Abnormalities in red blood cell production and pathogenesis of anemia in the progression of rock bream iridovirus (RBIV)

Highlights • RBIV causes systemic infections in rock bream.• Spleen reached its lethal level the fastest during RBIV infection.• Hemoglobins (α and β), CLNK and GALT were lowest expressed in RBIV-infected rock bream.• The RBC, HGB and HCT levels gradually decreased during RBV infection.

The spleen is the major site of pathogen growth and disease pathology in virus-affected rock bream (Jung and Oh, 2000) and is frequently used to diagnose damage caused by RBIV infections.The simplest method for diagnosing RBIV infection involves confirming the presence of abnormally enlarged cells on Giemsa-stained impression-smears of the spleen (Jung and Oh, 2000).Furthermore, when rock breams are infected, there is a marked increase in the weight of the spleen (up to approximately three-fold), indicating that spleen enlargement is positively correlated with RBIV replication (Jung et al., 2017a;Jung et al., 2017b;Jung and Jung, 2019).Although the genus Megalocytivirus comprises viruses associated with serious systemic infections that result in significant mortality, organs other than the spleen have not been commonly used to diagnose megalocytivirus infections.Moreover, although information regarding its etiology is limited, the clinical signs of the disease include anemia-related symptoms such as pale gills, increased respiration, lethargy, atypical swimming and negative buoyancy (Jung and Oh, 2000).
Massive deaths occur annually in Korean aquaculture fields owing to the combined effects of disease, low and high water temperature, typhoons, precipitation and red tides.Accordingly, minimizinge damage by inspecting fish health through regular surveillance is necessary; however, an established standard for determining these complex factors is lacking.Various indices have been used to evaluate the condition or well-being of fish, including the relative condition factor (Le Cren, 1951), gut index (Jensen, 1980), relative weight (Wege and Anderson, 1978), visceral somatic index (Adams et al., 1982;Delahunty and De Vlaming, 1980), RNA-DNA ratios of the liver and muscle (Bulow et al., 1981), and liver somatic index (Allen and Wootton, 1982;Bulow et al., 1978;Delahunty and De Vlaming, 1980;Edwards et al., 1972;Heidinger and Crawford, 1977;Tyler and Dunn, 1976;Valtonen, 1974).Them, liver somatic index is a useful biomarker for detecting environmental stressors, and is one of the most sensitive growth indicators (Allen and Wootton, 1982;Bulow et al., 1978;Edwards et al., 1972;Heidinger and Crawford, 1977;Tyler and Dunn, 1976;Valtonen, 1974).However, one disadvantage is that the fish must be sacrificed killed to test the indicators mentioned above.It is vital to obtain samples (e.g., blood) to measure the health of fish without killing them.
Therefore, various studies on blood cells, antibody titers, hemoglobin (HGB), hematocrit (HCT), pathogen-killing effects and immune responses can be conducted using the basic components of blood, white blood cells, red blood cells and plasma.However, the utility of these parameters in diagnosing and determining infections caused by pathogens is very low.This implies that barring a few pathogens that directly infect the blood, most pathogens target the spleen, kidneys, and heart as their primary target organs.There has been little interest in research using blood, where the proliferation of pathogens is relatively weak.Anemia may be useful among the different health measures that use blood.Anemia is not only caused specifically under a certain condition but also under multiple conditions, including water quality, dense diet, feed, environment and disease.Therefore, if the degree of anemia is quantified and compared to variations in a specific condition, and if a correlation is established between two, the condition may be used to assess health.
The aim of present study was to evaluate viral distribution in the organs of rock bream and the hematological changes following RBIV infection.Anemia was confirmed by measuring HCT, HGB and RBC levels according to the viral proliferation pattern, followed by a physiological analysis of blood.The expression levels of gene related to blood functions were also assessed in rock bream infected with RBIV.

Experimental fish and virus
Rock breams free from RBIV were obtained from a local farm and reared at Hanseo University.The virus used in the present study was originally isolated from RBIV-infected rock bream in 2010 (RBIV-HD 2011, KT031401) (Jung et al., 2014).The major capsid protein (MCP) gene copies of RBIV in the supernatant preparations were quantified using quantitative real-time polymerase chain reaction (qPCR).The RBIV titer was calculated as 1.1 × 10 7 /100 μl MCP gene copies.

Artificial infection
To determine mortality rate, 20 fish (10.4 ± 1.2 cm, 20.1 ± 2.4 g) in each group were divided randomly into two groups: the virus-injected group, which was intraperitoneally (i.p.) injected with RBIV (100 μl/ fish) containing 10 6 MCP gene copies, and the control group, which was administered phosphate-buffered saline (PBS).The fish were maintained at 23 • C in an aquarium containing 20 L seawater.
The gill, liver, intestine, spleen, heart, kidney, muscle, brain and red blood cells (RBCs) were collected from five virus-injected fish and five control fish at 1, 4, 7, 10, 14, 16 and 17 days post infection (dpi).Blood (over 150 μl/fish) was collected at each sampling point; 50 μl of the blood samples was used for biochemical examination, 40 μl for hematocrit examination, 20 μl for the examination of complete blood cell count (CBC) and 40 μl was purified for RBCs.The RBCs were purified using two consecutive density gradient centrifugations (7206 g, Ficoll 1.007: Sigma Aldrich).The samples were stored at -80 • C after being flash-frozen in liquid nitrogen.
Blood samples (40 μl/fish) were further used to analyze HCT.The HCT test measures the volume of RBCs compared to the total blood volume (red blood cells and plasma).The HCT tube was placed near the Eppendorf tube and blood was allowed to flow via capillary action into the HCT tube until two-thirds to three-fourths full or to a predesignated mark on the tube.The tubes were then placed in a microhematocrit centrifuge for 10 min at 10,000 rpm.

Biochemical examination of blood
Blood samples (50 μl/fish) were used to analyze biochemical components, including total protein (TP), gamma-glutamyl peptidase (GGT), aspartate aminotransferase (AST/GOT), alanine aminotransferase (ALT/ GPT), alkaline phosphatase (ALP), amylase (AMY), creatine (crea), blood urea nitrogen (BUN), Bun/Crea, total cholesterol (TC) and triglyceride (TG) using an Exigo C200 (Boule Medical AB, Sweden).Considering that the size of the fish was small, biochemical examination was performed using five volumes of blood and five volumes of PBS.All analyses were performed according to the manufacturer's instructions.

Quantitative real-time PCR analysis of expression levels of bloodrelated genes
To analyze the expression of blood-related genes, total RNA was extracted from the kidney and RBCs using the RNAiso Plus reagent (TaKaRa, Japan), following the standard protocol.Total RNA was treated with DNase I (TaKaRa, Japan) and reverse-transcribed using the ReverTra Ace qPCR RT Kit (Toyobo, Japan) according to manufacturer's protocol.Real-time PCR was carried out in a CFX Connect Real-Time PCR Detection System (Bio-Rad, USA) using an AccuPre®2x Greenstar qPCR Master Mix (Bioneer, Korea) as described in a previous study (Jung et al., 2014).Each assay was performed in duplicate with β-actin RNA as the control.The primers used to amplify immune-related genes are listed in Table 1.The quantitation of the mRNA was determined using the 2 − ΔΔCt method (Livak and Schmittgen 2001).

Determination of viral copy numbers
Viral loads in the gill, liver, intestine, spleen, heart, kidney, muscle and RBCs samples (20 to 160 mg) were measured using qRT-PCR.Genomic DNA was isolated using the High Pure PCR Template Preparation Kit (Roche, Germany).The qRT-PCR procedure followed was 95 • C for 10 min, 95 • C for 15 s, and 60 • C for 45 s for 40 cycles followed by scanning, as previously described (Jung et al., 2015).The details for establishing the standard curve have been previously described (Jung et al., 2015).The RBIV genome copy numbers were determined from 1 mg of organ weight of the gill, liver, intestine, spleen, heart, kidney, muscle and RBCs were quantified with specific MCP primer sets (Table 1).

Statistical analyses
Statistical analyses of gene expression and anemia related factor levels of rock bream were performed using unpaired t-tests using GraphPad Prism software version 5.0 for Windows (GraphPad Software, USA).Differences were considered statistically significant at p<0.001, p<0.05 and p<0.01.All the data are represented as the means ± the standard error.

Mortality rates
In rock bream reared at 23 • C, the mortality between 11 dpi and 18 dpi resulted in final mortality of 100 % (Fig. 1A).Dead rock bream exhibited typical signs of RBIV infection, such as enlarged basophilic cells in the spleen imprint and an enlarged spleen (Fig. 2).

Three infection stages of RBIV
The RBIV replication profiles were constructed and are presented in Fig. 1 and 3. Nine organs (gills, liver, intestine, spleen, heart, kidney, muscle, brain and RBCs) showed elevated viral replication following RBIV infection.Viral replication patterns were categorized into three infection stages: early (1 dpi), middle (4 to 10 dpi) and late (14 to 17 dpi) (Fig. 1 and 3).At the early stage of infection at 1 dpi, increases (range: 10 1 -10 2 ) in viral load were observed in all nine organs, and the infected fish appeared grossly healthy, demonstrating an increase in the number of infectious agents (Fig. 1).
The middle stage of the disease occurred between 4 and 10 dpi (viral load range: 10 1 -10 6 ).This was followed by active replication of the virus, wherein the infected fish demonstrated clinical symptoms.Changes in the virus replication (range: 10 2 -10 3 ) in the six organs (liver, muscle, brain, heart, gill and intestine) and RBCs showed elevated viral loads.However, the spleen and kidney showed high replication levels, with virus copy numbers of 1.70 × 10 6 and 3.67 × 10 4 /mg, respectively (Supplementary Table 1) (Fig. 1).
The disease proceeded to the last stage between 14 and 17 dpi (range: 10 3 -10 6 ).The viral load peaked exponentially and the fish demonstrated clinical symptoms with specific mortality.This period is regarded as the late stage of infection.Following initial RBIV infection, the number of replicated RBIV rapidly increased from 14 dpi.The RBIV load increased in all organs, indicating that intensive replication, transcription, and translation occurred to assemble and form viral progeny inside the infected host cells.At 16 dpi, the viral levels in the nine organs were significantly elevated.Among these organs, replication in six organs peaked at 16 dpi.In this study, the distribution of the virus based on the quantity in each organs was in the following order: spleen (average 9.92 × 10 6 ), heart (6.14 × 10 6 ), kidney (2.12 × 10 6 ), liver (1.87 × 10 6 ), gill (1.63 × 10 6 ), intestine (5.53 × 10 5 ), muscle (1.89 × 10 5 ), brain (1.05 × 10 5 ) and RBCs (2.62 × 10 3 ).This indicated that most RBIV virions were completely assembled at 16 dpi (Fig. 1).At 17 dpi, viral level were elevated in nine organs were elevated, including the muscle, intestine and RBCs, which exhibited strongly replication (Fig. 1).

Differential RBIV replication profiles
Specific mortality occurred between 11 and 18 dpi, and viral replication peaked between 14 and 17 dpi marking this period as the acute phase of infection (Fig. 1A and 3).Viral replication patterns at approximately 14 to 17 dpi, may be significant for evaluating the factor(s) involved in fish death.A lethal level was determined for each organ and all viral replication in the rock bream was clustered into three groups based on the similarity of replication patterns on 14 to 17 dpi as follows (Fig. 1 and 3): group 1, five organs (spleen, kidney, heart, gill and liver) showed high replication levels of >10 5 at 14 dpi and this was regarded as the 'lethal' virus copy number (Fig. 1B); group 2, three organs (muscle, brain and intestine) exhibited high replication levels of >10 4 at 14 dpi, which was regarded as the 'lethal' virus copy number (Fig. 1C); and group 3, which replication in one organ (RBCs) of >10 3 at 14 dpi considered the 'lethal' virus copy number (Fig. 1D).The average viral load distribution in the fish organs was in the following order: >10 6 (one organ: spleen), >10 5 (six organs: heart, kidney, liver, intestine, gill and muscle), >10 4 (one organ: brain) and >10 3 (one organ: RBCs) (Fig. 1B, 1C and 1D).
However, MCV, MCH, and MCHC values in the RBIV-infected group showed no significant differences compared to those in the control group  (Fig. 4B, 4E, and 4F).
The ALT values of the control groups of the PBS administered fish ranged from 3.0-8.0(U/L) and averaged 5.64 (U/L) (Supplementary Table 3).The ALP values of the virus-injected group were lower than those of the control fish and gradually increased from 14 (average 10.0) to 17 dpi (average 14.2) (p<0.001) (Fig. 6D).
The BUN values of the control groups of the PBS administered fish ranged between 0.7-1.5 (mmol/L) and averaged 1.0 (mmol/L) (Supplementary Table 3).In the virus-injected group, BUN values were lower than those of the control fish and gradually decreased from 1 dpi (average 0.74) to 10 dpi (average 0.71) followed by a gradual increase from 16 dpi (average 1.53) (p<0.01) to 17 dpi (average 1.90) (p<0.05) (Fig. 6H).
However, no differences were observed in the levels of TP, GGT, AMY, CREA, TC and TG in virus-injected fish compared to those in the control (Fig. 6A, 6B, 6D, 6F, 6G, 6J and 6K).

Discussion
Megalocytivirus includes viruses associated with serious systemic infections that can cause significant mortality.Organs, other than the spleen have not been commonly studied to evaluate megalocytivirus infections.The present study revealed similarities in RBIV replication patterns in nine organs (spleen, liver, muscle, brain, gill, kidney, heart, intestine and RBCs).The acute stage of RBIV infection was observed to occur from 4 to 10 dpi in all organs of the RBIV-infected rock bream, with viral replication reached its peak at 16 to 17 dpi.The RBIV copy numbers were not as high as those in the RBCs and other organs.Nonetheless, time-dependent increses were observed in the blood and Ficoll-purified RBCs of rock bream indicating that RBIV causes systemic infections in rock bream.
The patterns of viral copy number and spleen size in virus-infected rock bream have been evaluated previously to determine the lethal and safe levels of the viral copy number/or the spleen index, which may define disease progression (Jung et al., 2017b;Jung and Jung, 2019).In these previous studies, upon dead (100 % mortality, at 20-29 • C), the virus copy number and spleen index (spleen weight / fish weight × 100) were 10 8 -10 10 /100 μl and 3.00-5.38,respectively.Conversely, at survival (0 % mortality, at 17 • C), the virus copy number and spleen index were as low as those of the not-infected control (10 2 -10 3 /100 μl and 0.34-1.22,respectively).In the present study, a threshold (dangerous limit) was determined for the viral replication pattern in the nine organs including mortality points indicating a possible viral copy number, that leads to fish death.
Mortality was observed between 11 and 18 dpi and viral replication peaked in all organs between 14 and 17 dpi.Lethal levels were set for each organ at 14 and 17 dpi.The lethal level for each organ was in the order of: 10 5 /mg (spleen, kidney, heart, gill and liver), 10 4 /mg (muscle, brain and intestine) and 10 3 /mg (RBCs).The organs were organized according to the lethal levels in the following order: spleen, kidney, heart, liver, gill, intestine, muscle, brain and RBCs.Among them, the lethal level was achieved the quickest in the spleen, confirming its preference for the diagnosis of RBIV infection.Red blood cells (RBCs) play a key physiological role in the transport of oxygen from the lungs to tissues via hemoglobin.In mammals, mature RBCs lack nuclei, organelles and ribosomes (Moras et al., 2017).In contrast, with a few exceptions, non-mammalian RBCs (particularly in fish), nucleate and contain organelles in their cytoplasm (Nombela and Ortega-Villaizan, 2018).Nucleated RBCs offer numerous advantages over non-nucleated RBCs, and teleost RBCs can induce toll-like receptors (TLRs), phagocytosis by macrophages and cytokines or interferons (Dahle et al., 2015;Nombela et al., 2017;Nombela and Ortega-Villaizan, 2018;Passantino et al., 2007;Rodriguez et al., 2005).This indicates that in addition to their unique function in oxygen transport, the red blood cells of fish contribute to the host immune response to infectious diseases.Our previous study uncovered clues about the interaction between RBIV infection and the red blood cells or blood of rock bream: i) The rock bream administered with poly (I:C) demonstrated significantly elevated expression levels of interferon regulatory factor 3 (IRF3), interferon-stimulated gene 15 (ISG15) and protein kinase RNA-activated (PKR) genes in the blood; however, there was no significant up-regulation observed in the hematopoietic organ (spleen and kidney) (Jung and Jung 2017d), ii) Furthermore, the highest mRNA expression levels for the MHC class I gene were in the entire blood of the rock bream and was the lowest in the kidney and brain (Nikapitiya et al., 2014), iii) In RBIV-infected individuals, a total of 318 proteins were significantly regulated in the RBCs; 183 proteins were up-regulated and 135 proteins were down-regulated.These findings indicate that RBCs in rock bream infected with RBIV may induce a response to the infection.This response is characterized by the induction of apoptosis, MHC class I, and spliceosome pathways and downregulation of ISG15 antiviral mechanisms (Jung et al., 2019).Thus, RBCs may have a critical immune defense mechanism in rock bream.Massive deaths, occur if RBCs are unable to recover their function because of viral infections.
Whether blood function-related gene alterations occur owing to RBIV infection was investigated in the present study.The mRNA expression levels of the subunits of hemoglobin (Hb) protein, Hb-α and Hb-β in the RBCs, gradually decreased below 1-fold from 4 to 17 dpi.CLNK is an adaptor protein that regulates immune receptor signaling and signal transduction by interacting with various signaling proteins (Hakeim et al., 2021).The mRNA expression of CLNK increased approximately 2-fold at 1 dpi and which decreased after 4 dpi.Additionally, the mRNA expression of GATA, which plays a role in erythropoiesis (Fujiwara 2017), increased after 1 dpi and then decreased after 4 dpi.These results indicate that blood-related genes are unreactive in the early stages of RBIV infection, suggesting that RBC production is insufficient to activate immune response signals, especially hemoglobin synthesis, which is suppressed in RBIV-infected rock bream.Of note, the highest and lowest mRNA expression levels of CLNK and GATA were observed in the kidneys and RBCs, respectively.This indicates that various blood-related immune responses occur between hematopoietic organs (kidneys) and RBCs during RBIV infection.These findings may improve the current understanding of RBIV and host interactions and aid in the clarification of RBIV pathogenesis.
Analysis of blood parameters plays a key role in determining the health of an organism and its physiological status.Blood sampling and analyses are routinely performed in human, veterinary and physiological studies.Anemia in fish is the lack of oxygen-transporting red blood cells in the body (Hoffbrand et al., 2001) and has been described in several teleost fish (Witeska, 2015).The importance and relevance of investigating anemia can be explained as follows: i) The most apparent symptom of anemia in a fish is pale gills, resulting from a lack of red blood cell flow, ii) Anemia is an abnormal condition where the body lacks red blood cells, making it incapable of adequately transporting oxygen, iii) In blood tests, this appears as low hemoglobin concentration, the primary protein in red blood cells, iv) Low red blood cell counts may be observed when the body's HGB does not function; it can not produce enough hemoglobin when it breaks down RBCs too rapidly or lacks RBCs, vi) The symptoms of anemia, arise considering essential organs do not receive enough oxygen to function correctly.Thus, anemia is frequently used to diagnose the damage caused by pathogenic infections.
To the best of our knowledge, no published information is available to assess the nature of this recent health challenge, and no current hematological reference values are available for normally farmed rock bream or Megalocytivirus.
The most apparent symptom of anemia in RBIV-infected rock bream is pale gills, which may result from a lack of red blood cell flow.This was evident from the anemia-related indicators that showed a gradual increase in virus copy numbers (approximately 10 5 /mg) which was accompanied by a gradual decrease in HGB, RBC and HCT levels between 14 and 17 dpi.This indicated that anemia was induced, followed by a decrease in RBC count, HGB concentration and HCT level, ultimately leading to death.These findings may indicate that the hemoglobin ((<10 (g/dl)), RBC ((<2.0 (10 12 /L)) and HCT ((<20 ( %)) results in severe anemia when virus replication reached its peak.
In summary, this study demonstrated that blood-related indicators such as hemoglobin (α and β), CLNK and GATA may not be activated in RBIV infection and that their immune responses are critical factors for RBIV anemia.The results of this study provide a list of new hematological targets that may be used to further explore RBIV disease mechanisms.The role of blood or RBCs remains unclear, indicating that the absence of studies focusing on rock bream blood severely hampers elucidation of the pathogenic mechanisms underlying RBIV infection.More detailed studies are, required to obtain a better understanding of the morphological changes in blood cells (white blood cells, lymphocytes, monocytes, granulocytes, thrombocytes and RBCs) and the functions related to immune cell defense interactions in rock bream during RBIV infection.

Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Fig. 2 .
Fig. 2. Dead rock bream exhibiting typical signs of RBIV infection.(A) enlarged spleen.(B) enlarged basophilic cells in the spleen imprint.Triangle and round arrows indicate enlarged basophilic and red blood cells, respectively.

4.
Complete blood cell count (CBC) examination of RBC, MCV, RDW, HGB, MCH, MCHC and HCT in the blood after injection with RBIV (10 6 MCP gene copies/ 100 μl/fish) at 23 • C. Bars indicate the standard error (SE) of the mean of five individuals; *p<0.05,**p<0.01 and ***p<0.001.The viral copy number in the spleen at each sampling point is plotted alongside for reference.

Fig. 5 .
Fig. 5. Classification of anemia severity in RBIV-infected rock bream.Red and yellowish-brown boxes indicate high and low grade values, respectively.

Fig.
Fig. Biochemical examination of TP, GGT, AST, ALT, ALP, AMY, CREA, BUN, BUN/CREA, TC and TG in the blood after injection with RBIV (10 6 MCP gene copies/ 100 μl/fish) at 23 • C. Bars standard error (SE) of the mean of five individuals; *p<0.05,**p<0.01 and ***p<0.001.The viral copy number in the spleen at each sampling point is plotted alongside for reference.

Fig. 7 .
Fig. 7. Classification of hamotological parameter severity in RBIV-infected rock bream.Red and yellowish-brown boxes indicate high and low grade values, respectively.

Table 1
Primers used in this study.

Table 2
Anemia severity classification for the RBIV-infected rock bream.

Table 3
Hematological parameter classification for the RBIV-infected rock bream.