: one at local altitude of 1585 m (NORM) and one at 4300 m of simulated hypobaric hypoxia (HYP). Participants performed 60-minutes of cycling at a workload that elicited 65% of their NORM VO2 max. GI symptoms were assessed before and every 15-minutes during exercise. Pre- and post-exercise blood samples were assessed for intestinal fatty acid binding protein (I-FABP), claudin-3 (CLDN-3), and lipopolysaccharide binding protein (LBP). All participants reported at least one GI symptom in HYP compared to just 1 participant in NORM. I-FABP significantly increased from pre- to post-exercise in HYP (708±191 to 1215±518 pg mL-1 ; p = 0.011, d = 1.10) but not NORM (759±224 to 828±288 pg mL-1 ; p>0.99, d = 0.27). CLDN-3 significantly increased from pre- to post-exercise in HYP (13.8±0.9 to 15.3±1.2 ng mL-1 ; p = 0.003, d = 1.19) but not NORM (13.7±1.8 to 14.2±1.6 ng mL-1 ; p = .435, d = 0.45). LBP significantly increased from pre- to post-exercise in HYP (10.8±1.2 to 13.9±2.8 μg mL-1 ; p = 0.006, d = 1.12) but not NORM (11.3±1.1 to 11.7±0.9 μg mL-1 ; p>0.99, d = 0.32). I-FABP (d = 0.85), CLDN-3 (d = 0.95), and LBP (d = 0.69) were all significantly higher post-exercise in HYP compared to NORM (p≤0.05). Overall GI discomfort was significantly correlated to ΔI-FABP (r = 0.71), ΔCLDN-3 (r = 0.70), and ΔLBP (r = 0.86). These data indicate that cycling exercise performed in hypobaric hypoxia can cause intestinal injury, which might cause some commonly reported GI symptoms. This article is protected by copyright. All rights reserved.
Background Food processing alters diet digestibility and composition, thereby influencing interactions between host biology, diet, and the gut microbiota. The fecal metabolome offers insight into those relationships by providing a readout of diet-microbiota interactions impacting host health. Objectives To determine the effects of consuming a processed diet on the fecal metabolome, and to explore relationships between changes in the fecal metabolome with fecal microbiota composition and gastrointestinal health markers. Methods Secondary analysis of a randomized-controlled trial wherein healthy adults (94% male; 18–61 yr; BMI 26 ± 3 kg.m–2) consumed their usual diet (CON, n = 27) or a Meal, Ready-to-EatTM military ration diet comprised of processed, shelf-stable, ready-to-eat items for 21d (MRE, n = 27). Fecal metabolite profiles, fecal microbiota composition, biomarkers of intestinal barrier function, and gastrointestinal symptoms were measured before and after the intervention. Between-group differences and associations were assessed using nonparametric t-tests, partial least squares discriminant analysis, correlation and redundancy analysis. Results Fecal levels of multiple dipeptides (Mann-Whitney effect size (ES) = 0.27–0.50) and long-chain saturated fatty acids (ES = 0.35–0.58) increased, whereas, plant-derived compounds (ES = 0.31–0.60) decreased in MRE versus CON (P < 0.05; q < 0.20). Changes in dipeptides correlated positively with changes in fecal levels of Maillard-reaction products (ρ = 0.29–0.70; P < 0.05) and inversely with changes in serum prealbumin (ρ = -0.30 - -0.48; P ≤ 0.03). Multiple bile acids, coffee and caffeine metabolites, and plant-derived compounds were associated with both fecal microbiota composition and gastrointestinal health markers, with changes in fecal microbiota composition explaining 26% of the variability within changes in gastrointestinal health-associated fecal metabolites (P = 0.001). Conclusions Changes in the fecal metabolomes of adults consuming a Meal, Ready-to-EatTM diet implicate interactions between diet composition, diet digestibility and the gut microbiota as contributing to variability within gastrointestinal responses to the diet. Findings underscore the need to consider both food processing and nutrient composition when investigating the impact of diet-gut microbiota interactions on health outcomes. Clinical trials registration: NCT02423551.
Results from high altitude studies in humans and controlled animal experiments suggest that hypoxia exposure induces alterations in gut microbiota composition, which may in turn affect host metabolism. However, well-controlled studies investigating the effects of normobaric hypoxia exposure on gut microbiota composition in humans are lacking. The aim of this study was to explore the impact of mild intermittent hypoxia (MIH) exposure on gut microbiota composition in men with overweight and/or obesity. We performed a randomised, single-blind crossover study, in which participants were exposed to MIH (FiO 2 : 15%, 3×2 h per day) and normoxia (FiO 2 : 21%) for seven consecutive days. Following the MIH and normoxia exposure regimens, faecal samples were collected for determination of faecal microbiota composition using 16S rRNA gene-amplicon sequencing in the morning of day 8. Paired faecal samples were available for five individuals. Furthermore, tissue-specific insulin sensitivity was determined using the gold-standard two-step hyperinsulinemic-euglycemic clamp. MIH did not affect microbial alpha and beta-diversity but reduced the relative abundance of Christensenellaceae and Clostridiaceae bacterial families. MIH significantly increased the abundances of obligate anaerobic bacterial genera including Fusicatenibacter, Butyricicoccus and Holdemania, whilst reducing Christensenellaceae R-7 group and Clostridium sensu stricto 1, although these findings were not statistically significant after correction for multiple testing. Furthermore, MIH-induced alterations in abundances of several genera were associated with changes in metabolic parameters such as adipose and peripheral insulin sensitivity, plasma levels of insulin, fatty acids, triacylglycerol and lactate, and substrate oxidation. In conclusion, we demonstrate for the first time that MIH exposure induces modest effects on faecal microbiota composition in humans, shifting several bacterial families and genera towards higher abundances of anaerobic butyrate-producing bacteria. Moreover, MIH-induced effects on faecal microbial composition were associated with parameters related to glucose and lipid homeostasis, supporting a link between MIH-induced alterations in faecal microbiota composition and host metabolism. The study was registered at the Netherlands Trial Register: NL7120/NTR7325.
Full-text available
The aim of this study was to assess changes in mental health, gut microbiota composition, and stress marker serum cortisol due to COVID-19 lockdown in asymptomatic individuals. Healthy adults participated in anthropometric measurements, blood and stool sample collection pre-lockdown and post-lockdown (n = 38, 63.2% females), lifestyle and psychological questionnaires were included in pre-lockdown measurement and lockdown survey (n = 46, 67.4% females). Subjects reported significantly higher body dissatisfaction (p = 0.007) and anxiety (p = 0.002), and significantly lower positive affect (p = 0.001) during lockdown compared with pre-lockdown. According to perceived stress, 51.6% of females and 20% of males experienced moderate to high stress. This was reflected in serum cortisol levels that significantly increased only in females (p = 0.006) post-lockdown and correlated with perceived stress (p = 0.037) and anxiety (p = 0.031). In addition to psychological measures, changes in gut microbiota composition were observed. Gut microbial alpha diversity significantly decreased (p = 0.033), whereas relative abundance of Proteobacteria significantly increased (p = 0.043) post-lockdown. Depression during lockdown was moderately positively correlated with changes in Bacteroidetes abundance (p = 0.015) and negatively with changes in Firmicutes abundance (p = 0.008). Alistipes abundance post-lockdown was moderately positively correlated with anxiety (p = 0.004) and negative affect (p = 0.005) during lockdown. Despite a small sample size and not being able to perform objective measurements during lockdown, the results confirm the effect of lockdown on mental health and gut microbiota composition that could have a great impact on our health (ClinicalTrials identifier: NCT04347213).
The gut microbiota is involved in host responses to high altitude. However, the dynamics of intestinal microecology and their association with altitude-related illness are poorly understood. Here, we used a rat model of hypobaric hypoxia challenge to mimic plateau exposure and monitored the gut microbiome, short-chain fatty acids (SCFAs), and bile acids (BAs) over 28 d. We identified weight loss, polycythemia, and pathological cardiac hypertrophy in hypoxic rats, accompanied by a large compositional shift in the gut microbiota, which is mainly driven by the bacterial families of Prevotellaceae, Porphyromonadaceae, and Streptococcaceae. The aberrant gut microbiota was characterized by increased abundance of the Parabacteroides, Alistipes, and Lactococcus genera and a larger Bacteroides to Prevotella ratio. Trans-omics analyses showed that the gut microbiome was significantly correlated with the metabolic abnormalities of SCFAs and BAs in feces, suggesting an interaction network remodeling of the microbiome-metabolome after the hypobaric hypoxia challenge. Interestingly, the transplantation of fecal microbiota significantly increased the diversity of the gut microbiota, partially inhibited the increased abundance of the Bacteroides and Alistipes genera, restored the decrease of plasma propionate, and moderately ameliorated cardiac hypertrophy in hypoxic rats. Our results provide an insight into the longitudinal changes in intestinal microecology during the hypobaric hypoxia challenge. Abnormalities in the gut microbiota and microbial metabolites contribute to the development of high-altitude heart disease in rats.
The environmental conditions in high‐altitude areas can induce gastrointestinal disorders and changes in gut microbiota. Gut microbiota is closely related to a variety of gastrointestinal diseases, however, the underlying pathogenic mechanisms are not well‐identified. This study aimed to investigate the regulatory effect of high altitude on intestinal dysfunction via gut microbiota disturbance. Forty C57BL/6J mice were divided into four groups, i.e. one plain control group (CON) and three high‐altitude exposure groups (HAE) (altitude: 4000 m; oxygen content: 12.7%; 1‐, 2‐ and 4‐week exposure). Another set of forty mice was divided into two CON and two HAE subgroups. Antibiotic cocktails were administered to one CON and HAE groups, and autoclaved water to the second CON and HAE groups for 4 weeks, respectively. In the fecal microbiota transplantation experiment, there were four transplantation groups, which received respectively: phosphate‐buffered saline for 2 weeks, feces from CON for 2 weeks, feces from HAE‐4W for 2 weeks, and HAE‐4W for 4 weeks. Hematoxylin‐eosin staining, Periodic Acid‐Schiff staining, terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assay and qRT‐PCR were applied to detect changes in intestinal cellular structure, morphology, apoptosis, and intestinal inflammatory response. Fecal microbiota was analyzed using 16S rDNA Amplicon Sequencing. A high‐altitude environment changed the ecological balance of gut microbiota in mice and caused damage to intestinal structure and mucosal barrier in mice. Interestingly, similar damage, which was inhibited by antibiotic cocktails at high altitude, was observed in mice transplanted with fecal microbiota from HAE. High‐altitude environment contributes to dyshomeostasis of gut microbiota, thereby impairing the intestinal mucosal barrier, eventually inducing and exacerbating intestinal damage.
Full-text available
Stress, a ubiquitous part of daily human life, has varied biological effects which are increasingly recognized as including modulation of commensal microorganisms residing in the gastrointestinal tract, the gut microbiota. In turn, the gut microbiota influences the host stress response and associated sequelae, thereby implicating the gut microbiota as an important mediator of host health. This narrative review aims to summarize evidence concerning the impact of psychological, environmental, and physical stressors on gut microbiota composition and function. The stressors reviewed include psychological stress, circadian disruption, sleep deprivation, environmental extremes (high altitude, heat, and cold), environmental pathogens, toxicants, pollutants, and noise, physical activity, and diet (nutrient composition and food restriction). Stressors were selected for their direct relevance to military personnel, a population that is commonly exposed to these stressors, often at extremes, and in combination. However, the selected stressors are also common, alone or in combination, in some civilian populations. Evidence from preclinical studies collectively indicates that the reviewed stressors alter the composition, function and metabolic activity of the gut microbiota, but that effects vary across stressors, and can include effects that may be beneficial or detrimental to host health. Translation of these findings to humans is largely lacking at present. This gap precludes concluding with certainty that transient or cumulative exposures to psychological, environmental, and physical stressors have any consistent, meaningful impact on the human gut microbiota. However, provocative preclinical evidence highlights a need for translational research aiming to elucidate the impact of stressors on the human gut microbiota, and how the gut microbiota can be manipulated, for example by using nutrition, to mitigate adverse stress responses.
Full-text available
Intramuscular factors that modulate fat-free mass (FFM) loss in lowlanders exposed to energy deficit during high-altitude (HA) sojourns remain unclear. Muscle inflammation may contribute to FFM loss at HA by inducing atrophy and inhibiting myogenesis via the tumor necrosis factor (TNF)-like weak inducer of apoptosis (TWEAK) and its receptor, fibroblast growth factor-inducible protein 14 (Fn14). To explore whether muscle inflammation modulates FFM loss reportedly developing during HA sojourns, muscle inflammation, myogenesis, and proteolysis were assessed in 16 men at sea level (SL) and following 21 days of energy deficit (-1862 ± 525 kcal/days) at high altitude (HA, 4300 m). Total body mass (TBM), FFM, and fat mass (FM) were assessed using DEXA. Gene expression and proteolytic enzymatic activities were assessed in muscle samples collected at rest at SL and HA. Participants lost 7.2 ± 1.8 kg TBM (P < 0.05); 43 ± 30% and 57 ± 30% of the TBM lost was FFM and FM, respectively. Fn14, TWEAK, TNF alpha-receptor (TNFα-R), TNFα, MYOGENIN, and paired box protein-7 (PAX7) were upregulated (P < 0.05) at HA compared to SL. Stepwise linear regression identified that Fn14 explained the highest percentage of variance in FFM loss (r2 = 0.511, P < 0.05). Dichotomization of volunteers into HIGH and LOW Fn14 gene expression indicated HIGH lost less FFM and more FM (28 ± 28% and 72 ± 28%, respectively) as a proportion of TBM loss than LOW (58 ± 26% and 42 ± 26%; P < 0.05) at HA. MYOGENIN gene expression was also greater for HIGH versus LOW (P < 0.05). These data suggest that heightened Fn14 gene expression is not catabolic and may protect FFM during HA sojourns.
Full-text available
This study investigated how high-altitude (HA, 4300 m) acclimatization affected exogenous glucose oxidation during aerobic exercise. Sea-level (SL) residents (n = 14 men) performed 80-min, metabolically matched exercise ( V ˙ O2 ∼ 1.7 L/min) at SL and at HA < 5 h after arrival (acute HA, AHA) and following 22-d of HA acclimatization (chronic HA, CHA). During HA acclimatization, participants sustained a controlled negative energy balance (-40%) to simulate the "real world" conditions that lowlanders typically experience during HA sojourns. During exercise, participants consumed carbohydrate (CHO, n = 8, 65.25 g fructose + 79.75 g glucose, 1.8 g carbohydrate/min) or placebo (PLA, n = 6). Total carbohydrate oxidation was determined by indirect calorimetry and exogenous glucose oxidation by tracer technique with 13C. Participants lost (P ≤ 0.05, mean ± SD) 7.9 ± 1.9 kg body mass during the HA acclimatization and energy deficit period. In CHO, total exogenous glucose oxidized during the final 40 min of exercise was lower (P < 0.01) at AHA (7.4 ± 3.7 g) than SL (15.3 ± 2.2 g) and CHA (12.4 ± 2.3 g), but there were no differences between SL and CHA. Blood glucose and insulin increased (P ≤ 0.05) during the first 20 min of exercise in CHO, but not PLA. In CHO, glucose declined to pre-exercise concentrations as exercise continued at SL, but remained elevated (P ≤ 0.05) throughout exercise at AHA and CHA. Insulin increased during exercise in CHO, but the increase was greater (P ≤ 0.05) at AHA than at SL and CHA, which did not differ. Thus, while acute hypoxia suppressed exogenous glucose oxidation during steady-state aerobic exercise, that hypoxic suppression is alleviated following altitude acclimatization and concomitant negative energy balance.
Full-text available
We explored the metagenomic, metabolomic and trace metal makeup of intestinal microbiota and environment in healthy male participants during the run-in (5 day) and the following three 21-day interventions: normoxic bedrest (NBR), hypoxic bedrest (HBR) and hypoxic ambulation (HAmb) which were carried out within a controlled laboratory environment (circadian rhythm, fluid and dietary intakes, microbial bioburden, oxygen level, exercise). The fraction of inspired O2 (FiO2) and partial pressure of inspired O2 (PiO2) were 0.209 and 133.1 ± 0.3 mmHg for the NBR and 0.141 ± 0.004 and 90.0 ± 0.4 mmHg (~4,000 m simulated altitude) for HBR and HAmb interventions, respectively. Shotgun metagenomes were analyzed at various taxonomic and functional levels, 1H- and 13C -metabolomes were processed using standard quantitative and human expert approaches, whereas metals were assessed using X-ray fluorescence spectrometry. Inactivity and hypoxia resulted in a significant increase in the genus Bacteroides in HBR, in genes coding for proteins involved in iron acquisition and metabolism, cell wall, capsule, virulence, defense and mucin degradation, such as beta-galactosidase (EC3.2.1.23), α-L-fucosidase (EC3.2.1.51), Sialidase (EC3.2.1.18), and α-N-acetylglucosaminidase (EC3.2.1.50). In contrast, the microbial metabolomes, intestinal element and metal profiles, the diversity of bacterial, archaeal and fungal microbial communities were not significantly affected. The observed progressive decrease in defecation frequency and concomitant increase in the electrical conductivity (EC) preceded or took place in absence of significant changes at the taxonomic, functional gene, metabolome and intestinal metal profile levels. The fact that the genus Bacteroides and proteins involved in iron acquisition and metabolism, cell wall, capsule, virulence and mucin degradation were enriched at the end of HBR suggest that both constipation and EC decreased intestinal metal availability leading to modified expression of co-regulated genes in Bacteroides genomes. Bayesian network analysis was used to derive the first hierarchical model of initial inactivity mediated deconditioning steps over time. The PlanHab wash-out period corresponded to a profound life-style change (i.e., reintroduction of exercise) that resulted in stepwise amelioration of the negative physiological symptoms, indicating that exercise apparently prevented the crosstalk between the microbial physiology, mucin degradation and proinflammatory immune activities in the host.
Full-text available
Karl, J. Philip, Renee E. Cole, Claire E. Berryman, Graham Finlayson, Patrick N. Radcliffe, Matthew T. Kominsky, Nancy E. Murphy, John W. Carbone, Jennifer C. Rood, Andrew J. Young, and Stefan M. Pasiakos. Appetite Suppression and Altered Food Preferences Coincide with Changes in Appetite-Mediating Hormones During Energy Deficit at High Altitude, But Are Not Affected by Protein Intake. High Alt Med Biol. 00:000-000, 2018.-Anorexia and unintentional body weight loss are common during high altitude (HA) sojourn, but underlying mechanisms are not fully characterized, and the impact of dietary macronutrient composition on appetite regulation at HA is unknown. This study aimed to determine the effects of a hypocaloric higher protein diet on perceived appetite and food preferences during HA sojourn and to examine longitudinal changes in perceived appetite, appetite mediating hormones, and food preferences during acclimatization and weight loss at HA. Following a 21-day level (SL) period, 17 unacclimatized males ascended to and resided at HA (4300 m) for 22 days. At HA, participants were randomized to consume measured standard-protein (1.0 g protein/kg/d) or higher protein (2.0 g/kg/d) hypocaloric diets (45% carbohydrate, 30% energy restriction) and engaged in prescribed physical activity to induce an estimated 40% energy deficit. Appetite, food preferences, and appetite-mediating hormones were measured at SL and at the beginning and end of HA. Diet composition had no effect on any outcome. Relative to SL, appetite was lower during acute HA (days 0 and 1), but not different after acclimatization and weight loss (HA day 18), and food preferences indicated an increased preference for sweet- and low-protein foods during acute HA, but for high-fat foods after acclimatization and weight loss. Insulin, leptin, and cholecystokinin concentrations were elevated during acute HA, but not after acclimatization and weight loss, whereas acylated ghrelin concentrations were suppressed throughout HA. Findings suggest that appetite suppression and altered food preferences coincide with changes in appetite-mediating hormones during energy deficit at HA. Although dietary protein intake did not impact appetite, the possible incongruence with food preferences at HA warrants consideration when developing nutritional strategies for HA sojourn.
Full-text available
Population stratification is a useful approach for a better understanding of complex biological problems in human health and wellbeing. The proposal that such stratification applies to the human gut microbiome, in the form of distinct community composition types termed enterotypes, has been met with both excitement and controversy. In view of accumulated data and re-analyses since the original work, we revisit the concept of enterotypes, discuss different methods of dividing up the landscape of possible microbiome configurations, and put these concepts into functional, ecological and medical contexts. As enterotypes are of use in describing the gut microbial community landscape and may become relevant in clinical practice, we aim to reconcile differing views and encourage a balanced application of the concept.
Full-text available
High-altitude sojourn above 8000 ft is increasing day by day either for pilgrimage, mountaineering, holidaying or for strategic reasons. In India, soldiers are deployed to these high mountains for their duty or pilgrims visit to the holy places, which are located at very high altitude. A large population also resides permanently in high altitude regions. Every year thousands of pilgrims visit Holy cave of Shri Amarnath ji, which is above 15 000 ft. The poor acclimatization to high altitude may cause alteration in immunity. The low oxygen partial pressure may cause alterations in gut microbiota, which may cause changes in gut immunity. Effect of high altitude on gut-associated mucosal system is new area of research. Many studies have been carried out to understand the physiology and immunology behind the high-altitude-induced gut problems. Few interventions have also been discovered to circumvent the problems caused due to high-altitude conditions. In this review, we have discussed the effects of high-altitude-induced changes in gut immunity particularly peyer's patches, NK cells and inflammatory cytokines, secretary immunoglobulins and gut microbiota. The published articles from PubMed and Google scholar from year 1975 to 2017 on high-altitude hypoxia and gut immunity are cited in this review.
Full-text available
We explored the assembly of intestinal microbiota in healthy male participants during the randomized crossover design of run-in (5 day) and experimental phases (21-day normoxic bed rest (NBR), hypoxic bed rest (HBR) and hypoxic ambulation (HAmb) in a strictly controlled laboratory environment, with balanced fluid and dietary intakes, controlled circadian rhythm, microbial ambiental burden and 24/7 medical surveillance. The fraction of inspired O2 (FiO2) and partial pressure of inspired O2 (PiO2) were 0.209 and 133.1 ± 0.3 mmHg for NBR and 0.141 ± 0.004 and 90.0 ± 0.4 mmHg for both hypoxic variants (HBR and HAmb; ~4000 m simulated altitude), respectively. A number of parameters linked to intestinal environment such as defecation frequency, intestinal electrical conductivity (IEC), sterol and polyphenol content and diversity, indole, aromaticity and spectral characteristics of dissolved organic matter (DOM) were measured (64 variables). The structure and diversity of bacterial microbial community was assessed using 16S rRNA amplicon sequencing. Inactivity negatively affected frequency of defecation and in combination with hypoxia increased IEC (p < 0.05). In contrast, sterol and polyphenol diversity and content, various characteristics of DOM and aromatic compounds, the structure and diversity of bacterial microbial community were not significantly affected over time. A new in-house PlanHab database was established to integrate all measured variables on host physiology, diet, experiment, immune and metabolic markers (n = 231). The observed progressive decrease in defecation frequency and concomitant increase in IEC suggested that the transition from healthy physiological state towards the developed symptoms of low magnitude obesity-related syndromes was dose dependent on the extent of time spent in inactivity and preceded or took place in absence of significant rearrangements in bacterial microbial community. Species B. thetaiotamicron, B. fragilis, B. dorei and other Bacteroides with reported relevance for dysbiotic medical conditions were significantly enriched in HBR, characterized with most severe inflammation symptoms, indicating a shift towards host mucin degradation and proinflammatory immune crosstalk.
Full-text available
Recent studies have identified various bacterial groups that are altered in dogs with chronic inflammatory enteropathies (CE) compared to healthy dogs. The study aim was to use quantitative PCR (qPCR) assays to confirm these findings in a larger number of dogs, and to build a mathematical algorithm to report these microbiota changes as a dysbiosis index (DI). Fecal DNA from 95 healthy dogs and 106 dogs with histologically confirmed CE was analyzed. Samples were grouped into a training set and a validation set. Various mathematical models and combination of qPCR assays were evaluated to find a model with highest discriminatory power. The final qPCR panel consisted of eight bacterial groups: total bacteria, Faecalibacterium, Turicibacter, E. coli, Streptococcus, Blautia, Fusobacterium, and C. hiranonis. The qPCR based dysbiosis index (DI) was built based on the nearest centroid classifier, and reports the degree of dysbiosis in a single numerical value that measures the closeness in the$\ {l_2} - norm$ of the test sample to the mean prototype of each class. A negative DI indicates normobiosis, whereas a positive DI indicates dysbiosis. For a threshold of 0, the DI based on the combined dataset achieved 74% sensitivity and 95% specificity to separate healthy and CE dogs.
Muscle loss at high altitude (HA) is attributable to energy deficit and a potential dysregulation of anabolic signaling. Exercise and protein ingestion can attenuate the effects of energy deficit on muscle at sea level (SL). Whether these effects are observed when energy deficit occurs at HA is unknown. To address this, muscle obtained from lowlanders ( n = 8 males) at SL, acute HA (3 h, 4300 m), and chronic HA (21 d, -1766 kcal/d energy balance) before [baseline (Base)] and after 80 min of aerobic exercise followed by a 2-mile time trial [postexercise (Post)] and 3 h into recovery (Rec) after ingesting whey protein (25 g) were analyzed using standard molecular techniques. At SL, Post, and REC, p-mechanistic target of rapamycin (mTOR)Ser2448, p-p70 ribosomal protein S6 kinase (p70S6K)Ser424/421, and p-ribosomal protein S6 (rpS6)Ser235/236 were similar and higher ( P < 0.05) than Base. At acute HA, Post p-mTORSer2448 and Post and REC p-p70S6KSer424/421 were not different from Base and lower than SL ( P < 0.05). At chronic HA, Post and Rec p-mTORSer2448 and p-p70S6KSer424/421 were not different from Base and lower than SL, and, independent of time, p-rpS6Ser235/236 was lower than SL ( P < 0.05). Post proteasome activity was lower ( P < 0.05) than Base and Rec, independent of phase. Our findings suggest that HA exposure induces muscle anabolic resistance that is exacerbated by energy deficit during acclimatization, with no change in proteolysis.-Margolis, L. M., Carbone, J. W., Berryman, C. E., Carrigan, C. T., Murphy, N. E., Ferrando, A. A., Young, A. J., Pasiakos, S. M. Severe energy deficit at high altitude inhibits skeletal muscle mTORC1-mediated anabolic signaling without increased ubiquitin proteasome activity.