[Título] [Introducción] [Material y Métodos] [Resultados] [Discusión] [Iconografía] [Bibliografía]
Francisco Arrebola, MS.
Francisco Arrebola, Marco Masseroli, Francisco O'Valle, Mª Eugenia Reguero, Asunción Olmo, Mariano Aguilar, Raimundo G. Del Moral.
Department of Pathology, School of Medicine and University Hospital, University of Granada, Granada, Spain
Dpto. de Anatomía Patológica, Facultad de Medicina, Avda. de Madrid 11 - 18012 Granada, SPAIN
INTRODUCTION: Molecular determinations are gaining in importance as complementary diagnosis and research techniques in pathology. Identification of changes in mRNA expression requires internal controls on the quality of the reaction, normalization of the samples and an appropriate quantitation method. The most widely used of currently available techniques is the development of nucleic acid samples in agar gels labeled with ethidium bromide (EtBr). Given the high cost of methods for band quantification, and in some cases their complexity or inaccuracy, we have designed a simple method based on digital image analysis that allows the densitometric quantitation of bands of nucleic acids. The present work describes this method of image analysis and its validation.
MATERIAL AND METHODS: The image processing algorithms designed for the quantitation of bands of nucleic acids developed in agar gels were implemented using the program Visilog 4.0â for the development of image analysis applications. For the validation of the method we used a gel of "standards" labeled with EtBr and obtained by developing known serial dilutions of an RT-PRC product. A digital image of the gel placed on an ultra-violet light transilluminator was captured and the bands of nucleic acids on this image were quantitated densitometrically by means of the algorithms we designed. Of the quantitation values obtained, we evaluated their relationship with the amounts of "standard" truly present in each of the bands, and their reproducibility was assessed by statistical analysis of the results of four quantitations of the same image of the "standard" gel.
RESULTS AND DISCUSSION: The method presented here allows the densitometric quantitation of bands of nucleic acids developed in agar gels by means of digitization of an image of the gel and processing of the digital image for the segmentation of the areas of the bands present in the gel. The validation results showed that the densitometric values quantitated had an high linear correlation (r = 0.993) with the real amounts of "standard" present in each band. The analysis of proportions between amounts of "standard" and their values quantitated by optical density showed quantitation errors to be small and differences between replications not to be significant. Thus, this method offers an objective, reliable, accurate and reproducible quantitation of bands of nucleic acids.
CONCLUSIONS: The automated image analysis method illustrated here represents a simple instrument of great utility for the accurate and reproducible quantification of bands of nucleic acids developed in agar gel.
Key Words: RT-PCR, Image processing, Automatic quantification.
Arrebola is at the Pathology
Department of the School of Medicine, University of Granada, on a
predoctoral fellowship from the Spanish Ministry of Education and Culture. His main
research line is the study of drug nephrotoxicity and its relationship to drug
multiresistance, using in vitro experimental models.
We look forward to seeing you atThe Twenty-Second Annual Course on the Surgical Pathology of Neoplastic Diseases organized by the Memorial Sloan-Kettering Cancer Center in Granada (May 1999).