Abstract
We recommend the combination of performing globin mRNA reduction in peripheral whole blood samples and hybridizing on Illumina BeadChips as the practical approach for biomarker study.
Introduction
Peripheral blood is an ideal source for clinical biomarkers due to its ease of collection and its role in reflecting the body’s systemic immune and metabolic state. However, whole blood is a "noisy" medium for transcriptomics. Hemoglobin mRNA from red blood cells can account for up to 70% of the total mRNA population, creating a "hemoglobin dilution" effect. This overabundance masks the signals of lower-abundance transcripts—such as signaling molecules and transcription factors—that are often the most relevant for disease diagnosis. This study investigates a standardized solution to this interference using the Illumina BeadChip platform.
Methodology
The researchers analyzed samples from 11 healthy adults (8 females, 3 males) divided into two groups: pre-globin reduction (standard whole blood RNA) and post-globin reduction. The post-reduction group utilized the GLOBINclear™ Kit to physically remove hemoglobin $\alpha$ and $\beta$ mRNA via biotinylated oligonucleotides and magnetic beads.
All samples were hybridized onto Illumina Sentrix human-6 v2 BeadChips. The team employed the ReliefF algorithm for gene selection to identify gender-specific markers and used Real-time PCR (RT-PCR) to validate expression levels of four hemoglobin genes ($HBA1, HBB, HBD, HBE1$) and three differentially expressed genes ($AYTL2, CCL5, DYRK2$).
Key Insights and Results
Overcoming the "Pseudo-Binding" Effect
The study suggests that hemoglobin mRNA does more than just dilute other RNA; it likely causes "pseudo-binding" to microarray probes, creating high background noise that drowns out real data. Globin reduction addressed this by significantly lowering both intra-group and intra-sample variance, making the data much more reproducible.
Sensitivity Boost for Rare Transcripts
Globin reduction dramatically improved the detection of low-abundance genes. The average number of "present" gene calls rose from 8,987 (pre-reduction) to 11,921 (post-reduction)—an increase of nearly 3,000 detectable genes. Furthermore, 1,876 specific probes showed a consistent and statistically significant improvement in detection p-values after treatment.
Biological Signal Clarity
The most striking improvement was observed in gender separation. In untreated samples, gender-linked genes on the X and Y chromosomes were difficult to distinguish. After globin reduction, the biological signal became remarkably clear, allowing for near-perfect classification of male vs. female samples. RT-PCR confirmed that while the microarray showed some non-hemoglobin genes appearing to change expression after treatment, these were actually artifacts of the high hemoglobin noise in the untreated samples.
Conclusion
The combination of globin mRNA reduction and Illumina BeadChip microarrays establishes a robust and practical workflow for blood-based biomarker discovery. By stripping away the noise of abundant hemoglobin transcripts, researchers can significantly increase gene detection sensitivity and improve the resolution of biological signals. While this study was limited to 11 healthy subjects, the next logical step is to apply this platform to larger clinical cohorts, specifically for cancer biomarker detection in peripheral blood.
Summary updated