Fish Red Blood Cells Modulate Immune Genes in Response to Bacterial Inclusion Bodies Made of TNFα and a G-VHSV Fragment

Fish Red-Blood Cells (RBCs) are nucleated cells that can modulate the expression of different sets of genes in response to stimuli, playing an active role in the homeostasis of the fish immune system. Nowadays, vaccination is one of the main ways to control and prevent viral diseases in aquaculture and the development of novel vaccination approaches is a focal point in fish vaccinology. One of the strategies that has recently emerged is the use of nanostructured recombinant proteins. Nanostructured cytokines have already been shown to immunostimulate and protect fish against bacterial infections. To explore the role of RBCs in the immune response to two nanostructured recombinant proteins, TNFα and a G-VHSV protein fragment, we performed different in vitro and in vivo studies. We show for the first time that rainbow trout RBCs are able to endocytose nanostructured TNFα and G-VHSV protein fragment in vitro, despite not being phagocytic cells, and in response to nanostructured TNFα and G-VHSV fragment, the expression of different immune genes could be modulated.


INTRODUCTION
Fish red blood cells (RBCs) are nucleated cells that contain organelles in their cytoplasm unlike those of mammals (1). Apart from their well-known role in gas exchange, recently a set of new biological roles for nucleated RBCs related to the immune response have been reported. Nucleated RBCs are able to phagocytose and act as antigen presenting cells (2,3). They can respond to different pathogen associated molecular patterns (PAMPs), modulate leukocyte activity, release cytokine-like factors (4,5) and lately they have been implicated in the response to viral infections [reviewed in Nombela and Ortega-Villaizan (6)].Considering all of these findings, the potential role of RBCs in the immune system of fish takes on a new, interesting perspective.
To date, one of the best strategies for preventing and controlling viral diseases in aquaculture is DNA vaccination. However, it remains unclear which mechanisms are responsible for this protection (7). The search for new, safe and effective vaccines has become a priority in this field. Among fish viral diseases, viral hemorrhagic septicaemia (VHS) is a lethal infectious fish disease caused by viral hemorrhagic septicaemia virus (VHSV), which affects over 50 species of fish, freshwater and marine, in the northern hemisphere (8).
As an alternative to overcome the safety problems associated to live attenuated or DNA vaccines, bacterial inclusion bodies (IBs) nanostructured recombinant proteins have been presented as a new option for vaccination (9). IBs are per se strong stimulants of the fish immune system and have a set of characteristics which make them an attractive alternative: they are mechanically stable, production is scalable and costeffective, they are non-toxic biomaterials and are composed of recombinant proteins. The latter means they are an adaptable prototype, which could be a good platform for vaccination against a wide range of diseases (9,10). Such nanostructured recombinant proteins have already been shown to protect fish against bacterial infection (9).
In this paper, we show for the first time the response of rainbow trout RBCs in vitro and in vivo to two different nanostructured recombinant proteins, recombinant rainbow trout tumor necrosis factor alpha protein (IB TNFα ) and recombinant fragment 16 of the glycoprotein G of VHSV (11) (IB frg16G−VHSV ). In response to recombinant protein IBs, RBCs were able to modulate the expression of interferon related genes, the myxovirus resistance (mx) gene and genes related to antigen presentation (

MATERIALS AND METHODS
Production of IBs, Purification, Quantification , and Fluorescent Labeling Nanostructured proteins were produced in E. coli following the method described in Torrealba et al. (9) and Thwaite et al. (12). In short, E. coli transformed with the plasmid of interest was cultured in LB with the appropriate antibiotic and recombinant protein expression was induced at OD 550nm 0.5-0.8 with 1 mM IPTG (Panreac, Barcelona, Spain). IBs were isolated after 3 h additional incubation at 37 • C via enzymatic and mechanical disruption of the cells according to Torrealba et al. (10), followed by sterility monitoring (12). Purified nanoparticles, named here IB frg16G−VHSV , IB TNFα and IB iRFP [an inclusion body made of a non-immunogenic phytochrome-based near infra-red fluorescent protein (iRFP) with the excitation/emission maxima at 690/713 nm (13)], were stored at −80 • C until use. Quantification was performed by western blot using an anti-His-tag antibody (Genscript, Piscataway, NJ, USA) and calculating the protein concentration from a standard curve using Quantity One software (Biorad, Hercules, CA, the USA). For flow cytometry or confocal microscopy, IB frg16G−VHSV and IB TNFα were conjugated with fluorescent Atto-488 NHS ester (Sigma-Aldrich) following manufacturer's instructions.

Animals
Juvenile rainbow trout (Oncorhynchus mykiss) were obtained from a commercial farm (Piszolla S.L., Cimballa Fish Farm, Zaragoza, Spain), and maintained at the University Miguel Hernandez (UMH) facilities at 14 • C, fed daily with a commercial diet (Skretting, Burgos, Spain). Prior to experiments, fish ware acclimatized to laboratory conditions for 2 weeks. Separately, adult rainbow trout were maintained at the Universitat Autònoma de Barcelona (UAB) at 17 ± 1 • C, fed daily with a commercial diet. The number of individuals used in each experiment is indicated by an "n" in each figure legend.

Uptake of IB TNFα and IB frg16G-VHSV by RBCs
RBCs cultures were treated with fluorescent IB TNFα or IB frg16G−VHSV at different concentrations and uptake was analyzed by flow cytometry using a FACSCanto TM cytometer (BD Biosciences, Madrid, Spain) (10.000 total events), at different times post-treatment. For dose-response evaluation, IBs at concentrations of 10, 20 and 50 µg/mL were added to RBCs cultures for 24 h. For time-course experiments, RBCs were treated with 80 µg/mL IB TNFα or 160 µg/mL IB frg16G−VHSV for 6, 24 and 48 h. After incubation with IBs, the medium was removed and RBCs were washed with phosphate-buffered saline (PBS). RBCs were then resuspended in 200 µL of RPMI 2% FBS for flow cytometry analysis.
In addition, confocal microscopy was performed to evaluate the uptake of IBs by RBCs. RBCs were incubated with 80 µg /mL of IB TNFα or 160 µg /mL of IB frg16G−VHSV for 24 h. Then, medium was removed and RBCs were washed as indicated above. The RBC nucleus was labeled with 10 µg/mL Hoechst (Sigma-Aldrich) and RBC membrane was stained with 5 µg/mL of CellMask (Thermo Fischer Scientific). Images were taken with a Zeiss LSM 700 microscope (Zeiss, Oberkochen, Germany) and analyzed with Imaris Software v8.2.1 (Bitplane, Zurich, Switzerland).
RBCs Immune Response After in vitro Treatment With IB TNFα or IB frg16G-VHSV RBCs were treated in vitro with 50 µg/mL of each IB for 24 h. IB iRFP was used as a control. After treatment, RBCs were resuspended in TRK lysis buffer (Omega Bio-Tek Inc., Norcross, GA, USA) and stored at −80 • C until RNA extraction.
RBCs Immune Response After in vivo Treatment With IB TNFα or IB frg16G-vhsv Juvenile rainbow trout (15-20 g) were treated by intravenous injection in caudal vein with 50 µL of IBs (5.5 mg/kg) or 50 µL of PBS. At 24 and 48 h post-injection fish were sacrificed by overexposure to tricaine. Peripheral blood was sampled as described above and resuspended in RPMI 10% FBS. Then, RBCs were Ficoll-purified as explained above. Purified RBCs were either resuspended in TRK lysis buffer and stored at −80 • C until RNA extraction or fixed for immunofluorescence and flow cytometry, as described below.
In order to track the presence of IBs in vivo, IB TNFα was monitored in peripheral blood and head kidney from IB TNFα intravenously injected in caudal vein of rainbow trout by means of fluorescent microscopy using IN Cell Analyzer 6,000 Cell Imaging system (GE Healthcare, Little Chalfont, UK). Blood was extracted 3 h post-injection as described above. Head kidney was aseptically removed, placed in 24 well plates with RPMI 10%

Gene
Forward primer Reverse primer Probe Reference or accession number FBS and disaggregated with a Pasteur pipette and passed through a Falcon 40 µm nylon cell strainer (BD Biosciencies) using a plunger of a 5 ml syringe.

RNA Isolation, cDNA Synthesis, RT-qPCR, and Gene Expression Analysis
RBCs total RNA was extracted as previously described (14) using E.Z.N.A. R Total RNA Kit (Omega Bio-Tek Inc.). DNAse treatment was performed in order to eliminate residual genomic DNA using TURBO TM DNase (Ambion, Thermo Fischer Scientific Inc.). Then cDNA synthesis and RT-qPCR was performed as described in Nombela et al (14). Primers and probes used are listed in Table 1. Gene expression was analyzed by means of the 2 − Ct or 2 − Ct (23) using 18S rRNA (Applied Biosystems, Thermo Fischer Scientific Inc.) as endogenous gene. Principal component analysis (PCA) and clustering heatmap of immune-gene expression data (2 − Ct or 2 − Ct ) were performed using Clustvis software (24). For PCA, unit variance scaling was applied to rows and singular value decomposition (SVD) with imputation was used to calculate principal components. For clustering heatmap, columns were collapsed by taking mean inside each group, rows were centered, and unit variance scaling was applied to rows; then, imputation was used for missing value estimation; and, both rows and columns were clustered using correlation distance and average linkage.

Software and Statistics
Graphpad Prism 6.01 (www.graphpad.com) was used for statistics and graphic representation. Statistic tests and Pvalues associated with graphics are indicated in each assay. Flow cytometry data was processed and analyzed using Flowing Software 2.5.1 (www.flowingsoftware.com/). Principal component analysis (PCA) and clustering of gene expression analysis was performed using ClustVis software (https://biit.cs. ut.ee/clustvis/) (24).

Uptake of IB TNFα and IB frg16G-VHSV by RBCs
In order to evaluate the interaction between RBCs and IBs, we performed a dose-response and time-course evaluation by means of flow cytometry. According to our results, all IB concentrations assayed showed uptake or attachment to RBCs, which increased with IB concentration (Figure 1A). The percentage of IB positive cells ranged from 5 to 7% at 50 µg/mL after 24 h incubation. Time course evaluation at 6, 24, and 48 h showed no differences in IB load in RBCs ( Figure 1B) indicating that the maximum IB internalization or attachment occurred by 6 h of incubation. However, the time course was carried out with a higher dose and up to 17% of fluorescent positive cells were detected. This was maximum percentage uptake achieved under our experimental conditions. The level of uptake of IB TNFα by RBCs was observed to be higher than IB frg16G−VHSV when comparing the same concentration of both IBs ( Figure 1A). IB uptake was confirmed by confocal 3D images, which showed the internalization of IB TNFα (Figure 2A) and IB frg16G−VHSV (Figure 2B) in the cytosol of RBCs. The presence of IB TNFα in RBCs was monitored in vivo in peripheral blood and head kidney cells by fluorescent microscopy using intravenously injected IB TNFα . In blood, few RBCs were found to carry the IB TNFα (data not shown); however, RBCs carrying IB TNFα were easily found in head kidney cells extracts (Figure 2C).

Immune Response of RBCs Induced After Exposure to IB TNFα or IB frg16G-VHSV in vitro
To explore the immune response triggered by IBs in RBCs in vitro, RBCs were treated with 50 µg/mL of IB TNFα , IB frg16G−VHSV or IB iRFP and RNA was extracted at 24 h post-treatment. IB TNFα tended to down-regulate the genes tested in RBCs at 24 h post-treatment. This down-regulation was statistically significant in genes related to antigen presentation (cd83, mhcI) and antioxidant gene gstp1. On the other hand, only the antioxidant trx gene was significantly up-regulated in IB frg16G−VHSV treated RBCs at 24 h post-treatment ( Table 2).
In order to analyse the gene expression of RBCs in response to each treatment as a whole, multivariate analyses of the gene expression data matrix were performed. A principal component analysis (PCA) plot of the gene expression profile showed a differentiated population of RBCs treated with IB TNFα or IB frg16G−VHSV compared to IB iRFP (Figure 3A). This is also appreciable in the clustering heatmap (Figure 3B)

Immune-Gene and Protein Expression Modulation in RBCs From Peripheral Blood After in vivo Treatment With IB TNFα or IB frg16G-VHSV
Rainbow trout were intravenously injected to evaluate the immune response triggered by IBs in RBCs of peripheral blood in vivo. RBCs were sampled at 24 and 48 h post-injection. In general, the results showed, as in vitro, a down-regulatory trend in the gene expression of IB TNFα treated individuals compared to IB iRFP treated individuals. It should be noted that cd83 was significantly down-regulated at 24 h post-injection (Figure 4A), as occurred in vitro. On the other hand, il6 was significantly up-regulated at 24 h post-injection. Further, tlr9, ifn1, il1β, il2, mhcII and nkef genes were significantly down-regulated at 48 h post-injection ( Figure 4B). In contrast, IB frg16G−VHSV treated individuals showed an up-regulatory trend at both 24 and 48 h post-injection, compared to IB iRFP , with significant up-regulation of cytokines il2 and il6, and antioxidant gene nkef at 24 h post-injection, and of tlr3, interferon inducible  Table S1). However, mx gene appeared down-regulated at 24 h post-injection. Separately, most of the genes were upregulated with all the treatments in comparison with PBSinjection.
The gene expression profile PCA plot depicted differentiated populations for RBCs from individuals treated with IB TNFα or IB frg16G−VHSV compared to IB iRFP (Figures 5A, 6A, for 24 and 48 h post-injection, respectively), which was also observed in the clustering heatmap (Figures 5B, 6B, for 24 and 48 h post-injection respectively). In addition, at 48 h postinjection, MX and IL8 protein levels, evaluated by means of flow cytometry, showed an increment, but not statistically significant, in MX (Figures 7A,C) and IL8 (Figures 7B,D) in RBCs from rainbow trout treated with IB frg16G−VHSV in relation to PBS-injected or the other IBs assayed. This result correlates with the mx gene expression at 48 h in vivo. On the other hand, the protein levels of MX and IL8 in RBCs from IB TNFα treated rainbow trout were slightly lower than IB iRFP and PBS-injected individuals (only showing statistical significance for MX between IB TNFα and IB iRFP treatments), which is consistent with the down-regulatory trend observed in IB TNFα treated RBCs in vivo and in vitro at the transcriptional level. Moreover, in whole peripheral blood, a similar tendency was observed in MX protein expression, although more pronounced in this case. Note, however, for IL8 protein levels, we did not observe any difference among groups (Figures 8A,B).

DISCUSSION
Recently, IBs have been reported as new alternatives in fish prophylaxis as immunostimulants or adjuvants (10), thus potentially serving as a new platform for vaccine delivery. The uptake of IBs has been reported in rainbow trout macrophages (RT-HKM) and zebrafish liver cells (ZFL). In both cell types IBs made with cytokines stimulate the innate immune response (9). Moreover, IBs made with fish viral antigens have evoked an anti-viral innate immune response in ZFL and RT-HKM (12). However, the immune response of nucleated RBCs to nanostructured cytokine or viral antigen IBs has not been tested until now. Nucleated RBCs are the main cell in the blood and recently have been endorsed as immune cells mediators (6,28). In this work we show that the uptake or attachment of IBs by rainbow trout RBCs occurred in approximately 7% of cells counted. This contrasts to the near 40% and 80% reported for ZFL and RT-HKM, respectively, at same concentration (50 µg/mL) of IB TNFα (9). RBCs endocytosed both the IBs tested here reaching their maximum level at 6 h post-treatment, in contrast to RT-HKM and ZFL cell lines, which reached their maximum uptake at 24 h post-treatment in vitro (10). Besides, monitorization of IB TNFα in vivo demonstrated its presence on/in RBCs from head-kidney 3 h post-injection. The mechanism by which RBCs endocytose IBs is unknown. It may occur via the micropinocytosis endocytic pathway, as proposed for mammalian cells (29).
Significantly, with this work, we add to the growing body of data demonstrating nucleated RBCs can exercise a role in the immune response. RBCs are able to respond to virus (6,14,30), produce cytokines when exposed to stimuli (17), and endocytose pathogens (2). Here we show for the first time rainbow trout RBCs evoke an immune response to IBs made of cytokine TNFα and viral protein frg16G-VHSV in vitro and in vivo. We demonstrate this response at protein and transcript level. Rainbow trout Ficoll-purified RBCs treated with IBs in vitro and RBCs Ficoll-purified from blood extracts from IB-intravenously injected individuals modulated the expression of genes related to antigen presentation, cytokines and other genes involved in the immune response. PCA clearly clustered the RBCs' immune-gene expression profiles for each treatment.  As regards TNFα, RBCs from IB TNFα -treated rainbow trout individuals showed a down-regulatory trend for genes related to TNFα signaling such as tlr9, tnfα, il1β, il12β, and il2 genes transcripts, in vivo, at 24 and 48 h post-injection, compared to fish injected with the non-immunogenic protein IB iRFP . It is known that TNFα is a cytokine involved in the regulation of immune cells and inflammation. It is mainly produced by monocytes and macrophages along with additional producers including B and T lymphocytes, NK cells, polymorphonuclear leukocytes, and eosinophils in response to bacterial toxins, inflammatory products, and other invasive stimuli (31). Recently, nucleated RBCs have been also reported to modulate TNFα  protein in response to IPNV virus exposure (30). Here we observed that RBCs exposed to IB TNFα down-regulated the inflammatory response at 24 and 48 h post-treatment. TNFα is a pleiotropic cytokine with a diverse range of biological actions. TNF family members are known to represent a "doubleedged sword, " having both beneficial and detrimental activities  4). Mann-Whitney test was performed between each condition and control cells (treated with PBS or IB iRFP ). #P-value < 0.05, compared to PBS. (32). Systemic exposure to recombinant TNFα would cause a shock similar to septic shock syndrome (31). Further, TNFα inhibition of IFNγ-induced IL12 production exerts mechanisms by which TNFα and IL12 cytokines can elicit anti-inflammatory and repair functions, tightly modulated by positive and negative feedback signals for optimal immunity without manifested inflammation (33). Another important observation is that fish recombinant TNFα has been reported to regulate the expression of endothelial cells TLRs, including TLR9, but had negligible effects on macrophages (34). Therefore, taking into account that nucleated RBCs are the most abundant cell type in peripheral blood, it would make sense that RBCs were equipped to modulate inflammation in response to a systemic exposure to TNFα. Moreover, in the IB TNFα injected group, genes related to antigen presentation, cd83 and mhcII, were also down-regulated at 24 and 48 h, respectively. As well, RBCs treated in vitro with IB TNFα down-regulated the expression of cd83 and mhcI 24 h posttreatment. TNFα has been reported to modulate IFNγ-induced MHC class II expression in a cell type-specific mode (35). Therefore, TNFα treatment augments or blocks MHC class II induction depending on the cell type and cellular differentiation state (35). mhcII and cd83 gene expression has been previously reported for rainbow trout RBCs (3,36) and chicken RBCs (37). However, this is the first report that shows the regulation of cd83 and mhcII gene transcripts in response to an immunostimulant.
On the other hand, RBCs from rainbow trout injected with IB frg16G−VHSV showed an up-regulatory trend for most of the genes, specifically interleukins il2 and il6, and antioxidant enzyme nkef were significantly up-regulated, compared to IB iRFP , at 24 h post-injection. This is probably due to the effort of RBCs to compensate the inflammatory response triggered after the first treatment stimulus. Then, 48 h post-injection, the Type 1 IFN and antigen presentation responses were increased, since tlr3, mx, cd83, and mhcII genes transcripts were significantly up-regulated, compared to IB iRFP . MX protein production was consistent with gene expression levels.
G-VHSV is known to induce the expression of ifn1 and mx (25,38,39). Peptides derived from G-VHSV have also demonstrated their efficacy to induce type 1 IFN response (25,26,39). It is also noteworthy that IB frg16G−VHSV triggered the up-regulation of mhcII and cd83 gene expression in rainbow trout RBCs, thus endowing them the characteristics of antigen presenting cells (APCs). CD83 and MHCII are principally produced by professional APCs to process antigens and induce T cell priming. However, recently, the concept of non-professional APCs is emerging (40). These atypical APCs up-regulate the expression of MHC and related molecules under certain stimuli. However, there is not enough evidence about their functionality priming T cells (40).
Bacterial lipopolysaccharide has been reported to stimulate the innate immune response of RBCs in vitro (28). Bacterial IBs, which contain remnants of endotoxin, are therefore considered immunostimulants per se (41), which is shown by the global increment in the immune response of RBCs from rainbow trout injected with IB iRFP compared to PBS-injection. This, added to the utilization of IBs as delivery platforms to administrate cytokines, coadjuvants, or antigens, makes them a good candidate for future vaccines. In this context, RBCs have shown their ability to mount or modulate and immune-response to IBs made of cytokine TNFα and the viral protein frg16G-VHSV.
All these considerations provide a new perspective on the role and potential use of RBCs. Given the large amount of RBCs in the organism and their rapid distribution throughout the body they could be a promising target cell for the presentation or delivery of IBs or other types of vaccine carriers.  and by grants from the Spanish Ministry of Science, European commission and AGAUR funds to NR (AGL2015-65129-R MINECO/FEDER and 2014SGR-345 AGAUR). RT holds a pre-doctoral scholarship from AGAUR (Spain).