key: cord-0329458-ak3jpi5x
authors: Loveday, Emma Kate; Sanchez, Humberto S.; Thomas, Mallory M.; Chang, Connie B.
title: Single cell infection with influenza A virus using drop-based microfluidics
date: 2021-09-15
journal: bioRxiv
DOI: 10.1101/2021.09.14.460333
sha: 44373a596b27b998651f65c71b0c3910ab2afe4d
doc_id: 329458
cord_uid: ak3jpi5x

Influenza A virus (IAV) is an RNA virus with high genetic diversity which necessitates the development of new vaccines targeting emerging mutations each year. As IAV exists in genetically heterogeneous populations, current studies focus on understanding population dynamics at the single cell level. These studies include novel methodology that can be used for probing populations at the single cell level, such as single cell sequencing and microfluidics. Here, we introduce a drop-based microfluidics method to study IAV infection at a single cell level by isolating infected host cells in microscale drops. Single human alveolar basal epithelial (A549), Madin-Darby Canine Kidney cells (MDCK) and MDCK + human siat7e gene (Siat7e) cells infected with the pandemic A/California/07/2009 (H1N1) strain were encapsulated within 50 μm radii drops and incubated at 37°C. We demonstrate that drops remain stable over 24 hours, that 75% of cells remain viable, and that IAV virus can propagate within the drops. Drop-based microfluidics therefore enables single cell analysis of viral populations produced from individually infected cells.

Single-cell studies of viral infections enable high-resolution examination of 33 heterogeneous virus populations. Differential selective pressures in both the host cell and virus 34 population lead to variability in virus replication and production that enables antiviral escape, 35 zoonotic spillover events, and changes in virulence or pathogenesis (Dolan, Whitfield, and 36 Andino, 2018). Influenza A virus (IAV) is a negative-strand RNA virus with populations 37 containing high genetic diversity due to its segmented genome, rapid replication rate, and low 38 fidelity RNA-dependent RNA polymerase (RdRp) (Dolan et al., 2018) . As such, IAV infection 39 results in a diverse swarm of unique variants exhibiting heterogeneous genotypes and phenotypes 40 (Brooke, 2017; Petrova and Russell, 2018) . These studies have identified large variations in total viral mRNA expressed, infectious virus 46 produced, and host transcriptional response to infection. However, isolating single cells using 47 break the emulsion followed by vortexing. Centrifugation at 500 × g for five minutes was used to 154 separate the aqueous cell suspension from the oil phase, to isolate cells for further analysis. 155 156

Drop radii were measured after imaging a drop monolayer on a hemocytometer with a CCD 158 camera (FLIR Grasshopper3) on an inverted microscope (Nikon Ti-U). The drops were added to 159 a hemocytometer at a density where they are arranged in a monolayer with minimal packing. The 160 minimal packing allows the drops to relax in a spherical shape for better image analysis. footage of the drop formation junction using a high-speed camera (VEO 710L, Phantom). Two 172 high speed videos of the flow focusing junction were recorded at 6,000 frames per second (fps) 173 to capture the formation of individual drops over the course of several frames, with an average of 174 505 drops counted from each recording. A Poisson model, with cell lines as the mixed effects, 175 the date as the random effect, and the measurement method as a two-way interaction was applied 176 to determine if the cell encapsulation distributions were a function of the cell line or 177 measurement used. An adjusted p-value was then determined using a multiple comparison across 178 the cell lines and measurement methods. Following encapsulation, drops were incubated at 37 °C 179 with 95% relative humidity and 5% CO2. Cells were released from drops as previously 180 described. The cells in the aqueous phase were collected and pelleted at 1500 × g for five 181 minutes. The cell pellet was resuspended in 200 µL PBS. Cells were diluted 1:5 in PBS + 10% 182 trypan blue stain for a final volume of 50 µL to determine cell viability. An unpaired t-test was 183 used to determine if cell viability was significantly different between the cell lines. 184

Cells were seeded onto a 6-well plate at a concentration of 1 × 10 6 cells/well and infected with 187 replaced with fresh DMEM infection media (10 mL) for bulk infections or DMEM encapsulation 210 media as described in 2.6. Infected cells were either encapsulated in 100 µm drops, or replated as 211 a bulk control. For bulk infections, a 1-mL volume of cells was added to each well of a 6-well 212 plate (2 total) and placed on the shaker. Encapsulated cells were processed as described in 2.3. 213

Bulk and drop infections were incubated at 37°C with 95% relative humidity and 5% CO2, and 214 frozen at -20 °C at 0 and 24 hpi. copies/µL at 0 hpi, which increased to 3.9 × 10 8 genome copies/µL at 24 hpi, a 1000-fold 279 increase. Between 24 and 48 hpi, the amount of RNA detected fluctuated slightly but remained 280 between 1.6 and 6.1 × 10 8 genome copies/µL. MDCK cells demonstrated a slower increase in 281 viral replication with 1.1 × 10 5 genome copies/µL at 0 hpi, increasing to 3.5 × 10 7 genome 282 copies/µL at 24 hpi, a 100-fold increase, before reaching 1.9× 10 8 genome copies/ µL at 48 hpi. 283 IAV replication in Siat7e cells was limited with no exponential increase over the course of 48 284 hpi, with 6.0 × 10 5 genome copies/µL at 0 hpi and 5.4 × 10 6 genome copies/ µLat 48 hpi. The 285 average genomes/µL for A549, MDCK, and Siat7e cells from 6-48 hpi, the time frame that 286 represents active IAV replication where the 0 h measurement represents the inoculating dose, 287 was 2.6 × 10 8 , 6.4 × 10 7 , and 2.9 × 10 6 , respectively, demonstrating cell type dependent 288 differences in IAV replication. 289 (Fig 3A) loaded onto a hemocytometer (Fig 4A and 4B) . 353 A549 cells had 10.8 ± 1.9% and 13.1 ± 3.6% of 354 drops containing one cell on the hemocytometer 355 and high-speed camera, respectively (Fig 4C) . MDCK cells had 10.2 ± 1.9% and 5.8 ± 2.0% of 356 drops containing one cell on the hemocytometer and high-speed camera, respectively (Fig 4D) . 357 Siat7e cells had 5.7 ± 1.9% and 11.0 ± 1.3% of drops containing one cell on the hemocytometer 358 and high-speed camera, respectively (Fig 4E) . 359

The calculated λ was used to compare the cell loading across the three cell lines and the 360 two imaging methods to the theoretical Poisson distribution. The calculated λ ranged from 0.021 361 to 0.14, which corresponds to a theoretical Poisson distribution for a lower starting cell 362 concentration of 3 × 10 5 cells/mL. We hypothesize that this discrepancy was most likely due to Overall, the Siat7e cells demonstrated the greatest variability and unpredictability in cell 378 loading across both methods. We hypothesize that this is due to the cells adhering together, 379 which is also visible during normal cell growth in a shaker flask, in which large clumps of cells 380 arise during growth. Due to poor cell encapsulation and low IAV replication during infection, 381

Siat7e cells were excluded from further analysis. 

High-throughput single-cell kinetics of virus infections 539 in the presence of defective interfering particles

Formation of dispersions using "flow focusing" 541 in microchannels

Population Diversity and Collective Interactions during Influenza Virus 543 Infection

Conversion of 545 MDCK cell line to suspension culture by transfecting with human siat7e gene and its 546 application for influenza virus production

Production and 549 antigenic properties of influenza virus from suspension MDCK-siat7e cells in a bench-550 scale bioreactor

The Poisson distribution and 552 beyond: methods for microfluidic droplet production and single cell encapsulation

Single-Cell Analysis of 555 RNA Virus Infection Identifies Multiple Genetically Diverse Viral Genomes within 556 Single Infectious Units

The use of single-cell RNA-Seq to understand virus-host 558 interactions

Mapping the Evolutionary Potential of RNA 560 Viruses

HSV-1 single-cell analysis reveals the 562 activation of anti-viral and developmental programs in distinct sub-populations

Rapid Prototyping 564 of Microfluidic Systems in Poly(dimethylsiloxane)

Influence of Fluorinated Surfactant Composition 566 on the Stability of Emulsion Drops

A high-throughput drop microfluidic 570 system for virus culture and analysis

High-throughput single-cell activity-577 based screening and sequencing of antibodies using droplet microfluidics

Single-Cell Virology: On-Chip Investigation of Viral 581 Infection Dynamics

SARS-CoV-2 variants, spike mutations and immune escape

Single-cell analysis and 587 stochastic modelling unveil large cell-to-cell variability in influenza A virus infection

Murine norovirus, a recently discovered and highly prevalent viral agent 590 of mice

Single-Cell and Single-Cycle Analysis of HIV-592

Biocompatible surfactants for water-in-fluorocarbon emulsions

Drop-based microfluidic 599 devices for encapsulation of single cells

Single-Cell Analysis Uncovers a Vast Diversity in Intracellular Viral 602

Defective Interfering RNA Content Affecting the Large Cell-to-Cell Heterogeneity in 603 Influenza A Virus Replication

Single-cell landscape of bronchoalveolar 606 immune cells in patients with COVID-19

Ultra-sensitive digital quantification of proteins and mRNA in single 609 cells

Screening of Additive 611 Formulations Enables Off-Chip Drop Reverse Transcription Quantitative Polymerase 612 Chain Reaction of Single Influenza A Virus Genomes

Single-Cell Analysis Using Droplet 614 Microfluidics

Single-616 cell analysis and sorting using droplet-based microfluidics

The evolution of seasonal influenza viruses

Scaling by shrinking: empowering single-620 cell 'omics' with microfluidic devices

Evolution on the biophysical fitness 623 landscape of an RNA virus

Single-cell 625 virus sequencing of influenza infections that trigger innate immunity

Extreme heterogeneity of influenza virus 628 infection in single cells

Single-cell analysis uncovers extensive biological noise in 630 poliovirus replication

Innate immune response to homologous 633 rotavirus infection in the small intestinal villous epithelium at single-cell resolution

Design and performance of the CDC 637 real-time reverse transcriptase PCR swine flu panel for detection of 2009 A (H1N1) 638 pandemic influenza virus

Single cell 640 heterogeneity in influenza A virus gene expression shapes the innate antiviral response to 641 infection

Rapid, targeted and 644 culture-free viral infectivity assay in drop-based microfluidics

Artifact-Free Quantification and Sequencing of Rare Recombinant Viruses 648 by Using Drop-Based Microfluidics

Kinetics of virus production from single cells

Quantitative profiling of innate immune activation 651 by viral infection in single cells

Genetic and pathobiologic characterization of pandemic H1N1 2009 influenza 655 viruses from a naturally infected swine herd

Microfluidic Single-Cell Omics Analysis

Single-cell transcriptional 659 dynamics of flavivirus infection

Growth of an RNA virus in single cells reveals a broad 661 fitness distribution

Single-cell barcoding and sequencing using droplet microfluidics

increased to 2.3 × 10 7 genome copies/µL at 24 h ( Fig 5A) . Bulk infections of A549 cells had a 392 comparable increase with 1.8 × 10 6 genome copies at 0 h and 5.6 × 10 8 genome copies/µL at 24 393 h ( Fig 5A) . The log RNA concentration in drops compared to bulk at 0 h and at 24 h were 394 significantly different (p-value 0.0006 and 4.3E-09, respectively) which we hypothesize is due to 395 a lower number of cells associated with the in-drop samples due to calculated cell encapsulation. 396However, the log difference of RNA produced by cells from 0 to 24 h in drops compared to cells 397 in bulk was not significant (p-value 0.057) and suggests that viral replication was not impacted 398 by cell encapsulation in drops. Recovery of infectious virus from A549 cells over the same 24 h 399 incubation period was also consistent between IAV infection in drops and bulk culture with 1.1 × 400 10 5 PFU/mL and 3.6 × 10 5 PFU/mL recovered at 24 hpi, respectively ( Fig 5B) . Surprisingly, the 401 log PFU/mL concentration in drops compared to bulk at 0 h and at 24 h was not significantly 402 different (p-value 0.31 and 0.71, respectively). In addition, the amount of infectious virus 403 produced from 0 to 24 h in both drop and bulk infections was also not significantly different (p-404 value 0.84). The genome to PFU ratio results in 1 PFU per 2.97 × 10 2 genomes for A549 cells in 405 drops and 1 PFU per 1.63 × 10 3 genomes in bulk, suggesting that any significant difference in 406 RNA concentration is not impacting the amount of virus being produced by A549 cells in drops. 407For MDCK cells, the number of genome copies in drops was 8.9 × 10 3 genome copies/µL 408 at 0 h and increased to 2.3 × 10 7 genome copies/µL at 24 h. Whereas, bulk infections of MDCK 409 cells increased from 1.2 × 10 5 genome copies/µL at 0 h and 3.4 × 10 8 genome copies/µL at 24 h 410