The rapid progression of microarray technology was the focus of two recent conferences—Select Biosciences “Advances in Microarray Technology” held in Barcelona and the “World Microarray Congress” held in Vancouver. Both events emphasized the exploration of gene expression through microarray technology by advanced means that can provide enhanced results faster and more efficiently.
At the Barcelona meeting, Iain McWilliam, Ph.D., senior scientist at Arrayjet, described advances in microarray production using inkjet technology, which he characterized as having come of age. Once the preserve of specialist facilities, inkjet printing is now used by 20 Arrayjet customers globally in both research labs and microarray production facilities.
The technology uses noncontact printing within a flexible platform to produce diverse microarrays at high speed. The Arrayjet systems, Dr. McWilliam noted, are low maintenance and are used in both R&D and production settings. He listed nucleic acids, peptides, intact cells, and proteins—including “obnoxious proteins such as blood serum and cell lysates”—among the arrays that can be produced. Common substrates include coated glass slides, NC slides, SPR prisms, unbound NC membrane, and silicon wafers.
In another example, which concerned the application of microarray technology to hybridoma production for monoclonal antibodies, Dr. McWilliam described how protein microarrays improved throughput per technician 10-fold (from 20 to 200 targets per technician per year) by reducing processing time from four to seven months to just seven weeks. This method was developed by Federico De Masi, Ph.D., of Alan Sawyer’s group at the EMBL Monoclonal Antibody Core Facility in Italy.
Arrayjet microarrayers place spots in a precise and repeatable manner and, in the case of the Ultra-Marathon system, at a rate of one 384-well plate onto 1,000 slides in less than 90 minutes, reported Dr. McWilliams. The key to the system is the JetSpyder™, a nanoscale liquid-handling unit that enables aspiration of 12 or 32 samples simultaneously, he added.
A XaarJet inkjet print head (Xaar) mates to the JetSpyder via vacuum, and probes are aspirated from source plates (up to 48 x 96 or 384 wells). The JetSpyder is then undocked leaving the print head, containing 12 or 32 probes, primed and ready to print. Probes are delivered via noncontact, on-the-fly printing onto the substrate to the user’s specification. The accuracy of printing is such that in subsequent print runs probes can be overlaid onto the same spot locations to create multilayered assays. Additionally, the system aspirates each probe into multiple nozzles within the print head, enabling flexible dispensing from 100 pL to 600 pL per spot, according to the company.
Furthermore, no changes to the system are required when switching between sample types or when changing print layouts. The recently released Array Multiplier™ software enables the user to divide each slide or substrate into a number of areas, each of which can be independently addressed by the microarrayer during the production step, producing slides with multiple mini-arrays printed on them and thereby greatly increasing sample throughput, Dr. McWilliams said.
The system is automatically cleaned between aspirations during printing, as well as at the beginning and end of each print run: the print head and JetSpyder capillaries are stringently washed with buffer—both independently and when docked together—to eliminate carryover contamination, even when arraying sticky protein solutions or cell lysates.
aCGH Method for FFPE Samples
Also presenting at the Barcelona conference was Ernesto Guzmán, R&D scientist at Invitrogen, who described “An optimized aCGH method for use with FFPE tissue samples.”
Dr. Guzmán noted that there are several challenges posed in working with FFPE samples. The wide variety of fixing protocols affects DNA integrity, and formalin composition and fixing times vary. FFPE samples provide limited DNA quantities (5–100 ng, depending on tissue type) and DNA degradation occurs due to age and formalin treatment (overtreatment).
Dr. Guzmán’s group used a novel method to purify and label FFPE DNA for array comparative genomic hybridization (aCGH). The purification uses an FFPE-specific protocol with the Invitrogen PureLink Genomic DNA purification system coupled to a modified version of the company’s BioPrime Total Labeling kit.
In aCGH, test and reference DNAs are Cy5 and Cy3 fluorescent labeled, respectively, and hybridized to DNA arrays (BAC or oligo) that are scanned prior to analysis. In his presentation, Dr. Guzmán showed a variation at one end of chromosome 5 and noted that this technique can be applied to the whole genome to detect inherited defects such as Downs Syndrome where there is an extra copy of chromosome 21.
Citing the wealth of knowledge trapped in FFPE samples regarding cancer, for example, Invitrogen is working to isolate DNA from these samples as effectively and efficiently as possible. The results Dr. Guzmán presented indicate that significant progress is being made. The Invitrogen process uses Alexa dyes, so samples fluoresce as brightly as possible, and Random Prime Amplification, which requires less DNA and introduces less bias than other processes such as whole genome amplification.
In summary, Dr. Guzmán noted that BioPrime® Total genomic DNA labeling for FFPE samples results in higher yields and higher dye incorporation with a quality controlled, all-in-one kit. Array performance is equal to or better than competitor kits in a more cost-effective, convenient package, he claimed, and the Invitrogen process allows for a wider range of input genomic DNA in labeling reactions while maintaining performance oligo arrays.
After introducing Eppendorf Biochip System’s DualChip® Microarray System to the “Microarray World Congress,” principal scientist Peter Herzer, Ph.D., asked some basic questions: “Which genes and pathways are responsible for regulating a specific disease? What is the difference between a normal cell and a tumor cell?”
The DualChip system measures gene expression in two sets of identical reference RNA to differentiate between healthy tissue and, in the example provided by Dr. Herzer, human breast cancer tissue, classifying the cancer by type and stage to facilitate individualized therapy. The arrays are prespotted, low-density, pathway focused theme arrays with fewer than 300 genes and are selected based on actual research and publications.
DNA probes are from 200 to 400 nucleotides in length. Each probe is spotted in triplicate onto the array with a pre-attached chamber for easy handling, all bar-coded for documentation within an extensive control system. Arrays come in three formats: predefined (between 68 and 364 genes); semicustomized with the addition of up to 20 genes to predefined content; and customized, which provides the choice of up to 364 genes of interest. Currently, kits are available for 16 human and mouse disease-related conditions, Dr. Herzer added.
He also explained the functionality of the Biochip System’s transcription factor (TF) arrays, which differ from gene-expression arrays in that they are double-stranded DNA probes with TF binding sites. Transcription factors are usually inactive in resting cells. Upon activation, they usually undergo nuclear translocation and bind to specific dsDNA sequences in their target genes.
Activation mechanisms are fast events (as opposed to gene expression, which are slow events), allowing cells to react quickly to external stimulations or aggressions. Biochip System’s TF Chips exclusively detect the activated form of TFs. According to Dr. Herzer, no one else offers multiplexing to detect the activated form of TF.
Dr. Herzer also discussed Silverquant technology, which is up to 10-fold more sensitive than fluorescence detection in gene-expression experiments. In the Silverquant process, bound TF or labeled cDNA is detected with antibiotin gold-conjugates and subsequent staining using a silver staining solution. The gold particles catalyze the reduction of silver nitrate and metallic silver settles at the place of the reduction reaction.
The high sensitivity allows gene-expression analysis with as little as 1–2 µg of total RNA without the need for amplification, Dr. Herzer claimed. The process produces low background, has a dynamic range of approximately 3 logs, and is compatible with a wide range of substrates, including plastics, with no increase in background, he added. Arrays detected with Silverquant are not affected by light or ozone and can be archived indefinitely.
Yong Wu, Ph.D., staff scientist for GeXP development and applications at Beckman Coulter, spoke at the “World Microarray Conference” on the process “from microarray screening to multiplexed, quantitative biomarker validation.”
The trend in gene-expression studies after whole genome microarray assay, he noted, is to focus on genes that are biologically relevant—usually 10 to 500 associated with a particular disease and about 10 to 50 associated with a pathway or specific response. It is highly desirable in a routine testing environment to produce highly quantitative assays with increased sample throughput and substantial time and cost savings.
The GenomeLab™ GeXP Genetic Analysis System by Beckman Coulter, Dr. Wu stated, enables quantitative, multiplexed gene-expression analysis with up to 30 genes per reaction and up to 192 samples per run. Using only 5–50 ng RNA, the GeXP system delivers highly quantitative and highly reproducible results and is sensitive enough to detect changes as small as 0.5-fold.
An overall average %CV of less than 10% in a multiplex assay demonstrates “the superb reproducibility of this platform,” Dr. Wu said. Other than gene-expression studies, the GeXP system can also perform sequencing and SNP, STR, AFLP, and MLPA analyses. “This multifunction platform requires only one gel, one array, and one software package.”
Citing a specific study, Dr. Wu noted that a multiplex containing 21 functional genes and three housekeeping genes was developed for the quantitative validation of breast cancer biomarkers. These genes are associated with breast tumor progression including cell cycle, differentiation, apoptosis, DNA repair, angiogenesis, and modulation of extracellular matrix.
GeXP delivers high-quality linear correlation between the known input value (the experimental input RNA amount) and the measured output value (gene-expression level), with a typical linear correlation coefficient above 0.999, Dr. Wu said. In addition, ten consecutive 0.5-fold increases in RNA concentration were accurately quantified for all 24 genes in the multiplexed assay. RNA samples derived from normal breast tissue, invasive ductal carcinoma tumor, and tumor cell lines were analyzed, with the gene-expression-profile results displaying the capability of this multiplex assay to differentiate normal from tumor samples, he added.
A subset of the genes including KNTC2, PRC1, Col4A2, and RAB6B were identified as potential biomarkers of ductal carcinoma tumors. Summing up, Dr. Wu stated that the capacity to perform multiplexed, sensitive, and precise gene-expression analysis opens a new door for scientists to explore subtle, yet biologically meaningful changes in cancer biomarker studies in an effective manner.
Identifying Respiratory Pathogens
Finally, at the “World Microarray Congress,” Agnieszka Lichanska, staff scientist at Tessarae, described the company’s resequencing microarray technology for the simultaneous detection and definitive identification of respiratory pathogens. To date, TessArae has completed two large (>400 samples) clinical studies that demonstrate several unique benefits of the technology, Lichanska said, based on the generation of results as nucleotide sequence information rather than as a fluorescent signal from hybridization of a probe to a target.
TessArray Resequencing Pathogen Microarray (RPM) technology provides several unique capabilities versus immunoassay and RT-PCR including simultaneous detection of viral and bacterial pathogens and primary and co-infecting pathogens, definitive identification of the pathogen to the strain or type level as well as sequence variants, MLST-like epidemiological analysis, sensitivity equal to or greater than PCR, and same-day results with zero false positives, according to Dr. Lichanska.
Using the example of Hemagglutinin A/Goose/Guangdong/1/96/H5N1, Dr. Lichanska described the RPM technique. “For any given sequence, we tile eight probes for each possible base on both strands. This tiling of eight probes is marched down the sequence of interest, effectively sequencing and genotyping at every base position,” she told the conference. Dr. Lichanska went on to describe how the system works.
First, total pathogen nucleic acids are extracted from clinical samples, followed by reverse transcription of RNA genomes and locus-specific PCR amplification of all pathogen signatures and hybridization to an RPM array. The nucleotide sequence is called from the signal on the RPM array and compared to the TessArae database. Pathogens that are present in the sample are reported as highest match at genus, species, strain, and/or serotype levels.
Matthew Lorence, Ph.D., vp of marketing and sales, noted that one application of the RPM technology has involved examining the genomic diversity of Haemophilus influenzae, a typical bacterial commensal organism, at U.S. military training centers.
Recruits from around the country bring in geographic substrains of the organism, which can be detected by multiple base variations in the H. influenzae genomic sequence. Upon arrival, many recruits are exposed to Adenovirus Type 4, which is circulating within the training center. This is shown by the clonal nature of that virus exhibiting identical genomic sequence in all isolates from that same population.