Genetic diversity of tilapia lake virus genome segment 1 from 2011 to 2019 and a newly validated semi-nested RT-PCR method

The gene of RNA viruses, encoding RNA-directed RNA polymerase (RdRp) is relatively conserved due to its crucial function in viral genome replication and transcription making it a useful target for genetic diversity study and PCR detection. In this study, we investigated the genetic diversity of 21 tilapia lake virus (TiLV) genome segment 1 sequences predictively coding for RdRp subunit P1. Those sequences were obtained from infected fish samples collected in Ecuador, Israel, Peru, and Thailand between 2011 and 2019 (nine sequences from this study and 12 sequences from GenBank). Primers were then designed from the highly conserved regions among all 21 TiLV segment 1 sequences and used in semi-nested RT-PCR condition optimization. The result revealed that all 21 TiLV segment 1 sequences showed 95.00-99.94 and 99.00-100% nucleotide and amino acid sequence identity, respectively. These isolates were phylogenically clustered into three separate genetic clades, called i) Israeli-2011 clade (containing of TiLV isolates from Israel collected in 2011, Ecuador, and Peru isolates), ii) monophyletic Israel-2012 clade (containing only TiLV isolates collected from Israel in 2012), and iii) Thai clade (containing only sequences obtained from Thailand isolates). The newly established PCR protocol was 100 times more sensitive than our previous segment 3-based protocol when comparatively assayed with RNA extracted from infected fish. The assay was also shown to be specific when tested against negative control samples, i.e. RNA extracted from clinical healthy tilapia and from bacterial and viral pathogens (other than TiLV) commonly found in aquatic animals. Validation experiment with RNA extracted from naturally infected fish specimens collected in 2013-2019 yielded positive test results for all samples tested, confirming that our newly designed primers and detection protocol against TiLV segment 1, have a potential application for detection of all current genetic variants of TiLV.

TiLV is a segmented RNA virus that has been taxonomically assigned to Tilapia Tilapinevirus DNA sequence analysis 105 A total of 21 TiLV segment 1 sequences listed in Table 1 were used for comparison and 106 phylogenetic relationship analysis. Nine sequences were obtained from the present study and other 107 12 sequences were retrieved from the GenBank database. Multiple sequence alignments were 108 performed using MEGA 7 (Kumar et al. 2016) and a consensus sequence of 1,559 nucleotide 109 residues out of the putative 1,560 nucleotide-ORF segment 1 were obtained. Deduced amino 110 sequences of 519 nucleotide residues were translated from these consensus TiLV segment 1 using 111 ExPASy translate tool and then used for sequence comparison. Putative PB domain of the 112 translated sequences was predicted using InterPro database (https://www.ebi.ac.uk/interpro/).

113
Phylogenetic tree based on TiLV segment 1 nucleotide sequences was constructed using 114 Maximum-Likelihood with GTR+G (General Time Reversible model + Gamma distributed) 115 method as suggested by a best model feature of the MEGA 7 program. An ORF coding for a 116 putative PB1 of influenza virus C isolate Ann Arbor 1950 (NC_006308) was used as outgroup. 117 Pairwise distance analysis was also conducted using MEGA 7.

Development of a new semi-nested RT-PCR method based on segment 1 119
In this study, a new semi-nested RT-PCR detection of TiLV was developed. Primers were designed 120 from the highly conserved regions from the multiple sequence alignments of the 21 sequences of 121 TiLV segment 1. These sequences derived from TiLV isolates collected from the period from 2011 122 to 2019 (Table 1). Primer specificity was initially confirmed in silico with the NCBI primer-blast 123 tool. Primers TiLV/nSeg1F; 5'-TCT GAT CTA TAG TGT CTG GGC C-3' and TiLV/nSeg1R; 124 5'-AGT CAT GCT CGC TTA CAT GGT-3' with an expected amplified product of 620 bp were 125 used in the first round RT-PCR. Primers TiLV/nSeg1F and TiLV/nSeg1RN; 5'-CCA CTT GTG 126 ACT CTG AAA CAG -3' with an expected product of 274 bp were employed in the second round TiLV-infected red tilapia as template while no template added in the negative control. 5 µl of 135 product from the first round PCR was then used as template in the second round PCR reaction of 136 25 µl containing 500 nM of primer TiLV/nSeg1F, 600 nM of primer TiLV/nSeg1RN, 0.16 mM of 137 each dNTP, 0.8 mM MgCl2, 1 unit of Platinum Taq DNA polymerase (Invitrogen), and 1.2x 138 supplied buffer. Thermocycling conditions consisted of a 5 min initial denaturation step at 94 C 139 followed by 30 thermocycles and a final extension step described above. 10 µl of the amplified 140 products were analyzed by 1.5% agarose gel electrophoresis stained with ethidium bromide or 141 RedSafe DNA staining dye.

142
Validation of TiLV segment 1 semi-nested RT-PCR assay 143 Detection specificity of the assay was performed with RNA extracted from common viral and 144 bacterial pathogens found in aquatic animals as listed in Table 3. Viral infected fish tissues were 145 subjected to RNA extraction using Trizol reagent as mentioned above. For bacterial RNA isolation, 146 each bacterial strain was cultured in appropriated broth medium of 5 mL overnight at 30 C with 147 shaking at 200 rpm. After centrifugation, bacterial cell pellets were washed once with nuclease-148 free water and homogenized with Trizol reagent followed by the manufacturer's protocol. primers TiLV/nSeg1F and TiLV/nSeg1R; and TiLV/nSeg1F and TiLV/nSeg1RN, respectively 194 ( Fig. 2). Additionally, these primer pairs were shown, in silico, to be TiLV specific using primer-195 blast tool (https://www.ncbi.nlm.nih.gov/tools/primer-blast/). Thus, PCR condition optimization 196 for TiLV detection using the newly designed primers was conducted as described below. 198 The designed primer pairs described above were evaluated for their optimal annealing 199 temperatures which were found to be 60°C for both first and nested amplification (figure not revealed that the new protocol described in this study was 100 times more sensitive than the 207 previous one (Fig. 3a). This assay was done in two different laboratories i.e. 2 samples (i.e. fish 1 208 and 2) performed in Thailand and fish 3 in WorldFish Khulna laboratory, Bangladesh. Note the 209 presence of a 1.1 kb band predicted to be derived from cross hybridization of the amplified 210 products was visible at higher concentrated templates (Fig. 3a). Moreover, one set of specimens 211 from experimentally TiLV infected fish (Table 2, (Fig. 4). In addition, in one population, both clinically sick fish and apparently 220 healthy fish were all tested positive for TiLV segment 1 (  TiLV from a broader origin should be made available. Note that no correlation between sequence 234 variation or phylogenetic branches and disease severity was observed in this study.

235
Comparative nucleotide sequences revealed that TiLV segment 1 is relatively well conserved with 236 95.00-99.94% identity among all isolates used in this study, which is similar to that of standard serial dilutions of DNA plasmid in the laboratory. This study, therefore, compared the 256 sensitivity of two detection protocols together instead of using plasmid template to find out which 257 protocol was the best for field detection purpose. Interestingly, inter-laboratory tests consistently 258 proved that segment 1 PCR method was 100 times more sensitive than our previous segment 3 259 PCR protocol, when assayed with RNA extracted from diseased fish samples. In this study,       Details of bacterial isolates are shown in Table 2.