Title: Molecular characterization of infectious bursal disease viruses from Pakistan

Authors: Muhammad Zubair Shabbir1, Muhammad Ali2, Muhammad Abbas3, Umer Naveed

Chaudhry4, Zia-ur-Rehman5, Muhammad Munir6

Affiliations: 1University of Veterinary and Animal Sciences, Lahore Pakistan; 2Bahauddin

Zakariya University, Multan, Pakistan; 3Livestock and Dairy Development Department,

Punjab Pakistan; 4Roslin Institute, University of Edinburgh Scotland (UK); 5Institute of

Pathophysiology and Immunology, medical university of Graz, Austria; 6The Pirbright

Institute, Woking Surrey, GU240NF, United Kingdom

CORRESPONDING AUTHOR:

Muhammad Zubair Shabbir

Assistant Professor

Quality Operations Laboratory

University of Veterinary and Animal Sciences

Lahore 54600, Pakistan

E. mail: shabbirmz@uvas.edu.pk

1

ABSTRACT:

Since the first report of infectious bursal disease in Pakistan in 1987, outbreaks are common

even in vaccinated flocks. Despite appropriate administration of vaccines, concerns arise if

the circulating strains are different from the ones used in vaccine. Here, we sequenced the

hypervariable region (HVR) of the VP2 gene of circulating strains of infectious bursal

disease viruses (IBDV) originating from outbreaks (n = 4) in broiler flocks in Pakistan.

Nucleotide sequencing followed by phylogeny and deduced amino acid sequence analysis

showed the circulating strains to be very virulent (vv), and identified characteristic residues at

position 222(A), 242(I), 256(I), 294(I) and 299(S). In addition, a substitution at position

221(Q→H) was found exclusive to Pakistani strains in our analysis, though a larger dataset is

required to confirm this finding. Compared to commonly used vaccine strains in Pakistan,

substitution mutations were found at key amino-acid positions in VP2 that may be

responsible for potential changes in neutralization epitopes and vaccine failure.

Key words: Infectious bursal disease virus, Hypervariable region, VP2, vvIBDV, Pakistan

2

Infectious bursal disease (IBD) is a highly contagious viral disease of birds characterized by

lesions in the bursa of Fabricius, immunosuppression, severe morbidity and mortality. The

etiological agent, infectious bursal disease virus (IBDV), belongs to the family Birnaviridae

and has a double stranded bi-segmented genome (RNA) comprised of segment A [(3,300

nucleotides (nt) in length], and segment B (2,800nt in length). Segment A encodes for two

proteins named VP2 (capsid protein, 48kDa) and VP3 (ribonucleoprotein, 32-35kDa), the

viral protease VP4 (24kDa), and a non-structural protein VP5 (17-21kDa). Segment B, on the

other hand, encodes RNA-dependent RNA polymerase, VP1 (90kDa) (Mundt and Muller,

1995). There are two distinct serotypes of IBDV. Viruses in serotype 1 are classified as

classical virulent, antigenic variant, and very virulent (vv). The viruses in serotype 2 are

considered to be non-pathogenic for chickens (Ismail et al., 1988).

Following the first detection of infectious bursal disease in 1987 in Pakistan (Khan et al.,

1988), a number of disease outbreaks have been observed, even in vaccinated flocks.

Attempts have been made to characterize IBDV strains through restriction-endonuclease (RE)

assays in Pakistan (Lone et al., 2009; Zahoor et al., 2011). However, multiple genetic events

have been identified recently that have led to alterations in IBDV sequences, including

sequences used in the RE assay (Hoque et al., 2001; Parede et al., 2003; Jackwood and

Sommer-Wagner, 2007). RE-assays are therefore not adequate to determine the on-going

genetic diversity and molecular epidemiology of vvIBDVs. To better expand our

understanding of the molecular epidemiology of circulating strains of IBDV in Pakistan, we

determined the nucleotide and translated amino acid sequences of the VP2 gene from IBDV

viruses originating from four outbreaks in broiler flocks, and compared the sequences to

those reported elsewhere around the globe.

We studied four outbreaks that occurred between 2009-2010 in broiler flocks with a high

morbidity (70 – 90%) and mortality (40-60%). Farms were located in one of the poultry

3

dense areas in Punjab province, the area in and around Lahore and Multan district. Each farm

(n = 30,000 birds/farm) had a history of vaccination with intermediate and hot (intermediate

plus) strains of IBDV. The age of the birds ranged from 22–27 days, and birds exhibited

symptoms such as anorexia, fever, huddling, whitish diarrhoea and pasting around vent.

Necropsy showed typical inflammatory lesions in the bursa of Fabricius, and haemorrhages

on the breast and thigh muscles.

Pools of bursa were collected from each outbreak and minced using a sterile mortar and

pestle. A 20% suspension was made with sterile normal saline supplemented with penicillin

(10,000IU) and streptomycin (1,000µg) followed by centrifugation at 4,000rpm for 15

minutes. The filtered supernatants (0.22µm, Milipore, USA) were adsorbed on the FTA

QIAcard (Qiagen, Hilden, Germany) and shipped to Swedish University of Agricultural

Sciences, Uppsala, Sweden for processing. The genetic material was eluted from four 3mm

punches using 100uL of elution buffer that contained RNA processing buffer (10 mM Tris-

HCl and 0.1 mM EDTA with pH 8.0), 800U/ml RNaseOutTM (Invitrogen, Carlsbad, U.S.A.)

and 2mM DTT [(1,4-Dithiothreitol) (Sigma-Aldrich, St. Louis, U.S.A.)]. The extracted

genome was stored at -20 oC until use.

The VP2 gene was amplified using specific primers and protocols as described previously

(Kataria et al., 2001). The amplified fragments were visualized using a trans-illuminator,

excised from the gel, and purified using the Wizard® SV Gel and PCR Clean-Up System

(Promega, Co., Madison, WI) according to the manufacturer’s instructions. Using the same

primers as were used for amplification, the positive PCR products were sequenced with an

ABI PRISM BigDye Terminator version 3.1 (Applied Biosystems, Foster City, CA), as

recommended by the manufacturer. Each DNA fragment was sequenced in both directions to

generate a reliable consensus sequence. Sequence assembly and editing were performed using

the SEQMAN program of DNASTAR Lasergene 8 (version 8.0.2 13; Madison, WI).

4

Construction of a phylogenetic tree was done with the neighbor-joining method using the

Kimura two-parameter model in MEGA version 5.2 (Tamura et al., 2011).

We determined the nucleotide sequence and the translated amino acid sequence of the

hypervariable region (HVR) of the VP2 gene for four IBDVs, each one representing a

different outbreak. The HVR is a major conformational and neutralising antigenic domain

spanning amino acid residues between positions 211-350. Since there are the greatest

differences among serotype-1 strains of IBDV in this region of genome, the nucleotide and

deduced amino acid residues spanning this part of genome have been widely used for

diagnostics and typing as variant, classic and very virulent (Jackwood and Sommer, 1999).

The nucleotide sequence spanning the HVR of strains isolated in the present study were

compared to the genome sequences available in PubMed (AJ310185).

The Pakistani strains clustered to vvIBDVs (VV-3) reported from Asian, African and

European countries between 1997 and 2010 (Figure 1). We observed a number of

synonymous substitutions within the study strains; the strains

IBDV/Pak/Chicken/MM126/2010 (KU321592) and IBDV/Pak/Chicken/MM127/2010

(KU321594) while IBDV/Pak/Chicken/MM125/2009 (KU321593) and

IBDV/Pak/Chicken/MM128/2009 (KU321595) were 100% identical to each other. When

compared to representative strains of each known serotype and vaccine strains, the study

variants showed divergence varying from 3.8 – 18.2. Maximum similarity in nucleotide

sequences were observed (96.3 and 96.1%) between the strains in the present study and a

very virulent virus isolated from white-leghorn chicken from Japan, VV3 (OKYM, GI:

1669530). Compared to the vaccine strains used commonly in Pakistan, a divergence of 8.3 –

8.9 (Bursaplex) and 10.4 – 10.6 (Bursine) was observed (Figure 2). Despite being studied

over a period of two years representing four outbreaks at different locations, we observed a

high genetic homogeneity particularly in residue patterns (Figure 3) and clustering within one

5

clade close to vvIBDVs suggesting a single origin of IBDVs. This is not unexpected as

genetic stability within the HVR of VP2 gene of IBDV has previously been reported in a

number of studies (Owoade et al., 2004; Martin et al., 2007; Jenberie et al., 2014).

The deduced amino acid sequence of HVR of VP2 was determined and compared to

representative strains of vvIBDVs [(98/084, VV1), (97/102, VV2) and (OKYM, VV3),

classical virulent IBDV [(002/73, Aus-CV) and (GBF-1, CV)] and a number of vaccine

strains such as V887 (Malaysian Vaccines and Pharmaceuticals), Bursaplex (Embrex Inc.,

U.S.A.), IBA (Bestar Laboratories Pte Ltd., Singapore), NVRI-VOM (National Veterinary

Research Institute, NVRI, Vom, Nigeria) MB (Abic Ltd, Israel), Bursine Plus (Wyeth–Fort

DodgeAnimal Health, U.S.A.) and Ventri (Ventri Biologicals Venkateshwara Hatcheries Ltd,

Pune, India). The analyzed HVR included 145 amino acids residues from position 211 to 354

of the VP2 protein. We observed 100% homogeneity within the study strains with a pattern

characteristic of vvIBDV that includes the amino acid residues A222, I242, I256, I294 and

S299 (Brown et al., 1994; Jackwood and Sommer-Wagner, 2007). Since the presence of

H/N/Q at position 253 and the serine-rich heptapeptide 326SWSASGS332 next to major

hydrophilic region B has been found associated with cell tropism and virulence of isolates

(Jackwood et al., 2008; Qi et al., 2009), identification of 253Q and serine-rich heptapeptide

indicates the virulence of studied pathotype. Further, identification of histidine (H253),

asparagine (N279), threonine (T284) and Arginine (R330) has been previously linked to

propagation in cell culture (Mundt, 1999). Consistent with this, we found substitutions in the

study vvIBDVs as glutamine at 253 (H→Q), aspartic acid at 279 (N→D), alanine at 284

(T→A) and serine at 330 (R→S) indicating they are not cell-culture adapted but represent

four different outbreaks. Interestingly, though it needs large sample dataset to confirm, we

found histidine substitution at position 221 (Q→H) unique to strains from Pakistan when

compared to representative strains (Figure of 3D, label accordingly). We compared the

6

sequences from strains isolated in the present study to three vvIBDVs: VV1, VV2 and VV3

(Figure 3). The strains showed a pattern similar to VV3 (OKYM strain), except for one

substitution at position 221 (Q→H); Q at this position was found to be common in all

representative strains. In addition, compared to VV1 and VV2, there was a mutation at

position 300 (A→E), while serine was replaced by glycine in contrast to VV2 at position 254

(S→G) (Figure 3).

Differences in the relatively well-known or conserved regions/position considered exclusive

to vvIBDVs have been observed. For example, the vvIBDVs reported from outbreaks in

Indonesia and Singapore reported a substitution mutation 222(A→S) indicating that alanine

at this position is not a unique characteristic of vvIBDVs (Parede et al., 2003; Jackwood and

Sommer-Wagner, 2007). The residue S299 is thought to be conserved among vvIBDVs;

however studies have reported asparagine (N) for vvIBDVs (Etarradossi et al., 1999) and

serine (S) for non- vvIBDVs strains (Jackwood and Sommer-Wagner, 2007). Further, even

with characteristic pattern of vvIBDVs (A222, I242, I256, I294 and S299), an IBDV outbreak

from Malaysia produced only 10% mortality in susceptible chickens (Hoque et al., 2001).

This is contrary to our findings where we have observed high morbidity and mortality with

serine at position 299. Taken together, since the mechanism of IBDV virulence remains

poorly understood, the genetic changes in HVR should be considered as an outcome of

evolution pertaining to host-virus-environment interaction rather than virulence markers per

se. (Hoque et al., 2001; Parede et al., 2003; Jackwood and Sommer-Wagner, 2007).

The study flocks had a history of vaccination with commonly used vaccines in Pakistan;

however information on the extent of serological monitoring/evaluation of the flocks was not

available. Despite the use of attenuated vaccines to protect chickens, outbreaks in vaccinated

flocks are not uncommon. The vvIBDVs have the ability to break through high levels of

maternal antibodies and, in comparison to classical variant of IBDVs, can cause disease even

7

in the presence of high neutralising antibodies produced by classical attenuated vaccine

strains (Eterradossi et al., 1998; Jenberie et al., 2014). Therefore, highly attenuated vaccines

that induce low level of neutralising antibodies may not provide adequate protection to

vvIBDVs (Jenberie et al., 2014). Furthermore, beside variations at various places in deduced

residue analysis of study strains to commonly used vaccine strains in Pakistan, substitution

mutations in key epitope in the VP2 capsid at positions 221 (Q→H), 222 (P/L/Q/S→A), 249

(H→Q), 284 (T→A), 299 (N→S) and 317 (N→S) were in hydrophilic regions. Four

hydrophilic loops are identified in VP2 and the occurrence of these residue exchanges

indicates potential selective pressure for evolution of IBDV upon exposure to the immune

system (Durairaj et al., 2011). It might be possible that the change in amino acids may change

the polarity of the protein, or lead to modifications in protein folding or the interaction with

other/adjacent molecules resulting in a change in the topography of the neutralizing epitopes

and subsequent vaccine failure (Jackwood and Sommer-Wagner, 2011). Vaccination with

cell-culture attenuated vvIBDV have been found efficacious, however, risk of reversion of

vaccine strains always exists (Raue et al., 2004; Rasool and Hussain, 2006). Since virulent

classical viruses that are genetically related to classical vaccine strains have been identified

from outbreaks when users fail to follow manufacturer’s recommendations (Jackwood et al.,

2008; Jenberie et al., 2014), ensuring appropriate and correct administration of dosage while

maintaining cold-chain coupled with routine serological monitoring should be practised

adequately when using cell-culture attenuated vvIBDVs as vaccines.

In conclusion, we sequenced the HVR of the VP2 gene of circulating strains of IBDVs in

Pakistan, and found that they had a residue pattern typical of very virulent viruses closely

related to Asia/Europe lineage. Interestingly, we found a histidine (H) at position 221 that is

unique to strains from Pakistan when compared to representative strains in our analysis, and a

larger dataset would be beneficial to confirm this observation.

8

Acknowledgements

We acknowledge Dr Andrew Broadbent at The Pirbright Institute for helpful comments and

suggestions, and for reading and editing the manuscript prior to submission.

Author’s contributions

Conceived and designed the experiment: MZS, MA and MM. Field work and sample

collection from Pakistan: MA, MA, ZR. Genome processing in the collaborating institute at

Sweden: MM and UNC. Analysed the data: MZS, MA and ZR: Wrote the manuscript: MZS,

MA and MM. Author’s current affiliation are given in the manuscript.

Conflict of Interest

All authors declared no conflict of interest with respect to manuscript contents.

Ethical approval

All applicable international, national, and/or institutional guidelines for the care and use of

animals were followed.

References:

Brown MD, Green P, Skinner MA (1994) VP2 sequence of recent European ‘very virulent’

isolates of infectious bursal disease virus are closely related to each other but are

distinct from those of ‘classical’ strains. J Gen Virol 75: 675 – 680

Durairaj V, Sellers HS, Linnemann EG, Icard AH, Mundt E (2011) Investigation of the

antigenic evolution of field isolates using reverse genetics system of infectious bursal

disease virus (IBDV). Arch Virol 156(10):1717 – 1728

Eterradossi N, Arnauld C, Tekaia F, Toquin D, Le Coq H, Rivallan G, Guittet M, Domenech

J, van den Berg TP, Skinner MA (1999) Antigenic and genetic relationships between

European very virulent of infectious bursal disease viruses and an early West African

isolate. Avian Pathol 28: 36-46

9

Eterradossi N, Arnauld C, Toquin D, Rivallan G (1998) Critical amino acid changes in VP2

variable domain are associated with typical and atypical antigenicity in very virulent

infectious bursal disease viruses. Arch Virol 143: 1627 – 1636

Hoque MM, Omar AR, Chong LK, Hair-Bejo M, Aini I (2001) Pathogenicity of SspI-

positive infectious bursal disease virus and molecular characterization of the VP2

hypervariable region. Avian Pathol 30: 369–380

Ismail NM, Saif YM, Moorhead PD (1988) Lack of pathogenicity of five serotype 2

infectious bursal disease viruses in chickens. Avian Dis 32: 757–759

Jackwood DJ, Sommer SE (1999) Restriction fragment length polymorphisms in the VP2

gene of infectious bursal disease viruses from outside the United States. Avian Dis 43:

310–314

Jackwood DJ, Sommer-Wagner SE (2011) Amino acids contributing to antigenic drift in the

infectious bursal disease Birnavirus (IBDV). Virol 409 (1): 33 – 37

Jackwood DJ, Sreedevi B, LeFever LJ, Sommer-Wagner SE (2008) Studies on naturally

occurring infectious bursal disease viruses suggest that a single amino acid

substitution at position 253 in VP2 increases pathogenicity. Virol 377: 110 – 116

Jackwood DJJ, Sommer-Wagner S (2007) Genetic characteristics of infectious bursal disease

viruses from four continents. Virol 365: 369–375

Jenberie S, Lynch SE, Kebede F, Christley RM, Gelaye E, Negussie H, Asmare K, Ayelet G

(2014) Genetic characterization of infectious bursal disease virus isolates in Ethiopia.

Acta Tropica 130: 39 – 43

Kataria RS, Tiwari AK, Butchaiah G, Kataria JM, Skinner MA (2001) Sequence analysis of

the VP2 gene hypervariable region of infectious bursal disease viruses from India.

Avian Pathol 30(5): 501-507

10

Khan KNM, Shah SA, Afzal M (1988) Observations on Gumboro disease (infectious bursal

disease) in Pakistan. Rev Sci Tech Off Int Epiz 7(3): 625 – 629

Lone NA, Rehmani SF, Kazmi SU, Muzaffar R, Khan TA, Khan A, Khan SA, Ahmed A

(2009) Molecular characterization of Pakistani field isolates of infectious bursal

disease virus. Avian Dis 53(2): 306-309

Martin AM, Fallacara F, Barbeiri I, Tosi G, Rivallan G, Eterradossi N, Ceruti R, Cordioli P

(2007) Genetic and antigenic characterization of infectious bursal disease viruses

isolated in Italy during the period 2002 – 2005. Avian Dis 51: 863 – 872

Mundt E (1999) Tissue culture infectivity of different strains of infectious bursal disease

virus is determined by distinct amino acids in VP2. J Gen Virol 80: 2067 – 2076

Mundt E, Muller H (1995) Complete nucleotide sequence 5΄- and 3΄- non-coding region of

both genome segments of different strains of infectious bursal disease virus. Virol

209: 10-18

Owoade AA, Mulders MN, Kohnen J, Ammerlaan W, Muller CP (2004) High sequence

diversity in infectious bursal disease virus serotype 1 in poultry and turkey suggests

West-African origin of very virulent strains. Arch Virol 149: 653 – 672

Parede L, Sapats SI, Gould G, Rudd MF, Lowther S, Ignjatovic J (2003) Characterization of

infectious bursal disease virus isolates from Indonesia indicates the existence of very

virulent strains with unique genetic changes. Avian Pathol. 32: 511–518

Qi X, Gao H, Gao Y, Qin L, Wang Y, Gao L, Wang X (2009) Naturally occurring mutations

at residues 253 and 284 in VP2 contributes to cell tropism and virulence of very

virulent infectious bursal disease. Antiviral Res 84: 225 – 233

Rasool MH, Hussain I (2006) Preparation and evaluation of Vero-cell infectious bursal

disease vaccine in Pakistan. Vacc 24:2810–2814

11

Raue R, Islam MR, Islam MN, Islam KM, Badhy SC, Das PM, Muller H (2004) Reversion of

molecularly engineered, partially attenuated, very virulent infectious bursal disease

virus during infection of commercial chickens. Avian Pathol 33: 181–189

Tamura K, Peterson D, Peterson N, Stecher G, Nei M, Kumar S (2011) MEGA5: Molecular

Evolutionary Genetics Analysis using Maximum Likelihood, Evolutionary Distance,

and Maximum Parsimony Methods. Mol Biol Evol 28: 2731-2739

Zahoor MA, Abubakar M, naim S, Khan QM, Arshad MJ (2011) Molecular typing of field

isolates from two outbreaks of infectious bursal disease virus from Pakistan. Vet

World 4(7): 297 – 300

Legends:

Figure 1: Phylogenetic consensus tree for the study IBDVs representing four different

outbreaks in Pakistan. The hypervariable nucleotide sequences (VP2 gene) of the study

strains were compared with previously classified strains as very virulent (VV) subdivided in

VV-1, VV-2 and VV-3, classic variant (CV), very virulent atypical (VVA), Australian classic

(AC) and Australian variant (AV).

Figure 2: Percent similarity and divergence of nucleotide sequences of hypervariable regions

of study isolates with reference to representative strains of different pathotypes of IBDVs

including the vaccine strains

Figure 3: Alignment of deduced amino acid sequence of the hypervariable domain of VP2

spanning between amino acid positions 211 – 354. The residue profile of study viruses is

12

compared with strains of IBDV representing different pathotypes. The major and minor

hydrophilic regions are boxed for understanding.

Figure: X-ray crystal structures of the VP2 protein downloaded from the Protein Data Bank (PDB) ID 1WCD. Annotation and visualization were made in the MacPyMol v1.7.4.4. Residues 221, 222 and 317 are labelled green, blue and yellow, respectively.

13