Credit: Graham Colm
Storing blood samples at room temperature can induce changes that may cloud research findings, according to a group of investigators.
The team initially found that blood samples from leukemia patients had high levels of malformed RNA, a discovery they believed could explain leukemia’s origins.
But additional research showed this abnormality was a result of storing blood samples at room temperature for hours, or even days, prior to processing.
Heidi Dvinge, PhD, of the Fred Hutchinson Cancer Research Center in Seattle, and her colleagues reported these findings in PNAS.
The investigators had searched databases to collect genomic information on leukemia patients and healthy control subjects. In all but one of the datasets they analyzed, the team found high levels of abnormal RNA in leukemia cells.
Another finding was that samples from pediatric leukemia patients had the highest levels of abnormal RNA. And this led the researchers to speculate about the cause.
They realized that pediatric leukemias are rare, so blood samples can be difficult to obtain. Therefore, the samples in the databases were collected at facilities throughout the world and shipped to where they were needed. So the samples could be stored at room temperature for days at a time.
To confirm that this practice can affect blood samples, the investigators conducted an experiment. They collected samples from 4 healthy subjects (2 men and 2 women) and looked for differences between samples that were processed immediately and samples that sat in the lab at room temperature for up to 48 hours.
Sure enough, the team observed changes in the stored samples, even if they were only stored for 4 hours. Stored samples exhibited biased activation of biological pathways and upregulation of pseudogenes, antisense RNAs, and unannotated coding isoforms.
Storage affected a number of genes that play roles in biological pathways relevant to leukemia, including cytokine production, NF-κB signaling, chromatin modification, and RNA splicing. Additionally, storage inhibited RNA surveillance, leading to the genome-wide expression of normally degraded RNAs.
The researchers said these findings coincide with the database findings, as they observed inhibited RNA surveillance in all but one of the datasets they analyzed.
The samples from that dataset were processed immediately after collection. And samples from healthy subjects were processed immediately, which explains why those samples were normal as well.
The team also noted that they did not observe inhibited RNA surveillance in lymphoma or solid tumor datasets.
These results suggest previous research utilizing these types of databases may contain errors, and future work making use of these databases could be affected as well.
Fortunately, the investigators found that putting blood samples on ice can prevent the negative effects they observed. The group also identified biomarkers that indicate prolonged storage, so researchers can look for those biomarkers if chilling blood samples is not an option.
In addition to describing these findings in PNAS, Dr Dvinge and her colleagues are planning to present their work at the upcoming ASH Annual Meeting.
