FIND-seq: high-throughput nucleic acid cytometry for rare single-cell transcriptomics

Abstract

Rare cells have an important role in development and disease, and methods for isolating and studying cell subsets are therefore an essential part of biology research. Such methods traditionally rely on labeled antibodies targeted to cell surface proteins, but large public databases and sophisticated computational approaches increasingly define cell subsets on the basis of genomic, epigenomic and transcriptomic sequencing data. Methods for isolating cells on the basis of nucleic acid sequences powerfully complement these approaches by providing experimental access to cell subsets discovered in cell atlases, as well as those that cannot be otherwise isolated, including cells infected with pathogens, with specific DNA mutations or with unique transcriptional or splicing signatures. We recently developed a nucleic acid cytometry platform called 'focused interrogation of cells by nucleic acid detection and sequencing' (FIND-seq), capable of isolating rare cells on the basis of RNA or DNA markers, followed by bulk or single-cell transcriptomic analysis. This platform has previously been used to characterize the splicing-dependent activation of the transcription factor XBP1 in astrocytes and HIV persistence in memory CD4 T cells from people on long-term antiretroviral therapy. Here, we outline the molecular and microfluidic steps involved in performing FIND-seq, including protocol updates that allow detection and whole transcriptome sequencing of rare HIV-infected cells that harbor genetically intact virus genomes. FIND-seq requires knowledge of microfluidics, optics and molecular biology. We expect that FIND-seq, and this comprehensive protocol, will enable mechanistic studies of rare HIV⁺ cells, as well as other cell subsets that were previously difficult to recover and sequence.

Fig. 1 |. Major stages in the FIND-seq workflow.

a, The reaction between the acrydite modified oligo dT primer and allyl agarose is catalyzed by APS and TEMED and covalently links the agarose to primers. b, Heterogeneous cells are encapsulated in single droplets with oligo dT-conjugated agarose and lysis buffer by using a bubble-triggered device. c, Lysis buffer releases the genome and mRNA of the cell into the agarose bead. Beads are removed from oil and washed, and mRNA is reverse transcribed onto the agarose chain. d, Genome-capture beads are re-injected into droplets with TaqMan PCR master mix by using a re-injector device. e, The resulting emulsion is thermocycled, amplifying the TaqMan signal within positive droplets. f, The droplets are sorted on the basis of their TaqMan fluorescence. Depending on the collection method, FIND-seq can be used for either bulk or single-cell transcriptome analysis.

Fig. 2 |. Operation of three sequential microfluidic devices for cell lysis, nucleic acid detection and sorting.

a, The bubble-triggered device has five inlets: (i) cell inlet, (ii) agarose inlet, (iii) lysis buffer inlet, (iv) air inlet and (v) oil inlet. The bubble-triggered device co-flows cells, agarose and lysis buffer with oil. At high flow rates, this produces a jet that is broken into droplets by using air bubbles generated on a chip. b, The cell concentration is adjusted to an average loading of 1 cell per 10 drops. The streams of cells, agarose, lysis buffer and oil flow fast enough to create a jet, and air bubbles break the jet into monodisperse droplets. c, After droplet generation, encapsulated cells can be observed in the droplets. Staining with SYBR Green I reveals the genome released from the cells and captured in the agarose beads after droplet breaking. d, The re-injector device has three inlets: (i) agarose bead inlet, (ii) TaqMan PCR master mix inlet and (iii) oil inlet. e, The genome-capture beads and TaqMan PCR master mix co-flow to generate a re-injected single bead in a droplet. The anticipated re-injection efficiency is 50–70%, and checking this efficiency by using microscopy during the procedure is recommended. f, The re-injected droplets are thermocycled for target detection. g, The droplet sorter device has seven inlets: (i) emulsion inlet, (ii) oil 1 inlet, (iii) oil 2 inlet, (iv) oil 3 inlet, (v) saltwater inlet (for the electrode), (vi) saltwater inlet (for the moat) and (vii) pressurized air inlet. h, The thermocycled droplets are injected through the emulsion inlet (initial droplets). The droplets then proceed to the detection spot. If a droplet displays a positive signal, voltage is applied to the electrode to dialectrically sort the droplet to the positive outlet channel.

Fig. 3 |. Multiplexed detection of cells containing HIV provirus.

a, Location of the two target regions, Ψ and env, used in the intact proviral DNA assay (IPDA). Multiplexed Ψ and env droplet detection is used to differentiate intact from defective proviruses. b, Lysis within molten agarose releases the genome and transcriptome from HIV⁺ J.Lat 6.3 cells. The genome is captured as the agarose bead gels and stained with SYBR Green I. The proportion of stained beads is measured with flow cytometry (left) or microscopy (right) to check lysis and determine the cell loading percentage. c, The quality of cDNA on the genome-capture bead is checked with WTA, and size is determined by using a Bioanalyzer. d, Photograph of the emulsion generated from the re-injector device. This emulsion is collected in a PCR tube and distributed in small volumes to ensure that all the emulsion is covered by the heat block of the PCR machine. Inset: microscope image of agarose beads. Beads have been colored after image acquisition so that they are easily visualized in drops. e, Fluorescence microscope images of drops after thermocycling. Strong signal-to-noise of FAM (green; Ψ positive) and VIC (red; Env positive) intensities indicate successful TaqMan HIV DNA detection, and the merged image (yellow; double positive) indicates the presence of an intact provirus. f, Droplet cytometry plot after TaqMan PCR. The IPDA assay was tested by using J.Lat 6.3 (HIV⁺) and Jurkat (HIV⁻) cells separately, and the fluorescence intensity of droplets was plotted (n = 10,000). g, A schematic of cell and bead loading during the FIND-seq workflow illustrates that the efficiency of cell loading and bead encapsulation changes the frequency of double position drops observed in droplet cytometry. h, Detection efficiency as a function of target cell rarity after accounting for cell and bead loading (n = 1). FU, fluorescence units; UMI, unique molecular identifier.

Fig. 4 |. Accurate sorting of rare Ψ⁺ env⁺ cells.

a, Top: schematic of cell-mixing experiments (1% J.Lat 6.3 cells spiked into Jurkat cells) used to assess sorting accuracy and cell recovery. Bottom: fluorescence microscopy of double-positive droplets before and after sorting. b, Top: gag qPCR on genomic DNA and the resulting standard curve used from quantification of HIV⁺ cells (n = 3). Bottom: qPCR on sorted droplets from three rarity levels (0.1%, 1% and 10% J.Lat 6.3 into Jurkat). For qPCR curves, the color lines represent 100 sorted genomes, and the gray lines represent a 100-genome standard (std). Bar charts show the absolute quantification of HIV genomes by using gag qPCR sorted on 100 Ψ⁺env⁺ cells at different rarity levels. ‘Starting’ represents the theoretical percentage of drops that contain an HIV⁺ cell before sorting. ‘Sorted’ represents the measured percentage of 100 sorted drops with an HIV⁺ cell. All values represent mean ± s.d.