Design and application of a novel PNA probe for the detection at single cell level of JAK2^V617Fmutation in Myeloproliferative Neoplasms

Ph-negative MPNs include Essential Thrombocytemia, Polycythemia Vera and Primary Myelofibrosis. They share a common molecular signature represented by the mutation of JAK2[1‐4]. The detection JAK2 ^V617 is included in the diagnostic criteria and specific JAK2 inhibitors have been recently approved for the treatment of these patients [1]. MPNs are currently considered stem cell-related disorders of monoclonal origin although the presence of different co-existing subclones cannot be ruled out. Interestingly, in patients bearing JAK2 ^V617F within the CD34+ compartment a mosaicism of cells harbouring the JAK2 ^V617F can be detected alongside with the wild type counterparts, as elegantly reported by Scott and colleagues [5]. Nevertheless current approaches do not discriminate these two populations or directly quantify them in a easy and affordable format. In particular, the available methodology forecasts the use of colony formation assays followed by capillary electrophoresis sequencing [5, 6]. Overall, this method owns some pitfalls which are primarily due to: i) time consuming sample processing and ii) relatively low abundance of DNA isolated from colonies, iii) culture conditions (i.e. Epo +/−) may alter the proportion of colonies bearing the mutation. On the other hand, the method most frequently used for measuring the distribution of cell populations is indirect and based on JAK2 sequencing. The JAK2 ^V617F allele-burden is usually estimated by allele specific polymerase chain reaction (PCR). Although it is a sensitive assay, this is performed on the whole ‘white’ myeloid differentiated cell population (e.g. granulocytes) and for this reason it may be biased by ‘dilution effects’ on sample. For all this reasons, the chance of distinguishing the JAK2V617F at the single-cell level still represents a challenge, both in the diagnostic and research field. The PNA is a synthetic nucleic acid analogue in which the negatively charged sugar phosphate backbone is replaced by a neutral pseudo-peptide backbone [7]. Due to its (i) high degree of sequence selectivity, (ii) discrimination ability in binding to complementary DNA or RNA, and (iii) increased stability (relative to non-synthetic nucleic acids) and (iiii) low cost, PNA probes do have tremendous potential for therapeutic application as well as diagnostic and research use, especially when a highly specific matching is needed. The physical properties of PNA endow them with specific advantages over standard oligonucleotides probes: they are less polar than (natural) nucleic acids and -as a consequence- PNA/DNA heteroduplexes are thermodynamically favoured when compared to the DNA/DNA double helix [8]. When very short PNA are used, this greater specificity allows the PNA/DNA heteroduplex to become thermodynamically unstable even when a single base-pair mismatch occurs [9]. Taking advantage of these unique PNA features we set-up a fluorescently-labelled PNA probe, coupled to FISH technology, to identify the presence of JAK2 ^V617F at the single-cell level. This method allows to distinguish between CD34+ progenitor stem cells harbouring the mutated or wild type form of JAK2 (Figure 1).

In this article we report a method characterized by a high degree of specificity and sensitivity which allows to identify at a single cell level the presence of JAK2^V617F mutation.

CD34+ cells from JAK2 ^V617F positive patients (affected by ET, PV and PMF) displayed an heterogeneous staining pattern when probed with the JAK2 ^V617F /PNA. Indeed, in a single patient some CD34+ cells are clearly positive for JAK2 ^V617F /PNA-fluorescent staining, while others are negative (Figure 2). The analysis revealed that among JAK2 ^V617F PV patients the distribution pattern is fairly similar to that reported by Scott et al.[5] with a rather wide variability occurring among patients. We found a percentage of mutated CD34+ cells ranging from 40% to 100% in PV patients, from 15% to 80% in ET and from 25% to 100% in PMF. These findings are in agreement with previous data reporting that a variable proportion of progenitors from patients affected by JAK2 ^V617F positive PV are capable of generating JAK2 ^V617F negative colonies [5]. In addition these data indicate that fluorescinated JAK2 ^V617F /PNA probe displays a very high specificity towards a single base-pair mismatch.

Interestingly, when evaluating the presence of JAK2 ^V617F positive cells collected from JAK2 wild type subjects defined by sequencing and by Q-PCR we identified a small percentage of cells positive for the JAK2 ^V617F /PNA staining not exceeding 3% of the CD34+ cell population indicating a high level of sensitivity of the procedure. Interestingly, this apply only to patients with PV but not with PMF and ET. Importantly, the lack of positivity detected in CD34+ cells from 20 healthy subjects demonstrates a high specificity of this method. We conclude that the JAK2 ^V617F /PNA-FISH method displays high specificity and reliability in discriminating cell subpopulations harbouring the JAK2 ^V617F mutation. In addition, it allows to analyze the CD34+ population at the single cell level, avoiding the time consuming analysis of hematopoietic colonies. The fact that our results are in keeping with the data reported by Scott et al [5] corroborate the robustness of the technique although we think that the proposed approach is much easier and free from variability related to colony growth conditions [5, 6]. This approach allows to monitor longitudinally the evolution of a defined cell population over time in MPNs.