In a time of a global pandemic, with antivaxxer and anti-science sentiments running rife, and when combinations of fear, distrust, and paranoia are rampant, it is easy for important results from basic science to become fodder inaccurately supporting the doomsday memes that abound. This is the story of a scientific paper from two eminent scientists that, properly understood, explains a puzzling aspect of SARS-CoV-2 biology in COVID-19 patients: why can viral genetic information be detected in some patients months after they have recovered from their illness, using sensitive testing (polymerase chain reaction, or PCR; for reference see CSI). As it turns out, it is not evidence that SARS-CoV-2 is manmade, or that the various vaccines against the disease are toxic or disease-causing, or that (as Senator Rand Paul (R-TN) is (in)famous for charging) Dr. Anthony Fauci supported "gain-of-function" research in Wuhan China and then tried to cover it up. There are plenty of websites having that content; here we will strive to stick to the facts.
The paper, "Reverse-transcribed SARS-CoV-2 RNA can integrate into the genome of cultured human cells and can be expressed in patient-derived tissues," appeared in the Proceedings of the National Academy of Sciences USA this week, and was authored by Rudolf Jaenisch and Richard Young and their colleagues. The purpose of the paper as stated therein is to resolve "[a]n issue of SARS-CoV-2 disease is that patients often remain positive for viral RNA as detected by PCR many weeks after the initial infection in the absence of evidence for viral replication," i.e., in the absence of producing live virus. Their hypothesis: perhaps enzymes produced in human cells, capable of converting RNA (like messenger RNA) to DNA, was introducing some portions of the SARS-CoV-2 RNA into the genome of infected cells. This RNA would then be detected after the RNA was made as part of gene sequences surrounding the integration site.
It should be appreciated in this regard that a great deal of human DNA, colloquially termed "junk" DNA, does not encode proteins or regulatory sequences (which is the function of the remainder of genomic DNA and was what earlier workers believed to be the totality of the genome). This seemingly non-functional DNA is made up in significant part by the remnants of ancient retroviruses and retrotransposons, DNA elements "reverse transcribed" from RNA to DNA and inserted in the genome (perhaps randomly). Some of these species form so-called repetitive elements; one of the most prevalent in human DNA (17%) named Long Interspersed Nuclear Element 1 (LINE1):
These repeats, like their retrovirus and retrotransposon ancestors, can express a form of reverse transcriptase that can act on cellular RNA and result in reinsertion of the resulting DNA into the cellular chromosome (this is not an unknown phenomenon and is not necessarily associated with virus infection when it has been detected; see, Horie & Tomonaga, 2011, "Non-retroviral fossils in vertebrate genomes," Viruses 3, 1836–1848). Importantly to this work, other RNA viruses and even cellular genes have been known to have portions of their RNA retroinserted into cellular genomic DNA after reverse transcription by endogenous cellularly encoded reverse transcriptases.
Drs. Jaenisch and Moore first studied whether SARS-CoV-2 RNA could be reverse transcribed and reinserted into the genomic DNA from cultured human cells in a Petri dish; the cells they used are an established, immortal cell line known as HEK293T. These cells arose from human embryonic kidney and differ from normal cells by not experiencing aging, senescence, and cell death after a limited lifespan. They also contain DNA from a transforming adenovirus, having been made in 1973 and continuously grown (or "passaged") since then. Their chromosomal complement is abnormal, having on average 64 chromosomes and multiple copies of some of them (e.g., 3 copies of the X chromosome and 4 copies of chromosomes 17 and 22). In short, these are not normal cells but nevertheless have proven to be very useful for a variety of purposes in cell biology studies and biotechnology.
In the COVID-19 studies reported in this recent paper, these scientists increased the likelihood of observing what they expected to be extremely rare events by introducing into recipient HEK293T cells a cloned copy of human LINE1 genetically engineered to express the endogenous reverse transcriptase (it will be appreciated that such cells do not occur naturally). When assayed, using PCR detection or by genomic DNA sequencing methods, the largest SARS-CoV-2 derived DNA sequence was 1,632 basepairs (bp) encoding the protein (the nucleocapsid, NC) that acts as a "shell" for the viral RNA, in a genetic milieu that validated the hypothesis for LINE1-mediated reverse transcription and integration into the genome. These studies also showed that portions of human genes called exons were preferential (76%) targets for integration, providing an explanation for why PCR treatment of cellular RNA could detect portions of SARS-CoV-2 RNA in otherwise virus-free individuals post-infection. Similar results at lower frequencies were detected in HEK293T cells not having artificial LINE1-associated reverse transcriptase encoded by a genetically engineered LINE1 element.
When the presence of chimeric human/SARS-CoV-2 RNAs were assessed in samples from infected individuals they were found at very low frequency -- 0.004–0.14% of the total detected virus-related species, although they were found in multiple tissue types including lung, heart, and brain tissues. Most of these detected sequences were from NC-encoding RNA, consistent with the HEK293T cell experiments. That the SARS-CoV-2 RNA detected in human samples had been produced from integrated fragments of this RNA was supported by further findings that about 50% of these sample fragments had been produced by "negative-strand" (i.e., the complement in a double-stranded DNA) species that could only have come from such DNA (because the virus RNA is "positive stranded," i.e. the strand that contains the transcript in a form that can be directly translated into protein in the cell).
These results explain the scientific conundrum of finding evidence of SARS-CoV-2 RNAs in individuals who had recovered from COVID-19 and showed neither symptoms nor the presence of SARS-CoV-2 virus in clinical samples. It also suggested that continued expression of SARS-CoV-2 chimeric RNAs (and the proteins encoded thereby) might result in a persistence of immunity because the individual's immune system would continue to recognize them as antigenic and produce an immune response against the virus in the absence of later re-infection.
These results show the ability of scientific inquiry to resolve unknown and unsuspected aspects of SARS-CoV-2 biology, unsurprising for a virus only recently identified and (if current etiological explanations hold) only recently having humans as a host (see "Evolution of SARS-CoV-2 from Bat to Human Pathogen"). But it is clear that what these results do not support is a conclusion that the mRNA vaccines (from Pfizer/BioNTech or Moderna) are in any way unsafe; after all, besides being the consequence of natural processes existing through geological time the exceedingly small frequency of these events is inconsistent with realistic worries about vaccination (considering the much more severe and frequent alternative outcome for those infected with the virus). The advent of the first pandemic in over a century, particularly coming after a little more than a century of medical and scientific achievements that "conquered" infectious diseases, is disquieting if not frightening. But it is a mistake to let that fear inhibit use of the product of over a century of medical and scientific inquiry and successful intervention against infectious disease. As Dan Rather was known to say, "Courage."