Enrichment of circulating trophoblasts from maternal blood using laminar microscale vortices

Abstract Objective Enrichment of circulating trophoblasts (CTs) from maternal blood at week 11–13 of gestation, using laminar microscale vortices, and evaluation of the performance of the VTX‐1 Liquid Biopsy System in terms of CT recovery and purity. Method Eight mililiter of blood was collected from 15 pregnant women and processed with the VTX‐1 Liquid Biopsy System. Y‐chromosome specific quantitative PCR was performed to estimate the number of enriched male CTs. To evaluate the VTX‐1 performance, the target cell recovery was characterized by spiking experiments with a trophoblast cell line. Furthermore, the total quantity of DNA after enrichment was used to calculate the number of retained maternal cells. Results Successful recovery of male CTs was established in 7 out of 10 first trimester samples from pregnant women carrying a male fetus. The number of CTs, recovered from 8 ml of blood, was estimated between two and six. Spiking experiments resulted in a CT recovery of ±35 % with ±1524 retained maternal blood cells. Conclusion CTs can be enriched from maternal blood with high purity, using laminar microscale vortices, starting from 8 ml of blood.


What does this study add?
� Laminar microscale vortices allow size-based enrichment of circulating trophoblasts, starting from only 8 ml of maternal blood.
� CT recovery and purity after enrichment using the VTX-1 Liquid Biopsy System are reported.

| INTRODUCTION
Efficient prenatal genomic-based testing is essential for pregnant women to make informed reproductive decisions. 1 Since 2011, cellfree noninvasive prenatal testing (cfNIPT) has been clinically implemented as a first trimester screening test for Trisomy 21, 18, and 13. [2][3][4][5][6][7] The increased use of cfNIPT has led to a large reduction of invasive diagnostic procedures such as amniocentesis and chorionic villus sampling. 8 cfNIPT is based on massively parallel sequencing of a small fraction of cell-free fetal DNA fragments in the presence of a large fraction of cell-free maternal DNA. 7,9,10 Therefore, large genetic aberrations, such as aneuploidies, can be detected but smaller genetic aberrations, such as single-gene disorders caused by single nucleotide polymorphisms (SNPs) are far more challenging. 11,12 As an alternative, several research groups have engaged in the quest for the isolation of fetal cells from the blood stream of pregnant women to allow cell-based non-invasive prenatal testing (cbNIPT) during the first trimester of pregnancy. [13][14][15] As cbNIPT allows genetic analyses of the fetal genome in absence of contaminating maternal DNA, cbNIPT could be a valuable alternative for the current cfNIPT assay.
Different circulating fetal cell types such as nucleated red blood cells (nRBCs), trophoblasts, and lymphocytes have been reported in the past. 16,17 Unfortunately, all types of fetal cells are fragile and extremely rare with a frequency of approximately one fetal cell in 10 9 maternal blood cells. 16 Current cbNIPT methods are mainly focused on the isolation of circulating trophoblasts (CTs) because of the presence of relatively specific trophoblast markers 18 and their potentially larger cell size (>15 µm) compared to other maternal blood cells. 15,19 Numerous approaches have been investigated regarding the isolation of CTs from maternal blood, but only a few research groups have made significant progress in the quest for a feasible cbNIPT workflow. [13][14][15] The Danish research group Arcedi Biotech (Vejle, Denmark) discovered a unique combination of trophoblast markers through the characterization of CTs by means of microarray analysis. 18 Their results indicate that CTs express both the ectodermal marker Cytokeratin and the mesenchymal marker CD105. Based on these findings, in 2016, Arcedi Biotech succeeded in the enrichment of CTs with positive magnetic-activated-cell-sorting (MACS), using the surface marker CD105. 20 After enrichment, Cytokeratin positive CTs were selected and isolated by means of immuno-fluorescent staining. Finally, a recovery of on average 0.42 CTs/ml starting from 30 ml of maternal blood was reported. 20 Marker-based enrichment of CTs from maternal blood was also reported recently by Vossaert et al. 21   In contrast to the established cbNIPT methods, VTX-1 enrichment allows the processing of whole blood without pre-enrichments procedures such as formaldehyde fixation, red blood cell lysis, or density gradient centrifugation. Therefore, no sample manipulations are required before enrichment and consequent target cell loss is prevented. Figure 1C illustrates how the Vortex technology could be integrated in a beginning-to-end cbNIPT workflow. After VTX-1 enrichment, immunostaining and imaging can be performed to select putative CTs. Finally, single CTs can be isolated, the fetal genome can be amplified, and genetic analysis can be performed.
In this study, the enrichment of CTs from 8 ml of maternal blood by means of laminar microscale vortices is assessed at 11-13 weeks of gestation. The retrieval of male trophoblasts is determined using Y-chromosome specific quantitative PCR (qPCR). Furthermore, to evaluate the performance of the VTX-1 Liquid Biopsy System, both the recovery of VTX-1 enrichment and the purity of the enriched CTs

| Sample collection
Maternal blood samples were collected at Ghent University Hospital at 11-13 weeks of gestation. Each blood sample contained 12 ml maternal blood, collected in K 3 EDTA blood collection tubes (Thermo Fisher Scientific). All pregnant women were adult and signed informed consent. Ethical approval was obtained from the ethical review board of Ghent University Hospital, B670201524235. All methods were performed in accordance with the Declaration of Helsinki.

| Fetal sex determination from plasma
To determine the fetal sex based on circulating cell-free DNA in maternal plasma, 4 ml of maternal blood was processed within 4 h after blood collection. Following density gradient centrifugation for 15 min at 1200 x g, 400 µl of plasma was collected and cell-free DNA was extracted with the QIAamp DNA Blood Mini Kit (Qiagen, Hilden) following manufacturer's recommendations. DNA was eluted in 100 µl nuclease-free water at 65°C. Next, as described by Picchiassi et al. 28 quantitative PCR (qPCR) was executed to detect the Y-chromosome specific multicopy DYS14 sequence located within the TSPY gene.

| Vortex enrichment
Maternal blood samples from ten women carrying a male fetus and five women carrying a female fetus were used for Vortex enrichment.

| Detection of male cells after vortex enrichment
After the enrichment of target cells, DNA was extracted with the QIAamp DNA Blood Mini Kit. The extracted DNA was eluted in 50 µl sterile nuclease-free water at 65°C, followed by evaporation to a final volume of 30 µl. Next, qPCR was performed on this extracted DNA to detect the Y-chromosome specific multicopy DYS14 sequence located within the TSPY gene, as previously described.

| qPCR standard curve
To estimate the number of recovered male CTs after VTX-1 enrichment, a dilution series of male reference DNA was analyzed, using Fisher Scientific) and a mix of penicillin and streptomycin (Gibco, T A B L E 1 Overview of gestational age of the pregnant women at time of sample collection, confirmed fetal sex, C q value after VTX-1 enrichment, qPCR outcome, and estimated number of CTs, for all collected first trimester maternal blood samples -1175 Thermo Fisher Scientific) at a final concentration of 100 units/mL and 100 µg/ml, respectively. The cells were cultured in an humidified atmosphere containing 5 % CO 2 at 37°C.

| Performance of VTX-1 Liquid Biopsy System
All JEG-3 cells used for recovery calculation, were pre-stained with Hoechst 33342 Ready Flow TM Reagent (Invitrogen) for 25 min.  Table 1. In Figure 2, the fluorescent signal of each sample is plotted against the cycle number.  Table 1. This estimation demonstrates a recovery between 2 and 6 CTs from 8 ml of maternal blood in seven out of ten first trimester samples.

| Performance of VTX-1 Liquid Biopsy System
The target cell recovery of VTX-1 enrichment is estimated by means of spiking experiments with trophoblast cells from a JEG-3 cell line.
The results indicate that after VTX-1 processing of whole blood  Second, only 8 ml of maternal blood is processed, while other reported state-of-the-art CT isolation protocols start from 30 to 40 ml. 20,21 Increasing the volume of blood will probably decrease the number of samples for which no sufficient numbers of CTs are enriched.
Finally, the number of retained maternal cells after VTX-1 enrichment is estimated at �1524 cells. This low number of contaminating maternal blood cells will definitely simplify the downstream scanning and selection procedure for final single trophoblast isolation in cbNIPT.
In a beginning-to-end cbNIPT workflow, VTX-1 enrichment can be followed by fixation and intra-or extracellular staining in a collection recipient of choice. If needed, the stained cells can then be transferred to a microscopic glass slide and visualized using fluorescence microscopy. Finally, the target cells can be isolated individually, followed by whole genome amplification (WGA) and genetic analysis. Preliminary results of our research group indicate that VTX-1 enriched cells maintain good quality and allow downstream staining with CD45/ Cytokeratin/Hoechst. Moreover, DNA profiles can be obtained after single cell isolation using laser capture microdissection and WGA.

| CONCLUSION
This proof-of-concept study shows that CTs can be enriched from maternal blood based on their larger cell size, using laminar microscale vortices. Y-chromosome-specific qPCR results indicate that CTs are recovered from 8 ml of maternal blood, in seven out of ten first trimester samples, using VTX-1 enrichment. For these samples, the number of enriched CTs is estimated at 2-6 cells. For the other three samples, the presence of CTs could not be confirmed by qPCR. Only �1524 maternal blood cells are retained after VTX-1 enrichment, which simplifies subsequent CT isolation and cbNIPT analyses.