Aging

Aging Cell
Elevated Interferon Gamma Signaling Contributes to

Impaired Regeneration in the Aged Liver
Fo
Journal: Aging Cell
Manuscript ID: Draft

r
Manuscript Type: Original Paper

Pe
Date Submitted by the

n/a
Author:
Complete List of Authors: Singh, Pallavi; Columbia University, Department of Dermatology

Darlington, Gretchen; Baylor College of Medicine, Huffington Center
er
on Aging
Awad, Samir; Baylor College of Medicine, Surgery Division
Goode, Catriona; Baylor College of Medicine, Huffington Center on
Aging
Re
Dean, Adam; Baylor College of Medicine, Huffington Center on

Aging
Aging, inflammation, Liver Regeneration, Interferon gamma, Cell

Keywords:
Cycle
vi
ew
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901 Support: (434) 964 4100
Page 1 of 51 Aging Cell
Page |1
“Elevated Interferon gamma signaling contributes to impaired

Foregeneration in the aged liver. “
Pallavi Singh 1,2, Catriona Goode2, Adam Dean2, Samir S. Awad3, Gretchen
r
Darlington2
Pe
1
Columbia University, College of Physicians and Surgeons, New York, NY,
2
Huffington Center on Aging and 3Department of Surgery, Baylor College of
er
Medicine, Houston, TX,

Re
vi
ew
Aging Cell Page 2 of 51
Page |2
Corresponding Author: Pallavi Singh Ph.D.

Columbia University Medical Center
Department of Dermatology
Russ Berrie Medical Pavilion, Room 307
1150 St. Nicholas Avenue
New York, NY 10032, USA
ps2589@columbia.edu
Ph: 212-851-4855; Fax: 212-927-9704
Key Words: Aging, Inflammation, Gamma Interferon, Liver Regeneration, Partial

Hepatectomy
Fo
Word Count: 4436

r Pe
er
Re
vi
ew
Page |3
Summary
Our earlier study on immune related changes in the aged liver described immune
cell infiltration and elevation of inflammation with age. The levels of inflammatory
cytokine -IFN-γ, known to cause inhibition of cell cycle progression, were
elevated in the aging liver. Studies have shown that IFN-γ has a detrimental
effect on the regenerative capacity of the liver. In our current study, we
determine the effect of prevailing inflammatory conditions, including higher IFN-γ
levels, on the delayed regeneration observed in the aged livers after resection.
Interferon signaling is elevated in the aged hepatocytes and an even greater
increase in the levels of interferon related genes is observed after partial
hepatectomy. We demonstrate that the deletion of the major IFN-γ producing
immune cells– the macrophages and the natural killer cells, from the aged liver
Fo
prior to partial hepatectomy leads to restoration of the magnitude of BrdU
incorporation during DNA synthesis in old animals. Removal of these cells is
accompanied with a reduction in the IFN-γ levels. Eighteen month old IFN-γ -/-
mice were subjected to partial hepatectomy to determine the effects of deletion of
r
IFN-γ from the old livers. Despite increased frailty and high mortality after PH,
Pe
aged IFN-γ -/- mice exhibited an earlier entry into the cell cycle compared to age
matched wild type mice. Thus, our study strongly suggests that an age related
elevation in inflammatory conditions in the liver often dubbed as “inflammaging”
has a detrimental effect on the regenerative response of the liver.
er
Re
vi
ew
Page |4
1 Introduction
A decline in the tissue regeneration potential with age is a common theme seen
across several organ systems, including liver (Bucher et al. 1964). In young rats,
virtually all hepatocytes enters the cell cycle after 70% liver resection whereas
only one-third do so in the aged liver (Stocker & Heine 1971). Delayed entry into
the S-phase is also observed and the BrdU incorporation peak is shifted to later
time points. Consistent with the delay in DNA synthesis, aged rats have lower
activities of nucleotide metabolism enzymes - thymidylate synthetase and
thymidine kinase (Tsukamoto et al. 1993) (Beyer et al. 1991) post partial
hepatectomy (PH). Similarly, after liver transplant in humans, the allografts from
aged donors (> 50 yrs) display a lower liver growth-volume compared to the
Fo
younger donors (< 30yrs) (Ikegami et al. 2000).
Several studies have shown an upregulation of the inflammatory response with

age across various tissues, suggesting an age related increase in the
r
proinflammatory status (Bruunsgaard & Pedersen 2003); (Franceschi et al.

Pe
2000). Differential gene-expression profiling of young and aged livers using

microarrays reveals an age related upregulation of genes belonging to the
immune response category with 40% of the genes classified as inflammation
related (Cao et al. 2001),(Amador-Noguez et al. 2004). Elevated expression of
er
acute phase genes which are induced in liver during systemic inflammation along
with the presence of microgranulomas, is also observed in the aging hepatic
tissue (Haines et al. 2001). The total number of lymphocytes in the liver triples
Re
from the age of 7 weeks to 80 in mice (Mehal et al. 2001). Interestingly, chronic
inflammation induced by Concavalin A treatment or due to steatohepatitis causes
a marked inhibition of liver growth after partial hepatectomy (Hines et al. 2007),
(Vetelainen et al. 2007). Conversely, immunosuppressive drugs such as FK506
vi
and cyclosporine augment the regenerative response of the liver after resection
surgery. Hepatocytes, cultured in serum drawn from immunosuppressed mice
ew
display enhanced proliferation, indicating a detrimental effect of inflammatory

factors on the proliferative response (Francavilla et al. 1991). Thus, a negative
correlation seems to exist between chronic hepatic inflammation and the
regenerative capacity of the liver.
Our previous study described an elevation in the transcript levels of various

growth inhibitory inflammatory cytokines such as TGF-β, IFN-γ and IL-1 in the
hepatic tissue (Singh et al. 2008). Another study describes an increase in IFN-γ
production by splenic cells with advancing age (Takayama et al. 2000). IFN-γ is
known to be a potent inhibitor of the cell cycle and treated hepatocytic cell lines
exhibit growth arrest. IFN-γ inhibits the cell cycle through activation of cyclin-cdk
complex inhibitor, p21. It binds to its receptor on the cell surface and causes the
phosphorylation of the STAT1 protein which further activates the interferon
specific transcription factor IRF1 (Tura et al. 2001). In fact, the IFN-γ induced PH
repression is abolished in Stat1-/- and Irf1-/- mice (Sun et al. 2006). IRF1 causes
Page |5
nuclear accumulation of p53 (Kano et al. 1997; Kano et al. 1999) and increased
nitric oxide production via iNOS induction (Brooling et al. 2005; Lee et al. 2007)
which subsequently increase p21 transcription (Hofseth et al. 2003). Induced
elevation of IFN-γ levels artificially by dI:dC leads to impaired liver regeneration.
Gamma interferon knockout mice, show enhanced regeneration as compared to
controls (Sun & Gao 2004). Cellular components of the innate immunity – natural
killer cells and macrophages are known to be robust producers of IFN-γ. An
increase in the number of these cell types in the aged livers, as shown by our
previous study could augment IFN-γ levels both in the organ and in circulation.
We thus hypothesize that the increased presence of IFN-γ in the old livers can
potentially lead to inhibition of the cell cycle and impaired regeneration after PH.
In our current study, we report elevated IFN-γ levels in the aged liver tissue
compared to young. In addition, increased transcript levels of interferon-regulated
Fo
genes are observed in the aged animals both before and after partial
hepatectomy. We also tested whether the depletion of NK cells and
macrophages leads to a decrease in the production of IFN-γ in the aged liver and
r
whether the regeneration process in the liver improves in the absence of these
cell types. Lower levels of IFN-γ were observed in NK cell and macrophage
Pe
depleted animals. BrdU incorporation rates were comparable to young at 38 h,

which is the DNA synthesis peak in young animals. We further tested the effect
of the absence of IFN-γ on liver regeneration in the aged animals by studying the
er
response of 18 mo. old IFN-γ -/- mice to partial hepatectomy. Our study
addresses the impact of inflammatory mediators which have been shown to be
upregulated with age, on the regenerative process in the liver and can help in
Re
understanding their effect on recovery after liver resection surgeries in aged

patients.
vi
ew
Page |6
2 Results
2.1 Elevated levels of IFN-γ are present in old animals along with higher
IFN-γ signaling in hepatic tissue prior to and after partial hepatectomy.
To determine the age-related difference in the message levels of IFN-γ,
quantitative real time PCR (qRT-PCR) was carried out between cDNA obtained
from young (6 mo.) and old (24 mo.) CB6F1 mice livers (n=4). The levels of IFN-γ
mRNA were found to be 13 fold (p <0.05) higher in the old animals. Comparison
of 24 mo. versus 3 mo. old in another inbred strain C57Bl/6, (n=5) also showed
elevated message levels of IFN-γ with a fold change of 2.7 (p-value <0.001).
Fo
Higher IFN-γ levels as measured by multiplex ELISA, were also observed
although due to high variance in the old animals, the differences did not reach
statistical significance (p-value = 0.19) (Figure 1). IFN-γ signals to both
r
hepatocytes and non-parenchymal cells (NPC) which includes the lymphocytic
population of the liver. To dissect and study the effect of IFN-γ on the
Pe
hepatocytes, we carried out separation of the parenchymal fraction enriched in

hepatocytes from the NPC fraction by perfusing the liver followed by differential
centrifugation. qRT-PCR analysis was carried out for IFN-γ regulated genes on
the parenchymal fraction. The hepatocytes derived from livers of old mice
er
showed a general elevation of message levels of these IFN-γ targets compared

to the young mice with irf1, icam1, pml and ifngr1 displaying a statistically
significant difference between young and old (Table 1).
Re
Microarray gene expression dataset comparing the transcriptional profiles of

young and aged livers at 0.5h, 1h, 2h, 4h, 24h, 38h and 48h after resection
surgery were generated previously in our laboratory. To determine whether the
vi
interferon regulated genes are differentially over-expressed in the aged livers

after resection surgery, we selected two sets of genes from published studies
exhibiting a greater than fourfold increase in response to IFN-γ (Der et al. 1998;
ew
Zocco et al. 2006) and compared the differential expression profile between
young and old livers in our array data set at time points 0h, 0.5h, 1h, 2h and 4h
after PH (Figure 2A). The aged livers showed a general upregulation of the
chosen genes compared to young livers with statistically significant higher
expression (sign test; p-value < 0.05) at time points 0.5h, 1h, 2h and 4h in set A
and 0.5h and 2h for set B. In addition to the genes examined in the array dataset,
message levels of interferon regulated genes measured by qRT-PCR showed
elevated expression in the aged livers with significant (p-value < 0.05)
upregulation in message levels of ifitm3 at 0h, ly6e at 1h and irf7, ifi47, isgf3 and
ly6e at 2h after PH (Figure 2B).
We further analyzed the downstream IFN-γ signaling pathway in the regenerating
livers. Analysis of microarray gene profiling data showed that the expression
levels of cell cycle inhibitor p21 were elevated in the aged livers at 24 hrs
accompanied with lower expression levels of cell cycle genes – cyclin D1, cdk4,
E2F1 and PCNA involved in the G1/S transition phase, indicating a lag in cell
Page |7
cycle entry in old animals (Figure. 3) and (Supplementary Table). The expression
levels of Stat1 were significantly elevated at 0.5h and 2h, IRF1 at 1h and iNOS at
2h after PH (p-value < 0.05) indicating an increased signaling through the
pathway at the immediate early stage of regeneration (Supplementary Fig 1). To
further identify the genes which were significantly changing between young and
aged animals and to determine their association with various pathways, we
carried out network analysis using the software Ingenuity. (Figure. 3) Network
associated with G1/S cell cycle transition identified several genes which were
down regulated in the aged animals at 24h and 38h. Several of these genes such
as replication protein A3, DNA topoisomerase, chek1, ribonucleotide reductase,
high mobility group B2, replication factor C4 are for DNA synthesis, maintenance
and replication during S phase, explaining the delayed DNA synthesis in aged
animals post PH.
Total protein levels of Stat1 and p-Stat1 were also measured by western blot
Fo
(Figure 4). The levels of total Stat1 were significantly elevated in the aged
animals at time points 0h, 1h and 2h after PH (p<0.05). Phosphorylated Stat1
showed a significant elevation at 0h (p<0.05) indicating increased IFN-γ signaling
r
in the aged animals both prior to and post PH. The activated Stat1 homodimer
translocates to the nucleus and leads to transcriptional activation of interferon
Pe
specific transcription factors called IRFs or interferon regulatory factors (Boehm

et al. 1997). Thus, elevated IFN-γ and downstream regulated gene message
levels in the aged liver indicate an overall increased responsiveness to the IFN-γ
er
signaling pathway.
Re
2.2 Depletion of NK cells and macrophages leads to improved regeneration

in aged liver accompanied with a decrease in IFN-γ levels.
Our previous study showed increased infiltration of the aged liver with immune
cells such as macrophages, NK cells, T-cells, B-cells and neutrophils (Singh et
vi
al. 2008). Resident macrophage (Kupffer cells) and NK (pit cell) population of the
liver plays an important role in regulation of the immune functions of the liver.
ew
Kupffer cells are a part of the reticulo-endothelial system which aid in the
clearance of foreign antigens from the portal circulation whereas NK cells carry
out tumor surveillance functions. Both these cell types are potent inducers of
inflammation and secrete large amounts of IFN-γ when stimulated. To test
whether they play an inhibitory role in the hepatic regeneration process in the old
animals, we depleted the liver of NK cells or macrophages. CB6F1 mice, aged 18
mo. were injected 24 hrs prior to partial hepatectomy with either anti-asialoGM1
(AsGM1) antibody, which depletes the liver of NK cells, or normal rabbit serum
as a control. Depletion of NK cells in the old animals was assessed by measuring
the transcript levels of NK cell specific markers using qRT-PCR. Antibody
administration led to a significant decrease in the expression of NK cell specific
transcript levels of nkr-p1c, nkr-p1a, cd94 and a 2.8 fold decrease in the
message levels of IFN-γ compared to the control animals (Figure 5).
Macrophage depletion was carried out using tail vein injections of gadolinium
chloride (GdCl3). An age matched control group was injected with 0.9% saline.
Page |8
The success of depletion was determined by quantifying the message levels of

macrophage specific markers f4/80, microsialin, siglec and sr-a in the liver before
and after depletion. The transcript levels of macrophage markers were lower in
the liver 24 hrs after GdCl3 injection indicating successful depletion. A 4.8 fold
reduction in the message levels of IFN-γ was also observed after depletion
(Figure 5).
To further test the effect of NK cell depletion on the regeneration in aged livers,
PH was carried out on AsGM1 and normal rabbit serum (controls) injected 18
mo. mice (N=3). The regenerating livers were collected 38 hrs after surgery
which corresponds to the peak of DNA synthesis in untreated young mice. The
NK cell depleted animals showed higher levels of BrdU compared to the control
animals. Similar experiments were carried out on GdCl3 injected and saline
injected animals to study the effect of macrophage depletion on aged liver
Fo
regeneration. The macrophage depleted animals showed higher levels of BrdU

and PCNA compared to the control animals. Interestingly, comparison of the
levels of BrdU incorporation in NK and macrophage depleted aged animals with
r
young mice (6 mo.) showed no significant difference (Figure 5). These data
suggest that both NK cells and macrophages impede liver regeneration in aged
Pe
livers. The concomitant decrease in the levels of IFN-γ with the depletion of
macrophages and NK cells makes this cytokine a strong candidate for the growth
inhibition observed in the aged livers, yet we cannot exclude the possibility that
er
other cytokines/cell-products could also play a similar role in aged animals.
2.3 Liver regeneration in aged IFN-γ -/- mice

Re
IFN-γ signaling, known to arrest growth by induction of cell cycle inhibitors is

elevated both before and after partial hepatectomy in the aged livers. Thus we
further tested whether the absence of IFN-γ in the aged liver has a positive effect
on the regenerative process. Partial hepatectomy was carried out on 18 mo. old
vi
IFN-γ -/- mice and C57BL/6 age matched background controls. In general, the
liver tumor incidence in these mice was higher and the mice had very high
ew
mortality rates post hepatectomy. PCNA is expressed during early G1 and S

phase, acts as the processivity factor for polymerase delta during DNA synthesis
and is commonly used as a histological marker for actively dividing cells. The
percentage of PCNA positive cells in the IFN-γ -/- mice were significantly greater
than those of the wild type animals at time points 32h and 40h whereas at 48hrs
the levels were almost similar suggesting that IFN-γ -/- mice enter the cell cycle
earlier than the control mice. The average BrdU incorporation levels were higher
in IFN-γ -/- animals at early time points 32h and 40h compared to wildtype
controls although the differences were not significant (Figure 6). Hence the
absence of IFN-γ seems to have a positive effect on the cell cycle kinetics of the
aged liver.
Page |9
3 Discussion
Several lines of evidence suggest that innate immunity plays an important role in
the coordination of the regeneration process. As soon as the liver undergoes
resection, the flux of lipopolysaccharide (LPS) or the bacterial antigens in the
remnant liver increases, leading to increased activation of toll like receptor (TLR)
pathway, which seems to act as one of the triggers for regeneration. C3H/HeJ
mice defective in response to LPS as well as gram negative bacteria free mice
display a delayed liver regeneration response (Cornell et al. 1990). Toll like
receptors bind the lipopolysaccharide in circulation and through signaling via
MyD88 induce various proinflammatory cytokines by Kupffer cells. The MyD88 -/-
Fo
mice exhibit a marked delay in the expression of immediate early genes as well
as a delayed regeneration response post PH (Seki et al. 2005). LPS activation of
hepatic non-parenchymal cell population leads to an increased secretion of TNF-
α and IL-6 which in turn, induce the acute phase response in the liver and
r
activation of immune cells. During liver regeneration these cytokines signal

Pe
through their respective receptors on the hepatocytes and prepare them for
stimulation by growth factors; an event known as priming. The mice deficient in
IL-6 (IL-6 -/-) (Cressman et al. 1996) and TNF-α receptor deficient mice both
show delayed regeneration response and increased necrosis. Components of the
er
complement cascade which help in antigen clearance and mediation of

inflammation are also essential for repair and regeneration. Mice deficient in
complement components C3 and/or C5 exhibit parenchymal damage and
Re
impaired regeneration (Strey et al. 2003). Microarray study on the post PH tissue
from rats has yielded close to 85 genes whose levels change during liver
regeneration and which are classified functioning in innate immunity pathway,
suggesting its significant role in liver regeneration (Chen et al. 2006).
vi
Although innate immunity is an integral part of regeneration, the presence of

chronic inflammatory conditions, is detrimental to the regenerative process.
ew
Conditions of liver inflammation such as artificially induced hepatitis by

administration of concavalin A (Hines et al. 2007) and inflammation associated
with severe steatosis (Vetelainen et al. 2007) both show a decrease in the
proliferation and reduced levels of cell cycle genes in the mice. Inflammatory
conditions in liver are associated with increased levels of proinflammatory
cytokines such as IL-6, TNF-α, IFN-γ and TGF-β. Although IL-6 and TNF-α have
been shown in many studies to play a central role in the initiation of the priming
events, preexisting elevated levels of both IL-6 and TNF-α can be detrimental to
regeneration. Mice hyperstimulated by injection of IL-6 exhibit a strong activation
of stat3 accompanied with inhibition of hepatocyte proliferation as measured by
BrdU staining and Cyclin A and E expression (Wustefeld et al. 2000). In ob/ob
mice which exhibit chronically elevated IL-6 and p-stat3 levels, there is an
elevated accumulation of p-stat3 in the nucleus after liver resection leading to
induction of the cell cycle inhibitor p21 and cell cycle arrest (Torbenson et al.
2002). Similarly in a model of over-expression of TNF-α (TIMP3 -/-), a heightened
P a g e | 10
hepatic lymphocyte infiltration, necrosis and inhibition of regeneration was

observed after PH(Mohammed et al. 2004). The stop signal to cease proliferation
is initiated in part by the proinflammatory cytokine TGF-β which is a potent
inhibitor of hepatocyte growth. Deletion of TGF-β from liver tissue using
neutralizing antibodies or genetic knockout also leads to a better regenerative
response post PH (Enami et al. 2001), (Romero-Gallo et al. 2005). Thus chronic
inflammation seems to have an adverse effect on the regenerative process
despite an important role of innate immunity in the initiation of regeneration.
Macrophages are known to be central players in the process of “inflammaging”
which involves slow ramping up of inflammation with advancing age (Franceschi
et al. 2000). Kupffer cells form cell clusters in the parenchyma compared to the
younger liver, reminiscent of the inflammatory granulomatous disease (Singh et
al. 2008) and exhibit elevated production of proinflammatory cytokines IL-6 and
TNF-α. A 70 % knock down of macrophage numbers using gadolinium chloride
Fo
leads to improvement of regeneration due to reduction of production IL-10 which

inhibits TNF-α (Rai et al. 1997). Numbers as well as activity of NK and NKT cells
are also elevated with age in the liver (Abo et al. 2000). These cells have been
r
implicated in hepatic injury during liver regeneration along with increased

production of IFN-γ (Ito et al. 2003; Huang et al. 2006). Administration of poly
Pe
dI:dC which elicits IFN-γ production or direct IFN-γ treatment causes repression
of proliferation and increases the expression of several anti-proliferative proteins,
including STAT1, IRF-1, and p21 in mice livers after partial hepatectomy (Sun &
er
Gao 2004). The impairment of liver regeneration in poly dI:dC treated animals is
abolished in stat1-/- and IRF-/- mice implicating their role in downstream
signaling (Sun et al. 2006). We tested the effect of absence of IFN-γ on liver
Re
regeneration in the aged animals. PH experiments carried out on aged IFN-γ -/-
mice showed higher levels of the proliferative marker PCNA earlier in the cell
cycle compared to wild type controls. The levels of BrdU staining, although not
statistically significant, were also higher in the IFN-γ -/- mice at 32 hours after PH
vi
in suggesting an overall earlier entry into the cell cycle compared to WT mice.
Liver resection is commonly and successfully used for the treatment of malignant
ew
liver tumors with high long term survival rates. In elderly patients the mortality
rates associated with liver resection are higher (11%) compared to a younger
population (2-4%) and in cases of extensive resection these rates increase to
30%. Interestingly, when the elderly patients suffering with cirrhosis are excluded
from the dataset, the mortality rates drop to levels comparable to the younger
population (Fortner & Lincer 1990) suggesting cirrhotic patients have either
higher inflammation or fewer healthy heptocytes. Chronic liver disease has
been associated with higher risk of liver failure following partial hepatectomy due
to presence of extensive cirrhosis and ongoing inflammatory activity (Farges et
al. 1999; Eguchi et al. 2000). An increasingly common form of liver pathology,
nonalcoholic fatty liver disease and the accompanying steatohepatitis, has been
found to enhance the risk in major liver resections including postoperative liver
failure (Behrns et al. 1998) and can be detrimental to graft function in liver
transplantation. (Lalor et al. 2007) (Soejima et al. 2003). Fulminant hepatic
failure (FHF) is the development of massive acute liver injury with rapid clinical
P a g e | 11
deterioration in a previously healthy patient and in the elderly, the condition is

more acute with greater terminal complications compared to young patients
(Sawabe et al. 2000). Induction of IFN-γ and increased NK cell activity in balb/c
mice using the drug FS-112 causes FHF and an impaired regeneration response
(Ohnishi et al. 1993). In human cases of FHF, the serum levels of IFN-γ were
elevated along with increased numbers of CD8+ IFN-gamma+ T lymphocytes
(Kimura et al. 1999).
Our study indicates a similar deleterious effect of innate immunity and IFN-γ in
aged mice wherein the deletion of the inflammatory mediators elicits a better liver
proliferative response after liver resection. Thus studies addressing the impact of
age related increase in inflammatory conditions specifically IFN-γ on regenerative
process in the liver can help in understanding their effect on recovery after liver
surgeries in elderly patients.
Fo
r Pe
er
Re
vi
ew
P a g e | 12
4 Experimental procedures
4.1 Mice Aged CB6F1 mice were obtained from National Institute of Aging rodent
colony and C57BL/6 and IFN-/- mice from Jackson Laboratories. The animals
were housed in groups of 3–5 animals of the same gender in a room with
controlled photoperiod of 12 h light−12 h darkness (lights on from 07:00 to 19:00
h) and a temperature of 22 ± 2 °C. Animals were given free access to water and
pelleted diet (5010 rodent diet, LabDiet, PMI Nutrition International, Brentwood,
MO, USA). For tissue harvesting, the animals were anaesthetized with
isofluorane followed by cervical dislocation; livers were collected, flash frozen in
liquid nitrogen and stored at −70 °C. All procedures with mice were carried out in
accordance to the provided guidelines for laboratory animal welfare by the
Association of Assessment and Accreditation of Laboratory Animal Care
Fo
(AALAC)-approved Center of Comparative Medicine at Baylor College of

Medicine, Houston, TX, under specific pathogen-free (SPF) conditions in micro-
isolator cages.
r
4.2 Surgery and antibody injections. Mice anesthetized with avertin (1.2%
Pe
avertin in 0.9% saline 240mg/kg body weight) underwent partial hepatectomy

surgery as described (Higgins 1931). Median and lateral hepatic lobes were
separately ligated with monofilament suture (cat# Z421H, Ehticon, Somerville,
er
New Jersey) and excised through a 1-cm longitudinal abdominal incision, which
was then closed in two layers. Excised tissue was flash frozen in liquid nitrogen
as a 0 hour control as well as fixed in 10% formalin for histochemistry. Animals
Re
were given BrdU injections (50 mg/kg body weight) before tissue harvesting at
32h, 38h, 40h and 48hr after partial hepatectomy.
For NK cell depletion experiments, mice were injected intraperitoneally 24 h
before surgery with 100 µl of anti asialo GM1 antibody (catalog # 986-10001
vi
Waco chemicals, Richmond, VA ) in 200 µl PBS whereas the control mice were
given injections of normal rabbit serum (320mg /animal). For macrophage
ew
depletion experiments, mice were injected intravenously 24 h before surgery with

100 µl of 4.5 mg/ml solution of GdCl3 ( catalog # G7532 Sigma) in 200 µl 0.9%
saline whereas the control mice were given injections of 300 µl of 0.9% saline.
4.3 RT-PCR. Total RNA was extracted from frozen livers using the RNeasy
purification kit (Qiagen) in accordance with the manufacturer’s protocol. DNase-
treated total liver RNA was reverse transcribed using SuperScript II reverse
transcriptase (Invitrogen, Carlsbad, CA, USA). Realtime PCR (RTPCR) was
performed using SYBR Green Master Mix and the ABI Prism 7000 Sequence
Detection System (Applied Biosystems, Foster city, CA, USA). γ-tubulin and β-
actin were used to normalize the RNA concentration in samples. Thermal cycling
conditions consisted of an initial step at 95 °C for 10 min to activate the Taq DNA
polymerase and 40 cycles of sequential denaturation at 95 °C for 15 s and
annealing/extension at 60 °C for 60 s. Data analysis was performed using the
ABI Prism 7000 SDS Software (Applied Biosystems). The primers used are listed
P a g e | 13
in the supplementary table 1. The real-time PCR analysis was performed

according to the comparative CT method (Amador-Noguez et al. 2004). The p-
values reported for these changes refer to a two-tailed t-test between the
normalized CT values in old vs. young mice.
4.4 Immunohistochemistry Livers, along with spleen from the same animal,
were fixed in 10% buffered formalin, paraffin embedded and sectioned to a
thickness of (4 µm). The liver tissue sections were stained for BrdU incorporation
using mouse anti mouse BrdU (Catalog # M0744, Dako cytomation) along with
MOM kit (BMK-2202, Vector Laboratories) and counterstained with eosin.
Sections were also stained for PCNA and Histone3 phosphorylation levels using
rabbit anti mouse PCNA antibody (Catalog # sc7907, Santa Cruz) and rabbit anti
pH3 antibody ( Catalog # 06-570, Upstate). Simultaneous staining with spleen
was used as a positive control for proliferation markers. Immunohistochemistry
Fo
imaging was carried out using Zeiss Axioskop 2 plus microscope and image were
done using the AxioVision 3.1.2.1 image analysis software. The cells staining
positive for each of the cell cycle markers were counted manually per field for
r
multiple fields and two-tailed T-test for paired datasets was carried out using
excel software to assess statistical significance.
Pe
4.5 Protein isolation and Western analysis

er
Nuclear, cytoplasmic, or whole cell extracts were isolated according to the

standard lab protocol. Briefly, liver tissue was homogenized using a Teflon
mortar and pestle in buffer containing ‘COMPLETE’ protease inhibitors (Roche),
Re
and 1mM NaF (phosphatase inhibitor). Western analysis of protein isolates was
performed using either 50 µg or 100 µg of protein according to the standard lab
protocol. Briefly, SDS-polyacrylamide gels were transferred to PVDF membrane,
typically incubated overnight with primary antibody at 4oC (shaking), and then
vi
probed with secondary antibody for 1 hour at room temperature. Anti Stat1 (#
M22X, Santa Cruz) and anti pStat1 (# 9171, cell signaling) to detect the stat1
ew
levels in the liver.

4.6 ELISA
Small pieces of frozen liver tissue were homogenized by a dounce homogenizer
in PBS. The tissue was spun at 15000 g for 10 min to pellet the tissue debris.
The supernatant was used for cytokine analysis using the Bio-Plex Mouse
Cytokine 18-Plex Panel (cat # 171-304000, Biorad). The assay kit contained
microbead coupled antibodies against murine cytokines. The Bio-Plex cytokine
assay protocol was followed to carry out single well multiplex against the
cytokines using 50 µl of the sample along with the concentration standards. The
contents of the microplate well were analyzed using the Bio-Plex array reader.
The calculation of analyte concentration, standard deviation, and standard error
were carried out using the Bio-Plex Manager software.
4.7 Microarray Analysis – Supplementary Data
P a g e | 14
Acknowledgements: We thank the Digestive Disease Center, Baylor

College of Medicine, for help with immunohistochemical analysis. We also thank
Dr. Milton Finegold for his expertise in the histological analysis of the liver
samples.Supported by the NIH (PO1 AG20752), the Texas Medical Center
Digestive Disease Center (DK56338, DK53045), and the Ellison Medical
Foundation.
Author Contributions: P. Singh - Main author (animal surgeries, microarray

analysis and data analysis), C. Goode- animal surgeries early time points and
microarray analysis, S. Awad - animal survival surgery techniques, A Dean -
technical support, G. Darlington - Principle Investigator
Fo
r Pe
er
Re
vi
ew
P a g e | 15
References
Abo T, Kawamura T , Watanabe H (2000). Physiological responses of

extrathymic T cells in the liver. Immunol Rev. 174, 135-149.
Amador-Noguez D, Yagi K, Venable S , Darlington G (2004). Gene expression
profile of long-lived Ames dwarf mice and Little mice. Aging Cell. 3, 423-
441.
Behrns KE, Tsiotos GG, DeSouza NF, Krishna MK, Ludwig J , Nagorney DM
(1998). Hepatic steatosis as a potential risk factor for major hepatic
resection. J Gastrointest Surg. 2, 292-298.
Fo
Beyer HS, Sherman R , Zieve L (1991). Aging is associated with reduced liver
regeneration and diminished thymidine kinase mRNA content and enzyme
activity in the rat. J Lab Clin Med. 117, 101-108.
Boehm U, Klamp T, Groot M , Howard JC (1997). Cellular responses to
r
interferon-gamma. Annu Rev Immunol. 15, 749-795.

Pe
Brooling JT, Campbell JS, Mitchell C, Yeoh GC , Fausto N (2005). Differential

regulation of rodent hepatocyte and oval cell proliferation by interferon
gamma. Hepatology. 41, 906-915.
Bruunsgaard H , Pedersen BK (2003). Age-related inflammatory cytokines and
er
disease. Immunol Allergy Clin North Am. 23, 15-39.

Bucher NL, Swaffield MN , Ditroia JF (1964). The Influence of Age Upon the
Incorporation of Thymidine-2-C14 into the DNA of Regenerating Rat Liver.
Re
Cancer Res. 24, 509-512.

Cao SX, Dhahbi JM, Mote PL , Spindler SR (2001). Genomic profiling of short-
and long-term caloric restriction effects in the liver of aging mice. Proc Natl
Acad Sci U S A. 98, 10630-10635.
vi
Chen GW, Zhang MZ, Zhao LF , Xu CS (2006). Expression patterns and action
analysis of genes associated with physiological responses during rat liver
ew
regeneration: Innate immune response. World J Gastroenterol. 12, 7852-

7858.
Cornell RP, Liljequist BL , Bartizal KF (1990). Depressed liver regeneration after
partial hepatectomy of germ-free, athymic and lipopolysaccharide-
resistant mice. Hepatology. 11, 916-922.
Cressman DE, Greenbaum LE, DeAngelis RA, Ciliberto G, Furth EE, Poli V ,
Taub R (1996). Liver failure and defective hepatocyte regeneration in
interleukin-6-deficient mice. Science. 274, 1379-1383.
Der SD, Zhou A, Williams BR , Silverman RH (1998). Identification of genes
differentially regulated by interferon alpha, beta, or gamma using
oligonucleotide arrays. Proc Natl Acad Sci U S A. 95, 15623-15628.
Eguchi H, Umeshita K, Sakon M, Nagano H, Ito Y, Kishimoto SI, Dono K,
Nakamori S, Takeda T, Gotoh M, Wakasa K, Matsuura N , Monden M
(2000). Presence of active hepatitis associated with liver cirrhosis is a risk
P a g e | 16
factor for mortality caused by posthepatectomy liver failure. Dig Dis Sci.
45, 1383-1388.
Enami Y, Kato H, Murakami M, Fujioka T, Aoki T, Niiya T, Murai N, Ohtsuka K ,
Kusano M (2001). Anti-transforming growth factor-beta1 antibody
transiently enhances DNA synthesis during liver regeneration after partial
hepatectomy in rats. J Hepatobiliary Pancreat Surg. 8, 250-258.
Ernst J , Bar-Joseph Z (2006). STEM: a tool for the analysis of short time series
gene expression data. BMC Bioinformatics. 7, 191.
Farges O, Malassagne B, Flejou JF, Balzan S, Sauvanet A , Belghiti J (1999).
Risk of major liver resection in patients with underlying chronic liver
disease: a reappraisal. Ann Surg. 229, 210-215.
Fortner JG , Lincer RM (1990). Hepatic resection in the elderly. Ann Surg. 211,
141-145.
Francavilla A, Starzl TE, Barone M, Zeng QH, Porter KA, Zeevi A, Markus PM,
Fo
van den Brink MR , Todo S (1991). Studies on mechanisms of

augmentation of liver regeneration by cyclosporine and FK 506.
Hepatology. 14, 140-143.
r
Franceschi C, Bonafe M, Valensin S, Olivieri F, De Luca M, Ottaviani E , De

Benedictis G (2000). Inflamm-aging. An evolutionary perspective on
Pe
immunosenescence. Ann N Y Acad Sci. 908, 244-254.

Haines DC, Chattopadhyay S , Ward JM (2001). Pathology of aging B6;129
mice. Toxicol Pathol. 29, 653-661.
er
Higgins GMA, R.M (1931). Experimental pathology of the liver. Arch. pathol. 12,
186-202.
Hines IN, Kremer M, Isayama F, Perry AW, Milton RJ, Black AL, Byrd CL ,
Re
Wheeler MD (2007). Impaired liver regeneration and increased oval cell

numbers following T cell-mediated hepatitis. Hepatology. 46, 229-241.
Hofseth LJ, Saito S, Hussain SP, Espey MG, Miranda KM, Araki Y, Jhappan C,
Higashimoto Y, He P, Linke SP, Quezado MM, Zurer I, Rotter V, Wink DA,
vi
Appella E , Harris CC (2003). Nitric oxide-induced cellular stress and p53

activation in chronic inflammation. Proc Natl Acad Sci U S A. 100, 143-
ew
148.
Huang W, Dong Z, Wei H, Ding C, Sun R , Tian Z (2006). Selective elimination
of hepatic natural killer T cells with concanavalin A improves liver
regeneration in mice. Liver Int. 26, 339-345.
Ikegami T, Nishizaki T, Yanaga K, Shimada M, Kishikawa K, Nomoto K,
Uchiyama H , Sugimachi K (2000). The impact of donor age on living
donor liver transplantation. Transplantation. 70, 1703-1707.
Ito H, Ando K, Nakayama T, Taniguchi M, Ezaki T, Saito K, Takemura M,
Sekikawa K, Imawari M, Seishima M , Moriwaki H (2003). Role of Valpha
14 NKT cells in the development of impaired liver regeneration in vivo.
Hepatology. 38, 1116-1124.
Kano A, Haruyama T, Akaike T , Watanabe Y (1999). IRF-1 is an essential
mediator in IFN-gamma-induced cell cycle arrest and apoptosis of primary
cultured hepatocytes. Biochem Biophys Res Commun. 257, 672-677.
P a g e | 17
Kano A, Watanabe Y, Takeda N, Aizawa S , Akaike T (1997). Analysis of IFN-

gamma-induced cell cycle arrest and cell death in hepatocytes. J
Biochem. 121, 677-683.
Kimura K, Ando K, Tomita E, Ohnishi H, Ishikawa T, Kakumu S, Muto Y ,
Moriwaki H (1999). Elevated intracellular IFN-gamma levels in circulating
CD8+ lymphocytes in patients with fulminant hepatitis. J Hepatol. 31, 579-
583.
Lalor PF, Faint J, Aarbodem Y, Hubscher SG , Adams DH (2007). The role of
cytokines and chemokines in the development of steatohepatitis. Semin
Liver Dis. 27, 173-193.
Lee HJ, Oh YK, Rhee M, Lim JY, Hwang JY, Park YS, Kwon Y, Choi KH, Jo I,
Park SI, Gao B , Kim WH (2007). The role of STAT1/IRF-1 on synergistic
ROS production and loss of mitochondrial transmembrane potential during
hepatic cell death induced by LPS/d-GalN. J Mol Biol. 369, 967-984.
Fo
Mehal WZ, Azzaroli F , Crispe IN (2001). Immunology of the healthy liver: old
questions and new insights. Gastroenterology. 120, 250-260.
Mohammed FF, Smookler DS, Taylor SE, Fingleton B, Kassiri Z, Sanchez OH,
r
English JL, Matrisian LM, Au B, Yeh WC , Khokha R (2004). Abnormal

TNF activity in Timp3-/- mice leads to chronic hepatic inflammation and
Pe
failure of liver regeneration. Nat Genet. 36, 969-977.

Ohnishi H, Muto Y, Maeda T, Hayashi T, Nagaki M, Yamada T, Shimazaki M,
Yamada Y, Sugihara J , Moriwaki H (1993). Natural killer cell may impair
er
liver regeneration in fulminant hepatic failure. Gastroenterol Jpn. 28 Suppl

4, 40-44; discussion 53-46.
Rai RM, Loffreda S, Karp CL, Yang SQ, Lin HZ , Diehl AM (1997). Kupffer cell
Re
depletion abolishes induction of interleukin-10 and permits sustained

overexpression of tumor necrosis factor alpha messenger RNA in the
regenerating rat liver. Hepatology. 25, 889-895.
Romero-Gallo J, Sozmen EG, Chytil A, Russell WE, Whitehead R, Parks WT,
vi
Holdren MS, Her MF, Gautam S, Magnuson M, Moses HL , Grady WM

(2005). Inactivation of TGF-beta signaling in hepatocytes results in an
ew
increased proliferative response after partial hepatectomy. Oncogene. 24,

3028-3041.
Sawabe M, Arai T, Esaki Y, Fukazawa T, Takubo K , Hirokawa K (2000).
Fulminant hepatic failure in the elderly: a clinicopathological study of
autopsy cases aged over 65 years. Pathol Int. 50, 98-105.
Seki E, Tsutsui H, Iimuro Y, Naka T, Son G, Akira S, Kishimoto T, Nakanishi K ,
Fujimoto J (2005). Contribution of Toll-like receptor/myeloid differentiation
factor 88 signaling to murine liver regeneration. Hepatology. 41, 443-450.
Singh P, Coskun ZZ, Goode C, Dean A, Thompson-Snipes L , Darlington G
(2008). Lymphoid neogenesis and immune infiltration in aged liver.
Hepatology. 47, 1680-1690.
Smyth GK (2004). Linear models and empirical bayes methods for assessing
differential expression in microarray experiments. Stat Appl Genet Mol
Biol. 3, Article3.
P a g e | 18
Soejima Y, Shimada M, Suehiro T, Kishikawa K, Yoshizumi T, Hashimoto K,

Minagawa R, Hiroshige S, Terashi T, Ninomiya M, Shiotani S, Harada N ,
Sugimachi K (2003). Use of steatotic graft in living-donor liver
transplantation. Transplantation. 76, 344-348.
Stocker E , Heine WD (1971). Regeneration of liver parenchyma under normal
and pathological conditions. Beitr Pathol. 144, 400-408.
Strey CW, Markiewski M, Mastellos D, Tudoran R, Spruce LA, Greenbaum LE ,
Lambris JD (2003). The proinflammatory mediators C3a and C5a are
essential for liver regeneration. J Exp Med. 198, 913-923.
Sun R , Gao B (2004). Negative regulation of liver regeneration by innate
immunity (natural killer cells/interferon-gamma). Gastroenterology. 127,
1525-1539.
Sun R, Park O, Horiguchi N, Kulkarni S, Jeong WI, Sun HY, Radaeva S , Gao B
(2006). STAT1 contributes to dsRNA inhibition of liver regeneration after
Fo
partial hepatectomy in mice. Hepatology. 44, 955-966.

Takayama E, Seki S, Ohkawa T, Ami K, Habu Y, Yamaguchi T, Tadakuma T ,
Hiraide H (2000). Mouse CD8+ CD122+ T cells with intermediate TCR
r
increasing with age provide a source of early IFN-gamma production. J

Immunol. 164, 5652-5658.
Pe
Torbenson M, Yang SQ, Liu HZ, Huang J, Gage W , Diehl AM (2002). STAT-3
overexpression and p21 up-regulation accompany impaired regeneration
of fatty livers. Am J Pathol. 161, 155-161.
er
Tsukamoto I, Nakata R , Kojo S (1993). Effect of ageing on rat liver regeneration

after partial hepatectomy. Biochem Mol Biol Int. 30, 773-778.
Tura BJ, Bunyan KE , Harrison DJ (2001). The effect of IFNgamma on the
Re
hepatocyte: cell cycle and apoptosis. Int J Exp Pathol. 82, 317-326.
Vetelainen R, van Vliet AK , van Gulik TM (2007). Severe steatosis increases
hepatocellular injury and impairs liver regeneration in a rat model of partial
hepatectomy. Ann Surg. 245, 44-50.
vi
Wustefeld T, Rakemann T, Kubicka S, Manns MP , Trautwein C (2000).

Hyperstimulation with interleukin 6 inhibits cell cycle progression after
ew
hepatectomy in mice. Hepatology. 32, 514-522.

Zocco MA, Carloni E, Pescatori M, Saulnier N, Lupascu A, Nista EC, Novi M,
Candelli M, Cimica V, Mihm S, Gasbarrini G, Ramadori G , Gasbarrini A
(2006). Characterization of gene expression profile in rat Kupffer cells
stimulated with IFN-alpha or IFN-gamma. Dig Liver Dis. 38, 563-577.
P a g e | 19
5 Figures
Figure 1- Aged liver exhibits higher levels of IFN-γ expression in liver

tissue.
Elevated Message levels of IFN-γ in A) old 24 mos. CB6F1 mice compared to
young 6 mos. (n=4); p-value<0.005; Fold Change=13.63 and B) old (24 mo.)
C57BL/6 mice compared to 3 mo. (n=5); p-value < 0.001; (fold change= 2.72)
C) Tissue protein levels of IFN-γ in old (24 mos.) and young (3. mo.) C57B/L6
mice liver (n=4); p-value = 0.19. mRNA levels were normalized to γ-tubulin.
Fo
D and E) Frozen sections from liver tissue of old mice (24 mo.) and young mice
(3 mo.) were paraformaldehyde fixed and stained with DAPI (blue) and FITC
labeled anti-IFN-γ antibody (Green). Hepatocytes can be visualized with DAPI as
r
large round isolated nuclei whereas lymphocytic clusters can be seen as regions
with denser concentration of DAPI staining smaller elongated nuclei.
Pe
(Magnification levels - 10X; inset- 40X).

er
Re
vi
ew
P a g e | 20
Fo
r Pe
er
Re
vi
ew
P a g e | 21
Figure 2- Changes in the expression profile of IFN-γ responsive genes in

young and old animals at early time points after partial hepatectomy.
A) Expression fold change between aged and young livers at time points 0h,
0.5h,1h, 2h and 4h after PH in genes previously shown to exhibit elevated
expression in response to IFN-γ treatment in published studies – Set A (Der et al.
1998); Set B (Zocco et al. 2006). * denotes p < 0.05 (Significance value for
directional change - Sign test)
B) Real-time PCR analysis for interferon-regulated genes - (A) IRF7 - interferon
regulatory factor 7, (B) IFI47- interferon gamma inducible protein 47, (C) Ifitm3 -
interferon induced transmembrane protein 3 (D) ISGF3 - Interferon-stimulated
gene factor 3 and (E) Ly6E - Lymphocyte antigen 6 complex, locus E. mRNA
levels were normalized to β-actin mRNA and are graphed relative to the 0h
Fo
control time-point for young liver. * denotes p < 0.05

r Pe
er
Re
vi
ew
P a g e | 22
Fo
r Pe
er
Re
vi
ew
P a g e | 23
Figure 3- Changes in the G1/S transition between young and aged mice.
A) Figure shows the expression levels of cell cycle genes cyclinD1, cdk4, E2F1
and PCNA at time points 24h, 38h ad 48 after PH in aged and young mice.
(N=3), * denotes p-value < 0.05.
B) The G1/S transition network describes the Old/Young fold change in the
expression levels of the cell cycle genes in the regenerating liver at 24h post PH.
liver. The fold changes are colored on a scale ranging from green denoting
relative down regulation of genes to red indicating a relative upregulation. Solid
lines denote direct physical interactions, while dotted lines represent indirect
regulatory interactions between the genes. The genes in the network exhibit a
fold change regulation of 1.5 times or more and a p-value of interaction < 0.01.
Fo
r Pe
er
Re
vi
ew
P a g e | 24
Fo
r Pe
er
Re
vi
ew
P a g e | 25
Figure 4- shows the protein levels of Stat1 and phospho-Stat1 in the livers of
CB6F1 mice aged 24 mo. (old ) and 6 mo. (young ). Stat1 and pStat1 levels
were both normalized to β-actin. The levels of p-Stat1 were normalized relative to
total Stat1 levels at corresponding time points respectively. The composite charts
show the averaged levels of Stat1 and p-Stat1 for three sets of animals.
* denotes p-value < 0.05.
Fo
r Pe
er
Re
vi
ew
P a g e | 26
Fo
r Pe
er
Re
vi
ew
P a g e | 27
Figure 5- Changes in the levels of cell type specific markers and IFN-γ after
NK cell or macrophage depletion.
Changes in the message levels of immune cell specific genes 24hr after
depletion. A) -The levels of NK specific genes -NkrP1a, NkrP1c and CD94
showed a significant decrease in depleted animals. IFN-γ message levels were
also 2.8 times lower in the depleted animals. B) - Levels of macrophage specific
genes – F4/80, microsialin, siglec and SR-a were significantly reduced. The
levels of IFN-γ were 4.8 times lower. * denotes p < 0.05. Figure C) and D) show
the difference in BrdU incorporation and PCNA staining levels between untreated
- young (6mo) and old mice (24mo); NK cell depleted – AsGM1 treated and NRS
control mice (18mo.) and macrophage depleted – GdCl3 treated and saline
control mice (18 mo.) 38hrs after partial hepatectomy. BrdU was injected 1hr
Fo
prior to tissue collection. (N=3-5), * denotes p < 0.05

r Pe
er
Re
vi
ew
P a g e | 28
Fo
r Pe
er
Re
vi
ew
P a g e | 29
Figure 6- Differences in the distribution of cell markers between IFN-γ -/-

mice and C57BL/6 wild type 18 mo mice at different time points after partial
hepatectomy.
The figure shows the differences in the levels of – A) PCNA – cell cycle marker
for late G1/S phase and B) BrdU incorporation which indicates DNA synthesis
rates. The immunohistochemical analysis was carried out on livers harvested
after 32h, 40h and 48hrs post hepatectomy. BrdU injections were administered
1h prior to sacrifice. * denotes p < 0.05
Fo
r Pe
er
Re
vi
ew
P a g e | 30
Fo
r Pe
er
Re
vi
ew
P a g e | 31
Tables
Changes in the levels of IFN-γ upregulated genes with age in the hepatocyte
enriched fraction of the mouse liver.
Fo
r Pe
er
Re
vi
ew
P a g e | 32
Fold
Change P-
Gene Name Symbol (old/young) value Function
Interferon
inducible Induced by IFN-γ. Involved in pathogenic
GTPase 2 Iigp2 4.68 0.158 response against gram negative bacteria.
Interferon
consensus
sequence Transcription factor which confers resistance to
binding pathogenic infections and aids the
protein icsbp 3.46 0.151 differentiation of myeloid cells
Interferon
Binds to the upstream regulatory elements
regulatory called ICS ( interferon consensus sequence ) of
factor 1 irf1 2.47 0.042 IFN-inducible genes and activates them.
Fo
2',5'-
oligoadenylate Activates RNase L which degrades viral RNA
synthetase 1 Oas1 2.47 0.158 degradation and inhibits viral replication
r
Intercellular
Involved in leukocyte recruitment into the tissue
adhesion and inhibition of IL-4 production by naïve T-
Pe
molecule 1 icam-1 2.35 0.014 cells.
Interferon
Binds to the upstream regulatory region of
regulatory interferon regulated genes and activates their
er
factor 7 irf7 2.33 0.058 transcription

Growth arrest
and DNA-
Binds to PCNA and affects its interaction with
damage-
some CDK complexes; stimulates DNA excision
Re
inducible repair in vitro and inhibits entry of cells into S

protein gadd45 2.28 0.138 phase
Leads to CBP and ELF4 localization to PML

Promyelocytic nuclear bodies and enhances the transcriptional
vi
leukemia PML 2.24 0.036 activity of nuclear receptors.
N-myc (and
Enhances cytokine-mediated STAT
ew
STAT) transcription. Involved in recruitment of co-

interactor nmi 1.63 0.147 activators to transcription factors.
Interferon
gamma Encodes the ligand-binding chain (α) of the IFN-
receptor 1 Ifngr1 1.56 0.024 γ receptor
The table shows fold changes in the expression IFN-γ genes that were increased
in the parenchymal fraction of the liver obtained after liver perfusion. The gene
expression levels were assessed using real time PCR using cDNA from 3 mo.
and 24 mo. old C57BL/6 mice (n=4). The significance levels were determined
using a 2 tailed t-test between the sample groups.
Supplementary Data
Supplementary Figure 1 - The heat map shows the transcript levels of Stat1, Irf1 and
p21 of young and aged livers at different time points after PH. The normalized probe set
intensities for each gene were compared. The significant differences are indicated by
(p-values).
Fo
r Pe
er
Re
vi
ew
Supplementary Figure 2 – BrdU incorporation levels in the young and aged

CB6F1 mice after partial hepatectomy.
BrdU incorporation curve - CB6F1
50 *
40
Fo
% BrdU incorporation
30
r
old
20
Pe
young
10
er
0
24h 38h 48h
Re
-10
Time after PH
vi
ew
Figure 2 describes the percentage of BrdU staining positive hepatocytic nuclei present
in the regenerating liver 24, 38 and 48 hours after resection surgery in 24 mo. and 6 mo.
old mice. * denotes p-value < 0.05.
Microarray Analysis -
Animals, surgical treatment, and tissue preparation
Male mice of the CB6F1 hybrid strain at 5-6 months of age, and at 25-27 months of age
were obtained from Jackson Laboratories, and the National Institute on Aging,
respectively. Mice were anesthetized by use of an intraperitoneal injection of avertin
(approx. 400 mg/kg body weight), followed by supplementary inhalation of isoflurane, as
required. Surgical removal of 70% of the liver mass was performed, (Higgins 1931) and
animals were sacrificed (by cervical dislocation with prior isoflurane sedation) at 0.5, 1,
Fo
2, 4, 24,38 and 48 hour(s) following surgery (n=3 for each age at every time-point). The
liver tissue that was initially resected served as a 0h control time-point for each animal.
Sham surgeries were also performed whereby animals were subjected to similar
r
surgical conditions, however their livers were not resected but merely manipulated, and
tissue was retrieved at 0.5, 1, 2, 4, 24, 38 and 48 hour(s) following the sham operation
Pe
(n=2 for each age at every time-point). One hour prior to sacrifice, animals received an
intraperitoneal injection of BrdU (50 mg/kg body weight) (BrdU injections were not given
to the 0.5 h, or 1h PH animals, or to sham animals). For harvesting of tissue samples,
er
liver pieces were flash frozen in liquid nitrogen.

Re
vi
RNA isolation and quality control

ew
Total RNA was isolated using the RNeasy mini columns from Qiagen. Briefly, a small
amount of liver tissue was homogenized (polytron) in 1.3 ml of RLT lysis buffer. Only
600 µl of RLT homogenate was processed for RNA isolation (according to
manufacturer’s instructions), the remaining RLT homogenate was stored at –80oC.
Total RNA was eluted in 30 µl RNase-free water (Qiagen) and stored at –80oC. RNA
concentration was estimated by spectrophotometric quantitation (1:50 dilution) in
RNAse free water. Typical 260/280 ratios were 2.0 or greater. The quality/integrity of
RNA isolates was determined using the Agilent Bioanalyzer. In order to further
determine the ‘health’ status of animals, every 0h control liver sample was probed for
SAA1 mRNA expression by northern analysis. Subsequent to SAA1 northern results, 9
livers (7 x old, 2 x young) were removed from final analysis.
Microarray data quality control, normalization, statistical analysis, and annotation
of gene function.
Examination of the molecular profile of the priming phase of liver regeneration, involved
microarray analysis using 5 µg of total RNA from the 0, 0.5, 1, 2, and 4 hour(s) PH
samples (3 x replicates for each age, at every time-point). Analysis of mRNA
expression was achieved through use of Affymetrix MG_U74A version 2 and MOE430
high-density oligonucleotide arrays that contain 12,422 and 45000 targets for mouse
genes and expressed sequence tags (ESTs) respectively. Due to detection of elevated
SAA1 mRNA expression in the 0h control samples, chip numbers 6, 7 and 21 were
excluded from final analysis.
Fo
The sequential steps for quality control of microarray data were as follows: Firstly, the
.rpt files (obtained from Affymetrix Microarray Suite (MAS) 5.0) were examined for
quality control features such as overall signal intensity, and the 3’:5’ ratios of β-actin
r
probes were noted. Secondly, the .dat files were analyzed for gross defects. Following
the discovery of two ‘thumbprint-like’ defects in chip nos. 1 and 12, all chips were
Pe
processed using the ‘mask outliers’ function of MAS 5.0. Thirdly, the masked .cel files
were imported into dCHIP, where data was normalized, modeled (PM only), and log-
transformed. No array outlier warnings were found. Data was also normalized and
modeled using Robust Multichip Analysis (RMA).
er
Re
Data analysis
Linear modeling was carried out on the normalized data for time course analysis of the
gene expression in young and old animals for the partial hepatectomy and the sham
vi
group. Time course analysis in old –ph, young –ph, old –sham and young- sham as well
as interaction analysis between young and old was carried out using the LIMMA
ew
software data available through bio conductor package (Smyth 2004) for the
MGU74av2 and MOE430 gene chip datasets separately. Database for the normalized
expression values for genes in young and old animals along with Affymetrix annotation
and associated p-values for different analysis were organized in the Access software
from the Microsoft office package. Gene lists were generated for the datasets with a
negative or positive fold change of 1.5 times or more between young and old animals at
either 24h, 38h or 48h for MOE430 and 0h,05h, 1h,2h or 4h for MgU74av2 chips and an
intersection p-value <0.01 between young and old animals with PH treatment. Genes
showing laparotomy specific changes in the shams were excluded from the gene list.
Gene ontology analysis was carried out on the selected gene list using the online
resource David (david.abcc.ncifcrf.gov/) and pathways showing a significant enrichment
(p-value < 0.05) were selected. Pathways analysis and networks generation was done
using Ingenuity Pathways application. To study the differential temporal patterns of
genes between young and aged animals, list of genes changing with time in young and/
or old were selected at a p-value < 0.005 and a positive or negative fold change of 1.5
times or more. Selected genes were used for differential gene profile using the STEM
software (Ernst & Bar-Joseph 2006).
Supplementary Table 1 - List of primers used for Real Time PCR
Mouse Primers
Gene Primer Sequence
ICAM-1 Forward 5'-AATGCCAGCTCGGAGGATC-3'

Fo
Reverse 5'-CAGAGCGGCAGAGCAAAAG-3'
Iigp2 Forward 5' – TTGCCATCTCAGATCATGCC - 3'
Reverse 5' - CCTGGTGATAAACCCGGCTT- 3'
PML Forward 5'-TGGCGCCAGAGAGCTAGG-3'
r
Reverse 5'-TGTGTGGAAACGGGTATGGAC-3'
IRF1 Forward 5'-TTAGCCCGGACACTTTCTCTGATGG-3'
Reverse 5'-GTCCCCTCGAGGGCTGTCAATCTCT-3'
Pe
IFN-γ Forward 5' - CTGCTGATGGGAGGAGATGTC - 3'

Reverse 5' - ATGCTGGGACAGTGACCTGG - 3'
IRF7 Forward 5' - CAGCGAGTGCTGTTTGGAGAC - 3'
Reverse 5' - AAGTTCGTACACCTTATGCGG - 3'
er
Oas1 Forward 5' - GGCTGGAGGTTAAAATGCAT- 3'

Reverse 5' - GCATGCGGAAACACGTGTCTGG - 3'
Gadd45 Forward 5' - AATATGACTTTGGAGGAATTC- 3'
Reverse 5' - ATTCGGATGCCATCACCGTTC - 3'
Nmi Forward 5' - CGGCAGGATGCGCTACC - 3'
Re
Reverse 5' - TCAAGGCGCTCTCCATGG - 3'

CD49b Forward 5' - CCTTTCTATAAGGAGTGTGGCAGC - 3'
Reverse 5' - GGATCAGGTCAGAGATGCAAATTC - 3'
NKR-P1C Forward 5' - CTCCTGACTGCGAATCCCA - 3'
Reverse 5' - GAGATCTGGTGCGGATGGTAA - 3'
vi
NKR-P1A Forward 5' - CCTGTCTCTGCTGTCTGCCTCT - 3'

Reverse 5' - GTCCATGTGCTGTGCCACTAA - 3'
CD94 Forward 5' - GGCCAGAGCACTGCATTGT - 3'
ew
Reverse 5' - ATTCAGCAGAAACGCTTTTGC - 3'

γ-tubulin Forward 5' - TTCCGCAAGGAGGACATCTT - 3'
Reverse 5' - GGAGGTGTCCATCTCGTCAAAA - 3'
Ifngr1 Forward 5' - CCTGTCAGAGGTGTCCCTCG - 3'
Reverse 5' - GGGAGACCTTAGGACAGCTC - 3'
IFITM3 Forward 5' - TACAAGCAGCACCAGCCCGC - 3'
Reverse 5' - CGGTGCCTGGCCGTACTTGG - 3'
IFI47 Forward 5'- CTCGGACGGTTCTTCTTATC -3'
Reverse 5'- AGCACCCTCCTCTCTTCATG -3'
F4/80 Forward 5' - AGGACTGGAAGCCCATAGCCAA - 3'
Reverse 5' - GCATCTAGCAATGGACAGCTG - 3'
Microsialin Forward 5'- GGCGGTGGAATACAATGTGT -3'
Reverse 5'- TTATTGATTCCCCACCCCTA-3'
Siglec Forward 5'- TCGCCTCGTGGCCTCTACCC -3'
Reverse 5'- CGCACAGCCTGGGCATCGAA -3'
Sr-a Forward 5' - TAGGCACTTGGGATGTCTGA - 3'
Reverse 5' - GTCCTCAATTTGTATTGGTGCT - 3'
Supplementary Table 2 - List of G1/S checkpoint related cell cycle

genes differentially regulated between young and aged animals at
different time points after PH.
Fo
p-value
Description FC 24hO FC 38hO FC 48hO FC 24hY FC 38hY FC 48hY Affymetrix ID (interaction)
CCND1 cyclin D1 1.797 2.127 6.446 3.703 6.549 7.115 1417419_at 0.1280
r
CCNE2 cyclin E2 -1.175 1.78 1.889 2.948 9.671 1.842 1422535_at 5.20E-06
Pe
cyclin-
dependent
CDK2 kinase 2 1.094 1.17 1.24 1.019 1.713 1.279 1416873_a_at 4.90E-03
cyclin-
dependent
er
CDK4 kinase 4 1.265 1.316 1.27 1.25 1.681 1.482 1422441_x_at 4.73E-02
cyclin-
dependent
kinase inhibitor
CDKN1A 1A (p21, Cip1) 2.895 1.771 2.105 1.085 2.087 2.438 1421679_a_at 3.67E-04
Re
E2F
transcription
E2F2 factor 2 1.069 1.21 1.361 1.488 1.711 1.503 1455790_at 0.1446
E2F
transcription
vi
E2F1 factor 1 -1.135 1.459 1.597 1.503 2.76 1.87 1431875_a_at 5.80E-05
histone
HDAC5 deacetylase 5 -1.208 -1.363 -1.314 -1.543 -1.258 -1.29 1415743_at 7.91E-02
ew
histone
HDAC8 deacetylase 8 1.363 1.208 1.327 1.181 1.255 1.652 1428933_at 7.40E-02
histone
HDAC11 deacetylase 11 -2.441 -1.472 -1.521 1.087 -1.355 -1.956 1451229_at 2.46E-03
ErbB3 binding
EBP1 protein 1 1.711 1.574 1.318 1.557 1.5 1.618 1435372_a_at 0.4107
retinoblastoma-
RBL1 like 1 (p107) -1.02 1.608 6.041 2.218 4.964 4.54 1424156_at 1.77E-03
S-phase
kinase-
associated
SKP2 (SCF) protein 2 (p45) -1.374 1.292 1.489 1.348 2.441 1.927 1460247_a_at 3.39E-03
suppressor of
variegation 3-9
homolog 1
SUV39H1 (Drosophila) -1.038 1.254 2.014 1.025 1.86 2.136 1427382_a_at 3.73E-02
UBC ubiquitin C -1.472 -1.501 -1.783 -1.535 -1.316 -1.329 1454373_x_at 8.51E-02
Fo
r Pe
er
Re
vi
ew
Fo
r Pe
er
Re
vi
161x140mm (150 x 150 DPI)

ew
Fo
r Pe
er
Re
vi
159x134mm (150 x 150 DPI)

ew
Fo
r
218x98mm (150 x 150 DPI)

Pe
er
Re
vi
ew
Fo
r Pe
er
Re
227x170mm (150 x 150 DPI)

vi
ew
Fo
r Pe
er
218x141mm (150 x 150 DPI)

Re
vi
ew
Fo
r Pe
er
Re
vi
ew
87x115mm (150 x 150 DPI)
Page |1
Figures Legends
Figure 1- Aged liver exhibits higher levels of IFN-γ expression in liver tissue.
Elevated Message levels of IFN-γ in A) old 24 mos. CB6F1 mice compared to young 6
mos. (n=4); p-value<0.005; Fold Change=13.63 and B) old (24 mo.) C57BL/6 mice
compared to 3 mo. (n=5); p-value < 0.001; (fold change= 2.72) C) Tissue protein levels
of IFN-γ in old (24 mos.) and young (3. mo.) C57B/L6 mice liver (n=4); p-value = 0.19.
mRNA levels were normalized to γ-tubulin.
D and E) Frozen sections from liver tissue of old mice (24 mo.) and young mice (3 mo.)
were paraformaldehyde fixed and stained with DAPI (blue) and FITC labeled anti-IFN-γ
Fo
antibody (Green). Hepatocytes can be visualized with DAPI as large round isolated
nuclei whereas lymphocytic clusters can be seen as regions with denser concentration
of DAPI staining smaller elongated nuclei. (Magnification levels - 10X; inset- 40X).
r Pe
er
Re
vi
ew
Page |2
Figure 2- Changes in the expression profile of IFN-γ responsive genes in young

and old animals at early time points after partial hepatectomy.
A) Expression fold change between aged and young livers at time points 0h, 0.5h,1h, 2h
and 4h after PH in genes previously shown to exhibit elevated expression in response
to IFN-γ treatment in published studies – Set A (Der et al. 1998); Set B (Zocco et al.
2006). * denotes p < 0.05 (Significance value for directional change - Sign test)
B) Real-time PCR analysis for interferon-regulated genes - (A) IRF7 - interferon
regulatory factor 7, (B) IFI47- interferon gamma inducible protein 47, (C) Ifitm3 -
Fo
interferon induced transmembrane protein 3 (D) ISGF3 - Interferon-stimulated gene

factor 3 and (E) Ly6E - Lymphocyte antigen 6 complex, locus E. mRNA levels were
normalized to β-actin mRNA and are graphed relative to the 0h control time-point for
r
young liver. * denotes p < 0.05

Pe
er
Re
vi
ew
Page |3
Figure 3- Changes in the G1/S transition between young and aged mice.
A) Figure shows the expression levels of cell cycle genes cyclinD1, cdk4, E2F1 and
PCNA at time points 24h, 38h ad 48 after PH in aged and young mice. (N=3), * denotes
p-value < 0.05.
B) The G1/S transition network describes the Old/Young fold change in the expression
levels of the cell cycle genes in the regenerating liver at 24h post PH. liver. The fold
changes are colored on a scale ranging from green denoting relative down regulation of
genes to red indicating a relative upregulation. Solid lines denote direct physical
interactions, while dotted lines represent indirect regulatory interactions between the
genes. The genes in the network exhibit a fold change regulation of 1.5 times or more
and a p-value of interaction < 0.01.
Fo
r Pe
er
Re
vi
ew
Page |4
Figure 4- shows the protein levels of Stat1 and phospho-Stat1 in the livers of CB6F1
mice aged 24 mo. (old ) and 6 mo. (young ). Stat1 and pStat1 levels were both
normalized to β-actin. The levels of p-Stat1 were normalized relative to total Stat1 levels
at corresponding time points respectively. The composite charts show the averaged
levels of Stat1 and p-Stat1 for three sets of animals. * denotes p-value < 0.05.
Fo
r Pe
er
Re
vi
ew
Page |5
Figure 5- Changes in the levels of cell type specific markers and IFN-γ after NK
cell or macrophage depletion.
Changes in the message levels of immune cell specific genes 24hr after depletion. A) -
The levels of NK specific genes -NkrP1a, NkrP1c and CD94 showed a significant
decrease in depleted animals. IFN-γ message levels were also 2.8 times lower in the
depleted animals. B) - Levels of macrophage specific genes – F4/80, microsialin, siglec
and SR-a were significantly reduced. The levels of IFN-γ were 4.8 times lower. *
denotes p < 0.05. Figure C) and D) show the difference in BrdU incorporation and
PCNA staining levels between untreated - young (6mo) and old mice (24mo); NK cell
depleted – AsGM1 treated and NRS control mice (18mo.) and macrophage depleted –
GdCl3 treated and saline control mice (18 mo.) 38hrs after partial hepatectomy. BrdU
was injected 1hr prior to tissue collection. (N=3-5), * denotes p < 0.05
Fo
r Pe
er
Re
vi
ew
Page |6
Figure 6- Differences in the distribution of cell markers between IFN-γ -/- mice and
C57BL/6 wild type 18 mo mice at different time points after partial hepatectomy.
The figure shows the differences in the levels of – A) PCNA – cell cycle marker for late
G1/S phase and B) BrdU incorporation which indicates DNA synthesis rates. The
immunohistochemical analysis was carried out on livers harvested after 32h, 40h and
48hrs post hepatectomy. BrdU injections were administered 1h prior to sacrifice. *
denotes p < 0.05
Fo
r Pe
er
Re
vi
ew

Aging

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Aging

Uploaded by

Copyright:

Available Formats

Aging Cell

Elevated Interferon Gamma Signaling Contributes to

Journal: Aging Cell

Manuscript ID: Draft

Manuscript Type: Original Paper

Date Submitted by the

Complete List of Authors: Singh, Pallavi; Columbia University, Department of Dermatology

Dean, Adam; Baylor College of Medicine, Huffington Center on

Aging, inflammation, Liver Regeneration, Interferon gamma, Cell

“Elevated Interferon gamma signaling contributes to impaired

Medicine, Houston, TX,

Corresponding Author: Pallavi Singh Ph.D.

Key Words: Aging, Inflammation, Gamma Interferon, Liver Regeneration, Partial

Word Count: 4436

Several studies have shown an upregulation of the inflammatory response with

proinflammatory status (Bruunsgaard & Pedersen 2003); (Franceschi et al.

2000). Differential gene-expression profiling of young and aged livers using

display enhanced proliferation, indicating a detrimental effect of inflammatory

Our previous study described an elevation in the transcript levels of various

depleted animals. BrdU incorporation rates were comparable to young at 38 h,

understanding their effect on recovery after liver resection surgeries in aged

hepatocytes, we carried out separation of the parenchymal fraction enriched in

showed a general elevation of message levels of these IFN-γ targets compared

Microarray gene expression dataset comparing the transcriptional profiles of

interferon regulated genes are differentially over-expressed in the aged livers

specific transcription factors called IRFs or interferon regulatory factors (Boehm

2.2 Depletion of NK cells and macrophages leads to improved regeneration

The success of depletion was determined by quantifying the message levels of

regeneration. The macrophage depleted animals showed higher levels of BrdU

other cytokines/cell-products could also play a similar role in aged animals.

2.3 Liver regeneration in aged IFN-γ -/- mice

IFN-γ signaling, known to arrest growth by induction of cell cycle inhibitors is

mortality rates post hepatectomy. PCNA is expressed during early G1 and S

activation of immune cells. During liver regeneration these cytokines signal

complement cascade which help in antigen clearance and mediation of

Although innate immunity is an integral part of regeneration, the presence of

Conditions of liver inflammation such as artificially induced hepatitis by

hepatic lymphocyte infiltration, necrosis and inhibition of regeneration was

leads to improvement of regeneration due to reduction of production IL-10 which

implicated in hepatic injury during liver regeneration along with increased

deterioration in a previously healthy patient and in the elderly, the condition is

(AALAC)-approved Center of Comparative Medicine at Baylor College of

avertin in 0.9% saline 240mg/kg body weight) underwent partial hepatectomy

depletion experiments, mice were injected intravenously 24 h before surgery with

in the supplementary table 1. The real-time PCR analysis was performed

4.5 Protein isolation and Western analysis

Nuclear, cytoplasmic, or whole cell extracts were isolated according to the

levels in the liver.

4.7 Microarray Analysis – Supplementary Data

Acknowledgements: We thank the Digestive Disease Center, Baylor

Author Contributions: P. Singh - Main author (animal surgeries, microarray

Abo T, Kawamura T , Watanabe H (2000). Physiological responses of

interferon-gamma. Annu Rev Immunol. 15, 749-795.

Brooling JT, Campbell JS, Mitchell C, Yeoh GC , Fausto N (2005). Differential

disease. Immunol Allergy Clin North Am. 23, 15-39.

Cancer Res. 24, 509-512.

regeneration: Innate immune response. World J Gastroenterol. 12, 7852-

van den Brink MR , Todo S (1991). Studies on mechanisms of

Franceschi C, Bonafe M, Valensin S, Olivieri F, De Luca M, Ottaviani E , De

immunosenescence. Ann N Y Acad Sci. 908, 244-254.

Wheeler MD (2007). Impaired liver regeneration and increased oval cell

Appella E , Harris CC (2003). Nitric oxide-induced cellular stress and p53

Kano A, Watanabe Y, Takeda N, Aizawa S , Akaike T (1997). Analysis of IFN-