Siemens has developed a fully-automated nucleic acid purification method which supports routine clinical diagnostic analysis in the molecular pathology laboratory as well as paves the way for extensive retrospective biomarker studies of archived formalin-fixed, paraffin-embedded (FFPE) tissue material with long-term clinical follow-up data.

The key features of this fully-automated walk-away extraction system are:

Content

The identification and validation of molecular markers in formalin-fixed, paraffin-embedded (FFPE) tissue is currently an area of intense and exciting research activity. This is due in part to the fact that expression profiling, genotyping, and mutation analysis have been shown to aid in diagnosis and to provide guidance, particularly in the treatment of cancer. Despite this concentrated effort, to date, only a limited number of individual markers or panels of markers using extracted nucleic acids from histopathological tissue specimens have been introduced into routine clinical practice (e.g., OncotypeDX®; k-RAS mutation analysis for EGFR based targeted therapies). FFPE tissue is an invaluable tool in translational research of DNA, RNA and microRNA based biomarkers because FFPE tissue is readily available through routine clinical diagnosis. Both logistic facilities and easy access to large archives of FFPE tissue from various disease entities linked to clinical and follow-up databases render this material as particularly attractive.

Protein expression in pathological specimens by immunohistochemical (IHC) detection is well established and currently part of the diagnostic routine. In contrast, analyses using DNA and, in particular, mRNA extracted from FFPE tissue remains difficult because the fixation process introduces crosslinks between proteins and nucleic acids. This cross-linking chemically modifies numerous molecules and fragments the nucleic acids. Furthermore, many methods for extracting nucleic acid from FFPE tissue, particularly those that are highly manual, are not standardized protocols (e.g., they include variable lysis times and different chemistries for the extraction of RNA and DNA). In addition, these manual protocols are time consuming and require the use of hazardous and flammable materials like xylene and ethanol for deparaffinization. Such factors have required individualized adaptation to create standardized protocols. As a result, these isolation protocols have been incompatible with high-throughput formats.

The difficulties of isolating nucleic acids from FFPE tissue samples have been addressed by Siemens with the development of a fully-automated method for use in molecular pathology and companion diagnostic testing, both in routine clinical diagnostic applications as well as for research use.

The foundation of this technology is the proprietary silica-coated paramagnetic particles (PMP) developed by Siemens. These PMPs, produced by employing a simple and robust manufacturing procedure, simplify and improve automated nucleic acid isolation by virtue of the particles properties, particularly the nanolayer of silica absorbed to the particles. Figure 1 shows a comparison of the Siemens PMPs as compared to other commercially available beads. Nucleic acids are selectively bound under chaotropic conditions (positive selection); however, "dirt or tissue debris" is also bound under non-chaotropic conditions excluding target (negative selection; see below). The optimal suspension behavior of the particles is due to their very small size, the homogenous particle size distribution, and the strong magnetization properties. A universal extraction chemistry encompassing lysis, washing and elution steps has been developed to efficiently extract both DNA and RNA simultaneously with high recovery and purity from a broad range of specimen types (shown in Figure 2).

A major advantage to the Siemens FFPE tissue extraction method is the workflow which provides for the full integration of the de-paraffinization step into the extraction procedure providing for a process that is fully-automated. The de-paraffinization step works through a simple and efficient hydrophobic absorption of the molten paraffin to the inner polypropylene tube wall of the sample tube during the lysis process. This step not only replaces the harmful and laborious xylene/ethanol processing steps but also removes the need for centrifugation. Thus, the Siemens process is more amenable to full automation. It also allows for the loading of FFPE sections immediately following microtome cutting.

With "solid" FFPE samples, the tissue input is difficult to standardize because the cell/matrix composition, fixation, and storage conditions can be vastly different. As a result many other methods require variable incubation times to achieve complete lysis. Another key feature of the Siemens technology is the abililty of the PMPs, in the absence of chaotropic salts, to bind and eliminate any undigested tissue (debris) (referred to as negative selection). This negative selection is a key requirement for the effective and complete automation as any insoluble material can interfere with accurate liquid handling and result in clogging of pipette tips.

The Siemens platform uses specialized hardware which provides for the loading and lysis of FFPE sections in 1.5 mL plastic tubes that are loaded directly on the system. The system enables a throughput of up to 48 samples in less than 4 hours inclusive of a 30 minute incubation step for DNase I digestion for the isolation of purified RNA (with as few as 8 samples being processed in 3h 15min). If DNA is the desired result, the total time for processing of 48 samples is reduced to 2h 40 min because the DNase I step is not required. The universal extraction chemistry allows for the isolation of both DNA and RNA from the same sample. A simple software query applies the automated DNase I digestion step for accurate expression analysis. For typical DNA applications such as mutation and polymorphism analysis by classical or ultra deep sequencing, the total nucleic acid containing eluate can be applied.

The fully-automated extraction can process various sample input formats including standard FFPE sections, tiny tissues such as Tissue Micro Array (TMA) cores, and fixed sections from needle biopsies. Even laser capture microdissected cells, snap frozen tissue and cells from culture can be used for extraction.

In addition to the workflow advantages of complete automation with high throughput, the method also recovers high amounts of DNA and RNA with excellent purity. The success rate to extract PCR detectable RNA or DNA from several thousand FFPE samples of up to 30 years of age has been shown to be appr. 99% (ref 1,2,3,4). A standard eluate of 100 μl normally allows one to successfully run hundreds of PCR reactions from one tissue section.

Overall, the Siemens fully-automated nucleic acid extraction method provides a number of advantages when compared to state-of- the-art extraction methods from other manufacturers. Key advantages include full automation including an integrated xylene-free de-paraffinization step, high sample throughput of up to 48 samples in less than 4 hours, a consolidated and universal extraction chemistry for co-isolation of DNA and RNA, a standardized lysis time of one hour, a negative tissue binding/selection step circumventing centrifugation, flexible sample input with regards to tissue type and format, superior section-to-section reproducibility, reduction of carryover risk and compatibility with subsequent assay technologies like kinetic (RT-) PCR, endpoint PCR, sequencing and microarrays. The Siemens nucleic acid purification method paves the way for extensive retrospective biomarker studies of archived FFPE material with long-term clinical follow-up data. Furthermore and importantly, it supports routine analysis of this material in the clinical or molecular pathology laboratory.

Contacts:

Figures

Figure 1: Electron microscopic pictures of Siemens magnetic iron oxide beads coated with a nanolayer of silica (left) versus typical silica containing magnetic particles from other manufacturers (right). Note the different scale of the picture bars.