Genotyping with hybridization probes – Principle

The unique potential of hybridization probes unfolds when used to analyze sequence variations. This is the case not only for point mutations, or deletions and insertions, but also for other sequence variations such as related sequences or splice-variants.

The principle of the hybridization probes is conceivably simple – the ‘sensor’ probe covers the variable target sequence and the adjacent anchor probe has a distinctly higher melting temperature. The analysis of the melting point is, therefore, exclusively dependent on the "sensor" probe. Hybridized probes are slowly heated while the fluorescence is constantly being measured. Every difference of the sequence covered by the "sensor" probe will cause a lowering of the melting temperature. A homogenous full matching probe will produce a single defined peak; a heterozygote probe will result in two distinctive peaks, a homozygote mutant will again yield a single peak, albeit at a lower temperature. A single nucleotide difference will reduce the melting temperature by 2 to 10 degrees centigrade, depending on the nucleotide and its neighboring bases.

However, only variants represented in comparable amounts can be determined. This is the case of genetic (allelic) analysis. For determinations of contaminations, rare variants as in the case of cancer diagnostics (Minimal Residual Disease), or in mixed populations, the method has limited applicability. In these cases the selection of allele- or type-specific primers, to ensure amplification and quantification of underrepresented variants, should be applied. The area of the melting curve is not recommended to use as a calculation for the relative quantification, since probes often display different affinities to the target sequences. However, in established assays the area of the curves can be used to deduce simple relations like 1:1 or 2:3 common for the analysis of trisomy.

Genotyping with hybridization probes – positioning

The position to be analyzed should be more or less in the center of the "sensor" probe; at any rate not closer than four to five bases from either end. The "center" of the probe can also be defined by the binding strength; a GC-rich arm could be much shorter.

In principle there are not many placing variations for the probes, with the "anchor" to the left or the right or respectively on the opposite strand with the "anchor" left or right of the "sensor" probe. "Unsuitable" sequences, particularly surrounding palindrome or stem-loop forming sequence structures, are not easily avoidable. For example a mutation embedded in a strong binding region would be difficult to detect. Computer programs that allow the calculation of local binding stability definitely provide a tool for the optimization of the probes. Strong binding motifs can be weakened by the selective and deliberate substitution with weaker binding base analogues or even wrong bases.

Generally it is to be observed that a G-T mismatch is relatively good tolerated since these two bases are able to build a hydrogen bond. The temperature decrease will be only one – four degrees centigrade. Instead of designing a probe with "G" or "T" the mutant sequence should be selected or the probe based on the opposite strand resulting in a "C" or "A" mismatch.

Genotyping with hybridization probes – specificity

Deletion and insertions create smaller temperature differences than substitution mutations. Particularly in the case of monotonous sequences, e.g. the 4G/5G polymorphism in the PAI-1 gene (Nauck et al., 1999) or in the Connexin-26 35G del with five or six guanosine bases with a very difficult detection.

Other mutations covered by the probe will often lead to a different temperature profile as in the case of the HFE variant Ser65Asp, which was discovered because of the aberrant temperature profile (Bollhalder et al. 2000). Unknown mutations could, in principle display the same temperature profile, leading to a false conclusion. When using a mutant specific probe any additional mutation can result in a wild-type similar pattern, a wrong interpretation. In medical diagnostics it should be taken into consideration to carry out the analysis in the presence of both the wild-type and the mutant probe labeled with different fluorophores, thus allowing the simultaneous analysis by using the channels F2 and F3 of the LightCycler®.

Examples for critical targets:

Exon 10 of the human CTFR gene contains a deletion (Phe508del), a variant 508 Ser as well as other neighboring deletions and a non-pathogenic polymorphism. A wild-type specific probe will not easily resolve all these variants.

The human k-ras gene displays in codons 12 and 13 a variety of different mutations, all detectable with a single wild type specific probe. A mutant specific probe will not clearly distinguish the wild-type from the other mutations.

Both, the 3’-fluorescein as well as the 5’-LightCycler® Red labeled probes can be used as a "sensor". The former have the advantage of being less expensive and have a shorter turn around time. They are ideal for assay optimizations that require several different sensor probes or as in the case of rearrangement analysis in leukemia diagnostics, where patient specific probes are required. LightCycler®-dye labeled sensor probes have the advantage of allowing simultaneous single capillary analysis with different fluorophores.

In conclusion it should be considered that individual base exchanges may have an effect on the amplification of an allele, or that the probes will have different binding efficiencies on the variants. For example, due to secondary structures resulting from the mutation(s), which in turn may lead to a reduced binding efficiency or a binding inhibition and consequently no detection. Furthermore, mutations or deletions at the primer binding sites, particularly when located in intronic sequences and therefore potentially less conserved, thus preventing the amplification of an allele, may lead to the wrong conclusion of a homozygous situation (Intron 4 mutation of HFE, Jeffrey et al. 1999).

Remarks for genotyping probes

The melting curve is accomplished after the PCR - therefore the competition between primers and probes is less relevant. ***That mean for The positioning that that probes and primers can even overlap in extreme cases, means for the melting temperatures that they do not have to be higher than those of the primer.***

Probe Tm higher then primer Tm.

A lower Tm of the probes will not permit the observation of the amplification, though it will not have an effect on the melting curve analysis.

The position is not so relevant.

The distance to the primers has no effect on the melting curve analysis.

The anchor probe should have a noticeable higher Tm than its counterpart, the sensor probe (4-8 degrees centigrade).

The melting curve should only describe the binding of the "sensor" probe.

The sequence variation can be anywhere under the sensor probe, but not closer than 4 bases from either end.

A mismatch separates the cooperation between both parts of the sequence. Terminal positions have a smaller influence on the melting temperature.

Avoid G-T mismatches. Use instead the mutant specific probe or the opposite strand (C-A mismatch).

G-T mismatches are better tolerated and result in low Tm shifts (2-4 degrees centigrade in contrast to 5-10 degrees centigrade.

Avoid very strong-binding sequences within the sensor probe. Shift the probe position, introduce weak binding base analogues (such as inosine), or introduce mismatches to weaken a strong binding motif.

If the sequence alteration is near a tight-binding motif, smaller Tm shifts than expected may occur.

Avoid stem-loop-forming sequences.

Stem loops may preferentially affect binding to one allele, resulting in an unequal distribution of probe between wild type and mutant alleles.

The amplification primers should not be placed over variable sequence regions.

Stem loops may preferentially affect binding to one allele, resulting in an unequal distribution of probe between wild type and mutant alleles.

Unknown mutations.

Other sequence variations could cause the same effect, so that one should use mutated and wildtype probes for important questions parallel.

Testing and establishing a new assay

When designing a new Real-Time assay it is advisable to, either start from a conventional assay, where the efficiency and specificity of the primers have already been established; or, if it is unavoidable to design a complete new system, optimize the assay in two separate steps. Where step one is a conventional PCR analyzing the results on agarose gel electrophoresis. Once the functionality of the primers has been assessed the second optimization step with probes on the LightCycler® can be addressed. It is recommended to submit every amplification to a final melting curve analysis until an assay has been thoroughly validated. When fluorescence detection fails during the amplification and melting curve analysis the content of the capillary should be analyzed by agarose gel electrophoresis. The LightCycler® like any other Real-Time PCR instrument is not more sensitive than a conventional PCR. If a product is not detectable in a gel it will not be detectable with the LightCycler®.

  • use established primers. When using new or not validated primers test the amplification product in an agarose gel.
  • When designing a new assay test and compare several combinations of amplification primers
  • When using new probes review concept and label / channel
  • When testing new assays start with "medium" target concentrations. Avoid "single" copy concentrations as well as overloading the system. The "maximal" target concentration for 20 ul reaction volume is 1 ug DNA (corresponds to ca. 106 human genomes , 6 pg /diploid genome)
  • Lyophilized primers and probes should never be frozen. Store dark at ambient temperature
  • Reconstitute primers and probes in sterile ddH2O or in TE buffer (10 mM Tris-HCl, pH 8.0, 1 mM EDTA). Freeze aliquots at -20 degrees centigrade for long term storage. Multiple freeze/thaw cycles are discourage