HomeLarge Type Edition
HOME ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow A correction has been published
Services
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
PubMed
Right arrow PubMed Citation
The Journals of Gerontology Series A: Biological Sciences and Medical Sciences 59:1234-1243 (2004)
© 2004 The Gerontological Society of America

Mapping Development-Related and Age-Related Chromatin Remodeling by a High Throughput ChIP-HPLC Approach

Valya R. Russanova, Tazuko H. Hirai, Andrei V. Tchernov and Bruce H. Howard

National Institute for Child Health and Human Development, National Institutes of Health, Bethesda, Maryland.

Address correspondence to Bruce Howard, MD, National Institute for Child Health and Human Development, NIH, Bldg. 31, Rm. 2A25, 6 Center Dr., Bethesda, MD 20892. E-mail: howard{at}helix.nih.gov


   Abstract
Top
 Abstract
Methods
 Results
 Discussion
 References
 
Common to numerous differentiation pathways in vertebrate organisms is the regulation of key genes through epigenetic mechanisms. Less well studied is to what extent cells of a given differentiation state, but examined at different points within the life history of an organism, are distinct at the level of the epigenome. A few instances of such variation have been reported, and it would be of considerable value to have at hand a means to characterize additional examples more efficiently. We describe an integrated approach to this task, and further present evidence for regions of age-related histone H4 acetylation change extending over tens to hundreds of kilobases. Broad similarity between two distinct regions of such change suggests a previously unsuspected link between developmental programs and aging.

A question of interest in contemporary gerontology is to what extent age-related change in epigenome structure influences the life histories of humans and other vertebrates. The origins of this question are diverse. Three decades ago, in experiments to create a mouse model for position effect variegation, Cattanach serendipitously discovered X-chromosome reactivation with age (1). Other instances consistent with age-related loss of mouse X-chromosome silencing were subsequently reported, including a 50-fold reactivation of the X-linked ornithine carbamoyl transferase gene (2–4). Related and interesting early lines of investigation explored entropy-driven disorder in higher order nuclear architecture with aging (5,6), dysdifferentiation (7), and alterations in DNA methylation (8,9).

Epigenetic states that specify either reduced or elevated transcription competence are well studied in yeast and Drosophila melanogaster (10–13). In these genetically tractable organisms, epigenetic control is dependent on histone modifications and histone variants, as well as interacting proteins that form heritable chromatin-based structures. The genomes of vertebrates and plants also reflect regulation by DNA methylation-dependent mechanisms (14); thus, to examine epigenome patterns fully in vertebrates, both DNA methylation-based and chromatin-based regulatory arms should be considered.

Global aspects of epigenetic control have been repeatedly investigated in the contexts of postnatal development and aging. Most early studies in mammals failed to reveal robust changes in chromatin composition or acetylation of histones (15); however, clear age-related increases are documentable in trimethylation of histone H4 at lysine 20 in rat tissues (16). Also, histone replacement variants are more abundant in the livers of adult than young mice (17). Small age-related losses of global DNA methylation can be documented in mouse and human tissues (18), and more striking decreases occur with passage in nonimmortalized cells (8). Interestingly, telomerase expression can increase total genomic 5-methyl CpG content (19).

Numerous publications describe locus-specific alterations in DNA methylation patterns with age. In some instances, experiments focus on one or a small number of CpG dinucleotide positions (20,21). More extensive age-related DNA methylation appears to occur within mouse ribosomal rRNA gene elements (proximal 5' spacer domains) (22). A series of papers by Heidmann and colleagues documents age-related activation of intracisternal A particle-associated transcripts with concomitant loss of DNA methylation in mouse liver (23). Particularly influential are observations, originally by Issa and Baylin, that age-related CpG methylation occurs at some CpG islands (24). In an excellent brief overview, Issa stresses the increasing mosaicism in DNA methylation patterns with age (25). This investigator emphasizes the stochastic nature of the process, estimating that the level of abnormality ranges from 5% to 50% (though most often near the lower value). Overall, such alterations in DNA methylation patterns appear to fall within the framework of cumulative age-related disorder.

In comparison with the body of work on DNA methylation, little is known at the molecular level concerning age-related chromatin changes in vertebrates. This contrasts with the yeast model of replicative senescence, where gene expression changes consistent with chromatin remodeling have been documented (26). Likewise, progress has been rapid in delineating differentiation-associated remodeling (27–34), where genes encoding specialized differentiation functions provide a natural target for study. The task of finding development-related and age-related chromatin alterations is challenging, in part because such processes may be subtle and occur within seemingly unchanging cell states—in designing experimental approaches, it is simply not obvious in what gene-regulatory or gene-encoding regions to look. One strategy is to search for large-scale patterns by mapping microarray-based RNA data onto the genome (35). In previous work, we developed another solution to the epigenome search problem (36). A semirandom genome sampling/chromatin immunoprecipitation (gsChIP) approach was devised, validated, and applied to screen for differences in the histone H4 acetylation epigenome patterns of human fibroblasts. We looked for changes either related to in vitro passage or donor age. The searches revealed several loci subject to remodeling, including one located about 11 Mb from the 4p chromosome terminus and a second about 3 Kb from the 4q terminus.

An immediate question that arises on finding such loci is how far along the chromosome the observed acetylation changes might extend. To map a single region in two cell types is not difficult. However, if one envisages a systematic effort along these lines, it will entail the mapping of dozens to hundreds or more regions in numerous cell types, in multiple organisms, at various stages of aging, and so forth. A second major problem thus emerges, i.e., how to map flanking (or query potentially similar noncontiguous) regions efficiently in terms of time and effort. Mapping involves multiple steps, including primer design, testing of primers to verify that yields of a single product reflect template concentration, redundant amplification reactions from ChIP-derived genomic DNA fragments, and finally graphic display of the results for a substantial number of positions. We report here an integrated solution to these issues. Further, we propose from the results that new and unexpected links between development and aging may have been identified.


   METHODS
Top
 Abstract
 Methods
Results
 Discussion
 References
 
Cell Culture
HSC172 human lung embryo fibroblasts (37) were a gift from Sam Goldstein [formerly at the University of Arkansas]. Normal human skin fibroblasts from donors of different ages were obtained from the Coriell collection. Fetal cells used were: AG04392A, 16 weeks, population doubling level (PDL) 17; AG04449, 12 weeks, PDL 11; AG04431B, 15 weeks, PDL 18; AG04451, 16 weeks, PDL 13; AG04525, 17 weeks, PDL17. Newborn foreskin cells used were: AG01437B, 3 days, PDL 17; AG01440, 3 days, PDL 5. Cells from toddlers (very young) were: AG08498, 1 year, PDL 4; AG07095A, 2 years, PDL 11. Cells from young adults were: AG07720B, 23.9 years, PDL 16; AG07719A, 27.9 years, PDL 16; AG04441B, 29.4 years, PDL 16; AG13153, 30.2 years, PDL 5. Cells from old adults were: AG05274B, 83.4 years, PDL 14; AG13077, 84.6 years, PDL 5; AG09557, 83.5 years, PDL 9. Cells were grown in Dulbecco's modified Eagle medium (Gibco/BRL, Carlsbad, CA) supplemented with 10% fetal calf serum (FCS) or Eagle's MEM with Earle's salts with 15% FSC, 2 mM glutamine, and antibiotics (50 U of penicillin/ml and 50 mg of streptomycin/ml). Cultures from the Coriell AG collection were expanded for no more than several passages, the minimum number to obtain sufficient material for chromatin preparations (5–15 x 106).

Preparation of Nuclei and Chromatin
Nuclei and histone-H1 depleted chromatin were prepared by a procedure described previously (38). Micrococcal nuclease concentrations were titrated in order to obtain chromatin fragments consisting predominantly of mononucleosomes, dinucleosomes, and trinucleosomes. All solutions used starting from cell trypsinization through preparation of nuclei and chromatin contained 10 mM sodium butyrate to preserve histone acetylation.

Immunoprecipitation of Acetylated Chromatin
Antibody against acetylated H4 was prepared by immunizing rabbits with chemically acetylated histone H4 (39,40). Affinity purification was accomplished with acetylated BSA coupled to BrCN Sepharose and characterized by enzyme-linked immunosorbent assay (ELISA) and Western blotting as described previously (38). Affinity-purified antibody (70 µg) was incubated with magnetic beads coated with antirabbit immunoglobulin (IgG) (Dynal Biotech, Brown Deer, WI) as described by the manufacturer. Chromatin fragments (25 µg) were incubated with antibody-coated beads by rotating overnight at 4°C in 50 mM NaCl, 10 mM Tris-HCl pH 8.0, 1 mM EDTA, 10 mM Na-butyrate, and proteinase inhibitor cocktail (39,40). Supernatant from the first magnetic separation combined with the first wash was taken as the nonbound chromatin fraction. Beads were washed six times with 1 ml 150 mM NaCl, 10 mM Tris-HCl pH 8.0, 1 mM EDTA, 10 mM Na-butyrate, proteinase inhibitor cocktail (Roche Diagnostics, Indianapolis, IN). Samples were transferred to new tubes prior to a final wash. Bound DNA was recovered directly from the beads using Qiagen PCR (polymerase chain reaction) purification kits (Valencia, CA). DNA concentrations were determined fluorimetrically. The typical yield from 25 µg chromatin was 0.5–1 µg bound DNA.

PCR Amplification
Primer design across RepeatMasker (Smit, AFA & Green, P RepeatMasker at http://repeatmasker.org)-processed sequences was accomplished using an interface utility designed to work with Primer3 (41). Primers were obtained from Integrated DNA Technologies (Coralville, IA) in a normalized 96-well format. PCR amplification was performed in triplicate in a 20 µl reaction mix containing 0.5–1 ng DNA, 1X buffer (Perkin-Elmer, Wellesley, MA), 50 µM deoxynucleotides, 0.5 µM primers, 1 unit Taq polymerase (Boehringer-Mannheim/Roche). The reactions were set up using an automated liquid handling system (Multiprobe II HT, Perkin-Elmer) in 96-well plates (ABgene, Inc., Rochester, NY). The cycling parameters were: 5 minutes at 94°C, followed by 30 cycles of 30 seconds at 94°C, 30 seconds at the annealing temperature, and 30 seconds at 72°C. As a check for linearity, reactions containing 1.5, 0.5, or 0.17 ng DNA were performed initially for each primer pair. PCR products were fractionated by ion-pair reverse-phase high-pressure liquid chromatography (HPLC) on a polystyrene-divinylbenzene matrix at 50°C and detected by SYBR gold (WAVE system, Transgenomics, Inc., Omaha, NE). Gradients started at 56% buffer A (0.1 M triethylammonium acetate [TEAA]), 44% buffer B (0.1 M TEAA, 25% acetonitrile), and finished at 40.9% A, 59.1%B.

Real-time PCR was performed using SYBR Green Fast-start PCR kits and a LightCycler instrument (Roche). Each sample (0.5–1 ng) was analyzed in triplicate together with 0.1–3 ng input DNA used for standard curve calculation.

Computational Analysis
Bioinformatics software utilized included Primer3 (41), RepeatMasker, Tandem Repeats Finder (42), and TESS [TESS: Transcription Element Search Software on the WWW, Jonathan Schug and G. Christian Overton, Technical Report CBIL-TR-1997-1001-v0.0, Computational Biology and Informatics Laboratory, School of Medicine, University of Pennsylvania, 1997, URL: http://www.cbil.upenn.edu/tess]. Utilities developed in the course of this study provide a primer design interface for Primer3, generation of scalar vector graphics (SVG)-based virtual gel images, quantification of peak heights and associated standard deviations, and red-green-blue (RGB) graphic display of chromatin acetylation patterns [code available on request]. This study utilized the high-performance computational capabilities of the Biowulf PC/Linux cluster at the National Institutes of Health, Bethesda, MD (http://biowulf.nih.gov).


   RESULTS
Top
 Abstract
 Methods
 Results
Discussion
 References
 
In experiments that preceded this study, the gsChIP approach was applied to human skin fibroblasts derived from fetal and foreskin sources, as well as skin biopsies taken from children and adults of various ages. Chromatin from several individuals was typically pooled to minimize the possibility that histone H4 acetylation differences detected would reflect individual variation rather than age-related epigenome change. One of the first candidate sites identified, h9, is located in the chromosome 4p16.1 region, centered around position 10,967,271 in human genome build 34. Acetylation at this site was found to be lower in adult than fetal skin fibroblasts harvested at similar passage levels (36). To further characterize the nature of the apparent age-related change at h9, it was decided to map its flanking sequences and thus gain a rough estimate of the epigenome region involved.

Rapid ChIP Analysis Using an HPLC-Based Approach
For primer design and testing, the strategy adopted was to evaluate numerous primer pairs and to discard those that failed to yield informative data without optimization. Primer design across a RepeatMasker-processed region of interest was accomplished using a console-based Unix utility that constructs primer pair lists by means of consecutive calls to a locally installed copy of Primer3. Software-generated output lists were suitably formatted for direct ordering of plate arrays containing normalized primer pairs. PCR reactions were set up using an automated liquid handling system and run in a 96-well format. Products of the reactions were sampled directly and fractionated by ion-pair reverse-phase HPLC on a poly (styrene-divinylbenzene) matrix, with detection by postfractionation staining with SYBR gold. Notably, the sensitivity of this system permits PCR amplifications to be minimized so as to remain in or close to the linear range. XML-format result files were exported and processed using a locally written utility to both quantify the results and convert them to an SVG-based virtual gel format (see Figure 1A).



View larger version (31K):
[in this window]
[in a new window]
  		Figure 1. Virtual gel images of high-pressure liquid chromatography (HPLC)-fractionated polymerase chain reaction (PCR) products. A: Verification of primer pair products. Partial scalar vector graphic (SVG) image is shown. Numbers above lanes (A01, A02, ...) represent positions of PCRs done in a 96-well plate format. Numbers to left of lanes indicate primer pair candidates in the series being tested. B: Comparison of chromatin fractions from fetal and young adult fibroblasts. Original 96-lane SVG image is shown, with the exception that header (fetal vs adult, B [bound] vs NB [nonbound]) and primer pair annotations were added using a general purpose graphics program. Rows 01–03, bound fetal; rows 04–06, nonbound fetal; rows 07–09, bound adult; rows 10–12, nonbound adult. Sources of fetal and adult cells described in Methods

 
The above approach was implemented initially to construct a map of histone H4 acetylation patterns flanking the h9 locus. Triplicate PCR reactions were run on bound and nonbound ChIP fractions comparing fetal and adult fibroblasts in parallel. Figure 1B shows a representative SVG image. Real-time PCR assays, which were done for selected comparisons, verified the reliability of quantification by HPLC as described here (unpublished observations). Control ChIP experiments with nonimmune IgG confirmed that insignificant amounts of DNA were recovered in the absence of specific antibody (not shown). Rapid accumulation of results requires a simply constructed and easily interpreted graphics display format. This was accomplished using algorithms that represent acetylation levels, or differences in acetylation levels, according to arbitrarily chosen color scales (red-gray-blue and red-yellow-green for high vs low and increasing vs decreasing acetylation levels, respectively). In the former scale, gray corresponds to average epigenome histone H4 acetylation as determined by both chromatin-associated DNA recovery and multiple gsChIP comparisons.

A summary of the first-stage mapping results is provided in Figure 2A. As shown at the upper right of this figure, the region analyzed spans 2.15 Mb. A central subregion, denoted (b), is mapped over approximately 75 Kb in either direction flanking the original h9 locus, itself indicated by a black triangle. Fetal to adult ratios reach maximum values of 4.1 and 4.6 at two sites flanking h9 (gray triangles). Two more distal 1 Mb flanking regions, denoted (a) and (c), are sampled at intervals of about 40 Kb or greater. In the direction of the centromere, only minor fetal versus adult differences in acetylation are evident, whereas an extended region of acetylation difference is seen in the telomere-proximal direction.



View larger version (41K):
[in this window]
[in a new window]
  		Figure 2. Summary of chromatin mapping results. A: Continuous 2.15 Mb sequence depicted on three consecutive lines: (a), 1 Mb telomere-proximal region; (b), 150 Kb central region mapped in greater detail; (c), 1 Mb centromere-proximal region. See schematic diagram at upper right. Black vertical lines above central mapped region depict positions of primer pairs used. Vertical bars in sampled regions are not to scale. Color keys are shown at center right. For each region, red-gray-blue upper bar denotes relative histone acetylation levels, while lower red-yellow-green bar indicates change from fetal to adult pattern. Black triangle designates the position of the original h9 locus. Two gray triangles and asterisk show positions examined at multiple ages (see
Figure 3 and text). In control experiments, the mean of ratio values, represented by a set of five randomly selected chromosome positions, was 1.1 ± 0.2 (not shown). B: Summary of mapping results for subregion overlapping the MIST (mast cell immunoreceptor signal transducer) gene (see gray bar above telomere-proximal sampled sequence in (A), as well as red segment in schematic diagram at upper right). Note that the lower color key scale is adjusted to reflect greater changes in acetylation levels

 
Secondary mapping was done upstream from and overlapping the MIST gene (mast cell immunoreceptor signal transducer; LOC166522), where data from three adjacent sampled sites indicated a subregion of elevated histone acetylation in fetal fibroblasts and, in contrast, reduced acetylation in adult fibroblasts. The subregion examined, bounded by these sampled sites, is indicated by the gray bar at the top of Figure 2A. The results reveal further substructure in the remodeling process, with differences reaching over 10-fold at two positions (Figure 2B).

Timeframe for Chromatin Change at a Position Close to the h9 Site
Several loci exhibiting relatively large acetylation changes were examined further to relate these in an approximate way to developmental stages. Chromatin was prepared from skin fibroblasts at comparable low passage levels and from the following sources: additional fetal (3 samples), neonatal foreskin (2 samples), very young [ages 1 and 2 yr; (2 samples)], and young adult [ages 24 to 30 yr; (4 samples)]. Two positions closely flanking the original h9 locus (denoted by grey triangles in Figure 2A) displayed moderate acetylation only in fetal samples, with unchanging low acetylation in other cells (not shown). One position 85 Kb upstream of the MIST gene—indicated by asterisks in Figures 2A, 2B, and 3—exhibited a more gradual decrease in acetylation. In this case, baseline low acetylation levels appear to be reached by or just after the period of early childhood (Figure 3). No further acetylation change was observed on comparison of samples from young and old adults (unpublished observations).



View larger version (22K):
[in this window]
[in a new window]
  		Figure 3. Fetal-related and childhood-related chromatin remodeling. Bound/nonbound (B/NB) ratios for chromatin fractions derived from fetal (F), newborn foreskin (FS), toddler/very young (T), and adult (A) fibroblasts. Position examined is indicated by asterisks in telomere-proximal region in Figures 2A and 2B. Shown are averages obtained from three fetal, two newborn, two toddler, and five adult cultures

 
Mapping Chromatin Differences Between Young and Old Adults
An independent series of gsChIP experiments was initiated to compare chromatin preparations from preadolescent-derived and old donor-derived fibroblasts (36). This led to the discovery of another site of interest, termed here p14, in the 4q35.2 region (position 188,927,735 in human genome build 34). As in the case of h9, the observed difference suggests a decrease in histone H4 acetylation with age, and this prompted us to investigate the flanking regions. In order to more clearly define the time frame of apparent change, fibroblasts from young and old adults were used in the follow-up studies. Mapping over 30 Kb centered around p14 revealed differences predominantly in the direction of the telomere, so the adjacent telomere-proximal 30 Kb sequence and an additional 1 Mb were examined.

While further mapping remains to be done, a summary of the results is assembled in Figure 4. Near the centromere-proximal edge of the region, an area of high histone H4 acetylation is evident. As this is not associated with a CpG island or an identified gene, it could be an insulator, or have another function not yet discernible. Similarly to the 4p16.1 region, patchy subregions of acetylation change are seen over a near megabase-size chromosome segment. Young versus old adult ratio values are indicated by empty triangles at selected positions. There appears to be a trend toward higher ratio values as mapping and sampling proceed in the direction of the telomere. A histone H4 acetylation ratio of 4.3 near LOC389250, a hypothetical gene supported by expressed sequence tags (ESTs), is evident, and, like the subregion upstream from the MIST gene, will be of interest to examine in greater detail.



View larger version (31K):
[in this window]
[in a new window]
  		Figure 4. Chromatin mapping: comparisons of epigenome patterns in young versus old adults. Continuous 1.06 Mb region depicted on two consecutive lines. Notations are as in
Figure 2

 

   DISCUSSION
Top
 Abstract
 Methods
 Results
 Discussion
References
 
The results reported here focus on how epigenome patterns may change during development and aging. They are the first, to our knowledge, to describe regions of age-related chromatin remodeling that extend over multikilobase to near-megabase distances and span single-copy genes. One region, positioned approximately 11 Mb from the chromosome 4p terminus, exhibits diminishing H4 acetylation over a time interval spanning fetal development to early childhood. A second, less than 3 Mb from the 4q terminus, is evident on comparison between young and old adults. Despite marked time frame differences, the chromatin changes in these two regions are broadly similar. In this regard, we note an earlier proposal that the discovery of such similarity would constitute a first step toward linking programmed development and the aging process (43).

Mapping and sampling data for the 4p16.1 region extend over a distance of about 2 Mb. Most histone H4 acetylation changes occur within a roughly 0.7 Mb subregion delimited by clusters of 5 telomere-proximal and 12 centromere-proximal positions that exhibit little or no remodeling. Two pairs of genes are situated near or overlapping the edges of the 0.7 Mb subregion. Of particular interest is the MIST gene (LOC166522), since the most striking remodeling spans the 5' part of the gene and extends through the promoter region. MIST/Clnk and related proteins are adaptors involved in signal transduction, mostly studied in hematopoietic lineages (44,45). A murine homologue of the MIST gene is expressed in embryonic stem cells, but reverse transcriptase (RT)-PCR analysis revealed no expression in fetal fibroblasts (unpublished observations).

Yet to be determined are what sequence features in the MIST upstream region contribute to its striking age-related chromatin remodeling. As a first step to explore this issue, we generated a 160 Kb subregion map, and this revealed three more discrete areas where the magnitude of the histone H4 acetylation decrease reaches 8-fold to 11-fold. Each peak area of change still spans 8 Kb or more, so a third round submapping step should be informative. Possibly significant is the fact that the MIST promoter region is devoid of CpG islands. The nearest islands are located between this gene and KIAA1729 (near the 5' end of the latter), and close to the 5' end of LOC153027. Most gene-associated CpG islands are highly acetylated, and it is plausible that large islands might stabilize neighboring chromatin domains against development-related or age-related chromatin remodeling. Consistent with this, drift in CpG island methylation patterns is typically stochastic and occurs at low levels (25).

In mapping from the MIST gene toward the 4p telomere, a span of about 240 Kb is encountered, which exhibits minimal histone H4 acetylation change. At the most telomere-proximal position sampled, however, decreasing acetylation is again seen. Could this be the edge of another domain of chromatin remodeling? Imprinted domains with opposite maternal–paternal polarity are, in some instances, clustered into much larger regions. Quite possibly, the same may be true for the remodeling domains of the type described here. A very unusual tandem array of 4.7 Kb repeats, each copy of which contains a deubiquitinating enzyme 1A-like gene, is located approximately 0.8 Mb telomere-proximal to the edge of the presently mapped 4p16.1 domain. This large array merits comment, as tandem arrays are involved in several examples of striking age-related epigenetic change. At least two reports describe progressive DNA methylation within the tandemly arrayed ribosomal RNA genes (22,46). Another documents strong age-related silencing of transgene repeat arrays (47).

At the other end of the same chromosome, the 4q35.2 remodeling region has been subjected to preliminary mapping. While the boundaries of acetylation change are yet to be determined, the ZFP42 gene and several additional uncharacterized transcription units fall within the region already examined. Three of the transcription units are associated with minimal CpG islands. Several tandem arrays are to be found, including one located centromere-proximal at a distance of about 340 Kb, and a second located less than 50 Kb from the 4q telomere. Deletions of the latter array are clearly implicated in the etiology of facioscapulohumeral muscular dystrophy (48). The proximity of relatively large tandem arrays to both 4p16.1 and 4q35.2 remodeling regions may be coincidental; nevertheless, it is conceivable that silencing mechanisms acting on such arrays may exert clock-like influences on chromatin states of adjacent single-copy regions.

Beyond the above two domains, are there numerous other genome regions subject to age-related chromatin remodeling? And, if so, do any regions extend across and influence critical genes—i.e., those already linked to important diseases and pathophysiologies of human aging? This remains to be seen, but using the techniques described here, it should be straightforward to search for and map other examples. At the current basic level of automation, mapping across regions of megabase size can be accomplished inexpensively (relative to microarrays) and in a span of several days. Simple expedients (e.g., fractionation of bands from multiple PCR amplification reactions in a single column run), as well as further HPLC development and more-sensitive staining-detection technologies should increase throughput considerably. Moreover, when full genome-based microarrays become widely accessible, this approach should be of use for fine mapping of bacterial artificial chromosomes (BAC) or larger size domains identified by hybridization.

What significance should be ascribed to age-related changes in large chromatin domains? At this point, we cannot exclude the possibility that such chromatin remodeling is an epiphenomenon, of little or no consequence in the aging process. On the other hand, the remodeling of epigenetic patterns is a topic of broad interest relevant to both differentiation-associated gene regulation and the genome reprogramming required for successful animal cloning experiments. Further, it has been hypothesized that age-related change in chromatin-based epigenetic states could serve as one of the basic driving forces for aging (49,50). The results reported in this study are predicted by, and appear to provide important new experimental support for, that hypothesis.

Finally, while this article is focused solely on pattern changes in the epigenome, it is important to emphasize that the links between epigenetic mechanisms and aging are likely to be numerous and intricate. Histone-modifying and transcription factor-modifying enzymes are typically embedded in large (and presumably multifunctional) complexes. Disturbances in the functions of these complexes or DNA methylation mechanisms modulate cell growth and senescence, mortality pathways, and organismal aging (51–56). While such disturbances may prove to accelerate region-specific alterations in epigenome patterns, links can likewise be postulated to multiple damage-related pathways, metabolism, and other modulators of aging.


   Acknowledgments
 
We thank D. Landsman (NCBI/NIH), J. A. Epstein (NICHD/NIH), B. K. Lee (NCI/NIH), and S. Hornyak (NICHD/NIH) for information and suggestions regarding computational biology resources. We are grateful to Joseph Breen (Transgenomics, Inc.) for information on primer synthesis options, and V. J. Cristofalo (Lankenau Institute) for advice concerning human fibroblast resources.


   Footnotes
 
Decision Editor: James R. Smith, Jr.

Received August 3, 2004

Accepted October 1, 2004


   References
Top
 Abstract
 Methods
 Results
 Discussion
 References
 

  1. Cattanach BM. Position effect variegation in the mouse. Genet Res. 1974;23:291-306.[Medline]
  2. Wareham KA, Lyon MF, Glenister PH, Williams ED. Age related reactivation of an X-linked gene. Nature. 1987;327:725-727.[Medline]
  3. Gartler SM, Goldman MA. Reactivation of inactive X-linked genes. Dev Genet. 1994;15:504-514.[Medline]
  4. Bennett-Baker PE, Wilkowski J, Burke DT. Age-associated activation of epigenetically repressed genes in the mouse. Genetics. 2003;165:2055-2062.[Abstract/Free Full Text]
  5. Puvion-Dutilleul F, Macieira-Coelho A. Aging dependent nucleolar and chromatin changes in cultivated fibroblasts. Cell Biol Int Rep. 1983;7:61-71.[Medline]
  6. Macieira-Coelho A. Chromatin reorganization during senescence of proliferating cells. Mutat Res. 1991;256:81-104.[Medline]
  7. Cutler RG. The dysdifferentiation hypothesis of mammalian aging and longevity. In: Giacobini E, Filogamo F, Giacobini G, Vernadakis VA, eds. The Aging Brain: Cellular and Molecular Mechanisms of Aging in the Nervous System. New York: Raven Press; 1982:1–19.
  8. Wilson VL, Jones PA. DNA methylation decreases in aging but not in immortal cells. Science. 1983;220:1055-1057.[Abstract/Free Full Text]
  9. Holliday R. Understanding Ageing. Cambridge: Cambridge University Press; 1995.
  10. Kelley RL, Kuroda MI. The role of chromosomal RNAs in marking the X for dosage compensation. Curr Opin Genet Dev. 2000;10:555-561.[Medline]
  11. Richards EJ, Elgin SC. Epigenetic codes for heterochromatin formation and silencing: rounding up the usual suspects. Cell. 2002;108:489-500.[Medline]
  12. Rusche LN, Kirchmaier AL, Rine J. The establishment, inheritance, and function of silenced chromatin in Saccharomyces cerevisiae. Annu Rev Biochem. 2003;72:481-516.[Medline]
  13. Malik HS, Henikoff S. Phylogenomics of the nucleosome. Nat Struct Biol. 2003;10:882-891.[Medline]
  14. Bestor TH. The DNA methyltransferases of mammals. Hum Mol Genet. 2000;9:2395-2402.[Abstract/Free Full Text]
  15. Finch CE. Longevity, Senescence, and the Genome. Chicago: University of Chicago Press; 1990.
  16. Sarg B, Koutzamani E, Helliger W, Rundquist I, Lindner HH. Postsynthetic trimethylation of histone H4 at lysine 20 in mammalian tissues is associated with aging. J Biol Chem. 2002;277:39195-39201.[Abstract/Free Full Text]
  17. Russanova V, Stephanova E, Pashev I, Tsanev R. Histone variants in mouse centromeric chromatin. Mol Cell Biochem. 1989;90:1-7.[Medline]
  18. Wilson VL, Smith RA, Ma S, Cutler RG. Genomic 5-methyldeoxycytidine decreases with age. J Biol Chem. 1987;262:9948-9951.[Abstract/Free Full Text]
  19. Young JI, Sedivy JM, Smith JR. Telomerase expression in normal human fibroblasts stabilizes DNA 5-methylcytosine transferase I. J Biol Chem. 2003;278:19904-19908.[Abstract/Free Full Text]
  20. Issa JJ, Vertino PM, Boehm CD, Newsham IF, Baylin SB. Switch from monoallelic to biallelic human IGF2 promoter methylation during aging and carcinogenesis. Proc Natl Acad Sci U S A. 1996;93:11757-11762.[Abstract/Free Full Text]
  21. Ono T, Uehara Y, Kurishita A, Tawa R, Sakurai H. Biological significance of DNA methylation in the ageing process. Age Ageing. 1993;22:S34-S43.
  22. Swisshelm K, Disteche CM, Thorvaldsen J, Nelson A, Salk D. Age-related increase in methylation of ribosomal genes and inactivation of chromosome-specific rRNA gene clusters in mouse. Mutat Res. 1990;237:131-146.[Medline]
  23. Barbot W, Dupressoir A, Lazar V, Heidmann T. Epigenetic regulation of an IAP retrotransposon in the aging mouse: progressive demethylation and de-silencing of the element by its repetitive induction. Nucleic Acids Res. 2002;30:2365-2373.[Abstract/Free Full Text]
  24. Issa J-PJ, Ottaviano YL, Celano P, et al. Methylation of the oestrogen receptor CpG island links ageing and neoplasia in human colon. Nat Genet. 1994;7:536-540.[Medline]
  25. Issa JP. Living longer: the aging epigenome. In: Beck S, Olek A, eds. The Epigenome. Weinheim: Wiley-VCH; 2003:141–149.
  26. Kim S, Villeponteau B, Jazwinski SM. Effect of replicative age on transcriptional silencing near telomeres in Saccharomyces cerevisiae. Biochem Biophys Res Commun. 1996;219:370-376.[Medline]
  27. Hebbes TR, Clayton AL, Thorne AW, Crane-Robinson C. Core histone hyperacetylation co-maps with generalized DNase I sensitivity in the chicken beta-globin chromosomal domain. EMBO J. 1994;13:1823-1830.[Medline]
  28. Litt MD, Simpson M, Gaszner M, David Alis C, Felsenfeld G. Correlation Between histone lysine methylation and developmental changes at the chicken beta-globin locus. Science. 2001;293:2453-2455.[Abstract/Free Full Text]
  29. Turner BM. Cellular memory and the histone code. Cell. 2002;111:285-291.[Medline]
  30. Felsenfeld G, Groudine M. Controlling the double helix. Nature. 2003;421:448-453.[Medline]
  31. Bulger M, Sawado T, Schubeler D, Groudine M. ChIPs of the beta-globin locus: unraveling gene regulation within an active domain. Curr Opin Genet Dev. 2002;12:170-7.[Medline]
  32. Myers FA, Evans DR, Clayton AL, Thorne AW, Crane-Robinson C. Targeted and extended acetylation of histones H4 and H3 at active and inactive genes in chicken embryo erythrocytes. J Biol Chem. 2001;276:20197-20205.[Abstract/Free Full Text]
  33. O'Neill LP, Turner BM. Histone H4 acetylation distinguishes coding regions of the human genome from heterochromatin in a differentiation-dependent but transcription-independent manner. EMBO J. 1995;14:3946-3957.[Medline]
  34. Zhou W, Chang S, Aune TM. Long-range histone acetylation of the Ifng gene is an essential feature of T cell differentiation. Proc Natl Acad Sci U S A. 2004;101:2440-2445.[Abstract/Free Full Text]
  35. Zhang H, Pan KH, Cohen SN. Senescence-specific gene expression fingerprints reveal cell-type-dependent physical clustering of up-regulated chromosomal loci. Proc Natl Acad Sci U S A. 2003;100:3251-3256.[Abstract/Free Full Text]
  36. Russanova VR, Hirai TH, Howard BH. Semirandom sampling to detect differentiation-related and age-related epigenome remodeling. J Gerontol Biol Sci. 2004;59A:1221-1233.
  37. Gupta RS, Goldstein S. Diphtheria toxin resistance in human fibroblast cell strains from normal and cancer-prone individuals. Mutat Res. 1980;73:331-338.[Medline]
  38. Russanova VR, Driscoll CT, Howard BH. Adenovirus type 2 preferentially stimulates polymerase III transcription of Alu elements by relieving repression: a potential role for chromatin. Mol Cell Biol. 1995;15:4282-4290.[Abstract]
  39. Mutskov VJ, Russanova VR, Dimitrov SI, Pashev IG. Histones associated with non-nucleosomal rat ribosomal genes are acetylated while those bound to nucleosome-organized gene copies are not. J Biol Chem. 1996;271:11852-11857.[Abstract/Free Full Text]
  40. Hebbes TR, Thorne AW, Crane-Robinson C. A direct link between core histone acetylation and transcriptionally active chromatin. EMBO J. 1988;7:1395-1402.[Medline]
  41. Rozen S, Skaletsky H. Primer3 on the WWW for general users and for biologist programmers. In: Krawetz S, Misener S, eds. Bioinformatics Methods and Protocols: Methods in Molecular Biology. Totowa, NJ: Humana Press; 2000:365–386.
  42. Benson G. Tandem repeats finder: a program to analyze DNA sequences. Nucl Acids Res. 1999;27:573-580.[Abstract/Free Full Text]
  43. Russell RL. Evidence for and against the theory of developmentally programmed aging. In: Warner HR, Butler RN, Sprott RL, Schneider EL, eds. Modern Biological Theories of Aging. New York: Raven Press; 1987:35–61.
  44. Cao MY, Davidson D, Yu J, Latour S, Veillette A. Clnk, a novel SLP-76-related adaptor molecule expressed in cytokine-stimulated hematopoietic cells. J Exp Med. 1999;190:1527-1534.[Abstract/Free Full Text]
  45. Fujii Y, Wakahara S, Nakao T, et al. Targeting of MIST to Src-family kinases via SKAP55-SLAP-130 adaptor complex in mast cells. FEBS Lett. 2003;540:111-116.[Medline]
  46. Oakes CC, Smiraglia DJ, Plass C, Trasler JM, Robaire B. Aging results in hypermethylation of ribosomal DNA in sperm and liver of male rats. Proc Natl Acad Sci U S A. 2003;100:1775-1780.[Abstract/Free Full Text]
  47. Robertson G, Garrick D, Wilson M, Martin DIK, Whitelaw E. Age-dependent silencing of globin transgenes in the mouse. Nucleic Acids Res. 1996;24:1465-1471.[Abstract/Free Full Text]
  48. Clapp J, Bolland DJ, Hewitt JE. Genomic analysis of facioscapulohumeral muscular dystrophy. Brief Funct Genomic Proteomic. 2003;2:213-223.[Abstract/Free Full Text]
  49. Howard BH. Replicative senescence: considerations relating to the stability of heterochromatin domains. Exp Gerontol. 1996;31:281-293.[Medline]
  50. Howard BH. Chromatin, Aging, and Cellular Senescence. In: Hisama FM, Weissman SM, Martin GM, eds. Chromosomal Instability and Aging. New York: Marcel Dekker; 2003:149–166.
  51. Young J, Smith JR. Epigenetic aspects of cellular senescence. Exp Gerontol. 2000;35:23-32.[Medline]
  52. Tominaga K, Olgun A, Smith JR, Pereira-Smith OM. Genetics of cellular senescence. Mech Ageing Dev. 2002;123:927-936.[Medline]
  53. Bandyopadhyay D, Medrano EE. The emerging role of epigenetics in cellular and organismal aging. Exp Gerontol. 2003;38:1299-1307.[Medline]
  54. Sun LQ, Lee DW, Zhang Q, et al. Growth retardation and premature aging phenotypes in mice with disruption of the SNF2-like gene, PASG. Genes Dev. 2004;18:1035-1046.[Abstract/Free Full Text]
  55. Jiang JC, Wawryn J, Shantha Kumara HM, Jazwinski SM. Distinct roles of processes modulated by histone deacetylases Rpd3p, Hda1p, and Sir2p in life extension by caloric restriction in yeast. Exp Gerontol. 2002;37:1023-1030.[Medline]
  56. Jazwinski SM. Growing old: metabolic control and yeast aging. Annu Rev Microbiol. 2002;56:769-792.[Medline]



This article has been cited by other articles:


Home page
 PhysiologyHome page
J. P. de Magalhaes and G. M. Church
Genomes Optimize Reproduction: Aging as a Consequence of the Developmental Program
Physiology, August 1, 2005; 20(4): 252 - 259.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow A correction has been published
Services
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
PubMed
Right arrow PubMed Citation

HOME ARCHIVE SEARCH TABLE OF CONTENTS