Metagenomic Sequencing for Virus Discovery and Characterization - Dr. Ben Hause, Kansas State University, from the 2015 North American PRRS Symposium, December 4 - 5, 2015, Chicago, IL, USA. More presentations at http://www.swinecast.com/2015-north-american-prrs-symposium

1. Metagenomic sequencing for virus discovery and
characterization
Kansas State Veterinary Diagnostic Laboratory
Ben Hause, MS, PhD

2. Why NGS vs. Sanger? More Data!
• NGS generates massive amounts of sequence data
– Sanger= 1 read, ~800bp
– NGS=1M reads, 300bp
• NGS can be sequence dependent or independent
– Don’t need to know what your sequencing
– Unique to this technology
• Amount of data for $
– PRRS ORF5
• Sanger $182/600bp=$0.30/bp
• NGS: $300/15400bp=$0.02/bp
• More comprehensive picture of virus: whole genome vs. 1 gene
• Ability to detect multiple viruses or quasispecies
• Metagenomic sequencing: sequencing material directly recovered from the environment
– Nasal swab

3. Metagenomic viral
RNA and DNA
(sample pretreated
with DNase/RNase
cocktail
Random hexamer with
5’-20bp barcode
Reverse Transcription and
Second Strand Synthesis
(RNA -> cDNA->dsDNA)
PCR Amplification
using primer identical
to 20bp barcode
Amplicon pools
generated from
randomly amplified
virus nucleic acid

4. Full genome sequence of porcine parainfluenza 1 (PPIV1)
virus from a nasal swab
• 11 M reads
• 52,111 mapped to
PPIV1
– 0.45% reads
• 361x average
coverage

5. Next Generation Sequencing for
PRRSV Epidemiology

6. • Attempted sequencing from sera
NPB grant 14-204, Development of next generation sequencing
methodology for full genome characterization of PRRSV
• NGS on sera from 182 samples that were qRT-PCR positive
– 66 complete genomes
– Others had PRRSV sequence but not complete
• Parvoviruses

7. -phylogenetic analysis of glycoproteins identified 4-7 well supported clades
-<10% diversity within clade; >10% diversity between clades
RFLP174

8. Topological incongruences
Gene Combination Representative Strain Strains (n) GP2 GP3 GP4 GP5 M
1 14-79 (Inglevac ATP) 6 2.1 3.1 4.1 5.1 6.1
2 14-64 (Inglevac MLV) 4 2.2 3.2 4.2 5.1 6.2
3 ISU10 11 2.3 3.3 4.3 5.2 6.6
4 ISU49 1 2.3 3.3 4.3 5.5 6.6
5 ISU94 8 2.3 3.3 4.7 5.7 6.3
6 ISU81 5 2.4 3.4 4.4 5.6 6.4
7 ISU32 1 2.4 3.4 4.6 5.4 unassigned
8 ISU28 6 2.4 3.4 4.6 5.2 6.6
9 ISU3 7 2.5 3.3 4.3 5.2 6.6
10 ISU23 5 2.5 3.3 4.3 5.5 6.4
11 ISU39 7 2.5 3.3 4.3 5.5 6.6
12 ISU90 4 2.5 3.4 4.5 5.4 6.5
13 104194 1 unassigned unassigned unassigned 5.1 unassigned

9. Nsp2 phylogeny and deletions
VR2332-like
MN184-like
Novel
• 11/66 genomes (17%) identified
as PRRSV RFLP174
• >95% identity in GP2a, GP3,
GP4, GP5, M, N
• 74-100% identity in nsp2

10. Why perform PRRS full genome sequencing?
• GP5 is the major surface glycoprotein, is highly variable and contains a
neutralizing epitope, however…
– Delayed GP5 aby response after clearance of viremia
• Equivalent diversity in minor glycoproteins (GP2a/GP3/GP4)
– GP2a/GP4 are receptor binding proteins
– Known B-cell and neutralization epitopes
• Phylogenetic incongruence= which tree do you believe for epi studies?
– Viruses with same clade of GP5 but different clades for GP2a-4
• Nsp2 related to pathogenicity
• Can detect other viruses in serum (new pestivirus APPV, porcine parvovirus
6)
• Not significantly more money that Sanger

11. Atypical porcine pestivirus (APPV)
• Highly divergent pestivirus identified in pig sera sequenced
for NPB PRRSV genome sequencing project
• Very distantly related to BVDV, CSFV (~25% identity)
• More closely related to recently partially described bat
pestivirus in China (~68% identity)
• 5/182 positive serum samples (2.7%)
Hause et al, 2015, Journal of General Virology

12. BLASTP analysis of APPV

13. • MEGA6 phylogenetic trees: a) polyprotein
b) Npro c) Erns d) NS3

14. qPCR results
• Two qRT-PCR assays were developed targeting either E2 or the Erns region of the genome
• 3 of 5 were positive on both assays, one positive only on Erns assay and another sample on
tested on E2 due to qns
– Sample only positive for Erns showed only 88% aa identity for the partial E2 sequence
• Samples originated from NE, AZ, NC, MN, and KS

15. ELISA results
• ELISA to detect antibodies cross-reactive to Erns peptides produced in
E. coli

16. APPV and shaker pigs
• Starting in 2014, approx. 700 5-14 week old pigs exhibited intention
tremors, increased respiratory rates, and as disease progressed,
inability to swallow
– Mortality was 100% within 4 days of onset
• Metagenomic sequencing of brain homogenate mapped 195 reads
to APPV at 87% identity
• qRT-PCR from pig on brain, lymph nodes, heart tissue and serum were
positive for APPV
• August 2015 new group of pigs 10-16 weeks of age from same
system showed similar symptoms, lymph nodes from one pig was
APPV qRT-PCR positive
Transboundary and Emerging Diseases, submitted

18. IHC
• Anti-APPV Erns antibodies
were generated in mice
• IHC detected virus in the
liver, spleen and lymph
nodes
– Brain was negative
– Large number positive cells
in LN, localization in
cytoplasm

19. Recent Findings and future studies
• ISU was able to demonstrate APPV-like virus induced congenital shaker pigs after in utero
inoculation of fetuses
• How serious is this disease in the swine industry?
– Congenital tremors have been sporadically observed with no definitive etiologic agent
– Intention tremors and resulting mortality in older pigs new?
• Is there potential for persistent infection of piglets?
• Multiple subtypes resulting in different disease presentation?
– Observed ~15% E2 nt diversity
• Newly emerged virus or circulating unnoticed for some time?
– Virus X or pestivirus-like viruses isolated from pigs reported in the past

20. Porcine parainfluenza virus 1
• 10-21 day old pigs with acute respiratory disease
– Negative for influenza A virus and PRRSV
• Metagenomic sequencing of nasal swabs from IL and NE
– Complete PPIV1 genomes
U.S. PPIV1 genomes 90-95% identical to
Hong Kong PPIV1
Journal of General Virology, under revision
U.S.
H.K.

21. Hemagglutinin-Neuraminidase gene sequences
NE,
IL,
OK
HK
U.S. PPIV1 HN gene sequences 85-96% identical to PPIV1 from Hong Kong

22. PPIV1 incidence in U.S.
• 6.1% (17/279) lung/oral fluid/nasal swabs PCR+ for PPIV1
• Serology
– ELISA (recombinant F protein)=63% positive
– Immunocapture PCR= 55% positive

23. In situ hybridization
Nasal turbinates trachea

24. PPIV1
• Common in U.S. pigs
• Acute respiratory disease in pigs on sow farms/early
nursery
• Include in differential diagnostics for influenza/PRRS
• KState has qRT-PCR ($30/sample)
• No vaccines available
– Virus has not been isolated

25. Seneca Valley Virus
• Metagenomic sequencing of fecal/nasal swabs from 5 swine markets
in NC
– 2 slaughterhouses for healthy pigs (primary market)
– 2 cull slaughterhouses (secondary market)
– 1 buying station
• 5 pigs (nasal swab and fecal swab) per producer per site
– 5 producers per site
– 250 total swabs, analyzed in 50 pools (25 nasal, 25 rectal swab pools)
– 2 samplings, June/August, 2015
Emerging Infectious Diseases, submitted

26. • Large numbers of viruses identified=Virus Soup!
– Olivia Myers, NC State CVM at AASV, 2016
• Seneca valley virus identified from both samplings from 4/5
markets by sequencing
• qRT-PCR for SVV
– 26/50 (52%) June sampling
– 18/50 (36%) August sampling
• Primary market=1/40 (2.5%)
• Secondary market=43/60 (72%)
• Virus isolation, second sampling
– Positive for 5 samples (Ct values ~15-20)
• SVV much more common in lower health status pigs?
– Oral fluid testing for SVV, ~1% positive (ISU/UMN/SHIC)
– Cause versus effect???

27. N.C. 2015
Brazil 2015
2015 U.S., >99% identical
97% identical to Brazilian SVV, 2015
95% identical to Canadian SVV, 2011
87-92% identical to historical SVV

28. New circovirus species, PCV3
• NC sow farm, chronic poor reproductive performance
• Acutely dead sows with lesions consistent with PDNS
• Sows with PDNS-like skin lesions aborting
• Samples sent to ISU
– Lesions consistent with PCV2, however…
– PCV2 negative by PCR and IHC
• Samples sent to KState for metagenomic sequencing
Emerging Infectious Diseases, submitted

29. 2,000bp
Replicase=54% identical to bat circovirus
Capsid=36% identical to PCV2 and duck circovirus
ORF3=39% identical to Murid herpes protein (unknown function)

30. • Incidence determined by qPCR: 34/271 samples positive (12.5%)
• Quantified virus in symptomatic cases
– Sows with PDNS, Ct=28-30
– Mummies, Ct=17-21
• PCV2 history
– First identified in mid-1990’s
– Sporadic PMWS in Canada
– Severe systemic disease in Europe and Asia followed by U.S. in early/mid 2000’s
– Retrospective studies found evidence of PCV2 circulation for decades prior to
widespread disease
– Uncontrolled until commercial vaccines developed
• PCV3 on a similar trajectory???
– More research needed, including vaccine development!

31. NGS/Metagenomic Sequencing
• Powerful new method for veterinary diagnostics
• Complete viral genome sequencing
– Isolated virus
– Directly from clinical samples (with sufficient viral titer)
• Cases with unknown etiology
– Unusual clinical presentation
– Clinical symptoms with absence of usual suspects
• Profiling animal/herds
– Live exposure (rotavirus, PEDV, PDCoV, PRRSV)
• Affordable
– $300/sample
– Alternative: multiple PCRS, histopathology, culture, VI, etc., can easily
reach $300
• Need more widespread use!

32. Acknowledgements
• Kansas State University
• Dr. Dick Hesse
• Dr. Lalitha Peddireddi
• Dr. Jianfa Bai
• Dr. Erin Schirtzinger
• Dr. Emily Collin
• Rachel Palinski
• Dr. Namita Mitra
• South Dakota State University
• Travis Clement
• Iowa State University
• Dr. Phillip Gauger
• Swine Veterinarians
• Dr. Josh Duff
• Dr. Chad Smith
• Dr. Emily Byers
• Dr. John Prickett
• Those unnamed
• National Pork Board grant #14-204
• Zoetis
• Boehringer Ingelheim