Contents lists available at ScienceDirect

Infection, Genetics and Evolution

journal homepage: www.elsevier.com/locate/meegid

Research paper

Complete genome analysis and time scale evolution of very virulentinfectious bursal disease viruses isolated from recent outbreaks in Morocco

Charifa Drissi Touzania, Siham Fellahia, Ouafaa Fassi Fihria, Fatima Gabounb, Slimane Khayib,Rachid Mentagb, Chiara Licoc, Selene Baschieric, Mohammed El Houadfia, Mariette Ducatezd,⁎

aUnité de Pathologie Aviaire, Département de Pathologie et Santé Publique Vétérinaires, IAV Hassan II, BP 6202. Rabat- Instituts, 10000 Rabat, MoroccobUnité de Biotechnologie, CRRA-Rabat, Institut National de la Recherche Agronomique INRA, Avenue Mohamed Belarbi Alaoui B.P 6356, Rabat- Instituts, 10101 Rabat,Moroccoc Laboratory of Biotechnology, Agenzia Nazionale per le Nuove tecnologie, l'Energia e lo Sviluppo economico sostenibile (ENEA), C.R. Casaccia, Via Anguillarese 301,00123 Rome, ItalydUniversité de Toulouse, INRA, ENVT, IHAP, F- 31076 Toulouse, France

A R T I C L E I N F O

Keywords:Very virulent infectious bursal disease virusFull-length genome sequencingPhylogenetic treeEvolution ratePoultry flocksMorocco

A B S T R A C T

Emerging of very virulent infectious bursal disease virus (vvIBDV) genotype in poultry flocks in Morocco werecharacterized. VP2 sequence analysis showed that the strains of Moroccan vvIBDV genotypes clustered sepa-rately from classic and vaccine strains reference of IBDV. The full-length genome of four Moroccan vvIBDVstrains was determined, in order to get a more exhaustive molecular characterization allowing to conduct theevolution time scale and speculations on their origin. In a phylogenetic tree, nucleotide sequences of segment Aand B formed a common branch with those vvIBDV references strains published in GenBank, but they clearlygrouped into a distinct subcluster. An alignment of deduced amino acid sequences segment B, confirmed thepresence of the conserved TDN tripeptide found in all of the vvIBDV genotype and revealed the presence of 2substitutions I472L and E688D specific for the vvIBDV Moroccan isolates. The deduced amino acid sequences ofsegment A genes showed the presence of the “signature” typical of the vvIBDV genotype and revealed thepresence of 7 aa substitutions specific for the vvIBDV Moroccan strains. The evolution rate for IBDV VP2 genewas estimated at 5.875× 10−4 substitutions/site/year. The estimation of the time to most common recentancestor of Moroccan vvIBDV based on the VP2 sequences available was 31 years, corresponding to 3 yearsearlier than the first vvIBDV case detection in layers in the country.

1. Introduction

Infectious bursal disease (IBD) named also Gumboro disease is animmunosuppressive disease causing a lot of economic loses affectingthe industrial production of poultry worldwide. It was first described byCosgrove in 1960 in Gumboro village in Delware State (USA), as a viraldisease of young chicks of 3 to 6 weeks of the age (Cosgrove, 1962). Theetiological agent is infectious bursal disease virus (IBDV), a member ofthe Birnaviridae family, Avibirnavirus genus (Ozel and Gelderblom,1985). The virus replicates actively in the Bursa of Fabricious and de-stroys B-lymphocytes, which are in different maturity stages (Lasherand Shane, 1994; Berg, 2000). This destruction of B cells makeschickens immunosuppressed and vulnerable to opportunistic viral orbacterial pathogens (Sharma, 2000; Rautenschlein et al., 2002).

IBDV is a non-enveloped virus, containing double-stranded RNAdivided in two segments (A and B) (Dobos et al., 1979; Ozel andGelderblom, 1985). Segment A (3.3 kb) includes two open readingframes, ORF 1 and ORF2 (Mundt et al., 1995). ORF 1 encodes for themajority of viral proteins (VP) (VP2, VP3, VP4) while ORF2 encodes forVP5, a non-structural protein. VP2 is the most important protein ex-tensively studied in epidemiological and phylogenetics analyses as it isresponsible for the antigenic variation due to its hypervariable region(HVR) (206-350 amino acids, aa) (Bayliss et al., 1990; Heine et al.,1991; Jackwood and Sommer, 1999). Most neutralizing antibodiesconferring protective immunity are indeed directed against VP2 (Letzelet al., 2007). VP3, VP4, and VP5 (internal capsid protein, protease, andnon-structural protein, respectively) are also encoded by segment A(Caston et al., 2001; Ture and Saif, 1992; Mundt et al., 1995). Segment

https://doi.org/10.1016/j.meegid.2019.104097Received 27 August 2019; Received in revised form 2 October 2019; Accepted 27 October 2019

⁎ Corresponding author at: Mariette DUCATEZ, 23 Chemin des Capelles, 31076 Toulouse, France.E-mail addresses: charifadrissi@gmail.com (C. Drissi Touzani), fellahisiham2015@gmail.com (S. Fellahi), fassifihri.ouafaa@gmail.com (O. Fassi Fihri),

gabounf@gmail.com (F. Gaboun), slimane.khayi@gmail.com (S. Khayi), rachidmentag@yahoo.ca (R. Mentag), chiara.lico@enea.it (C. Lico),selene.baschieri@enea.it (S. Baschieri), elhouadfimohammed@yahoo.fr (M. El Houadfi), m.ducatez@envt.fr (M. Ducatez).

Infection, Genetics and Evolution 77 (2020) 104097

Available online 31 October 20191567-1348/ © 2019 Elsevier B.V. All rights reserved.

T

B (2.8 kb) encodes for VP1, an RNA dependent-RNA polymerase. It hasbeen suggested that VP1 also contributes to the virulence of IBDV(Escaffre et al., 2013). Two distinct IBDV serotypes exist: serotypes Iand II. Serotype I is pathogenic for chickens (Sharma et al., 1989), whileserotype II recovered from turkeys is not pathogenic for chickens(Ismail et al., 1988).

IBD viruses have recently been classified in 7 genogroups based onthe molecular structure of the HVR of VP2 and include the most im-portant strains: classical (cl), very virulent (vv), antigenic variant (av)as well as other strains circulating worldwide (Michel and Jackwood,2017).

The genetic and antigenic characteristics of IBD viruses are verydifferent. Both segment A and B play an important role in the patho-genicity of IBDV. The full genome characterization of IBDV remainstherefore essential to classify IBDV strains (Escaffre et al., 2013). Re-cently, Jackwood and co-workers (2018) have proposed a new classi-fication to eliminate this confusion between the genotype and the pa-thotype of IBDV and presented a new model to describe IBDV strains(Jackwood et al., 2018). Hence, to conduct molecular epidemiologystudies, it is advised to include sequence analysis for both genomicsegments. This molecular analysis is an important step to explain thelinks between emergence, spread and maintenance of vvIBDV strainsaround the world (Alfonso-Morales et al., 2015).

In Morocco, the first detection of IBDV in poultry flocks was in1978. In 1991, vvIBDV was first detected in layers (El Houadfi, personalcommunication). A recent study, focuses on the molecular character-ization of HVR VP2 protein from Moroccan IBDV isolates, revealed thatvvIBDV strains circulate in many parts of the country are clearly dis-tinct from classical and vaccine strains of IBDV (Drissi Touzani et al.,2019).

The objective of the present study is thus to thoroughly characterize4 Moroccan vvIBDV strains at the molecular level by full genome se-quencing. This will allow us to gain knowledge on Moroccan vvIBDVsand to conduct the time scale evolution analysis in order to better un-derstand the virus evolution in time.

2. Materials and methods

2.1. Case history

Four poultry flocks experiencing field outbreaks of IBD from dif-ferent regions of Morocco were investigated. Increased mortalities werereported and birds had ruffled feathers, whitish diarrhoea, and re-markable lesions in autopsy including oedematous or haemorrhagicBursa of Fabricious (BF). Samples of Bursa were screened by real timeRT-PCR for the presence of vvIBDV. The history of flocks used in thisstudy is presented in Table 1. The processing of the samples was per-formed as described by Drissi Touzani et al. (2019). In brief, the Bursasamples were chopped and minced using a mortar and pestle. Thehomogenates were placed in sterile tubes, centrifuged and the super-natants were used for viral extraction.

2.2. Viral extraction and full genome amplification

Viral RNA was extracted from the Bursa supernatant using theMacherey-Nagel™ NucleoSpin™ RNA Virus kit (Macherey-Nagel, KG,Düren, Allemagne) following the manufacturer's instructions. The onestep RT-PCR was performed using the SensiFAST Probe No-ROX One-Step Kit (Bioline) following the manufacturer instructions and using aPeqlab thermocycler. In brief, the reaction was run in a 25 μl final vo-lume containing 12.5 μl of 2× RT-PCR buffer mix, 5.85 μl of nuclease-free water, 0.25 μl of reverse transcriptase (200 U), 0.4 μl of RI (RNaseinhibitor), 0.5 μl of each primer (10 μM) and 5 μl of double stranded(ds) RNA template (denatured previously at 95 °C for 3min). The pro-gram was as follows: 20min incubation at 45 °C; 5min denaturation at95 °C; 40 amplification cycles with 1min at 95 °C, 1min at 53 to 60 °C(primer-specific parameters), and 2min at 72 °C; and a final elongationstep 10min at 72 °C. To amplify the IBDV full genome, segment A wascut into 6 fragments and segment B into 3 fragments. The primers usedin this study are described in Table 2. The primers were designed on thebasis of a Dutch IBDV strain (accession number AF240687), first blasthit of our first partial sequence data. PCR products of the expectedlength were purified with the QIAquick Gel Extraction kit (QiagenGmbH, Hilden, Germany) according to the manufacturer's instructions.

2.3. Sequencing and phylogenetic analyses

The Sanger sequencing for the purified PCR products was performedin both directions for segment A and B amplicons using the PCR primers

Table 1History of samples taken from flocks with signs of IBDV infection used in this study.

Samples Birds Clinical signs rtRT-PCR (Ct values) Accession no.

Name Date Age Type vvIBDV Non-vvIBDV Segment A Segment B

IB19 12/01/2017 30 d Broiler Witish diarrhoea, ruffled feather 15.53 N MK580160 MK580164IB61 15/01/2017 32 d Broiler Ruffled feather 22.37 N MK580162 MK580166IB80 30/05/2018 38 d Broiler Prostration, decrease in feed consumption 16.46 N MK580161 MK580165IB94 31/05/2018 34 d Broiler Depression 21.58 N MK580163 MK580167

d: day; N: negative.

Table 2Primers used to amplify and sequence Moroccan IBDV segments A and B.

Primers Positions Nucleotides Amplicon size (bp)

Segment A1AFa 1 CTATAGGGGATACGATCGGTCTGA 7371ARb 738 GCCTGTCACTGCTGTCACAT2AFd 737 GCC CAG AGT CTA CAC CAT 7432ARd 1479 CCC GGA TTA TGT CTT TGA3AFb 1254 GAACCTGGTCACAGAATACG 5323ARb 1786 ATAGCGTGGCACCCTCTCT4AFc 1700 CCGTAGTCGACGGGATTCT 4794ARc 2197 CAGTGGCGAGCTTGGTGC5AFb 2140 CYCAAYGCYTRTGGCGAGATT 4815ARb 2621 GCTGTCCCGTACTTGGCTCTT6AFc 2564 CTCGCAAACGCACCACAAGC 6866ARa 3250 CCACGCGTGGGGACCCGCGAACGGAT

Segment B1BFc 1 GGATACGATGGGTCTGACCCT 8851BRc 886 CATAGGTAGTCCACTTGATGAC2BFc 751 CACCCGGTGAGGATGACAAGC 11092BRc 1860 GATCCCGAGATCTTTGCTGTA3BFe 1705 AGGTCCATTGATGACATTAGG 10783BRe 2783 AGTGTCCTCTTCTTGAGTGG

a Wu et al., 2009.b Lojkić et al., 2008.cFelice et al., 2017.

dJackwood and Sommer-Wagner, 2007.

ePrimers manually designed.

C. Drissi Touzani, et al. Infection, Genetics and Evolution 77 (2020) 104097

2

and using a 16-capillary 3130XL genetic analyser sequencer (AppliedBiosystems) at the Plateau de Génomique GeT-Purpan, UDEAR UMR5165 CNRS/UPS, CHU PURPAN, Toulouse, France.

The complete sequences data of the four IBDV strains were manu-ally assembled using BioEdit software package version 7.2.5 (Hall,1999) and the alignments of the sequences were performed usingClustalW included in BioEdit. The sequences comparisons were carriedout using the open source BLAST program (National Center for Bio-technology Information, Bethesda MD, http://blast.ncbi.nlm.nih.gov/Blast.cgi). MEGA7 (molecular evolutionary genetics analysis version 7)(Kumar et al., 2016) was used for the phylogenetic analysis and treeconstruction using the maximum likelihood method, and 1000 boot-straps (bootstraps values above 50 were labelled on major tree branchesto assign confidence levels to branches). Evolutionary distances be-tween strains were calculated using the Neighbour-Joining distancesimplemented in Mega. The sequences of Moroccan IBDV strains werealigned and compared with representatives of all the molecular types ofIBDV strains retrieved from GenBank. The complete nucleotide se-quences of the Moroccan IBDV strains for both segments A and B weresubmitted to GenBank under accession numbers MK580160 toMK580163 for segment A and MK580164 to MK580167 for segment B.

2.4. Evolutionary and substitution rates

To estimate the time to the Most Recent Common Ancestors(tMRCA) and rates of nucleotide substitution per site and per year ofeach segment (A and B) and of the HVR-VP2, the BEAST package wasused. Only sequences with a known year of collection were included.Datasets of 159 sequences of segment A and 103 sequences of segmentB sequences were selected and used to generate the BEAST input filesusing BEAUti within the BEAST package v1.8.1 (Drummond andRambaut, 2007). We also analysed 557 sequences of HVR-VP2 fromseveral countries, including HVR-VP2 sequences recently obtained inMorocco (Drissi Touzani et al., 2019). The analysis was there afterperformed on 72 sequences of the HVR-VP2 from Algeria, Tunisia,Egypt, Spain, France, Germany, India, Brazil, Japan, Malaysia andMorocco to focus on IBDV close to our Moroccan sequences. The ob-jective was to determine the common ancestor of Moroccan IBDVstrains and its origin. The best nucleotide substitution model was esti-mated using MEGA 7 and the Hasegawa-Kishino-Yano (HKY) modelwas chosen. The Gamma-distributed rate including partition positions1+ 2, 3 and an initial value of 10−3 substitutions/nucleotide/year wasselected. The Markov chain Monte Carlo (MCMC) available in theBEAST package was run for each segment with 1 billion generationsand sampling every 1000 generations. Molecular clock branch rateswere derived from a log-normal distribution. The maximum cladecredibility (MCC) tree was generated using TreeAnnotator softwarewith 10% burn-in and 95% highest posterior density (HPD) intervals.The results were visualised with Figtree program version 1.4.2 andTracer version 1.6.0 included in the BEAST package.

3. Results

3.1. Case history and IBDV detection

The present study was focused on 4 Moroccan poultry flocksshowing clinical signs of Gumboro disease and pathological lesions suchas oedematous and haemorrhagic Bursa of Fabricius with mortalityrates higher than 0.5% per day during the onset of disease (“normal”mortality rate: 0.1% per day). The Bursa of Fabricius were then testedby real-time RT-PCR to quantify their viral loads. Cycle threshold (Ct)values of the bursa samples analysed ranged from 14.25 to 28.29(Table 1). The clinical and pathological features and Ct values con-firmed the presence of vvIBDV strains with high viral loads in Bursasamples (Table 1).

3.2. Sequences analyses

The amplification of A and B segments of the four IBDV strains wassuccessfully performed. The size of the IBDV full genomes obtained was3267 nucleotides (nt) for segment A and 2711 nt for segment B.

3.2.1. Segment ANucleotide sequence analysis of segment A showed that the

Moroccan viruses are highly homologous to vvIBDVs isolated fromdifferent countries: 98% nucleotide identity was found betweenMoroccan, Algerian and Tunisian strains. However, nucleotide identitywith the classical and attenuated strains was lower, reaching 94 to 95%.

The sequence of VP5 of the four Moroccan IBDV isolates such asthose of vvIBDV encode for 149 aa as described by Kong et al. (2004)unlike VP5 of classical strains that encode only for 145 aa. vvIBDVstrains (including Moroccan isolates), contain MLSL four aa at the startof the protein, compared to non-vvIBDV strains. In addition, VP5 ofMoroccan IBDV shared two residues with vvIBDV strains: 78F and137W. The 137W residue was previously identified by Boot et al.(2001) as being exclusive to vvIBDV. Two additional substitutions (49Rand 129P) unique for vvIBDV strains, were also found in MoroccanIBDV isolates.

In addition, segment A of Moroccan IBDV strains had 7 aa sub-stitutions (G254S, E300Q, V608I, N777S, T799 N, S991 T and T993A)specific to Moroccan IBDV isolates (Table 3). VP2 of Moroccan IBDVstrains shared five determinants of vvIBDV strains (222A, 242I, 256I,294I and 299S) (Table 3). The heptapeptide “SWSASGS” in positions326 to 332 was also found in Moroccan IBDV strains. Moroccan VP4harboured 652S and 692 K, common to all IBDV strains. They alsoharboured VP4 residues common to the majority of the vvIBDV strainspublished in Genbank: 680Y, 685 N, 715S, 541 N and 751D as describedby Kong et al., 2004. In VP3, we observed five aa specific of vvIBDVstrains (783Q, 919E, 981P, 990 V and 1005A).

3.2.2. Segment BMoroccan Segment B showed 97% nucleotide identity with re-

presentative vvIBDV strains and 89% with classical and attenuatedstrains. Triplet amino acids TDN (145–147) present in vvIBDV strainswas found in Moroccan IBDV strains (Table 3). Regarding amino acidposition 242, the Moroccan IBDV strains showed the signature 242Epresent in vvIBDV strains. Aa residues 390M, 393D, 511S, 562P and687P, common to vvIBDV (Kong et al., 2004), were also found inMoroccan IBDV strains (Table 3). Additionally, our sequences shared 4Iand 695 K with classical IBDV strains and had 13 K, 413Y, 561Q and756 K, which were detected in classical, variant and vvIBDV strains(Table 3) (Kong et al., 2004). Finally, two aa substitutions specific forMoroccan IBDV isolates were detected in segment B: I472L and E688D(Table 3).

3.3. Phylogenetic analysis

All Moroccan IBDV segment A sequences grouped with vvIBDVstrains and were clearly distinct from the classical, attenuated andvariant strains (Fig. 1). The phylogenetic analysis also showed thatsegment A of Moroccan IBDV viruses were closely related to each otherand very close to Algerian and Tunisian vvIBDV strains (Fig. 1). Simi-larly, the topology of the phylogeny tree showed that all the Moroccanfield IBD viruses fell into a unique sub-cluster within the representativevvIBDV strains cluster for segment B and very close to French, Algerianand Malaysian vvIBDV strains (Fig. 2).

3.4. Evolutionary rates and tMRCA

The estimated mean substitution rate of HVR VP2 of the populationsassessed of IBDV was 5.875×10−4 substitutions/site/year. The esti-mated mean substitution rates were 2.31× 10−4 substitutions/site/

C. Drissi Touzani, et al. Infection, Genetics and Evolution 77 (2020) 104097

3

Table 3Comparison of amino acid substitutions at different positions on polyprotein and VP1 protein between Moroccan IBDV strains and other published strains.

Amino acid substitutions

VP5 VP2 VP4 VP3

Strains Phenotype 49 78 129 137 222 242 253 254 256 294 299 300 330 608 680 685 715 751 777 783 799 918 981 990 991 993 1005

D6948 VV R F P W A I Q G I I S E S V Y N S D N Q T E P V S T AIB19 VV . . . . . . . S . . . Q . I . . . . S . N . . . T A .IB61 VV . . . . . . . S . . . Q . I . . . . S . N . . . T A .IB80 VV . . . . . . . S . . . Q . I . . . . S . N . . . T A .IB94 VV . . . . . . . S . . . Q . I . . . . S . N . . . T A .Cu-1wt Cl G I S R P V H . V L N . R . C K P H . . . D . A . . TCu-1 At G I S R P . . . V L N . . . C K P H . . . . L A . . T

Amino acid substitutions

VP1

Strains phenotype 4 13 61 145 146 147 242 287 390 393 413 472 508 511 561 562 646 687 688 695 756

D6948 VV V K I T D N E A M D Y I K S Q P S P E R KIB19 VV I . . . . . . . . . . L . . . . . . D K .IB61 VV I . . . . . . . . . . L . . . . . . D K .IB80 VV I . . . . . . . . . . L . . . . . . D K .IB94 VV I . . . . . . . . . . L . . . . . . D K .Cu-1wt Cl I . V N E G D T L E . . R R . S G S . K .Cu-1 At I T V N E G D T L E . . R R . S G S . K .

VV: very virulent IBDV strain; At: attenuated IBDV strain; Cl: classical IBDV strain.

Fig. 1. Phylogenetic tree analysing Moroccan IBDV nucleotides sequences of segment A with available IBDV references strains. (Moroccan IBDV isolates are markedin bold and red font size). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

C. Drissi Touzani, et al. Infection, Genetics and Evolution 77 (2020) 104097

4

year and 1.08×10−4 substitutions/site/year for segment A and B,respectively (Table 4). Thus, the substitution rate of segment B wasapproximately two times lower than the substitution rate of segment A.The estimation of the mean tMRCA for HVR-VP2 region of MoroccanIBDV strains suggested that the introduction of vvIBDV strain in Mor-occo was in.

1988 (95% HPD: 1978–2000) and its ancestor may have hadAlgerian origins. The mean tMRCA of the entire sequence of segment Aof Moroccan strains was dated around 1957 (95% HPD: 1937–1973).On the other hand, the mean tMRCA of segment B of Moroccan strainswas around 1910 (95% HPD: 1781–1964) (Table 4).

4. Discussion

Very virulent Infectious bursal disease virus is the cause of con-siderable economic losses in the world and the virus remains a major

problem in poultry flocks and consequently a major threat to thepoultry industry. Most studies based on segment A use the HVR-VP2 asa phylogenetic signature of vv or non-vvIBD viruses. However, mole-cular characterization of HVR-VP2 is not sufficient to determine thegenotype of IBDV strains, especially as both segments play an essentialrole in vvIBDV's pathogenesis (Escaffre et al., 2013). Therefore, se-quencing of the full IBDV genome is important to obtain complete in-formation on IBDV genome and to elucidate putative causes of itsemergence. In this study, we sequenced for the first time full genomesof Moroccan IBDV isolates and a time scale evolution of MoroccanvvIBDV was conducted.

In segment A, the presence of the heptapeptide 326 “SWSASGS” 332in Moroccan IBDV strains associated with 253Q was shown to increasethe virulence of the isolates (Jackwood et al., 2008; Qi et al., 2009). InVP3, an Alanine at position 990 is found in non-vvIBDV strains and ithas been proved that this mutation would lead to a reduced viral

Fig. 2. Phylogenetic tree analysing Moroccan IBDV nucleotides sequences of segment B with available IBDV references strains (Moroccan IBDV isolates are marked inbold and red font size). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Table 4Substitution rates values and time to the most recent common ancestor (tMRCA) for Moroccan IBDV strains.

Part of the genome Substitution rates (95% HPD)a tMRCA of Moroccan vvIBDV (95% HPD)

HVR-VP2 5.87× 10−4 (3.73× 10−4–8.30× 10−4) 1988 (1978–2000)Segment A 2.31× 10−4 (1.61× 10−4–3.04× 10−4) 1957 (1937–1973)Segment B 1.08× 10−4 (3.60× 10−5–1.74× 10−4) 1910 (1781–1964)

asubstitutions/site/year.

C. Drissi Touzani, et al. Infection, Genetics and Evolution 77 (2020) 104097

5

replication (Wang et al., 2010). The Moroccan IBDV strain shared 990 Vpositions with vvIBDV strains (Wang et al., 2007). Wang et al. alsodemonstrated that any change in aa positions 783, 918, 981, 990 and1005 in VP3 after attenuation process in SPF embryos or chicken em-bryo fibroblast cultures decreased the virulence of the strains (Wanget al., 2007). Moroccan VP4 harboured 652S and 692 K, common to allIBDV strains and have been found to have very important role forprotease activity of IBDV and were critical for this function (Boot et al.,2001; Lejal et al., 2000). VP5 nonstructural protein plays a significantrole in the release of viral progeny from infected cells (Lombardo et al.,2000; Wu et al., 2009) and also in the induction of apoptosis in vitro(Yao and Vakharia, 2001). VP5 of Moroccan IBDV isolates contains themarkers specific of vvIBDV strains. The same positions (78F and 137W)were found in vvIBDV strains isolates from different countries such asZambia (Kasanga et al., 2013), Croatia (Lojkić et al., 2008), Malaysia(Kong et al., 2004) and China (Wang et al., 2007).

In VP1, Moroccan IBDV strains harbour the triplet TDN at positions145/146/147, which is a unique motif of vvIBDV allowing for distin-guishing IBDV pathotypes (Gao et al., 2014). On the other hand, thepresence of this pattern triplet TDN in Moroccan IBDV strains asso-ciated with E residue at position 242 have been shown to increase thevirulence of IBDV (Kasanga et al., 2013; Gao et al., 2014). However, thesegment B containing VP1 is responsible for the virus polymerase ac-tivity, and mutations on segment B could therefore alter virus replica-tion (Escaffre et al., 2013).

The presence of all these markers of virulence in both segments Aand B, explain that in the phylogenetic tree of the two segments,Moroccan IBDV isolates were in the same clusters as vvIBDV strainsincluded in this study (Figs. 1 and 2). In fact, the segment A of Mor-occan IBDV isolates was phylogenetically related to Algerian and Tu-nisian vvIBDV isolates, which can easily be explained by the geo-graphical proximity and transboundary between Morocco and Algeriaand between Algeria and Tunisia. However, in the phylogeny of seg-ment B (Fig. 2), the French and Malaysian isolates are the closestviruses to Moroccan IBDV. For the French IBDV strains, this can beexplained by trade exchanges between the countries. Moreover, Franceexports day old chicks of breeder and layers to Morocco (exports areunder the control of National Office of Food Safety, http://www.onssa.gov.ma/). The relation to Malaysian IBDV is still unclear consideringthe very large geographic distance between the countries and the ab-sence of information on trade between Morocco and Malaysia. A gap insurveillance and sequencing may be a bias here is the understanding ofIBDV spread and origins as similar strains may circulate from South-East Asia to the Middle East without having been fully characterized.

The tMRCA of Moroccan HVR-VP2 was consistently estimated to bearound the 1988 (Table 4) and the substitution rates for the HVR-VP2dataset was 5.875×10−4 nucleotide substitutions per site per year,which is similar to the substitution rates of IBDV VP2 reported pre-viously (4,4× 10−4 nucleotide substitutions per site per year) and tentimes lower than the substitution rates of the capsid or envelope pro-teins of other RNA viruses (Jenkins et al., 2002). The estimation oftMRCA of segment A of vvIBDV dated about 31 years, corresponding to3 years earlier than the first vvIBDV case detection in layers in thecountry (1991, EL Houadfi, personal communication). Hon and co-workers reported that the estimated mean substitution rate of full VP2of vvIBDV strains was 0.67×10−3 nucleotide substitutions per site peryear and the tMRCA of these viruses was around 1960. They suggestedthat segment A of vvIBDV emerged at least 20 years before its expan-sion, while the first documented case of the emergence of vvIBDV wasin the late of 1980 (Hon et al., 2006). However, another study byAlfonso-Morales et al., 2015, looked at segment B of Cuban IBDV strainsand revealed that the substitutions rate for both lineages vvIBDV andnon-vvIBDV were 4.80×10−3 and 1.88× 10−4, respectively anddated the tMRCA for vvIBDV and non-vvIBDV in 1981 and 1917, re-spectively. Taken together these findings match the numbers obtainedhere.

In the future, an experimental study of the pathogenicity of ourMoroccan vvIBDV is necessary in order to determine the relationshipbetween the genetic and pathotype profile of these viruses.

Declaration of Competing Interest

All authors declare no conflict of interest.

Acknowledgments

This work was funded by AVIAMED project through the ARIMNet22015 Call by the following funding agencies: Ministry of HigherEducation, Scientific Research and Professional Training of Morocco(MESRSFC) and Italian Ministry of Agricultural, Food and ForestryPolicies (MIPAAF). ARIMNet2 (ERA-NET) has received funding fromthe European Union's Seventh Framework Programme for research,technological development and demonstration under grant agreementno. 618127.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.meegid.2019.104097.

References

Alfonso-Morales, A., Rios, L., Martínez-Pérez, O., Dolz, R., Valle, R., Perera, C.L., Bertran,K., Frías, M.T., Ganges, L., Díaz de Arce, H., Majó, N., Núñez, J.I., Pérez, L.J., 2015.Evaluation of a phylogenetic marker based on genomic segment B of infectious Bursaldisease virus: facilitating a feasible incorporation of this segment to the molecularepidemiology studies for this viral agent. PLoS One 10, e0125853. https://doi.org/10.1371/journal.pone.0125853.

Bayliss, C.D., Spies, U., Shaw, K., Peters, R.W., Papageorgiou, A., Muller, H., Boursnell,M.E.G., 1990. A comparison of the sequences of segment a of four infectious bursaldisease virus strains and identification of a variable region in VP2. J. Gen. Virol. 71,1303–1312. https://doi.org/10.1099/0022-1317-71-6-1303.

Berg, T.P.V.D., 2000. Acute infectious bursal disease in poultry: a review. Avian Pathol.29, 175–194. https://doi.org/10.1080/03079450050045431.

Boot, H.J., ter Huurne, A.A.H.M., Vastenhouw, S.A., Kant, A., Peeters, B.P.H., Gielkens,A.L.J., 2001. Rescue of infectious bursal disease virus from mosaic full-length clonescomposed of serotype I and II cDNA. Arch. Virol. 146, 1991–2007. https://doi.org/10.1007/s007050170047.

Caston, J.R., Martinez-Torrecuadrada, J.L., Maraver, A., Lombardo, E., Rodriguez, J.F.,Casal, J.I., Carrascosa, J.L., 2001. C terminus of infectious Bursal disease virus majorcapsid protein VP2 is involved in definition of the T number for capsid assembly. J.Virol. 75, 10815–10828. https://doi.org/10.1128/JVI.75.22.10815-10828.2001.

Cosgrove, A.S., 1962. An apparently new disease of chickens: avian Nephrosis. Avian Dis.6, 385. https://doi.org/10.2307/1587909.

Dobos, P., Hill, B.J., Hallett, R., Kells, D.T., Becht, H., Teninges, D., 1979. Biophysical andbiochemical characterization of five animal viruses with bisegmented double-stranded RNA genomes. J. Virol. 32, 593–605.

Drissi Touzani, C., Fellahi, S., Gaboun, F., Fassi Fihri, O., Baschieri, S., Mentag, R., ElHouadfi, M., 2019. Molecular characterization and phylogenetic analysis of veryvirulent infectious bursal disease virus circulating in Morocco during 2016-2017.Arch. Virol. 164, 381–390. https://doi.org/10.1007/s00705-018-4076-3.

Drummond, A.J., Rambaut, A., 2007. BEAST: Bayesian evolutionary analysis by samplingtrees. BMC Evol. Biol. 7, 214. https://doi.org/10.1186/1471-2148-7-214.

Escaffre, O., Le Nouen, C., Amelot, M., Ambroggio, X., Ogden, K.M., Guionie, O., Toquin,D., Muller, H., Islam, M.R., Eterradossi, N., 2013. Both genome segments contributeto the pathogenicity of very virulent infectious Bursal disease virus. J. Virol. 87,2767–2780. https://doi.org/10.1128/JVI.02360-12.

Felice, V., Franzo, G., Catelli, E., Di Francesco, A., Bonci, M., Cecchinato, M., Mescolini,G., Giovanardi, D., Pesente, P., Lupini, C., 2017. Genome sequence analysis of adistinctive Italian infectious bursal disease virus. Poult. Sci. 96, 4370–4377. https://doi.org/10.3382/ps/pex278.

Gao, L., Li, K., Qi, X., Gao, H., Gao, Y., Qin, L., Wang, Y., Shen, N., Kong, X., Wang, X.,2014. Triplet amino acids located at positions 145/146/147 of the RNA polymeraseof very virulent infectious bursal disease virus contribute to viral virulence. J. Gen.Virol. 95, 888–897. https://doi.org/10.1099/vir.0.060194-0.

Hall, T., 1999. BioEdit: A User-Friendly Biological Sequence Alignment Editor andAnalysis Program for Windows 95/98/NT.

Heine, H.-G., Haritou, M., Failla, P., Fahey, K., Azad, A., 1991. Sequence analysis andexpression of the host-protective Immunogen VP2 of a variant strain of infectiousBursal disease virus which can circumvent vaccination with standard type I strains. J.Gen. Virol. 72, 1835–1843. https://doi.org/10.1099/0022-1317-72-8-1835.

Hon, C.-C., Lam, T.-Y., Drummond, A., Rambaut, A., Lee, Y.-F., Yip, C.-W., Zeng, F., Lam,P.-Y., Ng, P.T.W., Leung, F.C.C., 2006. Phylogenetic analysis reveals a correlationbetween the expansion of very virulent infectious Bursal disease virus and

C. Drissi Touzani, et al. Infection, Genetics and Evolution 77 (2020) 104097

6

reassortment of its genome segment B. J. Virol. 80, 8503–8509. https://doi.org/10.1128/JVI.00585-06.

Ismail, N.M., Saif, Y.M., Moorhead, P.D., 1988. Lack of pathogenicity of five serotype 2infectious Bursal disease viruses in chickens. Avian Dis. 32, 757. https://doi.org/10.2307/1590995.

Jackwood, D.J., Sommer, S.E., 1999. Restriction fragment length polymorphisms in theVP2 gene of infectious Bursal disease viruses from outside the United States. AvianDis. 43, 310. https://doi.org/10.2307/1592622.

Jackwood, D.J., Sommer-Wagner, S., 2007. Genetic characteristics of infectious bursaldisease viruses from four continents. Virology 365, 369–375. https://doi.org/10.1016/j.virol.2007.03.046.

Jackwood, D.J., Sreedevi, B., LeFever, L.J., Sommer-Wagner, S.E., 2008. Studies onnaturally occurring infectious bursal disease viruses suggest that a single amino acidsubstitution at position 253 in VP2 increases pathogenicity. Virology 377, 110–116.https://doi.org/10.1016/j.virol.2008.04.018.

Jackwood, D.J., Schat, K.A., Michel, L.O., de Wit, S., 2018. A proposed nomenclature forinfectious bursal disease virus isolates. Avian Pathol. 47, 576–584. https://doi.org/10.1080/03079457.2018.1506092.

Jenkins, G.M., Rambaut, A., Pybus, O.G., Holmes, E.C., 2002. Rates of molecular evolu-tion in RNA viruses: a quantitative phylogenetic analysis. J. Mol. Evol. 54, 156–165.https://doi.org/10.1007/s00239-001-0064-3.

Kasanga, C.J., Yamaguchi, T., Munang’andu, H.M., Ohya, K., Fukushi, H., 2013. Genomicsequence of an infectious bursal disease virus isolate from Zambia: classical atte-nuated segment B reassortment in nature with existing very virulent segment A. Arch.Virol. 158, 685–689. https://doi.org/10.1007/s00705-012-1531-4.

Kong, L.L., Omar, A.R., Hair-Bejo, M., Aini, I., Seow, H.F., 2004. Sequence analysis ofboth genome segments of two very virulent infectious bursal disease virus field iso-lates with distinct pathogenicity. Arch. Virol. 149, 425–434. https://doi.org/10.1007/s00705-003-0206-6.

Kumar, S., Stecher, G., Tamura, K., 2016. MEGA7: molecular evolutionary geneticsanalysis version 7.0 for bigger datasets. Mol. Biol. Evol. 33, 1870–1874. https://doi.org/10.1093/molbev/msw054.

Lasher, H.N., Shane, S.M., 1994. Infectious bursal disease. World’s Poult. Sci. J. 50,133–166. https://doi.org/10.1079/WPS19940013.

Lejal, N., Da Costa, B., Huet, J.C., Delmas, B., 2000. Role of Ser-652 and Lys-692 in theprotease activity of infectious bursal disease virus VP4 and identification of its sub-strate cleavage sites. J. Gen. Virol. 81, 983–992. https://doi.org/10.1099/0022-1317-81-4-983.

Letzel, T., Coulibaly, F., Rey, F.A., Delmas, B., Jagt, E., van Loon, A.A.M.W., Mundt, E.,2007. Molecular and structural bases for the antigenicity of VP2 of infectious Bursaldisease virus. J. Virol. 81, 12827–12835. https://doi.org/10.1128/JVI.01501-07.

Lojkić, I., Bidin, Z., Pokrić, B., 2008. Sequence analysis of both genome segments of threeCroatian infectious bursal disease field viruses. Avian Dis. 52, 513–519. https://doi.org/10.1637/8272-022808-Reg.1.

Lombardo, E., Maraver, A., Espinosa, I., Fernández-Arias, A., Rodriguez, J.F., 2000. VP5,the nonstructural polypeptide of infectious bursal disease virus, accumulates withinthe host plasma membrane and induces cell lysis. Virology 277, 345–357. https://doi.org/10.1006/viro.2000.0595.

Michel, L.O., Jackwood, D.J., 2017. Classification of infectious bursal disease virus intogenogroups. Arch. Virol. 162, 3661–3670. https://doi.org/10.1007/s00705-017-3500-4.

Mundt, E., Beyer, J., Muller, H., 1995. Identification of a novel viral protein in infectiousbursal disease virus-infected cells. J. Gen. Virol. 76, 437–443. https://doi.org/10.1099/0022-1317-76-2-437.

Ozel, M., Gelderblom, H., 1985. Capsid symmetry of viruses of the proposed Birnavirusgroup. Arch. Virol. 84, 149–161. https://doi.org/10.1007/BF01378968.

Qi, X., Gao, H., Gao, Y., Qin, L., Wang, Y., Gao, L., Wang, X., 2009. Naturally occurringmutations at residues 253 and 284 in VP2 contribute to the cell tropism and virulenceof very virulent infectious bursal disease virus. Antivir. Res. 84, 225–233. https://doi.org/10.1016/j.antiviral.2009.09.006.

Rautenschlein, S., Yeh, H.-Y., Njenga, M.K., Sharma, J.M., 2002. Role of intrabursal Tcells in infectious bursal disease virus (IBDV) infection: T cells promote viral clear-ance but delay follicular recovery. Arch. Virol. 147, 285–304. https://doi.org/10.1007/s705-002-8320-2.

Sharma, J., 2000. Infectious bursal disease virus of chickens: pathogenesis and im-munosuppression. Dev. Comp. Immunol. 24, 223–235. https://doi.org/10.1016/S0145-305X(99)00074-9.

Sharma, J.M., Dohms, J.E., Metz, A.L., 1989. Comparative pathogenesis of serotype 1 andvariant serotype 1 isolates of infectious bursal disease virus and their effect on hu-moral and cellular immune competence of specific-pathogen-free chickens. Avian Dis.33, 112–124.

Ture, O., Saif, Y.M., 1992. Structural proteins of classic and variant strains of infectiousBursal disease viruses. Avian Dis. 36, 829. https://doi.org/10.2307/1591540.

Wang, X., Zhang, H., Gao, H., Fu, C., Gao, Y., Ju, Y., 2007. Changes in VP3 and VP5 genesduring the attenuation of the very virulent infectious bursal disease virus strain Gxisolated in China. Virus Genes 34, 67–73. https://doi.org/10.1007/s11262-006-0002-y.

Wang, Y., Qi, X., Kang, Z., Yu, F., Qin, L., Gao, H., Gao, Y., Wang, X., 2010. A single aminoacid in the C-terminus of VP3 protein influences the replication of attenuated in-fectious bursal disease virus in vitro and in vivo. Antivir. Res. 87, 223–229. https://doi.org/10.1016/j.antiviral.2010.05.004.

Wu, Y., Hong, L., Ye, J., Huang, Z., Zhou, J., 2009. The VP5 protein of infectious bursaldisease virus promotes virion release from infected cells and is not involved in celldeath. Arch. Virol. 154, 1873–1882. https://doi.org/10.1007/s00705-009-0524-4.

Yao, K., Vakharia, V.N., 2001. Induction of apoptosis in vitro by the 17-kDa nonstructuralprotein of infectious Bursal disease virus: possible role in viral pathogenesis. Virology285, 50–58. https://doi.org/10.1006/viro.2001.0947.

C. Drissi Touzani, et al. Infection, Genetics and Evolution 77 (2020) 104097

7