cDNA cloning of porcine interleukin 2 by polymerase chain reaction

Porcine interleukin 2 (IL-2) cDNA was cloned by polymerase chain reaction (PCR), using primers derived from the corresponding bovine sequence. The resulting porcine DNA sequence encodes a 154 residue IL-2 primary translation product. Comparison of the mature, secreted form of porcine IL-2 with those of other species was carried out in an attempt to identify differences that might contribute to the observed differing species specificities.     

Isolation and cDNA cloning of a novel galanin-like peptide (GALP) from porcine hypothalamus

Galanin is a widely distributed neuropeptide with a variety of physiological functions. Three galanin receptor subtypes, GALR1, GALR2, and GALR3, have been reported. We isolated a novel galanin-like peptide (GALP) from porcine hypothalamus by observing its activity for increasing [(35)S]GTPgammaS binding to a membrane preparation of GALR2-transfected cells. The peptide had 60 amino acid residues and a non-amidated C terminus. The amino acid sequence of GALP-(9-21) was completely identical to that of galanin-(1-13). A cloned porcine GALP cDNA indicated that GALP was processed from a 120-amino acid GALP precursor protein. The structures of rat and human GALP-(1-60) were deduced from cloned cDNA, which indicated that the amino acid sequences 1-24 and 41-53 were highly conserved between humans, rats, and pigs. Receptor binding studies revealed that porcine GALP-(1-60) had a high affinity for the GALR2 receptor (IC(50) = 0.24 nM) and a lower affinity for the GALR1 receptor (IC(50) = 4.3 nM). In contrast, galanin showed high affinity for the GALR1 (IC(50) = 0.097 nM) and GALR2 receptors (IC(50) = 0.48 nM). GALP is therefore an endogenous ligand that preferentially binds the GALR2 receptor, whereas galanin is relatively non-selective.     

Antigen-independent activation of memory cytotoxic T cells by interleukin 2

Culture supernatants from mitogen- or antigen-activated murine spleen cells are capable of causing reexpression of specific cytolytic activity from inactive memory cytotoxic T lymphocytes (CTL) in the absence of the original priming antigen. We have demonstrated that memory CTL from cytolytically inactive day 14 MLC cells are induced to reexpress high levels of specific cytotoxic activity after incubation with IL 2. Highly purified IL 2 was shown to induce levels of lytic activity comparable with that induced by supernatants from secondary mixed lymphocyte cultures (secondary MLC SN), suggesting that only IL 2 is necessary for the reactivation process. Moreover, only Lyt-2+ cells are necessary for reactivation inasmuch as inactive MLC cells depleted of Lyt-1+ cells by treatment with antibody and complement, followed by FACS selection of Lyt-2+ cells, were efficiently reactivated by IL 2. Because IL 2 is considered a proliferative signal, we examined whether proliferation was requisite for reactivation of memory CTL by IL 2. In the presence of cytosine arabinoside, which effectively inhibited proliferation, IL 2 was capable of reactivating memory CTL as efficiently as antigen, thus implying a differentiative role for IL 2 in secondary CTL activation. Reactivation of CTL by IL 2 and antigen appear to be functionally distinct events, because antigen but not IL 2 could trigger immune interferon release, although either IL 2 or antigen induced high levels of cytotoxicity. We propose that resting, memory CTL retain a heightened level of expression of IL 2 receptors as compared with naive CTL precursors, and thus are able to respond directly to exogenous IL 2. The consequences of this are proliferation and reexpression of specific killing activity, but this signal is not sufficient to induce immune interferon secretion. Rather, it appears that a signal via the antigen receptor is necessary for release of this lymphokine.     

Molecular cloning and characterization of a cDNA encoding a novel antibacterial peptide, defensin, from the mulberry longicorn beetle, Apriona germari.

A full-length cDNA clone with high homology (62% mature peptide sequence identity) to an Acalolepta luxuriosa antibacterial gene, possessing a conserved cysteine-stabilized alphabeta motif, was cloned by screening an Apriona germari cDNA library. This gene (AgCRP) had a total length of 360 bp with an open reading frame of 207 bp, and encoded a predicted peptide of 69 amino acid residues. The mature AgCRP peptide was 27 amino acid residues long and had a cysteine-stabilized alphabeta motif of C...CXXXC...C...CXC consensus sequence, similar to insect defensins. Northern blot analysis revealed that the AgCRP exhibited fat body-specific expression and was up-regulated by wounding, bacterial or fungal challenge.     

cDNA cloning and characterization of interleukin 2-induced genes in a cloned T helper lymphocyte

Clonal expansion of antigen-specific lymphocytes is an important aspect of the immune response. Interleukin 2 (IL2) is largely responsible for the amplification of antigen-specific T cells. In this study, the changes in gene expression accompanying interleukin 2 stimulation of T cells are examined, using a cloned T helper lymphocyte line as a model system. To isolate cDNA clones of IL2-induced genes, a cDNA library was screened by differential hybridization. Twenty-one different cDNA clones were isolated by this method, comprising six glycolytic enzymes, vimentin, alpha-tubulin, beta-actin, gamma-actin, ERp99, elongation factor 2, ribosomal phosphoprotein P1, the DNA-binding protein dbpB/YB-1, as well as seven clones which do not correspond to any previously described sequences. These clones are used to study the time course of expression and the sensitivity to cycloheximide inhibition of IL2-induced mRNAs. In addition, the tissue specificity of the unidentified mRNAs is examined, and two of these are shown to be expressed at high levels in normal mouse brain, with much lower or undetectable levels in the other tissues tested. These cDNA clones will be useful in future studies to determine the molecular basis of IL2-induced gene expression.     

Purification and cDNA cloning of luxuriosin, a novel antibacterial peptide with Kunitz domain from the longicorn beetle, Acalolepta luxuriosa

We have purified a novel antibacterial peptide from the hemolymph of the coleopteran insect Acalolepta luxuriosa, of the family Cerambyocidae, and named it luxuriosin. This peptide showed growth-inhibitory activity against Micrococcus luteus and germination- and/or growth-inhibitory activity against the conidia from rice blast fungus, Magnaporthe grisea. The amino acid sequence determined by cDNA cloning identified luxuriosin as a peptide of 88 amino acids with a theoretical molecular weight of 10,368.34, containing a Kunitz domain.     

Effect of interleukin 2 on cytotoxic effectors: II. Long-term culture of NK cells

The present study reports the establishment of an NK (natural killer) cell line, designated as IL2-CE1. This cell line was maintained in active growth in culture with the supplementation of interleukin 2 (IL2). The cells gave cytotoxicity against a variety of murine tumors. There was no detectable cytotoxicity against three human tumors tested. The reactivity obtained with the murine tumors was generally correlated with the sensitivity of these tumor cells to NK cytotoxicity. With few exceptions, the reactivity was much higher for NK-sensitive targets like YAC and RL♂ 1 cells. There was very little cytotoxicity against NK-resistant targets like P815 and HFL/d. The reactivity of the effectors against YAC was susceptible to trypsin treatment and this effect was reversible. In determining the surface markers of IL2-CE1 cells, it was shown that over 90% of the cells had Thy-1.2 antigen despite their resistance to the lytic effect of anti-Thy-1.2 antibody. There were no detectable surface Ig or Fc receptors. Therefore, the IL2-CE1 cells were consistent with being NK cells. Although these cells gave high levels of cytotoxicity against murine tumors, in the tumor neutralization tests the IL2-CE1 cells failed to give any protection against an immunosensitive leukemic line, FBL-3. After the IL2-CE1 cells were cloned, it was found that different NK clones showed variations in their fine specificity. Our finding supports the presence of different populations or subsets of NK cells. Establishment of these NK clones in long-term culture should be helpful for the detailed characterization of the NK cells and aid in the determination of their significance in the immune surveillance against neoplasia.     

5-Azacytidine treatment of a murine cytotoxic T cell line alters interferon-gamma gene induction by interleukin 2

Genetic susceptibility to multiple sclerosis (MS) in Caucasians was previously shown to be correlated to the presence of given alleles at the HLA-DR and Gm loci. We now demonstrate that the humoral immune response in MS central nervous system (CNS) is modulated by both loci: the levels of IgG1 subclass and IgG1 allotypes in cerebrospinal fluid of MS patients depend on both their Gm genotype and their HLA-DR2 or HLA-DR7 phenotype. That HLA-DR molecules may either participate in a preferential recruitment of IgG1 allotype-producing B cells in MS CNS or act after such a selective homing is discussed. These results demonstrate that both HLA and Gm loci are synergistically involved in the modulation of the humoral immune response.     

Stage-related, cell-mediated cytotoxicity with effector cell analysis and computation of cytotoxic activity of T cells and non-T cells.

Cell-mediated cytotoxicity (CMC) by lymphocytes from patients with oral squamous cell carcinoma, as well as from nonmalignant control donors, was tested by a microcytotoxicity assay against a cultured cell line derived from an oral squamous cell carcinoma. In terms of the degree of CMC, stage-related cytotoxicity was observed. A further study of the effector cell analysis revealed that the cytotoxic effects of lymphocytes from both patients and control donors were largely attributable to non-T cells. However, the effectors were also stage related, and in early stage patients, T-cell-mediated cytotoxicity was super-imposed on non-T-cell-mediated cytotoxicity. This conclusion was further supported by the evidence of elevated cytotoxic activity of T cells in early stage patients, which was computed from simultaneous equations proposed in the present paper for computing the cytotoxic activity of T cells (CTA_T) and non-T cells (CTA_non-T).     

Characterization of interleukin 2-dependent cytotoxic T-cell clones. IV. Production of alpha, beta and gamma interferons and interleukin 2 by Lyt-2+ T cells

The production of alpha, beta and gamma interferons (IFN) and interleukin 2 (IL-2) by Lyt-2+-dependent cytotoxic T-cell lines/clones was investigated. Cloned and uncloned T-cell lines specific for H-2Dd or the unique RL male 1 leukemia antigen were studied. After infection with Sendai virus (SV) or Newcastle disease virus (NDV) all cell lines produced IFN-alpha and -beta. Induction of IFN-gamma was attempted with the mitogens Con A, PHA, PWM, SEA, and SEB, with poly(I:C), with antibodies Lyt-1.2, -2.2, and Thy-1.2, or with the target cells Meth A (H-2Dd+) and RL male 1. All mitogens were effective inducers. However, the antibodies and poly(I:C) were not. One uncloned RL male 1-specific cell line CTLL-RP, produced IFN-gamma after induction with RL male 1. Production of IFN-alpha, beta depended on IL-2, whereas production of IFN-gamma did not, although addition of highly purified IL-2 increased IFN-gamma production even in the absence of other inducers. Crude IL-2 inhibited the production of IFN-gamma but not IFN-alpha, beta. In response to mitogens, some T-cell clones also produced IL-2. The results demonstrate that Lyt-2+ cells can produce a broad spectrum of lymphokine activities after appropriate stimulation. Their availability now affords us the opportunity to study the regulation of lymphokine production at the clonal level.     

Suppression of alloimmune cytotoxic T lymphocyte (CTL) generation by depletion of NK cells and restoration by interferon and/or interleukin 2

By using rabbit antiserum to a glycolipid, ganglio-n-tetraosylceramide (ASGM1), the accessory effect of natural killer (NK) cells on the generation of alloimmune CTL in mice was investigated. When normal C3H/He mice were immunized with C57BL/6 or BALB/c spleen cells, they generated alloimmune CTL with a surface marker phenotype of Thy-1+ Lyt-1-2+ ASGM1-, preceded by early augmentation of cytotoxic activity of NK cells with a Thy-1-Lyt-1-2-ASGM1+ phenotype. Administration of anti-ASGM1 (10 microliters) in mice resulted in a complete depletion of NK activity and ASGM1+ cells in the spleen even 1 day after injection, but no changes in the proportions of T (Thy-1+) cells and their Lyt-1 and Lyt-2 subsets as revealed by an immunofluorescence analyzer (FACS) and phagocytic cells. When these anti-ASGM1-treated mice were immunized with allogeneic cells, they showed neither augmented NK activity nor generation of alloimmune CTL, and spleen cells isolated from these anti-ASGM1-treated mice produced no CTL response to alloimmunization in vitro. Normal spleen cells treated with the antiserum and complement in vitro also showed a complete NK depletion without any deterioration of T cells and their Lyt-1 and Lyt-2 subsets, and when stimulated with allogeneic cells they generated no CTL. Spleen NK (ASGM1+) cells were purified by Percoll-gradient centrifugations followed by complement-dependent killing of T cells with the use of anti-Thy-1 monoclonal antibody, and were further purified by panning methods with anti-ASGM1, giving a preparation consisting of greater than 90% ASGM1+, Ly-5+ cells, and less than 0.5% of Thy-1+, Lyt-1+, and Lyt-2+ cells. These purified ASGM1+ Thy-1- cells alone generated no alloimmune CTL in response to alloantigens, suggesting that ASGM1+ NK cells contained no precursors of alloimmune CTL. When added into NK-depleted spleen cells, they restored the normal alloimmune CTL response of the spleen cells, indicating that ASGM1+ fractions contained cells to provide an accessory function for CTL generation. Lyt-1+ cells purified by panning methods did not restore the CTL response of NK-depleted spleen cells. These results indicate that ASGM1+ NK cells, but not Lyt-1+ helper T cells contaminating ASGM1+ fractions at undetectable levels, are responsible for the accessory function. When these purified ASGM1+ Thy-1- cells were stimulated with allogeneic cells, they produced IL 2 and IFN.(ABSTRACT TRUNCATED AT 400 WORDS)     

Up-regulation of NK cytotoxic activity via IL-15 induction by different viruses: a comparative study.

IL-15 is a recently identified cytokine that belongs to the four alpha-helix bundle cytokine family and possesses biological activities similar to those of IL-2. Its ability to induce effectors of NK activity suggests its involvement in innate immunity. In this study, we analyzed the effect of different viruses (HSV, EBV, respiratory syncitial virus, vesicular stomatitis virus, influenza virus, reovirus, and Sendai virus) on the up-regulation of NK activity in vitro. Exposure of human PBMC to the these viruses resulted in an immediate up-regulation of NK activity of PBMC via IL-15 induction; this effect was abrogated in the presence of mAbs to IL-15. Results of experiments conducted in parallel using mAbs to IL-15, as well as to other cytokines (IL-2, IL-12, IFN-gamma, and TNF-alpha), clearly indicated that IL-15 was specifically responsible for the observed effect. Furthermore, supernatants of virus-infected PBMC cultures significantly enhanced NK activity of uninfected PBMC in vitro. An increase of IL-15 protein levels 20 h postinfection was also confirmed in a bioassay using the IL-2-dependent cell line CTLL. Kinetic analysis of IL-15 mRNA expression using a semiquantitative RT-PCR revealed that the level of IL-15 messages peaked at different time points (up to 12 h) postinfection, depending on the nature of the virus. Taken together, these results suggest that the IL-15 response of the host to viral infection and the subsequent NK cell activation represent an important effector mechanism of the innate immune surveillance of the host against viral infections.     

Generation of nonspecific murine cytotoxic T cells in vitro by purified human interleukin 2

Murine spleen cells developed into nonspecific cytotoxic cells within 72 hr of culture in the presence of highly purified sources of human interleukin 2. In whole spleen cell cultures, human interleukin 2 generated effector cells which were Thy 1.2⁺, Lyt 2.2⁺, resistant to γ irradiation (1000 R), and capable of lysing both H-2 compatible and incompatible targets. The effector cells generated in this manner were not restricted to classical natural killer cell-sensitive targets. If thymus-derived cells (T cells) were depleted from the spleen cell population before culture with human interleukin 2, the effector cells generated were enriched in effectors capable of lysing natural killer cell-sensitive targets. Interferon was not produced in interleukin 2-stimulated spleen cell cultures. In addition, heterologous antibody to murine -γ-interferon did not abrogate the generation of cytotoxic cells by human interleukin 2. These and additional data suggest that human interleukin 2 is capable of stimulating γ-irradiation-sensitive Thy 1.2⁺ cell(s) capable of lysing a variety of target cells regardless of inherent sensitivities to classical natural killer cells. Thy 1.2^? cells were also stimulated by human interleukin 2 and lysed only natural killer cell-sensitive targets. Human interleukin 2 caused some Thy 1.2^? cells to become susceptible to lysis by anti-Thy 1.2 serum and complement.     

Natural killer activity in cloned cytotoxic T lymphocytes: regulation by interleukin 2, interferon, and specific antigen

It has previously been shown that monoclonal antigen-specific mouse CTL lines can be induced to express cytolytic activity with the same specificity as that of splenic natural killer (NK) cells following culture in high concentrations of concanavalin A-induced spleen cell supernatants. In the present experiments, we made use of this in vitro system to explore the regulation of NK activity at the clonal level. Interferon-alpha and interferon-beta and interleukin 2 (IL 2) were potent inducers of NK activity in CTL, demonstrating that these substances can activate NK functions directly without the participation of other cell types. By comparison, IFN-gamma was a poor activator of NK activity in CTL (and also in fresh spleen cells). Three major differences between induction of NK activity by IFN-alpha,beta and IL 2 were noted: IFN induced NK activity selectively without affecting specific cytolysis, whereas IL 2 also enhanced specific killing; IFN acted much more rapidly than IL 2; and IFN did not induce the cells to enter the cell cycle nor were there any obvious morphologic changes. Specific antigen was also a strong inducer of NK activity in CTL, but studies with antisera against the various classes of IFN revealed that this effect was mediated, at least in part, via the release of IFN-beta. By contrast, the same antisera had no effect on NK induction by crude TCGF or by highly purified IL 2, indicating that the regulation of NK activity by IL 2 occurs at the clonal level in an IFN-independent manner. Although, IL 2, IFN, and Ag could apparently act alone to induce NK activity, much greater (synergistic) induction was obtained by various combinations of these regulators, suggesting that the delivery of two (or more) signals to the responder cell was required for full expression of the NK state. As with fresh splenic NK cells, the induced NK state in cloned CTL was intrinsically labile as revealed by its rapid decay in the absence of inducers, but it could nonetheless be maintained indefinitely at very high levels in the continued presence of inducers. This clonal system thus displays a responsiveness to regulatory signals exactly analogous to that of splenic NK cells and provides a unique and exciting opportunity to evaluate the biochemistry of the regulation of NK activity.     

cDNA cloning of porcine PKD2 gene and RNA interference in LLC-PK1 cells

Mutations in the PKD2 gene cause autosomal dominant polycystic kidney disease (ADPKD), a common, inherited disease that frequently leads to end-stage renal disease (ESRD). Swine show substantial similarity to humans physiologically and anatomically, and are therefore a good model system in which to decipher the structure and function of the PKD2 gene and to identify potential therapeutic targets. Here we report the cloning and characterization of the porcine PKD2 cDNA showing that the full-length gene (3370 bases) is highly expressed in kidney, with minimal expression in the liver. RNA interference (RNAi) is a promising tool to enable identification of the essential components necessary for exploitation of the pathway involved in cellular processes. We therefore designed four shRNAs and nine siRNAs targeting the region of the porcine PKD2 gene from exons 3 to 9, which is supposed to be a critical region contributing to the severity of ADPKD. The results from HeLa cells with the dual-luciferase reporter system and porcine kidney cells (LLC-PK1) showed that sh12 could efficiently knock down the PKD2 gene with an efficiency of 51% and P1 and P2 were the most effective siRNAs inhibiting 85% and 77% respectively of PKD2 expression compared with untreated controls. A subsequent functional study of the transient receptor potential polycystic (TRPP) 2 channel protein indicated that the decreased expression of TRPP2 induced by siRNA P1 and P2 could release the arrest of the cell cycle from G0/G1 promoting progression to S and G2 phases. Our data, therefore, provides evidence of potential knock-down target sites in the PKD2 gene and paves the way for the future generation of transgenic ADPKD knock-down animal models.     

Brief antigenic stimulation generates effector CD8 T cells with low cytotoxic activity and high IL-2 production

It is currently believed that a brief antigenic stimulation is sufficient to induce CD8 T cells to complete their differentiation program, become effector T cells, and subsequently generate memory. Because this concept was derived from studies in which only a single effector function was analyzed (either IFN-gamma production or target cell lysis), we wondered whether monitoring for multiple effector functions might reveal novel characteristics of effector CD8 T cells elicited by brief or prolonged Ag exposure. Using an in vitro system to generate effector T cells and an in vivo adoptive transfer model to track donor CD8 T cells, we found that the differentiation programs acquired by CD8 T cells after brief or prolonged antigenic stimulation were different. Although the frequencies of IFN-gamma and TNF-alpha producers were comparable for both effector CD8 T cell populations, there were major differences in cytotoxic potential and IL-2 production. Whereas prolonged (>24 h) Ag exposure stimulated effector CD8 T cells with high cytotoxic activity and low IL-2 production, brief (<24 h) stimulation generated effector CD8 T cells with low cytotoxic activity and high IL-2 production. The latter effector T cells rapidly converted into central memory-like CD8 T cells, exhibited long-term survival in adoptively transferred hosts, and gave robust recall responses upon Ag challenge. These data suggest that not all functions of effector CD8 T cells are equally inherited after brief or prolonged antigenic stimulation.     

Lack of reconstitution of nude mice alloreactivity by purified interleukin 2 and induction of non-H-2-specific effector cells by crude supernatants

To verify or to challenge the reports indicating that IL-2 was the only molecule involved in the reconstitution of nu/nu mice alloreactivity in vitro, Balb/c (H-2d) nu/nu spleen cells were primed in culture against C57/B16 (H-2b) in the presence of crude IL-2-containing supernatants or purified IL-2. The generation of cytotoxic effectors was evaluated against a panel of 51Cr-labeled target cells. Although crude IL-2-containing supernatants sustained the generation of cytotoxic effectors, purified "natural" IL-2 (from different origins) and recombinant IL-2 were not able to do so. Con A or PHA were identified as cofactors synergizing with IL-2 to induce effectors from nu/nu spleen cells. These effectors efficiently lysed EL4 (H-2b, tumor line), but not mitogen-induced blast cells from the same strain. They also lysed targets bearing irrelevant allogenic H-2 specificities. Cold competition experiments confirmed the lack of H-2 specificity of such effectors: lysis of EL4 cells (H-2b) was inhibited strongly by YAC-1 cells (H-2a, very sensitive to NK lysis) or P815 cells (H-2d, autologous to the nu/nu effectors). Our results clearly challenge earlier conclusions and indicate that IL-2 alone does not reconstitute nude mice alloreactivity. Crude supernatants containing IL-2 and mitogen induce nonspecific effectors with patterns of reactivity similar to those of activated natural killers. We think that the cytotoxicity observed in these conditions in nude mice results from the mitogenic triggering of some kind of prethymic killer cells which subsequently are expanded by IL-2.     

Characterization and cDNA cloning of hinnavin II, a cecropin family antibacterial peptide from the cabbage butterfly, Artogeia rapae

Hinnavins, together with lysozymes, are the main types of antibacterial peptides/proteins previously isolated from the larval haemolymph of the cabbage butterfly, Artogeia rapae as part of the humoral immune response to a bacterial invasion. One of these antibacterial peptides, named hinnavin II, was purified and characterized after cDNA cloning. The purified hinnavin II was more active against Gram negative than against Gram positive bacteria. Hinnavin II also showed a powerful synergistic effect on the inhibition of bacterial growth with purified lysozyme. The cDNA has a total length of 186 bp with a 114 coding region. The deduced protein sequence contains 38 amino acids with a coding capacity of 4142.8 Da. The result of a multiple sequence alignment and phylogenetic analysis with Clustal W indicated that mature hinnavin II showed an approximately 78.9% amino acid sequence identity with cecropin A and originated from a group containing mostly lepidopteran cecropins.     

Purification and cDNA cloning of a novel protease inhibitor secreted into culture supernatant by MDCK cells.

The infectivity of influenza viruses to host cells depends on the activation of the viral glycoprotein hemagglutinin (HA) by proteases. Starting from the observation that influenza virus replication in MDCK (Madin Darby canine kidney) cells was impaired by inactivation of trypsin in the culture fluids, we demonstrated that the inhibitory activity was resolved into two Trypsin-inactivating factors (TF), TF A (15 kDa) and TF B (11 kDa). N-terminal protein sequences of the factors revealed that TF A was a known Submandibular Protease Inhibitor (SPI) secreted in dog saliva, while TF B was a novel protein (renamed CKPI; canine kidney protease inhibitor). Following peptide mapping and protein sequencing of CKPI we obtained a 390 bp cDNA encoding a 130-amino-acid protein from MDCK cell total RNA. Protein sequence comparison showed a 63.8% identity with human secretory leukocyte protease inhibitor (SLPI), the molecule containing two conserved whey acidic protein (WAP) motifs, and we suggest that CKPI is thought to be the canine analogue of human SLPI. These results suggest that the inhibitory factors are secreted from MDCK cells, which are involved in prevention of virus replication, and applicable to the protection of host cells from virus infection.