In this study we demonstrated that quantity and quality of DNA prepared from dried blood stored on FS are adequate to quantify circularized fragments of DNA, such as TRECs and KRECs, even if they are contained in peripheral blood in number as low as few copies per mL. Therefore, the proposed strategy for storing dried blood on FS and TREC and KREC quantification by dPCR can be successfully applied to adults and elderly subjects for a proper evaluation of new T- and B-lymphocyte generation.
It is well known that blood samples adsorbed on filter paper or nylon FS, and left to dry, represent a convenient alternative option to fresh blood specimens when collection and shipment of biological materials are difficult, when resources are limited, or if used for applications in large surveys. Indeed, dried blood carries a reduced biohazard risk as compared to liquid samples; it simplifies shipments to reference laboratories for testing, and importantly, it requires minimal storage facilities because nucleic acids are stable at room temperature. For these reasons, dried blood spotted as a drop onto filter paper is largely used in newborn screening for genetic diseases [
23]. The most recent improvement within this specific field of research is the newborn screening for severe combined immunodeficiency (SCID) through the detection of TRECs. In fact, the use of TREC test was introduced for early diagnosis of infants with SCID, following the evaluation of incidence and costs of early versus late treatment [
24,
25]. Several protocols for easy detection of TRECs in newborn blood spotted on filter paper have been developed and validated [
24,
26‐
28]. The absolute number of TRECs varies greatly, with several fold-differences, ranging from 119 to 1900 TRECs/μL, based on different laboratories [
25], depending on the parameters of the assays, such as DNA isolation method, qRT-PCR primers and probes, standard curve used, singleplex or multiplex PCR. The volume of total dried blood could be an important limiting factor leading to assay failure when, for instance, the biological substrates are contained at very low quantity and if a large representation of target molecules is mandatory. Indeed, while normal newborns have a high rate of new T-cell production, resulting in TREC numbers of about 10% of their total T-cell numbers, adults have progressively lower ratio of TRECs to T cells, reflecting peripheral T-cell expansion [
1]. Therefore, due to the lower representation of TRECs within their peripheral blood, several punches from filter paper would be required in order to obtain a sufficient DNA amount for TREC quantification. To our knowledge, data regarding the values of TRECs in the blood of adults spotted on filter paper are not available. In addition, those obtained from liquid and dried blood samples or PBMC [
1,
3,
4,
29] are of difficult comparison due the methods of TREC calculation or because TREC measurement has been presented in a number of different ways. Indeed, TRECs are reported as normalized amount (calculated as a difference between Ct values of reference gene and Ct of TRECs), as absolute number per unit number of cells, as TREC molecules per μg of DNA within PBMC or T-lymphocyte subsets, or as TRECs per mL of blood. The values that we have generated by directly quantifying TRECs from FS-derived dried blood amplified by dPCR alternative method are given per mL, are measured removing any confounding attributable to cell proliferation, and using very small quantities of starting whole blood (about 150 μL). The possibility to store dried blood on FS and to send it by regular mail (Guidance on Regulations for the Transport of Infectious Substances 2015–2016—WHO/HSE/GCR/2015.2) lends this approach to be easily used in clinical settings where limited tissue is available, for example in adult and ageing studies, in immunodeficient patients and/or where longitudinal information is required, and when the analysis cannot be performed locally.
We found that the extent of age-related TREC decrease determined by dPCR in DNA obtained from dried blood adsorbed on FS was comparable to that reported by Lorenzi et al. [
3], stating a 1.5-Log change of TRECs over the age range they have measured in liquid WB. Furthermore, newly produced data can be compared with retrospective ones because the mean values of dPCR-derived TRECs and KRECs perfectly matched those found in age-matched controls analyzed with the conventional “reference” method used in our diagnostic laboratory. This was quite unexpected, due to the following reasons:—DNA was extracted from differentially manipulated blood specimens (WB, PBMC, and dried blood) and using diverse methods (different buffers and reagents may potentially exert an effect on the end-result);—the amount of target DNA used for the two amplification protocols was different;—the methods of TREC and KREC evaluation were different (with qRT-PCR the values are extrapolated from a standard curve, while with dPCR the results are expressed as absolute values);—the mathematical formula used to calculate the number of TRECs and KRECs/mL of blood are diverse. In addition, it must be taken into account that the data were all obtained in different experimental procedures, in order to reproduce the “real world” conditions necessary for a test to be used for routine analyse. This introduces further variations in the results. Therefore, although the experiment performed with dilutions of plasmid DNA using dPCR allowed the detection of one copy of TRECs and one of KRECs, while 2.5 copies of the targets were detected by qRT-PCR, due to above reported differences, one cannot be confident enough to possibly state that dPCR is more sensitive than qRT-PCR for the quantification of TRECs and KRECs in peripheral blood. However, with qRT-PCR approach, the lowest levels of TRECs and KRECs are evidenced at Ct greater than 35. At values >35, the qRT-PCR-generated data have poor precision, are not always reproducible, and the method cannot be considered quantitative. Furthermore, we previously reported [
7] that the detection limit for qRT-PCR method for TRECs was approximately 10 TRECs or KRECs copies and Lorenzi et al. [
3] demonstrated that as the number of TRECs quantified per sample fell below 10, variations in the final result increased. Finally, an advantage of dPCR is that, when working at the limit of assay sensitivity, the probability of TREC and KREC detection could be easily increased by quantifying these targets from replicate aliquots of individual samples and adjusting the derived values accordingly.
To our knowledge, no data are available on KREC evaluation from adult dried blood samples, thus this is the first report in which both TRECs and KRECs have been simultaneously quantified starting from blood of adults adsorbed on FS.