100
FENG QIAN AND GREGORY G. GERMINO
Partial digests
Successive
hybridizations
Co-migration
PFGE, with its ability to resolve fragments of up to 10Mb, is a useful tool in
most mapping strategies. DNA generally is prepared in agarose blocks and
then digested with one of the rare cutting enzymes. The infrequency of sites
for these enzymes is based either on the length of their recognition sequence
(8 bp for Sfi I) or the presence of 1 or 2 CpGs within the recognition sequence. (The frequency of the dinucleotide CpG is generally reduced within
the mammalian genome.) After electrophoresis, the gel is blotted in the
usual manner and probed with radiolabeled fragments. There are a number
of features characteristic of PFGE mapping.
Single-copy sequences often detect multiple bands when hybridized to
Southern blots of pulsed field gels. This occurs because most of the infrequent cutter enzymes will not cleave at sites where cytosine is methylated.
Since different DNA molecules from the same DNA prep may have different
methylation patterns, a labeled probe may hybridize to several bands. This
frequently can be useful in linking markers that map to adjacent complete
digest fragments.
Genetic markers are said to be physically linked when they hybridize to at
least some ofthe same fragments. Since there is considerable gel-to-gel variation in the rate of migration of DNA samples relative to that of the molecular weight markers (as well as considerable differences in resolution),
probes must be successively hybridized to the same filter. Only identical
bands are considered co-migrating. Unfortunately, most blots lose signal
after several hybridizations, making it difficult to link up multiple loci.
One of the challenges in constructing a map is determining whether bands
detected by different probes that appear identical on an autoradiograph
represent the same fragment vs. co-migrating of separate fragments. Figure
1 illustrates this principle. In this example, all three probes hybridize to Nru
I and Pvu I fragments of identical size. Indeed, even the partial digestion
pattern with Cla I is consistent with the hypothesis that all three probes are
on the same Nru I fragment. The Not I hybridization data does not disprove
this since it is possible that there could be a Not I site between them. However, the Not 1/Nru I double digest disproves the hypothesis. The different
Not I fragments detected by Probes A and B/C cannot be simply due to a
single site between them because the Nru I fragment ofProbe A shifts down
when digested with Not I whereas the Nru I fragment of Probes Band C
remains unaltered. If all three were on the same Nru I fragment, each would
have detected a shift in size.
FENG QIAN AND GREGORY G. GERMINO
Partial digests
Successive
hybridizations
Co-migration
PFGE, with its ability to resolve fragments of up to 10Mb, is a useful tool in
most mapping strategies. DNA generally is prepared in agarose blocks and
then digested with one of the rare cutting enzymes. The infrequency of sites
for these enzymes is based either on the length of their recognition sequence
(8 bp for Sfi I) or the presence of 1 or 2 CpGs within the recognition sequence. (The frequency of the dinucleotide CpG is generally reduced within
the mammalian genome.) After electrophoresis, the gel is blotted in the
usual manner and probed with radiolabeled fragments. There are a number
of features characteristic of PFGE mapping.
Single-copy sequences often detect multiple bands when hybridized to
Southern blots of pulsed field gels. This occurs because most of the infrequent cutter enzymes will not cleave at sites where cytosine is methylated.
Since different DNA molecules from the same DNA prep may have different
methylation patterns, a labeled probe may hybridize to several bands. This
frequently can be useful in linking markers that map to adjacent complete
digest fragments.
Genetic markers are said to be physically linked when they hybridize to at
least some ofthe same fragments. Since there is considerable gel-to-gel variation in the rate of migration of DNA samples relative to that of the molecular weight markers (as well as considerable differences in resolution),
probes must be successively hybridized to the same filter. Only identical
bands are considered co-migrating. Unfortunately, most blots lose signal
after several hybridizations, making it difficult to link up multiple loci.
One of the challenges in constructing a map is determining whether bands
detected by different probes that appear identical on an autoradiograph
represent the same fragment vs. co-migrating of separate fragments. Figure
1 illustrates this principle. In this example, all three probes hybridize to Nru
I and Pvu I fragments of identical size. Indeed, even the partial digestion
pattern with Cla I is consistent with the hypothesis that all three probes are
on the same Nru I fragment. The Not I hybridization data does not disprove
this since it is possible that there could be a Not I site between them. However, the Not 1/Nru I double digest disproves the hypothesis. The different
Not I fragments detected by Probes A and B/C cannot be simply due to a
single site between them because the Nru I fragment ofProbe A shifts down
when digested with Not I whereas the Nru I fragment of Probes Band C
remains unaltered. If all three were on the same Nru I fragment, each would
have detected a shift in size.