Evaluation of Cell-Free RNA as Minimally Invasive Biomarkers of Target Engagement

Abstract

Extracellular vesicles (EVs) are secreted by nearly all cell types and play a vital role in intercellular communication (Kalluri & LeBleu, 2020). EVs contain proteins, lipids, nucleic acids, and other biomolecules that reflect the physiological state of their parent cells (Wang et al., 2022). EV-associated RNA, as a protected form of circulating cell-free RNA, has become a promising minimally invasive biomarker for assessing target engagement. However, isolating EVs from biofluids remains difficult because of their complex composition (Li et al., 2017). Current isolation techniques, including ultracentrifugation, polymer-based precipitation, and size-exclusion chromatography, yield variable results regarding purity, yield, and reproducibility (Gao et al., 2023). These inconsistencies can impact downstream analyses (Gao et al., 2023; Welsh et al., 2024). Therefore, standardized methods are needed to improve reproducibility and reliability in EV-based target engagement studies. Additionally, Macaca fascicularis, or Cynomolgus (Cyno) monkeys, are often used as model organisms in clinical trial applications; however, establishing target engagement biomarkers for non-secreted targets in preclinical pharmaceutical models remains a challenge. To address this gap, this study will directly compare multiple EV isolation methods for serum and CSF and evaluate their potential for target engagement. Previous research has examined individual techniques such as ultracentrifugation, polymer-based precipitation, and size-exclusion chromatography (SEC); however, there is a lack of comprehensive studies comparing the expression of different targets. Commercially available kits provide standardized methods that are easily accessible to researchers. Three isolation methods—qEV, ExoQuick/ExoQuick Ultra, and ExoEasy Maxi—were tested for their yield and suitability in isolating EVs from Cynomolgus serum. In CSF, we compared EV isolation using qEV columns with direct RNA extraction from neat CSF and from concentrated CSF (Centrivap) to assess the detectability of brain-derived transcripts. Jess automated capillary western blot analysis was performed to confirm the enrichment of EV markers, including CD81, flotillin-1, annexin A2, and HSP70. RNA was extracted from each sample and analyzed by RT-qPCR, both with and without pre- amplification or nested PCR, to evaluate detection sensitivity for tissue-specific mRNAs. Through this approach, we aim to establish a standardized method for obtaining EVs and identifying tissue-specific mRNA across various biofluids. EVs were detectable across multiple isolation methods by RT-qPCR; however, most target transcripts showed Ct values beyond acceptable thresholds without pre- amplification or nested PCR—techniques used to increase the abundance of target transcripts from cDNA before qPCR. Incorporating pre-amplification significantly improved detectability, especially for low-abundance targets, highlighting the extremely low RNA content typical of EVs across tested biofluids. Comparison of isolation strategies in serum indicated that ultracentrifugation, polymer-based precipitation, and membrane affinity methods produced detectable EV RNA. In contrast, size-exclusion chromatography (qEV) often resulted in lower RNA recovery, as evidenced by higher Ct values. In CSF, direct RNA extraction or concentration using centrivap yielded the strongest signals and was comparable. All methods required pre-amplification to produce measurable Ct values. Probe design influenced transcript detectability for Target A, with some exon-spanning probes resulting in lower Ct values, emphasizing the importance of assay-specific considerations. Jess automated capillary Western blot analysis confirmed the presence of EV markers (CD81, flotillin-1, annexin A2, HSP70) in concentrated CSF. Housekeeping genes ACTB, GAPDH, and UBC were consistently amplified and suitable as internal controls. Overall, these findings provide a framework for evaluating cell-free RNA isolation and detection in serum and CSF.