Hutchinson-Gilford progeria syndrome (HGPS) is a rare and fatal human premature aging
disease1–5, characterized by premature arteriosclerosis and degeneration of vascular
smooth muscle cells (SMCs)6–8. HGPS is caused by a single-point mutation in the LMNA
gene, resulting in the generation of progerin, a truncated splicing mutant of lamin
A. Accumulation of progerin leads to various aging-associated nuclear defects including
disorganization of nuclear lamina and loss of heterochromatin9–12. Here, we report
the generation of induced pluripotent stem cells (iPSCs) from fibroblasts obtained
from patients with HGPS. HGPS-iPSCs show absence of progerin, and more importantly,
lack the nuclear envelope and epigenetic alterations normally associated with premature
aging. Upon differentiation of HGPS-iPSCs, progerin and its aging-associated phenotypic
consequences are restored. Specifically, directed differentiation of HGPS-iPSCs to
SMCs leads to the appearance of premature senescence phenotypes associated with vascular
aging. Additionally, our studies identify DNA-dependent protein kinase catalytic subunit
(DNAPKcs) as a downstream target of progerin. The absence of nuclear DNAPK holoenzyme
correlates with premature as well as physiological aging. Since progerin also accumulates
during physiological aging6,12,13, our results provide an in vitro iPSC-based model
to study the pathogenesis of human premature and physiological vascular aging.
Three HGPS primary fibroblast lines, originally isolated from patients with the classical
LMNA mutation (Gly608Gly), were transduced with retroviruses encoding OCT4, SOX2,
KLF4, c-MYC, and GFP. Nanog-positive colonies were effectively obtained when using
early passage, but not late passage (>25), HGPS fibroblasts (Fig. S1a). We focused
on iPSC lines of a well characterized HGPS fibroblast line, AG019729–12. Compared
to normal fibroblasts, HGPS fibroblasts (AG01972) showed abnormal nuclear morphology,
reduced expression of the lamina components lamin B1 and LAP2β, loss of heterochromatin
markers H3K9Me3, HP1α, and HDAC1, and reduced expression of nuclear proliferation
marker Ki67 (Fig. 1a, S2). From HGPS fibroblasts, we derived six iPSC lines. In addition,
we generated control iPSC lines from wild type fibroblasts (BJ and IMR-90 cell lines).
Control and HGPS iPSC lines demonstrated pluripotent gene expression, demethylation
of the OCT4 promoter and transgene silencing (Fig. 1b, S1, S3, S4a, and data not shown).
They were maintained for more than 50 passages without a loss of pluripotency or the
acquisition of detectable morphological or growth abnormalities. The pluripotency
of each iPSC line was assessed by differentiation into the three embryonic germ layers
in vitro, using embryoid body (EB) formation, and/or in vivo, by teratoma formation
(Fig. S5a–d). Out of these lines, we primarily focused on HGPS-iPSC#4 and BJ-iPSC#3
for our studies (hereafter referred to as HGPS-iPSC or BJ-iPSC). Both BJ-iPSCs and
HGPS-iPSCs were able to differentiate towards specialized mesoderm-derivatives such
as smooth muscle cells (SMC), endothelial cells (Fig. S6a), or beating cardiomyocytes
(Movie S1–S2). Moreover, all analyzed iPSC lines showed normal chromosomal integrity
(Fig. S5e). Finally, LMNA sequencing confirmed the presence of the classical mutation
in HGPS-iPSCs (Fig. S5f). Altogether, these data indicate that the somatic cells from
HGPS patients, despite their significant premature senescence phenotypes and nuclear
defects, have been properly reprogrammed and can be effectively maintained in a pluripotent
state.
Lamin A/C protein is expressed in differentiated somatic cells but is absent in embryonic
stem cells (ESCs)11,14. Therefore, we next examined the expression of lamin A/C in
the generated iPSC lines. As shown in Fig 2a, lamin A/C expression is significantly
downregulated in iPSCs, compared to their parental fibroblasts, whereas lamin B1 transcripts
were upregulated. Although progerin should follow a similar pattern of expression
as observed for lamin A/C, LMNA expression is independent of promoter methylation
status (Fig. S4b)15. Indeed, a complete loss of progerin mRNA in HGPS-iPSCs was observed
(Fig. 2a). Furthermore, expression of lamin A/C and progerin proteins was practically
undetectable (Fig. 2b, S4c).
Since HGPS-iPSCs did not express progerin, we examined whether the nuclear abnormalities
observed in HGPS fibroblasts would also be absent at the pluripotent stage. Our results
indicate that all of the epigenetic, nuclear lamina and proliferation parameters analyzed
in HGPS-iPSCs were indistinguishable from BJ-iPSCs (Fig. 2b–c, S3, S7). In addition,
the nuclei of HGPS-iPSCs displayed the characteristic wrinkles and lobes observed
in hESCs and iPSCs (Fig. S8), indicative of a reprogramming of the nuclear envelope
components. Since the nuclear envelope associates with and regulates heterochromatin11,16,
we next examined genome-wide CpG methylation in HGPS fibroblasts, BJ fibroblasts,
HGPS-iPSCs, BJ-iPSCs, and H9 hESCs. Using bisulfite padlock probes and Illumina sequencing,
we captured and quantified the methylation level of an average of 95,932 CpG sites
within a set of 16,206 well annotated Differentially Methylated Regions (DMRs)17 per
cell line (Table S1). The correlation coefficient of the global methylation levels
between the pluripotent lines (BJ-iPSCs, HGPS-iPSCs and H9 hESCs) and the corresponding
fibroblasts indicated that the generated pluripotent lines are much more closely related
to each other and to hESCs than the two fibroblast lines (Fig. 2d). Interestingly,
586 autosome genes were found to be associated with regions that exhibited methylation
differences between HGPS and BJ fibroblasts (Table S2, Fig S9a). Furthermore, based
on DAVID18,19 analysis, we found that these genes were enriched for 21 Gene Ontology
terms, most of which were related to development and transcriptional regulation (Fig
S9a). In contrast, methylation differences between HGPS-iPSCs and BJ-iPSCs were only
found for 33 autosome genes (Table S3), which showed no significant functional enrichment.
Therefore, the presence of progerin in HGPS fibroblasts appears to lead to major epigenomic
changes in various pathways. These changes were no longer present in HGPS-iPSCs, coinciding
with the downregulation of progerin. Finally, genome-wide mRNA profiling demonstrated
that HGPS-iPSCs and BJ-iPSCs are closely related together with H9 hESCs, and discrete
from their parental fibroblasts (Fig. S9b–d). These results demonstrate the complete
resetting of the nuclear architecture, epigenome and global gene expression in HGPS
cells after being reprogrammed to pluripotency.
To test whether the expression of progerin could be re-activated, we first subjected
HGPS-iPSCs to in vitro differentiation via EB formation. Progerin mRNA was selectively
induced in differentiated HGPS-iPSCs, but not in differentiated BJ-iPSCs (Fig. S10a).
In contrast, lamin A was upregulated in both HGPS-iPSCs and BJ-iPSCs (Fig. S10a).
This reversible suppression of progerin expression by reprogramming, and subsequent
reactivation upon differentiation, provides a unique model system to study human premature
aging pathologies. Progerin is known to accumulate mainly in arterial SMCs of HGPS
patients, and vascular SMC degeneration is one of the characteristics of HGPS-associated
arteriosclerosis6,7,20. In fact, vascular SMC senescence has been involved in the
advanced arteriosclerosis of normal populations7,21,22. We therefore next asked whether
SMCs differentiated from HGPS-iPSCs exhibit premature senescence phenotypes. Using
a directed differentiation protocol, we obtained SMC populations from HGPS-iPSCs and
BJ-iPSCs, the majority of which expressed characteristic SMC markers such as smooth
muscle actin (SMA) and calponin (Fig. S6a). Immunoblotting and RT-PCR analyses confirmed
the expression of progerin in HGPS-iPSC, but not BJ-iPSC-derived SMCs (hereafter referred
to as “HGPS-SMC” and “BJ-SMC”, Fig. S6b–c). To model SMC senescence in vitro, the
differentiated SMCs were serially passaged in culture. As shown in Fig. 3a–c and S10b,
an increasing frequency of misshapen nuclei and a loss of the heterochromatin mark
H3K9Me3 were specifically observed in HGPS-SMCs after serial passaging. HGPS-SMCs
at later passages (i.e. passage 5) showed the typical characteristics of premature
senescence, including increased senescence-associated-β-Gal (SA-β-Gal) staining (Fig.
3d–e, S10c), reduced telomere length (Fig. 3f), reduced number of Ki67-positive cells
(Fig. 3g, S10d), and compromised cell proliferation (Fig. 3h, S10e). We also found
a selective upregulation of senescence-related transcripts in HGPS-SMCs (Fig. S10f).
To test whether the observed HGPS-related cell phenotypes were specific to SMCs, we
differentiated HGPS-iPSCs into fibroblasts and measured progerin-associated parameters.
Progerin expression in HGPS-iPSC-derived fibroblasts was detectable as early as passage
5 (Fig. S11a). However, we were unable to detect a loss of lamina or heterochromatin
markers before passage 10 (Fig. S11b–c). Nonetheless, these defects were present specifically
in HGPS-iPSC-derived fibroblasts, in contrast to control iPSC-derived fibroblasts
analyzed at similar passage (data not shown). Thus, even though direct comparison
of SMCs and fibroblasts is not possible due to their different culture conditions,
our observations demonstrate that mesoderm lineages derived from HGPS-iPSCs display
a characteristic HGPS phenotype.
We next investigated whether progerin accumulation is the direct cause of the accelerated
cell senescence observed in HGPS-SMCs. To this end, we induced ectopic expression
of progerin in human primary vascular SMCs. We found that introduction of progerin
in wild type SMCs resulted in compromised cell proliferation and nuclear defects,
as we had observed in HGPS-SMCs (Fig. 3i, S12). As a complementary approach, we transduced
HGPS-iPSCs with a lentiviral vector expressing a progerin-specific shRNA23. The modified
iPSCs showed normal karyotypes as well as normal expression of lamina/epigenetic and
pluripotent markers (Fig. S13a–b). After EB-based differentiation, both the mRNA and
protein levels of progerin, but not those of lamin A, were substantially downregulated
in the progerin-shRNA “corrected” HGPS-iPSCs compared to control cells (Fig. 3j, S13c–d).
We next differentiated these “progerin-free” HGPS-iPSCs into SMCs (Fig. S14a). A dramatic
improvement in the proliferation capability, as well as a downregulation of senescence-related
transcripts, was found in the SMCs differentiated from the corrected HGPS-iPSCs (Fig.
3k and S14b–c). Furthermore, transduction of progerin shRNA into early passage HGPS-iPSC-derived
fibroblasts resulted in a clear restoration of nuclear morphology and heterochromatin
markers after extended culture (Fig. S15). Taken together, these data identify progerin
as the key factor underlying the premature senescence phenotypes of HGPS-iPSC-derived
cells.
Since phenotypic characteristics of premature aging were able to be recapitulated
by directed differentiation of the HGPS-iPSCs, we next investigated whether this model
could serve to identify novel senescence-related markers. By using a sensitive MudPIT
proteomic approach24,25, we identified DNA-dependent protein kinase catalytic subunit
(DNAPKcs) as a hitherto unknown binding partner of progerin (Table S4). DNAPK holoenzyme,
comprising of DNAPKcs and its regulatory subunits Ku70/Ku80, is involved in various
aging-related cellular events26,27, and DNAPKcs or Ku70/Ku80-deficient mice exhibit
accelerated aging27,28. To further confirm the association of progerin with DNAPKcs,
we performed co-immunoprecipitation (co-IP) experiments. As shown in Fig 4a, ectopically
expressed progerin associated tightly with endogenous DNAPKcs. In contrast, lamin
A showed weak interaction with DNAPKcs, whereas both progerin and lamin A exhibited
similar binding to lamin B1. Neither progerin nor lamin A co-immunoprecipitated with
endogenous WRN protein. Since most of the nuclear proteins in complex with lamin A
are destabilized in HGPS cells9,29, we analyzed the status of DNAPKcs in primary HGPS
fibroblasts. We observed decreased nuclear DNAPKcs in HGPS fibroblasts when compared
to normal fibroblasts (Fig. 4b–c). In addition, the regulatory subunits Ku70/Ku80
were also downregulated in HGPS fibroblasts (Fig. S16a). Interestingly, we detected
a complete restoration of DNAPKcs/Ku70/Ku80 expression in HGPS-iPSCs (Fig. 4b, 4d,
S3, S16b), although a deficiency in the expression of these proteins reappeared after
differentiation into SMCs (Fig. 4e). These observations suggest that the downregulation
of DNAPKcs in HGPS cells is dependent on the accumulation of progerin in differentiated
cells. In fact, ectopic expression of progerin in primary vascular SMCs diminished
DNAPKcs/Ku80 protein expression (Fig. 4f, S17). We next tested whether inactivation
of DNAPKcs could partially contribute to some of the phenotypes observed in HGPS-SMCs.
We found that knockdown of DNAPKcs reduced the proliferation of primary vascular SMCs
(Fig. S18a). Finally, we extended our results and found that progressive loss of DNAPKcs/Ku70/Ku80
also occurs in fibroblasts isolated from normally aging individuals (Fig. S18b). Overall,
our data suggest that deficiency of the DNAPK holoenzyme may constitute a novel marker
for premature as well as physiological aging.
In summary, our results not only highlight the plasticity of the lamina-epigenetics
axis, but also point to the fact that the altered structure of the nuclear envelope,
as well as the epigenetic modifications that accumulate during physiological aging12
or under specific disease conditions1,11, can be restored to normalcy by reprogramming
(Fig. S19). The gradual onset and complexity of aging has impeded progress in understanding
the pathogenesis of aging-related cardiovascular disorders. Recently, striking similarities
between normal aging-associated and HGPS-associated arteriosclerosis have been reported6,16.
Indeed, the levels of progerin increase gradually during physiological aging6. Our
study provides the first evidence that, in a progerin-dependent manner, HGPS-iPSC-derived
SMCs reach senescence-related phenotypes earlier than their normal counterparts. The
iPSC-based accelerated aging model presented here and by Zhang et al30 may provide
an avenue to model and study the pathogenesis of human aging-related vascular diseases
as well as various human laminopathies1.
METHODS SUMMARY
iPSCs were generated from human fibroblasts with retroviruses encoding OCT4/SOX2/KLF4/c-MYC/GFP,
and cultured on MEF feeder cells or Matrigel. SMCs were differentiated from iPSCs-derived
CD34+ progenitor cells following an OP9-based protocol.
ONLINE METHODS
Cell culture
H9 hESCs (WiCell Research) and iPSCs were maintained on a layer of mitotically inactivated
mouse embryonic fibroblasts (MEFs) in hESC medium: DMEM/F12 (Invitrogen) supplemented
with 0.1 mM non-essential amino acids (Invitrogen), 1 mM glutamax (Invitrogen), 20%
Knockout Serum Replacement (Invitrogen), 55 μM β-mercaptoethanol (Invitrogen) and
10 ng/ml bFGF (Joint Protein hESCs and iPSCs were also cultured in Matrigel (BD Biosciences)
with mTeSR medium (Stem Cell Technologies). Human HGPS fibroblasts AG01972, AG11498,
AG06297, and normal fibroblasts GM00038 (9 year), AG05247 (87 year), and AG09602 (92
year) were purchased from Coriell Cell Repository. BJ normal human fibroblasts (CRL-2522)
were purchased from ATCC. All human fibroblasts were cultured at 37°C in DMEM containing
glutamax, non-essential amino acids, sodium pyruvate, and 15% fetal bovine serum (FBS).
Human aortic smooth muscle cells were purchased from Lonza and maintained in SmGM-2
medium (Lonza, Cat. # CC-3182).
Reagents
Antibodies were obtained from the following sources. Abcam: anti-NANOG (ab21624),
anti-H3K9Me3 (ab8898), anti-progerin (ab66587), anti-emerin (ab14208); anti-Ku70 (ab2172);
Santa Cruz Biotechnology: anti-Oct-3/4 (sc-5279), anti-SOX2 (sc-17320), anti-HDAC1
(sc-7872), anti-DNAPKcs (sc-9051), anti-lamin A/C (sc-6215), anti-lamin A/C (sc-7293),
anti-lamin B1 (sc-6217); Cell Signaling: anti-HP1α (2616); anti-Ku80 (2753); R&D systems:
anti-Foxa2 (AF2400); Millipore: anti-TRA-1-60 (MAB4360) ; Sigma: anti-β-Tubulin III
(T2200), anti-SMA(A5228), anti-Flag (M2), and anti-tubulin (T5168); Dako: anti-calponin
(clone CALP); anti-endoglin (clone SN6h); BD Transduction Laboratories: anti-LAP2β
(611000); MBL: Agarose-conjugated anti-GFP.
Plasmids
The pMXs vector containing the human cDNAs for OCT4, SOX2, KLF4 and c-MYC were purchased
from Addgene (17217, 17218, 17219 and 17220, respectively). pBABE-puro-GFP-progerin
and pBABE-puro-GFP-wt-lamin A were purchased from Addgene (17663 and 17662, respectively).
Flag-progerin lentiviral vector was kindly provided by Dr. Lucio Comai 31. For the
generation of the shRNA expression vectors against progerin and DNAPKcs, corresponding
oligos (see Table S5) were cloned into a MluI/ClaI-cleaved pLVTHM plasmid (Addgene,
12247). All the constructs generated were subjected to DNA sequencing to confirm accurate
shRNA target sequence.
Retrovirus and lentivirus production
For retrovirus production, 293T cells were transfected with the pMXs vectors carrying
OCT4, SOX2, c-MYC, KLF4, or GFP cDNAs, together with the packaging plasmids (pCMV-gag-pol-PA
and pCMV-VSVg, provided by Dr. Gerald Pao, The Salk Institute) using Lipofectamine
2000 (Invitrogen). Retroviruses were collected 36–48 h after transfection, and filtered
through a 0.45 μM filter. Lentiviruses were generated by co-transfecting the pLVTHM
vector together with the packaging plasmids (psPAX2 and pMD2.G, from Addgene, 12260
and 12259 respectively) into 293T cells using Lipofectamine 2000 (Invitrogen). Lentiviruses
were collected 36 hours after transfection and concentrated by ultracentrifugation.
iPSCs Generation
For the generation of human iPSCs, human fibroblasts were seeded in a 6-well plate
and spin-infected with a mix of high-quality retroviruses encoding OCT4, SOX2, KLF4,
c-MYC, and GFP in the presence of 4 μg/ml polybrene. Three infections on consecutive
days were performed. Six days after the first infection, fibroblasts were gently individualized
with TrypLE (Invitrogen) and seeded onto fresh MEFs in the fibroblast culture medium.
After 24 h, the medium was switched to hESC medium, and changed every 1–2 days depending
on cell density. To establish the iPSC lines, colonies were manually picked and transferred
onto MEF feeder cells for several passages before being transferred to Matrigel/mTesR
conditions.
Lentiviral infection of iPSCs
HGPS-iPSC#4 cell line cultured on Matrigel was treated with 10 μM ROCK inhibitor Y-27632
for 1 h and then individualized with TrypLE. Cells were infected in suspension with
either the concentrated lentivirus pLVTHM or pLVTHM-shRNA-Progerin in the presence
of ROCK inhibitor and polybrene (4 μg/ml) for 1 h. Cells were centrifuged to remove
the lentivirus and seeded back on fresh feeder MEFs in hESC media containing ROCK
inhibitor. After being cultured for a few days, small colonies were manually passaged
as a pool of colonies onto fresh MEFs to establish new iPSC lines. GFP expression
was used as an indicator to determine successful integration of the lentiviruses.
Cell differentiation
For embryoid bodies (EBs) based differentiation, the iPSC colonies growing on MEFs
were detached with dispase treatment, resuspended in DMEM/F12 medium supplemented
with 0.1 mM non-essential amino acids, 0.5 mM L-glutamine, 10% FBS (Atlanta Biologicals),
and 55 μM β-mercaptoethanol and cultured in low attachment 6-well plates for 4 days.
The EBs were then plated on gelatin-coated plates and maintained for another 10–17
days. Differentiation of iPSCs into fibroblasts was performed as previously described32.
Directed differentiation toward smooth muscle cells (SMCs) was performed essentially
as previously described32 with slight modifications. Irritated OP9 cells were plated
at 1 × 105 cells/well onto gelatinized 6-well plates in OP9 growth medium. After the
formation of confluent iPS cell cultures for 4 and 5 days, undifferentiated iPS cells
were harvested by treatment with 1 mg/mL dispase and dispersed by scraping to maintain
the cells in small clumps. Concurrently, iPS cultures growing under the same conditions
were used to obtain single cell suspension for counting. The iPSCs were added to OP9
cultures at a density of 3 × 105/2 mL per well of a 6-well plate in half TesR medium
and half hESC media. iPSCs were allowed to recover for 1–2 days in hESC media. At
day 0 of differentiation, the media was changed to Knockout DMEM supplemented with
10% FBS (HyClone), 10 mM β-mercaptoethanol, 1 mM L-glutamine, and 100 mM nonessential
amino acids. The iPSC/OP9 cocultures were incubated for up to 10 days at 37°C in 5%
CO2 conditions with medium change every other day. After 10 days of differentiation,
the coculture cells were harvested with TrypLE (Invitrogen) for single-cell suspension
and labeled with CD34 microbeads kit (Miltenyi Biotech, Cat. #130-046-702). Following
the manufacturer's protocol, cells were passed through MS separation column attached
to a Midi-MACS separation unit (Miltenyi Biotech) to obtain a magnet-retained fraction
of purified CD34+ cells. Isolated CD34+ cells were then plated in smooth muscle cell
media (SmGM-2 BulletKit, Lonza, Cat. # CC-3182) and maintained at 37°C in 5% CO2 conditions
with medium change every 2–3 days34. SMCs were passaged using TrypLE (diluted 1:4)
for 3 minutes at 37°C. To analyze early onset of senescence, cells were passaged at
a ratio of 1:3 (~6000–7500 cells/cm2) only when the cells reached confluence. To calculate
population doublings, SMCs seeded at 3500 cells/cm2 were passaged once culture reached
85–90% confluence. Cell growth was measured at every passage by calculation of accumulated
population doublings using the formula (logH -logS) / log2.0 (H=number of cells harvested;
S=number of cells seeded on the first day of each passage).
Protein and mRNA analysis
Cells were lysed and subjected to immunoblotting analysis according to the previously
described method35. Total RNA was extracted using Trizol (Invitrogen) followed by
cDNA synthesis using High capability RNA-to-cDNA Mater Mix (Invitrogen). Quantitative
RT-PCR was performed using SYBR Green PCR Master Mix (Applied Biosystems). Primer
sequences are listed in Table S5.
Immunofluorescence microscopy
Cells were fixed with 4% formaldehyde in PBS at room temperature (RT) for 20–30 min.
After fixation, cells were treated with 0.4% Triton X-100 in PBS for 5 min at RT.
After blocked with 10% FBS in PBS for 30 min, cells were incubated at RT for 1 h or
at 4°C overnight with the primary antibody, followed by washing in PBS and incubation
at RT for 1 h with the corresponding secondary antibody. Nuclei were stained with
Hoechst 33342 (Invitrogen). Quantitative microscopy measurements were carried out
as described previously9. Error bars represent standard deviations.
Immunohistochemical detection of NANOG
Cells were fixed with 4% formaldehyde in PBS at RT for 30 min, and permeabilized with
0.4% Triton-X100 in PBS for 10 min. Then the cells were incubated overnight with rabbit
anti-human NANOG antibody in 1% BSA/PBS, followed by incubation with a secondary biotin-conjugated
anti-rabbit antibody for 2 hours. Finally, cells were incubated with streptavidin-HRP
for 1 hour (Vector), and NANOG-positive cells were visualized with a DAB substrate
kit (Vector).
Teratoma analysis
To test pluripotency in vivo, NOD-SCID IL2Rgammanull mice (Jackson laboratories) were
injected with the indicated iPSC lines and teratoma formation assessed. Briefly, ~106
iPSCs in ~50uL of hESC medium were injected into the testis or kidney capsule of anesthetized
mice. Mice were monitored for teratoma formation and euthanized ~6–12 weeks after
injection. Teratomas were harvested, processed and analyzed by hematoxylin-eosin staining
and immunostaining. All animal experiments were performed with approval of The Salk
Institute Institutional Animal Care and Use Committee (IACUC).
Mutation validation
Primer sequences to amplify exon 11 of the LMNA gene are listed in Table S5. 50 μl
PCR reactions using 3 ng genomic DNA templates, 100 nM of the forward and reverse
primers with 25 μl Taq 2× Master Mix (NEB) was performed at 94°C for 2 min, 34 cycles
of 94°C 30 s, 55.5°C for 40 s, and 72°C for 40 s, and finally 72°C for 3 min. Products
were purified with 0.9× volume of AMPure beads (Agencourt). Amplicons were sequenced
by capillary Sanger sequencing (Genewiz). Results were visualized using an ABI Sequence
Scanner.
Genome–wide DNA methylation analysis
Genomic DNA was extracted using ALLPrep DNA/RNA Mini kit (Qiagen). Bisulfite conversion
and capture reaction was carried out on each sample (genomic DNA of fibroblasts, iPSCs,
or hESCs). The detailed protocol for genomic DNA methylation has been described previously36,
and the detailed information for DNA methylation is presented in Table S1.
Bisufite sequencing of OCT4 and LMNA promoters
Bisulfite conversion was carried out using 2 μg of purified genomic DNA using the
Zymo EZ-DNA Methylation Gold Kit (Zymo Research) following the manufacturer's instructions.
PCR was set up using previously published primers15. Cycling was terminated at 35
cycles. PCR products were purified using 2 % Size-Select E-gel (Invitrogen) and reamplified
for 10 cycles using Phusion HF enzyme (NEB). PCR products were cloned using Zero-blunt
PCR Cloning kit (Invitrogen) and heat transfected to TOP10 E. coli competent cells
(Invitrogen). Individual colonies were selected and sent for single pass sequencing.
DNA microarray and Bioinformatics analysis
The GeneChip microarray processing was performed by the Functional Genomica Core in
the Institute for Research in Biomedicine (Barcelona, Spain) according to the manufacturer's
protocols (Affymetrix, Santa Clara, CA). The amplification and labeling were processed
as indicated in Nugen protocol with 25ng starting RNA. For each sample, 3.75μg ssDNA
were labeled and hybridized to the Affymetrix HG-U133 Plus 2.0 chips. Expression signals
were scanned on an Affymetrix GeneChip Scanner (7G upgrade). The data extraction was
done by the Affymetrix GCOS software v.1.4. The statistical analysis of the data was
performed using ArrayStar 3. Briefly, raw CEL files were imported together with gene
annotation from NetAffix (from 11/13/2009) and after checking for top replication
quality for each of the 5 pairs of samples (R2>0.99), data was summarized at the gene
level (20,765 genes) and the median was used for each gene and sample type. As both
H9 hESCs and HGPS-iPSCs originate from female samples, and in order to remove any
possible bias introduced by the X and Y chromosome-coded genes, we performed the same
analysis with only autosome genes (19,884 genes). The result of the hierarchical clustering
is very similar to the one using all genes and is shown in Figure 2f. In addition,
a principal component analysis was performed on RMA-normalized probeset intensity
values for autosomes using the prcomp function in R (http://www.r-project.org/) (the
same figure including all genes gave highly similar results, data not shown). A figure
illustrating the two first principal components is shown in figure S9c. Differences
between some of the samples is shown using scatter plot of RMA-normalized intensity
values in figure S9b.
Multidimensional Protein Identification Technology (MudPIT) analysis of progerin-associated
proteins
The immunoprecipitation for MudPIT assay was performed as previously described24,25.
In brief, HEK293T cells were transfected with GFP-progerin or GFP and maintained in
culture for 48 h. After cells were lysed, the GFP-progerin, GFP, and their associated
proteins were immunoadsorbed to anti-GFP agarose. The immunoprecipitates were then
eluted with 8 M urea in 100 mM Tris, pH 8.5. The samples were reduced by adding 0.3
μL of 1M TCEP (for a final concentration of 5 mM TCEP) and incubated at RT. To alkylate,
1.2 μL of Iodoacetamide (10 mM final concentration) was added and the samples were
subsequently incubated at RT in the dark for 15 minutes. The addition of 180 μL of
100 mM Tris pH 8.5 diluted the solutions to 2 M Urea. Calcium chloride (100 mM) was
then added (2.4 μL) for a final concentration of 1 mM CaCl2. Trypsin (0.5 μg/μL) was
added in the amount of 7.0 μL. The resulting mixtures were then shaken for 18 hours
and incubated in the dark at 37 °C. To neutralize 13.5 μL of Formic Acid (90%) was
added for a final concentration of 5% Formic Acid. The tubes were centrifuged for
30 minutes at 2 °C in a table-top centrifuge. Upon completion of the digestion, the
proteins were pressure-loaded onto a fused silica capillary desalting column containing
3 cm of 5-μm strong cation exchange (SCX) followed by 3 cm of 5-μm C18 (reverse phase
or RP material) packed into an undeactivated 250-μm i.d capillary. Using 1.5 mL of
buffer A (95% water, 5% acetonitrile, and 0.1% formic acid) the desalting columns
were washed overnight. Following the desalting process, a 100-μm i.d capillary consisting
of a 10-μm laser pulled tip packed with 10 cm 3-μm Aqua C18 material (Phenomenex,
Ventura, CA) was attached to the filter union (desalting column-filter union-analytical
column) and the entire split-column (desalting column-filter union-analytical column)
was placed in line with an Agilent 1100 quaternary HPLC (Palo Alto, CA) and analyzed
using a modified 6-step separation, described previously25. The buffer solutions used
were 5% acetonitrile/0.1% formic acid (buffer A), 80% acetonitrile/0.1% formic acid
(buffer B), and 500 mM ammonium acetate/5% acetonitrile/0.1% formic acid (buffer C).
Step 1 consisted of a 90 min gradient from 0–100% buffer B. Steps 2–5 had the following
profile: 3 min of 100% buffer A, 2 min of X% buffer C, a 10 min gradient from 0–15%
buffer B, and a 97 min gradient from 15–45% buffer B. The 2 min buffer C percentages
(X) were 20, 40, 60, 80% respectively for the 6-step analysis. In the final step,
the gradient contained: 3 min of 100% buffer A, 20 min of 100% buffer C, a 10 min
gradient from 0–15% buffer B, and a 107 min gradient from 15–70% buffer B. As peptides
eluted from the microcapillary column, they were electrosprayed directly into an LTQ
2-dimensional ion trap mass spectrometer (ThermoFinnigan, Palo Alto, CA) with the
application of a distal 2.4 kV spray voltage. A cycle of one full-scan mass spectrum
(400–1400 m/z) followed by 8 data-dependent MS/MS spectra at a 35% normalized collision
energy was repeated continuously throughout each step of the multidimensional separation.
Application of mass spectrometer scan functions and HPLC solvent gradients were controlled
by the Xcalibur datasystem.
As each step was executed, its spectra were recorded to a RAW file. This data was
then converted into .ms2 format through the use of RawXtract (Version 1.9). From the
.ms2 files, poor quality spectra were removed from the dataset using an automated
spectral quality assessment algorithm37. MS/MS spectra remaining after filtering were
searched with the SEQUEST™ algorithm38 against the NCBI RefSeq Human (04-23-2010)
protein database concatenated to a decoy database in which the sequence for each entry
in the original database was reversed25,39. All searches were parallelized and performed
on a Beowulf computer cluster consisting of 100 1.2 GHz Athlon CPUs40. No enzyme specificity
was considered for any search. SEQUEST results were assembled and filtered using the
DTASelect (version 2.0) program. DTASelect 2.0 uses a linear discriminant analysis
to dynamically set XCorr and DeltaCN thresholds for the entire dataset to achieve
a user-specified false positive rate. The false positive rates are estimated by the
program from the number and quality of spectral matches to the decoy database. The
hits detected uniquely in the GFP-progerin sample but not in GFP sample represent
proteins that are specifically associated with progerin, by either direct or indirect
interactions.
Co-Immunoprecipitation
BJ human fibroblasts were transduced with retrovirus encoding GFP-progerin, GFP-lamin
A or GFP, and maintained in culture for 72h. For immunoprecipitation, cells were lysed
in ice-cold lysis buffer (250 mM NaCl, 0.5% Triton X-100, 50 mM Tris, pH 7.5, 1 mM
EGTA, 1 mM EDTA, 10% glycerol, and complete protease inhibitor cocktail (Roche Diagnostics)).
Samples were briefly sonicated and immunoprecipitated by incubating with anti-GFP
agarose. The immunoprecipitates were washed extensively in lysis buffer, eluted in
SDS sample buffer, and subjected to immunoblotting.
Senescence-associated beta-galactosidase (SA-βgal) assay
SA-βgal assay were performed based as previously described methods41.
Measurement of telomere length
Genomic DNA was isolated from 1 × 106 cells. The telomere-specific oligonucleotide
probe (5'-TTAGGGTTAGGGTTAGGGTTAGGG-3'; ValueGene) was end-labeled using γ-32P-ATP
(MP Biomedicals) and T4 polynucleotide kinase (NEB). Two μg of genomic DNA for each
sample was digested with AluI (NEB) and MboI (NEB) and subjected to Southern analysis
with the telomere-specific probe. Mean telomere length was calculated from ΣODi/(ΣODi/ΣMWi).
ODi and MWi are optical density and molecular weight at a given position i, respectively.
Cell proliferation assay
Cell proliferation was determined with CellTiter 96® AQueous One Solution Cell Proliferation
Assay (MTS (3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium)), according to the
protocol provided by the manufacturer (Promega).
Statistical analysis
Results are presented as mean±s.d. Comparisons were performed with student's t-test
or one-way anova. p<0.05 was defined as statistically significant.
Supplementary Material
1
2
5
6
7
8
9