Optimized workflow for high-throughput whole genome surveillance of Influenza A virus

doi:10.21203/rs.3.rs-5216720/v1

Download PDF

Method Article

Optimized workflow for high-throughput whole genome surveillance of Influenza A virus

https://doi.org/10.21203/rs.3.rs-5216720/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Whole genome sequencing (WGS) is crucial for studying influenza A virus (IAV) genomic diversity in various host species to mitigate its impact on human and animal health. While the multisegment RT-PCR (mRT-PCR) efficiently amplifies all genomic segments in a single reaction, its sensitivity for larger segments is suboptimal. To improve WGS sensitivity, we optimized the mRT-PCR protocol by adjusting RT and PCR cycling conditions, achieving a 1000-fold increase in sensitivity. Additionally, we developed a dual-barcoding approach for the Oxford Nanopore platform, enabling the multiplexing of multiple IAV-positive samples without compromising sensitivity, thereby creating a scalable, high-throughput workflow for IAV surveillance.

Influenza A virus

whole genome sequencing

Oxford Nanopore sequencing

swine IAV

avian IAV

human IAV

high-throughput

Influenza A viruses (IAVs), classified under the Alphainfluenzavirus genus and Orthomyxoviridae family, are negative-sense, single-stranded RNA viruses with a segmented genome of approximately 13.6 kb in size, encoding up to 17 proteins ¹. These viruses share antigenically related nucleocapsid and matrix proteins, but are classified based on their two surface glycoproteins, haemagglutinin (HA), with 19 recognized subtypes, and neuraminidase (NA), with 11 recognized subtypes ^2,3. IAVs can be shed from their main reservoir, aquatic wild birds, to a wide spectrum of avian and mammalian species, including pigs and humans. This transmission can lead to severe consequences for both animal and human health, as evidenced by the ongoing H5N1 panzootic, the increasing number of sporadic zoonotic spillovers, and the multiple human influenza pandemics that have occurred throughout history ^4–9.

It is well established that the error-prone viral polymerase (point mutations) or simultaneous infection with two or more viral subtypes in the same host (reassortment) drives the genetic diversity and evolution of IAVs ¹⁰. This genetic diversity can influence viral fitness, antiviral resistance, and interspecies transmissibility and is often associated with specific changes in one or more IAV genes or with the overall “constellation“ of the genome segments. Therefore, whole genome sequencing plays a pivotal role in monitoring IAV evolution in both humans and animals, enabling the detection of new variants and/or transmission patterns and assessing the efficacy of current vaccines and antiviral treatments. In addition, assessments of virulence, tropisms, and zoonotic propensity can be conducted using whole genome sequences. The one-step multisegment RT-PCR (mRT-PCR) approach by Zhou et al., or derivatives thereof, is often used for whole genome surveillance at the human-animal interface ^11–17. However, the recovery of sequence information for the genomic segments encoding the largest IAV genes, i.e. encoding the polymerase (PB1, PB2, and PA) from clinical material with a low viral load can be challenging, especially on third-generation sequencing platforms that generally have a lower throughput than second-generation sequencing technologies, but crucially provide portable real-time long-read sequencing ¹³.

To increase the sensitivity of recovering whole genome sequences from avian, swine, and human IAV-positive clinical samples, we further optimized the existing approach from Rambo-Martin et al. by using a different reverse transcription (RT) enzyme and adapting the RT and PCR cycling conditions. This resulted in an approximately 1000-fold increase in sensitivity compared to the mRT-PCR protocol initially developed and described by Zhou et al. ¹¹. Following these results, we also developed and tested a novel dual-barcoding approach to increase the sequencing throughput on portable third-generation sequencing platforms. This approach allowed multiplexing of at least eight samples of avian, swine, or human origin per sequencing library barcode without a significant loss in sensitivity, creating an optimized workflow for portable high-throughput whole genome surveillance of Influenza A viruses at the human-animal interface.

Viral RNA

To optimize and test the sensitivity of the IAV whole genome amplification protocol developed by Zhou et al., we extracted viral RNA from the supernatant of Madin-Darby canine kidney II (MDCKII) cells infected with IAV (A(H1N1)pdm09) using the Quick-RNA Viral Kit (Zymo Research, Irvine, CA, USA) according to the manufacturer's instructions ¹¹. For human IAV-positive clinical samples, viral RNA was extracted from previously screened samples sent in for viral diagnostics by the treating physicians, per national regulations (Cantonal Ethics Commission Bern, Req-2020-00167). This process utilized 200 µL of virus transport medium (VTM) (provide company´s details here) from IAV-positive samples and was performed using the Kingfisher Apex automated extraction instrument (ThermoFisher Scientific, Waltham, MA, USA) in combination with the NucleoMag®VET kit (Macherey-Nagel, Düren, DE) following the manufacturer's protocol. The viral RNA from avian and swine IAV was extracted previously as part of routine molecular diagnostics of IAV at the FLI by using the QIamp Viral RNA Mini Kit (Qiagen, Hilden, DE) from 140 µL volume of each avian field sample (combined swab) or by using 100 µL volume of each swine sample (nasal swab) within the NucleoMaq®VET Kit (Macherey-Nagel), according to the manufacturer´s instructions. Origin and some metadata of these samples are summarized in supplementary Table S2.

RT-qPCR

To determine the relative IAV viral load, we quantified the viral RNA by RT-qPCR using the LightCycler Multiplex RNA Virus Master (Roche, Basel, CH) combined with previously described primers ¹⁸. Briefly, 2 µL of RNA template was added to the reaction mixture composed of 1x RT-qPCR Reaction Mix, 1x RT Enzyme Solution, 0.8 µM of each forward (SVIP-MP-F) and reverse (SVIP-MP-R) primers, 0.2 µM of the probe (SVIP-MP_P2-MGB), supplemented to a total volume of 10 µL with PCR grade water. The RT-qPCR analysis was performed on a LightCycler 480 (Roche) with an RT step for 10 minutes at 50°C, followed by a heat-inactivation step for 30 seconds at 95°C. Followed by 45 cycles of denaturation (5 seconds, 95°C) and annealing and elongation steps (30 seconds, 60°C) with fluorescence readout. Finally, the cycling was concluded with a cooling step (30 seconds, 40°C).

RT-PCR amplification

To avoid potential buffer composition incompatibility between the cDNA synthesis subsequent PCR, we chose to use the LunaScript RT Master Mix Kit (Primer-free) (New England BioLabs Inc. (NEB), Ipswich, MA, USA) for the reverse transcription (RT). For this, we used the previously described MBTuni-12 and MBTuni-12.4 primers in a ratio of 1:4 at final molarity of 0.5 µM, and 7.5 µL of RNA eluate as input ¹³. The cDNA synthesis consisted of two steps (2 minutes at 25°C and 30 minutes at 55°C) followed by heat inactivation of the enzyme (1 minute, 95°C). Following the RT, we used 2.5 µL of cDNA as a template for a 25 µL PCR reaction with the Q5 Hot Start High-Fidelity DNA Polymerase (NEB) and 200 µM dNTP mix (Promega, Madison, WI, USA), 0.02 U/µL of Q5 Hot Start High-Fidelity DNA Polymerase (NEB), 0.2 µM of each primer MBTuni-13 and MBTuni-12.4R (5'-ACG CGT GAT CAG CRA AAG CAG G-3') or with barcoded primer pairs (Uni13-BCxx, Uni12-BCxx; Table S1). The PCR protocol included an initial denaturation step of 30 seconds at 98°C, followed by 35 cycles of denaturation (10 seconds at 98°C), annealing (20 seconds at 64°C), and elongation (105 seconds at 72°C), concluding with a final elongation step of 5 minutes at 72°C.

Library preparation and sequencing

Following PCR amplification, amplicons were subjected to a size selection using AMPure XP Bead-Based Reagent (Beckman Coulter, Brea, CA, USA) in a 0.5x ratio to omit PCR amplicons smaller than 500 bp. After adding the beads to the PCR amplicons, the plate was loaded in the KingFisher Apex automated extraction instrument (ThermoFisher Scientific) using a custom in-house protocol. After the initial mixing and binding of the amplicons with the beads, two sequential washing steps with 80% ethanol were performed, followed by a bead drying step, and finished with the elution in the same volume of the input of nuclease-free water. At this point, the samples amplified using the barcoded primers were pooled. Following the size selection, individual samples of sample pools were quantified using the Qubit 1X dsDNA HS Assay Kit (ThermoFisher Scientific) on a Qubit 4 fluorometer (ThermoFisher Scientific). For the ligation-based nanopore sequencing library preparation (SQK-NBD114.96, Oxford Nanopore Technologies (ONT), Oxford, UK), a total of 210 ng (200 fmol) was used as input for library preparation according to the manufacturer's instructions. The sequencing was done on a MinION Mk1B or GridION X5 device (ONT) in combination with a MinION flow cell (R10.4.1, FLO-MIN114, ONT) and resulting POD5 files were re-basecalled on the high-performance cluster (HPC) of the University of Bern using the Dorado (> v0.5.0, ONT) basecaller with the Super accurate (SUP) basecalling model from ONT ²⁰.

Data analysis

Before analysis, the output from the Dorado basecaller was demultiplexed into per-barcode Binary Alignment Map (BAM) files based on the native barcoding sequencing library kit used (SQK-NBD114.96), without trimming and requiring barcodes at both ends of each sequence read. For individual barcoded samples, the samples were subsequently emitted as individual Fastq files. For dual-barcoded samples, the barcode classification header (BC) in the initial per-barcode BAM files was modified using Samtools (v1.13) to allow reclassification of reads during another round of demultiplexing into per-barcode BAM files based on a custom barcode kit arrangement, without trimming and requiring barcodes at both ends of each sequence read ²². Thereafter, custom barcodes were emitted as individual Fastq-files. The individual emitted Fastq-files were analyzed using the EPI2ME-labs (ONT) wf-flu pipeline (v1.0.1, ONT), using a minimal coverage cutoff of 10 ²⁴. Thereafter, to visualize the sequencing data, the depth for each nucleotide position was extracted from the BAM files generated by the wf-flu pipeline using Samtools. The depth files were subsequently used to generate the heatmaps using the R package ComplexHeatmap (v2.16.0) in R (v4.3.1) ^26,28.

A modified two-step approach increases IAV whole-genome detection sensitivity

The original mRT-PCR developed by Zhou et al. targeting the conserved nucleotide termini of each IAV genomic segment is often used for whole genome surveillance of IAV at the human-animal interface. However, recovering whole genome sequence information remains challenging for samples with a low viral load, even using the two-step approach by Rambo-Martin et al. with modified primer ratios to accommodate increased sequencing coverage of the polymerase gene segments (i.e., PB1, PB2, and PA) ¹³. Since both approaches use PCR reaction temperatures below the optimal reaction temperatures to facilitate primer binding at the conserved nucleotide termini (personal communication Bin Zhou), we first sought to assess if elevating the individual reaction temperatures of the Reverse Transcription (RT) and PCR stages of the workflow increases the overall detection sensitivity (i.e., full genome recovery for samples with a low viral load).

To circumvent any potential buffer composition incompatibility between the separate cDNA synthesis and PCR reactions, we selected the LunaScript Master Mix Kit and the Q5 DNA polymerase ¹⁹. Similar to Rambo-Martin et al., instead of two independent Uni12 primers (i.e., MBTuni12 and MBTuni12.4), we used a degenerative Uni12 primer (MBTuni12.4R) for the PCR to compensate for U/C variation at the fourth position of the 3′ terminus (Fig. 1A). We then modified the two-step approach by Rambo-Martin et al. by increasing the incubation temperatures during the RT (55°C) and using a single annealing temperature during PCR (64°C) with 35 cycles. Finally, we used six 10-fold dilutions of RNA extracted from Madin-Darby canine kidney II (MDCKII) cell culture supernatant of an IAV stock (A(H1N1)pdm09) as a template to compare the overall whole-genome detection sensitivity of our modified two-step approach to the initial single-step protocol described by Zhou et al. (Figure S1) ¹¹. Following the individual reactions, we resolved the products on an agarose gel. This demonstrated that for the initial single-step protocol all amplicons corresponding to the eight genomic IAV segments could be detected up to the first ten-fold dilution; however, in subsequent dilutions one or more amplicons were missing. Also, the HA and M segments appeared to be disproportionately amplified. In contrast, using the modified two-step approach, all eight genomic IAV segments were more evenly amplified and could be detected up to the fourth ten-fold dilution, whereas in the highest dilution (10^− 5) only the amplicon corresponding to the NP segment was missing (Fig. 1B). These results indicate that our modified two-step approach has a higher overall whole-genome detection sensitivity than the initial protocol.

We observed that regardless of the amplification protocol used, a relatively high number of smaller amplicons were visible on the agarose gels (Fig. 1B). Because these smaller amplicons were in equal or higher molecular abundance than the genomic segments of interest, we added a magnetic bead-based size selection step to remove amplicons smaller than 500 bp prior to the sequence library preparation (Figure S1). Following the size selection, we sequenced all samples and assessed the coverage and depth of each genomic segment (Fig. 1C). This corroborated the previous results, where the full IAV genome could be recovered in the first five dilutions using the optimized two-step approach, whereas with the original one-step protocol full genome recovery was limited to the first dilution. Furthermore, in contrast to the original one-step protocol, our optimized protocol appears to result in a more even distribution of sequencing depth among all eight IAV segments. Overall, these results show that our modified two-step approach increases whole-genome detection sensitivity by approximately 1000-fold and evens the coverage distribution across all eight genomic segments.

Dual-barcodes for high-throughput whole genome sequencing

Although the portable third-generation sequencing platform from ONT is often used for the genomic surveillance of IAV, its throughput, namely the number of samples per run, is currently restricted by the number of available barcodes provided by the sequencing libraries (i.e., 24 or 96). Because Rambo-Martin et al. previously used modified primers targeting the 3’ and 5’ termini of the eight genomic segments that readily included the ONT native sequencing barcodes, we evaluated whether it would be possible to modify this approach using a dual-barcoding strategy for whole genome sequencing of IAV ¹³. Thus, we created different barcoded primer sets targeting the 3’ and 5’ termini of the eight genomic segments using the first 24 barcode sequences of the ONT PCR barcoding expansion kit (EXP-PBC096) with additional flanking sequences for later demultiplexing (Fig. 2A), which are compatible with the ONT native barcoding kit (SQK-NBD114.24 or SQK-NBD114.96). Using the enhanced two-step approach described above, we used cDNA from a single IAV sample to verify that all 24 primer sets resulted in similar amplification and sequencing coverage of the eight genomic segments. Following the amplification, we pooled the amplicons of eight individual primer pairs together for subsequent barcoding with the ONT native barcoding kit and sequencing. This revealed that all eight segments of the IAV genome could be recovered with at least 100x coverage for all 24 samples (Fig. 2B). However, four primer sets showed a slightly reduced amplification for certain genomic segments (BC09, BC15, BC16, and BC20). Nonetheless, our results show that it is possible to use a dual-barcoding approach for whole genome sequencing of IAV, this has the potential of increasing the throughput for one library preparation by at least eight-fold.

High-throughput WGS of clinical samples

To evaluate whether our modified two-step approach (single and dual-barcoding strategies) could be used to successfully generate whole genome sequences from clinical samples derived from different host species with a wide range of Cp-values (16–36) and multiple IAV subtypes (i.e., avian (H4N6, H5N1, H6N1, H7N7, H9N9, H11N9); human (H3N2); swine (H1avN2, H1pdmN1, H1huN2, and H3huN2)), we tested this approach on 24 IAV-positive clinical samples of human, swine, and avian origin. Following RNA extraction and the RT step, IAV was amplified from the samples with either the MBTuni primer set designed by Zhou et al. or with our dual-barcoded primer set (Table S1). For the latter, eight individual barcoded samples were pooled together for subsequent barcoding with the ONT native barcoding kit. After sequencing and sequence assembly, we generated heatmaps to compare the coverage and depth of each segment for all samples.

Overall, 16 whole genomes were recovered using both single and dual-barcoding strategies. Of the 24 IAV-positive samples (8 per host species), 18 could be completely subtyped and 3 partially subtyped using our modified two-step approach with one or both barcoding strategies (Fig. 2C). For the avian samples, the full genome was recovered from four samples with both barcoding strategies (AIV-11, AIV-12, AIV-17, AIV-23), while for two additional samples (AIV-19, AIV-20) the full genome was recovered using only one of the two barcoding strategies. Only the smaller genomic segments were recovered from the two remaining samples (AIV-02, AIV-14), which had Cp-values of 27.7 or higher (Fig. 2C) and could only be partially subtyped. Full genomes were recovered from five samples with both barcoding strategies for the swine samples, whereas for one sample (swIAV-05) one or more segments were missing in either approach. This included the HA segment, and therefore this sample could only be partially subtyped based on the NA. Finally, there were two samples with a Cp-value above 35 (Fig. 2C), where no segments could be recovered with either barcoding approach. In contrast to the avian and swine samples, seven full genomes were recovered for the human IAV samples regardless of the barcoding strategy up to a Cp-value 34 (Fig. 2C). However, one sample tested negative in the qPCR, and no full genome was recovered from this sample (Table S2). Since third-generation sequencing platforms enable sequencing of the entire individual DNA amplicons, in principle, it is possible to detect defective viral genomes (DVGs) that are generated during infection. We therefore also evaluated if DVGs can be detected in clinical samples. For this we allowed the aligner to be splice aware and assembled the 5’ and 3’ termini of possible DVGs, as done previously for influenza D virus, and then determined the sequencing coverage of the individual IAV genomic segments ²¹. This revealed that for several samples there was a sharp drop in the coverage of the middle region of especially the PB1, PB2, and PA segments, whereas in some samples this is absent (Figure S2). Although the underlying reason for the presence/absence of DVGs in certain samples remains unclear, our analysis indicates that our approach can readily detect DVGs in clinical samples from avian, swine, and human origin.

Combined, these results demonstrate that with our revised two-step PCR protocol, both the single and dual-barcoding strategies can be used for WGS of IAV-positive samples originating from diverse host species using a portable third-generation sequencing platform.

In the present study we demonstrate that by adapting the RT and PCR cycling conditions we can increase the sensitivity of recovering IAV whole genome sequences by approximately 1000-fold compared to the original multisegment RT-PCR (mRT-PCR) protocol for IAV. Furthermore, we show that with the inclusion of barcoded primers we can multiplex at least eight clinical samples of avian, swine, or human origin per sequencing library preparation without a significant loss in sensitivity. Collectively, this results in an enhanced workflow for high-throughput whole genome surveillance of influenza A virus in various host species.

Here, we demonstrate that dual-barcoding is possible on the ONT sequencing platform with up to eight samples per native ONT sequencing barcode library. Based on the number of barcodes available for the ONT PCR barcoding expansion kit (EXP-PBC096), this can, in principle, be expanded to 96 samples per native ONT sequencing barcode library. However, in our experimental setting we did not quantify individual barcoded samples, but only the pool of samples to ensure that that each native ONT sequencing barcode library was made with 200 fmol. Therefore, including more barcodes could dilute out samples from which the amplicon concentration is low. To overcome this, one needs to quantify individual barcoded samples, normalize them during pooling, and empirically determine the maximum number of individual barcoded samples that can be pooled per native ONT sequencing barcode library. This likely will further reduce the number of native ONT sequencing barcode libraries needed, without compromising full genome recovery detection sensitivity, but would be more laborious compared to the “blinded” pooling approach described in the current study, and alternatively can be resolved with a deeper sequencing depth. Finally, although the dual-barcoding approach has thus far only been evaluated for WGS, it is conceivable to adapt this methodology to other multisegment and multiplex approaches for genomic surveillance of seasonal human influenza A and B viruses ^23,25.

The multisegment RT-PCR is an essential tool to perform whole genome sequencing of Influenza A viruses (IAVs). We show that our optimized protocol can be used to sequence IAV-positive samples from avian, swine, and human origin up to a Cp-value of approximately 34. However, we observed that regardless of the barcoding strategy, the samples of avian origin had a reduced genomic recovery compared to the samples of swine or human origin, which had even lower viral loads than the avian IAV samples (Fig. 2C). This discrepancy suggests that sample origin can significantly influence detection sensitivity. However, we cannot exclude that this might be influenced by the fact that the avian and swine samples, in contrast to the human samples, underwent several freeze/thaw cycles prior to the experiments, which can negatively affect RNA quality and integrity. Similarly, the avian samples were extracted manually with a column-based kit, whereas the human and swine samples were extracted using an automated magnetic bead-based method, resulting in the avian samples having a lower sequencing sensitivity, which aligns with a previous report ³¹. Moreover, given that influenza A viruses cause acute infections, the time of sampling in relation to the time of infection is important, as towards the end of the illness the viral load in clinical samples decreases ^32,33. Therefore, it would be interesting to evaluate our protocol in a more controlled study to determine whether full genome recovery rates are influenced by the sample processing as well as the time of sampling.

Because the exact role of defective viral genomes (DVGs) in vivo and in vitro during virus infection remains largely elusive ²⁷, there have been several short-read based bioinformatic pipelines established to analyse this phenomenon ^29,30. Thus, while the detection of DVGs for IAV is not novel per se, we demonstrate that in addition to recovering whole genome information our WGS approach can also be used to detect IAV DVGs in clinical specimens, without any potential fragment partitioning during sequence library preparation ³⁰ However, because DVGs of PB1, PB2, and PA are on average 400–500 nt in size, and HA, NA, NP, M, and NS DVGs are on average around 400 nt, it is possible that the size-selection step of our protocol prior to sequencing depletes some of the DVG amplicons shorter than 500 bp. This likely reflects why most DVGs we detected originated from the largest genomic segments of IAV (e.g., PB1, PB2, and PA) (Figure S2). Therefore, it remains to be determined if our approach can reliably detect DVGs of the small genomic segments of IAV, as well as how long-read NGS methodologies compare to short-read NGS approaches.

In conclusion, the optimized and scalable whole genome sequencing workflow for influenza A virus, combined with the availability of a portable third-generation sequencing platform, enables genomic sequence surveillance at the human-animal interface. This significantly enhances sensitivity and throughput, facilitating the early detection and monitoring of IAV evolution and zoonotic spillovers in clinical samples from diverse origins.

Acknowledgments

We want to thank Alban Ramette, Loïc Borcard, and Sonja Gempeler from the Institute for Infectious Diseases, University of Bern, Switzerland, for the sequencing of the samples and the Multidisciplinary Center for Infectious Diseases (MCID), University of Bern, Switzerland, and the Swiss National Science Foundation (IZCOZ0_220329) for providing the funding for this project to Jenna N. Kelly and Ronald Dijkman. RNA isolation and virus characterization of porcine samples have been funded by the European ICRAD project “PIGIE” (number 2821ERA24).

Author contributions

Conceptualization, R.D.; methodology, R.D. and M.L.; formal analysis, R.D., M.M., and M.L.; investigation, M.L. and M.F.L.; resources, F.S., P.B., C.S., A.G., and T.H.; writing—original draft, M.L.; writing—review and editing, M.L., A.G., T.H., J.K., and R.D.; visualization, M.L.; supervision, R.D.; funding acquisition, T.H., J.K., and R.D.

Declaration of interests

The authors declare no competing interests.

Data Availability Statement: The data presented in the study have been deposited in the European Nucleotide Archive (ENA) at EMBL-EBI under the accession number PRJEB79943 (https://www.ebi.ac.uk/ena/browser/view/ PRJEB79943).

Chauhan, R.P., and Gordon, M.L. (2022). An overview of influenza A virus genes, protein functions, and replication cycle highlighting important updates. Virus Genes 2022 58:4 58, 255–269. https://doi.org/10.1007/S11262-022-01904-W.
Krammer, F., Smith, G.J.D., Fouchier, R.A.M., Peiris, M., Kedzierska, K., Doherty, P.C., Palese, P., Shaw, M.L., Treanor, J., Webster, R.G., et al. (2018). Influenza. Nat Rev Dis Primers 4, 1–21. https://doi.org/10.1038/s41572-018-0002-y.
Fereidouni, S., Starick, E., Karamendin, K., Genova, C. Di, Scott, S.D., Khan, Y., Harder, T., and Kydyrmanov, A. (2023). Genetic characterization of a new candidate hemagglutinin subtype of influenza A viruses. Emerg Microbes Infect 12. https://doi.org/10.1080/22221751.2023.2225645.
Chan, P.K.S. (2002). Outbreak of Avian Influenza A(H5N1) Virus Infection in Hong Kong in 1997. Clinical Infectious Diseases 34, S58–S64. https://doi.org/10.1086/338820.
Castillo, A., Fasce, R., Parra, B., Andrade, W., Covarrubias, P., Hueche, A., Campano, C., Tambley, C., Rojas, M., Araya, M., et al. (2023). The first case of human infection with H5N1 avian Influenza A virus in Chile. J Travel Med 30, 1–3. https://doi.org/10.1093/JTM/TAAD083.
Technical Report: Highly Pathogenic Avian Influenza A(H5N1) Viruses | Avian Influenza (Flu) https://www.cdc.gov/flu/avianflu/spotlights/2023-2024/h5n1-technical-report_april-2024.htm.
AbuBakar, U., Amrani, L., Kamarulzaman, F.A., Karsani, S.A., Hassandarvish, P., and Khairat, J.E. (2023). Avian Influenza Virus Tropism in Humans. Viruses 2023, Vol. 15, Page 833 15, 833. https://doi.org/10.3390/V15040833.
Gao, R., Cao, B., Hu, Y., Feng, Z., Wang, D., Hu, W., Chen, J., Jie, Z., Qiu, H., Xu, K., et al. (2013). Human Infection with a Novel Avian-Origin Influenza A (H7N9) Virus. https://doi.org/10.1056/NEJMoa1304459 368, 1888–1897. https://doi.org/10.1056/NEJMOA1304459.
Garten, R.J., Davis, C.T., Russell, C.A., Shu, B., Lindstrom, S., Balish, A., Sessions, W.M., Xu, X., Skepner, E., Deyde, V., et al. (2009). Antigenic and genetic characteristics of swine-origin 2009 A(H1N1) influenza viruses circulating in humans. Science (1979) 325, 197–201. https://doi.org/10.1126/SCIENCE.1176225/SUPPL_FILE/GARTEN.SOM.REV.PDF.
Shao, W., Li, X., Goraya, M.U., Wang, S., and Chen, J.L. (2017). Evolution of Influenza A Virus by Mutation and Re-Assortment. Int J Mol Sci 18. https://doi.org/10.3390/IJMS18081650.
Zhou, B., Donnelly, M.E., Scholes, D.T., St George, K., Hatta, M., Kawaoka, Y., and Wentworth, D.E. (2009). Single-reaction genomic amplification accelerates sequencing and vaccine production for classical and Swine origin human influenza a viruses. J Virol 83, 10309–10313. https://doi.org/10.1128/JVI.01109-09.
Miah, M., Hossain, M.E., Hasan, R., Alam, M.S., Puspo, J.A., Hasan, M.M., Islam, A., Chowdhury, S., and Rahman, M.Z. (2023). Culture-Independent Workflow for Nanopore MinION-Based Sequencing of Influenza A Virus. Microbiol Spectr 11. https://doi.org/10.1128/SPECTRUM.04946-22.
Rambo-Martin, B.L., Keller, M.W., Wilson, M.M., Nolting, J.M., Anderson, T.K., Vincent, A.L., Bagal, U.R., Jang, Y., Neuhaus, E.B., Davis, C.T., et al. (2020). Influenza A Virus Field Surveillance at a Swine-Human Interface. mSphere 5, e00822-19. https://doi.org/10.1128/mSphere.00822-19.
Leguia, M., Garcia-Glaessner, A., Muñoz-Saavedra, B., Juarez, D., Barrera, P., Calvo-Mac, C., Jara, J., Silva, W., Ploog, K., Amaro, Lady, et al. (2023). Highly pathogenic avian influenza A (H5N1) in marine mammals and seabirds in Peru. Nature Communications 2023 14:1 14, 1–11. https://doi.org/10.1038/s41467-023-41182-0.
Kandeil, A., Patton, C., Jones, J.C., Jeevan, T., Harrington, W.N., Trifkovic, S., Seiler, J.P., Fabrizio, T., Woodard, K., Turner, J.C., et al. (2023). Rapid evolution of A(H5N1) influenza viruses after intercontinental spread to North America. Nature Communications 2023 14:1 14, 1–13. https://doi.org/10.1038/s41467-023-38415-7.
Watson, S.J., Langat, P., Reid, S.M., Lam, T.T.-Y., Cotten, M., Kelly, M., Van Reeth, K., Qiu, Y., Simon, G., Bonin, E., et al. (2015). Molecular Epidemiology and Evolution of Influenza Viruses Circulating within European Swine between 2009 and 2013. J Virol 89, 9920–9931. https://doi.org/10.1128/JVI.00840-15.
Jackson, M.L., Chung, J.R., Jackson, L.A., Phillips, C.H., Benoit, J., Monto, A.S., Martin, E.T., Belongia, E.A., McLean, H.Q., Gaglani, M., et al. (2017). Influenza Vaccine Effectiveness in the United States during the 2015–2016 Season. New England Journal of Medicine 377, 534–543. https://doi.org/10.1056/NEJMOA1700153.
Nagy, A., Cerníková, L., Kunteová, K., Dirbáková, Z., Thomas, S.S., Slomka, M.J., Dán, Á., Varga, T., Máté, M., Jirincová, H., et al. (2021). A universal RT-qPCR assay for “One Health” detection of influenza A viruses. PLoS One 16, e0244669. https://doi.org/10.1371/JOURNAL.PONE.0244669.
LunaScript® RT Master Mix Kit (Primer-free) | NEB https://www.neb.com/en/products/e3025-lunascript-rt-master-mix-kit-primer-free.
GitHub - nanoporetech/dorado: Oxford Nanopore’s Basecaller https://github.com/nanoporetech/dorado.
Probst, L., Laloli, L., Licheri, M.F., Licheri, M., Gultom, M., Holwerda, M., V’kovski, P., and Dijkman, R. (2023). Generation and Characterization of an Influenza D Reporter Virus. Viruses 15. https://doi.org/10.3390/V15122444.
Danecek, P., Bonfield, J.K., Liddle, J., Marshall, J., Ohan, V., Pollard, M.O., Whitwham, A., Keane, T., McCarthy, S.A., and Davies, R.M. (2021). Twelve years of SAMtools and BCFtools. Gigascience 10, 1–4. https://doi.org/10.1093/GIGASCIENCE/GIAB008.
Zhou, B., Deng, Y.-M., Barnes, J.R., Sessions, O.M., Chou, T.-W., Wilson, M., Stark, T.J., Volk, M., Spirason, N., Halpin, R.A., et al. (2017). Multiplex Reverse Transcription-PCR for Simultaneous Surveillance of Influenza A and B Viruses. J Clin Microbiol 55, 3492–3501. https://doi.org/10.1128/JCM.00957-17.
GitHub - epi2me-labs/wf-flu https://github.com/epi2me-labs/wf-flu.
Zhou, B., Lin, X., Wang, W., Halpin, R.A., Bera, J., Stockwell, T.B., Barr, I.G., and Wentworth, D.E. (2014). Universal Influenza B Virus Genomic Amplification Facilitates Sequencing, Diagnostics, and Reverse Genetics. J Clin Microbiol 52, 1330. https://doi.org/10.1128/JCM.03265-13.
Gu, Z. (2022). Complex heatmap visualization. iMeta 1, e43. https://doi.org/10.1002/IMT2.43.
Vignuzzi, M., and López, C.B. (2019). Defective viral genomes are key drivers of the virus-host interaction. Nat Microbiol 4, 1075–1087. https://doi.org/10.1038/S41564-019-0465-Y.
Gu, Z., Eils, R., and Schlesner, M. (2016). Complex heatmaps reveal patterns and correlations in multidimensional genomic data. Bioinformatics 32, 2847–2849. https://doi.org/10.1093/BIOINFORMATICS/BTW313.
Routh, A., and Johnson, J.E. (2014). Discovery of functional genomic motifs in viruses with ViReMa–a Virus Recombination Mapper–for analysis of next-generation sequencing data. Nucleic Acids Res 42, e11. https://doi.org/10.1093/NAR/GKT916.
Alnaji, F.G., Holmes, J.R., Rendon, G., Vera, J.C., Fields, C.J., Martin, B.E., and Brooke, C.B. (2019). Sequencing Framework for the Sensitive Detection and Precise Mapping of Defective Interfering Particle-Associated Deletions across Influenza A and B Viruses. J Virol 93. https://doi.org/10.1128/JVI.00354-19.
Di, H., Thor, S.W., Trujillo, A.A., Stark, T.J., Marinova-Petkova, A., Jones, J., Wentworth, D.E., Barnes, J.R., and Davis, C.T. (2019). Comparison of nucleic acid extraction methods for next-generation sequencing of avian influenza A virus from ferret respiratory samples. J Virol Methods 270, 95–105. https://doi.org/10.1016/J.JVIROMET.2019.04.014.
Houlihan, C.F., Frampton, D., Bridget Ferns, R., Raffle, J., Grant, P., Reidy, M., Hail, L., Thomson, K., Mattes, F., Kozlakidis, Z., et al. (2018). Use of Whole-Genome Sequencing in the Investigation of a Nosocomial Influenza Virus Outbreak. J Infect Dis 218, 1485. https://doi.org/10.1093/INFDIS/JIY335.
Richman, D.D., Murphy, B.R., Baron, S., and Uhlendorf, C. (1976). Three strains of influenza A virus (H3N2): interferon sensitivity in vitro and interferon production in volunteers. J Clin Microbiol 3, 223. https://doi.org/10.1128/JCM.3.3.223-226.1976.

No competing interests reported.

FigureS1.pdf
Figure S1. Comparison of the step used in the Zhou et al. and the optimized approaches. The protocol established by Zhou et al. (top) is compared to the one optimized during this study (bottom) by showing the different steps and the primers used in the RT and PCR (Created in BioRender. Dijkman, R. (2024) BioRender.com/l62f505).
FigureS2.pdf
Figure S2. Sequencing coverage plots for each genomic segment recovered from the avian, human, and swine samples. To assess the coverage for each segment from the IAV of each species, we plotted the coverage depth in the function of each nucleotide position. The depth was normalized it to the maximum depth value per segment.
TableS1.xlsx
Table S1. Barcoded primers table. The table includes all the barcoded primers designed and used in this study. The underlined nucleotides at the end of each primer sequence correspond to the conserved nucleotides at each IAV segment end (i.e., Uni12 and Uni13).
TableS2.xlsx
Table S2. Metadata of RNA samples extracted from IAV-positive clinical samples or virus isolates of avian and porcine origin.

Download PDF

Reviews received at journal
08 Nov, 2024
Reviews received at journal
07 Nov, 2024
Reviewers agreed at journal
31 Oct, 2024
Reviewers agreed at journal
29 Oct, 2024
Reviewers invited by journal
15 Oct, 2024
Editor assigned by journal
08 Oct, 2024
Submission checks completed at journal
07 Oct, 2024
First submitted to journal
07 Oct, 2024

You are reading this latest preprint version

Optimized workflow for high-throughput whole genome surveillance of Influenza A virus

Status:

Version 1

Abstract

Figures

Background

Material and methods

Viral RNA

RT-qPCR

RT-PCR amplification

Library preparation and sequencing

Data analysis

Results

A modified two-step approach increases IAV whole-genome detection sensitivity

Dual-barcodes for high-throughput whole genome sequencing

High-throughput WGS of clinical samples

Discussion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1