Functional characterization of TNF-α in species other than mammalian vertebrates is limited, and TNF-α has been studied in a limited number of fish species, primarily in vitro using recombinant proteins. Studies on TNF-α from different fish species so far pointed to several inconsistencies, in particular with respect to some receptor-mediated activities of fish TNF-α, such as the ability to directly activate phagocytes. In the present study a comprehensive analysis of in vitro as well as in vivo biological activities of two isoforms of carp TNF-α was performed. Our results show that carp TNF-α directly primes carp phagocytes and indirectly promotes typical receptor-mediated activities such as phagocyte activation by acting via endothelial cells. Additionally, for the first time in nonmammalian vertebrate species, the lectin-like activity of fish TNF-α homologs was investigated. Our results show an evolutionary conservation of function of this receptor-independent activity of TNF-α not only in cyprinid fish, but also in perciform and salmonid fish. The role of TNF-α in vivo, during infections of carp with the blood parasite Trypanoplasma borreli, was examined using three fundamentally different but complementary approaches: (1) inhibition of TNF-α expression, (2) overexpression of TNF-α, and (3) inhibition of shedding of membrane-bound TNF-α. Our results show that, also in fish, a tight regulation of TNF-α expression is important, since depletion or excess of TNF-α can make an important difference to survival of infection. Finally, we demonstrate a crucial protective role for membrane-bound TNF-α, which has a yet unexploited function in fish.

Tumor necrosis factor-α is a pleiotropic cytokine and a major regulator of leukocyte trafficking and inflammation (1). Under most conditions monocytes/macrophages are the major producers of TNF-α, but in response to specific stimuli such as polyclonal activation, PMA, or LPS stimulation, also T and B lymphocytes, NK cells, endothelial cells, and other immune cells can produce TNF-α (2, 3, 4, 5, 6).

So far, TNF-α homologs have been identified only in the mammalian warm-blooded vertebrates. In birds, neither TNF-α nor lymphotoxin-α has been identified (7, 8). In amphibians, only very recently the first TNF-α sequence was published (9), with a limited functional characterization. In contrast, TNF-α homologs have been reported for at least 10 different teleost fish species (10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23) with two or sometimes three isoforms of TNF-α found in several fish species, including the common carp (Cyprinus carpio) (11, 13, 14, 16, 17, 18, 22). In vitro studies using recombinant proteins have shown that fish TNF-α is active at concentrations at least 1000 times higher than its mammalian counterpart (20, 21, 22, 24, 25, 26, 27) and have evidenced crucial differences between TNF-α of cold- and warm-blooded vertebrates. Furthermore, activities of fish TNF-α appear species-specific in particular with respect to the ability of fish TNF-α to activate phagocytes directly. For example, both in rainbow trout (Oncorhynchus mykiss) (24) and goldfish (Carassius auratus) (22), TNF-α stimulates phagocyte activity in vitro, albeit at different concentrations. In contrast, seabream (Sparus aurata) TNF-α is a potent activator of endothelial cells but does not activate phagocytes directly (27). Instead, supernatant from seabream endothelial cells treated with TNF-α promoted the adhesion, migration, and activation of leukocytes. This is in line with receptor-mediated functions of TNF-α in mammalian vertebrates, where stimulation of endothelial cells with TNF-α increases the expression of adhesion molecules and the synthesis of chemotactic mediators such as IL-8. Thereby, TNF-α promotes leukocyte recruitment and activation and contributes to the inflammatory process. In the present study we show that carp (C. carpio) TNF-α primes but does not directly activate phagocytes and thereby promotes phagocyte activation, indirectly, via stimulation of endothelial cells.

TNF-α fulfills important functions in host-pathogen interactions and is required for protective immunity against intracellular bacteria and fungi but also against intracellular and extracellular protozoan parasites (28, 29). An additional, but perhaps less well-known activity of TNF-α, is its lectin-like activity first recognized in 1988 (30) when TNF-α was reported to directly interact with specific oligosaccharides through its lectin-like domain. The lectin domain is located at the top of the pyramid-shaped TNF-α molecule, also referred to as TIP domain, and is spatially and functionally distinct from the receptor binding region. TNF-α binds through its lectin-like domain to mannose moieties present in the flagellar pocket of some African trypanosomes and can cause direct parasite lysis (31, 32, 33, 34). The lectin-like activity of TNF-α, a yet unexamined receptor-independent function of fish TNF-α, was investigated in a trypanolytic assay in vitro. Our results clearly show that not only carp and zebrafish (Danio rerio) (both Cyprinidae), but also rainbow trout (Salmonidae) and seabream (Perciformes) TNF-α directly lyse Trypanosoma brucei, indicating that the lectin-like activity of TNF-α is evolutionary conserved.

Studies with TNF-α-deficient knockout mice, with transgenic mice overexpressing TNF-α, or studies administrating recombinant TNF-α all pointed out a dual role of TNF-α in immune protection as well as pathology (35). The “double-edged sword” role of TNF-α has most clearly been shown in studies on malaria and trypanosome infections where TNF-α was identified as a key mediator in both control of parasitemia as well as infection-associated pathology (28, 29, 36, 37, 38). In fish, due to a general lack of knockout and transgenic animals, the function of TNF-α in vivo is largely unknown. Only very recently, by the use of TNFR1 morphant zebrafish, TNF-α signaling has been shown to mediate resistance to mycobacteria by inhibiting bacterial growth and preventing macrophage death (39). To further investigate the in vivo role played by TNF-α with respect to protection or infection-associated pathology, we used a natural host-parasite infection model. Experimental infection of carp with the extracellular blood parasite Trypanoplasma borreli (Kinetoplastida) presents most of the pathological features associated with trypanosome infections in mammalian vertebrates such as anemia, splenomegaly, polyclonal B cell activation, as well as extremely elevated serum nitrite levels and associated tissue nitration (40, 41, 42, 43, 44). Similarly, T. borreli infection of carp can be treated with the human anti-trypanosome drug melarsoprol (Arsobal) (44). In the present study we show that inhibition of TNF-α gene expression in T. borreli-infected carp treated with the TNF-α inhibitor pentoxifylline results in extremely high parasitemia and increased mortality, whereas overexpression of TNF-α by injection of plasmid DNA leads to increased mortality, possibly owing to an exacerbation of the inflammatory response. Our results indicate a functional conservation in fish of the dual role of TNF-α in control of parasitemia and in infection-associated pathology.

TNF-α is present as a soluble and membrane-bound (mTNF-α)4 form, with both exhibiting unique and overlapping activities. While solunle TNF-α mainly signals via TNFRI (p55), mTNF-α is the prime activating ligand of TNFR2 (p75) (45). Membrane-bound TNF-α has been shown to be involved in several biological activities, such as cytotoxicity, polyclonal activation of B cells, induction of IL-10 by monocytes, induction of chemokines, and ICAM-1 expression on endothelial cells (46, 47, 48). mTNF-α has been implicated in the control of Listeria monocytogenes (49), mycobacterial infection (50, 51), and possibly in the pathogenesis of experimental cerebral malaria (52). In the present study, we used an inhibitor of TNF-α-converting enzyme (TACE) to investigate the contribution of mTNF-α to the clearance of T. borreli in infected carp. In vitro, treatment of LPS-stimulated carp leukocytes with TACE inhibitor increased the number of cells bearing TNF-α on their surface. In vivo, fish treated with the TACE inhibitor were fully protected and cleared the parasite within days after treatment. Additionally, splenomegaly, plasma nitrite levels, and tissue nitration were all considerably reduced in fish treated with the TACE inhibitor, suggesting a so far unrecognized protective role for membrane-bound TNF-α in fish.

European common carp (Cyprinus carpio carpio L.) were bred and raised in the central fish facility at Wageningen University, The Netherlands, at 23°C in recirculating UV-treated tap water and fed pelleted dry food (Trouvit; Nutreco) daily. R3xR8 carp, which are the offspring of a cross between fish of Hungarian origin (R8 strain) and of Polish origin (R3 strain), were used (53). All experiments were performed with the approval of the animal experimental committee of Wageningen University.

Fish were killed by an overdose of anesthetic (0.3 g/L tricaine methane sulfonate; Crescent Research Chemicals) in aquarium water buffered with 0.6 g/L sodium bicarbonate (Sigma-Aldrich). PBL, head kidney leukocytes (HKL), or enriched phagocyte fractions (macrophages and neutrophilic granulocytes) were isolated as described previously (54, 55). After isolation, cells were adjusted to the appropriate concentration with carp complete RPMI 1640 medium (cRPMI (Invitrogen) adjusted to 280 mOsmol/kg, supplemented with 1.5% (v/v) heat inactivated-pooled carp serum, 2 mM l-glutamine, 100 U/ml penicillin G, and 50 mg/ml streptomycin sulfate).

Isolation of cardiac endothelial cells (EC) was performed as published (56, 57) with some modifications. Briefly, hearts from 9- to 12-mo-old carp were aseptically removed and transferred to 20 ml of ice-cold cRPMI supplemented with 5 IU/ml heparin (Leo Pharmaceutical Products). All subsequent steps were performed at room temperature. Excess blood was removed by slowly injecting into the atrium 10 ml of perfusion buffer (56) supplemented with 5 IU/ml heparin. Atria were dissected and several incisions were made to ensure complete access of the following buffers. Atria were incubated by shaking at 250 rpm for 30 min in carp PBS (cPBS, 150 mM NaCl, 2 mM KCl, 20 mM Na2HPO4, 2 mM KH2PO4) for 5 min in cPBS containing 0.5 mg/ml trypsin and 0.1 mg/ml EDTA and for 30 min in 0.7% saline solution containing 0.5 mg/ml collagenase and 0.7 mg/ml CaCl2. After each incubation step, cells were centrifuged for 5 min at 400 × g. After collagenase treatment, the suspension was transferred to a sterile petri dish and flushed 5–10 times through the jet of a 10-ml syringe. Large fragments were discarded by sedimentation at 1 × g for 8–10 min. The resulting cell suspension was centrifuged for 5 min at 400 × g and cell pellets were resuspended in complete EC medium (RPMI 1640 supplemented with 10% FCS, 2 mM l-glutamine, 100 U/ml penicillin G, 50 mg/ml streptomycin sulfate, and 50 μg/ml gentamicin). The cell suspension obtained from n = 4 atria was seeded in two collagen type I-coated 6-well plates or two 24-well plates (Greiner, catalog nos. 657950 and 662950) and incubated at 27°C in a humidified atmosphere of 5% CO2 to allow cells to adhere. After 24–48 h, cell debris and nonadherent cells were removed by washing twice with complete EC medium and then cultures were left undisturbed.

T. borreli was cloned and characterized by Steinhagen et al. (58) and maintained by syringe passage through carp following i.p. injection with 1 × 104 parasites/fish. Parasitemia was monitored using a Bürker counting chamber after dilution of blood 1/1 in RPMI 1649 medium adjusted to 280 mOsmol/kg containing 50 IU/ml of heparin. For parasite isolation, blood was collected from heavily infected carp and, after centrifugation, T. borreli were collected from the buffy coat and purified on a 1 × 12-cm ion-exchange column chromatography using DEAE cellulose (DE-52; Whatman International) (59). After isolation, parasites were harvested by centrifugation and resuspended in fresh complete HML medium (60) supplemented with 5% pooled carp serum, l-glutamine (2 mM), penicillin G (100 IU/ml), and streptomycin sulfate (50 mg/L)).

For bacterial expression of soluble TNF-α, the sequence corresponding to the mature peptide of carp TNF-α1 (accession no. AJ311800, aa 77–237) or TNF-α2 (accession no. AJ311801, aa 70–231) was amplified by standard PCR using a proofreading Taq polymerase (Expand high-fidelity Taq polymerase; Roche) and cDNA from LPS-stimulated HKL as template. All primers used for cloning are listed in Table I. PCR products were directly cloned into pQE-30UA plasmid (Qiagen) downstream of the sequence coding for a 6× histidine tag. The primers used for the amplification were pQETNF1_FW or pQETNF2_FW in combination with pQETNF_RV.

Table I.

Primers used for TNF-α cloning in bacterial or eukaryotic plasmid

PrimerSequence 5′ → 3′a
pQETNF1_FW ctttcaaaagcaaatgtc 
pQETNF2_FW ctttcaaaagaaaatgtc 
pQETNF_RV taaagcaaacaccccaaa 
IRES_EcoRI_TNF1_FW aagagaattcctttcaaaagcaaatgtc 
IRES_EcoRI_TNF2_FW aagagaattcctttcaaaag+aaaatgtc 
IRES_BamHI_TNF RV agagggatcctcataaagcaaacaccccaa 
XhoI_TGFleader_FW cggacgctcgagatgagggtggagagtttatta 
EcoRIHisLeader_RV ctccaggaattcgtgatggtgatggtgatgtgctccgctatagtgcacaaatccc 
PrimerSequence 5′ → 3′a
pQETNF1_FW ctttcaaaagcaaatgtc 
pQETNF2_FW ctttcaaaagaaaatgtc 
pQETNF_RV taaagcaaacaccccaaa 
IRES_EcoRI_TNF1_FW aagagaattcctttcaaaagcaaatgtc 
IRES_EcoRI_TNF2_FW aagagaattcctttcaaaag+aaaatgtc 
IRES_BamHI_TNF RV agagggatcctcataaagcaaacaccccaa 
XhoI_TGFleader_FW cggacgctcgagatgagggtggagagtttatta 
EcoRIHisLeader_RV ctccaggaattcgtgatggtgatggtgatgtgctccgctatagtgcacaaatccc 
a

+, Locked nucleic acid modifications.

For eukaryotic expression of soluble TNF-α1 or TNF-α2, the mature peptides (excluding the TNF-α leader peptide and transmembrane sequence) were cloned in the pIRES-EGFP plasmid (Promega), downstream of the leader peptide region of the carp TGFβ (AF136947). The transmembrane region was removed to favor direct TNF-α secretion. The TGFβ leader peptide was inserted upstream of the mature TNF-α peptide to increase the rate of TNF-α secretion, as the use of a TGFβ leader peptide has been previously shown to effectively increase protein secretion in carp cells (61). TNF-α1 or TNF-α2 sequences were amplified as described above using IRES_EcoRI_TNF1_FW or IRES_EcoRI_TNF2_FW in combination with IRES_BamHI_TNF_RV (see Table I). The nucleotide sequence encoding for the leader peptide region of the carp TGFβ was amplified using XhoI_TGFleader_FW and EcoRIHisLeader_RV, with part of the RV primers containing the sequence encoding for a 6× histidine tag. The leader sequence was first cloned into the XhoI site of the pIRES-EGFP plasmid, followed by restriction and ligation of the TNF-α1 or TNF-α2 sequence between the EcoRI site, downstream of the leader-His sequence, and the BamHI site of the pIRES-EGFP plasmid. All constructs were transformed into M15 competent Escherichia coli cells, and positive clones were selected on agar plates containing ampicillin (100 μg/ml) and kanamicin (25 μg/ml). At least eight clones for each construct were sequenced. Expression of mature carp TNF-α was confirmed by Western blot analysis of supernatants from transfected Epithelioma papulosum cyprini (EPC) cells (62) using an anti-histidine Ab (Qiagen).

Transformed E. coli M15 cells were grown on agar plates supplemented with antibiotics (100 μg/ml ampicillin and 25 μg/ml kanamicin) overnight at 37°C. One single colony was transferred to 20 ml of Luria-Bertani medium (1% (w/v) tryptone, 0.5% (w/v) yeast extract, 0.5% (w/v) NaCl (pH 7.5)) supplemented with antibiotic and grown overnight at 37°C with shaking. The overnight culture was transferred to 1 liter of prewarmed TB medium (tryptone 2.4% (w/v), yeast extract 1.2% (w/v), NaCl 0.5% (w/v), glycerol 4% (v/v) (pH 7.5)) and incubated at 30°C with shaking. At OD600 of 0.6–0.8, protein production was induced with 1 mM isopropyl β-D-thiogalactoside and the culture was incubated for a further 4 h at 30°C. Bacterial pellets were collected by centrifugation at 5270 × g for 15 min at 4°C and stored at −80°C until use. Proteins were purified under native conditions and all steps were conducted at 4°C unless stated otherwise. Bacterial pellets from 1-liter culture were resuspended in 20–40 ml of lysis buffer (LB; 50 mM Tris-Cl (pH 7.5), 500 mM NaCl, 5 mM MgCl2, 10 mM imidazole, 10 mM 2-ME, 10% glycerol (v/v), 0.1% Triton X-114 (v/v)) supplemented with protease inhibitor cocktail (Sigma-Aldrich, catalog no. P8849) and lysozyme (1 mg/ml; Merck) and the suspension was incubated for 30 min on ice. Lysates were sonicated and subsequently centrifuged at 10,000 × g for 20 min at 4°C. Supernatants were collected and combined with 500 μl (for TNF-α1) or 750 μl (for TNF-α2) of Ni-NTA matrix (Qiagen). Samples were incubated for 30 min at 4°C while rotating and subsequently loaded on a 12-ml Poly-Prep chromatography column (Bio-Rad, catalog no. 7311550). All washing steps were performed under the following conditions: 300 × g for 4.5 min at 4°C. Columns were washed five times with 5 ml of wash buffer A (LB containing 20 mM imidazole), five times with 5 ml of wash buffer B (LB containing 30 mM imidazole), three times with 5 ml of wash buffer C (LB containing 50 mM imidazole, without Triton X-114), once with 5 ml of wash buffer D (LB containing 100 mM imidazole, without Triton X-114), and three times with 5 ml of wash buffer E (500 mM NaCl, 2 mM KCl, 20 mM Na2HPO4, 2 mM KH2PO4, 100 mM imidazole). After an empty centrifugation step, proteins were eluted in elution buffer (wash buffer E containing 250 mM imidazole). Eluted proteins were subsequently dialyzed against cPBS, and endotoxins were measured by end-point chromogenic Limulus amebocyte lysate test (Charles River Laboratories). If necessary, further removal of endotoxin was achieved by Triton X-114 phase separation (63). All preparations contained <0.015 EU/ml. Unless stated otherwise, heat-treated TNF-α samples were used as negative control in all stimulation assays.

To confirm overexpression of carp TNF-α in plasmid-injected fish, ∼100 mg of tissue collected from the injection site was resuspended in 1 ml of RIPA lysis buffer (50 mM Tris-HCl (pH 7.4), 1% Nonidet P-40, 0.25% sodium deoxycholate, 150 mM NaCl, 1 mM EDTA, 1 mM PMSF, and protease inhibitor cocktail (Sigma-Aldrich, catalog no. P8849) and homogenized on ice by sonication (3 min, on for 30 s, off for 30 s, power 5, duty 30%). Lysates were cleared by centrifugation at 21,000 × g for 20 min at 4°C. Supernatants were carefully collected and 25 μl was resolved on a 12.5% SDS-PAGE. Proteins were transferred to nitrocellulose membranes (Protran; Schleicher & Schuell BioScience) and incubated overnight at 4°C with a 1/2000 dilution of affinity-purified polyclonal rabbit-anti-cTNFα Ab with gentile agitation. HRP-conjugated goat-anti-rabbit Ab (1/2000; Dako) was used as secondary Ab. Proteins were visualized by chemiluminescence detection (ECL detection system for Western blot; Amersham Biosciences) on x-ray films.

Affinity-purified polyclonal rabbit IgG anti-carp TNF-α1 or anti-TNF-α2 was produced by immunization of rabbits with purified bacterial recombinant protein, according to a 3-mo standard protocol (Eurogentec). Affinity-purified IgG recognized both TNF-α1 and TNF-α2. Anti-TNF-α Abs were used for subsequent protein detection.

PBL, HKL, or enriched phagocyte fractions were resuspended in complete medium at a concentration of 1 × 107 cells/ml and 100 μl was transferred to 96-well plates. Cells were incubated overnight at 27°C in a humidified atmosphere of 5% CO2. The following day cells were stimulated with various concentration of TNF-α, in the presence or absence of LPS in a final volume of 200 μl.

EC were cultured in 24-well plates until confluent and subsequently stimulated with various concentrations of TNF-α in a final volume of 500 μl. After 4 h cells were harvested for gene expression analysis. Alternatively, EC were cultured in 6-well plates in complete EC medium until confluent. The day before stimulation the medium was replaced with complete cRPMI containing 1.5% pooled carp serum, and cells were subsequently stimulated with 1 μg/ml TNF-α in a final volume of 1.5 ml. After 24 h cell supernatants were collected and cleared through a 0.45-μm filter. Supernatants were used immediately or stored at −80°C until use.

Total RNA was isolated from spleen, PBL, or enriched phagocyte fractions using the RNeasy Mini Kit according to the manufacturer’s instructions, including an on-column DNase treatment with the RNase-free DNase set (Qiagen), and stored at −80°C until use. Before cDNA synthesis, to 0.25–1 μg of total RNA, a second DNase treatment was performed using DNase I, amplification grade (Invitrogen). Synthesis of cDNA was performed using SuperScript III first-strand synthesis systems for RT-PCR (Invitrogen) using random primers. A nonreverse transcriptase control was included for each sample. cDNA samples were further diluted 1/5 in nuclease-free water before real-time quantitative PCR analysis.

Real-time quantitative PCR using SYBR Green I technology was performed with Rotor-Gene 6000 (Corbett Research) and the Brilliant SYBR Green quantitative PCR (Stratagene) as detection chemistry as described previously (55). Primers used for real-time quantitative PCR are listed in Table II. Fluorescence data from real-time quantitative PCR experiments were analyzed using Rotor-Gene version 6.0.21 software and exported to Microsoft Excel. The cycle threshold (Ct) for each sample and the reaction efficiencies (E) for each primer set were obtained upon comparative quantitation analysis from the Rotor-Gene version 6.0.21 software. The relative expression ratio (R) of a target gene was calculated based on the E and Ct deviation of sample vs control (64, 65), and it was expressed relative to the S11 protein of the 40S subunit as reference gene.

Table II.

Primers used for quantitative real-time PCR analysis

PrimerSequence 5′ → 3′Accession No.
q40S.FW1 ccgtgggtgacatcgttaca AB012087 
q40S.RV1 tcaggacattgaacctcactgtct 
qTNFa1_FW gagcttcacgaggactaatagacagt AJ311800 
qTNFa1_RV ctgcggtaagggcagcaatc 
qTNFa2_FW cggcacgaggagaaaccgagc AJ311801 
qTNFa2_RV catcgttgtgtctgttagtaagttc 
qIL_10_FW cgccagcataaagaactcgt AB110780 
qIL-10_RV tgccaaatactgctcgatgt 
qIL_1β_FW aaggaggccagtggctctgt CCA245635 
qIL-1β_RV cctgaagaagaggaggctgtca 
qiNOS_FW aacaggtctgaaagggaatcca AJ242906 
qiNOS_RV cattatctctcatgtccagagtctcttct 
qCXCa_FW ctgggattcctgaccattggt AJ421443 
qCXCa_RV gttggctctctgtttcaatgca 
qIL1RI_FW acgccaccaagagcctttta AJ843873 
qIL1RI_RV gcagcccatatttggtcaga 
qsele_FW ggaaagaataatgaagactgtgtgg GQ231486 
qsele_RV caggatgccgtgtagcagag 
PrimerSequence 5′ → 3′Accession No.
q40S.FW1 ccgtgggtgacatcgttaca AB012087 
q40S.RV1 tcaggacattgaacctcactgtct 
qTNFa1_FW gagcttcacgaggactaatagacagt AJ311800 
qTNFa1_RV ctgcggtaagggcagcaatc 
qTNFa2_FW cggcacgaggagaaaccgagc AJ311801 
qTNFa2_RV catcgttgtgtctgttagtaagttc 
qIL_10_FW cgccagcataaagaactcgt AB110780 
qIL-10_RV tgccaaatactgctcgatgt 
qIL_1β_FW aaggaggccagtggctctgt CCA245635 
qIL-1β_RV cctgaagaagaggaggctgtca 
qiNOS_FW aacaggtctgaaagggaatcca AJ242906 
qiNOS_RV cattatctctcatgtccagagtctcttct 
qCXCa_FW ctgggattcctgaccattggt AJ421443 
qCXCa_RV gttggctctctgtttcaatgca 
qIL1RI_FW acgccaccaagagcctttta AJ843873 
qIL1RI_RV gcagcccatatttggtcaga 
qsele_FW ggaaagaataatgaagactgtgtgg GQ231486 
qsele_RV caggatgccgtgtagcagag 

Enriched phagocyte fractions (1 × 106) were stimulated for 1 h at 27°C in a humidified atmosphere of 5% CO2 with increasing concentrations of recombinant TNF-α in the presence of 0.1 μg/ml dihydrorhodamine (DHR; Sigma-Aldrich). In all set-ups, PMA (Sigma-Aldrich) was added 15 min before measurement at a final concentration of 0.05 μg/ml. Each treatment was conducted in quadruplicate. Cells were analyzed by flow cytometry in the presence of 1 μg/ml propidium iodide to exclude necrotic cells. Forward scatter and side scatter characteristics of 1 × 104 events were acquired in linear mode and fluorescence intensities were acquired at log scale using a Beckman Coulter Epics XL-MCL flow cytometer.

Heat-killed, formalin-fixed bacteria (Staphylococcus aureus; Pansorbin cells standardized; Calbiochem/EMD Biosciences, catalog no. 507861) were incubated overnight with 5 μg/ml FITC in carbonate/bicarbonate buffer (pH 9.4) at room temperature, while rotating and protected from light. Excess FITC was removed by washing five times in cPBS, and bacteria were resuspended in cRPMI and were ready to use in the phagocytosis assay.

Total HKL (0.5 × 106) were transferred to flow cytometry tubes and stimulated for 30 min at 27°C in a humidified atmosphere of 5% CO2 with increasing concentrations of recombinant carp TNF-α in the presence of FITC-labeled bacteria (5 × 106). Each treatment was conducted in quadruplicate. Phagocytosis was stopped by placing the tubes on ice and by adding 800 μl of ice-cold 1.5% paraformaldehyde. Two-hundred microliters of cell suspension was combined with 200 μl of cPBS containing 1.5% PFA and 200 μg/ml trypan blue and measured by flow cytometry as described above. Live cell gating excluded free bacteria from the measurement.

Total HKL (1 × 106) were stimulated with various concentrations of TNF-α in the presence or absence of LPS (20 μg/ml). Each treatment was conducted in triplicate. After 72 h of incubation at 27°C in a humidified atmosphere of 5% CO2, 75 μl of cell culture supernatant was combined with 100 μl of 1% (w/v) sulfanilamide in 2.5% (v/v) phosphoric acid and 100 μl of 0.1% (w/v) N-naphthyl-ethylenediamine in 2.5% (v/v) phosphoric acid. Absorbance values at 540 nm were acquired and nitrite concentration was measured using a sodium nitrite standard curve. In plasma samples, total nitrite plus nitrate was quantified as previously described (44) using a nitrite/nitrate colorimetric method (Roche Diagnostics, catalog no. 1746081) according to the manufacturer’s instruction.

Cell migration was analyzed using a 10-well transmigration chamber (Neuroprobe, catalog no. AA10) as described previously (66). Total HKL (4 × 106) were layered in the upper well of a transmigration chamber and TNF-α-treated EC supernatants were layered in the lower well. The upper and lower wells were separated by a polyvinylpyrrolidone-pretreated polycarbonate filter with 3-μm pores. Pooled carp serum (at 10%) was used as positive control, and EC supernatants stimulated with heat-treated TNF-α served as negative controls. After 3 h cells were harvested and an absolute cell count was performed by flow cytometry, as described previously (44).

The structure of mouse TNF-α with a resolution of 1.41 Å (Protein Data Bank entry 2tnf) was used as a template to model the carp TNF-α1 and TNF-α2 with the program MODELLER (version 9v5) (67, 68) using the consistent valence force field (69). The models were verified after several rounds of sequence alignment adjustments and energy minimization. Stereochemical quality of the homology models was assessed using the program PROCHECK (70). Protein folding quality was verified using the program PROSAII (71), which independently evaluates the compatibility of each residue to its environment.

Trypanolysis assay was performed as described previously (32). Briefly, AnTat1.1 T. brucei parasites were purified on a DE-52 ion-exchange column and resuspended to a concentration of 2 × 106 parasites/ml in PSG (PBS (pH 8.0) supplemented with 0.1% glucose). One hundred microliters of parasite suspension was combined with recombinant TNF-α or with TNF-α preincubated for 30 min with increasing concentration of N,N′-diacetylchitobiose (Sigma-Aldrich, catalog no. D1523) and then incubated at 30°C. Percentage cell death was calculated by light microscopy counts of remaining parasites after 4 h of incubation. Recombinant rainbow trout TNF-α (accession no. NM_001124357) was provided by Dr. J. Zou (Department of Zoology, University of Aberdeen, Aberdeen, U.K.) and recombinant zebrafish (accession no. NM_212859) and seabream (accession no. AJ413189) TNF-α were provided by Dr. F. Roca (Department of Cell Biology and Histology, University of Murcia, Murcia, Spain).

Nine-month-old carp (weighing 65 ± 0.3 g) were infected with 1 × 104T. borreli per fish. Starting at 1 wk after infection, and for a period of 4 wk, fish (n = 11) received a daily dose (i.p.) of pentoxifylline (PTX; Sigma-Aldrich, catalog no. P1784) of 50 mg/ml dissolved in 100 μl cPBS. PBS-injected fish (n = 11) served as infected control, and noninfected fish (n = 5) injected with PTX only (same concentration, same volume) served as negative control for adverse side effects due to PTX administration. In a parallel experiment, spleen tissue from n = 5 fish from each group was collected to confirm inhibition of TNF-α expression in PTX-treated fish.

Six-month-old carp (n = 77, weighing 16 ± 0.2 g) were infected with 1 × 104T. borreli per fish; at the same time, fish received two i.m. injections in the dorsal muscle of 25 μl of PBS containing plasmid DNA: fish were divided over seven groups (n = 11) and injected with a high (640 ng/μl) or low dose (64 ng/μl) of plasmid DNA encoding either TNF-α1 or TNF-α2 (pIRES-TNF-α1-EGFP or pIRES-TNF-α2-EGFP) or with equivalent doses of a mixture of the two plasmids. Control fish received the equivalent of a high dose of empty plasmid (pIRES-EGFP). To confirm overexpression of carp TNF-α in plasmid-injected fish, muscle tissue from the injection site was collected from n = 3 fish from each group at 48 h after injection and used for TNF-α detection by Western blot.

Nine-month-old carp (weighing 70 ± 2.2 g) were infected with 1 × 104T. borreli per fish. After 2.5 wk when parasitemia levels reached 2 × 106 parasites/ml of blood, fish (n = 9) received three consecutive daily injections of 100 μl (i.p.) of TACE inhibitor (PKF242-484, 10 mg/ml). PKF242–484 was provided by Dr. A. Trifilieff and was synthesized at the Novartis Institute for BioMedical Research, Basel, Switzerland. As PKF242–484 was shown to inhibit various matrix metalloproteinases, FN-439 (10 mg/ml, 100 μl, i.p.), a broad-range matrix metalloproteinase inhibitor that has no activity on TACE, was used as negative control (72, 73). After 3 days parasitemia was measured and plasma collected for subsequent measurement of nitrite levels. Spleen tissues were collected and snap-frozen in liquid nitrogen for subsequent histological analysis.

In a parallel in vitro experiment, head kidney leukocytes were stimulated with LPS (50 μg/ml) in the presence or absence of PKF242-hyphen]484 (50 μM). At various time points after stimulation, cells were collected and washed once in cold cRPMI medium. To remove any receptor-bound soluble TNF-α, cells were first acid-treated as described previously (74) and subsequently analyzed by flow cytometry. Surface-bound TNF-α was detected using affinity-purified rabbit-anti-carp TNF-α IgG.

For the detection of nitrotyrosine, anti-nitrotyrosine rabbit polyclonal immunoaffinity-purified IgG (Bio-connect; Upstate Biotechnology, catalog no. 06-284) was used. Cryosections (7 μm) of spleen tissue were mounted on poly-l-lysine-coated glass slides (BDH Laboratory Supplies) and treated as described previously (44) Anti-nitrotyrosine Ab was used in a 1/20 dilution and alkaline phosphatase-conjugated goat-anti-mouse Ab (Dako) was used in a 1/200 dilution.

For gene expression analysis, relative expression ratios (R) were calculated as described. Transformed (ln(R)) values were used for statistical analysis in SPSS software (15.0). Significant differences (p < 0.05) were determined by independent sample Student’s t test for the in vitro gene expression study and by a two-way ANOVA followed by a Sidak’s test for the in vivo gene expression study. A one-way ANOVA followed by a Sidak’s test was used for the migration and respiratory burst studies.

One of the typical signs associated with T. borreli infections is splenomegaly, and we therefore focused on the spleen to study the kinetics of TNF-α gene expression in vivo by real-time quantitative PCR (Fig. 1). Two isoforms of carp TNF-α (TNF-α1 and TNF-α2) were identified previously (14) and will be referred to as TNF-α when differences between the two isoforms are not relevant. TNF-α2 fold change in gene expression was consistently higher than the fold change observed for TNF-α1. This difference can be explained by the dissimilar basal expression of the two genes in carp PBL, where TNF-α1 is ∼50-fold more expressed than TNF-α2. Hence, TNF-α2, although being expressed very low at the basal level, is up-regulated to a larger extent than TNF-α1, possibly leading to a comparable protein expression. At peak levels of parasitemia, both TNF-α1 and TNF-α2 were up-regulated in spleen from T. borreli-infected fish, concomitantly with the peak of expression of inducible NO synthase (iNOS), but not IL-1β (Fig. 1).

FIGURE 1.

Real-time quantitative PCR analysis of gene expression in spleen after in vivo infection with T. borreli. Fish (9 mo old, weighing 160 ± 20 g) were infected with 1 × 104 parasites per fish. At time point 0 h, n = 5 control fish and at various time points after infection, n = 5 infected and n = 3 noninfected fish were sacrificed. Parasitemia was monitored during infection and is shown in the upper right plot. Gene expression was normalized relative to the 40S ribosomal protein S11 as internal reference gene and expressed relative to noninfected fish at time point 0. Values are given as means and SD. ∗, Significant difference relative to noninfected fish at the same time point. Note the difference in fold change (y-axis) between the genes.

FIGURE 1.

Real-time quantitative PCR analysis of gene expression in spleen after in vivo infection with T. borreli. Fish (9 mo old, weighing 160 ± 20 g) were infected with 1 × 104 parasites per fish. At time point 0 h, n = 5 control fish and at various time points after infection, n = 5 infected and n = 3 noninfected fish were sacrificed. Parasitemia was monitored during infection and is shown in the upper right plot. Gene expression was normalized relative to the 40S ribosomal protein S11 as internal reference gene and expressed relative to noninfected fish at time point 0. Values are given as means and SD. ∗, Significant difference relative to noninfected fish at the same time point. Note the difference in fold change (y-axis) between the genes.

Close modal

Recombinant TNF-α1 and TNF-α2 (Fig. 2) were produced in a bacterial system and the biological activity was initially tested in vitro with respect to the ability of recombinant carp TNF-α to promote classical proinflammatory activities in carp phagocytes. Throughout the study, heat-treated recombinant carp TNF-α was used as a negative control. Stimulation with various concentrations of recombinant TNF-α1 and TNF-α2 resulted in only a moderate up-regulation of proinflammatory molecules (Fig. 3, A and B). Both TNF-α1 and TNF-α2 failed to induce the production of oxygen (DHR oxidation, Fig. 3, C and D) and nitrogen (nitrite) radicals (Fig. 3, E and F), as well as phagocytosis in carp phagocytes (Fig. 3, G and H). Longer or shorter incubation times with carp TNF-α also failed to promote phagocyte activity under the condition tested (data not shown). Additionally, carp TNF-α was produced in two eukaryotic systems: in insect cells and in EPC cells (62). Recombinant carp TNF-α, affinity purified from insect cell supernatants or TNF-α-containing supernatants from transfected EPC also failed to promote phagocyte activity (data not shown).

FIGURE 2.

Expression and purification of carp TNF-α1 and TNF-α2. A, Total bacterial cell lysates from noninduced (−) and isopropyl β-D-thiogalactoside (IPTG)-induced (+) cultures and (B) elution fractions (lanes 1–6) of purified proteins were analyzed on 12.5% SDS-PAGE and visualized by Coomassie brilliant blue staining.

FIGURE 2.

Expression and purification of carp TNF-α1 and TNF-α2. A, Total bacterial cell lysates from noninduced (−) and isopropyl β-D-thiogalactoside (IPTG)-induced (+) cultures and (B) elution fractions (lanes 1–6) of purified proteins were analyzed on 12.5% SDS-PAGE and visualized by Coomassie brilliant blue staining.

Close modal
FIGURE 3.

Effects of recombinant carp TNF-α on phagocytes: A and B, Real-time quantitative PCR analysis of gene expression. Phagocytes (1 × 106) were stimulated for 3 h with the indicated concentrations of recombinant carp TNF-α1 and TNF-α2. 40S was used as internal reference gene, and expression was normalized against the respective heat-treated control. C and D, Respiratory burst activity (DHR oxidation). Phagocytes (1 × 106) were stimulated with increasing concentration of native or heat-treated TNF-α. Respiratory burst activity was measured fluorometrically after 1 h. E and F, NO production. Total HKL (1 × 106) were stimulated with a suboptimal LPS concentration (20 μg/ml) in the presence or absence of increasing concentrations of native or heat-treated TNF-α. After 72 h supernatants were collected and nitrite concentrations measured by Griess reaction. G and H, Percentage of phagocytosis. Total HKL (0.5 × 106) were stimulated with FITC-labeled bacteria (5 × 106) in the presence or absence of increasing concentrations of native or heat-treated TNF-α. Percentage of phagocytosis was determined fluorometrically after 30 min.

FIGURE 3.

Effects of recombinant carp TNF-α on phagocytes: A and B, Real-time quantitative PCR analysis of gene expression. Phagocytes (1 × 106) were stimulated for 3 h with the indicated concentrations of recombinant carp TNF-α1 and TNF-α2. 40S was used as internal reference gene, and expression was normalized against the respective heat-treated control. C and D, Respiratory burst activity (DHR oxidation). Phagocytes (1 × 106) were stimulated with increasing concentration of native or heat-treated TNF-α. Respiratory burst activity was measured fluorometrically after 1 h. E and F, NO production. Total HKL (1 × 106) were stimulated with a suboptimal LPS concentration (20 μg/ml) in the presence or absence of increasing concentrations of native or heat-treated TNF-α. After 72 h supernatants were collected and nitrite concentrations measured by Griess reaction. G and H, Percentage of phagocytosis. Total HKL (0.5 × 106) were stimulated with FITC-labeled bacteria (5 × 106) in the presence or absence of increasing concentrations of native or heat-treated TNF-α. Percentage of phagocytosis was determined fluorometrically after 30 min.

Close modal

Next, the ability of recombinant TNF-α to stimulate carp EC was analyzed. TNF-α1 and TNF-α2 were both able to promote the expression of several proinflammatory genes, CXCa (a fish chemokine with IL-8 characteristics), and adhesion molecules in EC (Fig. 4 A). TNF-α2 was more potent than TNF-α1 in promoting gene expression. iNOS gene expression was present at low basal level (R = 0.0045, relative to reference gene) in carp EC and could be up-regulated after stimulation, only with TNF-α2. Additionally, based on homology sequencing, we identified a partial sequence of carp P-selectin (Sele, accession no. GQ231486) with extremely high (R = 1.395, relative to reference gene) basal gene expression in EC.

FIGURE 4.

Effects of recombinant carp TNF-α on EC. A, Real-time quantitative PCR analysis of gene expression. Endothelial cells were cultured in 24-well plates until confluent and then stimulated for 4 h with the indicated concentrations of carp TNF-α1 and TNF-α2. Gene expression was normalized relative to the 40S ribosomal protein S11 as internal reference gene and was expressed relative to the respective heat-treated control. Values are given as means and SD (n = 4). ∗, Statistical differences relative to the heat-treated control. B, Migration of leukocytes. Total HKL (4 × 106) were layered on the upper well of a transmigration chamber, and migration toward TNF-α alone or toward supernatants from TNF-α-treated EC was recorded after 3 h. Pooled carp serum (10%) was used as positive control and complete cRPMI as negative control. The total number of cells that migrated toward supernatants from EC stimulated with native TNF-α was corrected for the number of cells that migrated toward supernatants from EC stimulated with heat-treated TNF-α. ∗, Significant differences relative to the medium control or to supernatants form samples stimulated with the respective heat-treated control; #, significant differences between supernatants treated with TNF-α1 and TNF-α2. P-sele indicates carp P-selectin; SPN, supernatant.

FIGURE 4.

Effects of recombinant carp TNF-α on EC. A, Real-time quantitative PCR analysis of gene expression. Endothelial cells were cultured in 24-well plates until confluent and then stimulated for 4 h with the indicated concentrations of carp TNF-α1 and TNF-α2. Gene expression was normalized relative to the 40S ribosomal protein S11 as internal reference gene and was expressed relative to the respective heat-treated control. Values are given as means and SD (n = 4). ∗, Statistical differences relative to the heat-treated control. B, Migration of leukocytes. Total HKL (4 × 106) were layered on the upper well of a transmigration chamber, and migration toward TNF-α alone or toward supernatants from TNF-α-treated EC was recorded after 3 h. Pooled carp serum (10%) was used as positive control and complete cRPMI as negative control. The total number of cells that migrated toward supernatants from EC stimulated with native TNF-α was corrected for the number of cells that migrated toward supernatants from EC stimulated with heat-treated TNF-α. ∗, Significant differences relative to the medium control or to supernatants form samples stimulated with the respective heat-treated control; #, significant differences between supernatants treated with TNF-α1 and TNF-α2. P-sele indicates carp P-selectin; SPN, supernatant.

Close modal

Given the up-regulation of chemokines and adhesion molecules in EC after stimulation with TNF-α, the ability of supernatants from TNF-α-treated EC to promote leukocyte migration was investigated. Supernatants collected from EC treated with either TNF-α1 or TNF-α2 were able to promote leukocyte migration (Fig. 4 B), whereas recombinant TNF-α alone did not. In line with the difference observed at the gene expression level, TNF-α2-treated supernatants were more potent than TNF-α1-treated supernatants in promoting leukocyte migration.

To investigate the indirect activation of phagocytes by carp TNF-α, supernatants from TNF-α-treated EC were used to promote the production of oxygen radicals (DHR oxidation) in carp phagocytes (Fig. 5). Clearly, increased DHR oxidation was observed in granulocytes (Fig. 5,A, gate B) in samples incubated for 30–60 min with supernatants from TNF-α-treated EC (Fig. 5, A and B, gray bars). Preincubation of phagocytes for as short as 10 min with carp TNF-α1 or TNF-α2 resulted in a more rapid (15 min) DHR oxidation of granulocytes (Fig. 5, A and B, black bars). Preincubation with TNF-α2 resulted in a higher production of oxygen radicals than did preincubation with TNF-α1. The difference in oxidation of DHR in phagocytes prestimulated, or not, with TNF-α was less pronounced after 30 min. However, after 60 min, oxidation of DHR in prestimulated cells was significantly lower than in non-prestimulated phagocytes, with a greater effect visible in TNF-α1-prestimulated cells (compare gray and black bars in Fig. 5 B). This indicates that although carp TNF-α was not able to directly promote the production of oxygen radicals in phagocytes, TNF-α can prime phagocytes to respond faster to EC-derived mediators.

FIGURE 5.

Activation of phagocytes by supernatants from TNF-α-treated EC. Carp phagocytes (1 × 106) were incubated for the indicated time with supernatants collected from TNF-α-treated EC in the presence of DHR (0.1 μg/ml). Alternatively, cells were first preincubated for 10 min with 1 μg/ml carp TNF-α1 or TNF-α2. For the last 15 min, PMA was added to the culture and fluorescence values of 104 events were acquired fluorometrically. A, Representative scatter dot plot of phagocytes stimulated with TNF-α2-treated EC supernatants. Cells were directly incubated with TNF-α2-treated EC supernatants (upper panel) or pretreated with TNF-α2 and subsequently stimulated with the respective supernatant (lower panel). Gate A indicates total DHR+ cells and gate B indicates a granulocyte subpopulation with increased fluorescence intensity. B, Positive cells in gate B in phagocyte cultures directly stimulated with EC supernatants (gray bars) or preincubated with TNF-α and subsequently stimulated with the respective supernatant (black bars). Measurements were performed in quadruplicate. Values are given as means and SD. ∗, Significant differences with respect to supernatant from EC stimulated with heat-treated TNF-α; #, significant differences between prestimulated and non-prestimulated phagocytes. Shown is one representative experiment of two independent experiments performed.

FIGURE 5.

Activation of phagocytes by supernatants from TNF-α-treated EC. Carp phagocytes (1 × 106) were incubated for the indicated time with supernatants collected from TNF-α-treated EC in the presence of DHR (0.1 μg/ml). Alternatively, cells were first preincubated for 10 min with 1 μg/ml carp TNF-α1 or TNF-α2. For the last 15 min, PMA was added to the culture and fluorescence values of 104 events were acquired fluorometrically. A, Representative scatter dot plot of phagocytes stimulated with TNF-α2-treated EC supernatants. Cells were directly incubated with TNF-α2-treated EC supernatants (upper panel) or pretreated with TNF-α2 and subsequently stimulated with the respective supernatant (lower panel). Gate A indicates total DHR+ cells and gate B indicates a granulocyte subpopulation with increased fluorescence intensity. B, Positive cells in gate B in phagocyte cultures directly stimulated with EC supernatants (gray bars) or preincubated with TNF-α and subsequently stimulated with the respective supernatant (black bars). Measurements were performed in quadruplicate. Values are given as means and SD. ∗, Significant differences with respect to supernatant from EC stimulated with heat-treated TNF-α; #, significant differences between prestimulated and non-prestimulated phagocytes. Shown is one representative experiment of two independent experiments performed.

Close modal

Carp TNF-α is homologous to mammalian TNF-α and presents many conserved structural features of its mammalian counterpart (14). We produced a putative three-dimensional model based on the crystal structure of mouse TNF-α (75), confirming that also carp TNF-α might exist as a homotrimer, with each subunit consisting of anti-parallel β-sheets organized in a “jelly-roll” motif with a pyramid shape (Fig. 6). Mammalian TNF-α has a lectin domain located at the top of the pyramid-shaped molecule also referred to as TIP domain. Carp TNF-α has a conserved three-dimensional structure clearly identifying a TIP domain (Fig. 6,A). More detailed analysis of the three-dimensional structure and of the primary sequence corresponding to the TIP region evidenced considerable differences between TNF-α1 and TNF-α2 (Fig. 6, B and C).

FIGURE 6.

Three-dimensional analysis of carp TNF-α. A, Side view of carp TNF-α1 with the individual monomers shown in magenta, blue, and green. At the top of the pyramid-shaped molecule is the TIP domain. B, Top view of superimposed TNF-α1 (orange) and TNF-α2 (gray) trimeric molecules showing the different orientation of the loops forming the TIP domain in each of the two molecules. C, ClustalW alignment of the amino acid portion corresponding to the TIP region of TNF-α1 and TNF-α2. Despite the high similarity in the surrounding region, the TIP region (yellow) shows a considerable degree of variation.

FIGURE 6.

Three-dimensional analysis of carp TNF-α. A, Side view of carp TNF-α1 with the individual monomers shown in magenta, blue, and green. At the top of the pyramid-shaped molecule is the TIP domain. B, Top view of superimposed TNF-α1 (orange) and TNF-α2 (gray) trimeric molecules showing the different orientation of the loops forming the TIP domain in each of the two molecules. C, ClustalW alignment of the amino acid portion corresponding to the TIP region of TNF-α1 and TNF-α2. Despite the high similarity in the surrounding region, the TIP region (yellow) shows a considerable degree of variation.

Close modal

We investigated whether the lectin-like activity ascribed to mammalian TNF-α, and known to act through its TIP domain (31, 32), would be conserved in the homologous carp TNF-α molecule. The ability of carp TNF-α1 and TNF-α2 to kill the bloodstream form of T. brucei (AnTat 1.1 clone) was investigated in a trypanolytic assay. Both TNF-α1 and TNF-α2 killed T. brucei in a dose- and time-dependent manner (data not shown) where, maximally, 1 μg/ml carp TNF-α killed 100% of parasites within 4 h (Fig. 7, A and B). Trypanolytic activity of carp TNF-α against T. brucei could be blocked by preincubation with increasing concentrations of N,N′-diacetylchitobiose (Fig. 7, A and B), confirming that TNF-α activity acted via a lectin-like interaction. In parasite cultures incubated with TNF-α1, N,N′-diacetylchitobiose completely blocked TNF-α1 activity at concentrations >1 ng/ml, whereas in cultures incubated with TNF-α2 even the highest concentration of N,N′-diacetylchitobiose could not completely block TNF-α2 activity. Such a difference could be due to the dissimilarities observed in the TIP region. Collectively, these results indicate that both isoforms of carp TNF-α have lectin-like activities and that TNF-α2 might have a stronger activity than TNF-α1.

FIGURE 7.

Lysis of bloodstream forms of T. brucei by fish TNF-α. Freshly isolated parasites (2 × 106/ml) were incubated in PSG (pH 8.0) with TNF-α (1 μg/ml) for 4 h at 37°C. Parasites were incubated with carp TNF-α1 (A) or carp TNF-α2 (B) alone or after preincubation with increasing concentrations of N,N′-diacetylchitobiose or (C) with zebrafish, seabream, or trout.

FIGURE 7.

Lysis of bloodstream forms of T. brucei by fish TNF-α. Freshly isolated parasites (2 × 106/ml) were incubated in PSG (pH 8.0) with TNF-α (1 μg/ml) for 4 h at 37°C. Parasites were incubated with carp TNF-α1 (A) or carp TNF-α2 (B) alone or after preincubation with increasing concentrations of N,N′-diacetylchitobiose or (C) with zebrafish, seabream, or trout.

Close modal

To investigate whether the lectin-like activity would be limited to carp TNF-α only, we investigated the trypanolytic activity of TNF-α from zebrafish (both Cyprinidae) but also from two more distantly related fish species, seabream (Perciformes) and rainbow trout (Salmonidae). All investigated TNF-α were effective in killing T. brucei at concentrations and time comparable to that observed for carp TNF-α (Fig. 7 C), further confirming that the lectin-like activity is highly conserved.

We investigated the role of TNF-α in vivo particularly with respect to fish survival and control of parasitemia. Starting at 1 wk postinfection, infected fish received a daily dose of PTX, a TNF-α inhibitor that acts at the transcriptional level (14, 76). Fish injected with PTX showed lower survival and higher parasitemia at all time points after infection when compared with the infected control group (Fig. 8, A and B), indicating that TNF-α is critically involved in parasitemia control. Reduced TNF-α, but not iNOS, gene expression was observed in spleen from infected fish treated with PTX, but not in PBS-injected fish infected with T. borreli (Fig. 8 C). Noninfected PTX-injected fish did not show any side effects due to PTX administration (data not shown).

FIGURE 8.

Immunomodulation by soluble and mTNF-α during T. borreli infection in vivo. At time 0 fish were infected i.p. with 104 parasites. Mortality (top panels) and parasitemia (middle panels) were monitored. Control experiments are shown in the lower panels. A–C, Inhibition of TNF-α expression by PTX administration. A and B, Mortality and parasitemia in infected fish. Starting 1 wk after infection, fish (n = 12) received a daily dose of PTX (50 mg/kg) or PBS. C, TNF-α and iNOS gene expression (quantitative RT-PCR) in spleen of nontreated and PTX-treated infected fish. Spleen from n = 5 fish was collected at 24 days postinfection. Shown are expression data relative to the noninfected fish at the same time point. Note that for clarity, fold changes of TNF-α1 and TNF-α2 are indicated on the left axis and iNOS of the right axis. D–F, Overexpression of TNF-α by recombinant plasmid DNA administration. D and E, Mortality and parasitemia in infected fish. At time 0, fish (n = 11) were injected i.m. with 25 μl of PBS containing a high dose (10 μg/10 g), a low dose (1 μg/10 g), or a mixture of the two of plasmid encoding for carp TNF-α1 (a1) or TNF-α2 (a2). Fish injected with a high dose of empty plasmid served as infected negative control. F, TNF-α protein expression at the injection site in muscle of plasmid-injected fish. Muscle from fish injected with the empty plasmid or with a high dose of plasmid encoding for TNF-α1 and TNF-α2 was collected for protein expression analysis. Carp TNF-α was detected using an affinity-purified rabbit anti-carp TNF-α IgG and HRP-conjugated goat-anti-rabbit IgG as secondary Ab by Western blot. Shown are the results from one fish out of n = 3 tested showing similar results. G–I, Overexpression of mTNF-α by administration of TACE inhibitor. G and H, Mortality and parasitemia in infected fish. After 18 days, when parasitemia reached ∼2 × 106 parasites/ml of blood, fish (n = 9) received three consecutive daily injections of the TACE inhibitor (PKF242-484, 10 mg/ml) or equivalent doses of the control compound (FN439). Values are given as mean ± SD of at least n = 7 measurements. ∗, Significant differences with respect to the infected control group. I, Relative increase of the number of cells bearing surface-bound TNF-α. Carp leukocytes were stimulated in triplicate wells with LPS (50 μg/ml) in the presence of PKF242-484 (50 μM) and values are expressed relative to leukocytes stimulated with LPS only. Shown are results from one representative experiment out of three performed independently.

FIGURE 8.

Immunomodulation by soluble and mTNF-α during T. borreli infection in vivo. At time 0 fish were infected i.p. with 104 parasites. Mortality (top panels) and parasitemia (middle panels) were monitored. Control experiments are shown in the lower panels. A–C, Inhibition of TNF-α expression by PTX administration. A and B, Mortality and parasitemia in infected fish. Starting 1 wk after infection, fish (n = 12) received a daily dose of PTX (50 mg/kg) or PBS. C, TNF-α and iNOS gene expression (quantitative RT-PCR) in spleen of nontreated and PTX-treated infected fish. Spleen from n = 5 fish was collected at 24 days postinfection. Shown are expression data relative to the noninfected fish at the same time point. Note that for clarity, fold changes of TNF-α1 and TNF-α2 are indicated on the left axis and iNOS of the right axis. D–F, Overexpression of TNF-α by recombinant plasmid DNA administration. D and E, Mortality and parasitemia in infected fish. At time 0, fish (n = 11) were injected i.m. with 25 μl of PBS containing a high dose (10 μg/10 g), a low dose (1 μg/10 g), or a mixture of the two of plasmid encoding for carp TNF-α1 (a1) or TNF-α2 (a2). Fish injected with a high dose of empty plasmid served as infected negative control. F, TNF-α protein expression at the injection site in muscle of plasmid-injected fish. Muscle from fish injected with the empty plasmid or with a high dose of plasmid encoding for TNF-α1 and TNF-α2 was collected for protein expression analysis. Carp TNF-α was detected using an affinity-purified rabbit anti-carp TNF-α IgG and HRP-conjugated goat-anti-rabbit IgG as secondary Ab by Western blot. Shown are the results from one fish out of n = 3 tested showing similar results. G–I, Overexpression of mTNF-α by administration of TACE inhibitor. G and H, Mortality and parasitemia in infected fish. After 18 days, when parasitemia reached ∼2 × 106 parasites/ml of blood, fish (n = 9) received three consecutive daily injections of the TACE inhibitor (PKF242-484, 10 mg/ml) or equivalent doses of the control compound (FN439). Values are given as mean ± SD of at least n = 7 measurements. ∗, Significant differences with respect to the infected control group. I, Relative increase of the number of cells bearing surface-bound TNF-α. Carp leukocytes were stimulated in triplicate wells with LPS (50 μg/ml) in the presence of PKF242-484 (50 μM) and values are expressed relative to leukocytes stimulated with LPS only. Shown are results from one representative experiment out of three performed independently.

Close modal

A recombinant plasmid expressing TNF-α1 or TNF-α2 was constructed to investigate the immunomodulatory activity of excess TNF-α during in vivo T. borreli infection. Fish treated with various doses of recombinant plasmid encoding either TNF-α1 or TNF-α2 all showed lower survival than the infected control group, with the group injected with a high dose of TNF-α2 plasmid showing the lowest survival (Fig. 8,D). Despite the clear effect on survival, parasitemia was not significantly affected. For clarity, only the parasitemia curve of the group injected with a high dose of TNF-α2 plasmid is shown (Fig. 8,E). TNF-α plasmid-injected fish showed high TNF-α protein expression in muscle tissue at the injection site as detected by Western blot (Fig. 8 F). Parallel experiments performed by i.p. injection of high and low doses of recombinant carp TNF-α1 and TNF-α2 showed similar results, where all groups treated with TNF-α had lower survival than did the infected control group (data not shown).

An inhibitor of TACE was used in vivo during T. borreli infection to prevent TNF-α shedding, thereby increasing mTNF-α levels (73). About 2.5 wk (18 days) after infection fish reached a parasitemia of 2 × 106 parasites/ml of blood and received TACE inhibitor. After 3 days of treatment only, in all fish treated with the TACE inhibitor, complete clearance of the parasite was observed, while in fish treated with a general matrix metalloproteinase inhibitor, which has no effect on TACE, parasitemia progressed normally (Fig. 8, G and H). In vitro, stimulation of carp leukocytes with LPS in the presence of the TACE inhibitor significantly increased the number of positive cells with detectable surface-bound TNF-α in a time-dependent manner (Fig. 8,I). Additionally, fish treated with the TACE inhibitor displayed extremely reduced spleen size as compared with the control group (data not shown). Neither the TACE inhibitor nor the control compound showed direct parasite toxicity in vitro in a concentration range from 0.2 to 100 μM (data not shown). The effects of the TACE inhibitor on parasitemia and spleen size led us to investigate whether other parameters generally associated with pathology of infection, such as plasma nitrite levels and tissue nitration, would also be affected. As expected, plasma nitrite levels were strongly reduced, although not significantly owing to high variation between individuals, in fish treated with the TACE inhibitor (Fig. 9A). Tissue nitration, as measured by immunohistochemistry using an anti-nitrotyrosine Ab (44), was also reduced by treatment with TACE inhibitor, as indicated by the lower nitrotyrosine staining in the spleen of treated fish (Fig. 9 B).

FIGURE 9.

Plasma nitrite levels and tissue nitration in T. borreli-infected fish treated with TACE inhibitor. At time 0 fish were infected i.p. with 104 parasites. After 18 days, when parasitemia reached ∼2 × 106 parasites/ml of blood, fish (n = 9) received three consecutive daily injections of the TACE inhibitor (PKF242-484, 10 mg/ml) or equivalent doses of the control compound (FN439). At 21 days (3 days after treatment) plasma and spleen tissue samples were collected. A, Plasma nitrite levels. B, Anti-nitrotyrosine immunoreactivity (purple). Incubation of anti-nitrotyrosine Ab with a solution of 10 mM 3-nitrotyrosine completely abrogated the reaction (not shown).

FIGURE 9.

Plasma nitrite levels and tissue nitration in T. borreli-infected fish treated with TACE inhibitor. At time 0 fish were infected i.p. with 104 parasites. After 18 days, when parasitemia reached ∼2 × 106 parasites/ml of blood, fish (n = 9) received three consecutive daily injections of the TACE inhibitor (PKF242-484, 10 mg/ml) or equivalent doses of the control compound (FN439). At 21 days (3 days after treatment) plasma and spleen tissue samples were collected. A, Plasma nitrite levels. B, Anti-nitrotyrosine immunoreactivity (purple). Incubation of anti-nitrotyrosine Ab with a solution of 10 mM 3-nitrotyrosine completely abrogated the reaction (not shown).

Close modal

To date, TNF-α homologs have been identified in mammalian vertebrates, one amphibian species (Xenopus laevis) (9), and several teleost species (10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23) but not in birds (7, 8). Although TNF-α has been identified in several teleost fish species its biological activities and role particularly during infections are largely unknown. In fish, a limited number of functional studies have been performed (20, 21, 22, 24, 25, 26, 27), mostly using recombinant TNF-α in vitro, and results point out several inconsistencies particularly with respect to some receptor-mediated activities of fish TNF-α, such as the ability of fish TNF-α to directly activate phagocytes. In the present study, carp TNF-α directly stimulated the expression of proinflammatory cytokines, chemokines, and adhesion molecules in endothelial cells but not in phagocytes, and supernatants from TNF-α-treated endothelial cells were able to promote leukocyte migration and respiratory burst activity. Interestingly, although TNF-α failed to directly stimulate the production of nitrogen and oxygen radicals, TNF-α-primed phagocytes were able to respond faster to TNF-α-induced mediators from endothelial cells. Our findings are in line with a recent study in zebrafish embryos on mycobacterial pathogenesis using a TNFR1 knock-down approach (39). In this study, both control and TNFR1 morphant embryos displayed iNOS staining that colocalized with a subset of infected macrophages, suggesting that TNF-α signaling is not required for iNOS expression. In the same study, where the authors showed that TNF-α is not required for tuberculous granuloma formation but does maintain granuloma integrity, macrophage trafficking across epithelial and endothelial barriers was shown to be independent from TNF-α signaling. Another study in zebrafish (77), however, did show a role for TNF-α signaling in neutrophil influx into the intestine in response to proinflammatory stimuli induced by LPS. Most recently, a study in seabream (27) showed that endothelial cells, more than phagocytes, might be the primary target of fish TNF-α, suggesting that TNF-α is primarily involved in the recruitment of phagocytes to inflammatory sites rather than in the direct activation of phagocytes.

Besides the numerous receptor-dependent activities exerted by TNF-α, at least one receptor-independent activity, that is, the lectin-like recognition of specific oligosaccharides, has been described. In mammals, TNF-α has been shown to directly bind to the variant specific glycoprotein present in the flagellar pocket of some African trypanosomes and to cause direct lysis (31, 32, 33, 34). An invertebrate functional analog of mammalian TNF-α, named coelomic cytolytic factor (CCF-1), has been described in the earthworm Eisenia foetida foetida (78). CCF-1 has been shown to exert lectin-like activities similar to mammalian TNF-α and is able to directly lyse T. brucei and Trypanosoma cruzi in vitro (79, 80). However, despite the functional similarities based on their lectin-like activity, CCF-1 and mammalian TNF-α do not share any sequence similarity and are not homologous genes, indicating a convergent evolution of function of two genetically unrelated cytokines (79). In the present study, the lectin-like activity of fish TNF-α homologs was investigated and results showed an evolutionary conservation of function of this receptor-independent activity of TNF-α not only in cyprinid fish, but also in salmonids and perciforms. To our knowledge this is the first report of lectin-like activities of TNF-α homologs in lower vertebrate species.

The conservation of the lectin-like activity of TNF-α among different fish species is opposed to the differences between fish species with regard to the receptor-dependent proinflammatory functions of TNF-α (24, 25, 22, 27), most of which have been studied in vitro. The role played by fish TNF-α during immune responses in vivo is largely unknown owing to the lack of suitable knockout or transgenic animal models. In the present study, we examined the role of TNF-α in vivo using three fundamentally different but complementary approaches: (1) inhibition of TNF-α expression, (2) overexpression of TNF-α, and (3) inhibition of mTNF-α shedding.

Inhibition of TNF-α gene expression during T. borreli infections was achieved by administration of PTX. PTX-treated fish showed impaired TNF-α but not iNOS gene expression and high plasma nitrite levels with kinetics corresponding to parasitemia (our unpublished data), indicating that parasite-derived components but not TNF-α directly contribute to the high nitrite levels typically associated with T. borreli infections. This is in line with our in vitro results showing that carp TNF-α did not induce NO production in phagocytes and corresponds to findings in TNFR1 morphant zebrafish embryos displaying normal iNOS staining during mycobacterium infections (39). Results showed that TNF-α is essential to control parasitemia since PTX-treated fish showed extremely high parasitemia numbers and succumbed faster to the infection. In mice, a crucial role for TNF-α in parasitemia control has been described in several studies on African trypanosomes where TNF-α-deficient mice show severely shortened survival times and fail to control parasitemia (36, 81, 82). The mechanisms underlining TNF-α activity are diverse and parasite species-specific. In T. congolense infections, TNFRI (TNFp55) signaling and soluble TNF-α have been shown to be crucial for NO-mediated parasite killing (82). In T. brucei infections, parasite control is independent from NO and it is TNF-α that might play a central role owing to its possible direct trypanolytic effects (32). In carp, the mechanisms by which TNF-α might contribute to the control of parasite load remain to be investigated. Although a direct lytic effect of carp TNF-α on T. borreli could not be detected in vitro (data not shown), the possibility that TNF-α might directly interact with T. borreli in vivo through a lectin-like interaction cannot be excluded. Independent of the exact mechanism involved, our results indicate that also in fish, TNF-α deficiency can make a difference in survival to parasitic infections.

Overexpression of carp TNF-α during T. borreli infections was achieved by i.m. injection of DNA plasmid encoding for carp TNF-α. Fish overexpressing TNF-α, regardless of the dose or isoform, all succumbed faster to the infection than did the control group, indicating that excess of TNF-α is detrimental to the host. Our results are in line with studies in mice showing that overexpression of TNF-α in vivo during malaria or T. cruzi infections result in increased mortality (29). In fact, as opposed to its beneficial effects on parasiteamia control, TNF-α has also been implicated in the immunosuppression and immunopathology typically associated with protozoan infections (29, 38). In our study, the adverse effects observed after administration of carp TNF-α could not be ascribed to increased parasitemia or plasma nitrite levels (data not shown). In contrast, treated fish were able to control the parasite load, since parasitemia decreased during the late phase of infection in treated as well as control fish. We have previously shown that Abs and complement are the main mechanisms responsible for parasite control and clearance of T. borreli (41), and the present results suggest that TNF-α does not interfere with this process. However, excess of TNF-α might directly contribute to pathology possibly through an exacerbation of the inflammatory response.

Inhibition of mTNF-α shedding during T. borreli infections was achieved by administration of an inhibitor of TACE. The compound used in this study (PFK242-484) has previously been shown to effectively reduce the inflammatory response associated with intestinal ischemia and reperfusion by inhibiting the production of soluble TNF-α (73). PFK242-484 has inhibitory effects not only on TACE but also on other matrix metalloproteinases (72); therefore, also in our study, FN439, a general matrix metalloproteinase inhibitor that has no effect on TACE was used as negative control. TNF-α is active not only as soluble but also as membrane-bound molecules, with both exerting unique and overlapping activities. mTNF-α has been reported to play a role during L. monocytogenes and mycobacterial infections (49, 50, 51), while in T. congolense infections soluble but not mTNF-α was necessary to control infection (82). In fish, the role of mTNF-α has not been investigated except for our preliminary studies. The TACE inhibitor was used in the period immediately before the peak of TNF-α transcription and to the peak of parasitemia, and results show that mTNF-α is a determining factor in protection since fish treated with the TACE inhibitor had extremely reduced splenomegaly and cleared the parasites within a period as short as 3 days. Although we cannot exclude that also in fish TACE might be involved in the shedding of other molecules besides TNF-α (83), given the clear involvement of TNF-α during T. borreli infections, the implications of increased levels of mTNF-α must have a great impact on the immune response to this parasite. The mechanisms responsible for the protective effects mediated by mTNF-α are unknown at present and could be several: (1) mTNF-α, similar to soluble TNF-α, might directly interact with the parasite through the lectin-like domain and cause direct parasite lysis. This mechanism does not provide the most likely explanation, however, because TNF-α-mediated trypanolysis of African trypanosomes requires accumulation in the endocytotic vesicles (34), and it seems unlikely that mTNF-α would be as easily endocytosed by the parasite as soluble TNF-α. (2) At the time of treatment, parasite-specific Abs are already present (data not shown) and TNF-α has almost reached maximum mRNA levels (this study). Thus, Abs present on leukocytes (via binding to Fc receptors) or on the parasite itself (opsonization) together with the high levels of mTNF-α on leukocytes might lead to a tight adhesion of the parasite to the surface of effector cells. In this context, mTNF-α would contribute to adhesion more than to direct lysis of the parasite. Ab-dependent cell-mediated cytotoxicity by IgM “armed” NK-like cells, as has been described for the channel catfish (Ictalurus punctatus) (84), potentially could play a role in this mechanism. In mice infected with T. congolense, parasitaemia is effectively cleared by macrophage/neutrophil-derived soluble TNF but not mTNF-α, as well as by intact TNFR1 signaling, which induces trypanolytic NO (82). This contrasts with our results in two ways: in our infection model mTNF-α does seem crucial for protection, and NO is not trypanolytic for T. borreli (44). (3) mTNF-α itself can mediate a “bidirectional signal” (85) whereby mTNF-α alone, independent from Abs, might act as receptor upon interaction with the parasite and transmit positive and/or negative feedback signals into the bearing cell. At present, we cannot firmly exclude nor confirm any of the above-mentioned mechanisms without extensive experimentation. Without doubt our results provide clear evidence for a yet unexploited functional role for membrane-bound TNF-α in fish that warrants further investigation.

Throughout this study we analyzed the biological activity of two isoforms of carp TNF-α. The most obvious difference was detected at the transcription level where TNF-α2, although expressed at very low levels, could be up-regulated to a greater extent than TNF-α1, both in vitro and in vivo. In general, recombinant TNF-α2 was more potent than TNF-α1 in promoting endothelial cell activation based on gene expression and on the ability of TNF-α-treated endothelial cell supernatant to promote leukocyte migration. In vivo, overexpression of TNF-α led to a lower survival in all groups, whereas fish injected with a high dose of plasmid encoding for TNF-α2 had the lowest survival and succumbed faster to the infection. Three-dimensional analysis of carp TNF-α showed major differences between the two isoforms, particularly in the region corresponding to the TIP domain. With regard to the lectin-like activity of TNF-α, at maximum concentrations, TNF-α1 and TNF-α2 were equally effective in killing T. brucei in vitro, whereas TNF-α2, in the presence of N,N′-diacetylchitobiose, constantly showed a slightly higher trypanolytic activity than did TNF-α1. Carp (14, 16), rainbow trout (13), goldfish (22), and zebrafish (17) all have multiple copies of TNF-α. At present, the evolutionary relationship between these TNF-α family members is not resolved (86). A recent study using the zebrafish ducttrip (dtp) mutant phenotype model showed the involvement of zebrafish TNF-α, but not TNF-β, in hepatic steatosis and liver degeneration (87). Our results show that both carp TNF-α isoforms examined in this study are biologically active. Further characterization of the biological relevance of the differences between TNF-α1 and TNF-α2 would require identification of the TNF-α receptors in carp.

Collectively, this study provides a comprehensive analysis, not only in vitro but also in vivo, of the biological activities of TNF-α in one fish species. Our data provide important insights in the functional conservation of TNF-α activities in teleost fish and point out similarities but also differences not only between mammals and fish, but also among different fish species. For the first time the lectin-like activity of TNF-α homologs in vertebrate species other than mammals has been investigated. Additionally, we provide evidence that, also in fish, a tight regulation of TNF-α expression is important, because depletion or excess of TNF-α can make an important difference to survival of infection. Finally, we propose a crucial protective role for mTNF-α, a yet unexploited function of TNF-α in fish.

We thank the central fish facilities, “De Haar-Vissen”, for taking care of the carp. We also thank Dr. F. Roca for the help provided with the endothelial cell cultures, Dr. V. Mulero for providing recombinant seabream TNF-α, and Dr. J. Zou and Dr. C. J. Secombes for providing recombinant trout TNF-α. Trudi Hermsen and Anja J. Taverne-Thiele are acknowledged for technical support and Carla S. Ribeiro for the help provided with the trypanolysis assay.

The authors have no financial conflicts of interest.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

This work was partly supported by the European Commission’s Improving Human Potential Program under Contract HPRN-CT-2001-00214 and QLK5-CT-2001-50988 and by the European Commission through Contract FP6007103 (Improved Immunity of Aquacultured Animals, IMAQUANIM).

2

The sequence presented in this article has been submitted to GenBank under accession no. GQ231486.

4

Abbreviations used in this paper: mTNF-α, membrane-bound TNF-α; cPBS, carp PBS; cRPMI, carp complete RPMI 1640 medium; DHR, dihydrorhodamine; EC, endothelial cell; EPC, Epithelioma papulosum cyprinid; HKL, head kidney leukocyte; iNOS, inducible NO synthase; LB, lysis buffer; PTX, pentoxifylline; TACE, TNF-α-converting enzyme.

1
Grell, M., P. Scheurich.
1997
. Tumor necrosis factor. D. LeRoith, and C. Bondy, eds. In
Growth Factors and Cytokines in Health and Disease
Vol. 2B
:
669
-726. JAI Press, Greenwich.
2
Peters, P. M., J. R. Ortaldo, M. R. Shalaby, L. P. Svedersky, G. E. Nedwin, T. S. Bringman, P. E. Hass, B. B. Aggarwal, R. B. Herberman, D. V. Goeddel, et al
1986
. Natural killer-sensitive targets stimulate production of TNF-α but not TNF-β (lymphotoxin) by highly purified human peripheral blood large granular lymphocytes.
J. Immunol.
137
:
2592
-2598.
3
Cuturi, M. C., M. Murphy, M. P. Costa-Giomi, R. Weinmann, B. Perussia, G. Trinchieri.
1987
. Independent regulation of tumor necrosis factor and lymphotoxin production by human peripheral blood lymphocytes.
J. Exp. Med.
165
:
1581
-1594.
4
Sung, S. S., J. M. Bjorndahl, C. Y. Wang, H. T. Kao, S. M. Fu.
1988
. Production of tumor necrosis factor/cachectin by human T cell lines and peripheral blood T lymphocytes stimulated by phorbol myristate acetate and anti-CD3 antibody.
J. Exp. Med.
167
:
937
-953.
5
Andersson, U., G. Adolf, M. Dohlsten, G. Moller, H. O. Sjogren.
1989
. Characterization of individual tumor necrosis factor α- and β-producing cells after polyclonal T cell activation.
J. Immunol. Methods
123
:
233
-240.
6
Chan, E. L., S. B. Haudek, B. P. Giroir, J. T. Murphy.
2001
. Human coronary endothelial cell activation by endotoxin is characterized by NF-κB activation and TNF-α synthesis.
Shock
16
:
349
-354.
7
Takimoto, T., K. Takahashi, K. Sato, Y. Akiba.
2005
. Molecular cloning and functional characterizations of chicken TL1A.
Dev. Comp. Immunol.
29
:
895
-905.
8
Takimoto, T., K. Sato, Y. Akiba, K. Takahashi.
2008
. Role of chicken TL1A on inflammatory responses and partial characterization of its receptor.
J. Immunol.
180
:
8327
-8332.
9
Mawaribuchi, S., K. Tamura, S. Okano, S. Takayama, Y. Yaoita, T. Shiba, N. Takamatsu, M. Ito.
2008
. Tumor necrosis factor-α attenuates thyroid hormone-induced apoptosis in vascular endothelial cell line XLgoo established from Xenopus tadpole tails.
Endocrinology
149
:
3379
-3389.
10
Hirono, I., B. H. Nam, T. Kurobe, T. Aoki.
2000
. Molecular cloning, characterization, and expression of TNF cDNA and gene from Japanese flounder Paralychthys olivaceus.
J. Immunol.
165
:
4423
-4427.
11
Laing, K. J., T. Wang, J. Zou, J. Holland, S. Hong, N. Bols, I. Hirono, T. Aoki, C. J. Secombes.
2001
. Cloning and expression analysis of rainbow trout Oncorhynchus mykiss tumour necrosis factor-α.
Eur J. Biochem.
268
:
1315
-1322.
12
Garcia-Castillo, J., P. Pelegrin, V. Mulero, J. Meseguer.
2002
. Molecular cloning and expression analysis of tumor necrosis factor α from a marine fish reveal its constitutive expression and ubiquitous nature.
Immunogenetics
54
:
200
-207.
13
Zou, J., T. Wang, I. Hirono, T. Aoki, H. Inagawa, T. Honda, G. I. Soma, M. Ototake, T. Nakanishi, A. E. Ellis, C. J. Secombes.
2002
. Differential expression of two tumor necrosis factor genes in rainbow trout, Oncorhynchus mykiss.
Dev. Comp. Immunol.
26
:
161
-172.
14
Saeij, J. P., R. J. Stet, B. J. de Vries, W. B. van Muiswinkel, G. F. Wiegertjes.
2003
. Molecular and functional characterization of carp TNF: a link between TNF polymorphism and trypanotolerance?.
Dev. Comp. Immunol.
27
:
29
-41.
15
Zou, J., C. J. Secombes, S. Long, N. Miller, L. W. Clem, V. G. Chinchar.
2003
. Molecular identification and expression analysis of tumor necrosis factor in channel catfish (Ictalurus punctatus).
Dev. Comp. Immunol.
27
:
845
-858.
16
Savan, R., M. Sakai.
2004
. Presence of multiple isoforms of TNF-α in carp (Cyprinus carpio L.): genomic and expression analysis.
Fish Shellfish Immunol.
17
:
87
-94.
17
Savan, R., T. Kono, D. Igawa, M. Sakai.
2005
. A novel tumor necrosis factor (TNF) gene present in tandem with the TNF-α gene on the same chromosome in teleosts.
Immunogenetics
57
:
140
-150.
18
Haugland, Ø., I. S. Mercy, K. Romøren, J. Torgersen, Ø. Evensen.
2007
. Differential expression profiles and gene structure of two tumor necrosis factor-α variants in Atlantic salmon (Salmo salar L.).
Mol. Immunol.
44
:
1652
-1663.
19
Nascimento, D. S., P. J. B. Pereira, M. I. R. Reis, A. do Vale, J. Zou, M. T. Silva, C. J. Secombes, N. M. S. dos Santos.
2007
. Molecular cloning and expression analysis of sea bass (Dicentrarchus labrax L.) tumor necrosis factor-α (TNF-α).
Fish Shellfish Immunol.
23
:
701
-710.
20
Ordas, M. C., M. M. Costa, F. J. Roca, G. Lopez-Castejon, V. Mulero, J. Meseguer, A. Figueras, B. Novoa.
2007
. Turbot TNFα gene: molecular characterization and biological activity of the recombinant protein.
Mol. Immunol.
44
:
389
-400.
21
Xiao, J., Z. C. Zhou, C. Chen, W. L. Huo, Z. X. Yin, S. P. Weng, S. M. Chan, X. Q. Yu, J. G. He.
2007
. Tumor necrosis factor-α gene from mandarin fish, Siniperca chuatsi: molecular cloning, cytotoxicity analysis and expression profile.
Mol. Immunol.
44
:
3615
-3622.
22
Grayfer, L., J. G. Walsh, M. Belosevic.
2008
. Characterization and functional analysis of goldfish (Carassius auratus L.) tumor necrosis factor-α.
Dev. Comp. Immunol.
32
:
532
-543.
23
Covello, J. M., S. Bird, R. N. Morrison, S. C. Battaglene, C. J. Secombes, B. F. Nowak.
2009
. Cloning and expression analysis of three striped trumpeter (Latris lineata) pro-inflammatory cytokines, TNF-α, IL-1β and IL-8, in response to infection by the ectoparasitic, Chondracanthus goldsmidi.
Fish Shellfish Immunol.
26
:
773
-786.
24
Zou, J., S. Peddie, G. Scapigliati, Y. Zhang, N. C. Bols, A. E. Ellis, C. J. Secombes.
2003
. Functional characterisation of the recombinant tumor necrosis factors in rainbow trout, Oncorhynchus mykiss.
Dev. Comp. Immunol.
27
:
813
-822.
25
García-Castillo, J., E. Chaves-Pozo, P. Olivares, P. Pelegín, J. Meseguer, V. Mulero.
2004
. The tumor necrosis factor-α of the bony fish seabream exhibits the in vivo proinflammatory and proliferative activities of its mammalian counterparts, yet it functions in a species-specific manner.
Cell. Mol. Life Sci.
61
:
1331
-1340.
26
Hino, K., O. Nakamura, Y. Yoshiura, H. Suetake, Y. Suzuki, T. Watanabe.
2006
. TNF induces the growth of thymocytes in rainbow trout.
Dev. Comp. Immunol.
30
:
639
-647.
27
Roca, F. J., I. Mulero, A. Lopez-Munoz, M. P. Sepulcre, S. A. Renshaw, J. Meseguer, V. Mulero.
2008
. Evolution of the inflammatory response in vertebrates: fish TNF-α is a powerful activator of endothelial cells but hardly activates phagocytes.
J. Immunol.
181
:
5071
-5081.
28
Pfeffer, K..
2003
. Biological functions of tumor necrosis factor cytokines and their receptors.
Cytokine Growth Factor Rev.
14
:
185
-191.
29
Clark, I. A..
2007
. How TNF was recognized as a key mechanism of disease.
Cytokine Growth Factor Rev.
18
:
335
-343.
30
Sherblom, A. P., J. M. Decker, A. V. Muchmore.
1988
. The lectin-like interaction between recombinant tumor necrosis factor and uromodulin.
J. Biol. Chem.
263
:
5418
-5424.
31
Lucas, R., S. Magez, R. De Leys, L. Fransen, J. P. Scheerlinck, M. Rampelberg, E. Sablon, P. De Baetselier.
1994
. Mapping the lectin-like activity of tumor necrosis factor.
Science
263
:
814
-817.
32
Magez, S., M. Geuskens, A. Beschin, H. del Favero, H. Verschueren, R. Lucas, E. Pays, P. de Baetselier.
1997
. Specific uptake of tumor necrosis factor-α is involved in growth control of Trypanosoma brucei.
J. Cell Biol.
137
:
715
-727.
33
Daulouede, S., B. Bouteille, D. Moynet, P. De Baetselier, P. Courtois, J. L. Lemesre, A. Buguet, R. Cespuglio, P. Vincendeau.
2001
. Human macrophage tumor necrosis factor (TNF)-α production induced by Trypanosoma brucei gambiense and the role of TNF-α in parasite control.
J. Infect. Dis.
183
:
988
-991.
34
Magez, S., M. Radwanska, B. Stijlemans, H. V. Xong, E. Pays, P. De Baetselier.
2001
. A conserved flagellar pocket exposed high mannose moiety is used by African trypanosomes as a host cytokine binding molecule.
J. Biol. Chem.
276
:
33458
-33464.
35
Aggarwal, B. B..
2003
. Signalling pathways of the TNF superfamily: a double-edged sword.
Nat. Rev. Immunol.
3
:
745
-756.
36
Magez, S., M. Radwanska, A. Beschin, K. Sekikawa, P. De Baetselier.
1999
. Tumor necrosis factor α is a key mediator in the regulation of experimental Trypanosoma brucei infections.
Infect. Immun.
67
:
3128
-3132.
37
Magez, S., C. Truyens, M. Merimi, M. Radwanska, B. Stijlemans, P. Brouckaert, F. Brombacher, E. Pays, P. De Baetselier.
2004
. P75 tumor necrosis factor-receptor shedding occurs as a protective host response during African trypanosomiasis.
J. Infect. Dis.
189
:
527
-539.
38
Antoine-Moussiaux, N., S. Magez, D. Desmecht.
2008
. Contributions of experimental mouse models to the understanding of African trypanosomiasis.
Trends Parasitol.
24
:
411
-418.
39
Clay, H., H. E. Volkman, L. Ramakrishnan.
2008
. Tumor necrosis factor signaling mediates resistance to mycobacteria by inhibiting bacterial growth and macrophage death.
Immunity
29
:
283
-294.
40
Saeij, J. P., W. B. Van Muiswinkel, A. Groeneveld, G. F. Wiegertjes.
2002
. Immune modulation by fish kinetoplastid parasites: a role for nitric oxide.
Parasitology
124
:
77
-86.
41
Saeij, J. P. J., B. J. de Vries, G. F. Wiegertjes.
2003
. The immune response of carp to Trypanoplasma borreli: kinetics of immune gene expression and polyclonal lymphocyte activation.
Dev. Comp. Immunol.
27
:
859
-874.
42
Saeij, J. P. J., W. B. van Muiswinkel, M. van de Meent, C. Amaral, G. F. Wiegertjes.
2003
. Different capacities of carp leukocytes to encounter nitric oxide-mediated stress: a role for the intracellular reduced glutathione pool.
Dev. Comp. Immunol.
27
:
555
-568.
43
Scharsack, J. P., D. Steinhagen, C. Kleczka, J. O. Schmidt, W. Körting, R. D. Michael, W. Leibold, H. J. Schuberth.
2003
. The haemoflagellate Trypanoplasma borreli induces the production of nitric oxide, which is associated with modulation of carp (Cyprinus carpio L.) leucocyte functions.
Fish Shellfish Immunol.
14
:
207
-222.
44
Forlenza, M., J. P. Scharsack, N. M. Kachamakova, A. J. Taverne-Thiele, J. H. Rombout, G. F. Wiegertjes.
2008
. Differential contribution of neutrophilic granulocytes and macrophages to nitrosative stress in a host-parasite animal model.
Mol. Immunol.
45
:
3178
-3189.
45
Grell, M., E. Douni, H. Wajant, M. Lohden, M. Clauss, B. Maxeiner, S. Georgopoulos, W. Lesslauer, G. Kollias, K. Pfizenmaier, P. Scheurich.
1995
. The transmembrane form of tumor necrosis factor is the prime activating ligand of the 80 kDa tumor necrosis factor receptor.
Cell
83
:
793
-802.
46
Decker, T., M. L. Lohmann-Matthes, G. E. Gifford.
1987
. Cell-associated tumor necrosis factor (TNF) as a killing mechanism of activated cytotoxic macrophages.
J. Immunol.
138
:
957
-962.
47
Higuchi, M., K. Nagasawa, T. Horiuchi, M. Oike, Y. Ito, M. Yasukawa, Y. Niho.
1997
. Membrane tumor necrosis factor-α (TNF-α) expressed on HTLV-I-infected T cells mediates a costimulatory signal for B cell activation: characterization of membrane TNF-α.
Clin. Immunol. Immunopathol.
82
:
133
-140.
48
Xu, J., A. K. Chakrabarti, J. L. Tan, L. Ge, A. Gambotto, N. L. Vujanovic.
2007
. Essential role of the TNF-TNFR2 cognate interaction in mouse dendritic cell-natural killer cell crosstalk.
Blood
109
:
3333
-3341.
49
Torres, D., L. Janot, V. F. Quesniaux, S. I. Grivennikov, I. Maillet, J. D. Sedgwick, B. Ryffel, F. Erard.
2005
. Membrane tumor necrosis factor confers partial protection to Listeria infection.
Am. J. Pathol.
167
:
1677
-1687.
50
Fremond, C., N. Allie, I. Dambuza, S. I. Grivennikov, V. Yeremeev, V. F. Quesniaux, M. Jacobs, B. Ryffel.
2005
. Membrane TNF confers protection to acute mycobacterial infection.
Respir. Res.
6
:
136
51
Allie, N., L. Alexopoulou, V. J. Quesniaux, L. Fick, K. Kranidioti, G. Kollias, B. Ryffel, M. Jacobs.
2008
. Protective role of membrane tumour necrosis factor in the host’s resistance to mycobacterial infection.
Immunology
125
:
522
-534.
52
Lucas, R., P. Juillard, E. Decoster, M. Redard, D. Burger, Y. Donati, C. Giroud, C. Monso-Hinard, T. De Kesel, W. A. Buurman, et al
1997
. Crucial role of tumor necrosis factor (TNF) receptor 2 and membrane-bound TNF in experimental cerebral malaria.
Eur. J. Immunol.
27
:
1719
-1725.
53
Irnazarow, I..
1995
. Genetic variability of Polish and Hungarian carp lines.
Aquaculture
129
:
215
-219.
54
Kemenade, B., A. Groeneveld, B. Rens, J. Rombout.
1994
. Characterization of macrophages and neutrophilic granulocytes from the pronephros of carp (Cyprinus carpio).
J. Exp. Biol.
187
:
143
-158.
55
Forlenza, M., P. D. Walker, B. J. de Vries, S. E. Wendelaar Bonga, G. F. Wiegertjes.
2008
. Transcriptional analysis of the common carp (Cyprinus carpio L.) immune response to the fish louse Argulus japonicus Thiele (Crustacea: Branchiura).
Fish Shellfish Immunol.
25
:
76
-83.
56
Farrell, A. P..
1987
. Coronary flow in a perfused rainbow trout heart.
J. Exp. Biol.
129
:
107
-123.
57
Koren, C. W., B. Sveinbjornsson, B. Smedsrod.
1997
. Isolation and culture of endocardial endothelial cells from Atlantic salmon (Salmo salar) and Atlantic cod (Gadus morhua).
Cell Tissue Res.
290
:
89
-99.
58
Steinhagen, D., P. Kruse, W. Körting.
1989
. The parasitemia of cloned Trypanoplasma borreli Laveran and Mesnil, 1901, in laboratory-infected common carp (Cyprinus carpio L.).
J. Parasitol.
75
:
685
-689.
59
Overath, P., J. Ruoff, Y.-D. Stierhof, J. Haag, H. Tichy, I. Dyková, J. Lom.
1998
. Cultivation of bloodstream forms of Trypanosoma carassii, a common parasite of freshwater fish.
Parasitol. Res.
84
:
343
-347.
60
Steinhagen, D., W. Hedderich, A. Skouras, J. P. Scharsack, J. Schuberth, W. Leibold, W. Körting.
2000
. In vitro cultivation of Trypanoplasma borreli (Protozoa: Kinetoplastida), a parasite from the blood of common carp Cyprinus carpio.
Dis. Aquat. Org.
41
:
195
-201.
61
Collet, B., C. J. Secombes.
2005
. Construction and analysis of a secreting expression vector for fish cells.
Vaccine
23
:
1534
-1539.
62
Fijan, N., D. Sulimanovic, M. Bearzotti, D. Muzinic, L. O. Zwillenberg, S. Chilmonczyk, J. F. Vautherot, P. de kinkelin.
1983
. Some properties of the Epitherlioma papulosum cyprini (EPC) cell line from carp cyprinus carpio.
Ann. Virol. (Institue Pasteur)
134
:
207
-220.
63
Liu, S., R. Tobias, S. McClure, G. Styba, Q. Shi, G. Jackowski.
1997
. Removal of endotoxin from recombinant protein preparations.
Clin. Biochem.
30
:
455
-463.
64
Pfaffl, M. W..
2001
. A new mathematical model for relative quantification in real-time RT-PCR.
Nucleic Acids Res.
29
:
e45
65
Tichopad, A., M. Dilger, G. Schwarz, M. W. Pfaffl.
2003
. Standardized determination of real-time PCR efficiency from a single reaction set-up.
Nucleic Acids Res.
31
:
e122
66
Gonzalez, S. F., M. O. Huising, R. Stakauskas, M. Forlenza, B. M. Lidy Verburg-van Kemenade, K. Buchmann, M. E. Nielsen, G. F. Wiegertjes.
2007
. Real-time gene expression analysis in carp (Cyprinus carpio L.) skin: inflammatory responses to injury mimicking infection with ectoparasites.
Dev. Comp. Immunol.
31
:
244
-254.
67
Sali, A., T. L. Blundell.
1993
. Comparative protein modelling by satisfaction of spatial restraints.
J. Mol. Biol.
234
:
779
-815.
68
Eswar, N., B. John, N. Mirkovic, A. Fiser, V. A. Ilyin, U. Pieper, A. C. Stuart, M. A. Marti-Renom, M. S. Madhusudhan, B. Yerkovich, A. Sali.
2003
. Tools for comparative protein structure modeling and analysis.
Nucleic Acids Res.
31
:
3375
-3380.
69
Dauber-Osguthorpe, P., V. A. Roberts, D. J. Osguthorpe, J. Wolff, M. Genest, A. T. Hagler.
1988
. Structure and energetics of ligand binding to proteins: Escherichia coli dihydrofolate reductase-trimethoprim, a drug-receptor system.
Proteins
4
:
31
-47.
70
Laskowski, R. A., M. W. Macarthur, D. S. Moss, J. M. Thornton.
1993
. PROCHECK: a program to check the stereochemical quality of protein structures.
J. Appl. Crystallogr.
26
:
283
-291.
71
Sippl, M. J..
1993
. Recognition of errors in three-dimensional structures of proteins.
Proteins
17
:
355
-362.
72
Kottirsch, G., G. Koch, R. Feifel, U. Neumann.
2002
. Beta-aryl-succinic acid hydroxamates as dual inhibitors of matrix metalloproteinases and tumor necrosis factor alpha converting enzyme.
J. Med. Chem.
45
:
2289
-2293.
73
Souza, D. G., F. L. Ferreira, C. T. Fagundes, F. A. Amaral, A. T. Vieira, R. A. Lisboa, M. V. Andrade, A. Trifilieff, M. M. Teixeira.
2007
. Effects of PKF242-484 and PKF241-466, novel dual inhibitors of TNF-α converting enzyme and matrix metalloproteinases, in a model of intestinal reperfusion injury in mice.
Eur. J. Pharmacol.
571
:
72
-80.
74
Van Ommen, R., A. E. Vredendaal, H. F. Savelkoul.
1994
. Secondary IgE responses in vivo are predominantly generated via γ1ε-double positive B cells.
Scand. J. Immunol.
40
:
491
-501.
75
Baeyens, K. J., H. L. De Bondt, A. Raeymaekers, W. Fiers, C. J. De Ranter.
1999
. The structure of mouse tumour-necrosis factor at 1.4 Å resolution: towards modulation of its selectivity and trimerization.
Acta Crystallogr. D Biol. Crystallogr.
55
:
772
-778.
76
Doherty, G. M., J. C. Jensen, H. R. Alexander, C. M. Buresh, J. A. Norton.
1991
. Pentoxifylline suppression of tumor necrosis factor gene transcription.
Surgery
110
:
192
-198.
77
Bates, J. M., J. Akerlund, E. Mittge, K. Guillemin.
2007
. Intestinal alkaline phosphatase detoxifies lipopolysaccharide and prevents inflammation in zebrafish in response to the gut microbiota.
Cell Host Microbe
2
:
371
-382.
78
Beschin, A., M. Bilej, F. Hanssens, J. Raymakers, E. Van Dyck, H. Revets, L. Brys, J. Gomez, P. De Baetselier, M. Timmermans.
1998
. Identification and cloning of a glucan- and lipopolysaccharide-binding protein from Eisenia foetida earthworm involved in the activation of prophenoloxidase cascade.
J. Biol. Chem.
273
:
24948
-24954.
79
Beschin, A., M. Bilej, L. Brys, E. Torreele, R. Lucas, S. Magez, P. De Baetselier.
1999
. Convergent evolution of cytokines.
Nature
400
:
627
-628.
80
Olivares Fontt, E., A. Beschin, E. Van Dijck, V. Vercruysse, M. Bilej, R. Lucas, P. De Baetselier, B. Vray.
2002
. Trypanosoma cruzi is lysed by coelomic cytolytic factor-1, an invertebrate analogue of tumor necrosis factor, and induces phenoloxidase activity in the coelomic fluid of Eisenia foetida foetida.
Dev. Comp. Immunol.
26
:
27
-34.
81
Iraqi, F., K. Sekikawa, J. Rowlands, A. Teale.
2001
. Susceptibility of tumour necrosis factor-α genetically deficient mice to Trypanosoma congolense infection.
Parasite Immunol.
23
:
445
-451.
82
Magez, S., M. Radwanska, M. Drennan, L. Fick, T. N. Baral, N. Allie, M. Jacobs, S. Nedospasov, F. Brombacher, B. Ryffel, P. De Baetselier.
2007
. Tumor necrosis factor (TNF) receptor-1 (TNFp55) signal transduction and macrophage-derived soluble TNF are crucial for nitric oxide-mediated Trypanosoma congolense parasite killing.
J. Infect. Dis.
196
:
954
-962.
83
Moss, M. L., L. Sklair-Tavron, R. Nudelman.
2008
. Drug insight: tumor necrosis factor-converting enzyme as a pharmaceutical target for rheumatoid arthritis.
Nat. Clin. Pract. Rheumatol.
4
:
300
-309.
84
Shen, L., T. B. Stuge, J. P. Evenhuis, E. Bengtén, M. Wilson, V. G. Chinchar, L. W. Clem, N. W. Miller.
2003
. Channel catfish NK-like cells are armed with IgM via a putative FcμR.
Dev. Comp. Immunol.
27
:
699
-714.
85
Eissner, G., W. Kolch, P. Scheurich.
2004
. Ligands working as receptors: reverse signaling by members of the TNF superfamily enhance the plasticity of the immune system.
Cytokine Growth Factor Rev.
15
:
353
-366.
86
Glenney, G. W., G. D. Wiens.
2007
. Early diversification of the TNF superfamily in teleosts: genomic characterization and expression analysis.
J. Immunol.
178
:
7955
-7973.
87
Matthews, R. P., K. Lorent, R. Manoral-Mobias, Y. Huang, W. Gong, I. V. Murray, I. A. Blair, M. Pack.
2009
. TNFα-dependent hepatic steatosis and liver degeneration caused by mutation of zebrafish S-adenosylhomocysteine hydrolase.
Development
136
:
865
-875.