Rolling-circle amplification (RCA) is an isothermal amplification method where circular – endless – DNA strands are being replicated. The reaction generates strands that can contain hundreds or more copies of a sequence complementary to the DNA circle that templates replication. The single-stranded amplification products can be visualized by hybridizing oligonucleotides labeled with fluorophores or enzymes that generate colored products.  

RCA reactions can be used to enhance and record detection reactions using molecular tools that generate circular reaction products, such as in situ proximity ligation assays and padlock probes. RCA has the effect to locally magnify detection events into highly visible signals.

Rolling-circle amplification (RCA) is an isothermal amplification method where circular – endless – DNA strands are being replicated. The reaction generates strands that can contain hundreds or more copies of a sequence complementary to the DNA circle that templates replication. The single-stranded amplification products can be visualized by hybridizing oligonucleotides labeled with fluorophores or enzymes that generate colored products.  

RCA reactions can be used to enhance and record detection reactions using molecular tools that generate circular reaction products, such as in situ proximity ligation assays and padlock probes. RCA has the effect to locally magnify detection events into highly visible signals.

A padlock probe is a DNA oligonucleotide with target-complementary segments at both the 5′ and 3′ ends. The sequences at the ends of a padlock probe are designed to hybridize in juxtaposition to a target DNA strand, leaving a nick site in the double-stranded structure. This missing internucleotide link is enzymatically sealed by a DNA ligase. Thereby, the padlock probe is converted to a DNA circle, which is wound around and locked on the target strand. The DNA ligase activity is sensitive to mismatched base pairs at the nick site, allowing single nucleotide variants of the target sequence to be distinguished by the padlock probes.

The central part of the padlock probe is not complementary to the target but can include sequences useful for hybridization of detection probes or amplification primers. Padlock probes have been used in many applications, for example for analysis of single nucleotide polymorphisms (SNPs), copy number variants (CNVs), and for profiling gene expression and detection of pathogens.

In superRCA a first RCA reaction replicates a circular DNA strand, generating a long strand composed of complements of the circular DNA strand. The resulting RCA product is next interrogated with padlock probes that recognize segments in the repeated sequence of this first RCA product. Pairs of padlock probes may be used that are specific either for normal or mutant variants of a DNA sequence captured in the initial circular strand. Once ligated and forming circles wound around the first RCA product, the reacted padlock probes may in turn be subjected to a second RCA. The result of the two consecutive RCA reactions is that for each starting DNA circle, a large bundle of DNA is created with many thousand complements for detection probes. The superRCA products are easily visible by microscopy, and large numbers of them can be rapidly distinguished and counted in a standard flow cytometer. 

Techniques for detecting rare mutant DNA variants

Tumor tissue typically contains somatic mutations that are not present in normal tissues. Such mutations can be investigated in blood or bone marrow cells from leukemic patients to follow the course of disease or its possible recurrence. Besides the genomic DNA (gDNA) in nucleated cells in blood, blood plasma contains small amounts of cell-free DNA (cfDNA). This pool of short, double-stranded cfDNA provides minimally invasive access to genotypes of cells throughout the body. In patients with tumors, some fraction of this cfDNA in blood plasma may derive from the malignant cells.  

Detection of low frequencies of mutant DNA is a demanding challenge, particularly for the common class of mutations that only differ from normal variants in single nucleotide positions. Before superRCA there have been two principally different approaches to achieve the selectivity required to detect and enumerate low proportions of mutant DNA, namely DNA sequencing and PCR with Tagman probes. 

DNA sequencing as a means to detect and quantify low frequencies of mutant DNA 

Several approaches for DNA sequencing are all associated with some proportion of error in assigning the nucleotide identity in each position of a DNA strand. There is also a risk that artificial mutations will have been introduced in the course of amplifying the target sequence before sequencing reactions. The risk of recording artificially introduced mutations can be minimized by attaching DNA segments bearing unique barcodes to all DNA fragments in a sample. That way it is possible to tell which sequenced molecules represent different amplified copies from the same starting molecules, to identify amplification artefacts. 

The related risk of misidentifying nucleotides in the course of sequencing is mitigated by ensuring that amplification products of individual starting target strands are sequenced enough times. If the same deviation from a normal sequence is observed repeatedly, then it is unlikely to be accounted for by errors in the sequencing reactions but represent real sequence variants in a sample.  

Sufficiently deep sequencing, that is analyzing the same target sequence enough times to exclude artifactual sequence changes, can allow detection of mutations at very low frequencies. The method is costly, however, and associated with a long turn-around time. By contrast, a number of relatively rapid assays involve PCR in the search for rare mutant DNA sequences.

PCR-based detection of low-frequency mutations

Taqman probes are a fundamental element in several PCR-based mutation detection technologies. In digital PCR (dPCR), emulsions or fabricated microcompartments are used together with quantitative PCR (qPCR), where Taqman probes serve to distinguish alleles during PCR amplification. The hydrolysis feature in Taqman probing relies on the 5’ to 3’ exonuclease activity of the PCR polymerase, thus limiting the selection of polymerase to ones that may be suboptimal for PCR replication fidelity.   

Another technology for allele distinction – BEAMing assays – employs a bead-based amplification which allows an initial use of a high-fidelity DNA polymerase to generate enough target molecules in beads before a Taqman probe-based allele distinction after the PCR amplification.  

In superRCA assays for detecting even very low proportions of mutations, padlock probes are used to distinguish the sequence variants. Padlock probes accurately distinguish between targets that are mismatched versus matched against the probe. This target selectivity is further greatly enhanced by a so-called majority vote mechanism. The basis of this mechanism is that padlock probes are not used to genotype single target molecules, but RCA products that each contain hundreds of copies of the sequence of interest, that is hundreds of copies of either the mutant or normal DNA sequence. If a minority of padlock probes should misidentify the target sequence, such incorrect reactions escape detection against a majority of correct reactions for a given RCA product, ensuring extremely accurate genotyping.

The reacted genotyping padlock probes are in turn replicated by a second RCA, before the final large reaction products, generated via two consecutive RCA reactions, are identified via fluorescent probes hybridizing to barcodes specific for mutant or normal amplification products. In the end, a million or so reaction products are evaluated over a few minutes using generally available flow cytometers that record the identity of individual reaction products, in order to identify and count even very rare mutant reaction products.

Blood typically yields between 15-50 µg gDNA/ml from the approximately 5 million nucleated blood cells per ml in healthy individuals. From a 10 ml venous draw around 350 µg gDNA can be extracted, equivalent to around 50 million diploid cells.

As mentioned above, blood plasma also includes low levels of cell-free DNA (cfDNA) that may derive from any cells in the body. The cfDNA may for instance include low proportions of fetal DNA in blood from pregnant women, or the plasma DNA may include tumor-specific mutations in blood plasma from tumor patients. The concentration typically varies between 0-100 ng cfDNA/ml with an average around 30 ng/ml. cfDNA is unstable with a reported half-life between 16 min to 2.5 h in circulation.

From a 10 ml venous draw around 150 ng cfDNA may be isolated, typically corresponding to a little more than 20.000 diploid cells. To achieve a reliable sensitivity of 1 mutant copy among100,000 normal sequence variants, one would at a minimum need 200,000 copies and hence 100.000 diploid cells, or 660 ng cfDNA and hence around 40 ml whole blood would be needed.

Biopsies from mammalian solid tissues typically yield 0.2-0.4 µg genomic DNA (gDNA)/mg tissue. The amount depends on how many nucleated cells the tissue contains per mg. Fatty tissues such as brain, bone marrow or tissues with high levels of extracellular matrix such as connective tissues typically fall in the lower range.

References:

1. Elazezy M, Joosse SA. Techniques of using circulating tumor DNA as a liquid biopsy component in cancer management. Comput Struct Biotechnol J. 2018 Oct 9;16:370-378. doi: 10.1016/j.csbj.2018.10.002. PMID: 30364656; PMCID: PMC6197739.
2. Anatoli Kustanovich, Ruth Schwartz, Tamar Peretz & Albert Grinshpun (2019) Life and death of circulating cell-free DNA, Cancer Biology & Therapy, 20:8, 1057-1067, doi: 10.1080/15384047.2019.1598759
3. Alborelli, I., Generali, D., Jermann, P. et al. Cell-free DNA analysis in healthy individuals by next-generation sequencing: a proof of concept and technical validation study. Cell Death Dis 10, 534 (2019). https://doi.org/10.1038/s41419-019-1770-3