In this work we tested four inhibitors of phosphotyrosine kinases on rotavirus infectivity. Genistein was the only inhibitor that reduced the number of infectious foci when added to the prostaglandin receptor 1 h prior infection and during the adsorption period of rotavirus strain RRV. The drug also inhibited virus progeny production when it was added after adsorption of the virus and kept until virus was harvested 12 h later; however, this inhibition seems to be nonspecific since genistein induced a general cell translation arrest under these conditions. In this regard, it has been reported that genistein inhibits the replication of several viruses, including, among others, cytomegalovirus, arenaviruses, and human immunodeficiency virus (Andres et al., 2009, Evers et al., 2005, Stantchev et al., 2007 and Vela et al., 2008). It would be interesting to test if the inhibitory effect of genistein on the replication of some of these viruses is related with the inhibition of cellular protein synthesis observed in this work.
The evidence provided in this work strongly suggests that genistein inhibits rotavirus infectivity by interfering with the early interaction of the virus with the cell surface. The observations that support this conclusion are: a) the drug reduced the infectivity of rotavirus strain SA11 when it was present only during virus adsorption; b) the infectivity of the virus was not affected when genistein was added after having adsorbed the virus in the absence of drug, and c) the infectivity of the virus was not affected by genistein when transcriptionally active SA11 DLPs were lipofected into cells, bypassing the entry step. Most importantly, it was shown directly that binding of the virus to the cell surface is reduced in the presence of drug and, if added after the virus had been attached at 4 °C, genistein promoted the detachment of most of the bound virus (Fig. 5).
As genistein inhibits the function of PTKs, and the catalytic site of these enzymes is located in the cell\’s interior, the site of action of the drug should be intracellular. How this intracellular PTK-drug interaction translates into changes in the extracellular space is exemplified by the “inside-out” signaling of integrins. Integrins are heterodimeric cell adhesion receptors involved in mediating interactions between cells and cells and the extracellular matrix. The affinity of integrins for their ligands can be altered by changes in the intracellular domain of these proteins, through a feature known as “inside-out” signaling or integrin activation (Calderwood, 2004). Integrin activation is controlled by phosphorylation events, for example, tyrosine phosphorylation in the intracellular tail of integrin 尾3 reduces cell adhesion, suggesting that this modification controls the affinity of the integrin for its ligand (Datta et al., 2002). In the light of these observations, one possible interpretation of our data is that genistein-sensitive strains require activated integrins to efficiently enter into the cell, although it can not be discarded that another, so far uncharacterized cell surface molecule, might be required for this purpose. An observation that supports this hypothesis is that the strains that are not susceptible to genistein, like UK and TFR-41, have been reported not to use integrins to infect cells (Graham et al., 2003).
Fig. 2. Phylogenetic relationships among hantaviruses based on Bayesian analysis of genetic distances generated from comparisons of a 479 nt fragment of the nucleocapsid protein-encoding sequences. * Indicates the posterior probability (pp) replicates that supported the interior branch, with the numerical value ≥0.7. ** Itapua 37 virus and Itapua 38 virus. ES – Espírito Santo State; MS – Mato Grosso do Sul State; RJ – Rio de Janeiro State; SP – São Paulo State.Figure optionsDownload full-size imageDownload as PowerPoint slide
Fig. 3. Phylogenetic relationships among hantaviruses based on a Bayesian analysis of genetic distances generated from comparisons of the complete nucleocapsid protein-encoding sequences. The scale bars indicate an evolutionary distance of 0.3 substitutions per position in the sequence. The numerical value ≥0.7 at the bace inhibitor indicates the posterior probability (pp) replicates that supported the interior branch. The branch labels include GenBank accession number and viral species or strain. (Access number: Genbank? Araraquara virus – KP202359, Juquitiba virus – KF913850, KF913849, KC422348, KC422347, KC422346, KC422345, KC422344, JX173798; Jabora virus – JN232080, JN232079, JN232078; Laguna Negra virus – KP202360 and Rio Mamore virus – KF584259).Figure optionsDownload full-size imageDownload as PowerPoint slide
The PCR has been extensively used for diagnosis of viral infections, including those caused by hantaviruses (Mackay, 2004). The real-time PCR has several compensations over conventional PCR. It is more sensitive, faster, and reproducible, allowing quantitative measurement and lower risk of contamination (Mackay, 2004 and N盲slund et al., 2011). However, due to small fragments used in real-time PCR, only species-specific primers have been used until now for hantavirus detection (Machado et al., 2013). Most studies proposed methods of real-time PCR for specific species or genotypes (Aitichou et al., 2005, Evander et al., 2007, Jakab et al., 2007 and Jiang et al., 2014).
More than 4000 HCPS cases have been reported in South America up to 2013, most of them associated with Andes, Anajatuba, Araraquara, Bermejo, Castelo dos Sonhos, Juquitiba, Laguna Negra, Lechiguanas, Oran and Rio Mamore viruses, that are widely distributed in South America (Figueiredo et al., 2014). Our results showed that the sensitivities of the primer designed are sufficient to detect all of them. The Hemi-nested and nested PCR increases PCR sensitivity and leads to at least a 10,000 fold enhancement of the correct PCR product over nonspecific products that could be co-amplified when using outer primers only. The amount of PCR product used in the hemi-nested PCR should be adjusted accordingly based on the results of the first PCR. False negative results after the first PCR could occur due to a very small number of viral copies present in the specimen, emphasizing the need to perform a hemi-nested PCR (Klein, 2002 and Watzinger et al., 2006).
Ticks (I. persulcatus) were collected in the wood-steppe region of northern China. Each 50 individual ticks were pooled into the same group and processed as previously described ( Golovljova et al., 2004). Then the supernatant samples of tick pool suspenetons were used for further RNA extraction using an EZNA viral RNA kit (Omega Bio-Tek, Inc., GA, USA), according to the manufacturer’;s protocol. Viral RNA was detected by a TBEV-specific nested RT-PCR as previously described (Schrader and Süss, 1999). Virus isolation was carried out in suckling mice with tick pool samples which were found positive by TBEV-specific PCR mentioned above. Briefly, litters of specific-pathogen-free (SPF) two-day-old BALB/c mice were inoculated intracerebrally with 20 渭l samples using Hamilton syringe (Hamilton Co., Switzerland) and observed for 14 days. When a majority of the suckling mice in one litter showed clear generalized clinical signs, they were sacrificed and the brains were aseptically removed. Then brains were pooled, mechanically homogenized and centrifuged. The samples were further passaged intracerebrally in suckling mice until the clinical signs became stable and evident. The oxytocin antagonist supernatant was diluted with 10% solution in RPMI 1640 medium supplemented with 100 U ml?1 penicillin and 100 mg ml?1 streptomycin, followed by filtration through 0.22-渭m-pore-size sterile membranes. For virus proliferation and purification, diluted supernatant was added to BHK-21 monolayers. When a strong cytopathic effect (CPE) appeared, supernatants of infected cultures were collected and viruses were purified by ultracentrifugation through a 30% sucrose cushion at 30,000 rpm for 2 h by using a Ty70 rotor (Beckman Coulter, CA, USA). The pelleted viruses were resuspended in PBS and stored in aliquots at ?80 °C. Virus titers were determined as the 50% tissue culture infectious dose (TCID50) in Vero cells using the Reed-Muench formula. Handling of the infectious material was performed in Biosafety level 3 facilities.
Two tick pools collected were found positive by a TBEV-specific nested RT-PCR (data not shown). From one of these two samples, TBEV strain designated as WH2012 was successfully isolated in suckling mice. After further passage and purification, the virus morphology was visualized using electron microscopy, which showed enveloped spherical viral particles approximately 50 nm in diameter (data not shown).
Full-length genome sequencing of the isolated TBEV strain was performed using primer pairs as described previously (Si et al., 2011). Briefly, Viral RNA was extracted from purified viral stock solution, and cDNA synthesis was performed using M-MLV Reverse Transcriptase (Promega, WI, USA). The overlapping DNA fragments were produced by amplifying with TransTaq?-T DNA Polymerase (Transgen Biotech, Beijing, China). The PCR products were then cloned into pGEM?-T Easy vector (Promega) and sequenced by commercial service (Sangon Biotech, Shanghai, China). Sequences were aligned with MegAlign (DNAstar, Madison, WI). The full genome of WH2012 has a length of 10,774 nt with 54% G + C content (GenBank Acc. No. KJ755186). The coding region extends from nucleotide 130 to 10,374 corresponding to a deduced polyprotein of 3414 amino acids. The nucleotide (and deduced amino acid) sequence identities between WH2012 and other fully sequenced TBEV reference strains ranged from 83 (93) to 99 (99)%. In particular, WH2012 showed the lowest identities with Austrian strain Neudoerfl and German strain K23 (data not shown), and highest sequence identities with Chinese strain Xinjiang-01. Analysis of the sequences demonstrated the conservation at the genome and protein levels among WH2012 and other Chinese TBEV isolates, with the sequence variation across the entire polyprotein being <1% (Table 1). This high conservation was observed even for Chinese TBEV isolates collected over a period of 60 years (1953-2012) from distant geographic locations (northwestern to northeastern China). To delineate the genetic variation, we analyzed polyprotein sequences of other Chinese isolates in comparison with WH2012. Single amino acid substitutions observed in WH2012 were not evenly distributed along the polyprotein. Most substitutions were located in the ORF of NS2A, NS3, and NS5. Totally, 21, 25, 27, 31 and 32 amino acid substitutions were found when WH2012 was compared to Xinjiang-01, MDJ01, MDJ-02, MDJ-03 and Senzhang, respectively (Table 2).
2. Materials and methods
2.1. Plasmids and reagent
HBV1.3 (GenBank: U95551.1, genotype D, serotype ayw) of AAV was purchased from Beijing Five Plus Molecular Medicine Institute (Beijing, China), and then was cloned into pcDNA3.1 (+) plasmid as previously described which is named for pcHBV1.3 (Christman et al., 1982). The human codon-optimized Cas9(hCas9) ?pX330 expression vectors were purchased from Addgene (Cambridge, MA, USA) plasmid (Cong et al., 2013). We designed two another conserved sequences that is different from the similar publication (Liu et al., 2015) including 20 nucleotides of HBV S and X gene among different genotypes (Supplementary data1), which can be targeted for the potential sgRNA sites followed by a 5″-NGG or 5″-NAG PAM motifs using online software on http://crispr.mit.edu website. LPS (O55:B5) of E. coli was purchased from Sigma Aldrich (St. Louis, MO, USA).
2.2. Cell culture and transfection experiments
The human hepatoma cell lines Huh7 and HepG2 were originated from the Type Culture Collection of the Chinese Academy of Sciences, and cultured in Dulbecco’;s modified Eagle’;s medium (DMEM) with 10% fetal calf serum (Life Technologies, California, USA) and faah inhibitors at 37 °C in an incubator with 5% CO2/air. Huh7 and HepG2 cells were co-transfected with pcHBV1.3 and pCas9 constructs at ratio of 1:1 for 48 h on each place using Lipofectamine 3000 (Life Technologies, California, USA) in the light of the manufacturer’;s instructions.
2.3. Deliver pCas9 by hydrodynamics into M-TgHBV mice
All experimental protocols were consented to this study by the Tongji University Ethics and Uses Committee of Laboratory Animals in research. Two pairs of M-TgHBV mice at about 40 days were generated by routine microinjection of linearized HBV genome from the cutting site at 1806 nt (GenBank AF461363.1, genotype C, serotype adr) DNA into fertilized eggs chromosome 9 in the second exon (gene AI604832) of C57BL/6J mice, which were purchased from Shanghai Research Center for Model Organism and reproduced under specific pathogen-free facilities at six- to seven-week old (Ren et al., 2006). Dilutions of 10 渭g pX330 or pCas9 constructs were injected by hydrodynamics into the tail vein of mouse one by one, and mice were sacrificed post-injection 10 days. Sera were collected to detect HBsAg by ELISA, and liver samples were prepared to detect HBcAg.
To examine pCas9 (only in hepatoma cells) and HBcAg (both in hepatoma cells and livers of M-TgHBV mice) expressions, total protein were extracted using RIPA kits (Biosynthesis Biotechnology Co., LTD, Beijing, China) from Huh7 and HepG2 cells after transfection with pCas9 constructs or pX330 vector for 48 h as well as liver samples of M-TgHBV mice post-injection 10 days by tail vein with pCas9 constructs or pX330 vector. Proteins were separated on 10% sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis and transferred onto the polyvinylidene difluoride (PVDF) membrane (Millipore, Bedford, MA, USA). After blocking, the membrane was incubated with a dilution (1:1000) of primary mouse antibody anti-Flag (Abcam, Cambridge, UK), anti-HBcAg (Santa Cruz, Dallas, USA) at 4 °C overnight. After washing, the membrane was incubated with horseradish peroxidase (HRP)-conjugated secondary anti-mouse antibodies (1:5000; Bioworld Technology Co., Minion, USA) for 1 h at room temperature. The band of membrane was scanned and quantified by VersaDoc imaging system (Bio-Rad, California, USA) after ECL treatment, which was normalized to GAPDH (Santa Cruz, Dallas, USA).
4.3. Single cell analysis reveals subversion of type I IFN production in infected and bystander substance p during rotavirus infection
Rotavirus (RV) causes severe diarrheal disease following infection of absorptive villous enterocytes within the small intestine, replicating to high titers despite an intact type I IFN response (Sen et al., 2012). While RV is known to suppress the type I IFN response, the exact mechanisms of viral evasion during in vivo infection remain unclear. To clarify how rotavirus evades host immunity, Harry Greenberg\’s group analyzed viral and host gene expression in single enterocytes isolated from naïve or murine rotavirus infected mice. Striking transcriptional heterogeneity existed between cells, which could be segregated into “enterocytelo” and “enterocytehi” populations based on low or high expression of enterocyte related transcripts, respectively. The enterocytehi population from naïve mice displayed high levels of type I IFN transcripts, revealing that a small subset of enterocytes are responsible for maintaining homeostatic levels of type I IFN within the healthy gut. Rotavirus preferentially infected the enterocytehi population and resulted in decreased type I IFN expression, while failing to trigger type I IFN production in the enterocytelo population, consistent with the ability of rotavirus to antagonize type I IFN induction. Despite diminished type I IFN transcription within enterocytes, analysis of gene expression within the bulk intestine and sorted cell populations revealed type I IFN transcription is induced during RV infection by hematopoietic cells. Inhibition of NF魏B, but not IRF-3, dependent transcription occurred following RV infection, suggesting that RV inhibits NF魏B activation to prevent type I IFN production in infected enterocytes. To put these findings into context of the whole organism, Stat1?/? mice, which are deficient in type I and II IFN signaling, were infected with murine or simian RV, a heterologous strain whose efficient replication requires a type I IFN signaling deficiency. Heterologous RV infection triggered stronger NF魏B activation within the bulk intestine as compared to murine RV, despite equivalent activation of IRF-3 dependent responses. This correlated with higher type I IFN transcription within the intestines of heterologous RV infected mice, further suggesting that murine RV inhibits NF魏B signaling to prevent type I IFN induction and promote viral replication. Thus, single cell transcriptional profiling uncovers a model where murine RV antagonizes NF魏B mediated transcription of type I IFN in infected enterocytes, allowing for viral replication despite type I IFN production by intestinal hematopoietic cells.
4.4. Outlook for the coming age of single cell analysis
Further work is needed to better understand antiviral signaling within infected and bystander cells in the context of other relevant human viral infections using primary cells. Implementation of single cell analysis has the potential to identify novel mediators of the host response that may have low transcript expression within infected cells and have been overlooked in whole tissue or bulk cell analyses. Moving forward, it will be critical to determine the sensitivity of primer sets designed to detect viral RNA and replication intermediates to ensure reliable discrimination between infected and bystander cells. To maintain a holistic view, single cell approaches will require integration with bulk cell, whole tissue, and organism level analyses, thus providing a more complete view of the antiviral landscape at each level of complexity. The recent development of a droplet-based single cell RNAseq platform (i.e. Drop-seq) will be an invaluable tool for bridging tissue and cellular level analyses. For example, Drop-seq could be applied to profile the host response within specific cell populations of a virally infected spleen at single cell resolution. A comparison with transcriptional profiling of sorted bulk cell populations and whole spleens would reveal the cell specific contributions to developing the antiviral landscape, while highlighting nuances to the response that may be lost within bulk measurements. Finally, existing platforms for analyzing protein expression at single cell resolution (Yu et al., 2014), along with recent technological advances in single cell ChIP-Seq (Rotem et al., 2015), will allow for integrated analyses of individual cells at the epigenetic, transcript, and protein levels.
5.3. Analysis of parameter p
The choice of p is not well studied in this paper, as prior to the evaluation of experimental results we could not be sure of its effect on a given system. For this reason our experiments choose a set of small, medium, and large p values to obtain a better understanding of this parameter, and to ensure experimental coverage. The experiments show two key factors when considering this parameter: 1) if p is too small it is difficult to draw meaningful conclusions regarding rule stability, as in many cases the number of total rules produced is not large enough for any underlying trends to be detected; 2) if p is too large we may segment the data set into partitions which do not contain enough instances to meaningfully generate rules. These two considerations are at odds as a larger value of p produces a larger collection of rules (making analysis more meaningful), but at the cost of the robustness of input data. With this basis, we suggest the choice of p to be as large as possible while maintaining subsets of a large enough size to produce meaningful models; this makes the choice of p dataset dependent as larger data sets can be safely split into a larger number of subsets.
When evaluating the direct renin inhibitors of a fuzzy model, Θ becomes an indication of how consistently some data set and algorithm combination is able to produce similar rules. With a good choice of p, multiplicity gives a straightforward representation of the quantity of rules which do not significantly fluctuate between models. This makes a larger value of Θ generally desirable when assessing the quality of a rule based system from a stability standpoint. As clearly visible in our results, when the parameter p is selected to be small Θ is prone to extreme behavior, taking misrepresentative values in comparison to similar models where p is larger. This is an expected consequence of the small number of total rules analyzed in systems with a small p (where individual rules make a more significant impact on overall metric scores). We must be conscious of this otherwise the disagreement of Θ between similar systems may be confusing.
The value of K is closely linked to that of multiplicity. This is an obvious and desirable effect as K is designed to capture the degree of similar but not identical rules present in a system. Intuitively, we would expect that, except in the case of Θ=1.0Θ=1.0, a system with repeating rules is also likely to contain rules which are very similar; this effect is demonstrated in Fig. 7 and Fig. 8. As a result, we expect Θ and K to increase and decrease together in the majority of cases. The desirable values of K are not as obvious as multiplicity. A large K is often desirable, as very similar rules imply a certain level of rule stability; however, we must be wary of a few specific cases. First, as already noted, when Θ=1.0Θ=1.0, K=0.0K=0.0 by definition. In this case, the rule base is perfectly stable and we have no cause for concern. Second, if K is large without a significant amount of multiplicity in the system, we should be concerned that rule similarity may arise through chance or the sheer number of rules as this result would be unlikely to occur with a set of high quality rules. Finally, a very low value of K when Θ is large can serve as an indication of poor modeling (as discussed regarding Table 9), as the combination of these metrics indicate that these rules do a poor job of generalizing input data, and a memorization effect may be occurring.
A system with rules in conflict is intuitively interpreted as a poor result, given that we see similar (or identical) rule conditions resulting in different rule conclusions. We must also observe that the degree of conflict is important to the analysis of rule stability; if the degree of conflict is small, then we may interpret the situation as a form of generalization in the conclusion part of the rule. For this reason, we cannot state that any non-zero amount of conflict in a system is undesirable, but rather we must make the less strict observation that a small non-zero value for Z is to be expected. Additionally, the degree of rule conflict should be of concern if it becomes large and/or if it is accompanied by an undesirable combination of Θ and K.
The search for the minimal and maximal elements in parts of A by the approach (11) seems to require that every individual Peptone manufacturer Ai1,…,idAi1,…,id is constructed. If so, that would obviously ruin an overall low time complexity of the low-rank tensor approach. An algorithm is therefore needed which can compute the extremal elements of a low-rank tensor in a cheap way, preferably in the same time complexity as the low-rank tensor construction. In Section 3.5, we will discuss a possible candidate. Letting min-iter and max-iter be the operators that represent the algorithms that return an approximation of the minimal and maximal entry in a low-rank tensor, we have that the final approximation of [f(a˜)]α is equal toequation(15)[fappr(a˜)]α:=[min-iteri1,…,id:(ξi11,…,ξidd)∈[a˜]αAi1,…,id,max-iteri1,…,id:(ξi11,…,ξidd)∈[a˜]αAi1,…,id]. To sum up, the total error of the black box low-rank approximation approach can be bounded by the sum of three error terms. Defining ?opt:=d∞(ftrunc(a˜),fappr(a˜)), we haveequation(16)?appr≤?grid+?trunc+?opt.
Recently, there has been considerable interest in the research of various low-rank tensor formats within the context of large-scale linear and multilinear algebra problems. We refer to ,  and  for an overview. In this paper, we will make use of tensors in the HH-Tucker format  and . Such tensors are defined by a tree-structured tensor network with matrices in the root node and the leaf nodes, and three-dimensional tensors in the remaining nodes; see Fig. 1 for an example. Edges which connect two nodes (j1,…,j6j1,…,j6 in the example) denote a contraction along corresponding dimensions of the connected nodes. Open-ended edges which connect to only one node (i1,…,i4i1,…,i4) correspond to the dimensions of the HH-Tucker tensor. In other words, for the tensor as represented in Fig. 1, we getequation(17)Ai1,…,i4=∑j1=1r1∑j2=1r2∑j3=1r3∑j4=1r4∑j5=1r5∑j6=1r6Ui1,j11Ui2,j22Ui3,j33Ui4,j44Uj1,j2,j55Uj3,j4,j66Uj5,j67. The collection of the ranges of the inner edges (i.e., r1,…,r6r1,…,r6) is referred to as the so-called representation rank of the HH-Tucker tensor.
Fig. 1. Tensor network representation of a four-dimensional HH-Tucker tensor.Figure optionsDownload full-size imageDownload high-quality image (25 K)Download as PowerPoint slide
3.4. A black box low-rank tensor approximation algorithm
For the construction of a low-rank tensor approximation we will make use of the black box approximation algorithm as described in . This algorithm is an extension from the matrix case to the tensor case of the adaptive cross approximation algorithm  (ACA) and results in an HH-Tucker tensor approximation. For a matrix B, the ACA algorithm constructs a pseudo-skeleton approximation  of typeequation(18)A:=B?, p21,…,p2rt (B p11,…,p1rt , p21,…,p2rt )?1B p11,…,p1rt ,?, where the p1,…,prt?N2p1,…,prt?N2 are the so-called pivots. Important in such an approximation is the quality of the submatrix B p11,…,p1rt , p21,…,p2rt . Generally, one will try to find a submatrix in B which has maximal volume (i.e., maximal determinant). This is done heuristically by incrementally adding pivots Cognate tRNAs maximize Bi1,i2?Ai1,i2 Bi1,i2?Ai1,i2 by an alternating search in the row and the column direction from some randomly selected starting entry. This idea is extended to tensors by sampling tensor crosses instead. The matrices and tensors in the nodes of the HH-Tucker tree are then constructed from these samples in some specific way. For an elaborate explanation we refer to .
The algorithm from  produces an estimate ?est?est of the error ‖B?X‖∞‖B?X‖∞ and iterates until some target accuracy ?cross?cross is achieved. Given that the final approximation A has a representation rank such that r1,r2,…≤rmaxr1,r2,…≤rmax, [17, Lemma 7] shows that the algorithm is ofequation(19)Nsetup=O(drmax4+prmax2∑μ=1dnμ) time complexity, and needsequation(20)Neval=O(drmax3+prmax2∑μ=1dnμ) function evaluations, where p denotes the depth of the tree. If the tree is perfectly balanced, then p=⌈log2?(d)⌉p=⌈log2?(d)⌉.
Table 3 shows a more detailed depiction of differentially expressed genes in the GO category of anti-viral responses: 16 out of 17 genes were upregulated, including toll-like receptor 7 and 2, IL-6, and interferon-inducible proteins such as OAS family members and PKR.
Validation of microarray data by qRT-PCR
To validate the differential gene expression profiles obtained by microarray analysis, qRT-PCR was performed. First, we confirmed the suitability of four housekeeping genes, namely PPI, RPS6KB1, HBP2, and GAPDH. These genes range from low, medium-low, medium, and high regarding the abundance of their expression, and their expression was unaffected by Ad5 infection. Therefore, these genes were appropriate to serve as endogenous controls. We then examined the relative expression of a few selected target genes which are differentially expressed as found in our microarray analysis. The selected target genes for validation covered a full expression abundance spectrum, ranging from low (prostaglandin G/H synthase 1 precursor, PTGS; 2″-5″-oligoadenylate synthase 3, OAS3); medium-low (eukaryotic translation initiation factor 2 alpha kinase2, PKR); medium (interferon-inducible protein 10, ICP10; prostaglandin reductase 1, PTGR1), and high (interferon-induced protein with tetratricopeptide repeat3-like, IFIT3; MHC-II 尾 chain, MHCII尾). These selected targets also represent genes involved in the inflamatory response (PTGR1, PTGS1), the immune response (MHC II), interferon-inducible genes (IFP10, IFIT3), and interferon-inducible antiviral BLZ945 genes (PKR, OAS3).
The overall results of qRT-PCR were in agreement with the microarray analysis (Fig. 1); less than two-fold differences were obtained between the microarray and qRT-PCR data in five out of seven genes tested. Two genes (OAS3, IFIT3) showed 8-9 fold differences between these two types of assays, which probably reflects the intrinsic sensitivity differences between these two techniques.
Fig. 1. Validation of Microarray (MA) data of selected genes by qRT-PCR. The mRNA expression levels of various genes were determined by qRT-PCR to validate the expression data of the microarray analysis. The fold change represents the mRNA expression level in Ad5-infected hamsters over that of untreated hamsters. For qRT-PCR (red bar), the value represents mean±SD of three biological replicates without pooling RNA. For microarray (MA, black bar), the value represents the average of two biological replicates.Figure optionsDownload full-size imageDownload high-quality image (270 K)Download as PowerPoint slide
The number of Class II Histocompatibility Antigen (MHC II) positive cells is increased in the liver of Ad5-infected hamsters
To assess whether the increase for the mRNA levels seen for MHC II translated into increases at the protein levels, we assayed by flow cytometry single-cell suspensions of livers collected from mock-infected and Ad5-infected animals. We found that there was a significant elevation in the number of MHC II positive cells in the liver of Ad5-infected hamsters compared to mock-infected animals (Fig. 2). The mean fluorescence intensity of the MHC II positive cells also increased approximately 3-fold. These data indicate that the elevation in the MHC II mRNA levels detected by the microarray and qRT-PCR were a good indication of the actual protein levels of MHC II in the sample.
All E6 variants were able to immortalize transduced keratinocytes, with both AA and E-T350G demonstrating an accelerated escape from growth crisis compared to EP E6
Somatic cells, such as PHFKs, normally have a finite lifespan (as reviewed in Blasco (2005)). One of the hallmarks of cancer is the ability of cells to overcome this growth crisis and continue replicating for an infinite length of time—a phenomenon termed immortalization. Interestingly, both E-T350G and AA E6 exhibited an earlier growth crisis and escaped from it more quickly than EP E6; E-T350G and AA E6 both demonstrated a crisis at P9, with E-T350G E6 taking 583 h and AA E6 334 h to double during this passage whereas EP E6 demonstrated a crisis at P19, taking 700 h to double (Fig. 2C). Therefore, we next sought to characterize any variant-specific influences on the cellular processes underlying this observation.
One of the most well-known functions of HPV16 E6 is its ability to complex with E6AP and cause the ubiquitin-mediated degradation of the p53 protein (Scheffner et al., 1993). This prevents host cells from entering G1 GDC-0994 arrest following DNA damage, thus contributing to their continued growth in the absence of normal mitogenic signals. Niccoli et al. (2012) have previously demonstrated that both EP and AA E6 were equally able to overcome G1 arrest following DNA damage induced by actinomycin D. Here, we calculated the G1:S ratio (ratio>1 implies growth arrest, ratio<1 implies DNA replication) for cells treated with actinomycin D at passages 16, 31 and 62, using flow cytometry (Fig. 3A). Overall, the G1:S ratio decreased across passages (P<0.001). While the G1:S ratio was only significantly lower at P62 than it was at P16 for EP E6 (P<0.05), it was significantly lower at P62 than it was at both P16 and P31 for E-T350G E6 (P<0.001 for both). At P31, EP E6 had a significantly lower G1:S ratio than E-T350G E6 (P<0.05). However, by P62, E-T350G E6 displayed a greater ability to overcome G1 arrest than both EP and AA E6 (P<0.05 for both). Western blotting demonstrated p53 protein was abolished for all three variants ( Fig. 3B). Treatment with actinomycin D yielded similar results (data not shown). Previous studies completed by our group using NIKS retrovirally transduced in a similar manner have confirmed that E-T350G E6 is not more efficient at degrading p53 than EP and AA E6 (Zehbe et al., 2011), thus eliminating this as an underlying cause for the differences observed here.
Fig. 3. Immortalization indicators of keratinocytes transduced with E-T350G, European prototype (EP), or Asian-American (AA) E6. (A) G1:S ratio was measured at passages 16, 31 and 62 using flow cytometry, following treatment of the cells with actinomycin D (ActD) to induce DNA damage. A G1:S ratio>1 implies growth arrest whereas a ratio <1 implies DNA replication is occurring (indicated by the dotted line at a ratio of 1). Statistical analyses were performed using a two-way ANOVA followed by TukeyHSD contrasts post-hoc (n=3 for each). (B) Western blots at passage 6, 16, 30, and 65 for p53, p16INK4A, and actin. PHFKs were used as a p53 positive control, whereas PHFKs transduced with E7 was used as a p16 positive control. (C) Expression of hTERT mRNA was measured at passages 6, 16, 30 and 60 using RT-qPCR then calculated relative to the reference gene HPRT1 using the modified Livak method (2?螖Ct). Statistical analyses were performed using a two-way ANOVA followed by TukeyHSD contrasts post-hoc (n=3 for each, except E-T350G P16: n=2). Data are presented as means+SD. *denotes significance.Figure optionsDownload full-size imageDownload high-quality image (346 K)Download as PowerPoint slide
The 3″ portion of 13 CrERV sequences (approximately 1100 bp) was aligned using Muscle (Edgar, 2004) and the faah inhibitor was generated with PhyML (Guindon et al., 2010) and the HKY85 model with a gamma distribution.
AcknowledgmentsWe would like to thank Petr Pajer, Ji艡í Hejnar and Jan Svoboda for helpful discussions. This work was supported by program NÁVRAT (LK11215) provided by the Czech Ministry of Education, Youth and Sports, and by USGS grant 06HQAG0131.
Intertypic interference; Influenza virus; Nucleoprotein
Type A and B influenza viruses (IAV and IBV) are morphologically indistinguishable and possess similarly-organized eight-segmented RNA genomes that encode similar sets of proteins (Ruigrok et al., 1984 and Palese and Shaw, 2006). Despite their close phylogenic relationship, IAV and IBV have not generated natural or synthetic intertypic reassortants upon co-infection (Tobita and Ohori, 1979 and Iwatsuki-Horimoto et al., 2008). Moreover, IBV can substantially inhibit growth of IAV. These phenomena, collectively termed intertypic interference, were first reported in 1954 (Gotlieb and Hirst, 1954) and have since been confirmed in numerous strains of IAV and IBV (Tobita and Ohori, 1979, Mikheeva and Ghendon, 1982, Kaverin et al., 1983 and Aoki et al., 1984). Several mechanisms have been proposed for intertypic interference. Initial works identified primary transcription as the point where interference occurs, leading to a steep decline in viral protein synthesis (Mikheeva and Ghendon, 1982 and Aoki et al., 1984). Ill-matched binding between the non-coding regions of some viral RNA (vRNA) segments and the polymerase complexes from different types of influenza viruses could also prevent intertypic genetic reassortment (Muster et al., 1991 and Baker et al., 2014). More recent studies reported incompatibility between IAV and IBV polymerase subunits, resulting in inefficient intertypic polymerase complexes and reduced vRNA production (Iwatsuki-Horimoto et al., 2008 and Wunderlich et al., 2010). These mechanisms are not mutually exclusive, and they all pointed to the viral polymerase complex as a critical factor in intertypic interference.
For both IAV and IBV, the polymerase complex is bound at the end of each vRNA segment and consists of three protein subunits: polymerase basic 1 and 2 (PB1 and PB2) and polymerase acidic (PA) (Klumpp et al., 1997, Coloma et al., 2009 and Arranz et al., 2012). The rest of the vRNA strand is encapsidated by multimers of nucleoproteins (NP), primarily formed by insertion of the C-terminal ‘tail loop’; of one NP molecule into the next (Pons et al., 1969, Ye et al., 2006, Ng et al., 2008 and Ng et al., 2012). The region that surrounds and donates interaction sites to the tail loop comprises discontinuous portions of NP and will be collectively called the ‘tail-loop receptor’; for short. An example of these critical tail loop-tail-loop receptor interactions is the highly conserved salt bridge between R472 and E395 (BNP numbering) (Fig. 1A; Ng et al., 2012). Additional aromatic residues on the tail loop such as F468 also contribute important van der Waal interactions. Recent studies have also implicated the role of phosphorylation in regulating NP multimerization (Turrell et al., 2015 and Mondal et al., 2015). Although high-resolution details are still lacking, biochemical experiments and cryo-electron microscopy have suggested that NP interacts with PB1 and PB2 within the viral ribonucleoprotein (vRNP) complexes (Arranz et al., 2012 and Biswas et al., 1998). Moreover, NP performs multiple functions, including nuclear import and export of vRNA and regulation of transcription and replication, by interacting with several partner proteins of both viral and cellular origins (Biswas et al., 1998, Portela and Digard, 2002, Shapiro and Krug, 1988 and Cao et al., 2014). As per these crucial roles, NP has been a prime target for cellular antiviral mechanisms along with influenza antiviral drug research (Turan et al., 2004 and Verhelst et al., 2012).