Pomegranate Juice and Extract Consumption Increases the Resistance to UVB-induced Erythema and Changes the Skin Microbiome in Healthy Women: a Randomized Controlled Trial

Abstract

In vitro and animal studies have demonstrated that topical application and oral consumption of pomegranate reduces UVB-induced skin damage. We therefore investigated if oral pomegranate consumption will reduce photodamage from UVB irradiation and alter the composition of the skin microbiota in a randomized controlled, parallel, three-arm, open label study. Seventy-four female participants (30–45 years) with Fitzpatrick skin type II-IV were randomly assigned (1:1:1) to 1000 mg of pomegranate extract (PomX), 8 oz of pomegranate juice (PomJ) or placebo for 12 weeks. Minimal erythema dose (MED) and melanin index were determined using a cutometer (mexameter probe). Skin microbiota was determined using 16S rRNA sequencing. The MED was significantly increased in the PomX and PomJ group compared to placebo. There was no significant difference on phylum, but on family and genus level bacterial composition of skin samples collected at baseline and after 12 week intervention showed significant differences between PomJ, PomX and placebo. Members of the Methylobacteriaceae family contain pigments absorbing UV irradiation and might contribute to UVB skin protection. However, we were not able to establish a direct correlation between increased MED and bacterial abundance. In summary daily oral pomegranate consumption may lead to enhanced protection from UV photodamage.

Full text

1
SCIENTIFIC REPORTS | (2019) 9:14528 | https://doi.org/10.1038/s41598-019-50926-2
www.nature.com/scientificreports
Pomegranate Juice and Extract
Consumption Increases the
Resistance to UVB-induced
Erythema and Changes the Skin
Microbiome in Healthy Women: a
Randomized Controlled Trial
Susanne M. Henning
1, Jieping Yang1, Ru-Po Lee1, Jianjun Huang1, Mark Hsu1, Gail Thames1,
Irene Gilbuena1, Jianfeng Long2, Yunhui Xu1, Esther HaeIn Park1, Chi-Hong Tseng3,
Jenny Kim1,4, David Heber1 & Zhaoping Li1
In vitro and animal studies have demonstrated that topical application and oral consumption of
pomegranate reduces UVB-induced skin damage. We therefore investigated if oral pomegranate
consumption will reduce photodamage from UVB irradiation and alter the composition of the skin
microbiota in a randomized controlled, parallel, three-arm, open label study. Seventy-four female
participants (30–45 years) with Fitzpatrick skin type II-IV were randomly assigned (1:1:1) to 1000 mg
of pomegranate extract (PomX), 8 oz of pomegranate juice (PomJ) or placebo for 12 weeks. Minimal
erythema dose (MED) and melanin index were determined using a cutometer (mexameter probe).
Skin microbiota was determined using 16S rRNA sequencing. The MED was signicantly increased
in the PomX and PomJ group compared to placebo. There was no signicant dierence on phylum,
but on family and genus level bacterial composition of skin samples collected at baseline and after 12
week intervention showed signicant dierences between PomJ, PomX and placebo. Members of the
Methylobacteriaceae family contain pigments absorbing UV irradiation and might contribute to UVB
skin protection. However, we were not able to establish a direct correlation between increased MED
and bacterial abundance. In summary daily oral pomegranate consumption may lead to enhanced
protection from UV photodamage.
Exposure of human skin to UV radiation is a major factor for skin pathologies including erythema, inamma-
tion, degenerative age-related changes and cancer1. UV radiation is mostly composed of UVA (315–400 nm) and
UVB (290–320 nm). Overexposure of the skin to UVA and to a lesser extent to UVB leads to oxidative stress that
increases the generation of reactive oxygen species (ROS) causing lipid peroxidation of cell membranes, DNA
and protein damage to tissue, inammation and keratinocyte apoptosis2–4. ROS also trigger the expression of
matrix metalloproteinases (MMP), that degrade extracellular matrix such as collagen maintaining cell and skin
integrity3.
Pomegranate fruits have been used for centuries in ancient cultures for its medicinal purposes5. e health
benet of pomegranate has been attributed to the content of hydrolysable tannins (ellagitannins) including
punicalagins and ellagic acid (EA) as well as anthocyanins and other polyphenols found in pomegranate extract
1Center for Human Nutrition, David Geen School of Medicine, Department of Medicine, Los Angeles, CA, 90095,
USA. 2Department of Clinical Nutrition, 2nd Xiangya Hospital of Central South University, Changsha, Hunan, 410011,
China. 3Department of Statistics Core, David Geen School of Medicine, University of California Los Angeles, Los
Angeles, CA, 90095, USA. 4Division of Dermatology, David Geen School of Medicine, University of California Los
Angeles, Los Angeles, CA, 90095, USA. Correspondence and requests for materials should be addressed to S.M.H.
(email: shenning@mednet.ucla.edu)
Received: 19 June 2018
Accepted: 20 September 2019
Published: xx xx xxxx
OPEN
Content courtesy of Springer Nature, terms of use apply. Rights reserved

2
SCIENTIFIC REPORTS | (2019) 9:14528 | https://doi.org/10.1038/s41598-019-50926-2
www.nature.com/scientificreports
www.nature.com/scientificreports/
and juice6. Although POM ellagitannins are highly bioactive in vitro, they are not absorbed intact in the small
intestine and undergo partial hydrolysis to form EA7. Ellagitannins and EA remaining in the large intestine are
further metabolized to urolithins A –D by the microbiota in the large intestine and are absorbed into the blood
stream8–10. erefore, pomegranate properties could be mediated by the metabolites produced in the intestine in
addition to the original phenolic compounds present in the food matrix. Due to the sugar content of PomJ, which
might be of health concern, we included both PomJ and PomX intervention in this investigation.
Several in vitro and animal studies provide evidence that either topical application or oral consumption of
pomegranate juice (PomJ) or extract (PomX) or EA reduce damage from UVB irradiation11–17. One human
study demonstrated that oral consumption of an EA rich pomegranate extract in healthy women was asso-
ciated with a protective eect on slight sunburn caused by UV irradiation even at a low dose resulting in a
decrease in pigmentation17. In vitro antioxidant and anti-inammatory activity of ellagitannins and EA have
been widely demonstrated18–20, while oral ingestion of pomegranate in humans has been less investigated. Some
human studies provide mechanistic evidence that pomegranate ingestion leads to an increase in antioxidant and
anti-inammatory activity9,21.
Inammation caused by UV radiation activates various matrix-degrading matrix metalloproteases (MMPs),
which leads to collagen degradation and cellular apoptosis. MMP-1, especially, was the main endogenous factor
that degraded dermal collagen in the process of human skin senility22.
Previously published ndings from our laboratory demonstrated that pomegranate extract inhibited in vitro
growth of bacteria involved in the pathogenesis of acne including Propionibacterium acnes, Propionibacterium
granulosum, Staphylococcus aureus and Staphylococcus epidermidis23. Although the concentration of circulating
Pom ellagitannin metabolites is very low, there is a possibility that Pom ellagitannin metabolites in the skin might
alter the composition of the skin microbiota8. In addition, pigment forming bacteria can contribute to skin UV
protection24. erefore, it was our hypothesis that pomegranate consumption might alter the skin microbiota
contributing to protection from UVB.
e objective of the current study was to determine whether pomegranate extract (PomX) or pomegranate
juice (PomJ) can decrease UVB-induced skin photoaging, alter inammatory markers and the skin microbiota.
Results
Characteristics of participants. Seventy-four participants completed the study. ere was no statistically
signicant dierence at baseline for average age, height, weight, BMI, race, ethnicity and skin type (Table1).
Although there were more Asian women in the PomJ group compared to PomX and placebo, there was no dier-
ence in skin type distribution among the three groups (Table1).
Urolithin formation. In the intestine pomegranate ellagitannins are broken down to ellagic acid (EA), which
can be absorbed and converted to methylellagic acid glucuronide (DMEAG). Urinary DMEAG can be used to
determine compliance. EA remaining in the intestine is further metabolized by bacteria to urolithin A (UA). Aer
absorption UA circulates in form of UA glucuronide (UAG). In the PomJ group, 16 participants (67%) showed
UAG in urine and 8 (33%) were UA non-producers, while in the PomX group 19 participants (83%) formed
UA and 4 (17%) were UA non-producers. In the PomX group two participants had neither UAG nor DMEAG
Pom Juice (n = 24) Pom extract (n = 25) Placebo (n = 25) P value
Age (years) 35.1 ± 4.3 35.9 ± 4.1 37.9 ± 4.2 0.063
Height (inches) 63.5 ± 2.8 63.9 ± 2.6 63.8 ± 2.5 0.843
Weight (lbs) 153 ± 30.8 158.7 ± 34.5 170.2 ± 40.7 0.253
BMI 26.6 ± 5.0 27.1 ± 5.1 29.9 ± 6.7 0.092
Female 24 (100) 25 (100) 25 (100)
Race:White 14 (58) 20 (80) 14 (56) 0.284
Black 01 (4) 1 (4)
Asian 9 (38) 3 (12) 3 (12)
Bi-racial 1 (4) 1 (4) 2 (8)
Ethnicity:Hispanic 9 (37) 16 (64) 10 (40) 0.119
Non-Hispanic 15 (63) 9 (36) 15 (60)
Skin Type 0.552
II 5 (20.8) 2 (8.0) 3 (12.0)
III 6 (25.0) 9 (36.0) 11 (44.0)
IV 13 (54.2) 14 (56.0) 11 (44.0)
Melanin index (RU) 264.5 ± 212.6 242.7 ± 91.1 202.2 ± 75.4 0.59
UAG producer 16 (67) 19 (83) n/a
UAG non-producer 8 (33) 4 (17)# n/a
Table 1. Baseline demographics of study participants (n = 24–25). Data are mean ± SD. Numbers in
parenthesis are percent. #two participants did not produce urolithin A glucuronide (UAG) or dimethylellagic
acid glucuronide (DMEAG), relative unit RU.
Content courtesy of Springer Nature, terms of use apply. Rights reserved

3
SCIENTIFIC REPORTS | (2019) 9:14528 | https://doi.org/10.1038/s41598-019-50926-2
www.nature.com/scientificreports
www.nature.com/scientificreports/
in urine, which may indicate that on the day prior to the blood collection the participants did not consume the
pomegranate product. ey were excluded from the statistical analyses of the skin measurements. ere was no
signicant dierence in urine UAG and DMEAG between individuals consuming PomJ or PomX (Table2).
Eects of POM extract consumption on minimal erythema dose and melanin index. To e val -
uate the minimal erythema dose (MED) the inner arm skin was exposed to increasing dose and exposure time
to determine the minimal UVB dose that induces erythema. In women consuming PomJ (p = 0.038) or PomX
(p = 0.011) for 12 weeks the MED was increased signicantly when analyzing the dierence of baseline to nal
erythema dose compared to placebo group. In the PomX group UA producers experienced a signicant increase
in MED, while in UA non-producers MED was not increased signicantly (Table2). In the PomJ group, there
was no dierence between UA producers and non-producers in regards to MED. e time of exposure showed
a trend to increase in both groups compared to the placebo group (p = 0.083 for PomJ and p = 0.089 for PomX).
Upon exposure to UVB irradiation epidermal melanocytes produce melanin that is transferred to neighboring
keratinocytes protecting the cells from UV radiation damage25. However, hyperpigmentation will lead to prema-
ture photoaging and senescence in melanocytes25. Melanin formation also can be used to evaluate the eect of
UVB exposure. In the present study in women consuming PomJ and PomX melanin concentration was decreased
but did not reach signicance (Table3).
Eects of POM extract consumption on the skin microbiota. Pomegranate ellagitannins and EA
have anti-bacterial activity in vitro and topical application of a Pom extract ointment has been shown to decrease
the growth of Propionibacterium acnes and reduce edema in Wistar rat ears23,26,27. In the present study we deter-
mined whether oral consumption of PomJ or PomX altered the microbiota on the skin surface. Microbiota anal-
ysis showed the following composition of the skin microbiota on the phylum level: 36–38% Firmicutes, 25–31%
Proteobacteria, 18–30% Actinobateria, 9–17% Bacteroidetes, 0.2–1.9% Fusobacteria, 0.2–1.1% Cyanobacteria,
0.003–0.4% Tenericutes and 0.2–1% others (Fig.1A). Intergroup analysis showed no signicant dierence in
change of phylum composition of skin samples collected at baseline and aer 12 week intervention from women
consuming PomJ compared to placebo and PomX and placebo.
Intergroup analysis comparing the change in relative abundance of skin bacteria from baseline to week 12
between each intervention group (placebo, PomX and PomJ) showed multiple signicant changes, which were
not always in the same direction for PomX and PomJ intervention. We focused our data analysis on bacteria with
relative abundance that was signicantly dierent between the pomegranate and placebo groups.
Pom Juice UA
producer (n = 16) Pom Juice nonUA
producer (n = 8) Pom extract UA
producer (n = 19) Pom extract nonUA
producer (n = 4) Placebo
(n = 25)
Urine UAG (ng/mL) 3147 ± 1861 0 4068 ± 2049 0 n/a
Urine DMEAG (ng/mL) 163 ± 147 199 ± 118 210 ± 155 200 ± 101 n/a
MED BL (mJ/cm2) 385.7 ± 97.7 379.3 ± 100.1 409.2 ± 80.6 347.5 ± 111.4 384.2 ± 105.6
MED F (mJ/cm2) 417.5 ± 126.6 420.0 ± 87.6 440.0 ± 88.1*373.8 ± 41.1 367.2 ± 90.2
Table 2. Minimal erythema dose and urine urolithin A glucuronide and dimethylellagic acid glucuronide
in UA producers and non producers consuming PomJ, PomX or placebo. Data are mean ± SD, n = 24–25.
ANCOVA model was used to compare outcomes adjusted for baseline value between placebo and Pom extract
or placebo and Pom Juice groups; *p < 0.05. Urolithin A glucuronide (UAG); dimethylellagic acid glucuronide
(DMEAG); minimal erythema dose (MED); baseline BL; nal (F).
Pom Juice (n = 24) Pom extract (n = 25) Placebo (n = 25)
Compliance F (%) 96.3 ± 5.1 97.5 ± 3.9 98.9 ± 2.1
MED BL (mJ/cm2) 383.6 ± 95.8 396.8 ± 83.6 384.2 ± 105.6
MED F (mJ/cm2) 418.3 ± 113.1*429.6 ± 81.8*367.2 ± 90.2
Time BL (sec) 175.5 ± 42.9 193.7 ± 46.7 185.6 ± 52.5
Time F (sec) 199.1 ± 58.6 203.6 ± 49.0 177.2 ± 43.6
Melanin index BL (RU) 264.5 ± 212.6 242.7 ± 91.1 202.2 ± 75.4
Melanin index F (RU) 195.0 ± 63.9 219.6 ± 61.0 198.9 ± 67.4
Sebum BL (µg/cm2) 17.0 ± 19.7 14.8 ± 17.5 22.1 ± 36.8
Sebum F (µg/cm2) 15.3 ± 24.4 15.2 ± 18.2 26.9 ± 37.7
Hydration BL (RU) 41.8 ± 14.4 36.4 ± 12.6 41.9 ± 14.5
Hydration F (RU) 38.7 ± 9.8 41.4 ± 11.4 40.9 ± 10.8
Table 3. UVB-induced minimal erythema dose, time of exposure and skin characteristics determined before
and aer PomX, PomJ and placebo intervention. Data are mean ± SD, n = 24–25. BL = baseline, F = nal,
RU = relative unit. ANCOVA model was used to compare outcomes adjusted for baseline value. *p < 0.05.
Minimal erythema dose (MED); baseline BL; nal (F).
Content courtesy of Springer Nature, terms of use apply. Rights reserved

4
SCIENTIFIC REPORTS | (2019) 9:14528 | https://doi.org/10.1038/s41598-019-50926-2
www.nature.com/scientificreports
www.nature.com/scientificreports/
On the family level bacteria from the family Aerococcaceae, Methylobacteriaceae and Campylobacteraceae
were altered signicantly with PomX consumption compared to placebo (Fig.2). Intergroup analysis comparing
the change of bacterial abundance from week 12 to baseline showed that Campylobacteraceae (Proteobacteria)
was increased and Methylobacteriaceae not changed in the PomX group, while a decrease was observed in the
placebo group (Fig.2). e relative abundance of Aerococcaceae was increased in the placebo group, but there
was no change in the PomX group (Fig.2). e changes for PomJ compared to placebo group were not sig-
nicant. On the genus level the most frequently occurring bacteria included Propionibacterium (Actinobacte
ria) > Staphylococcus (Firmicutes) > Prevotella (Bacteroidetes) > Streptococcus (Firmicutes) > Lactobacillus
(Firmicutes) > Corynebacterium (Actinobacter) > Veillonella (Firmicutes) > Haemophilus (Proteobacteria) > A
cinetobacter (Proteobacteria) (Fig.1B). On the genus level comparing the PomX to placebo group, ve genera
were changed in opposite direction compared to the placebo group: Coprococcus (Firmicutes) was decreased,
Alicycliphilus, Conchiformibius and Campylobacter (Proteobacteria) were increased and Geodermatophilaceae_
unclassied was not changed comparing week 12 to baseline (Fig.3). Comparing PomJ group to placebo we
determined that the unclassied genus from the family Rhizobiaceae was changed signicantly (Fig.3).
Pomegranate consumption had no eect on alpha diversity and no clusters were observed in the principal
coordinate analysis of weighted and unweighted beta-diversity (Figs4 and 5).
Discussion
e main ndings observed in the present study were a signicant increase in minimal UVB dose to induce ery-
thema in women consuming 1000 mg of PomX or 8 oz of PomJ daily for 12 weeks. e minimal erythema dose is
determined by the lowest UVB dose and time of exposure to induce skin erythema. In the present study we also
observed a non-signicant trend to increase in time of UVB exposure (Pom X p = 0.08 and PomJ p = 0.088) and
a non-signicant decrease in melanin formation (Pom J p = 0.16). Together the results demonstrate that pome-
granate consumption may lead to increased protection to UVB-induced damage to skin.
Possibly the MED might have been increased due to systemic pomegranate metabolites such as dimethylel-
lagic acid glucuronide or urolithin A glucuronide circulating in the blood stream. Since we found that in the Pom
X group urolithin A producers showed a signicant increase in MED compared to non-producers, we suggest that
circulating urolithin A glucuronide is involved in the UVB protection.
In previous in vitro and animal studies the topical application of pomegranate and EA improved the resist-
ance of skin to UVB exposure12. For example, the study by Bae JY et al. demonstrated that topical application of
EA reduced collagen breakdown by inhibiting matrix metalloproteinase (MMP) activity and inammation in
UVB-irradiated human skin cells and hairless mice12. In addition, animal studies also showed that oral consump-
tion of pomegranate and EA prevented UVB-induced skin damage4,28,29. For example oral consumption of PomJ
concentrated powder in hairless mice resulted in reduction of UVB-induced skin wrinkles through increased skin
water content, collagen type I and hyaluronan content4. Other processes reported in the literature, contributing to
pomegranate’s UVB photoprotection of the skin include oxidation, inammation, melanin formation, apoptosis
Figure 1. Relative abundance of skin microbiota before and aer pomegranate and placebo intervention.
Stacked column bar graphs depict the average relative abundance and distribution of the most abundant
resolved taxa at the phylum (A) and genus (B) level before (BL) and aer (12 week, F) PomX, PomJ and placebo
control (Cntr) intervention.
Content courtesy of Springer Nature, terms of use apply. Rights reserved

5
SCIENTIFIC REPORTS | (2019) 9:14528 | https://doi.org/10.1038/s41598-019-50926-2
www.nature.com/scientificreports
www.nature.com/scientificreports/
Figure 2. Relative abundance of skin bacteria at the family level that were signicantly dierent comparing
PomX to placebo control (Cntr) and PomJ to placebo control groups (change from BL to 12 weeks [F]). Data
are mean ± SD, n = 24–25. Non-parametric Kruskal–Wallis with Mann-Whitney test was used. Bonferroni
correction was used to correct the probability for multiple comparisons.
Figure 3. Relative abundance of skin bacteria at the genus level that were signicantly dierent comparing
PomX to placebo control (Cntr) and PomJ to placebo control groups (change from BL to 12 weeks [F]). Data
are mean ± SD, n = 24–25. Non-parametric Kruskal–Wallis with Mann-Whitney test was used. Bonferroni
correction was used to correct the probability for multiple comparisons.
Content courtesy of Springer Nature, terms of use apply. Rights reserved

6
SCIENTIFIC REPORTS | (2019) 9:14528 | https://doi.org/10.1038/s41598-019-50926-2
www.nature.com/scientificreports
www.nature.com/scientificreports/
of keratinocytes, activity of matrix metalloproteinases (MMPs), collagen and elastin formation and were identi-
ed in in vitro studies in human skin broblasts, keratinocytes or reconstituted skin13,14,30.
Previously, one human study has been performed testing the eect of oral consumption of a high EA pome-
granate extract for 4 weeks on pigmentation in the skin caused by UV irradiation and showed a trend, but no
signicant change in erythema17. Possibly, in this study by Kasai et al. 4 weeks of pomegranate consumption was
not long enough to see a signicant change. Another human study demonstrated that oral consumption of a phy-
tonutrient blend containing omega-3 fatty acids, resveratrol, quercetin, and other polyphenols led to protection
against UVR-induced skin damage31.
Skin exposed to UVB irradiation develops symptoms of a mild sunburn associated with inflammatory
response and characterized clinically by redness, which is mediated by increased dermal vascular permeability,
vasodilation, edema and inammatory cell inltration32. In the current study we did not observe a change in skin
hydration at the end of pomegranate consumption for 12 weeks.
A likely target of Pom metabolites is melanin formation28,30. We observed a trend to decrease of melanin
formation comparing melanin concentration before and aer pomegranate consumption. It has been previously
demonstrated that EA and pomegranate concentrate inhibited tyrosinase activity, the enzyme necessary for mel-
anin formation28. Administration of pomegranate extract inhibited pigmentation, in a dose-dependent manner,
Figure 4. Diversity analyses of skin microbiota before (BL) and aer (F) PomX, PomJ and placebo control
(Cntr) intervention. Alpha diversity using Chao1 index (A) and Whole tree index (B) was evaluated using
QIIME soware package. Data are means ± SD (n = 24–25).
Figure 5. Beta-diversity analyses of skin microbiota before (BL) and aer (F) PomX, PomJ and placebo control
(Cntr) intervention. Unweighted (A) and weighted (B) UniFrac PCoA plots were created using QIIME soware
package.
Content courtesy of Springer Nature, terms of use apply. Rights reserved

7
SCIENTIFIC REPORTS | (2019) 9:14528 | https://doi.org/10.1038/s41598-019-50926-2
www.nature.com/scientificreports
www.nature.com/scientificreports/
on the skin of brownish guinea pigs receiving UV irradiation, where the number of melanocytes in the epidermis
was decreased in a dose-dependent manner28. ese results suggest that EA orally administered is absorbed into
the body and the EA and/or its metabolites inhibit proliferation of melanocytes in the skin, resulting in inhibited
synthesis of melanin by tyrosinase in melanocytes.
Other potential mechanisms of protection from photoaging are through anti-apoptotic eects, inhibition of
MMP activity, inhibition and extracellular matrix (ECM) (COL1 and hyaluronan) synthesis- related moistur-
izing, anti-inammatory and antioxidative eects, which have been observed in healthy female SKH-1 hairless
mice receiving oral gavage of 100–400 mg/kg body weight of a dried PomJ concentrated powder4. In the present
study, we did not nd any changes in gene expression of markers of inammation (Smad3, Tgfβ) and photoaging
(MMP1, β-integrin, stratin and IGF1R) (data not included). Samples were collected by tape stripping of skin
locations not exposed to UVB since UVB exposed skin was irritated and tape stripping could cause considerable
skin irritations. However, since no photo challenge was induced to the skin prior to tape stripping we did not
observe any protection with pomegranate consumption.
Human skin microbiota composition depends on the location of the skin on the body33. Skin swabbing sam-
ples for microbiota analysis were collected at the inner elbow (antecubital fossa), which is considered a moist but
not oily habitat. is region is usually enriched for Corynebacteria species and Staphylococci species34. In healthy
individuals, the most common skin bacteria are categorized into four dierent phyla: Actinobacteria (most dom-
inated by Propionibacterium spp., and Corynebacterium spp.), Firmicutes (major genus is Staphyloccocus spp.),
Proteobacteria and Bacteroidetes33,35. Previous studies have demonstrated that consumption of pomegranate will
aect the intestinal microbiota36 and in vitro studies demonstrated the antibacterial eect of pomegranate on
bacteria commonly found on skin, such as Propionibacterium and Staphylococcus23. However, the present study
is rst to demonstrate that oral consumption of pomegranate altered the skin microbiota. Skin swipes for the
microbiota assessment were performed in skin areas of the inner arm, which were not exposed to photoaging. No
changes on the phylum level were observed. On the family and genus level several bacteria with minor abundance
were changed when comparing PomX and PomJ to placebo. Bacteria in the family Methylobacteriaceae (phy-
lum Proteobacteria) have been found to form UVA-absorbing compounds and are frequently found on plant’s
phyllosphere (above ground), exposed to harmful UV irradiation24. e proportion of Methylobacteriaceae was
decreased in the placebo and PomJ groups and not changed in the PomX group comparing 12 weeks to base-
line. e abundance of Methylobacteriaceae was very low and little is known about pigments in other bacteria.
Bacterial pigments, however, can potentially contribute to UV skin protection. Other bacteria with altered abun-
dance such as Campylobacter (Proteobacteria) and Coprococcus (Firmicutes) are not commonly found on skin.
Our intergroup evaluation of bacteria abundance showed dierent eects of PomX and PomJ consumption on the
skin microbiota. However, both extract and juice intervention resulted in an increase in UVB protection. Possibly
the skin microbiota does not contribute to the UVB protection or since the microbiota functions in a network of
many bacteria possibly changes in individual bacteria may not aect the microbiota function.
The evaluation of the skin microbiota was performed in healthy skin at a site without UVB exposure.
erefore, we do not know if the observed changes in the skin microbiota contributed to the increase in UVB
protection.
One shortcoming of this study was that it is not possible to sample skin sites shortly aer UV exposure. In future
studies we will collect skin samples near the site of UV exposure to evaluate the mechanism of skin UVB protection
from UVB-induced photodamage and to determine if the skin microbiota contributes to UVB protection.
In summary, PomX and PomJ consumption resulted in an increase in skin protection to UVB exposure
as shown by an increase in minimal erythema dose and a trend to decrease melanin formation, indicating an
enhancement of UVB protection. Our mechanistic studies did not provide insights into specic targets of the
pomegranate induced skin protection mostly because skin and microbiota samples were collected at non-UVB
exposed sites. In future studies, we will collect samples near exposed sites to test if changes in gene expression and
composition of the microbiota contribute to the UVB protection by pomegranate consumption.
Methods
is study was a randomized controlled, parallel, three arm open label study completed at the Center for Human
Nutrition, University of California Los Angeles, California, USA. e study was carried out in accordance with
the guidelines of the Human Subjects Protection Committee of the University of California, Los Angeles. e
clinical protocol was approved by the internal review board (IRB) of the University of California, Los Angeles.
All subjects gave written informed consent before the study began. e study was registered in ClinicalTrials.gov
under the following identier: NCT02258776 on 10/07/2014.
Study participants. Seventy-seven healthy women were enrolled. Seventy-four women completed the
12-week pomegranate intervention study. ree participants dropped out related to pregnancy (one participant
in the PomX group and one in the PomJ group) and related to moving out of state (one particiant in the the PomJ
group). eir data was not included in the statistical evaluation. No adverse eects were reported. Inclusion criteria
were 30–40 years of age, female, be in good health. Exclusion criteria were: no use of topical antibiotic or topical
steroid on the face, scalp, neck, arms, forearms or hands in the previous 7 days or with any skin condition in the
target area, no skin irritations, dry skin or rash and no intake of antibiotics. roughout the study participants were
instructed not to consume pomegranate products, walnuts, or polyphenol-rich fruits (strawberry, raspberry, etc.).
Study design. Participants were randomized to consuming either pomegranate juice (PomJ 8 oz)(n = 24),
pomegranate extract capsules (PomX 1000 mg) (n = 25), or placebo (n = 25) capsules for 12 weeks. e random
permuted block design was implemented to carry out the randomization using our standard random number
program. Block size was 4 or 6 carried out in a random way. Subjects were instructed to take a daily dose of
Content courtesy of Springer Nature, terms of use apply. Rights reserved

8
SCIENTIFIC REPORTS | (2019) 9:14528 | https://doi.org/10.1038/s41598-019-50926-2
www.nature.com/scientificreports
www.nature.com/scientificreports/
1000 mg of the pomegranate extract (POMx®, POM Wonderful, Inc., Los Angeles), which delivers pomegranate
polyphenols in an amount equivalent to about 8 oz of pomegranate juice. PomX was developed to be used as a
dietary supplement and has Generally Recognized as Safe status37. Pomegranate placebo capsules contained inac-
tive excipients (dextran). PomJ contained 100 mg of punicalagin A/B and 23 mg ellagic acid in 240 ml of juice, and
PomX contained 100 mg punicalagin A/B and 44 mg ellagic acid in 1000 mg extract, as determined by HPLC38.
Aer subjects signed the informed consent form at baseline, weight and height were measured and skin type
was evaluated. At baseline prior to taking pomegranate and at the end of 12 weeks of pomegranate consumption,
evaluation of UVB-induced erythema, microbiota, gene expression of epithelial cells and blood collection was
performed as described below.
Study outcomes. e primary outcome of this intervention study was the quantication of the minimal
erythema dose (MED) of skin response to UVB exposure and the measure of melanin index in skin. For the
secondary outcome changes in the skin microbiota were determined. Both outcomes were determined before
(baseline) and aer (12 weeks) of PomJ, PomX or placebo consumption.
Minimal erythema dose, melanin index, hydration and sebum evaluation. We determined
the lowest dose of UVB radiation capable of inducing erythema (minimal erythema dose [MED]). e MED
was determined for each subject before (week 0) and aer (week 12) the intervention. Prior to testing, the skin
type was evaluated based on the Fitzpatrick Skin Type scale39. Participants with Fitzpatrick skin type 2–4 were
included in the study. Background erythema (T0) was measured in all test areas before treatment using a mex-
ameter probe attached to Cutometer dual MPA 580 (Courage&Khazaka electronic GmbH, Koeln, Germany).
Skin UVB dose and treatment time were determined based on overall skin type classication. Using the dosing
guideline for NB-UVB and the National Biological UVB mJ chart, we determined the sequential exposure times
for each skin patch. A sleeve with 6 cut out patches was placed on the subjects arm. Using the Dermalight 90
handheld device (National Biological, Beachwood, OH) the test area on the inner arm of subjects was irradiated
with a dened dose of narrow band ultraviolet B (NB-UVB) light delivered by the UV radiation between 270 and
400 nm, peaking at 310 nm was delivered from a uorescent UV-B lamp (Philips TL20 W/12). Depending on skin
type, a dose range of 220–550 mJ/cm2 for a time of 100–290 seconds was used. is dose range is typically used
for treatment of skin conditions like psoriasis and vitiligo. To evaluate MED the subject returned 24 hours later
to determine which skin patch showed minimal erythema (pink color). Photographs were taken before irradia-
tion and aer 24 hours. e lowest dose and time of the occurrence of pink were determined and used as MED.
Melanin index, hydration and sebum on skin surface were evaluated at baseline (prior to Pom intake) and aer
12 week intervention using the mexameter MA18, corneometer CM825 and sebumeter SM815 probes attached
to the Cutometer. Sebumeter SM 815 uses the dierence of light intensity through a plastic strip to indicate the
amount of absorbed sebum. e sebum level is expressed in μg/cm2 40. Corneometer CM 825 uses the high dielec-
tric constant of water for analyzing the water-related changes in the electrical capacitance of the skin. It displays
hydration measurements in system-specic arbitrary units40. A melanin index is calculated by Mexameter MX 18
from the strength of the absorbed and the reected light at, respectively, 660 and 880 nm. An erythema index is
processed similarly at, respectively, 568 and 660 nm40.
Skin surface microbiota collection. Skin sampling using the wet swab method was performed at base-
line and 12 weeks. Samples were collected as described by the Human Microbiome Project41. A sterile 4 cm
square template was placed on the inner arm to mark the sampling area. e collection swab (CatchAll®Sample
Collection Swab (Epicenter, Illumina, Madison, WI) was moistened with buer (50 mM Tris buer [pH 7.6],
1 mM EDTA [pH 8.0], and 0.5% Tween-20) and the area within the template was swabbed for 30 s rubbing the
swab back and forth about 50 times applying rm pressure. e swabs were placed into bead solution for DNA
extraction using DNeasy Powerlyzer microbial kit (Qiagen, Valencia, CA) and vortexed for 30 sec. e quality of
the extracted DNA was conrmed using the Nanodrop 1000 (ermo Fisher Scientic, Wilmington, DE).
Skin microbiological analyses. MiSeq sequencing. Microbial sequencing of the V1 to V3 region of 16 S
bacterial rDNA was performed using primer pair 27 F (AGA GTT TGA TCC TGG CTC AG) and 534 R (ATT
ACC GCG GCT GCT GG)42. 30 cycle PCR using the HotStarTaq Plus Master Mix Kit (Qiagen, USA) was per-
formed under the following conditions: 94 °C for 3 min, followed by 28 cycles of 94 °C for 30 s, 53 °C for 40 s and
72 °C for 1 min, aer which a nal elongation step at 72 °C for 5 min was performed. Aer amplication, PCR
products were checked in 2% agarose gel to determine the success of amplication and the relative intensity of
bands. Multiple samples are pooled together (e.g., 100 samples) in equal proportions based on their molecular
weight and DNA concentrations. Pooled samples were puried using calibrated Ampure XP beads. en the
pooled and puried PCR product was used to prepare DNA library by following Illumina TruSeq DNA library
preparation protocol. Sequencing was performed at MR DNA (www.mrdnalab.com, Shallowater, TX, USA)
on a MiSeq (Illumina, San Diego, CA) following the manufacturer’s guidelines. Sequence data were processed
using a proprietary analysis pipeline (MR DNA, Shallowater, TX, USA). In summary, sequences were depleted
of barcodes then sequences <150 bp removed, sequences with ambiguous base calls removed. Sequences were
denoised, OTUs generated and chimeras removed. Operational taxonomic units (OTUs) were dened by cluster-
ing at 3% divergence (97% similarity). Final OTUs were taxonomically classied using BLASTn against a curated
GreenGenes database43. Within community diversity (α-diversity) was calculated using Quantitative Insights
Into Microbial Ecology (QIIME) soware package44. Analysis of α-diversity (Shannon index) was performed by a
one-way ANOVA. β-diversity was measured by calculating the weighted UniFrac distances45 using QIIME default
scripts, and weighted UniFrac PCoA biplot was visualized using EMPeror46.
Content courtesy of Springer Nature, terms of use apply. Rights reserved

9
SCIENTIFIC REPORTS | (2019) 9:14528 | https://doi.org/10.1038/s41598-019-50926-2
www.nature.com/scientificreports
www.nature.com/scientificreports/
Identication of pomegranate metabolites by high performance liquid chromatography and
mass spectrometry. All solvents were HPLC grade from Fisher Scientic. Ellagic, formic and phosphoric
acid were purchased from Sigma-Aldrich (St Louis, MO). Pure punicalagin A/B was purchased from ChromaDex
(Irvine, CA) and urolithin A was purchased from Jinan Feiteng Technology (Jinan Shandon, China). e com-
position of the pomegranate extract was analyzed by HPLC and diode array detection. To determine the concen-
tration of urolithin A glucuronide and dimethylellagic acid glucuronide in urine, samples (1 mL) were diluted
with 1 ml of 2% formic acid MeOH, vortexed for 30 s and centrifuged at 20,000xg for 10 min at 4 °C. e super-
natant was analyzed by LC-MS/MS8. e concentration was estimated based on urolithin A standard. e con-
version of urolithin A glucuronide to urolithin was estimated by using β-glucuronidase to catalyze hydrolysis of
β-D-glucuronic acid residues from urolithin A glucuronide in human urine samples.
Statistical analysis. To obtain an estimate of the power of this study to detect a treatment eect for the pri-
mary outcome we use data from Kasai et al.17. On the basis of this data we estimated power for this study assum-
ing a treatment eect of similar magnitude and we use a two sample t-test. Based on these assumptions, a nal
sample size of 20 per group will have 85% power to detect a dierence in mean over time among the groups with a
0.050 two sided signicance level. We assume that there will be a 20% drop out rate and the goal was to randomize
24 subjects per group. e study was stopped when at least 24 subjects for each group completed the intervention.
Summary statistics (mean, standard deviation and frequency distribution) were generated for baseline demo-
graphic and clinical information for each study group to characterize the study population. ANOVA (analysis
of variance) and Chi-square test were used to evaluate the dierence between treatment groups for continuous
variable and categorical variables, respectively. e 12 week outcomes were compared between study groups,
using ANCOVA (analysis of covariance) with the adjustment of baseline values. Data management, variable
transformations, and other statistical analyses were conducted using SAS 9.2 (Statistical Analysis System, Cary,
NC, 2008). Dierence in changes of bacterial relative abundances over 12 week intervention among the treatment
groups were compared using the non-parametric Kruskal–Wallis test in IBM SPSS Statistics version 23. Only
the signicant genera and species from the Kruskal–Wallis test were further tested with Mann–Whitney test to
assess the dierences between treatments. Bonferroni correction was used to correct the probability for multiple
comparisons. P values < 0.05 were considered statistically signicant.
Data Availability
e datasets generated during the current study are available from the corresponding author on reasonable request.
References
1. Amaro-Ortiz, A., Yan, B. & D’Orazio, J. A. Ultraviolet radiation, aging and the sin: prevention of damage by topical cAMP
manipulation. Molecules 19, 6202–6219, https://doi.org/10.3390/molecules19056202 (2014).
2. Cadet, J., Doui, T. & avanat, J. L. Oxidatively generated damage to cellular DNA by UVB and UVA radiation. Photochemistry and
photobiology 91, 140–155, https://doi.org/10.1111/php.12368 (2015).
3. Che, D. N. et al. Protective eects of grape stem extract against UVB-induced damage in C57BL mice sin. Journal of photochemistry
and photobiology. B, Biology 173, 551–559, https://doi.org/10.1016/j.jphotobiol.2017.06.042 (2017).
4. ang, S. J. et al. Benecial eects of dried pomegranate juice concentrated powder on ultraviolet B-induced sin photoaging in
hairless mice. Experimental and therapeutic medicine 14, 1023–1036, https://doi.org/10.3892/etm.2017.4626 (2017).
5. Seeram, N. P. S. .N. & Heber, D. Pomegranates: Ancient roots modern medicine., (CC Press & Taylor and Francis Group, 2006).
6. Gomez-Caravaca, A. M. et al. Determination of the major phenolic compounds in pomegranate juices by HPLC-DAD-ESI-MS.
Journal of agricultural and food chemistry 61, 5328–5337, https://doi.org/10.1021/jf400684n (2013).
7. Cerda, B., Periago, P., Espin, J. C. & Tomas-Barberan, F. A. Identication of urolithin a as a metabolite produced by human colon
microflora from ellagic acid and related compounds. Journal of agricultural and food chemistry 53, 5571–5576, https://doi.
org/10.1021/jf050384i (2005).
8. Seeram, N. P. et al. Pomegranate juice ellagitannin metabolites are present in human plasma and some persist in urine for up to 48
hours. e Journal of nutrition 136, 2481–2485 (2006).
9. Mertens-Talcott, S. U., Jilma-Stohlawetz, P., ios, J., Hingorani, L. & Derendorf, H. Absorption, metabolism, and antioxidant eects
of pomegranate (Punica granatum l.) polyphenols aer ingestion of a standardized extract in healthy human volunteers. Journal of
agricultural and food chemistry 54, 8956–8961, https://doi.org/10.1021/jf061674h (2006).
10. Espin, J. C., Larrosa, M., Garcia-Conesa, M. T. & Tomas-Barberan, F. Biological signicance of urolithins, the gut microbial ellagic
Acid-derived metabolites: the evidence so far. Evidence-based complementary and alternative medicine: eCAM 2013, 270418, https://
doi.org/10.1155/2013/270418 (2013).
11. Pacheco-Palencia, L. A., Noratto, G., Hingorani, L., Talcott, S. T. & Mertens-Talcott, S. U. Protective effects of standardized
pomegranate (Punica granatum L.) polyphenolic extract in ultraviolet-irradiated human sin broblasts. Journal of agricultural and
food chemistry 56, 8434–8441, https://doi.org/10.1021/jf8005307 (2008).
12. Bae, J. Y. et al. Dietary compound ellagic acid alleviates sin wrinle and inammation induced by UV-B irradiation. Experimental
dermatology 19, e182–190, https://doi.org/10.1111/j.1600-0625.2009.01044.x (2010).
13. Par, H. M. et al. Extract of Punica granatum inhibits sin photoaging induced by UVB irradiation. International journal of
dermatology 49, 276–282, https://doi.org/10.1111/j.1365-4632.2009.04269.x (2010).
14. Afaq, F., Zaid, M. A., han, N., Dreher, M. & Muhtar, H. Protective eect of pomegranate-derived products on UVB-mediated damage
in human reconstituted sin. Experimental dermatology 18, 553–561, https://doi.org/10.1111/j.1600-0625.2008.00829.x (2009).
15. han, N., Syed, D. N., Pal, H. C., Muhtar, H. & Afaq, F. Pomegranate fruit extract inhibits UVB-induced inammation and
proliferation by modulating NF-appaB and MAP signaling pathways in mouse sin. Photochemistry and photobiology 88,
1126–1134, https://doi.org/10.1111/j.1751-1097.2011.01063.x (2012).
16. Hayouni, E. A. et al. Hydroalcoholic extract based-ointment from Punica granatum L. peels with enhanced in vivo healing potential
on dermal wounds. Phytomedicine: international journal of phytotherapy and phytopharmacology 18, 976–984, https://doi.
org/10.1016/j.phymed.2011.02.011 (2011).
17. asai, ., Yoshimura, M., oga, T., Arii, M. & awasai, S. Eects of oral administration of ellagic acid-rich pomegranate extract on
ultraviolet-induced pigmentation in the human sin. Journal of nutritional science and vitaminology 52, 383–388 (2006).
18. Henning, S. M. et al. Variability in the antioxidant activity of dietary supplements from pomegranate, mil thistle, green tea, grape
seed, goji, and acai: eects of in vitro digestion. Journal of agricultural and food chemistry 62, 4313–4321, https://doi.org/10.1021/
jf500106r (2014).
Content courtesy of Springer Nature, terms of use apply. Rights reserved

10
SCIENTIFIC REPORTS | (2019) 9:14528 | https://doi.org/10.1038/s41598-019-50926-2
www.nature.com/scientificreports
www.nature.com/scientificreports/
19. usso, M. et al. Analysis of phenolic compounds in dierent parts of pomegranate (Punica granatum) fruit by HPLC-PDA-ESI/MS
and evaluation of their antioxidant activity: application to dierent Italian varieties. Analytical and bioanalytical chemistry, https://
doi.org/10.1007/s00216-018-0854-8 (2018).
20. Larrosa, M. et al. Anti-inammatory properties of a pomegranate extract and its metabolite urolithin-A in a colitis rat model and
the effect of colon inflammation on phenolic metabolism. The Journal of nutritional biochemistry 21, 717–725, https://doi.
org/10.1016/j.jnutbio.2009.04.012 (2010).
21. Ghavipour, M. et al. Pomegranate extract alleviates disease activity and some blood biomarers of inammation and oxidative stress
in heumatoid Arthritis patients. European journal of clinical nutrition 71, 92–96, https://doi.org/10.1038/ejcn.2016.151 (2017).
22. Gao, C., Chen, H., Niu, C., Hu, J. & Cao, B. Protective eect of Schizandrin B against damage of UVB irradiated sin cells depend on
inhibition of inammatory pathways. Bioengineered 8, 36–44, https://doi.org/10.1080/21655979.2016.1227572 (2017).
23. Li, Z. et al. Antimicrobial Activity of Pomegranate and Green Tea Extract on Propionibacterium Acnes, Propionibacterium
Granulosum, Staphylococcus Aureus and Staphylococcus Epidermidis. Journal of drugs in dermatology: JDD 14, 574–578 (2015).
24. Yoshida, S., Hiradate, S., oitabashi, M., amo, T. & Tsushima, S. Phyllosphere Methylobacterium bacteria contain UVA-absorbing
compounds. Journal of photochemistry and photobiology. B, Biology 167, 168–175, https://doi.org/10.1016/j.jphotobiol.2016.12.019
(2017).
25. Choi, S. Y. et al. Exposure of human melanocytes to UVB twice and subsequent incubation leads to cellular senescence and
senescence-associated pigmentation through the prolonged p53 expression. Journal of dermatological science. https://doi.
org/10.1016/j.jdermsci.2018.02.016 (2018).
26. Finegold, S. M. et al. Pomegranate extract exhibits in vitro activity against Clostridium dicile. Nutrition 30, 1210–1212, https://doi.
org/10.1016/j.nut.2014.02.029 (2014).
27. L ee, C. J., Chen, L. G., Liang, W. L. & Wang, C. C. Multiple Activities of Punica granatum Linne against Acne Vulgaris. International
journal of molecular sciences 18, https://doi.org/10.3390/ijms18010141 (2017).
28. Yoshimura, M., Watanabe, Y., asai, ., Yamaoshi, J. & oga, T. Inhibitory eect of an ellagic acid-rich pomegranate extract on
tyrosinase activity and ultraviolet-induced pigmentation. Bioscience, biotechnology, and biochemistry 69, 2368–2373, https://doi.
org/10.1271/bbb.69.2368 (2005).
29. Afaq, F., han, N., Syed, D. N. & Muhtar, H. Oral feeding of pomegranate fruit extract inhibits early biomarers of UVB radiation-
induced carcinogenesis in SH-1 hairless mouse epidermis. Photochemistry and photobiology 86, 1318–1326, https://doi.
org/10.1111/j.1751-1097.2010.00815.x (2010).
30. ang, S. J. et al. Inhibitory effects of pomegranate concentrated solution on the activities of hyaluronidase, tyrosinase, and
metalloproteinase. Journal of cosmetic science 66, 145–159 (2015).
31. Wood, S. M. et al. Protective eects of a novel nutritional and phytonutrient blend on ultraviolet radiation-induced sin damage and
inammatory response through aging defense mechanisms. Journal of cosmetic dermatology 16, 491–499, https://doi.org/10.1111/
jocd.12295 (2017).
32. Egoumenides, L. et al. A Specic Melon Concentrate Exhibits Photoprotective Eects from Antioxidant Activity in Healthy Adults.
Nutrients 10, https://doi.org/10.3390/nu10040437 (2018).
33. Byrd, A. L., Belaid, Y. & Segre, J. A. The human sin microbiome. Nature reviews. Microbiology 16, 143–155, https://doi.
org/10.1038/nrmicro.2017.157 (2018).
34. Grice, E. A. et al. Topographical and temporal diversity of the human sin microbiome. Science 324, 1190–1192, https://doi.
org/10.1126/science.1171700 (2009).
35. Ze euwen, P. L. et al. Microbiome dynamics of human epidermis following sin barrier disruption. Genome biology 13, 101, https://
doi.org/10.1186/gb-2012-13-11-r101 (2012).
36. Li, Z. et al. Pomegranate extract induces ellagitannin metabolite formation and changes stool microbiota in healthy volunteers. Food
& function 6, 2487–2495, https://doi.org/10.1039/c5fo00669d (2015).
37. Freedland, S. J. et al. A double-blind, randomized, neoadjuvant study of the tissue eects of POMx pills in men with prostate cancer
before radical prostatectomy. Cancer prevention research 6, 1120–1127, https://doi.org/10.1158/1940-6207.CAP-12-0423 (2013).
38. Yang, J. et al. Soy protein isolate does not aect ellagitannin bioavailability and urolithin formation when mixed with pomegranate
juice in humans. Food chemistry 194, 1300–1303, https://doi.org/10.1016/j.foodchem.2015.08.099 (2016).
39. Fitzpatric, T. B. e validity and practicality of sun-reactive sin types I through VI. Archives of dermatology 124, 869–871 (1988).
40. Firooz, A. et al. Variation of biophysical parameters of the sin with age, gender, and body region. eScienticWorldJournal 2012,
386936, https://doi.org/10.1100/2012/386936 (2012).
41. Group, N. H. W. et al. e NIH Human Microbiome Project. Genome research 19, 2317–2323, https://doi.org/10.1101/gr.096651.109
(2009).
42. Conlan, S., ong, H. H. & Segre, J. A. Species-level analysis of DNA sequence data from the NIH Human Microbiome Project. PloS
one 7, e47075, https://doi.org/10.1371/journal.pone.0047075 (2012).
43. DeSantis, T. Z. et al. Greengenes, a chimera-checed 16S rNA gene database and worbench compatible with AB. Applied and
environmental microbiology 72, 5069–5072, https://doi.org/10.1128/AEM.03006-05 (2006).
44. Caporaso, J. G. et al. QIIME allows analysis of high-throughput community sequencing data. Nature methods 7, 335–336, https://
doi.org/10.1038/nmeth.f.303 (2010).
45. Lozupone, C. & night, . UniFrac: a new phylogenetic method for comparing microbial communities. Applied and environmental
microbiology 71, 8228–8235, https://doi.org/10.1128/AEM.71.12.8228-8235.2005 (2005).
46. Langille, M. G. et al. Predictive functional proling of microbial communities using 16S rNA marer gene sequences. Nature
biotechnology 31, 814–821, https://doi.org/10.1038/nbt.2676 (2013).
Acknowledgements
Supported by departmental funds from the Center for Human Nutrition, Department of Medicine, David Geen
School of Medicine, University of California, Los Angeles.
Author Contributions
S.M.H. wrote the main manuscript text, J.Y., Y.X., J.L., E.H.P. and M.H. performed QIIME and microbiota data
analysis, R.P.L. and J.H. performed chemical analyses, I.G. and G.T. coordinated clinical study, J.K., Z.L. and
D.H. participated in manuscript preparation and C.H.T. performed statistical analysis. All authors reviewed the
manuscript.
Additional Information
Competing Interests: e authors declare no competing interests.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional aliations.
Content courtesy of Springer Nature, terms of use apply. Rights reserved

11
SCIENTIFIC REPORTS | (2019) 9:14528 | https://doi.org/10.1038/s41598-019-50926-2
www.nature.com/scientificreports
www.nature.com/scientificreports/
Open Access This article is licensed under a Creative Commons Attribution 4.0 International
License, which permits use, sharing, adaptation, distribution and reproduction in any medium or
format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Cre-
ative Commons license, and indicate if changes were made. e images or other third party material in this
article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the
material. If material is not included in the article’s Creative Commons license and your intended use is not per-
mitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the
copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.
© e Author(s) 2019
Content courtesy of Springer Nature, terms of use apply. Rights reserved

1.
2.
3.
4.
5.
6.
Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”), for small-
scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are maintained. By
accessing, sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use (“Terms”). For these
purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or a personal
subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription
(to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the Creative Commons license used will
apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data internally within
ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not
otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of companies unless we have your permission as
detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that Users may
not:
use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to circumvent access
control;
use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil liability, or is
otherwise unlawful;
falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by Springer Nature in
writing;
use bots or other automated methods to access the content or redirect messages
override any security feature or exclusionary protocol; or
share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer Nature journal
content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates revenue,
royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain. Springer Nature journal
content cannot be used for inter-library loans and librarians may not upload Springer Nature journal content on a large scale into their, or any
other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any information or
content on this website and may remove it or features or functionality at our sole discretion, at any time with or without notice. Springer Nature
may revoke this licence to you at any time and remove access to any copies of the Springer Nature journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express or implied
with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or warranties imposed by law,
including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be licensed
from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other manner not
expressly permitted by these Terms, please contact Springer Nature at
onlineservice@springernature.com