First report of invertebrate Mx: cloning, characterization and expression analysis of Mx cDNA in disk abalone (Haliotis discus discus)

Myxovirus resistance (Mx) protein is one of the most studied antiviral proteins. It is induced by the type I interferon system (IFN alpha/beta) in various vertebrates, but its expression has not been identified or characterized in mollusks or other multi-cellular invertebrates to date. In this study, we isolated the Mx gene from a disk abalone (Haliotis discus discus) normalized cDNA library. Mx cDNA was sequenced, cloned and compared to other known Mx proteins. The full-length 1664 bp of abalone Mx cDNA contained a 1533-bp open reading frame that codes for 511 amino acids. Within the coding sequence of abalone Mx, characteristic features were found, such as a tripartite guanosine-5'-triphosphate (GTP)-binding motif and a dynamin family signature. In addition, leucine residues in the C-terminal region displayed a special leucine domain at L(468), L(475), L(489) and L(510), suggesting that abalone Mx may have a similar oligomerization function as other leucine zipper motifs. Abalone Mx protein exhibited 44% amino acid similarity with channel catfish Mx1, rainbow trout Mx2 and Atlantic halibut Mx. Abalones were injected intramuscularly with the known IFN inducer poly I:C and RT-PCR was performed for Mx mRNA analysis. The results showed enhanced Mx expression in abalone gill and digestive tissues 24h as well as 48 h after injection of poly I:C. Mx mRNA was expressed in gill, digestive gland, mantle and foot tissues in healthy abalone, suggesting that the basal level of Mx expressed is tissue-specific. There is no known Mx protein closely related to abalone Mx according to phylogenetic analysis. Abalone Mx may have diverged from a common gene ancestor of fish and mammalian Mx proteins, since abalone Mx showed high similarity in terms of conserved tripartite GTP-binding, dynamin family signature motifs and poly I:C enhancement of Mx mRNA expression.     

Quantification of Atlantic salmon type-I interferon using an Mx1 promoter reporter gene assay

We here describe an assay for the detection of interferon-like activity in Atlantic salmon based on the transient transfection of chinook salmon embryo cells (CHSE-214 cells) with a rainbow trout Mx1 promoter linked to a luciferase reporter. A beta-galactosidase gene under the control of a constitutively expressed beta-actin promoter was used as a transfection standard, and luciferase and beta gal expression were measured by a commercially available kit. Interferon containing supernatants from poly I:C- or CpG-stimulated leucocytes added to transfected CHSE-cells induced high luciferase expression (>60-fold induction compared to supernatants from non-stimulated cells). There was no response to supernatants from LPS- and ConA/PMA-stimulated leucocytes, demonstrating the specificity for type I interferon-like activity. Duplicate samples analysed using a cell protection assay for detection of antiviral activity correlated well with levels obtained by the Mx1 promoter reporter gene assay (R2=0.97), confirming the reporter assay as a reliable substitute for the standard antiviral assay. The Mx reporter gene assay also has advantages in terms of sensitivity, high dynamic range and reliability over the conventional cell protection assay.     

An antiviral state induced in Chinook salmon embryo cells (CHSE-214) by transfection with the double-stranded RNA poly I:C

Type I interferons (IFN alpha and beta) convert vertebrate cells into an antiviral state by inducing expression of proteins that inhibit virus replication. In humans and mice, Mx proteins constitute one family of interferon-induced antiviral proteins. Mx genes have recently been cloned from Atlantic salmon and rainbow trout. Moreover, double-stranded RNA (dsRNA) and type I IFN-like activity have been shown to induce Mx protein in salmonid cells. Chinook salmon embryo cells (CHSE-214 cells) have been suggested to have a defect in the IFN-system because the dsRNA polyinosinic polycytidylic acid (poly I:C) failed to induce an antiviral state in the cells. We have studied this phenomenon more closely in the present work. CHSE-214 cells were either transfected with poly I:C or incubated with poly I:C without transfection reagent. The cells were then studied for Mx protein expression and protection against infectious pancreatic necrosis virus (IPNV) infection. The results showed that cells transfected with poly I:C were protected from IPNV infection, whilst cells incubated with poly I:C were not protected. Cells transfected with the double-stranded DNA poly dI:dC were also not protected against IPNV. Mx protein was expressed in CHSE-214 cells upon transfection with poly I:C, but not after incubation with poly I:C alone. Stimulation of CHSE-214 cells with supernatants from cells transfected with poly I:C, induced protection against IPNV, indicating production of type I IFN-like activity. These results suggest that CHSE-214 cells in fact are able to produce type I IFN, but may have defects in the mechanisms mediating uptake of poly I:C or may degrade unprotected poly I:C.     

Cloning and characterisation of a channel catfish (Ictalurus punctatus) Mx gene

A 2.5 kb full-length cDNA clone of a channel catfish (Ictalurus punctatus) Mx gene was obtained using RACE (rapid amplification of cDNA ends) polymerase chain reaction (PCR) from RNA extracted from the liver of poly I:C stimulated channel catfish. The gene consists of an open reading frame of 1905 nucleotides encoding a 635 amino acid protein. The predicted protein is 72.5 kDa and contains the dynamin family signature, a tripartite GTP binding motif and a leucine zipper, characteristic of all known Mx proteins. The catfish Mx protein exhibited 79% identity with perch Mx and between 71% and 74% identity with the three Atlantic salmon and the three rainbow trout Mx proteins. Mx mRNA was constitutively expressed in channel catfish ovary (CCO) cells, but in higher quantities in response to poly I:C treatment. Mx was induced in channel catfish following injection with channel catfish virus (CCV) and poly I:C.     

Effect of double-stranded RNA and interferon on the antiviral activity of Atlantic salmon cells against infectious salmon anemia virus and infectious pancreatic necrosis virus

Type I interferons (IFN) establish an antiviral state in vertebrate cells by inducing expression of Mx and other antiviral proteins. We have studied the effect of Atlantic salmon interferon-like activity (AS-IFN) and poly I:C on the Mx protein expression and antiviral activity against infectious salmon anaemia virus (ISAV) and infectious pancreatic necrosis virus (IPNV) in the Atlantic salmon cell lines SHK-1 and TO. The double-stranded RNA poly I:C is an inducer of type I IFN in vertebrates. A cell cytotoxicity assay and measurements of virus yield were used to measure protection of cells against virus infection. Maximal induction of Mx protein in TO and SHK-1 cells occurred 48 h after poly I:C stimulation and 24 h after AS-IFN stimulation. TO cells pretreated with AS-IFN or poly I:C were protected from infection with IPNV 24 to 96 h after stimulation. Poly I:C or AS-IFN induced a minor protection against ISAV infection in SHK-1 cells, but no protection was induced against ISAV in TO cells. Western blot analysis showed that ISAV induced expression of Mx protein in TO and SHK-1 cells whereas IPNV did not induce Mx protein expression. These results suggest that ISAV and IPNV have very different sensitivities to IFN-induced antiviral activity and have developed different strategies to avoid the IFN-system of Atlantic salmon. Moreover, Atlantic salmon Mx protein appears not to inhibit replication of ISAV.     

Kinetics of Mx expression in rainbow trout (Oncorhynchus mykiss) and Atlantic salmon (Salmo salar L.) parr in response to VHS-DNA vaccination

The duration of the Mx mRNA response to an intramuscular injection of the viral haemorrhagic septicaemia virus (VHSV) glycoprotein (G) gene DNA vaccine as well as to the control plasmid was determined in rainbow trout at 14 degrees C over a period of 11 weeks. The Mx response was detectable on day 7, peaked on day 14 and returned to pretreatment levels on day 21 and thereafter. No increase in Mx expression was detectable to the control plasmid. In further experiments, the kinetics of the Mx response were compared in rainbow trout and Atlantic salmon parr kept at 10 degrees C and injected with the DNA vaccine or the synthetic double-stranded RNA, poly I:C. In both species there was a rapid response to poly I:C detectable from day 1, reaching maximum from days 3 to 9 and decreasing to background level by day 12. The peak level and return to background was reached slightly later in salmon. In both species the response to the VHS/DNA vaccine was slower to begin, not being detectable on days 1 and 3, but elevated levels were found on day 6. However, in the salmon parr, the peak level was on day 6 and the signal disappeared by day 12, while in the rainbow trout, the response peaked at day 12 and lasted until day 21. The kinetics of the Mx response to the VHS/DNA vaccine in rainbow trout correlate with the early non-specific protection against VHS in this species following vaccination. It is speculated that the more transient Mx response in Atlantic salmon parr to the DNA vaccine may be related to the innate resistance of salmon to VHS.     

Effects of NV gene knock-out recombinant viral hemorrhagic septicemia virus (VHSV) on Mx gene expression in Epithelioma papulosum cyprini (EPC) cells and olive flounder (Paralichthys olivaceus)

To determine whether the NV gene of viral hemorrhagic septicemia virus (VHSV) is related to the type I interferon response of hosts, expression of Mx gene in Epithelioma papulosum cyprini (EPC) cells and in olive flounder (Paralichthys olivaceus) in response to infection with either wild-type VHSV or recombinant VHSVs (rVHSV-ΔNV-EGFP and rVHSV-wild) was investigated. A reporter vector was constructed for measuring Mx gene expression using olive flounder Mx promoter, in which the reporter Metridia luciferase was designed to be excreted to culture medium to facilitate measurement. The highest increase of luciferase activity was detected from supernatant of cells infected with rVHSV-ΔNV-EGFP. In contrast cells infected with wild-type VHSV showed a slight increase of the luciferase activity. Interestingly, cells infected with rVHSV-wild that has artificially changed nucleotides just before and after the NV gene ORF, also showed highly increased luciferase activity, but the increased amplitude was lower than that by rVHSV-ΔNV-EGFP. These results strongly suggest that the NV protein of VHSV plays an important role in suppressing interferon response in host cells, which provides a condition for the viruses to efficiently proliferate in host cells. In an in vivo experiment, the Mx gene expression in olive flounder challenged with the rVHSV-ΔNV-EGFP was clearly higher than fish challenged with rVHSV-wild or wild-type VHSV, suggesting that lacking of the NV gene in the genome of rVHSV-ΔNV-EGFP brought to strong interferon response that subsequently inhibit viral replication in fish.     

Effects of salmonid fish viruses on Mx gene expression and resistance to single or dual viral infections

We examined the ability of several fish viruses to induce protection against homologous or heterologous viruses in single or double infections, and assessed whether such protection is correlated with innate immunity or expression of the Mx gene. Monolayers of BF2 cells pre-treated with supernatants of brown trout (Salmo trutta L.) macrophage cultures that had been stimulated with either polyinosinic polycytidylic acid (poly I:C) or viruses, such as infectious pancreatic necrosis virus (IPNV), infectious haematopoietic necrosis virus (IHNV) or a mixture of the two, showed varying degrees of protection against viral infections. The virus showing the strongest induction was IPNV, and the antiviral activity against IHNV was also high: around 6 log(10) reduction of virus yield. Consequently, the IPNV-IHNV co-infection yield was also reduced by varying amounts. In vivo, the cumulative mortality observed in the IPNV-IHNV co-infected fish was always less than that in those with a single infection. Stimulation with poly I:C for 7 days significantly reduced cumulative mortality in single-infected fish, but not in the double-infected, in which the IPNV was the only virus isolated from moribund animals. By RT-PCR, Mx was expressed in all the organ samples tested (kidney, liver and spleen) from virus-stimulated fish at 1, 2 and 3 days. By qRT-PCR the extent and timing of Mx expression was shown to differ in the poly I:C and the single or dual viral infections. The highest increase in Mx expression (21.6-fold above basal levels) occurred (after 24 h) in fish infected with the IHNV, and expression remained high until day 7. Mx expression in fish infected with IPNV peaked later, at 2 days post infection, and also remained high until day 7. The dual infection with IPNV-IHNV induced high Mx expression on day 1, which peaked on day 2 and remained high until day 7. These results indicate that activation of the immune system could explain the interference and loss of IHNV in the IPNV-IHNV co-infections.     

Induction of Mx protein by interferon and double-stranded RNA in salmonid cells

Mx protein is one of several antiviral proteins that are induced by the type I interferons (IFN), IFNalpha and beta, in mammals. In this work induction of a 76 kDa Mx protein by double-stranded RNA (dsRNA) or type I IFN-like activity in Atlantic salmon macrophages, Atlantic salmon fibroblast cells (AS cells) and in Chinook salmon embryo cells (CHSE-214) is reported. Type I IFN-like activity was produced by the stimulation of Atlantic salmon macrophages with the synthetic dsRNA polyinosinic polycytidylic acid (poly I:C). A correlation appeared to exist between Mx protein expression and protection against infectious pancreatic necrosis virus (IPNV) induced by IFN in CHSE-214 cells. Several observations in the present work suggest that, as in mammals, the induction of Mx protein by dsRNA in fish cells primarily occurs via induction of type I IFN. First, type I IFN-like activity but not poly I:C, induced Mx protein expression in CHSE-214 cells. These cells apparently lack the ability to produce IFN in response to poly I:C. Second, the putative IFN induced maximal Mx protein expression 48 h earlier than poly I:C in AS cells. Third, the peak expression of Mx protein in macrophages induced by poly I:C occurred after 48 h whereas peak in IFN-like activity was observed by 24 h after addition of poly I:C. The present work supports the notion of using Mx protein as a molecular marker for the production of putative type I IFN in fish.