In situ hybridization method and buffer
An improved method of in situ hybridization which relies on an improved formulation of the in situ hybridization buffer is described. In at least some formulations the buffer are non-toxic. The combination of Locked Nucleic Acid (LNA) comprising ISH probes and the improved ISH buffer are useful for detection of small non-coding RNA as well as in the manufacturing of ISH kits directed to the detection of such small non-coding RNA. Further disclosed is a method of semi-quantitative ISH and demonstration of the semi-quantitative ISHs diagnostic potential.
1. A method for detection of nucleic acids by in situ hybridization of a fixed cellular specimen comprising a hybridization step which is performed in a formamide-free, hybridization buffer that comprises a chaotropic component of either urea, salts of guanidinium, salts of guanidine, or a mixture of two or more selected from the group consisting of urea, salts of guanidinium, and salts of guanidine, wherein the nucleic acids are small, non-coding RNA molecules.
2. The method according to claim 1 , wherein the cellular specimen is a formalin-fixed and paraffin-embedded (FFPE) tissue section.
3. The method according to claim 2 , wherein the chaotropic component is urea.
4. The method according to claim 3 , wherein the in situ hybridization utilizes a probe contains one or more LNA monomers.
5. The method according to claim 4 , wherein the method comprise a hybridization step wherein said cellular specimen is contacted with a hybridization-solution comprising:
at least one non-radioactive labelled probe comprising from 7 to 22 nucleotides which are capable of hybridizing to a specific RNA sequence and which comprise one or more LNA monomers;
a hybrid stabilizing agent selected from the group of salts of mono- and di-valent cations; and
urea in a concentration between 0.5 and 5 M.
6. The method according to claim 5 , wherein the hybridization probe is labeled at both the 3′ end and the 5′ end with digoxigenin.
7. The method according to claim 6 , wherein the method further comprises a step wherein a hybridization signal is visualized by formation of a dark-blue NBT-formazan precipitate, the specimens are optionally counterstained with nuclear fast red, and further comprises a quantification comprising the steps of:
taking between about 8 to about 17 random images from within the tumor area,
excluding random images which do not contain evident cancer cells,
excluding random images with tissue artifacts, staining artifacts, or both tissue and staining artifacts,
employing a supervised segmentation based on Bayesian classification trained to recognize blue pixels areas (NBT-formazan precipitate) and, if stained with nuclear fast red, red pixel areas (nuclear fast red coloration) and purple pixel areas (both NBT-formazan precipitate and nuclear fast red coloration), to estimate the blue pixel areas (B) and, if stained with nuclear fast red, the red pixel areas (R) and purple pixel areas (P), and
quantifying the relative level of the RNA by estimating either a total blue area (TB=B+P) and/or, if stained with nuclear fast red, a relative total blue area (TBR=TB/TR, wherein total red area TR=R+P
wherein the TB and/or the TBR is a measurement of the specific RNA level in the sample.
8. A method of predicting a disease-free survival of a stage II colon cancer patient comprising:
a) determining a relative level of miR-21 in at least one representative tissue section from a stage II colon cancer of said patient determined by the method of claim 7 ,
b) comparing the relative level of miR-21 in the patient to a set of relative levels of miR-21 from a reference panel of stage II colon cancer samples of patients with known disease history, and
c) grouping the reference panel in tertiles according to the relative level of miR-21;
wherein the tertile in which the miR-21 level of said at least one representative tissue section falls is predictive of the disease-free survival probability of said patient.