Next-generation RNA sequencing has expedited the identification of new non-coding RNA species (ncRNAs), thus ushering in the emerging field of ncRNA biology. The goals of this study were to catalogue the spectrum of different ncRNAs in serum and liver of patients with NAFLD and to compare expression of serum exRNAs between NAFLD patients and healthy control subjects.

1. James E. Nelson1, Ed Sendler2, Stephen A. Krawetz2, Kris V. Kowdley1
1 Liver Center of Excellence, Digestive Disease and Benaroya Research Institutes at Virginia Mason Medical Center, Seattle WA
2 Department of Obstetrics and Gynecology, Center for Molecular Medicine and Genetics, Wayne State University School of Medicine, Detroit, MI.
We have performed RNA-seq on 23 matched serum and liver tissue
specimens from patients with biopsy-proven NAFLD and on serum of 22
healthy non-diabetic persons using the Illumina HiScanSQ platform.
RNA was isolated using the miRNeasy kit (Qiagen). The TruSeq Small
RNA Sample Prep Kit (Illumina) was used for cDNA library construction.
Samples were then pooled and purified using gel electrophoresis based
size-fractionation. Sequence alignment, differential expression (DE)
analysis and bioinformatics were performed using the Genomatix
Software Suite.
DEEP SEQUENCING IDENTIFIES NOVEL CIRCULATING
AND HEPATIC ncRNA PROFILES IN NONALCOHOLIC
FATTY LIVER DISEASE PATIENTS
Next-generation RNA sequencing has expedited the identification of
new non-coding RNA species (ncRNAs), thus ushering in the emerging
field of ncRNA biology. Together non-coding RNAs are poorly
understood, but have been shown to have widely diverse regulatory
functions including for transcription, translation, splicing, imprinting,
protein localization and cellular structural integrity, among others.
Extracellular ncRNAs or exRNAs, including miRNAs, the best known
and most widely studied class of exRNAs, represent a newly discovered
means of cell-to-cell communication.
The goals of this study were to catalogue the spectrum of different
ncRNAs in serum and liver of patients with NAFLD and to compare
expression of serum exRNAs between NAFLD patients and healthy
control subjects. .
This work was funded by a Pioneer Award from the Wilske Center for Translational Research at Virginia Mason Medical Center to JEN
Disclosures: Kris V. Kowdley –Grant/Research Support: BMS, Merck/Schering Plough, Intercept,
Pharmasett, Abbott, Ikaria, Mochida, Zymogenetics, Conatus. All other authors have nothing to disclose.
Figure 2. Distribution of exRNAs detected by RNA-seq
in serum of NAFLD patients.
5 Most
abundant
serum
miRs
(healthy)
% total
reads
5 Most
abundant
serum miRs
(NAFLD)
% total
reads
NAFLD
liver rank
5 Most
abundant
liver miRs
% total
reads
NAFLD
serum
rank
486-3p 45.0% 486-5p 38.0% 33 192-5p 13.4% 6
486-5p 31.4% 486-3p 24.1% 39 122-5p 12.5% 166
92a-3p 4.5% 92a-3p 6.2% 7 148a-3p 10.6% 12
16-5p 2.3% 16-5p 3.1% 27 22-3p 7.1% 5
3184-3p 1.9% 22-3p 2.9% 4 3591-3p 6.6% 200
61%
5%
35%
Serum
35%
59%
6%
Liver
Unique hits Multiple hits
Insufficient quality reads
Figure 1. Comparison of RNA-seq read quality in
serum and liver of NAFLD patients
Type
Number
analyzed
Up regulated
exRNAs, n(%)
Down
regulated
exRNAs, n(%)
Log2 fold
change
(range)
Adj. P value
(range)
piRNA 32,194 500 (67%) 249 (33%) -8.3 to 12.1 .049 to 8.8e-18
miRNA 2019 43 (34%) 85 (66%) -6.9 to 12.4 0.046 to 7.2e-34
snoRNA 583 2 (3%) 57 (97%) -8.6 to 4.6 .046 to 6.6e-10
snRNA 474 1 (100%) 0 (0%) 9.1 5.8e-5
Figure 3. Distribution of ncRNAs detected by RNA-seq
in liver of NAFLD patients.
Type Name
Adj. P
value
Log2
fold
change
Known Functions
miRNA miR-5787 7.2e-34 7.8
represses cellular growth by targeting
eukaryotic translation initiation factor 5
miRNA miR-5586-5p 1.26e-25 12.4 ?
miRNA miR-512-3p 8.38e-23 8.1 differentiation, ECM, apoptosis
miRNA miR-107 4.31e-22 6.2 glucose homeostasis, insulin signaling
piRNA piR-43435 8.8e-18 11.7 ?
piRNA piR-30597 8.59e-17 10.5 ?
piRNA piR-59342 4.70e-15 10.6 ?
piRNA piR-47259 4.70e-15 9.2 ?
snoRNA SNORD47 (U47) 6.6e-10 -8.6
C/D box; 2'O-ribose methylation of 28S
rRNA C3866
snoRNA SNORA74 (U19) 1.25e-09 4.6
H/ACA Box; pre-rRNA processing,
pseudouridylation of residues U3741
and U3743 of 28S rRNA
snoRNA SNORA24 9.48e-08 -7.4
H/ACA Box; pseudouridylation of
residue U863 and U609 of 18S rRNA
snoRNA SNORA29 (ACA29) 2.28e-07 -5.5 H/ACA Box, target unknown
snRNA U3 5.8e-5 9.1 mRNA splicing
Table 3. Relationship of serum and liver miRNA levels
Table 1. Serum exRNAs are differentially expressed
in NAFLD vs healthy subjects
Table 2. Most significant differentially expressed exRNAs
in NAFLD vs healthy serum
 Numerous ncRNA classes are detectable in liver and serum, but in different proportions.
 10%, 6% and 2% of snoRNA, miRNA and piRNAs, detected in serum, respectively, were
differentially expressed between NAFLD and healthy subjects.
 The majority of miRNAs (66%) and snoRNAs (97%) are down regulated in NAFLD,
while most piRNAs (67%) are up regulated. One snRNA, U3, was up regulated.
 The most abundant miRs in serum of NAFLD patients are similar to that of healthy persons,
and are enriched in RBCs. Differential mechanisms of secretion among miRs likely also
contributes to the observed differences between hepatic and serum miR profiles.
 The differential expression of numerous snoRNA and piRNAs , suggest, like miRNAs, that
these ncRNAs may have a role in the pathophysiology of NAFLD and/or its
associated comorbidities.