Evaluation of Anticoagulant Solutions in the Flow Cytometric Analyses of Fish Blood and Larval Cells for Assessing Ploidy and Other Cell Characteristics

Artificially induced, functionally sterile triploid grass carp (Ctenopharyngodon idella, Valenciennes, 1844) are used globally to control populations of aquatic macrophytes (Chilton & Muoneke, 1992; Wittmann et al., 2014), avoiding ecological risk (Fraser et al., 2012), and diploids serve as a major food source globally, as well (FAO, 2020). States within the U.S. utilizing triploid grass carp can monitor diploid contamination in the triploid supply chain (Kinter et al., 2018) and in natural water bodies (Clemens et al., 2016) by determining ploidy level of these individuals. Flow cytometry has been used to replace time-consuming and empirical methods (Krynak et al., 2015; Mc Carter, 1988; Strommen et al., 1975; Thorgaard, 1983) for ploidy determination. In flow cytometry, nuclear DNA of target cells is stained by intercalating fluorescent dyes, and the resulting signal is quantified by a flow cytometer (Jenkins et al., 2019; Kinter et al., 2018). While compatible with large-scale analysis, DNA flow cytometry also allows for the identification of fine DNA quantitative differences, i.e. genetic mosaicism (Jenkins et al., 2017). Dorsal fin-based genome size and ploidy level determination are non-invasive sampling options (Lamatsch et al., 2000) with flow cytometry; however, both databases and literature consistency demand red blood cell-based data. There is no bleeding protocol proposed in the literature targeting juvenile grass carp that suggests type and concentration of anticoagulant for a given blood volume, fulfilling the quality requirements of flow cytometry. Previously, 500 IU/ml sodium heparin was used with juvenile (~6.67 g body weight) grass carp; however, the exact technical details such as blood volume and any effects on red blood cells were not described (Liu et al., 2019). Another study used a robust protocol for sample preparation for flow cytometry; however, it was adapted for adult individuals blood from the optic nerve vein (Kinter et al., 2018). In another study, Na2EDTA had an adverse effect compared to sodium heparin in the range of haemolysis and in other blood cell-associated parameters in anticoagulated samples obtained from common carp (Witeska & Wargocka, 2011). The objective of this study was to perform a comparative analysis of a variety of anticoagulants on the cytometric properties of blood cells of juvenile grass carps, using common carp (Cyprinus carpio L 1758) and European catfish (Silurus glanis L 1758) as reference species. Furthermore, we report the identification of a genetically mosaic individual from a small stock of putative triploid grass carp larvae. Research was conducted following the National Food Chain Safety Office approved experimental layout with the reference number: GK-2675/2012. EDTA (Panreac Applichem; A5097) solutions were diluted in the presence of NaOH (Sigma-Aldrich; 1310-73-2). TEVA HEPARIBENE NA 25000 NE (ATC code: B01AB01) and VACUETTE® 2 9NC tubes were the source of heparin and sodium citrate. Details can be found in Table S1. Ploidy of a total of 36 grass carp larvae was determined: 28 heat shock-treated (Cassani & Caton, 1985) putative triploids and eight untreated diploid controls (3 days post-hatch). Larval samples were prepared according to Jenkins et al. (2017). The suspension was filtered through a 30 µm filter to remove debris and was fixed with 70% ethanol (diluted in physiological saline). Anticoagulant tests were performed on 29 grass carp individuals (L: 8.58 cm ± 0.56; W: 13.05 g ± 2.188); 24 common carp (L: 7.85 cm ± 0.76; W: 15.52 g ± 4.27) and 30 European catfish (L: 16.2 cm ± 2.52; W: 38.91 g ± 18.91) fingerlings. Experimental fish were collected from commercial farms and were kept in 350-L tanks (20 ± 0.5℃) supplied with external filter and oxygen saturation was maintained over 80%. Fish were fed twice daily with dry pellets. Concentration of anticoagulants was calculated based on Walencik and Witeska (2007) and Witeska and Wargocka (2011). Due to the small size of the fish, 100 µl of blood was taken from each individual. Anticoagulants, 25 G needles and 1-ml syringes were kept at 4℃ before use. Prior to blood collection, ¾ part of the anticoagulant was aspirated into the syringe and the remaining ¼ part was transferred into the 1.5-ml Eppendorf tube. Blood samples were fixed with 25% ethanol in physiological saline, added dropwise. In the case of grass carp, complete coagulation was observed with 1 mg/ml EDTA, whereas 2 mg/ml EDTA-induced vigorous haemolysis, indicated by high amount of cell debris as detected either on the flow cytometric histograms or by microscopy. Applied anticoagulant–blood ratios at the desired volumes are shown in Table S1. Larval suspensions and anticoagulated blood samples for flow cytometry were prepared with the FXCycle PI/RNASE kit (Thermo Fisher, F10797) according to the manufacturer's protocol. DNA quantification of cells was performed on a Beckman Coulter FC 500 flow cytometer with a 488 nm 20 mW argon ion laser, recorded on FL3 detector (650 nm LP). Acquisitions were stopped after counting five thousand cells in the selected gate region (Figure S1) or after 5 min. Median fluorescence intensity (MFI) and coefficient of variance (CV) values of histograms were recorded. Duplicate measurements were conducted on all samples (Figure 1). Data were evaluated by Flowing software (www.flowingsoftware.com, v2.5.1) and WinMDI 2.8 software. Blood cell morphology was imaged at 1000× magnification with an Olympus BX 43 biological microscope. Diploid (mean MFI: 227.9; SD: 9.28) and triploid (mean MFI: 323.5; SD: 11.42) larval specimens were clearly distinguished with 1.46 times MFI difference. A genetically mosaic individual was identified among the 26 triploidized grass carp larvae. In our histogram profile, the deviation between the size of the diploid and triploid peaks is different compared to peaks of previously published mosaic grass carp data (Jenkins et al., 2017) which indicates that the level of mosaicism can vary among individuals at larval stage. In agreement with previous findings (Jenkins et al., 2017), our results highlight the intriguing presence of genetic mosaics with unknown reproductive status among polyploids. CV value determines the histogram peak width as percentage of the peak position, referring to the structural integrity of cells in the sample, at a maximum acceptable value of 8% (Givan, 2001). G0/G1 cell phase event number correlates with the number of PI-stained blood cells gated and recorded. In the case of grass carp, the CV of the histogram peak was not significantly affected by anticoagulant type and concentration (p = 0.612), and in all cases, the CV value remained under the 8% threshold, indicating structurally intact cells (Figure 2a). Furthermore, 1.5 mg/ml EDTA anticoagulant resulted in the highest number of cells recorded (G0/G1 cell phase event number), whereas in 2 mg/ml EDTA the number of events was significantly lower (p = 0.004). Common carp blood yielded mixed results when treated with different anticoagulants: CV values fell outside of the 8% range, except for EDTA samples (Figure 2b). Similar to grass carp, the highest number of events in the G0/G1 cell phase was also measured with samples anticoagulated with 1.5 mg/ml EDTA and 150 IU Na-heparin decreased the event number significantly (p = 0.016). Due to extreme haemolysis of European catfish samples, histogram peaks were unacceptably wide, CV values were over 8% and lower event numbers were recorded. The application of 1.5 mg/ml EDTA yielded the highest results in event numbers (ANOVA, p = 0.022) (Figure 2c). The addition of 1.5 mg/ml of EDTA caused no, or a slight level of haemolysis in grass carp blood samples, leaving blood cells structurally intact. Therefore, histogram profiles could be evaluated with acceptable histogram parameters (Figure 3a), which is a novel result of our study. During sample preparation of whole blood nuclei-based DNA ploidy assay, the presence of haemolysis and clot alone do not exclude the possibility of successful PI-based flow cytometric measurements. In the case of the evolutionarily closely related common carp, partial haemolysis of blood cells was observed similar to Witeska and Wargocka (2011), yet a sufficient number of structurally intact blood cells remained allowing for successful PI-based ploidy analysis by flow cytometry (Figure 3b). Catfish samples were mostly haemolysed with 1.5 mg/ml EDTA, and cells with intact morphology were scarce (Figure 3c). The number of blood cells and those with intact DNA in the sample were both lower, reflected by the flow cytometric histogram analysis—the PI fluorescence histogram profile could not be used (Figure 2c). Our results should improve the flow cytometric analyses of polyploids, not only in grass carp, but potentially in other teleosts as well. We would like to thank Aranyponty Zrt. and Miklós Bercsényi for providing their fish for our experiment. Special thanks to László Orbán and Ildikó Szeverényi for their helpful comments on earlier versions of the manuscript. The authors declare no conflict of interest. SzN developed the idea and concept; AGy and AB contributed to experimental work on fish; SzN and AGy acquired flow cytometric data and analysed the data; AGy prepared the manuscript; SzN, AGy and AB finalized the manuscript. Not applicable as fish were sampled at a commercial farm. The data are available from the corresponding author upon reasonable request. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.