Unlike studies focused on analysis of sequence-specific differences between two populations (e.g. mRNA profiling, genotyping, copy number analysis etc.,), methylation profiling presents an additional layer of complexity during sample preparation in that the nucleic acid sequence itself is not sufficient to discriminate between two populations.  As a result, some sort of manipulation must first be performed that differentiates methylated samples from unmethylated for downstream analysis.

Over the years many methods have been introduced to accomplish this including: bisulfite conversion, restriction digestion with methyl sensitive and methyl dependant enzymes, protein binding/capture, and immunoprecipitation.  After tackling the sample preparation, many of same methods can be used for global profiling and/or locus specific methylation analysis with modifications in probe and primer design for microarrays and PCR respectively.

Each has their own relative strengths and weaknesses so we recommend reviewing some of these protocols from the companies and labs that have tried to streamline each method. 

METHYLATION ANAYLISIS SAMPLE PREPARATION

Bisulfite Conversion

• Issa Lab (MD Anderson Cancer Center)
• Ushijima/Gratchev
• Fan Lab (University of California Los Angeles)

Methyl-Sensitive Enzymatic Digestion Combos

• BstUI and HpaII  

Reference: Yan, P.S. et al. (2002) Methods 27,162–169

• BstUI, HhaI, and HpaII  

  Reference: Ibrahim, Ashraf et al. (2006) Nucleic Acids Research Vol. 34, No. 20 e136

• AciI, HinP1I, HpyCH4IV, and HpaII 

  Reference: Ibrahim, Ashraf et al. (2006) Nucleic Acids Research Vol. 34, No. 20 e136

Methyl-Dependent Enzymatic Digestion

• McrBC

  Reference: Ibrahim, Ashraf et al. (2006) Nucleic Acids Research Vol. 34, No. 20 e136

Methylated DNA Immunoprecipitation (MeDIP)

• 5 Methylcytosine Antibody

Reference: Weber, M. (2005) Nature Genetics Aug. Volume 37 Number 8

Genomic DNA Amplification

• FlyChip Group LM PCR
• Friedman Lab University of Pennsylvania LM PCR

Fluorescent Genomic DNA Labeling

•  Brown Lab (Stanford)
• FlyChip Group
• University of Wisconsin E.coli Genome Project
• McClelland Lab (Sidney Kimmel Cancer Center) 

 
METHYLATION PROFILING AND VALIDATION

 CpG Island Microarrays

• University Health Network Microarray Center
   

Methylation Specific PCR (MSP)

• Issa Lab (MD Anderson Cancer Center)
   

Chromosome immunoprecipitation (ChIP) has proven to be an invaluable method for researchers studying protein-DNA interactions. Recent advances in high density microarrays and deep sequencing technologies have made it possible to use ChIP to survey genome-wide protein-DNA interactions at unprecedented resolution, however, the efficacy of these downstream applications is often dependant on the upstream sample processing.

Stabilization of protein-DNA interactions and subsequent enrichment are extremely critical in ChIP protocols. High quality antibodies are ideal when attempting to enrich for a specific protein nucleic acid complex, although alternative vector-based approaches have been employed in cell lines.

Two slightly different approaches are commonly used in ChIP, Native-ChIP (N-ChIP) and Cross-Linking ChIP (X-ChIP).  The primary difference between the N-ChIP and X-ChIP protocols is the preliminary processing step. N-ChIP protocols typically employ a micrococcal nuclease digestion to fragment chromatin, whereas X-ChIP protocols utilize formaldehyde cross-linking to stabilize protein-DNA interactions prior to sonication to fragment chromatin and immunoprecipitation to pull down the specific complexes
Some of the more popular uses for ChIP on Chip, the common term for applying Chromosome Immunoprecipitation to microarrays, include studying the impact of various histone modifications on chromatin structure and function, the role of trans acting regulatory proteins, and the impact of DNA methylation on gene regulation.

SAMPLE PREPARATION

Chromosome Immunoprecipitation (ChIP)

• Fan Lab (University of California Los Angeles)
• University Health Network (Bead-Based)
• University Health Network (Staph A)

Genomic DNA Amplification

• FlyChip Group LM PCR
• Friedman Lab University of Pennsylvania LM PCR

Fluorescent Genomic DNA Labeling

• Brown Lab (Stanford)
• FlyChip Group
• University of Wisconsin E.coli Genome Project
• McClelland Lab (Sidney Kimmel Cancer Center)

CHIP ON CHIP

CpG Island Microarrays

• University Health Network Microarray Center

Non-coding RNAs (ncRNAs) play diverse roles in gene regulation and many other cellular processes. They can be very diverse in size, expression, and function. Some such as microRNAs (miRNAs), Piwi-interacting RNA (piRNAs), and small nucleolar RNA (snoRNAs) are quite small, ranging in size from 20 nt to 200-300 nt. Other ncRNAs, such as Xist and AIR, approach 20-100 kb in length in mammals. The temporal and spatial expression patterns of ncRNAs are very dynamic. This diversity of physical and functional attributes of ncRNAs has driven the development and introduction of a variety of research tools to facilitate analysis of ncRNAs.

Currently miRNAs are more widely studied than other types of ncRNA. There are many optimized commercial protocols for miRNA enrichment, target labeling, profiling, qRT-PCR, cellular localization, and functional analysis. A mixture of approaches has been taken for target labeling and profiling.

The most significant difference amongst target labeling methodologies is the approach for fluorescent addition. Some methods ligate a single fluorophore onto the end of the miRNA with T4 RNA ligase. Other approaches use poly-A polymerase to tail the end of the miRNA. During this step modified amino-allyl nucleotides are introduced for subsequent coupling to fluorescent amino esters as described with previous mRNA labeling protocols.  Another method combines components of both protocols by first tailing the miRNAs with poly A polymerase, then ligating a “capture sequence” for detection with a branched DNA molecule containing hundreds of fluorophores. The number of human miRNAs in the Sanger Institute’s miRBase is approximately 500 (Version 9.2), making profiling by bead-based methods and qRT-PCR more feasible, however, microarrays are more widely used. Opinions on the number of miRNAs vary, but many believe that there are thousands if not tens of thousands hinting that microarrays may prove useful in the near future for global profiling experiments.

miRNA SAMPLE PREPARATION

miRNA Cloning

• Bartel Lab (MIT)

miRNA Labeling

• Miska Lab (Gurdon Institute) T4 RNA Ligase

miRNA Northern Blot

• Bartel Lab (MIT)