Clin Rheumatol
DOI 10.1007/s10067-016-3238-5
ORIGINAL ARTICLE
Changes in fecal microbiota and metabolomics in a child
with juvenile idiopathic arthritis (JIA) responding to two
treatment periods with exclusive enteral nutrition (EEN)
Lillemor Berntson 1,4 & Peter Agback 2 & Johan Dicksved 3
Received: 18 February 2016 / Revised: 7 March 2016 / Accepted: 15 March 2016
# International League of Associations for Rheumatology (ILAR) 2016
Abstract The microbiome and immune system of the diges-
tive tract are highly important in both health and disease.
Exclusive enteral nutrition (EEN) is a common anti-
inflammatory treatment in children with Crohn’s disease in
the European countries, and the mechanism is most likely
linked to changes in the intestinal microbiome. In the present
study, EEN was given in two treatment periods several months
apart to a patient with very severe, disabling juvenile idiopath-
ic arthritis (JIA), with a remarkable clinical response as the
result. The aim of the present study was to study how the EEN
treatment influenced the microbiome and metabolome of this
patient. Fecal samples from before, during, and between treat-
ments with EEN were studied. The microbiome was analyzed
by sequencing of 16S rRNA amplicons using Illumina MiSeq,
and the metabolome was analyzed using nuclear magnetic
resonance. The microbiome changed markedly from treatment
with EEN, with a strong reduction of the Bacteroidetes phy-
lum. Metabolic profiles showed clear differences before,
Key messages In this patient with JIA, exclusive enteral nutrition had a
remarkable anti-inflammatory effect in two treatment periods, corre-
sponding to changes in fecal microbiota and metabolites.
* Lillemor Berntson
lillemor.berntson@telia.com
1
Department of Women’s and Children’s Health, Uppsala University,
Uppsala, Sweden
2
Department of Chemistry and Biotechnology, Swedish University of
Agricultural Sciences, Uppsala, Sweden
3
Department of Animal Nutrition and Management, Swedish
University of Agricultural Sciences, Uppsala, Sweden
4
Department of Pediatrics, Unit for Pediatric Rheumatology, Uppsala
University Hospital, Uppsala S-75185, Sweden
during, and between treatment with EEN, where butyrate,
propionate, and acetate followed a cyclic pattern with the low-
est levels at the end of each treatment period. This patient with
JIA showed remarkable clinical improvement after EEN treat-
ment, and we found corresponding changes in both the fecal
microbiome and the metabolome. Further studies are needed
to explore the pathophysiological role of the intestinal canal in
children with JIA.
Keywords Exclusive enteral nutrition . Juvenile idiopathic
arthritis . Metabolomics . Microbiota
Introduction
It has become widely known that the symbiosis between the
gastrointestinal microbiota and the immune system has an
important role in health. This research field has expanded
dramatically within the last decade, not least in regard to in-
flammatory diseases. Juvenile idiopathic arthritis (JIA) is the
most common inflammatory joint disease in children.
There is very little knowledge about the gastrointestinal
immune system in JIA and the disease is heterogeneous, com-
prising seven categories [1]. For one of the seven categories of
the disease, enthesitis-related arthritis (ERA), studies have in-
dicated an overrepresentation of subclinical inflammation in
the gut [2–4]. In Finnish patients with JIA and gastrointestinal
symptoms without proven IBD, increased HLA-DR expres-
sion was found in the ileal mucosa, irrespective of disease
category, suggesting an enhanced intestinal antigen presenta-
tion [5]. The same group found a correlation between clinical
disease activity and low gene expression of tolerogenic medi-
ators in the ileum [5]. This supports the hypothesis that a link
exists between the gut immune system and JIA.

The intestinal microbiota is essential for the balance of
the immune system. When operating optimally, the immune
system and microbiota interact in a dialog that controls
misdirected immune responses [6]. The microbiota in adults
with rheumatoid arthritis (RA) [7], ankylosing spondylitis
[8], or psoriatic arthritis [9] has been shown to be different
than that in healthy individuals, but knowledge is scarce
about the immunological consequences. There are two stud-
ies so far on microbiota in children with JIA, one focuses on
ERA [10], the other has included mostly oligoarticular and
rheumatoid factor-negative polyarticular JIA [11]. Both stud-
ies show a different fecal microbiota in children with JIA
compared with controls [10, 11]. Studies in children with
Crohn’s disease (CD) have shown a reduced diversity in
intestinal microbiota and a composition that differs from that
in healthy children [12, 13].
The main theory behind exclusive enteral nutrition (EEN),
one of the most common treatments in children with CD, is
that it changes the microbiota to an immunologically more
preferable composition [14]. In many countries, EEN is the
most common treatment for CD in children today. EEN has
been shown to have an anti-inflammatory effect on intestinal
mucosa and to lead to improved nutritional status and a re-
duced need for corticosteroids [15].
In children with CD, EEN has been proven to influence
levels of short chain fatty acids (SCFAs) in fecal samples
[16, 17]. Under the anaerobic conditions of the large intestine,
undigested carbohydrates are fermented mainly to SCFAs
(such as acetate, propionate, and butyrate). A mounting body
of evidence indicates that those microbial metabolites have
profound effects on T cells and directly and indirectly regulate
their differentiation [18]. Thus, a change in microbiota may
influence levels of SCFAs.
Levels of SCFAs can be assessed as part of metabolomics,
an analysis approach to study the complete set of metabolites
present in biological samples. It describes the final products of
cellular processes, both qualitatively and quantitatively.
Among other applications, metabolomics enables mapping
of biochemical alterations involved in the pathogenesis of
diseases [19]. Nuclear magnetic resonance (NMR) in particu-
lar has the ability to simultaneously detect and structurally
characterize an abundance of metabolic components, even
when their identities are unknown [20].
At the Pediatric Rheumatology Unit at Uppsala University
Children’s Hospital, an investigation of the anti-inflammatory
effect of EEN in JIA is being carried out following approval
from the local ethics committee. The first patient in our study
showed a remarkable anti-inflammatory response following
two treatment periods with EEN (Fig 1) [21].
The aims of this study were to identify changes in fecal
microbiota as well as fecal metabolites and to further explore
the obvious anti-inflammatory effect of two treatment periods
with EEN in this patient.
Clin Rheumatol
Materials and methods
Study design
This was the first patient enrolled in a pilot study on exclusive
enteral nutrition as an anti-inflammatory treatment for chil-
dren with JIA. The patient had onset of severe polyarticular
disease at 3.2 years of age and was negative for RF (rheuma-
toid factor), anti-CCP (anti-cyclic citrullinated peptides),
ANA (anti-nuclear antibodies), and HLA-B27 (human leuko-
cyte antigen). The patient had no heredity for psoriasis, CD, or
ankylosing spondylitis. She was included in the study at
7.4 years of age. At that time, she had been through several
treatment periods with DMARDs (disease-modifying anti-
rheumatic drugs) and bDMARDs (biological DMARDs) with
disappointing results.
Exclusive enteral nutrition was given in two periods
of 6.5 weeks each, several months apart. During treat-
ment, the patient and parents were in close contact with a
dietician and a physician, who performed regular follow-
ups. The dietician kept close control of the nutritional
need and patient compliance. Clinical and laboratory sta-
tus were assessed before, during, and after treatment pe-
riods. The clinical results have been published earlier
(Fig 1) [21]. Due to clinical symptoms of gastritis, fol-
lowing the first period of EEN, a gastro-colonoscopy was
performed 2 weeks after the first period of EEN. The
mucosa was normal, without signs of recent inflamma-
tion, and microscopic investigation of biopsies showed
normal results. Fecal samples were collected before the
first treatment period with EEN (S0), after 3 weeks of
EEN (S1), after 5 weeks of EEN (S2), between the two
periods of EEN (S3 and S4), before the second EEN
treatment (S5), and finally, after 5 weeks of the second
EEN treatment (S6) (Fig 1).
The fecal microbiota was analyzed using sequencing of the
V3 and V4 regions of the gene encoding 16S rRNA [22], and
fecal metabolites were analyzed using NMR portions from the
same samples. In addition, fecal calprotectin was analyzed
before and after treatment through routine ELISA analysis,
reference value <50 mg/kg.
Ethical consideration
The study was approved by the regional ethics committee in
Uppsala County (Dnr 2012/378). Oral as well as written con-
sent was obtained from the patient’s parents.
16S rRNA analysis
DNA was extracted from 200 mg of fecal samples using a
procedure that has been described elsewhere [22]. In brief,
bead-beating was used to disrupt bacterial cell walls using
Clin Rheumatol
Fig. 1 Assessments of clinical
variables during more than 1 year
in a patient with juvenile
idiopathic arthritis, treated with
exclusive enteral nutrition for two
Affected joints (n)
6.5-week periods. Active joints:
filled circles. Joints with limited
range of movement: open circles.
CHAQ (child health assessment
questionnaire (0–3): filled
squares. EEN exclusive enteral
nutrition S0–S6 shows time
points of fecal sampling. Adapted
from [Berntson L, Clin
Rheumatol (2014)33:1173-75],
with permission
0.1-mm zirconia/silica beads (BioSpec products,
Bartlesville, USA) for 3 min at 20 Hz on a TissueLyser
Adapter Set 2X24 (Qiagen, Düsseldorf, Germany). The
DNA was then isolated and purified using an easyMAG
NucliSENS extractor off-board protocol (bioMèrieux,
Marcy l’Étoile, France). Sequencing libraries were pre-
pared by amplifying the V3–V4 regions of the 16S
rRNA gene using the 341f-805r primers, described by
Hugerth et al. [23]. After initial amplification, a second
PCR was performed to attach Illumina adapters, as well as
barcodes that allowed for multiplexing. Details on 16S
rRNA gene primer sequences, amplification conditions,
and sample barcodes have been published earlier [24].
The processed samples were sequenced using the
Illumina MiSeq platform producing 300 base pair reads.
Bioinformatic processing of sequences
The PCR primer sequences were trimmed off, and paired-
end reads were merged using SeqPrep version 1.1 (https://
github.com/jstjohn/SeqPrep) with default parameters. The
sequences were further processed in the QIIME 1.8.0
pipeline (Quantitative Insights into Microbial Ecology)
[25]. Assignment of sequences to operational taxonomic
units (otu) was based on a closed reference otu strategy.
Using the UCLUST [26] algorithm built into the QIIME
pipeline, sequences were clustered at 97 % identity against
the Greengenes reference database [27]. Principal coordinate
analysis (PCoA) based on a Bray Curtis distance metrics
was used to identify clustering of samples, and the microbial
alpha diversity was assessed using Shannon’s index.
Nuclear magnetic resonance
Metabolic profiles from feces were generated with nuclear
magnetic resonance (NMR). Preparation of the samples
10 1.0
EEN EEN
8 0.8
6 0.6
CHAQ
4 0.4
2 0.2
0 0.0
0 10 20 30 40 50 60 Weeks
S0 S1 S2 S3 S4 S5 S6
was performed by mixing 0.2 g of feces with 700-ml
phosphate buffer (150 mM, pH 7.4, 0.01 % TSP). The
samples were centrifuged for 5 min at 17, 000g, and the
supernatant was collected and filtered through 0.2-μm sy-
ringe filters. Metabolic fingerprints were generated using
a Bruker Avance III 600-MHz spectrometer (Bruker
BioSpin, Solna, Sweden), and measurements were taken
using a 5-mm tube in a cryoprobe (QCI-P) at 25 °C. The
chemical shifts were referenced to the internal standard
(TSP) added to each sample. 3D shimming and pulse cal-
ibration were performed on each sample before recording
the spectra. A pulse sequence with the removal of the
water peak by excitation sculpting was used for 1H-
NMR with a relaxation delay of 4 s and 64 scans.
Spectral phase and baseline were manually corrected.
Forty-seven metabolite concentrations were measured for
each sample using Chenomx nmr suite 7.51 (Chenomx
Inc., Edmonton, Canada).
Principal component analysis (PCA) was used to evaluate
clustering of samples based on their metabolic profiles.
Results
The clinical improvement after treatment with EEN in this
patient has already been described and published [21] and is
presented in Fig 1. Clinical signs like the number of in-
flamed joints, morning stiffness, global assessment of pain
(VAS 0-100 mm), and global assessment of disease activity
by parents and doctor (VAS 0-100 mm), as well as motor
function as assessed using the Child Health Assessment
Questionnaire (CHAQ), improved remarkably both times.
Levels of fecal calprotectin were never raised.
The sequence analysis generated 55,084 sequences per
sample on average (range 39 978–106 631). Sample S0 had
to be excluded due to an error in its sequencing. The microbial

diversity did not differ between samples. However, principal
coordinate analysis (PCoA) on the microbiota structure
showed a dramatic effect of the EEN treatment (Fig 2). The
samples collected after 5 weeks of EEN (S2 and S6) clustered
together despite different treatment periods and were clearly
separated from the non-EEN samples.
The distribution of the main phyla Firmicutes and
Bacteroidetes changed markedly with EEN treatment (Fig 3).
The proportion of Firmicutes was highest and Bacteroidetes
lowest in the two samples taken at 5 weeks in both treatment
periods (S2, S6). Conversely, Bacteroidetes was highest in
between treatment periods and was also higher after 3 weeks
of EEN (S1) as well as before a new treatment period (S5)
compared to samples at 5 weeks of treatment (S2, S6).
Much like the microbiota results, metabolic profiling of the
samples showed a clear effect of the EEN treatment, with
distinct clustering of samples from before (S0, S5), during
(S1, S2, S6), and in between EEN treatment periods (S3, S4)
(Fig 4). Interestingly, the metabolic profiles were different also
between symptomatic (S0, S5) and asymptomatic (S3, S4)
stages of the disease (Fig 4). Short-chain fatty acids are the
main metabolites produced from microbial fermentation and
are normally the most abundant metabolites in stool samples.
The EEN treatment influenced the levels of butyrate, propio-
nate, and acetate, which decreased during both treatment pe-
riods (Fig 5).
Discussion
This young patient with severe JIA had a remarkable anti-
inflammatory response following two treatment periods with
Fig. 2 A principal coordinate analysis plot that illustrates how samples
cluster based on composition of the microbiome. Circles represent
samples collected before (S5) or in between (S3, S4) treatment periods
with exclusive enteral nutrition (EEN), whereas triangles and squares
represent samples collected after three (S1) or five (S2, S6) weeks of
EEN treatment. Percent values in parentheses show the variation in data
explained by each principal component (PC)
Clin Rheumatol
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
S1 S2 S3 S4 S5 S6
Bacteroidetes Firmicutes Other
Fig. 3 A bar chart showing the proportions of Bacteroidetes and
Firmicutes in relation to exclusive enteral nutrition (EEN) treatment. S1
shows the proportions after 3 weeks with EEN, S2 after 5 weeks with
EEN, S3 2 weeks after the first finished EEN treatment, S4 12 weeks after
the first finished EEN treatment, S5 days before start of second EEN
treatment period, and S6 after 5 weeks of the second EEN treatment
exclusive enteral nutrition. The first treatment period resulted
in approximately 6 months without any disease activity or
morning stiffness.
Treatment with EEN changed the fecal microbiota con-
siderably, with a clear distinction between samples col-
lected during treatment periods compared with both be-
fore and between treatment periods (Fig 2). The EEN
treatment contributed to changed proportions between
the Bacteroidetes and Firmicutes phyla, with a similar
pattern after two separate 6.5-week treatment periods with
EEN. The proportion of Bacteroidetes was higher before
and between treatment periods. In a recent study on fecal
Fig. 4 A principal component analysis plot that illustrates how samples
cluster based on metabolic profiles. Circles represent samples collected
before (S0, S5) or in between (S3, S4) treatment periods with exclusive
enteral nutrition (EEN), whereas triangles and squares represents
samples collected after three (S1) or five (S2, S6) weeks of EEN
treatment. Symbols colored in gray represent samples with active
disease stage and those in black, passive disease stage. Percent values
in parentheses show the variation in data explained by each principal
component (PC)
Clin Rheumatol
800
EEN EEN Propionate
Butyrate
600
Acetate
Conc (mM)
400
200
0
0 10 20 30 40 50 Weeks
S0 S1 S2 S3 S4 S5 S6
Fig. 5 Levels of the short-chain fatty acids butyrate, propionate, and
acetate (in millimolar, mM) before, between, and after treatment periods
of exclusive enteral nutrition (EEN). S0–S6 show time points of fecal
sampling
microbiome in new-onset JIA, the fecal flora was charac-
terized by a low level of Firmicutes and an abundance of
Bacteroidetes [11]. Our result, with a lower proportion of
Bacteroidetes during treatment with EEN, is in line with
the results of studies in pediatric CD patients [14, 17]
where the major groups within Bacteroidetes phylum
were suggested to be associated with disease activity
[14]. In addition, the fecal microbiota in children with
type 1 diabetes coincided with a high abundance of
Bacteroidetes and low abundance of Firmicutes [28, 29].
Prevotella copri is one of the main bacterial taxa of the
Bacteroidetes phylum and has been found in increased
abundance in adults with RA [7]. It has emerged as a
possible candidate responsible for the priming of aberrant
systemic immunity in RA and discussed as a driver for
progression of the disease. However, we could not find
Prevotella copri in the present study. The diversity of the
microbiota has also been shown to correlate with disease
activity in pediatric CD patients [13]. There are no equiv-
alent studies in JIA, but our data did not reveal any dif-
ference in diversity between samples.
In two studies of children with CD, 8 weeks of treatment
with EEN induced a significant change in the microbiota
that persisted for 2–4 months [14, 17]. In the present study,
the shift in microbiota from EEN treatment was not sustain-
able and had changed dramatically 2 weeks after normal
eating was reintroduced. Earlier studies have shown that
gut microbiota in healthy individuals alters within days of
a changed diet [30]. The articular symptoms recurred ap-
proximately 6 months after the first treatment period and
10 months after the second. One could only speculate that
the shift in microbiota induces a more preferable immuno-
logical state with significant advantages that are sustained
for a longer time than the shift itself.
The metabolic fingerprint in this patient, analyzed
using metabolomics, was also altered by EEN treatment.
Me tabo lo mics ena bles m ap ping of bio che mical
alterations involved in the pathogenesis of diseases and
offers the opportunity to non-invasively identify diagnos-
tic and prognostic markers [31]. Data analysis of the
metabolic profiles showed that samples clustered in three
groups related to both treatment and symptoms (Fig 4).
There is, thus far, no other study presenting data on
metabolomics in JIA. In a study of adults with RA, the
pattern of metabolites in patients differed significantly
from that in controls, presenting a possibility to differen-
tiate between these groups with high sensitivity and
specificity [32].
The levels of SCFAs, e.g., butyrate, propionate, and
acetate, declined with EEN treatment in this patient
(Fig 5). SCFAs have garnered large attention due to both
their metabolic value and their function as signaling mol-
ecules. Butyrate in particular is known to have anti-
inflammatory and immuno-regulatory properties [14].
EEN treatment contributed to an anti-inflammatory effect,
but this was likely not linked to fecal levels of SCFAs in
our study, since the treatment contributed to decreased
levels of SCFAs. Fermentation of dietary fiber by the
colonic microbiota is the primary source for production
of SCFAs, and the decrease of SCFAs during EEN treat-
ment could be explained by the lack of fibers in the colon
during treatment. A similar result has been shown previ-
ously after EEN treatment in children with CD [17], but
contradictory results in CD have also been reported [16].
Our results are based on only one patient, and if further
conclusions are to be drawn regarding how SCFA levels
change in response to EEN, the study needs to be repeated
in a larger cohort.
In conclusion, this study demonstrates an example of a
patient benefitting from EEN as an anti-inflammatory treat-
ment for JIA. The treatment is challenging for both child
and parents since the patient fulfills their nutritional require-
ments exclusively through a liquid formula taken orally or
via a nasogastric tube. In order to treat a patient with EEN, a
close teamwork with a dietician, physician, and nurse is
mandatory. Also, we currently lack the knowledge of wheth-
er the beneficial effect is generalizable to all JIA patients and
what duration of EEN is needed. Our results demonstrated
that EEN treatment induced remission of JIA, which coin-
cided with a dramatic change in both the microbiome and
the metabolome. Use of Bmeta-omics^ techniques such as
metagenomics and metabolomics can provide important les-
sons regarding the link between diet, microbiome, gut mu-
cosal immunology, and inflammation in the joints, contrib-
uting to a better understanding of the etiology of the disease
as well as providing tools to reduce symptoms.
Acknowledgments This work was supported by grants from the
Department of Women’s and Children’s Health, Uppsala University
Hospital, Uppsala, and the Gillbergska Foundation, Uppsala.

Compliance with ethical standard The study was approved by the
regional ethics committee in Uppsala County (Dnr 2012/378). Oral as
well as written consent was obtained from the patient’s parents.
Disclosures None.
References
1. Petty RE et al (2004) International League of Associations for
Rheumatology classification of juvenile idiopathic arthritis: second
revision, Edmonton, 2001. J Rheumatol 31(2):390–392
2. Stoll ML, Punaro M, Patel AS (2011) Fecal calprotectin in children
with the enthesitis-related arthritis subtype of juvenile idiopathic
arthritis. J Rheumatol 38(10):2274–2275
3. Conti F et al (2005) Chronic intestinal inflammation and
seronegative spondyloarthropathy in children. Dig Liver Dis
37(10):761–767
4. Lionetti P et al (2000) Evidence of subclinical intestinal inflamma-
tion by 99m technetium leukocyte scintigraphy in patients with
HLA-B27 positive juvenile onset active spondyloarthropathy. J
Rheumatol 27(6):1538–1541
5. Arvonen M et al (2012) Altered expression of intestinal human
leucocyte antigen D-related and immune signalling molecules in
juvenile idiopathic arthritis. Clin Exp Immunol 170(3):266–273
6. Belkaid Y, Hand TW (2014) Role of the microbiota in immunity
and inflammation. Cell 157(1):121–141
7. Scher JU et al (2013) Expansion of intestinal Prevotella copri cor-
relates with enhanced susceptibility to arthritis. Elife 2:e01202
8. Costello ME et al (2014) Intestinal dysbiosis in ankylosing spon-
dylitis. Arthritis Rheumatol 67(3):686–691
9. Scher JU et al (2015) Decreased bacterial diversity characterizes the
altered gut microbiota in patients with psoriatic arthritis, resembling
dysbiosis in inflammatory bowel disease. Arthritis Rheumatol
67(1):128–139
10. Stoll ML et al (2014) Altered microbiota associated with abnormal
humoral immune responses to commensal organisms in enthesitis-
related arthritis. Arthritis Res Ther 16(6):486
11. Tejesvi MV et al (2015) Faecal microbiome in new-onset juvenile
idiopathic arthritis. Eur J Clin Microbiol Infect Dis
12. Gevers D et al (2014) The treatment-naive microbiome in new-
onset Crohn’s disease. Cell Host Microbe 15(3):382–392
13. Kolho KL et al (2015) Fecal microbiota in pediatric inflammatory
bowel disease and its relation to inflammation. Am J Gastroenterol
110(6):921–930
14. Leach ST, Mitchell HM, Eng WR, Zhang L, Day AS (2008)
Sustained modulation of intestinal bacteria by exclusive enteral
nutrition used to treat children with Crohn’s disease. Aliment
Pharmacol Ther 28(6):724–733
Clin Rheumatol
15. Zachos M, Tondeur M, Griffiths AM (2007) Enteral nutritional
therapy for induction of remission in Crohn’s disease. Cochrane
Database Syst Rev 1:CD000542
16. Tjellstrom B et al (2012) Effect of exclusive enteral nutrition on gut
microflora function in children with Crohn’s disease. Scand J
Gastroenterol 47(12):1454–1459
17. Gerasimidis K et al (2014) Decline in presumptively protective gut
bacterial species and metabolites are paradoxically associated with
disease improvement in pediatric Crohn’s disease during enteral
nutrition. Inflamm Bowel Dis 20(5):861–871
18. Furusawa Y et al (2013) Commensal microbe-derived butyrate in-
duces the differentiation of colonic regulatory T cells. Nature
504(7480):446–450
19. Fischer R, Bowness P, Kessler BM (2013) Two birds with one
stone: doing metabolomics with your proteomics kit. Proteomics
13(23-24):3371–3386
20. Puchades-Carrasco L, Palomino-Schatzlein M, Perez-Rambla C,
Pineda-Lucena A (2015) Bioinformatics tools for the analysis of
NMR metabolomics studies focused on the identification of clini-
cally relevant biomarkers. Brief Bioinform
21. Berntson L (2014) Anti-inflammatory effect by exclusive enteral
nutrition (EEN) in a patient with juvenile idiopathic arthritis (JIA):
brief report. Clin Rheumatol 33(8):1173–1175
22. Kampmann C, Dicksved J, Engstrand L, Rautelin H (2015)
Composition of human faecal microbiota in resistance to
Campylobacter infection. Clin Microbiol Infect 22 (1):
61.e61–61.e68
23. Hugerth LW et al (2014) DegePrime, a program for degenerate
primer design for broad-taxonomic-range PCR in microbial ecolo-
gy studies. Appl Environ Microbiol 80(16):5116–5123
24. Engstrand Lilja H, Wefer H, Nystrom N, Finkel Y, Engstrand L
(2015) Intestinal dysbiosis in children with short bowel syndrome
is associated with impaired outcome. Microbiome 3:18
25. Caporaso JG et al (2010) QIIME allows analysis of high-
throughput community sequencing data. Nat Methods 7(5):
335–336
26. Edgar RC (2010) Search and clustering orders of magnitude faster
than BLAST. Bioinformatics 26(19):2460–2461
27. Caporaso JG, Bittinger K, Bushman FD, DeSantis TZ, Andersen
GL, Knight R (2010) PyNAST: a flexible tool for aligning se-
quences to a template alignment. Bioinformatics 26(2):266–267
28. Giongo A et al (2011) Toward defining the autoimmune
microbiome for type 1 diabetes. ISME J 5(1):82–91
29. Murri M et al (2013) Gut microbiota in children with type 1 diabe-
tes differs from that in healthy children: a case-control study. BMC
Med 11:46
30. David LA et al (2014) Diet rapidly and reproducibly alters the
human gut microbiome. Nature 505(7484):559–563
31. Holmes E, Li JV, Athanasiou T, Ashrafian H, Nicholson JK (2011)
Understanding the role of gut microbiome-host metabolic signal
disruption in health and disease. Trends Microbiol 19(7):349–359
32. Madsen RK et al (2011) Diagnostic properties of metabolic pertur-
bations in rheumatoid arthritis. Arthritis Res Ther 13(1):R19