Is 1 mL of plasma enough to extract exosomes for RNA sequencing?

We previously processed 1 mL of clinical plasma using a membrane affinity column, diluted it to 2 mL before loading onto the column, and were able to extract RNA, but the Qubit values were usually near the detection limit. After library preparation, the main problem was a relatively high proportion of adapter dimers, requiring strict gel cutting or magnetic bead selection. In the sequencing results of hemolyzed samples, the proportion of red blood cell miRNA was very high, and during analysis the differential gene list would be skewed. It is recommended to clearly record in the experimental notes the time from blood collection to centrifugation, hemolysis status, RNA concentration, and library preparation input amount. For precious clinical samples, it is best to first take out a portion for a pilot experiment rather than extracting all at once. If conditions permit, follow-up samples from the same patient can be collected for validation.

1 mL of plasma for small RNA sequencing is theoretically worth trying, but the risk is not low, especially for clinical samples that have undergone freeze-thaw cycles. The specification of more than 2 mL for many kits is not an absolute lower limit, but rather to ensure the total amount of RNA required for library preparation. It is recommended to first conduct a pilot experiment with one or two samples of good quality: extract from 200–500 µL plasma, assess small RNA fragment yield using Qubit and capillary electrophoresis, and then decide whether to proceed with sequencing on all samples. If the total RNA amount is near the detection limit, the number of PCR cycles during library preparation will increase, and the proportion of duplicates and adapter dimers may also rise. For the extraction method, it is best to choose a membrane affinity column to avoid carrying over large amounts of lipoproteins and plasma proteins with precipitation methods. Hemolyzed samples must be recorded, as this will affect the subsequent miRNA ratio.

For small RNA sequencing of 1 mL plasma, a failure contingency plan is essential. Insufficient total RNA and inadequate input for library preparation can lead to low effective reads and low alignment rates in the sequencing data. It is recommended to first validate using healthy plasma to confirm that the extracted RNA can reliably enter the library preparation workflow; when processing clinical samples, it is best to include one replicate or QC per batch. You can ask the library preparation service provider about the minimum total RNA amount and volume they accept. Some low-input small RNA library preparation kits can handle lower concentrations, but the cost will increase. For extraction, try to choose membrane affinity columns with small elution volumes. If RNA quality is suboptimal, stepping back to detect 3-5 candidate miRNAs by qPCR may be more reliable than forcing it into sequencing.

I would not recommend using a precipitation method for 1 mL of plasma. HDL, lipoproteins, and contaminating proteins in plasma will co-precipitate, and the extracted exosomal RNA is prone to contamination by free miRNA and protein-bound miRNA. You may prioritize membrane affinity columns/size exclusion chromatography, as the sample volume requirements are generally more flexible than those of precipitation methods. During processing, first centrifuge at 3000g to remove platelets, then bring up to the minimum working volume with PBS; if the column volume is relatively large, 1 mL of plasma can be diluted to 2 mL with PBS before loading. When eluting RNA, use the minimum elution volume to reduce losses during the concentration step. After extraction, take only a portion for QC and immediately freeze the remainder. Focus on the small RNA fragment distribution rather than total RNA concentration, because exosomal RNA inherently lacks clear 18S/28S peaks.

Hemolysis can severely affect plasma exosomal RNA results. Erythrocyte-derived miRNAs such as miR-451a and miR-144 are substantially elevated in hemolyzed samples, and after library preparation they can dilute or interfere with true exosomal miRNA differences. If a 1 mL sample is already visibly yellowish or reddish, the RNA yield after extraction may appear normal, but it is in fact contributed by erythrocyte RNA. It is recommended to record the hemolysis grade at sample collection, or measure OD414, and to remove hemolysis-related miRNAs or include them as covariates during analysis. If a sample is severely hemolyzed, prioritize qPCR validation of a small number of candidate miRNAs rather than blindly proceeding to whole small RNA sequencing. Extraction kits cannot "correct" hemolysis-induced bias; this can only be addressed through sample quality control and analysis strategy.