Finding the Hypermutation Consensus Sequence: The Next Step
A brief description of the annotated loci

    The publicly available sequence has been obtained and annotated. In other words, the entire Ig kappa and Ig lambda loci are now annotated, as is approximately half of the Ig heavy locus.
    The heavy locus was found on two separate contigs. There was a gap between the two contigs, and it was calculated to be approximately 75000 base pairs. The 3' end of the Ig heavy locus is also missing. Thus, only the IgM, IgD, IgG3, IgG1, IgG2 and IgG4 regions are available. The gap encompases the pseudoE, IgA1 and pseugoG genes, whereas the IgG and IgA2 regions are in the missing 3' end.
    In the heavy locus, I was able to obtain the following complete intergenic regions: IgM-IgD, IgD-IgG3, IgG3-IgG1 and IgG2-IgG4. There are two incomplete intergenic regions: 8 kB in the 5' end out of 19 kB of the IgG1-IgPseudoE region, and 25 kB in the 3' end out of a possible 40 kB of  the IgPseudoG-IgG2 region. As one of the incomplete regions contains only the 5' end, while the other has the 3' end, it is likely that at least one of these regions will not contain the HCS, if we assume that the HCS has a conserved position in the intergenic sequence. Or for that matter, if we assume that the HCS is actually present in the intergenic region. Or for that matter we have to assume tthat the HCS exists at all... That's a lot of assumptions indeed...

Summary of the loci, and the proposed experiments

    In this section, I will only deal with the experiments concerning the region 3' of the IgC, and not the intronic regions. This is not to say that the intronic regions will not be analyzed at some point, but at this point I feel that there is a higher probability that the HCS is in the 3' region (see Scharff and Neuberger articles).



    The hypothesis of this project is that the HCS is present in the region 3' of the constant gene. In this locus, there are 4 such complete sequences, and 2 incomplete sequences. However, IgM and IgD are translated from the same transcript, and it is thus likely that there is one HCS that controls the hypermutation of both transcripts. Hever, it's unknown whether the HCS is in the IgM-IgD region, or 3' of IgD. Therefore, 3 parallel experiments need to be performed: one comparing both sequences to the rest, and the other two comparing only one of the sequences to the rest (Expts 1-3; see table below for summary of experiments).
    Another question that arises is whether the incomplete sequences (3' of IgG1 and 5' of IgG2) should be used in the comparison. As the regions are incomplete, the HCS might not be present, even if the hypothesis is correct. This introduces even more uncertainty, and these regions ill not be used in the preliminary experiments (Expts 4-7).



    The IgK locus is complete. The 3' region contains a potential transcription unit similar to the BENE gene. The NKG2E gene is even further 3' downstream. This poses a question of whether the last part of the 3' region (between BENE and NKG2E) should be used in the analysis. The IgKC-BENE region encompases around 40 Kb.  While it's possible that the HCS is even further downstream of that, it's unlikely, and therefore only the 40 Kb will be anayzed initially.
    In Entrez, the size of the MAR was said to be 1000 bp. This seems more like an estimate, rather than an actual analysis, and it should be verified at some point in time...



    There are 4 IgL C genes and 3 pseudogenes in this locus. The "real" genes have relatively conserved intronic sequences, different from the pseudogenes. This suggests a duplication event in the locus. The region analyzed will consist of the 3' end of the locus, from C7 to the end, and including the enhancer. There is a question of whether the enhancer should be included, since it may contain repetitive sequence, or other regulatory sequences that might be found in the oher loci. However, it will be included for now, and if it turns out that there's too much info, I will exclude it.

Table of intended experiments

EXPT #
IgM-IgD
IgD-IgG3
IgG3-IgG1
3' IgG1
5' IgG2
IgG2-IgG4
IgLC7-enh
IgL ENH
3'IgL ENH
IgK-BENE
3' BENE 
1
X
X
 
 
 
 
 
    
 
 
2
X
-
 
 
 
 
 
    
 
 
3
-
X
 
 
 
 
 
    
 
 
4 (NO)
 
 
X
X
X
X
 
    
 
 
5 (NO)
 
 
X
X
-
X
 
    
 
 
6 (NO)
 
 
X
-
X
X
 
    
 
 
7
 
 
X
-
-
X
 
    
 
 
8 (NO)
 
 
 
  
 
 
 
    
X
X
9
 
 
 
 
 
 
 
    
X
-
10 (NO)
 
 
 
 
 
 
 
    
-
X
11
            
X
X
X
    
12 (NO)
            
X
-
-
    
13 (NO)
            
X
-
X
    
    Thus there are 4 groups of variable experiments, and each experiment in the group should be performed with all the possible alternatives. This gives a total numbr of experiments of : (3) x (4) x (3) x (3) = 108! This is a bit much, and some of the scenarios are more likely that others.
    For example, experiments 8 and 10 are largely unnecessary, as BENE transcript likely marks the end of the 3' region of IgK. However, BENE is a hypothetical transcipt, and perhaps it is inside the 3' IgK region. This, however, is unlikely. Therefore, experiments 8 and 10 will not be performed.
    Another point of question is the validity of including the incomplete regions (3' of IgGG1 and 5' of IgG2). As the regions are incomplete, the HCS might not be present, even if the hypothesis is correct. Thus, experiments 4, 5 and 6 will also be put on the proverbial shelf for now.
    With regard to IgL, the enhances might contain some repetitive sequence, or conseved regulatory elements, and thus, it could be left out. However, the strcture of the IgL locus is unorthodox, and thus, I feel uncomfortable making too many assumptions with very little ground to stand on. Thus, I will include the whole 3' region of IgL, and expts. 12 and 13 will not be performed.

    This reduces the number of experiments to 3, as the only variables are the IgM and IgD regions. To me, it is most likely that the HCS will not be present in the IgM-IgD region, as IgM and IgD are a part of the same primary transcript, and there is no switch region between them (and thus IgM cannot be excised from the genome without excising IgD - it thus follows that the two are under the control of the same HCS). Further, the IgM-IgD region is more like an intron, and I'll go out on a limb and say that the HCS is likely 3' of IgD. Nonetheless, this logic is shady at best, and thus I will perform the three alternate experiments.

Now, to prepare the sequence...