CGH, CGH array, micro-macro array - (Mar/10/2006 )

Ciao, I am new to this forum..I hope somebody can help me to understand the diffrence between CGH, CGH array and micro and macro array.


is CGH only used to check structural anomalies and micro-macro array only gene expression because the probes can be smaller?

why I only find info on the web about micro arrays and never about macro arrays?

Thanx millions

Francy

---

Hi Francy,

Some general answers to your questions:

Macroarrays constist of DNA probes (normally gene fragments / PCR products - sometimes long oligos) immobilised on nylon membranes, normally between 8x12cm and 22x22cm square. The spots are normally ~1mm diameter, so probe density is quite limited. Macroarray experiments generally use separate membranes/hybridisations for each experimental condition (control vs. induced, for example), with normalisation within each array using signals from 'housekeeping' genes.

Advantages: Tend to be cheaper to buy and use than microarrays, as each array can be 'stripped' and reused 2-3 times (unlike microarrays, which are single-use). Target cDNA (for gene expression) or genomic DNA (for CGH) is often radiolabelled using 32P or 33P, so very sensitive detection is possible. There's lots of probe on each spot, so they tend to give good reproducible and quantitative data over a large range (several logs).

Disadvantages: Number of probes per membrane can be quite limited (maybe ~1500 on a ~10cm square membrane), so maybe better suited to prokaryotic genomics than eukaryotic (unless you're interested in a limited subset of genes). Radiolabelling can also present problems, and you need to use X-ray film or have access to a storage-phosphor scanner to 'image' the arrays. Fluorescence labelling is generally not possible due to difficulty scanning membranes (also lots of 'autofluoresence') - although you can use chemiluminescent labelling.

For web links to gene expression experiments using macroarrays google "clontech Atlas array".

Microarrays come in 2 broad types:

1. Spotted glass slide arrays (made using either PCR-product or oligo probes). These are used in a 'competitive 2-colour hybridisation' format where a mixture of 2 differentially labelled targets are hybridised onto the array at the same time. This is essential to obtain reproducible data from arrays of this type.

2. In-situ synthesised oligo arrays (Affymetrix / Agilent / Febit). Agilent oligo arrays consist of long (60mer) probes, so they're used just like spotted arrays (but give nicer results). Affymetrix-type arrays consist of short (~25mer) probes, with an overlapping set of perfect-match and mismatched probes for each target gene. Unlike spotted array experiments, Affymetrix arrays use a 'single colour' format, with one array per sample (rather than 2 samples co-hybridised onto the same array) - just like the macroarray format described above. This is possible because of the highly reproducible data which these arrays generate.

Here's a general review of spotted and Affymetrix microarray formats - a few years old now, but useful:
Monitoring gene expression using DNA microarrays (2000).
CA Harrington, C Rosenow, J Retief. Current Opinion in Microbiology. 3:285-291.

The main advantage of microarrays vs. macroarrays are really the much higher probe density (typically 20-30,000 probes on a glass slide array; many more on Affy arrays).

CGH stands for comparative genome hybridisation, which aims to compare the presence / absence / number of similar genes in 2 genomes (i.e. it's a 'survey' of genetic differences between 2 organisms). So, a CGH experiment might compare gemomic DNA from 2 closely related bacterial species or strains. The difference between CGH array experiments and gene expression array experiments is just the target which is hybridised to the array: labelled genomic DNA for CGH; labelled cDNA (or cRNA - derived from mRNA in either case) for gene expression. You can use exactly the same arrays for both types of experiment, although the objective of CGH is normally 'genome-wide' comparison of 2 genomes, so CGH normally uses whole-genome microarrays (for prokaryotes, at least). In general, you can use any genomic array for either gene expression or CGH.

hope that helps to clarify things!

Del.

---

Much more clear now Del, thanx!

Just two questions though:

1) are micro and macro arrays for mRNA always based on competitve hybridization of test DNA and control that can be radiolabelled in two different ways or bound to two different fluophores as in CGH for the DNA?

2) micro and macro arrays that are not used for CGH are used to see gene expression but mRNA level and protein expression do not necessarily correlate. Protein functionality is often dependent on posttranslational processing of the precursor protein and regulation of cellular pathways frequently occurs by specific interaction between proteins and/or by reversible covalent modifications such as phosphorylation.......so how can micro and macro arrays be used to see gene expression?

Francy

---

Hi Francy,

Macroarrays and Affymetrix microarrays don't use a competitive hyb format. Macroarrays use radioactive (or chemiluminescent) labelling, so you can't do differential labelling - think of them as 'one colour' or 'one channel' systems. That's okay though, because there's a huge excess of probe on every spot (relative to the amount of target which hybridises), and the hybs themselves are generally well mixed etc., so they tend to be both quantitative and reproducible. So... there's no need to use a competitive hyb format.
Affymetrix arrays also use a one-colour fluorescent labelling system - again, that's okay because the data are so reproducible. In principle you *could* do a competitive 2-colour hyb on an Affy array, but there's no need to (also Affy don't provide protocols for this, and their scanners are single-channel only).
Spotted glass slide arrays *have* to use a competitive 2-colour format, as the results of hybs with these are more variable. The 2-colour competitive hyb is just an elegant solution to reduce this inherent variability.
So...
For a single 'test-vs-control' gene-expression expression experiment:

• Maroarray format with radiolabel (32P or 33P); 2 separate labeling reactions followed by 2 separate array hybs - one macroarray for 'test' & one for 'control'.
• Affy format (single fluor label); 2 separate labelling reactions followed by 2 separate array hybs - one GeneChip for 'test' & one for 'control'.
• Spotted slide format (2-colour competitive hyb); 2 separate labelling reactions with different fluors; labelled products combined and hybridised to the same array.

For a gene expression experiment, the labelled targets would be derived from mRNA (normally cDNA or aRNA). For a CGH experiment the labelled targets would be genomic DNA. Otherwise, the experimental details are the same.


Quite right - genome-wide gene expression methods (microarrays, SAGE, whatever) only give information about relative mRNA abundance (the 'transcriptome'). Obviously this is important, but it's certainly not the whole story. This information has to be combined with 'proteome' data (from 2D-PAGE & MALDITOF-MS, LC-MS-MS etc.) to infer whether particular transcripts are really being expressed, and in what form. Then there's the question of the biological activity of the protein products and their interaction with exogenous factors, like pharmaceuticals (the 'metabolome' or 'metabonome')... Gene expression work is just one part of this 'systems biology' approach.

hope that's clearer.

D.

---

your explanation was wandefully clear Del....i think I finally have the picture of the arrays: thank you for the patience in writing all the details!!!

---