Journals of Gerontology: Biological Sciences

cite as: J Gerontol A Biol Sci Med Sci, 2019, Vol. 74, No. 1, 1–8
doi:10.1093/gerona/gly048
Advance Access publication March 15, 2018

Original Article

Molecular Aging of Human Liver: An Epigenetic/

Transcriptomic Signature

Downloaded from https://academic.oup.com/biomedgerontology/article/74/1/1/4937953 by guest on 29 March 2023

Maria Giulia  Bacalini, PhD,1 Claudio  Franceschi, MD, PhD,1–3 Davide  Gentilini, PhD,4
Francesco  Ravaioli, MSc,2 Xiaoyuan  Zhou, PhD,5,6,7 Daniel  Remondini, PhD,8
Chiara  Pirazzini, PhD, 1 Cristina  Giuliani, PhD, 9 Elena  Marasco, MSc, 3
Noémie  Gensous, MSc,2 Anna Maria  Di Blasio, MD,10 Ewa  Ellis, PhD,11
Roberto  Gramignoli, PhD,12 Gastone  Castellani, PhD,9,3 Miriam  Capri, PhD,2,3
Stephen  Strom, PhD,13 Christine  Nardini, PhD,14–16 Matteo  Cescon, MD, PhD,17
Gian Luca Grazi, PhD,18,# and Paolo Garagnani, PhD2,3,14,19–21,#
1
IRCCS Istituto delle Scienze Neurologiche di Bologna, Italy. 2DIMES-Department of Experimental, Diagnostic and Specialty Medicine,
Alma Mater Studiorum, Bologna, Italy. 3CIG, Interdepartmental Center ‘L. Galvani’, Alma Mater Studiorum, Bologna, Italy. 4Department of
Brain and Behavioral Sciences, University of Pavia, Italy. 5Group of Clinical Genomic Networks, Key Laboratory of Computational Biology,
CAS-MPG Partner Institute for Computational Biology, Shanghai Institutes for Biological Sciences, PR China. 6University of Chinese
Academy of Sciences, Beijing, PR China. 7Department of Neurology, University of San Francisco, California. 8Department of Physics
and Astronomy (DIFA) and INFN Sez. Bologna, Alma Mater Studiorum, Italy. 9Department of Biological Geological and Environmental
Sciences, Laboratory of Molecular Anthropology and Centre for Genome Biology, University of Bologna, Italy. 10Istituto Auxologico Italiano
IRCCS, Milan, Italy. 11Department of Clinical Science, Intervention and Technology (CLINTEC), Karolinska Institutet, Stockholm, Sweden.
12
Department of Laboratory Medicine, Division of Pathology, Karolinska Institutet, Stockholm, Sweden. 13Department of Laboratory Medicine,
Karolinska Institute and Karolinska Universitetssjukhuset, Stockholm, Sweden. 14Department of Laboratory Medicine, Karolinska Institutet,
Stockholm, Sweden. 15CNR IAC “Mauro Picone”, Roma, Italy. 16Personal Genomics S.r.l., Verona, Italy. 17General Surgery and Transplant
Unit, Department of Medical and Surgical Sciences, Sant’Orsola-Malpighi Hospital, Bologna, Italy. 18Istituto Nazionale Tumori ‘Regina
Elena’, Roma, Italy. 19Applied Biomedical Research Center, S. Orsola-Malpighi Polyclinic, Bologna, Italy. 20Institute of Molecular Genetics
(IGM)-CNR, Unit of Bologna, Italy. 21Laboratory of Musculoskeletal Cell Biology, Rizzoli Orthopaedic Institute, Bologna, Italy.

Denotes co-senior authorship.

#

Address correspondence to: Claudio Franceschi, MD, PhD, DIMES-Department of Experimental, Diagnostic and Specialty Medicine,Alma Mater
Studiorum, Via San Giacomo 12, Bologna 40126, Italy. E-mail: [email protected]

Received: December 31, 2017; Editorial Decision Date: March 6, 2018

Decision Editor: David Le Couteur, MBBS, FRACP, PhD

Abstract
The feasibility of liver transplantation from old healthy donors suggests that this organ is able to preserve its functionality during aging. To
explore the biological basis of this phenomenon, we characterized the epigenetic profile of liver biopsies collected from 45 healthy liver donors
ranging from 13 to 90 years old using the Infinium HumanMethylation450 BeadChip. The analysis indicates that a large remodeling in DNA
methylation patterns occurs, with 8,823 age-associated differentially methylated CpG probes. Notably, these age-associated changes tended
to level off after the age of 60, as confirmed by Horvath’s clock. Using stringent selection criteria, we further identified a DNA methylation
signature of aging liver including 75 genomic regions. We demonstrated that this signature is specific for liver compared to other tissues and
that it is able to detect biological age-acceleration effects associated with obesity. Finally, we combined DNA methylation measurements with
available expression data. Although the intersection between the two omic characterizations was low, both approaches suggested a previously
unappreciated role of epithelial–mesenchymal transition and Wnt-signaling pathways in the aging of human liver.

© The Author(s) 2018. Published by Oxford University Press on behalf of The Gerontological Society of America. All rights reserved.
1
For permissions, please e-mail: [email protected].
2 Journals of Gerontology: BIOLOGICAL SCIENCES, 2019, Vol. 74, No. 1

Keywords: DNA methylation, Epigenetic clock, Epithelial–mesenchymal transition

The liver is a highly sophisticated metabolic factory. It performs a characterization, 10 with nonalcoholic fatty liver disease, and 15
vast array of biochemical functions necessary to maintain whole- patients with nonalcoholic steatohepatitis). Liver biopsies follow-up
body homeostasis and it is pivotal in integrating and elaborating sig- available for 23 individuals were discarded from our analysis.
nals that originate from peripheral tissues. Still, details on how and
to what extent liver physiology is affected during aging is an open Genome-Wide DNA Methylation Analysis
question in basic biological research. Compared to other organs, Genomic DNA was extracted from 45 liver biopsies (27 males and 18
liver has long been known to be unique in terms of extended func- females, age range 13–90 years) using the Qiagen kit QiAmp mini Kit,
tionality. Despite some age-related changes in morphology and func- following manufacturer’s instructions. DNA was bisulfite-converted
tion, liver appears to preserve its functionality at older ages much

Downloaded from https://academic.oup.com/biomedgerontology/article/74/1/1/4937953 by guest on 29 March 2023

using the EZ DNA Methylation-Gold Kit (Zymo Research) and ana-
better than other tissues (1), with important implications in transla- lyzed on the Infinium HumanMethylation450 BeadChip (Illumina)
tions research, including the selection of donors in liver transplants. following manufacturer’s instructions. Arrays were scanned by
Indeed, available data suggest that transplants from older donors HiScan (Illumina). Data processing and analysis are described in
have duration and success rates comparable to those from young Supplementary Data. DNA methylation data discussed in this publi-
donors (2,3). Collectively, these evidences suggest that liver aging cation have been deposited in NCBI’s Gene Expression Omnibus and
rate is decelerated compared to other organs. are accessible through GEO Series accession number GSE107039
In this frame, we recently reported that the proteasomes’ func- (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE107039),
tion and the microRNA expression in liver biopsies from healthy which includes also expression data (see below).
liver donors ranging from 12 to 92  years are largely preserved at
least until the age of 60 (4,5).
Gene-Targeted DNA Methylation analysis
To deepen and complement these findings, and on the basis of
DNA was extracted using the AllPrep Mini Kit (Illumina) following
recent literature (6–8), we here investigate genome-wide DNA methy-
manufacturer’s instructions. Site-specific DNA methylation analysis
lation in liver biopsies from 45 healthy donors with age ranging from
was performed by the EpiTYPER Assay. Briefly, 500 ng DNA were
13 to 90 years. In fact, DNA methylation is the epigenetic change that
converted via bisulfite treatment using EZ-96 DNA Methylation
received most attention in the study of human aging, due to the large
Kit (Zymo Research Corporation). Ten ng bisulfite-converted DNA
remodeling that it undergoes in life span (9) and to its ability to detect
were amplified using bisulfite-specific primers for ELOVL2, ZIC1,
acceleration/deceleration effects in the biological age of an individual
ZIC1_Shore, MACROD1, CCNJ, and CYP1B1 (Supplementary
(10,11). Finally, to get further insight into the molecular pathways
Table  3) using a step-down amplification routine as indicated in
altered during liver aging, we corroborated the epigenetic analysis
Suchiman and colleagues (15) (Supplementary Table 4). Methylation
with the matched characterization of gene expression changes.
data analysis was performed using the R stats package.

Methods Gene Expression Analysis

Total RNA (including microRNAs) was extracted from 33 liver
Samples
biopsies belonging to the same collection analyzed for genome-
Liver biopsies were selected from 45 heart-beating and brain-dead
wide DNA methylation (4) (11 females and 22 males, age range
donors (age range 13–90  years), as described in Capri and col-
13–90 years) using the mirVana microRNA Isolation Kit (Ambion),
leagues (4), following approval of the local ethical committee (code:
following manufacturer’s instructions. Of these 33 biopsies, 26 had
44/2008/Tess) and obtainment of written informed consent. No
also been analyzed for DNA methylation profiles. The HG-U133
donor organs were obtained from executed prisoners or other insti-
plus 2.0 GeneChip (Affymetrix) platform was used to evaluate gene
tutionalized persons. Donors’ death causes are described in Capri
expression, as already described (4). Analysis of gene expression data
and colleagues (4); no significant differences in bilirubin, alanine
is described in Supplementary Data.
aminotransferase, aspartate aminotransferase, and gamma-glutamyl
transferase between donors younger than 70  years and older than
70 years were observed.
Results
Human hepatocytes were derived from an independent set of
22 patients undergoing either partial or total hepatectomy and Identification of Age-Associated Changes in Liver
from 10 organ donors (Supplementary Table  1). Cells were iso- DNA Methylation
lated as reported in Gramignoli and colleagues (12), centrifuged We used the Illumina Infinium 450 k microarray to characterize
at 80 g for 6 minutes, and frozen as dry cell pellets for long-term DNA methylation of liver tissues collected from 45 individuals rang-
storage at −80°C. ing from 13 to 90 years of age.
DNA methylation data sets used to investigate DNA methyla- We run two types of analyses (see Methods section). The
tion in healthy tissues other than liver are reported in Supplementary first identified 8,823 differentially methylated positions (DMPs,
Data (Supplementary Table  2). Two data sets (GSE61256 and Figure 1A and Supplementary File 1). Most of these probes (5,772)
GSE48325) were used to assess the effects of obesity on age-depend- were positively associated with age. The overlap between our results
ent DNA methylation changes. GSE61256 includes 79 liver sam- and those recently published (8) (3,518/8,823 probes) is reported
ples from patients undergoing liver biopsy for assessment of liver in Supplementary File 1 and commented in Supplementary Data.
histology or for suspected nonalcoholic fatty liver disease and from The analysis of the chromatin states of the genomic regions harbor-
normal controls to exclude liver malignancy during oncological sur- ing age-DMPs (Supplementary Data and Supplementary Figure  1)
gery (13). GSE48325 (14) includes 85 liver samples from 62 patients showed an enrichment of hyper-age-DMPs in bivalent chromatin
(17 normal controls, 17 healthy obese, 3 obese without clinical domains, consistently with previous reports on other tissues (16).
Journals of Gerontology: BIOLOGICAL SCIENCES, 2019, Vol. 74, No. 1 3

Validation of the DNA Methylation Signature on

Isolated Hepatocytes
Although hepatocytes are the predominant cell types in human liver,
we cannot exclude a priori that the observed epigenetic alterations in
aging liver are in fact related to changes in cell types abundances in
the tissue. Therefore, we evaluated age-dependent DNA methylation
of six randomly selected age-DMRs (ELOVL2 island, ZIC1 island,
ZIC1 Shore, MACROD1 island, CCNJ island, and CYP1B1 island)
in a existing collection of hepatocytes isolated from liver tissues in an
independent cohort of 32 subjects (age range: 0–82 years). Pearson
correlation between DNA methylation and age was significant (p
value <0.05) for multiple CpG sites within each of the six regions

Downloaded from https://academic.oup.com/biomedgerontology/article/74/1/1/4937953 by guest on 29 March 2023

analyzed (Figure 2B, Supplementary Files 4 and 5) indicating that,
at least for these DMRs, the observed epigenetic changes occur in
hepatocytes.

Tissue Specificity of the DNA Methylation Signature

Subsequently, we investigated whether the signature identified in liver
could show similar age-dependent trends in other human tissues.
We used 14 publicly available data sets where DNA methylation is
measured in healthy tissues (Supplementary Data) from subjects of
different ages. An independent liver data set [GSE61258, considering
only subjects with a body mass index (BMI) <25] was included in
Figure  1.  Identification of age-DMPs (differentially methylated positions) in
this analysis in order to verify the reproducibility of the age-DMRs
human liver. (A) Volcano plot of age-DMPs. The difference between mean
identified in our study. For each of the 687 probes included in the 75
DNA methylation values in older (age 71–90 years) and younger individuals
(age 13–30  years) is plotted on the x-axis, while the nominal p value for age-DMRs, we considered the slope values of the linear regression
the regression between DNA methylation and age, corrected for sex and between DNA methylation and age in each tissue. Hierarchical clus-
batch, is on the y-axis (−1  × log10 scale). The dotted line corresponds to a tering using these tissue-specific slope values (Supplementary Data)
BH-corrected p value of 0.001. Probes with a mean differential methylation showed that the two liver data sets form a distinct group compared
of at least 0.15 and 0.30 are highlighted. (B) MDS plot of methylation values to the other tissues, suggesting that the age-dependent DNA meth-
of the 8,823 significant age-DMPs. (C) Scatter plot of the first dimension of
ylation signature that we identified is liver-specific (Figure 3A).
the MDS against chronological age of the subjects. The line corresponds to
the LOWESS regression between MDS1 and age. (D) Scatter plot between
A deeper analysis (Figure 3B; Supplementary Files 6 and 7) indi-
chronological age (x-axis) and DNAmAge (DNA methylation age) (y-axis). cates that although in some regions trends in age-association are
Lines correspond to the regression between age and DNAmAge in subjects similar between liver and several tissues (for example, ELOVL2
younger than 60  years of age  and to the LOWESS regression between island or OTUD7A island), other regions harbor a completely liver-
DNAmAge and age, considering the whole cohort. specific age-association (for example, SRCIN1 or MAP6D1 islands).
In particular, we observed that for some regions liver and mesenchy-
mal stem cells display opposite trends in age-associations. In most
Multidimensional scaling (MDS) on the list of 8,823 DMPs
of the cases (MIB2, MACROD1, and FAM18A islands), the same
(age-DMPs) showed a clear separation of the samples along the
CpG probes were positively (liver) and negatively (mesenchymal
first dimension (Figure  1B), linearly correlated with chronological
stem cells) associated with age, whereas in the B3GALT4 island, the
age until 60 years of age, while a trend to a plateau was observed
5′ CpG probes showed positive age-association in liver and no asso-
at older ages (Figure 1C). The same trend was confirmed when we
ciation in mesenchymal stem cells, while the 3′ CpG were negatively
estimated DNA methylation age (DNAmAge) of our samples using
associated with age in mesenchymal stem cells but not age associate
Horvath’s epigenetic clock (17) (Figure 1D), suggesting a slower epi-
in liver.
genetic aging rate of liver after 60 years of age.
As a second type of analysis, we focused on the regions in
which multiple adjacent CpG sites show age-dependent changes Validation of the Liver Aging Signature in Obese
(age-dependent differentially methylated regions, age-DMRs; see Subjects
Materials and Methods section) owing to their more interpret- To assess the ability of our physiological aging signature to capture
able biological meaning compared to DMPs (18). We identified pathological conditions of the liver, we revisited the liver Infinium
a signature of 75 age-DMRs, including 687 probes mapping 450k data sets used by Horvath (GSE61258 and GSE48325, see
on 89 genes, all positively associated with aging (Figure  2A, Materials and Methods section) where the authors demonstrated
Supplementary Files 2 and 3). Unsurprisingly, the top-ranking that human livers from obese subjects have higher DNAmAge val-
DMR mapped in the CpG island of gene ELOVL2, already well ues than livers from nonobese subjects (13). We therefore evaluated
documented and known to undergo age-associated hypermeth- whether the DNA methylation status of our shortlist of age-DMRs
ylation in several tissues (19,20). MDS analysis on the 75 age- was also affected by BMI using the same liver Infinium 450k data
DMRs signature gave results similar to those achieved using the sets used by Horvath (GSE61256 and GSE48325, see Materials
list of age-DMPs, with a clear association of MDS1 with chrono- and Methods). For each CpG probe in the 75 age-DMRs signature,
logical age until 60 years and a leveling off after this threshold we calculated the association with BMI, correcting for age. Eighty
(Supplementary Figure 2). probes, mapping in 25 DMRs, resulted significantly associated with
4 Journals of Gerontology: BIOLOGICAL SCIENCES, 2019, Vol. 74, No. 1

BMI in both data sets (p value <0.01; Supplementary Files 8 and 9;

and Figure 4) and in all cases lean subjects (BMI < 25) tended to have
lower DNA methylation levels, for the same age, compared to obese
subjects. Given that all the 75 DMRs are hypermethylated with age,
this result indicates that the liver aging signature that we identified is
able to detect acceleration effects in the epigenetic age of liver from
obese subjects.

Identification of Age-Associated Changes in Liver

mRNA Expression
To gain insight into the molecular activities altered over liver aging,
we selected a different and more appropriate type of omic technol-

Downloaded from https://academic.oup.com/biomedgerontology/article/74/1/1/4937953 by guest on 29 March 2023

ogy and associated analysis, i.e., transcriptomics and functional
analysis. Pearson correlation between expression signal and age was
computed for each probe along with p values corrected for multi-
ple hypothesis testing (Supplementary File 10), leading to 56 probes
significantly differentially expressed (age-DE; adjusted p value <
0.001), with a high prevalence of probes positively associated with
the age of the donor (47 out 56; Figure 5A). MDS was computed on
the expression values of the selected 56 age-DE probes, returning
well-separated groups along the first component (Figure 5B) which,
differently from DNA methylation data, presented an almost linear
association with chronological age (Figure 5C).
Gene Set Enrichment Analysis (GSEA) was performed on gene
expression data to gain insight into the biological processes involved
in aging liver. We identified several gene sets significantly enriched
at a False Discovery Rate of 0.05 (Supplementary Table  5). Genes
with a positive correlation with age were enriched in allograft rejec-
tion, interferon gamma response, inflammatory response, epithelial-
mesenchymal transition (EMT), and myogenesis, whereas genes with
a negative correlation with age were enriched in metabolic processes.

Multi-omic analysis
As it is now well recognized, the joint analysis of multiple omics
allows the identification of properties of a system that are not vis-
ible when exploring single layers (21), and this has specifically been
shown for the exploration of phenomena as diverse as cellular trans-
differentiation and response to vaccination (21). In this context, we
searched for the genes that displayed age-associated changes in both
DNA methylation (using the list of 8823 age-DMPs) and mRNA
expression (using the list of 56 age-DE genes) and identified 11
genes: FZD2, ICMT, KCNIP4, LGALS4, PTGDS, SDK1, SORCS2,
TDRD10, TSPYL5, VASH1, and ZIC1. For each of these genes, we
calculated Pearson correlation between DNA methylation values of
the probes in the age-DMPs list and gene expression values of the
probes in the list of age-DE (Supplementary File 11). Pearson cor-
relation was significant (nominal p value <0.01) for 3 genes: ZIC1,
TSPYL5, and FZD2. The vast majority of the age-DMPs probes
mapping on gene ZIC1 were positively correlated with the signal
from the three age-DE Affymetrix probes mapping on the same gene.
For gene TSPYL5, a negative correlation between Affymetrix probe
213122_at and both Infinium cg04917181 and cg00032205 probes,
located on a CpG island could be identified; for gene FZD2, a posi-
tive correlation was found between Affymetrix probe 210220_at
Figure  2.  Human liver age-DMRs (differentially methylated regions). (A)
DNA methylation profiles of the CpG islands located in ZIC1 and MACROD1 and Infinium probe cg01684881, located on a CpG island.
genes according to Illumina Infinium 450k measurements. The dotted gray
lines indicate the regions assessed by EpiTYPER. Mean methylation values Discussion
and standard deviations are reported for four age classes (0–30  years old;
31–50 years old; 51–70 years old; 71–90 years old). (B) Pearson correlations
In the present study, we explored the genome-wide epigenomic
between age and DNA methylation, measured by EpiTYPER, in the CpG and transcriptional molecular changes that accompany aging in
islands located in ZIC1 and MACROD1 genes. human liver.
Journals of Gerontology: BIOLOGICAL SCIENCES, 2019, Vol. 74, No. 1 5

Our data highlight for the first time that age-associated epigen- agreement with experimental evidences showing that healthy liver
etic changes level off after 60 years of age in human liver. This trend undergoes a successful aging, preserving liver’s functionality at
is confirmed, although less strikingly, when the DNAmAge is calcu- older ages (1,4,5). The integration of DNA methylation data with
lated using Horvath’s epigenetic clock. This is expected as Horvath’s results from further studies on cell morphology, cellular functions,
clock is optimized to be a multitissue predictor of age, while our
analysis is specifically focused on liver and therefore it is more likely
to detect the specific epigenetic remodeling that occurs in this tissue
during aging. Although quantified for the first time, our finding is in

Downloaded from https://academic.oup.com/biomedgerontology/article/74/1/1/4937953 by guest on 29 March 2023

Figure  4.  Methylation of cg23449696 within the N-Shore of ZIC1 in lean
and overweight/obese subjects. Black and red lines represent the linear
regression between methylation and age in lean and overweight/obese
subjects, respectively.

Figure  3.  Identification of age-DMRs (differentially methylated regions) in

human liver (A) Heatmap of the slope values of the regressions between
DNA methylation of each CpG probe within the 75 age-DMRs and age in Figure  5.  Identification of age-DE (differentially expressed) in human liver.
the analyzed tissues. CpG probes are ordered according to their genomic (A) Pearson correlation between expression and age for the Affymetrix
localization. (B) For the CpG probes within the six regions in ELOVL2, HG-U133_Plus_2 array probes, plotted according to their chromosomal
OTUD7A, SRCIN1, MAP6D1, MIB2, and B3GALT4 genes, the plots report position. (B) Multidimensional scaling (MDS) plot of expression values of
the slope values of the regression between DNA methylation and age in the the 56 significant age-DE. (C) Scatter plot of the first dimension of the MDS
analyzed tissues. For the sake of clarity, the names of the CpG probes are not against chronological age of the subjects. The blue line corresponds to the
reported in the figure, but can be found in Supplementary File 10. LOWESS regression between MDS1 and age.
6 Journals of Gerontology: BIOLOGICAL SCIENCES, 2019, Vol. 74, No. 1

proteomic, and enzymatic profiles will provide a more complete pic- still a controversial issue (33). The role of EMT in kidney fibrosis is
ture of the biological basis of successful aging in liver. increasingly accepted (34), and it is noteworthy that caloric restric-
The decrease in the epigenetic aging rate after 60 years is evident tion alleviates age-related EMT in kidneys from aged mice (35).
using both the list of 8,823 age-DMPs and the shorter list including We noted that a number of genes in the 75 age-DMRs signature is
687 probes mapping on 75 age-DMRs. In addition to the robustness involved in the Wnt pathway (ZIC1, NEFM, FOXD3, MIR155HG,
emerging from confirming the results with two different approaches, CELS3, HEYL) (25,36–39) and in the regulation of the epithelial-
the latter signature is defined using a region-centric approach (18), to-mesenchymal transition (PRDM14) (39). A role of Wnt signaling
more likely to guarantee biologically interpretable results compared in aging has been recently proposed, prompted from the observation
to a single-probe analysis. We demonstrated that the 75 age-DMRs that it is downregulated in cellular senescence (40). Age-dependent
list is: (i) validated in hepatocytes, (ii) liver specific, and (iii) able variations in the expression of Wnt pathway genes could affect tis-
to detect epigenetic age-acceleration effects associated with obesity. sue homeostasis and fibrosis (41). Although the results from our
With respect to the first feature, the signature on hepatocytes was DNA methylation analysis are not conclusive, as they do not allow

Downloaded from https://academic.oup.com/biomedgerontology/article/74/1/1/4937953 by guest on 29 March 2023

confirmed on six regions tested by gene-targeted analysis of DNA to clearly infer an activation/reduction of Wnt signaling in the aging
methylation in a collection of hepatocytes isolated from subjects at of human liver, they sustain previous evidences and, to the best of
different ages. our knowledge, for the first time suggest an epigenetic regulation
The second feature was assessed by testing and confirming the of this pathway in aging. Further experiments should investigate
75 age-DMRs signature in an independent liver data set including EMT-related changes in expression and epigenetic profiles during
tissues from nonobese subjects (GSE61256). In the same analysis, we liver aging.
demonstrated that the signature is liver-specific compared to other The integration of the two levels of information (DNA meth-
tissues, suggesting that the peculiar aging process characteristic of ylation and expression) highlighted three genes sharing differential
this tissue has a specific epigenetic landscape. This refines on the methylation and expression: ZIC1, TSPYL5, and FZD2.
work of Steve Horvath who has shown how methylation changes are ZIC1, which is also part of the 75 age-DMRs list and which is
a ubiquitous phenomenon in human tissues (17), with however lim- one of the genes affected by obesity, encodes for a transcription fac-
ited emphasis on the tissue specificity of epigenetic remodeling dur- tor belonging to the family of the C2H2-type zinc finger proteins.
ing aging. Bysani and colleagues demonstrated that a large fraction ZIC1 plays an important role during development, in particular dur-
of age-associated DNA methylation changes in liver were reflected ing neurogenesis (42), and has been implicated in liver regeneration
also in blood, while the overlap with age-associated changes in (43). ZIC1 has been shown to be hypermethylation with aging in
pancreatic islets and adipose tissue was smaller (8). Our work com- peripheral blood mononucleated cells and naïve CD4+ cells (44). It
plements these findings, as we were able to identify a liver-specific is interesting to note that in a recent report Slieker and colleagues
epigenetic signature of aging, in line with the apparently lower aging demonstrated that ZIC1 methylation encounters an increase in inter-
rate of liver compared to other body districts. individual variability with aging in whole blood, and that this is asso-
We then evaluated the effect of BMI on our epigenetic signature ciated to a higher variability in the expression of the gene (45). On
of aging liver. Eighty probes, mapping in 25 age-DMRs analyzed, the basis of this results and on our observation that ZIC1 is hyper-
were hypermethylated, for the same age, in overweight or obese sub- methylated in liver tissues from obese subjects, it would be interest-
jects compared to lean subjects in two independent data sets. This ing to investigate whether ZIC1 methylation and/or expression are
result indicates that, similarly to Horvath’s epigenetic clock, specific associated with health status and, more in general, with aging quality.
components of our signature are able to capture the already known TSPYL5 was encodes for the testis-specific Y-like protein 5 and
age-acceleration effects in liver from obese subjects. Furthermore, is frequently hypermethylated and silenced in different cancer types,
literature search indicates that several of the genes we identified including hepatocellular carcinoma (HCC) (46,47). Furthermore,
are hypermethylated in hepatocellular carcinoma (ZIC1, FOXD3) its methylation is increased in fetal cortex of Down Syndrome, a
(22,23) or in other cancer types (NEFM, KCNG3, PRDM14, disease which shows a profound remodeling of DNA methylation
FOXD3, CELSR3, HEYL) (24–30), suggesting a role of these genes patterns (19,48).
in liver homeostasis. The protein encoded by FZD2, Frizzled2, is a Wnt recep-
Finally, we investigated gene expression changes in our cohort tor overexpressed in tumors and correlates with markers of EMT
and correlated them with DNA methylation variations. Differently (49). Down-regulation of FZD2 expression through a short-hairpin
from the epigenetic signature of liver aging, we observed a linear RNA suppresses the proliferation of HCC and gastric cancer cells
change in gene expression from young to old subjects. This confirms (50). Although also in this case, it is not easy to infer the role of
the different nature and informativity of omic screens, and empha- DNA methylation and expression changes of FZD2 during aging,
sizes the necessity to integrate them to achieve higher levels of under- it is intriguing that the gene is involved in Wnt-signaling pathway
standing of the underlying molecular events. In particular, although and EMT.
transcriptomics cannot capture the aging deceleration typical of A possible limitation of our study is that the liver biopsies were
liver, it allows to explore functional alterations thanks to the ample collected from brain-dead, heart-beating donors (death by accidents
and well-validated number of tools designed for enrichment analysis. and exclusion of major chronic disease). We cannot exclude that
Accordingly, GSEA highlighted several biological processes associ- intensive care unit stay could have had an impact on liver transcrip-
ated to liver aging. Genes up-regulated with aging are enriched in tion, while methylation patterns are likely to be more stable. This
inflammatory response and in EMT. Although alterations in immune observation could also explain, at least in part, the poor correlation
and inflammatory responses have been described also in murine between transcription and methylation that we observed. In any case,
models of aging liver (31) and are in accordance with the accrual of the use of livers considered suitable for organ transplantation from
chronic inflammation during aging (inflammaging) (32), a possible brain-dead, heart-beating donors, allowed us to investigate a unique
role of EMT in liver aging has not been investigated so far, probably collections of livers from donors of different ages, ranging from very
because the contribution of hepatocytes to EMT in liver fibrosis is young to very old subjects that would be otherwise impossible owing
Journals of Gerontology: BIOLOGICAL SCIENCES, 2019, Vol. 74, No. 1 7

to the strict clinical and ethical limitations on biopsies from healthy, 10. Chen BH, Marioni RE, Colicino E, et al. DNA methylation-based meas-
nonobese subjects. ures of biological age: meta-analysis predicting time to death. Aging
In conclusion, in this study, we characterized the epigenomic (Albany NY). 2016;8: 1844–1865. doi:10.18632/aging.101020
11. Horvath S, Pirazzini C, Bacalini MG, et  al. Decreased epigenetic age of
and transcriptomic remodeling that occurs in healthy liver tissues
PBMCs from Italian semi-supercentenarians and their offspring. Aging
during aging. Epigenomic data indicate a leveling off in the epi-
(Albany NY). 2015;7: 1159–1170. doi:10.18632/aging.100861
genetic aging rate of liver after 60  years, supporting the clinical
12. Gramignoli R, Green ML, Tahan V, et al. Development and application of
and molecular observations that liver can preserve its functionality purified tissue dissociation enzyme mixtures for human hepatocyte isolation.
also at older ages. Functional studies are needed to characterize the Cell Transplant. 2012;21: 1245–1260. doi:10.3727/096368911X600939
contribution of the genes that we identified in DNA methylation 13. Horvath S, Erhart W, Brosch M, et al. Obesity accelerates epigenetic aging
and expression analysis to verify whether the observed changes of human liver. Proc Natl Acad Sci U S A. 2014;111:15538–15543. doi:
are associated with the balance between age-related decline and 10.1073/pnas.1412759111
maintenance of functionality. We suggest that in particular the reg- 14. Ahrens M, Ammerpohl O, von Schönfels W, et al. DNA methylation ana-

Downloaded from https://academic.oup.com/biomedgerontology/article/74/1/1/4937953 by guest on 29 March 2023

ulation of EMT and Wnt-signaling pathways can play a yet unap- lysis in nonalcoholic fatty liver disease suggests distinct disease-specific
and remodeling signatures after bariatric surgery. Cell Metab. 2013;18:
preciated role in liver aging.
296–302. doi:10.1016/j.cmet.2013.07.004
15. Suchiman HE, Slieker RC, Kremer D, Slagboom PE, Heijmans BT, Tobi
EW. Design, measurement and processing of region-specific DNA methy-
Supplementary Material
lation assays: the mass spectrometry-based method EpiTYPER. Front
Supplementary data is available at The Journals of Gerontology, Genet. 2015;6: 287. doi:10.3389/fgene.2015.00287
Series A: Biological Sciences and Medical Sciences online. 16. Day K, Waite LL, Thalacker-Mercer A, et al. Differential DNA methyla-
tion with age displays both common and dynamic features across human
tissues that are influenced by CpG landscape. Genome Biol. 2013;14:
Funding R102. doi:10.1186/gb-2013-14-9-r102
17. Horvath S. DNA methylation age of human tissues and cell types. Genome
This work was supported by Ministero dell’Istruzione, dell’Università e della
Biol. 2013;14: R115. doi:10.1186/gb-2013-14-10-r115
Ricerca (MIUR) (PRIN2008) to GLG; by the Seventh Framework Programme
18. Bacalini MG, Boattini A, Gentilini D, et al. A meta-analysis on age-asso-
to CF (grant number 602757, HUMAN); by the Horizon 2020 Framework
ciated changes in blood DNA methylation: results from an original ana-
Programme to CF and PG (grant number 634821, PROPAG-AGING); by
lysis pipeline for Infinium 450k data. Aging (Albany NY). 2015;7: 97–109.
JPco-fuND to CF (ADAGE). This project has received funding from the
doi:10.18632/aging.100718
H2020 Marie Skłodowska-Curie Actions, grant agreement No 675003. http://
19. Bacalini MG, Deelen J, Pirazzini C, et  al. Systemic age-associated DNA
www.birmingham.ac.uk/panini.
hypermethylation of ELOVL2 gene: in vivo and in vitro evidences of a cell
replication process. J Gerontol A Biol Sci Med Sci. 2017;72: 1015–1023.
doi:10.1093/gerona/glw185
Conflict of interest
20. Giuliani C, Cilli E, Bacalini MG, et al. Inferring chronological age from
None reported. DNA methylation patterns of human teeth. Am J Phys Anthropol.
2016;159: 585–595. doi:10.1002/ajpa.22921
21. Fronza R, Tramonti M, Atchley WR, Nardini C. Joint analysis of tran-
References scriptional and post- transcriptional brain tumor data: searching for emer-
1. Anantharaju A, Feller A, Chedid A. Aging liver. A  review. Gerontology. gent properties of cellular systems. BMC Bioinformatics. 2011;12: 86.
2002;48: 343–353. doi:10.1159/000065506 doi:10.1186/1471-2105-12-86
2. Cescon M, Grazi GL, Cucchetti A, et al. Improving the outcome of liver 22. He G, Hu S, Zhang D, et al. Hypermethylation of FOXD3 suppresses cell
transplantation with very old donors with updated selection and manage- proliferation, invasion and metastasis in hepatocellular carcinoma. Exp
ment criteria. Liver Transpl. 2008;14: 672–679. doi:10.1002/lt.21433 Mol Pathol. 2015;99: 374–382. doi:10.1016/j.yexmp.2015.06.017
3. Gastaca M, Guerra M, Alvarez Martinez L, et  al. Octogenarian donors 23. Wang YY, Jiang JX, Ma H, et  al. Role of ZIC1 methylation in hepato-
in liver transplantation. Transplant Proc. 2016;48: 2856–2858. cellular carcinoma and its clinical significance. Tumour Biol. 2014;35:
doi:10.1016/j.transproceed.2016.06.063 7429–7433. doi:10.1007/s13277-014-1971-4
4. Capri M, Olivieri F, Lanzarini C, et al. Identification of miR-31-5p, miR- 24. Li J, Xing X, Li D, et al. Whole-genome DNA methylation profiling identi-
141-3p, miR-200c-3p, and GLT1 as human liver aging markers sensitive fies epigenetic signatures of uterine carcinosarcoma. Neoplasia. 2017;19:
to donor-recipient age-mismatch in transplants. Aging Cell. 2017;16: 100–111. doi:10.1016/j.neo.2016.12.009
262–272. doi:10.1111/acel.12549 25. Ashktorab H, Shakoori A, Zarnogi S, et al. Reduced representation bisulfite
5. Bellavista E, Martucci M, Vasuri F, et  al. Lifelong maintenance of com- sequencing determination of distinctive DNA hypermethylated genes in
position, function and cellular/subcellular distribution of proteasomes the progression to colon cancer in African Americans. Gastroenterol Res
in human liver. Mech Ageing Dev. 2014;141–142: 26–34. doi:10.1016/ Pract. 2016;2016: 2102674. doi:10.1155/2016/2102674
j.mad.2014.09.003 26. Calmon MF, Jeschke J, Zhang W, et al. Epigenetic silencing of neurofila-
6. Huse SM, Gruppuso PA, Boekelheide K, Sanders JA. Patterns of gene ment genes promotes an aggressive phenotype in breast cancer. Epigenetics.
expression and DNA methylation in human fetal and adult liver. BMC 2015;10: 622–632. doi:10.1080/15592294.2015.1050173
Genomics. 2015;16: 981. doi:10.1186/s12864-015-2066-3 27. Rauch TA, Wang Z, Wu X, Kernstine KH, Riggs AD, Pfeifer GP. DNA
7. Bonder MJ, Kasela S, Kals M, et al. Genetic and epigenetic regulation of methylation biomarkers for lung cancer. Tumour Biol. 2012;33: 287–296.
gene expression in fetal and adult human livers. BMC Genomics. 2014;15: doi:10.1007/s13277-011-0282-2
860. doi:10.1186/1471-2164-15-860 28. Li K, Guo Q, Yang J, et al. FOXD3 is a tumor suppressor of colon can-
8. Bysani M, Perfilyev A, de Mello VD, et al. Epigenetic alterations in blood cer by inhibiting EGFR-Ras-Raf-MEK-ERK signal pathway. Oncotarget.
mirror age-associated DNA methylation and gene expression changes in 2017;8: 5048–5056. doi:10.18632/oncotarget.13790
human liver. Epigenomics. 2017;9: 105–122. doi:10.2217/epi-2016-0087 29. Khor GH, Froemming GR, Zain RB, Abraham TM, Lin TK. Involvement
9. Bacalini MG, D’Aquila P, Marasco E, et  al. The methylation of nuclear of CELSR3 hypermethylation in primary oral squamous cell car-
and mitochondrial DNA in ageing phenotypes and longevity. Mech Ageing cinoma. Asian Pac J Cancer Prev. 2016;17: 219–223. doi:10.7314/
Dev. 2017;165(Pt B): 156–161. doi:10.1016/j.mad.2017.01.006 APJCP.2016.17.1.219
8 Journals of Gerontology: BIOLOGICAL SCIENCES, 2019, Vol. 74, No. 1

30. Kuo KK, Jian SF, Li YJ, et  al. Epigenetic inactivation of transforming 41. Hofmann JW, McBryan T, Adams PD, Sedivy JM. The effects of aging
growth factor-β1 target gene HEYL, a novel tumor suppressor, is involved on the expression of Wnt pathway genes in mouse tissues. Age (Dordr).
in the P53-induced apoptotic pathway in hepatocellular carcinoma. 2014;36: 9618. doi:10.1007/s11357-014-9618-3
Hepatol Res. 2015;45: 782–793. doi:10.1111/hepr.12414 42. Aruga J. The role of Zic genes in neural development. Mol Cell Neurosci.
31. White RR, Milholland B, MacRae SL, Lin M, Zheng D, Vijg J.
2004;26: 205–221. doi:10.1016/j.mcn.2004.01.004
Comprehensive transcriptional landscape of aging mouse liver. BMC 43. Jochheim-Richter A, Rüdrich U, Koczan D, et  al. Gene expression ana-
Genomics. 2015;16: 899. doi:10.1186/s12864-015-2061-8 lysis identifies novel genes participating in early murine liver develop-
32. Sheedfar F, Di Biase S, Koonen D, Vinciguerra M. Liver diseases and aging: ment and adult liver regeneration. Differentiation. 2006;74: 167–173.
friends or foes? Aging Cell. 2013;12: 950–954. doi:10.1111/acel.12128 doi:10.1111/j.1432-0436.2006.00066.x
33. Ikegami T, Zhang Y, Matsuzaki Y. Liver fibrosis: possible involvement of 44. Steegenga WT, Boekschoten MV, Lute C, et al. Genome-wide age-related
EMT. Cells Tissues Organs. 2007;185: 213–221. doi:10.1159/000101322 changes in DNA methylation and gene expression in human PBMCs. Age
34. Liu Y. New insights into epithelial-mesenchymal transition in kidney fibrosis. (Dordr). 2014;36: 9648. doi:10.1007/s11357-014-9648-x
J Am Soc Nephrol JASN. 2010;21: 212–222. doi:10.1681/ASN.2008121226 45. Slieker RC, van Iterson M, Luijk R, et al.; BIOS consortium. Age-related

Downloaded from https://academic.oup.com/biomedgerontology/article/74/1/1/4937953 by guest on 29 March 2023

35. Dong D, Cai G-Y, Ning Y-C, et al. Alleviation of senescence and epithelial- accrual of methylomic variability is linked to fundamental ageing mecha-
mesenchymal transition in aging kidney by short-term caloric restriction nisms. Genome Biol. 2016;17: 191. doi:10.1186/s13059-016-1053-6
and caloric restriction mimetics via modulation of AMPK/mTOR signal- 46. Qiu X, Hu B, Huang Y, Deng Y, Wang X, Zheng F. Hypermethylation of
ing. Oncotarget. 2017;8: 16109–16121. doi:10.18632/oncotarget.14884 ACP1, BMP4, and TSPYL5 in hepatocellular carcinoma and their poten-
36. Merzdorf CS, Sive HL. The zic1 gene is an activator of Wnt signaling. Int tial clinical significance. Dig Dis Sci. 2016;61: 149–157. doi:10.1007/
J Dev Biol. 2006;50: 611–617. doi:10.1387/ijdb.052110cm s10620-015-3878-3
37. Wang W, Jossin Y, Chai G, Lien W-H, Tissir F, Goffinet AM. Feedback regu- 47. Shen J, LeFave C, Sirosh I, Siegel AB, Tycko B, Santella RM. Integrative
lation of apical progenitor fate by immature neurons through Wnt7-Celsr3- epigenomic and genomic filtering for methylation markers in hepato-
Fzd3 signalling. Nat Commun. 2016;7: 10936. doi:10.1038/ncomms10936 cellular carcinomas. BMC Med Genomics. 2015;8: 28. doi:10.1186/
38. Li C, Song G, Zhang S, Wang E, Cui Z. Wnt3a increases the meta-
s12920-015-0105-1
static potential of non-small cell lung cancer cells in vitro in part via its 48. Lu J, Mccarter M, Lian G, et al. Global hypermethylation in fetal cortex
upregulation of Notch3. Oncol Rep. 2015;33: 1207–1214. doi:10.3892/ of Down syndrome due to DNMT3L overexpression. Hum Mol Genet.
or.2014.3700 2016;25: 1714–1727. doi:10.1093/hmg/ddw043
39. Chou YS, Yang MH. Epithelial-mesenchymal transition-related factors in 49. Gujral TS, Chan M, Peshkin L, Sorger PK, Kirschner MW, MacBeath
solid tumor and hematological malignancy. J Chin Med Assoc. 2015;78: G. A noncanonical Frizzled2 pathway regulates epithelial-mesenchy-
438–445. doi:10.1016/j.jcma.2015.05.002 mal transition and metastasis. Cell. 2014;159: 844–856. doi:10.1016/
40. Ye X, Zerlanko B, Kennedy A, Banumathy G, Zhang R, Adams PD.
j.cell.2014.10.032
Downregulation of Wnt signaling is a trigger for formation of facultative 50. Tomizawa M, Shinozaki F, Motoyoshi Y, Sugiyama T, Yamamoto S, Ishige
heterochromatin and onset of cell senescence in primary human cells. Mol N. Gastric cancer cell proliferation is suppressed by frizzled-2 short hair-
Cell. 2007;27: 183–196. doi:10.1016/j.molcel.2007.05.034 pin RNA. Int J Oncol. 2015;46: 1018–1024. doi:10.3892/ijo.2015.2830