Concentration and quantification of Tilapia tilapinevirus from water using a simple iron flocculation coupled with probe-based RT-qPCR

Introduction

Tilapia tilapinevirus, commonly known as tilapia lake virus (TiLV), has been recently discovered and is the sole virus in the genus Tilapinevirus and the family Amnoonviridae (International Committee on Taxonomy of Viruses, 2019). TiLV is an RNA virus with a 10 segmented negative sense single stranded genome of approximately 10.323 kb (Bacharach et al., 2016). Since its discovery in 2014, the virus has significantly impacted tilapia aquaculture worldwide (Eyngor et al., 2014; Ferguson et al., 2014; Jansen, Dong & Mohan, 2019). TiLV typically results in a cumulative mortality from 20% to 90% (Behera et al., 2018; Dong et al., 2017a; Eyngor et al., 2014; Ferguson et al., 2014; Surachetpong et al., 2017). There are 16 countries that have confirmed the presence of TiLV, to date (Jansen, Dong & Mohan, 2019; Surachetpong, Roy & Nicholson, 2020). However, it is believed that the disease has a wider geographical spread due to the active movements of live tilapia between countries (Dong et al., 2017b). The waterborne spread of TiLV may contribute to the dissemination of the pathogen to new areas and other fish species (Chiamkunakorn et al., 2019; Eyngor et al., 2014; Jaemwimol et al., 2018; Piamsomboon & Wongtavatchai, 2021). Experimental evidence has already demonstrated that TiLV is transmitted horizontally through water (Eyngor et al., 2014).

Environmental DNA and environmental RNA (eDNA and eRNA) have been demonstrated to be an effective tool in the surveillance of fish pathogens (Haramoto et al., 2007; Kawato et al., 2016; Minamoto et al., 2009; Nishi et al., 2016). Environmental RNA may aid in disease surveillance, as described by Kawato et al. (2016) who used an iron flocculation method to concentrate red sea bream iridovirus (RSIV) in a challenge model with Japanese amberjack (Seriola quinqueradiata). Results from that study showed that detection RSIV by qPCR in fish-rearing water samples peaked more than 5 days before fish mortality occurred, suggesting the potential benefit of the iron flocculation method for forecasting disease outbreaks. Others studies used a cation-coated filter method to detect the DNA of cyprinid herpesvirus 3 (CyHV-3), also known as koi herpesvirus (KHV), from concentrated river water samples 3 to 4 months before mass mortality events occurred in wild carp in Japan (Haramoto et al., 2007; Minamoto et al., 2009). The virus continued to be detectable in river water samples for at least 3 months after the outbreaks (Minamoto et al., 2009). These findings helped local authorities and farmers to make rapid decisions for emergency harvesting, the implementation of biosecurity measures, and appropriate disinfection procedures and fallowing periods.

Several molecular methods have been developed for the detection of TiLV, including RT-PCR (Eyngor et al., 2014), nested and semi-nested PCR (Dong et al., 2017a; Kembou Tsofack et al., 2017; Taengphu et al., 2020), RT-qPCR (Tattiyapong, Sirikanchana & Surachetpong, 2018; Waiyamitra et al., 2018), loop-mediated isothermal amplification (LAMP) (Kampeera et al., 2021; Phusantisampan et al., 2019; Yin et al., 2019) and the nanopore-based PCR amplicon approach (Delamare-Deboutteville et al., 2021). However, all of these methods require fish tissues for diagnosis and none reported any use for TiLV detection from environmental water samples. Previous probe-based RT-qPCR methods have been developed to detect TiLV from tilapia clinical samples with detection limits of 2.7 × 10⁴ or ~70,000 copies (Kembou Tsofack et al., 2017; Waiyamitra et al., 2018). However, these may not be sensitive enough to detect low viral loads of TiLV in environmental water samples. Moreover, at the time of earlier primer and probe design, there were a limited number of TiLV genome sequences in the NCBI database. As a result, sequence variation among viral isolates and within the genome segments may not be accounted for in the design of those previous methods. The objective of this study was to develop a new RT-qPCR assay based on updated publicly available TiLV genomic sequences data, in order to detect and quantify TiLV in fish tissues and in environmental RNA (eRNA) concentrated from fish-rearing water samples using an iron flocculation method.

Materials and Methods

Optimization for viral concentration protocol

Virus preparation

The viral stock used in this study was isolated from TiLV-infected Nile tilapia using the E-11 cell line, a clone of the cell line SSN-1 derived from the whole fry tissue of snakehead fish (Cat no. 01110916-1VL; Sigma-Aldrich, St. Louis, MO, USA). The virus was propagated as described by Dong et al. (2020). Briefly, 200 µL of TiLV stock (~10⁸ copies/mL) was added into a 75 mL cell culture flask containing a monolayer of E-11 cell and 15 mL of L15 medium (Leibovitz), incubated at 27 °C for 5 days. The culture supernatant containing viral particles was collected after centrifugation at 15,000g for 10 min at 4 °C. The viral stock was kept in aliquots of 1 mL at −80 °C until used.

Iron flocculation

The viral concentration was determined using the iron flocculation method, following the protocol previously described by Kawato et al. (2016) with some modifications. The workflow of this method is illustrated in Fig. 1. Briefly, 100 µL of TiLV viral stock containing ~10⁷–10⁸ viral copies was added to 500 mL of sterile water that contained 1% marine salt and 36 µM ferric chloride. The viral copy numbers were quantified by RT-qPCR using RNA extracted from viral stock vials. The suspension was stirred at room temperature for 1 h before being mechanically filtered through a 0.4-μm pore size polycarbonate filter (Advantec, Chiba, Japan) with a vacuum pump connected to a filter holder KG-47 (Advantec, Chiba, Japan) under <15 psi pressure. The flocculate-trapped filters were then subjected to nucleic acid extraction using the Patho Gene-spin DNA/RNA extraction kit (iNtRON Biotechnology, Gyeonggi-do, South Korea). In comparison studies, the flocculate-trapped filters were soaked in an oxalate-EDTA buffer to re-suspend the trapped particles (John et al., 2011) prior to nucleic acid extraction. Experiments were carried out in two to four replicates. The viral concentration and percentage (%) recovery of the virus copies were calculated from Cq values after flocculation and compared to that of the starting viral stock.

Results

A new probe-based RT-qPCR method for detection and quantification of TiLV

The Seg9 RT-qPCR method developed in this study had a sensitivity of 10 copies/µL with mean Cq ± SD values of the detection limit at 38.24 ± 0.09 (Fig. 2A). Therefore, samples with a Cq value ≥38.15 were considered TiLV negative or under the limit of this detection method. The established RT-qPCR was found to be highly efficient with slope = −3.476, R² = 0.998, and E (amplification efficiency) = 94.0% based on the standard curve analysis (Fig. 2B). The formula, copy number = 10^{(Cq - Intercept)/Slope} i.e., 10^{(Cq – (−42.295))/−3.476}, may be used to calculate TiLV copy numbers found in the assayed samples. The analytical specificity test revealed that the method was highly specific to TiLV alone since no amplifications were found when the method was assayed with RNA templates extracted from three other viruses, 15 bacterial species, and healthy tilapia (Fig. 2C, Table S1). The method had a 100% diagnostic specificity and a 100% diagnostic sensitivity when assayed with previously diagnosed TiLV infected and non-infected fish samples (n = 65 with Cq value ranges 13.02–34.85) (Table 1).

Table 1:

Diagnostic specificity and sensitivity of the Seg9 probe-based RT-qPCR method.

Test results	Diseased samples (n = 44)	Non-diseased samples (n = 21)
Positive (+)	True positive 44	False positive 0
Negative (−)	False negative 0	True negative 21
Diagnostic sensitivity (%)	100
Diagnostic specificity (%)	100

DOI: 10.7717/peerj.13157/table-1

Discussion

Methods to concentrate and recover viral particles from environmental water samples have been applied in human health studies especially with waterborne diseases caused by enteric viruses (Cashdollar & Wymer, 2013; Haramoto et al., 2018). The process has now become essential in the study of aquatic environments (Jacquet et al., 2010). Several techniques have been used for viral concentration from aquatic environment, including coagulation/flocculation, filtration/ultrafiltration, and centrifugation/ultracentrifugation (Cashdollar & Wymer, 2013; Ikner, Gerba & Bright, 2012). Our study used an iron flocculation method which was initially described for virus removal from freshwaters (Chang et al., 1958) and virus concentration from marine waters (John et al., 2011). It was later adapted to detect and quantify two fish viruses: nervous necrosis virus (NNV) (an RNA virus) and red sea bream iridovirus (RSIV) (a DNA virus) that were experimentally spiked in fish-rearing water (Kawato et al., 2016; Nishi et al., 2016). The recovery rate was estimated by qPCR and yielded >50 and >80% for NNV and RSIV, respectively. In this study, the recovery rate of TiLV (an RNA virus) from spiked-water was considerably lower (16.11 ± 3.3%), however, similar practical methods have been used for concentrating and detecting human viruses from water environments (Haramoto et al., 2018). For example, murine norovirus-1 (MNV-1) used as a viral model in viral concentration assay of human enteric viruses was recovered from spiked-water at 5.8–21.9% using the electronegative hydroxyapatite (HA)-filtration combined with polyethylene glycol (PEG) concentration method. This protocol was then used for the detection of human noroviruses (NoV) and hepatitis A virus (HAV) in all water types (De Keuckelaere et al., 2013). More recently, researchers used porcine coronavirus (porcine epidemic diarrhea virus, PEDV) and mengovirus (MgV) as model viruses to concentrate severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) from water samples (Randazzo et al., 2020). The use of an aluminum hydroxide adsorption-precipitation concentration method, allowed PEDV and MgV spiked in water to be recovered at 3.3–11.0%. The method can then be applied to detect SARS-CoV-2 RNA in untreated wastewater samples of ~10^5.4 genomic copies/L (Randazzo et al., 2020).

Despite a low recovery rate from water samples in this study, we confirmed the usefulness of the iron flocculation and RT-qPCR approach to concentrate and determine the concentration of TiLV from fish-rearing water and other water sources from two aquaculture production systems during disease outbreaks. The inherent nature of DNA and RNA viruses and their ability to persist outside their hosts may also contribute to the differences observed in the recovery rates (Cashdollar & Wymer, 2013; Pinon & Vialette, 2018). Other viral concentration techniques using different coagulant/flocculant chemicals, as well as more efficient RNA extraction methods, should be tested for further improvement of TiLV recovery from water.

Following the viral concentration and recovery processes, viral detection is generally performed using PCR-based assays, cell culture methods, or viral metagenomics analysis (Haramoto et al. (2018)). Here, we employed the RT-qPCR technique for the detection and quantification of TiLV, although the detected amounts did not represent the viral viability. Using all TiLV genomic sequences publicly available, we designed a new set of conserved primers and a probe targeting the viral genomic segment 9. The newly established RT-qPCR protocol was highly specific to TiLV and did not cross-amplify RNA extracted from other common bacterial and viral aquatic pathogens. The method is very sensitive as it can detect as low as 10 viral copies per µL of template, which is >2,700 times more sensitive than previous probe-based RT-qPCR methods (Kembou Tsofack et al., 2017; Waiyamitra et al., 2018). Our RT-qPCR method has a 100% diagnostic specificity and sensitivity in agreement with previous results (n = 65) obtained using semi-nested RT-PCR protocols (Dong et al., 2017a; Taengphu et al., 2020). An increased number of sample sizes with diverse geographical sources may be required for further investigation. Most importantly, this new Seg 9 RT-qPCR assay was able to detect and quantify the TiLV load from various types of field samples, including clinically sick fish, asymptomatic fish, and water samples, as opposed to other molecular diagnostic methods optimized solely for use in fish specimens.

The viral loads from water samples collected during the two disease events were approximately ~10³ viral copies/L from earthen ponds and ~10⁴ viral copies/L from open-cages. However, these concentrations may be significantly higher due to substantial losses during the concentration and recovery process. Higher viral loads observed in some of the water samples collected during the disease outbreak may be due to the active shedding of the virus from diseased fish into the environment, and may confirm the waterborne transmission nature of TiLV that was reported previously (Eyngor et al., 2014; Yamkasem et al., 2019). The potential to forecast TiLV outbreaks should be further investigated by experimental infection to monitor viral loads in water in relation to fish morbidity and mortality as previously described for other fish pathogens (Haramoto et al., 2007; Kawato et al., 2016; Minamoto et al., 2009; Nishi et al., 2016).

Conclusions

In summary, the viral concentration method by iron flocculation used in concert with a newly developed probe-based RT-qPCR was not only successful for the detection and quantification of TiLV from water in diseased pond/cages, but also from unaffected ponds, reservoir, and sewage water. This method, apart from its potential practical use for future monitoring programs of TiLV viral load in water samples from various culturing units, may be useful to detect possible TiLV contamination from incoming and outgoing waste water as well as to test the systems after disinfection treatments. Such applications will support health professionals and farmers to design appropriate biosecurity interventions to reduce the loss caused by TiLV in tilapia farms and hatcheries.

Supplemental Information