The JI
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     
 

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Rouhi, A.
Right arrow Articles by Mager, D. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Rouhi, A.
Right arrow Articles by Mager, D. L.
The Journal of Immunology, 2006, 176: 2991-2999.
Copyright © 2006 by The American Association of Immunologists

Evidence for Epigenetic Maintenance of Ly49a Monoallelic Gene Expression1

Arefeh Rouhi*,{dagger}, Liane Gagnier*, Fumio Takei*,{ddagger} and Dixie L. Mager2,*,{dagger}

* The Terry Fox Laboratory, British Columbia Cancer Agency, Vancouver, British Columbia, Canada; and {dagger} Department of Medical Genetics and {ddagger} Department of Pathology and Laboratory Medicine, University of British Columbia, Vancouver, British Columbia, Canada


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
Although structurally unrelated, the human killer cell Ig-like (KIR) genes and the rodent lectin-like Ly49 genes serve similar functional roles in NK cells. Moreover, both gene families display variegated, monoallelic expression patterns established at the transcriptional level. DNA methylation has been shown to play an important role in maintenance of expression patterns of KIR genes, which have CpG island promoters. The potential role of DNA methylation in expression of Ly49 genes, which have CpG-poor promoters, is unknown. In this study, we show that hypomethylation of the region encompassing the Pro-2 promoter of Ly49a and Ly49c in primary C57BL/6 NK cells correlates with expression of the gene. Using C57BL/6 x BALB/c F1 hybrid mice, we demonstrate that the expressed allele of Ly49a is hypomethylated while the nonexpressed allele is heavily methylated, indicating a role for epigenetics in maintaining monoallelic Ly49 gene expression. Furthermore, the Ly49a Pro-2 region is heavily methylated in fetal NK cells but variably methylated in nonlymphoid tissues. Finally, in apparent contrast to the KIR genes, we show that DNA methylation and the histone acetylation state of the Pro-2 region are strictly linked with Ly49a expression status.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
Natural killer cells are large granular lymphocytes that are part of the innate immune system. They exert their function primarily through the detection of changes in expression of MHC class I molecules on target cells, mediated via families of receptors on NK cells (1). In primates and some other mammals, the killer cell Ig-like (KIR) genes3 provide this function (1, 2). However, in rodents the Ly49 family is the main receptor gene family for the recognition of classical MHC class I (3, 4). A third and smaller receptor family common to possibly all mammals is the CD94/NKG2 family that recognizes nonclassical MHC class I (3, 5). Although the KIR and Ly49 genes are not orthologs, they share many commonalties. Both gene families are composed of closely related genes with unidirectional transcription and monoallelic yet independent and stochastic expression (5, 6, 7). This unique pattern of gene expression is established at the transcriptional level (7, 8).

Transcriptional repression can generally be achieved by limitation of transcription factors and/or by inaccessibility of regulatory sequences such as the promoter and enhancer. An emerging theme for transcriptional control of a number of gene families and tissue-specific genes is the role of epigenetics. Chromatin remodeling through histone acetylation/methylation and DNA methylation is one of the major mechanisms of transcriptional control for tissue-specific genes and the main mechanism for allele-specific gene expression (9). The olfactory genes, that number more than a thousand, have variegated monoallelic transcription just like the KIR and Ly49 genes. It is believed that maintenance of the stable "one allele per one neuron" expression pattern of the olfactory receptors is via epigenetic mechanisms (10, 11, 12). The expression of IL-2 was shown to correlate with the DNA methylation pattern of its promoter in naive and activated T cells (13). This correlation of transcription and epigenetic states seems to be a common feature of immune system genes (14).

The Ly49 and KIR genes share the same variegated pattern of expression. A recent study presented evidence that an upstream, bidirectional Ly49 promoter, Pro-1, acts as a probabilistic switch that establishes activity of the downstream major promoter, Pro-2 (15, 16). However, the molecular mechanisms that maintain stable Pro-2 activity in mature NK cells are unknown. It has been established that KIR transcription is in part regulated by epigenetic mechanisms primarily through differential DNA methylation of their CpG island promoters (17, 18). Held et al. (19) suggested that the maintenance of monoallelic Ly49 expression may be achieved via DNA methylation. This suggestion was proposed on the basis of the relative stability of monoallelic expression of Ly49 receptors on cultured NK cells (20), but no studies to examine the methylation state of Ly49 genes or alleles have been published. In this study, we have investigated the epigenetic state of the Ly49a Pro-2 region in primary NK cells and cell lines and show a strong link between DNA hypomethylation, histone acetylation, and transcriptional activity of the gene. These findings support the view that DNA methylation plays a role in maintaining Ly49 expression patterns.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 Disclosures
 References
 
Mice

All mice were bred and maintained in the animal facility of the British Columbia Cancer Research Centre (Vancouver, British Columbia, Canada). F1 hybrids were generated by crossing female BALB/c with C57BL/6 (B6) males. All mice used in this study were 6–10 wk old unless specified otherwise. All experiments were according to a protocol approved by the Committee on Animal Care of the University of British Columbia.

Abs, cell separation, and flow cytometry

The mAbs anti-FcR{gamma} (2.4G2), biotin-conjugated YE1/48 (anti-Ly49A BALB/c and B6), biotin-conjugated 4LO3311 (anti-Ly49C BALB/c and B6) have been described (21, 22). Biotin-conjugated A1 (anti-Ly49A B6-specific), anti-CD3{epsilon}-PerCP-Cy5.5 (145-2C11), anti-NK1.1-PE, anti-NK1.1-allophycocyanin, anti-Ter-119-PE, and fluorochrome-conjugated streptavidin were purchased from BD Biosciences. FITC-conjugated anti-Ly49E/C (4D12) Ab was provided by Dr. G. Leclercq (University of Ghent, Ghent, Belgium). Flow cytometry for cell sorting was performed on BD FACSAria Cell Sorting System. All sorted samples were >95% pure.

Fetal liver NK cell retrieval and sort

Embryonic day 17.5 B6 embryos were dissected for extraction of whole liver in ice-cold 1x PBS. Single-cell suspension of liver cells was stained with anti-Ter-119-PE and immunomagnetically depleted of embryonic erythrocytes via the Easysep kit (StemCell Technologies) according to the manufacturer’s protocol. The remaining cells were stained with anti-CD3{epsilon}-PerCP-Cy5.5 (145-2C11), anti-NK1.1-allophycocyanin, and FITC-conjugated anti-Ly49E/C (4D12) and sorted for NK1.1+CD3{epsilon}– and Ly49E positive and negative populations on the BD FACSAria Cell Sorting System. The sorted populations were resorted for >95% purity.

Primary cell and tissue genomic DNA (gDNA) extraction

gDNA was obtained by pelleting FACS-sorted cells via centrifugation and lysis by addition of 50 µl of water. The lysate was incubated for 10 min at 98°C, after which proteinase K was added to a final concentration of 1 mg/ml. The lysate was further incubated for 130 min at 55°C and 10 min at 98°C.

gDNA was extracted from fresh B6 mouse kidney and pancreas using DNAzol reagent (Invitrogen Life Technologies) per manufacturer’s instructions. Further proteinase K digestion and phenol-chloroform extraction was performed.

Cell culture

The B6 NKT cell line, EL4, the IL-2-independent T cell line, CTLL2 (obtained from Dr. T. Gonda, Hanson Center for Cancer Research, Adelaide, Australia) (23), and the pancreatic endothelium cell line, MS1 (ATCC CRL-2279; American Type Culture Collection), were cultured in DMEM supplemented with 10% FBS, 100 U/ml penicillin, and 100 U/ml streptomycin.

Chromatin remodeling drug treatment and RT-PCR

EL4 cells were cultured at an initial concentration 104 cells/ml for 24 h and treated with varying amounts and combinations of 5-aza-cytidine (5-aza-C) and trichostatin-A (TSA) for 96 h before RNA extraction. Total RNA was extracted from EL4 cells with TRIzol Reagent (Invitrogen Life Technologies) per the manufacturer’s instruction. RT-PCR was performed as described before (24) with the gene-specific primers for Ly49a, forward primer: 5'-GGTCACTTATTCAATGGTG-3', reverse primer: 5'-AGATAACAACATACATCCC-3'; Ly49d, forward primer: 5'-CGGAAGCCTGAAAAAGCTCG-3', reverse primer: 5'-TCACACAGTATGTTTTGATCCC-3'; Ly49g, forward primer: 5'-TTGCCACGATAACTGCAGCC-3', reverse primer: 5'-ATGTCTGAAGGAGCCAGGTTC-3'; beta-actin, forward primer: 5'-GAGGGCTATGCTCTCCCTCA-3', reverse primer: 5'-GCGCAAGTTAGGTTTTGTCAA-3' and a general primer pair for Ly49c-like (Ly49c/e/f/i/j) transcripts, forward primer: 5'-TCATAAGTCTTCAGGGTTG-3', reverse primer: 5'-ATCATAAGACAATCCAATCC-3'.

Sodium bisulfite conversion and PCR

Bisulfite conversion of DNA leads to conversion of all unmethylated cytosines into uracils, while methylated cytosines remain unchanged. Bisulfite conversion was performed using the CpGenome DNA Modification kit (Chemicon International) according to the manufacturer’s protocol with the following exceptions: gDNA was resuspended in 0.3 µM NaOH and heated at 50°C for 10 min. The gDNA was incubated with reagent-I at 50°C for a total of nearly 4 h with 30-s pulses to 95°C every 15 min. The conversion rate was >98%. First-round PCR amplification of Ly49a 5' region was performed using forward flanking Ly49a-specific primer: 5'-GTGTTTTTGTTTTTTTTGTAGGAGTT-3' and reverse flanking Ly49a-specific primer: 5'-AAAAAATCACAATTATCACATACTC-3'.

Converted DNA was used as template in a 45-µl reaction volume, containing 30 pmol of each primer, 1 mM dNTPs, 3 mM MgCl2, and 0.5 U Taq Platinum DNA polymerase (Qiagen). After initial denaturation for 7 min at 95°C, 30–40 cycles were performed, each consisting of 90 s at 95°C, 55 s at 53°C, and 30 s at 72°C followed by a final extension of 7 min at 72°C. Two microliters of the first PCR was used for seminested amplification using forward nested primer: 5'-TGTTTTGAGGGTTAGGTTTTATTAA-3' and the same reverse primer used in the first round. The same amplification conditions were chosen as for first-round PCR with the exception that the annealing temperature was raised to 54°C and the extension time was reduced to 20 s. For the Ly49c 5' region, the following primers were used: forward flanking Ly49c-specific primer: 5'-TTAAAGATAATGTTTTTTTTTTTTTGTAGT-3'; reverse flanking Ly49c-specific primer: 5'-CAATTATCACATACTACCAAAATT-3'; forward nested Ly49c-specific primer: 5'-AATAAGTAATTTTTTTTTTTTGTTTTGG-3'; reverse nested Ly49c-specific primer: 5'-TTCAATATATTTAATCATTTAATAAAAAC-3'.

Bisulfite sequencing and combined bisulfite and restriction enzyme analysis (COBRA)

The PCR products were electrophoresed on 1% agarose gel and correct size bands were extracted using the MinElute gel extraction kit (Qiagen). To exclude any Ly49g fragments that might have amplified along with the Ly49a fragments, the purified bands were subjected to EarI (NEB) restriction enzyme digest, per manufacturer’s instructions, which selectively cuts Ly49g in the amplified region. The digested products were gel extracted using the MinElute gel extraction kit (Qiagen). The purified products were cloned into the T vector using the pGEMT-vector kit (Promega). Sequencing was performed using the SP6 primer by McGill University and the Genome Québec Innovation Centre sequencing facility. All clones included in the figures are unique. Each clone was considered unique if it satisfied at least one of the following criteria: 1) if it contained a unique CpG pattern. 2) If the clone contained any unique unconverted (i.e., from C to T) non-CpG cytosines assuming that the clone’s total C to T conversion was above the random 95% cut-off limit. 3) If the clone contained any unique substitution of a noncytosine (non-CpG) base due to Taq-polymerase errors assuming that there were two or less of such events within a clone. 4) If there were identical clones in sequence that were derived from independent original gDNA samples. In cases where variation among various COBRA batches was observed (in particular Ly49A+ and Ly49C+ samples), due to random sampling variability, the bisulfite sequencing data was derived from 10 to 16 independent PCRs (two to three independent cell sorting events) for a more accurate estimation of the DNA methylation state.

For COBRA, the EarI-digested gel-purified fragments were digested with Taq{alpha}I and/or Acl-I restriction enzymes (NEB) to distinguish between methylated (CpG) and unmethylated (TpG). Taq{alpha}I and Acl-I recognize and cut 5'-TCGA-3'and 5'-AACGTT-3', respectively, and therefore will only cut fragments that were originally methylated in the gDNA and hence have not been converted by sodium bisulfite treatment.

Chromatin immunoprecipitation (ChIP) assays and quantitative real-time PCR

ChIP assays were performed using a Chromatin Immunoprecipitation Assay kit (Upstate Biotechnology) according to the manufacturer’s instructions. Briefly, cells were cross-linked with 1% formaldehyde at 37°C for 10 min. Cells were washed twice with ice-cold PBS. Cell pellets were lysed in the SDS lysis buffer provided in the kit. Lysates were sonicated at 30% power, 6 x 5 s using a Sonic Dismembrator model 300 (Fisher Scientific) on ice to shear gDNA. The following Abs were used to perform immunoprecipitations, polyclonal anti-acetyl-histone H3 (Lysine-9) (Upstate Biotechnology) and polyclonal anti-acetyl-histone H4 (multiple residues) (Upstate Biotechnology). Subsequent washes, elution and de-cross-linking were done according to manufacturer’s protocol. The DNA was purified via QIAquick PCR purification (Qiagen) and resuspended in 40 µl of deionized water. Quantitative PCR amplification of the promoter-2 (Pro-2) region of Ly49a was performed using Ly49a-specific forward primer 5'-CAACTTTTTCCTCCACCAGAAC-3' and Ly49a-specific reverse primer 5'-CGAGCGCTCAGATAACACTAC-3'. The housekeeping gene beta-glucuronidase (gus-b) was used as a positive control and for normalization of the data (primers provided by Dr. L. Palmqvist, Sahlgren’s University Hospital, Gotenberg, Sweden). Forty-eight rounds of amplification with SYBR Green PCR Master Mix (Applied Biosystems) were performed. The default 7500 System SDS software version 1.2.10 (7500 RealTime PCR System; Applied Biosystems) cycle was used with the exception that amplification was performed at 62°C in a total volume of 20 µl. Dissociation curve analysis was performed after the end of the PCR to confirm the presence of a single and specific product.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 Disclosures
 References
 
CpG distribution of the Ly49 Pro-2 region

The human KIR genes have classical CpG island promoters, where the CpG density is much higher than the rest of the human genome (25), and CpG sites are generally well-conserved among the promoters (17). In contrast, both promoter-1 (Pro-1) and promoter-2 (Pro-2) of the mouse Ly49 genes are CpG-poor. That is, the CpG density of these regions is on par with the average CpG distribution of the mouse genome. The distribution and conservation of CpG dinucleotides in the Pro-2 region of several Ly49 genes is shown in Fig. 1. Pro-2 is the major promoter in mature NK cells and gives rise to multiple transcriptional start sites in the genes that have been analyzed (26). In general, Ly49 genes that are closely related, namely Ly49c, i, e, and f, show a more conserved CpG distribution among the genes as would be expected. However, compared with the highly related KIR genes, there is much less CpG conservation among the Ly49 genes in this region. Nevertheless, because DNA methylation of even a single CpG residue has been shown to be functionally important (27), the low density of CpGs does not preclude a role for DNA methylation in this gene cluster.


Figure 1
View larger version (13K):
[in this window]
[in a new window]
  		FIGURE 1. Location of CpG dinucleotides in the Pro-2 regions of inhibitory Ly49 genes. A 340-bp region surrounding Pro-2 regions of Ly49a, c, e, f, g, and i was aligned. CpG dinucleotides are represented by vertical lines. The approximate locations of transcription start sites mapped by 5' RACE are indicated by braces for Ly49a and c. The bent arrows indicate the 5' ends of the following mRNA sequences Ly49i mRNA (U4986); Ly49e mRNA (U10091); Ly49f mRNA (U10092). For Ly49g, transcription was shown to be initiated mostly from promoter-3 (Pro-3) located near exon2 (in primary NK cells) which is not shown in this figure (26 ). A CpG dinucleotide conserved between Ly49a and c is indicated with a crescent symbol. Two CpG dinucleotides conserved among Ly49c, i, e, or f are shown with star symbols. A single CpG site conserved among Ly49c, e, f, and i is shown with a sun symbol.

 
Differential DNA methylation of the Ly49a and Ly49c Pro-2 regions in primary NK cells

To investigate the role of DNA methylation in Ly49 gene expression, we examined the methylation state of the Ly49a regulatory region in Ly49A-positive and negative primary NK cells from C57BL/6 (B6) mice. As mentioned above, both Pro-1 and Pro-2 of Ly49a are CpG-poor (Fig. 2A). The Pro-1 region of Ly49a contains only two CpG dinucleotides but the Pro-2 region, where transcription in mature NK cells originates (15, 26), contains six CpG dinucleotides within an ~340-bp range. This region, which we collectively call the Ly49a Pro-2 region, covers Pro-2, exon 1 and part of intron 1 (Fig. 2B). To determine the DNA methylation status of this region, we collected primary ex vivo splenic NK cells (CD3{epsilon}– NK1.1+) by sorting, according to Ly49A surface expression, into expressing and nonexpressing fractions. DNA was isolated from each fraction and sodium bisulfite sequencing and COBRA was performed. As shown in Fig. 3A, the CpG dinucleotides located in the Pro-2 region of Ly49a are heavily methylated in Ly49A– NK cells. In contrast, in DNA from Ly49A+ cells, half of the clones sequenced were completely unmethylated and the other half were heavily methylated as in the Ly49A– fraction (Fig. 3B). COBRA of Ly49A+ NK cells, which assays one CpG in the region (Fig. 2B), also showed the nearly equal presence of methylated and unmethylated sites in DNA from Ly49A+ cells (Fig. 3C). Hence, this methylation pattern is likely not due to patchy methylation or partial hypomethylation of this region but rather may reflect the monoallelic expression of Ly49a, a possibility tested directly below.


Figure 2
View larger version (22K):
[in this window]
[in a new window]
  		FIGURE 2. Ly49a Pro-2 region. A, Locations of all CpG dinucleotides relative to the Pro-1 and the Pro-2 regions of Ly49a. CpGs are represented by vertical lines, black boxes represent exons, and the bent arrows indicate promoter regions and the direction of transcription. Ly49 genes typically have two promoters. The putative Promoter-1 (Pro-1) is located upstream of exon1a and is active in immature NK cells. Promoter-2 (Pro-2) is located upstream of exon 1 and is the main promoter in mature NK cells. Exon 2 includes the ATG start codon. The CpGs within the boxed region of Pro-2 were assayed for methylation in this study. B, Nucleotide sequence of the boxed area in A. The CpG dinucleotides are shaded. The CpG dinucleotide assayed in COBRA by Taq{alpha}I is shown by a star symbol; the two TCF-1 sites (29 ) are indicated with boxes; the ATF-2 region (35 ) is shown with a dashed line above the sequence; the underlined region indicates the interval in which transcriptional start sites for Ly49a have been mapped (26 ). The horizontal arrows represent the nested bisulfite primers and the dashed horizontal arrows the location of the ChIP primers. The exon 1/intron 1 boundary is indicated with a horizontal line. This sequence is from the B6 Ly49a and the nucleotides shown below each line are the polymorphisms from the BALB/c sequence (48 ). There are two polymorphisms within the region amplified by the nested bisulfite primers and a third polymorphism within the 5' primer that cannot be detected using this primer pair.

 

Figure 3
View larger version (26K):
[in this window]
[in a new window]
  		FIGURE 3. Methylation status of Ly49a Pro-2 region in primary C57BL/6 Ly49A-nonexpressing and -expressing NK cells. Each line represents the sequence of an independent clone. The white and black circles represent unmethylated and methylated CpGs, respectively. A, Bisulfite sequencing of Ly49A-nonexpressing NK cells (YE1/48-negative). B, Bisulfite sequencing of Ly49A-expressing NK cells (YE1/48-positive) assayed from 10 independent samples. C, COBRA of Ly49A+ NK cells in four independent samples. The location of the CpG dinucleotide assayed by COBRA is indicated by a star symbol in Fig. 2B. Fragments that contain a CpG dinucleotide at this location are digested by Taq{alpha}I restriction endonuclease indicating methylation in the original genomic DNA. Fragments that remain uncut contain a TpG instead of a CpG which indicates that in the original genomic DNA template this CpG was unmethylated. The overrepresentation of the unmethylated signal in the COBRA samples might be due to some heteroduplex formation during PCR and hence inhibition of the restriction enzyme digestion (49 ).

 
Ly49c is also shown to be monoallelically expressed (6), so we investigated the methylation pattern of the Ly49c Pro-2 region in Ly49C+ cells. As with Ly49a, the Ly49c Pro-1 and Pro-2 regions are CpG-poor (Fig. 4A). We observed the same pattern of DNA methylation for the Ly49C+ NK cells, where the presence of unmethylated and methylated clones was detected via bisulfite sequencing and COBRA (Fig. 4, B and C, respectively). In Ly49C– adult NK cells, this region is highly methylated as indicated by bisulfite sequencing (Fig. 4D).


Figure 4
View larger version (17K):
[in this window]
[in a new window]
  		FIGURE 4. Methylation status of Ly49c Pro-2 region in primary ex vivo C57BL/6 Ly49C-expressing NK cells. A, Vertical lines depict CpG sites relative to Pro-1 and Pro-2 (arrows) of Ly49c. The positions of exons 1a, 1, and 2 (small boxes) are also shown. The Pro-2 region of Ly49c analyzed is indicated with the open box. Methylation state of the seven CpG residues located within the open box in A from Ly49C-positive (4LO3311) NK cells assayed by bisulfite sequencing from 12 independent PCRs (B) and COBRA on four independent samples with Taq{alpha}I and another two independent samples with Acl-I (C). The locations of the CpG dinucleotides assayed with COBRA by Taq{alpha}I and Acl-I restriction endonucleases are shown by arrows in B. Digestion by the Acl-I enzyme at either of its two recognition sites would generate a band labeled "methylated". The slight overrepresentation of the methylated clones might be due to a small amount of PCR bias for the methylated allele (50 ). Differences in the independent COBRA assays are due to random sampling variability. D, Methylation of Ly49c Pro-2 in Ly49C-negative NK cells. Each line represents the sequence of an independent clone. The white and black circles represent unmethylated and methylated CpGs, respectively.

 
Ly49A monoallelic gene expression is linked with DNA methylation profile

To investigate the link between the DNA methylation pattern of the Ly49a Pro-2 region and monoallelic expression of Ly49A, we generated C57BL/6 x BALB/c F1 hybrid mice. For the Ly49A– population, F1 NK cells were sorted using the YE1/48 Ab, which recognizes both B6 and BALB/c Ly49A, and DNA was subjected to bisulfite analysis. In this Ly49A– population, the region was heavily methylated as evident from bisulfite sequencing and COBRA (Fig. 5, A and B). For the Ly49AB6+ fraction, we sorted F1 NK cells that stained positive with the A1 Ab, which detects only the B6 allele of Ly49A (28). As expected, COBRA of DNA from Ly49AB6+ cells showed the presence of both methylated and unmethylated CpG residues (data not shown). Two polymorphisms between B6 and BALB/c Ly49a in the region of interest (shown in Fig. 2B) were used to distinguish the allelic origin of templates analyzed by bisulfite sequencing. In DNA from Ly49AB6+ F1 NK cells, all BALB/c alleles sequenced were heavily methylated and all B6 alleles were largely unmethylated (Fig. 5C). This result demonstrates that transcriptional activity of Ly49a alleles is associated with lack of DNA methylation in the Pro-2 region. Although a small percentage of Ly49A+ cells (~%5) should express Ly49A from both alleles (6, 20), we did not detect any unmethylated BALB alleles in the Ly49AB6+ (stained only with the A1 Ab) fraction. The failure to detect such clones may be due to the low percentage of double-positive cells.


Figure 5
View larger version (18K):
[in this window]
[in a new window]
  		FIGURE 5. Methylation status of Ly49a Pro-2 region in primary F1 hybrid Ly49A double-negative and Ly49AB6+ NK cells. A, Bisulfite sequencing and, B, COBRA (four independent samples) of Ly49A-nonexpressing (YE1/48-negative) NK cells. The location of the site assayed by COBRA is shown with a star in Fig. 2B. C, Bisulfite sequencing of Ly49AB6-expressing NK cells (A1-positive) derived from 16 independent samples. The methylation profiles of B6 and BALB/c alleles are shown individually.

 
Variable DNA methylation of Ly49a in nonlymphoid tissues

Ly49A expression is limited to lymphocytes probably because nonlymphoid cells do not possess the transcription factor repertoire needed to express this gene (29). We therefore hypothesized that there would be less need for maintenance of a hypermethylated state at the Ly49a locus in nonlymphoid cells. To test this hypothesis, we analyzed the methylation status of the Ly49a Pro-2 region in DNA from freshly isolated B6 kidney and pancreas (pooled from three mice) via COBRA. We found a mix of methylated and unmethylated CpGs in both the pancreas and kidney (Fig. 6A). There is a significant level of hypomethylation in the pancreas and to a lesser level in the kidney at least for one CpG site. This hypomethylation is not due to the incomplete digestion of the fragments by the enzyme (data not shown). To further investigate the DNA methylation state of the Ly49a Pro-2 region in nonlymphoid tissues, we performed bisulfite sequencing on pancreatic gDNA (Fig. 6B). The DNA methylation of this region in pancreatic cells is generally lower than that observed in Ly49A– NK cells.


Figure 6
View larger version (18K):
[in this window]
[in a new window]
  		FIGURE 6. Methylation status of Ly49a Pro-2 region in fresh B6 nonlymphoid tissue and fetal NK cells. A, COBRA of the Ly49a Pro-2 region of fresh pancreas and kidney (three and two independent samples, respectively) from the B6 mouse. B, Methylation of the Ly49a Pro-2 region in pancreas assayed by bisulfite sequencing. C, COBRA (four independent samples) and, D, bisulfite sequencing of the same region in primary ex vivo fetal (day 17.5) liver NK cells. The location of the site assayed by COBRA is shown with a star in Fig. 2B.

 
Ly49a Pro-2 is methylated in primary ex vivo fetal NK1.1+ NK cells

To gain insight into DNA methylation of the Ly49a Pro-2 region and Ly49 ontogeny, we examined the methylation state of Ly49a in fetal NK cells. Primary fetal NK cells from B6 mice do not express any Ly49 genes except Ly49e (30, 31). IL-2-dependent expansion of fetal day 17 (FD17) splenic and thymic NK cells also does not lead to expression of Ly49A (31). As well, we did not detect Ly49A on the surface of IL-2-expanded B6 FD17.5 liver NK cells via flow cytometry (data not shown). We sorted fresh liver NK cells from FD17.5 embryos, isolated DNA, and performed bisulfite sequencing and COBRA at the Ly49a Pro-2 region. As evident from the COBRA analysis, there is heavy methylation of the assayable CpG site in FD17.5 NK cells (Fig. 6C). Bisulfite sequencing revealed nearly complete methylation of all CpG dinucleotides (Fig. 6D). To address the possibility that the heavy DNA methylation of the Ly49a Pro-2 region is a characteristic of fetal NK cells rather than a phenomenon based on transcription, we analyzed the methylation status of Ly49e in Ly49E-expressing FD17.5 NK cells. As expected, the Ly49e Pro-2 region contains both methylated and unmethylated CpGs as assayed with COBRA (our unpublished data).

Linkage of DNA methylation status with histone acetylation of the Ly49a promoter region

DNA hypomethylation and histone acetylation usually correlate with transcriptional activity but this is not always the case (32, 33). To determine whether these two epigenetic marks are correlated with Ly49 expression, we used cell lines positive and negative for Ly49A expression. As judged by bisulfite sequencing and/or COBRA analysis, the methylation profile of Ly49a in B6 lymphoid and nonlymphoid cell lines mimics that seen in primary cells. The Pro-2 region of Ly49a is mostly unmethylated in the NKT cell line EL4, which expresses Ly49A (>97% by FACS), but is heavily methylated in the Ly49A– T cell line CTLL2 (negative by FACS and RT-PCR) and partially methylated in the pancreatic endothelium cell line MS1 (data not shown).

We examined the histone acetylation status of the Ly49a Pro-2 region in these cell lines using ChIP. Primers for the specific region analyzed are shown by dashed arrows in Fig. 2B. We observed a significant enrichment of Ly49a in acetyl-lysine-9-H3 and acetyl-H4 (multiple residues) fractions for EL4 compared with the Ly49A– cell lines CTLL2 and MS1 (Fig. 7A). The Ly49a region in EL4 cells is more than 100-fold enriched in the H3-acetyl-lysine-9 fraction compared with CTLL2 and to a slightly lesser extent compared with MS1. The same differential levels of Ly49a enrichment were detected in the acetyl-H4 fractions. Interestingly, there is a slight enrichment (<3-fold) of this region in the H3-acetyl-lysine-9 fraction of MS1 compared with CTLL2 which correlates with the less dense DNA methylation in MS1 compared with CTLL2.


Figure 7
View larger version (19K):
[in this window]
[in a new window]
  		FIGURE 7. Histone acetylation and effect of chromatin remodeling drugs. A, ChIP of the Ly49a Pro-2 region in lymphoid and nonlymphoid cell lines. ChIP analysis with anti-acetyl-lysine-9 H3 and anti-acetyl-H4 (multiple residues) Abs on 1) EL4, 2) CTLL2, and 3) MS1. The fold enrichments relative to the endogenous control are presented as 1/2{Delta}Ct; where {Delta}Ct is the difference between the threshold cycles (Ct) of Ly49a and the endogenous control (gus-b) of the same immunoprecipitation (IP). Enrichment is calculated as 2{Delta}Ct where approximately every 3.3 cycle difference is equivalent to 10-fold enrichment. The data is the average of three independent ChIP experiments ({Sigma}1/2{Delta}Ct) plus SD performed in duplicate. The enrichment of the control gene was equal among the three cell lines in all fractions and the nonspecific enrichment (no-Ab control) of both the control gene and Ly49a was very low in all three cell lines (data not shown). B, Induction of Ly49 transcription in EL4 by 5-aza-C and/or TSA treatment. Cells were cultured in various drug conditions for 96 h. Total RNA was isolated and RT-PCR was performed using gene-specific primers for beta-actin and Ly49a (35 cycles), Ly49d and g (37 cycles) and a general primer pair for Ly49c-like transcripts, Ly49c, e, f, i, and j (40 cycles).

 
Effect of chromatin remodeling drugs on Ly49 transcription

Although the Ly49 genes are similar in sequence and may share the same transcription factor repertoire, it is unknown whether they share a similar link between epigenetic modifications and transcription. As a preliminary step to answering this question, we treated EL4 cells with the DNA methyltransferase inhibitor 5-aza-C and the histone deacetylase inhibitor TSA. Because the Ly49a locus in EL4 is already in a state of open chromatin, no changes in Ly49a transcription were expected and none were detected. However, these treatments induced the detectable transcription of a number of other Ly49 genes. Most transcripts were detected with the combination of the two drugs. Ly49d, g, and at least one of Ly49c, e, f, i, and j were detected in EL4 (Fig. 7B).


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
Disclosures
 References
 
In this study, we have extensively analyzed the DNA methylation state of the Pro-2 region of Ly49a in B6 and F1 hybrid NK cells. Ly49A expression and DNA hypomethylation correlate where only the transcribed Ly49a allele is unmethylated. In NK cells that do not express Ly49A, both alleles are hypermethylated. Apart from the general hypo- or hypermethylation, it seems that the four most 3' CpG dinucleotides, which are located in exon-1 and intron-1 (Fig. 2B), are most indicative of the expression state of the allele. In the Ly49A– NK cells of adult B6, fetal B6, and the BALB/c allele of Ly49AB6+ F1 hybrid NK cells, these CpGs are nearly completely methylated. This pattern of DNA methylation may indicate an intronic control element or a general mechanism for the prevention of transcription from Pro-2 via the creation of transcriptionally restrictive chromatin. Previous work by our group showed that a region including the end of exon-1, all of intron-1, and the beginning of exon-2, has insignificant promoter activity in transient transfection assays in lymphoid cell lines (24). Even though possible enhancer activity by this region cannot be ruled out, it is likely that when this region is methylated, the formation of closed chromatin is induced which spreads into the promoter region and inhibits transcription from Pro-2. Indeed, we have shown a significant difference in histone acetylation levels of the Ly49a Pro-2 region between Ly49A-positive and -negative cell lines. Because the DNA methylation profile of the cell lines reflects that of the primary cells, it is quite probable that the primary cell histone acetylation profile also reflects that of the cell lines. A closed chromatin structure of downstream regions has also been shown to inhibit transcriptional elongation by RNA polymerase II (34).

Ly49a expression was shown to be dependent on the transcription factor, T cell factor-1 (TCF-1) (29). However, the TCF-1 protein is present in equal amount in both Ly49A-expressing and -nonexpressing primary NK cells. Therefore, the decision to express Ly49A or which allele to express it from, is not dependent on the presence of TCF-1. Rather, we propose that epigenetic mechanisms are responsible for the maintenance of the stable clonal monoallelic gene expression of Ly49a and possibly other Ly49 genes. We observed the same correlation between Ly49c expression and DNA methylation of the Pro-2 region of this gene in B6 NK cells as we have seen with Ly49a.

Interestingly, the Ly49a Pro-2 region is significantly less methylated (p < 0.001 as calculated by {chi}2) in primary nonlymphoid pancreatic cells compared with the region in all Ly49A-nonexpressing NK cells (B6 fetal, B6 adult, F1 double-negative and F1 BALB alleles). In fresh pancreatic cells, the Ly49a Pro-2 region is patch-methylated and this pattern is quite different from that of Ly49A– NK cells. Additionally, we observed a slight increase in the enrichment of the acetylated histones for this region from MS1 cells. This apparent DNA hypomethylation and possibly a less restrictive chromatin structure in nonlymphoid tissue at the Ly49a Pro-2 region may indicate a lack of need for the maintenance of a closed chromatin structure due to the absence of lymphoid-specific transcription factors. We speculate that the maintenance of heavy methylation of the Ly49a Pro-2 region is only necessary in Ly49A-nonexpressing lymphoid cells because they contain the transactivating factors to express this gene (29, 35). In accord with this idea is the finding that the IL-4 gene region in fibroblast cells, which lack the ability to express this gene, was shown to be hypomethylated compared with the IL-4-nonexpressing T cells (36).

There are differing views on the acquisition order of Ly49A by NK cells. Depending on the developmental stage of the NK progenitor or the culture system used, Ly49A was measured to be the first (37) or one of the last (38) Ly49 receptors to be expressed. However, there are no reports of Ly49A expression on fetal NK cells and our own analysis has also confirmed this. It seems that regardless of its acquisition order, Ly49A expression is restricted to postnatal and adult NK and NKT cells. Indeed, we have found that the Ly49a Pro-2 region of FD17.5 liver NK cells is highly methylated which not only correlates with the expression pattern of this receptor but also may indicate that at this stage in mouse development Ly49a transcription has not yet been initiated.

An upstream promoter, Pro-1, that seems to be active only in immature NK cells was identified for the inhibitory Ly49 genes (15). Pro-1 is essential for the expression of Ly49a as its deletion in an Ly49a transgene model abolished Ly49 expression in NK, T, and B cells (39). This transgenic Ly49a construct was expressed with the same ontogeny as the native wild-type gene in NK cells regardless of gene copy number and genomic position suggesting that the initiation of Ly49a expression is primarily dependent on the developmental stage of the NK cell and not on genomic location (39). However, the deregulated expression of Ly49a in B cells, which normally do not express it, may suggest aberrant epigenetic regulation due to multiple gene copies and change in the genomic location of the gene.

Pro-1 has been shown to act as a binary switch with the ability to transcribe in two directions and the strength of forward and reverse transcription from Pro-1 correlated well with the percentage of NK cells expressing a given Ly49 (16). It is possible that the variegated expression of Ly49a is due to the variegated expression from Pro-1 but the mechanism that ultimately stabilizes and maintains the final expression pattern is likely epigenetic. Pro-1 is hypersensitive to DNaseI in Ly49A-expressing and -nonexpressing cells but not in nonlymphoid tissue (39), suggesting that the lymphoid-specific control is applied at another location, possibly Pro-2. We believe that the presence of the Pro-1 region is not sufficient for maintenance of tissue-specific and monoallelic Ly49a expression. In the V(D)J regions of the Ig H chain locus, noncoding genic and intergenic transcripts appear before the rearrangement of each region and act to open up the chromatin to allow accessibility of the recombination machinery (40). In the immature NK cells, forward transcripts originating from Pro-1 might lead to the opening of the closed chromatin of Pro-2 and subsequent transcription from Pro-2 (16). In every individual NK cell, the combination of Ly49 alleles expressed may be further reinforced and maintained by the chromatin structure of the Pro-2 region.

In the KIR system of human NK cells, there is a generally high acetylation level for H3 and H4 of all the genes in the cluster regardless of expression state (41). A recent study has shown that the histone acetylation level of KIR3DL1 is similar in KIR3DL1-expressing and -nonexpressing NK cell lines and IL-2-expanded primary NK cells (42). This is contrary to what we have found in mouse cell lines for Ly49a where we observe a significant difference between acetylation levels of the Ly49a Pro-2 region among expressing and nonexpressing cell lines. We observed very low levels of histone acetylation for the Ly49e Pro-2 region which is not expressed by any of the cell lines tested (our unpublished data). This difference in types of epigenetic control between KIR and Ly49 genes may be due to distance between the genes in the cluster. The KIR genes are located very close together in a small 150-kb cluster where as the Ly49 genes (excluding Ly49b) are spread out over a 650-kb region (5, 17). It is possible that the Ly49 genes have insulators between adjacent genes that would prevent spreading of open or closed chromatin from one gene to the other where as such insulators may not exist in the KIR cluster. Indeed, differential DNA methylation may be sufficient for maintenance of KIR gene expression (17, 18, 42). This proposed mechanism is supported by the fact that transcription of KIR genes can be readily induced in NK cells by DNA methyltransferase inhibitors but not by the histone deacetylase inhibitor TSA (17). We have detected transcripts from the majority of Ly49 genes after treatment of EL4 cells with chromatin remodeling drugs, but the response of the genes to these drugs seems to be variable. Ly49g seems to be the most responsive to these drugs and we are currently investigating the link between Ly49g transcription and epigenetic mechanisms (A. Rouhi, C. G. Brooks, F. Takei, and D. L. Mager, submitted for publication). It is tempting to speculate that the different Ly49 genes are controlled by somewhat different epigenetic mechanisms and that the density of CpG dinucleotides in the promoter regions affects their maintenance capacity. In light of new evidence on the plasticity of the Ly49 repertoire of lymphoid cells in response to various cytokines (43), the involvement of epigenetic mechanisms seems plausible.

The EL4 cell line is the most permissive lymphoid cell line to Ly49 transcription. Promoter constructs of Ly49c, f, g, i, and j have activity in EL4 even though these genes are not usually expressed by this cell line (24, 44). In accordance with the hypothesis of epigenetic control of Ly49 transcription, our chromatin-remodeling drug treatment of EL4 led to the detection of transcripts from these genes. However, we also detected Ly49d transcripts in EL4 after drug treatment. This is contrary to the promoter construct experiment performed by Kunz and Held (44) where their Ly49d Pro-2 construct had no activity in the EL4 cell line. Experiments with chromatin remodeling drugs are known to produce false-positive and false-negative results, as these drugs affect the whole genome. However, it is also possible that the Ly49d transcripts detected were transcribed from a promoter other than Pro-2. Further work is needed to determine the role of epigenetic mechanisms in the transcriptional control of various Ly49 genes. In general, we expected the Ly49 genes to be less responsive to DNA methyltransferase inhibitors compared with the KIR genes because the promoter structure and CpG content of the two gene families are quite different. CpG island promoters, such as those of the KIR genes, tend to respond much more readily to such agents (45, 46). The Ly49 genes responded best to combinations of DNA methyltransferase and histone deacetylase inhibitors. Although it is not possible to make any firm conclusions about the variable effect of epigenetic mechanisms on the regulation of all Ly49 genes from our chromatin remodeling drug experiments, we can deduce a general trend underlining the effect of histone acetylation levels on transcriptional control of these genes.

In the present study, we have shown a strong association between lack of DNA methylation and expression of the Ly49a gene in primary NK cells. We have also determined a positive link between monoallelic expression and monoallelic DNA methylation pattern. We propose that differential epigenetic modifications are likely major factors in maintenance of the stable clonal and monoallelic Ly49 repertoire of NK cells. It would be interesting to determine whether the transcribed and the nontranscribed Ly49 alleles have different replication timings and nuclear organization as these seem to be additional epigenetic features related to random monoallelic gene expression (47).


   Acknowledgments
 
We thank Dr. Motoi Maeda, Dr. Brian Wilhelm, Dr. Karina McQueen, Dr. Lars Palmqvist, Dr. Randy Mottus, and Mona Wu for valuable advice, Dr. Sally Rogers for comments on the manuscript, and the Terry Fox Laboratory FACS facility staff for cell sorting. We also thank Dr. Georges Leclercq for the 4D12 Ab and Dr. Tom Gonda for the CTLL-2 cell line. Finally, we thank the anonymous reviewers for their helpful comments.


   Disclosures
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
References
 
The authors have no financial conflict of interest.


   Footnotes
 
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1 This work was supported by a grant from the Canadian Institutes for Health Research (to D.L.M.). Core support was provided by the British Columbia Cancer Agency. A.R. was supported by studentships from the Natural Sciences and Engineering Research Council of Canada and from the Michael Smith Foundation for Health Research. Back

2 Address correspondence and reprint requests to Dr. Dixie L. Mager, Terry Fox Laboratory, British Columbia Cancer Agency, 675 West 10th Avenue, Vancouver, British Columbia V5Z 1L3, Canada. E-mail address: dmager{at}bccrc.ca Back

3 Abbreviations used in this paper: KIR, killer cell Ig-like receptor; gDNA, genomic DNA; 5-aza-C, 5-aza-cytidine; TSA, trichostatin-A; ChIP, chromatin immunoprecipitation; COBRA, combined bisulfite and restriction enzyme analysis; TCF, T cell factor; FD, fetal day. Back

Received for publication August 11, 2005. Accepted for publication December 23, 2005.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 

  1. McQueen, K. L., P. Parham. 2002. Variable receptors controlling activation and inhibition of NK cells. Curr. Opin. Immunol. 14: 615-621. [Medline]
  2. McQueen, K. L., B. T. Wilhelm, K. D. Harden, D. L. Mager. 2002. Evolution of NK receptors: a single Ly49 and multiple KIR genes in the cow. Eur. J. Immunol. 32: 810-817. [Medline]
  3. Yokoyama, W. M., B. F. Plougastel. 2003. Immune functions encoded by the natural killer gene complex. Nat. Rev. Immunol. 3: 304-316. [Medline]
  4. Veinotte, L. L., B. T. Wilhelm, D. L. Mager, F. Takei. 2003. Acquisition of MHC-specific receptors on murine natural killer cells. Crit. Rev. Immunol. 23: 251-266. [Medline]
  5. Takei, F., K. L. McQueen, M. Maeda, B. T. Wilhelm, S. Lohwasser, R. H. Lian, D. L. Mager. 2001. Ly49 and CD94/NKG2: developmentally regulated expression and evolution. Immunol. Rev. 181: 90-103. [Medline]
  6. Held, W., B. Kunz. 1998. An allele-specific, stochastic gene expression process controls the expression of multiple Ly49 family genes and generates a diverse, MHC-specific NK cell receptor repertoire. Eur. J. Immunol. 28: 2407-2416. [Medline]
  7. Kubota, A., S. Kubota, S. Lohwasser, D. L. Mager, F. Takei. 1999. Diversity of NK cell receptor repertoire in adult and neonatal mice. J. Immunol. 163: 212-216. [Abstract/Free Full Text]
  8. Valiante, N. M., K. Lienert, H. G. Shilling, B. J. Smits, P. Parham. 1997. Killer cell receptors: keeping pace with MHC class I evolution. Immunol. Rev. 155: 155-164. [Medline]
  9. Attwood, J. T., R. L. Yung, B. C. Richardson. 2002. DNA methylation and the regulation of gene transcription. Cell. Mol. Life Sci. 59: 241-257. [Medline]
  10. Schwarzenbacher, K., J. Fleischer, H. Breer, S. Conzelmann. 2004. Expression of olfactory receptors in the cribriform mesenchyme during prenatal development. Gene Expr. Patterns 4: 543-552. [Medline]
  11. Chess, A., I. Simon, H. Cedar, R. Axel. 1994. Allelic inactivation regulates olfactory receptor gene expression. Cell 78: 823-834. [Medline]
  12. Shykind, B. M.. 2005. Regulation of odorant receptors: one allele at a time. Hum. Mol. Genet. 14 Spec No. 1: R33-R39.
  13. Bruniquel, D., R. H. Schwartz. 2003. Selective, stable demethylation of the interleukin-2 gene enhances transcription by an active process. Nat. Immunol. 4: 235-240. [Medline]
  14. Smale, S. T., A. G. Fisher. 2002. Chromatin structure and gene regulation in the immune system. Annu. Rev. Immunol. 20: 427-462. [Medline]
  15. Saleh, A., A. P. Makrigiannis, D. L. Hodge, S. K. Anderson. 2002. Identification of a novel Ly49 promoter that is active in bone marrow and fetal thymus. J. Immunol. 168: 5163-5169. [Abstract/Free Full Text]
  16. Saleh, A., G. E. Davies, V. Pascal, P. W. Wright, D. L. Hodge, E. H. Cho, S. J. Lockett, M. Abshari, S. K. Anderson. 2004. Identification of probabilistic transcriptional switches in the Ly49 gene cluster: a eukaryotic mechanism for selective gene activation. Immunity 21: 55-66. [Medline]
  17. Santourlidis, S., H. I. Trompeter, S. Weinhold, B. Eisermann, K. L. Meyer, P. Wernet, M. Uhrberg. 2002. Crucial role of DNA methylation in determination of clonally distributed killer cell Ig-like receptor expression patterns in NK cells. J. Immunol. 169: 4253-4261. [Abstract/Free Full Text]
  18. Chan, H. W., Z. B. Kurago, C. A. Stewart, M. J. Wilson, M. P. Martin, B. E. Mace, M. Carrington, J. Trowsdale, C. T. Lutz. 2003. DNA methylation maintains allele-specific KIR gene expression in human natural killer cells. J. Exp. Med. 197: 245-255. [Abstract/Free Full Text]
  19. Held, W., B. Kunz, V. Ioannidis, B. Lowin-Kropf. 1999. Mono-allelic Ly49 NK cell receptor expression. Semin. Immunol. 11: 349-355. [Medline]
  20. Held, W., D. H. Raulet. 1997. Expression of the Ly49A gene in murine natural killer cell clones is predominantly but not exclusively mono-allelic. Eur. J. Immunol. 27: 2876-2884. [Medline]
  21. Takei, F., J. Brennan, D. L. Mager. 1997. The Ly-49 family: genes, proteins and recognition of class I MHC. Immunol. Rev. 155: 67-77. [Medline]
  22. Brennan, J., S. Lemieux, J. D. Freeman, D. L. Mager, F. Takei. 1996. Heterogeneity among Ly-49C natural killer (NK) cells: characterization of highly related receptors with differing functions and expression patterns. J. Exp. Med. 184: 2085-2090. [Abstract/Free Full Text]
  23. Wilhelm, B. T., J. R. Landry, F. Takei, D. L. Mager. 2003. Transcriptional control of murine CD94 gene: differential usage of dual promoters by lymphoid cell types. J. Immunol. 171: 4219-4226. [Abstract/Free Full Text]
  24. McQueen, K. L., B. T. Wilhelm, F. Takei, D. L. Mager. 2001. Functional analysis of 5' and 3' regions of the closely related Ly49c and j genes. Immunogenetics 52: 212-223. [Medline]
  25. Gardiner-Garden, M., M. Frommer. 1987. CpG islands in vertebrate genomes. J. Mol. Biol. 196: 261-282. [Medline]
  26. Wilhelm, B. T., K. L. McQueen, J. D. Freeman, F. Takei, D. L. Mager. 2001. Comparative analysis of the promoter regions and transcriptional start sites of mouse Ly49 genes. Immunogenetics 53: 215-224. [Medline]
  27. Santoro, R., I. Grummt. 2001. Molecular mechanisms mediating methylation-dependent silencing of ribosomal gene transcription. Mol. Cell 8: 719-725. [Medline]
  28. Held, W., J. Roland, D. H. Raulet. 1995. Allelic exclusion of Ly49-family genes encoding class I MHC-specific receptors on NK cells. Nature 376: 355-358. [Medline]
  29. Held, W., B. Kunz, B. Lowin-Kropf, M. van de Wetering, H. Clevers. 1999. Clonal acquisition of the Ly49A NK cell receptor is dependent on the trans-acting factor TCF-1. Immunity 11: 433-442. [Medline]
  30. Toomey, J. A., S. Shrestha, S. A. de la Rue, F. Gays, J. H. Robinson, Z. M. Chrzanowska-Lightowlers, C. G. Brooks. 1998. MHC class I expression protects target cells from lysis by Ly49-deficient fetal NK cells. Eur. J. Immunol. 28: 47-56. [Medline]
  31. Van Beneden, K., A. De Creus, V. Debacker, J. De Boever, J. Plum, G. Leclercq. 1999. Murine fetal natural killer cells are functionally and structurally distinct from adult natural killer cells. J. Leukocyte Biol. 66: 625-633. [Abstract]
  32. Walsh, C. P., T. H. Bestor. 1999. Cytosine methylation and mammalian development. Genes Dev. 13: 26-34. [Abstract/Free Full Text]
  33. Fu, X. H., D. P. Liu, X. B. Tang, G. Liu, X. Lv, Y. J. Li, C. C. Liang. 2005. A conserved, extended chromatin opening within {alpha}-globulin locus during development. Exp. Cell Res. 309: 174-184. [Medline]
  34. Lorincz, M. C., D. R. Dickerson, M. Schmitt, M. Groudine. 2004. Intragenic DNA methylation alters chromatin structure and elongation efficiency in mammalian cells. Nat. Struct. Mol. Biol. 11: 1068-1075. [Medline]
  35. Kubo, S., R. Nagasawa, H. Nishimura, K. Shigemoto, N. Maruyama. 1999. ATF-2-binding regulatory element is responsible for the Ly49A expression in murine T lymphoid line, EL-4. Biochim. Biophys. Acta 1444: 191-200. [Medline]
  36. Santangelo, S., D. J. Cousins, N. E. Winkelmann, D. Z. Staynov. 2002. DNA methylation changes at human Th2 cytokine genes coincide with DNase I hypersensitive site formation during CD4+ T cell differentiation. J. Immunol. 169: 1893-1903. [Abstract/Free Full Text]
  37. Dorfman, J. R., D. H. Raulet. 1998. Acquisition of Ly49 receptor expression by developing natural killer cells. J. Exp. Med. 187: 609-618. [Abstract/Free Full Text]
  38. Williams, N. S., A. Kubota, M. Bennett, V. Kumar, F. Takei. 2000. Clonal analysis of NK cell development from bone marrow progenitors in vitro: orderly acquisition of receptor gene expression. Eur. J. Immunol. 30: 2074-2082. [Medline]
  39. Tanamachi, D. M., D. C. Moniot, D. Cado, S. D. Liu, J. K. Hsia, D. H. Raulet. 2004. Genomic Ly49A transgenes: basis of variegated Ly49A gene expression and identification of a critical regulatory element. J. Immunol. 172: 1074-1082. [Abstract/Free Full Text]
  40. Bolland, D. J., A. L. Wood, C. M. Johnston, S. F. Bunting, G. Morgan, L. Chakalova, P. J. Fraser, A. E. Corcoran. 2004. Antisense intergenic transcription in V(D)J recombination. Nat. Immunol. 5: 630-637. [Medline]
  41. Uhrberg, M.. 2005. Shaping the human NK cell repertoire: an epigenetic glance at KIR gene regulation. Mol. Immunol. 42: 471-475. [Medline]
  42. Chan, H. W., J. S. Miller, M. B. Moore, C. T. Lutz. 2005. Epigenetic control of highly homologous killer Ig-like receptor gene alleles. J. Immunol. 175: 5966-5974. [Abstract/Free Full Text]
  43. Gays, F., K. Martin, R. Kenefeck, J. G. Aust, C. G. Brooks. 2005. Multiple cytokines regulate the NK gene complex-encoded receptor repertoire of mature NK cells and T cells. J. Immunol. 175: 2938-2947. [Abstract/Free Full Text]
  44. Kunz, B., W. Held. 2001. Positive and negative roles of the trans-acting T cell factor-1 for the acquisition of distinct Ly-49 MHC class I receptors by NK cells. J. Immunol. 166: 6181-6187. [Abstract/Free Full Text]
  45. Lodygin, D., A. Epanchintsev, A. Menssen, J. Diebold, H. Hermeking. 2005. Functional epigenomics identifies genes frequently silenced in prostate cancer. Cancer Res. 65: 4218-4227. [Abstract/Free Full Text]
  46. Brueckner, B., R. G. Boy, P. Siedlecki, T. Musch, H. C. Kliem, P. Zielenkiewicz, S. Suhai, M. Wiessler, F. Lyko. 2005. Epigenetic reactivation of tumor suppressor genes by a novel small-molecule inhibitor of human DNA methyltransferases. Cancer Res. 65: 6305-6311. [Abstract/Free Full Text]
  47. Singh, N., F. A. Ebrahimi, A. A. Gimelbrant, A. W. Ensminger, M. R. Tackett, P. Qi, J. Gribnau, A. Chess. 2003. Coordination of the random asynchronous replication of autosomal loci. Nat. Genet. 33: 339-341. [Medline]
  48. Anderson, S. K., K. Dewar, M. L. Goulet, G. Leveque, A. P. Makrigiannis. 2005. Complete elucidation of a minimal class I MHC natural killer cell receptor haplotype. Genes Immun. 6: 481-492. [Medline]
  49. Warnecke, P. M., C. Stirzaker, J. Song, C. Grunau, J. R. Melki, S. J. Clark. 2002. Identification and resolution of artifacts in bisulfite sequencing. Methods 27: 101-107. [Medline]
  50. Warnecke, P. M., C. Stirzaker, J. R. Melki, D. S. Millar, C. L. Paul, S. J. Clark. 1997. Detection and measurement of PCR bias in quantitative methylation analysis of bisulphite-treated DNA. Nucleic Acids Res. 25: 4422-4426. [Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 Nucleic Acids ResHome page
A. Rouhi, C. B. Lai, T. P. Cheng, F. Takei, W. M. Yokoyama, and D. L. Mager
Evidence for high bi-allelic expression of activating Ly49 receptors
Nucleic Acids Res., September 1, 2009; 37(16): 5331 - 5342.
[Abstract] [Full Text] [PDF]

Home page
 Mol. Cell. Biol.Home page
I. A. Maksakova, Y. Zhang, and D. L. Mager
Preferential Epigenetic Suppression of the Autonomous MusD over the Nonautonomous ETn Mouse Retrotransposons
Mol. Cell. Biol., May 1, 2009; 29(9): 2456 - 2468.
[Abstract] [Full Text] [PDF]

Home page
 Nucleic Acids ResHome page
C. B. Lai, Y. Zhang, S. L. Rogers, and D. L. Mager
Creation of the two isoforms of rodent NKG2D was driven by a B1 retrotransposon insertion
Nucleic Acids Res., May 1, 2009; 37(9): 3032 - 3043.
[Abstract] [Full Text] [PDF]

Home page
 ScienceHome page
A. Gimelbrant, J. N. Hutchinson, B. R. Thompson, and A. Chess
Widespread Monoallelic Expression on Human Autosomes
Science, November 16, 2007; 318(5853): 1136 - 1140.
[Abstract] [Full Text] [PDF]

Home page
 J. Immunol.Home page
S. L. Rogers, A. Rouhi, F. Takei, and D. L. Mager
A Role for DNA Hypomethylation and Histone Acetylation in Maintaining Allele-Specific Expression of Mouse NKG2A in Developing and Mature NK Cells
J. Immunol., July 1, 2006; 177(1): 414 - 421.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Rouhi, A.
Right arrow Articles by Mager, D. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Rouhi, A.
Right arrow Articles by Mager, D. L.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS