Skip Navigation

Cord Blood Gene Expression in Infants Hospitalized with Respiratory Syncytial Virus Bronchiolitis

  1. Hans-Olav Fjaerli1,
  2. Geir Bukholm2,
  3. Camilla Skjaeret2,
  4. Marit Holden3 and
  5. Britt Nakstad1
  1. 1 Faculty Division, Department of Pediatrics, Oslo, Norway
  2. 2 Institute of Clinical Epidemiology and Molecular Biology, Akershus University Hospital, University of Oslo, Oslo, Norway
  3. 3 Norwegian Computing Center, Oslo, Norway
  1. Reprints or correspondence: Hans-Olav Fjaerli, University of Oslo, Faculty Div., Akershus University Hospital, Dept. of Pediatrics, Akershus University Hospital, Oslo, Norway (h.o.fjarli{at}medisin.uio.no).
 
Next Section

Abstract

Background. Only a few infants develop acute bronchiolitis when exposed to respiratory syncytial virus (RSV), and host, environmental, and viral properties are probably all of importance in determining the severity of infection.

Methods. Microarray analysis was used to identify differentially expressed single genes and gene sets in cord blood from 5 infants hospitalized with RSV bronchiolitis versus cord blood from 5 control infants exposed to RSV without bronchiolitis during infancy. Quantitative real-time polymerase chain reaction (QRT-PCR) was performed on single genes in both the 5 infants selected for microarray analysis and 13 more infants hospitalized with the same disease. Gene set enrichment analysis (GSEA) was performed to identify differentially expressed gene sets within the microarray experiments.

Results. Microarray analysis identified 15 single genes to be significantly differentially expressed between case and control infants. Eleven of these genes were evaluated with QRT-PCR, and the genes FAM102A, TNFRSF25, and STMN3 were down-regulated in all but 1 of the 18 infants. A pathway involved in regulation of the actin cytoskeleton was found to be clearly down-regulated when analyzed with GSEA.

Conclusions. FAM102A, TNFRSF25, and STMN3 and a pathway involved in regulation of the actin cytoskeleton are down-regulated in cord blood from infants hospitalized with RSV bronchiolitis.

Respiratory syncytial virus (RSV) is an important cause of childhood respiratory tract infection worldwide, and most children are infected by age 2 years. The clinical course varies from mild upper respiratory tract symptoms (“common cold”) to bronchiolitis or pneumonia. Reinfections are common, but the first infection is usually the most severe and leads to hospitalization in 2%- 3% of cases, with the majority occurring in otherwise healthy infants [1, 2]. Why some infants develop bronchiolitis when exposed to RSV is poorly understood, and several aspects of environmental and host immunity have been extensively studied [3]. Genetic studies have shown an association between variations in the interleukin-8 and -10 genes and susceptibility to severe RSV bronchiolitis [4, 5]. Other studies have proposed a possible protective role of antenatal RSV sensitization to bronchiolitis during early life [6].

The aim of the present study was to identify genes expressed in cord blood from infants hospitalized with RSV bronchiolitis during infancy and to discuss the impact of these genes on severity of infection.

Microarray analysis can detect the simultaneous expression and interactions of thousands of genes [7]. This technology was used as a hypothesis-generating tool to identify the most differentially expressed genes and gene sets in cord blood samples from infants hospitalized with RSV bronchiolitis. To evaluate our microarray gene-expression results, quantitative real-time polymerase chain reaction (QRT-PCR) was performed [8]. To do this, cord blood samples from 13 infants hospitalized with the same disease, as well as cord blood samples from 5 infants selected for microarray analysis, were analyzed.

Previous SectionNext Section

Materials and Methods

Patients and Control Infants

From a total of 3500 deliveries in Akershus University Hospital during the period 20 January 2002 to 15 Feb

ruary 2003, a basic birth cohort of 2108 healthy infants was established. For each delivery, a venous blood sample for isolation of total RNA was collected from the umbilical cord, and informed consent for further studies was obtained from the parents. If a child was hospitalized with acute bronchiolitis during infancy, another venous blood sample for RNA isolation and a sample of mucous from nasopharyngeal aspirate (NPA) for microbiological diagnosis were collected as part of the admission routine. When the follow-up period ended on 15 February 2004, a total of 57 infants from the basic birth cohort had been hospitalized with acute bronchiolitis. Of these, 27 had RSV subtype B bronchiolitis and were selected for further evaluation. Eighteen of the 27 cord blood samples had acceptable RNA quality and were selected for final analyses. Cord blood samples from 5 boys in this group were randomly selected for microarray analysis, whereas the remaining 13 infants took part in the QRT-PCR studies (tables 1 and 2).

Table 1.
View larger version:
Table 1.

Medical history of 18 infants hospitalized with respiratory syncytial virus bronchiolitis and of five 1-year-old control infants without prior hospitalization or treatment for bronchiolitis.

Table 2.
View larger version:
Table 2.

Clinical observations, routine laboratory evaluations, and treatment in 18 infants hospitalized with respiratory syncytial virus bronchiolitis.

By random telephone interview of parents, we invited a limited number of children from the same basic birth cohort to a follow-up visit at age 1 year. These children had not been hospitalized andtreated for acute bronchiolitis during infancy. Venous blood, including a sample for isolation of total RNA and a sample for serological testing, was collected, and the parents were once again asked whether the child had been hospitalized andtreated for acute bronchiolitis during infancy. All clinical examinations and interviews with the parents of these children were performed by the principal investigator, and answers given by the parents were checked against hospital discharge records. Five boys with cord blood of acceptable RNA quality were used as matched control infants in the cord blood microarray experiments, and a pooled cord blood RNA sample from the same 5 boys was used as an exogenous control in the QRT-PCR studies (tables 1 and 2). All 5 boys tested positive at age 1 year for previous RSV exposure when their samples were analyzed for RSV antibodies by use of a standard complement-fixation test [9].

Clinical and Virological Diagnosis

The infants received a diagnosis of bronchiolitis if they had symptoms in the lower airways characterized by wheezing, dyspnea, respiratory distress, poor feeding, tachypnea, and fine crepitations on auscultation when examined by the doctor on call in the emergency room [10]. The doctor conducted a structured clinical interview with the parents, and NPAs were obtained for viral examination by multiplex RT-PCR [11].

Microarray Experiments

RNA extraction, amplification, labeling, and cRNA synthesis. Cord blood samples were collected in PaxGene RNA collection tubes (PreAnalytiX) in accordance with the manufacturer's recommendations and were stored at 80C until RNA extraction [12]. Total RNA was extracted from the PaxGene RNA collection tubes using the PaxGene blood RNA isolation kit (Pre-AnalytiX). RNA quality was assessed using the RNA integrity number of each sample as determined using the Agilent 2100 Bioanalyzer [13]. Fluorescent cRNA was generated by RNA amplification using the Agilent Low Input linear amplification kit (Agilent Technologies) in accordance with the manufacturer's instructions, and each sample was labeled with either Cy3 or Cy5 (PerkinElmer). The final cRNA concentration was determined using the ND-1000 spectrophotometer (NanoDrop Technologies).

Hybridization, scanning, and storage of raw data. One microgram of Cy3-labeled cRNA from each of the 5 infants was mixed with the same amount of reverse-color, Cy5-labeled cRNA from the corresponding control infant. All experiments were also dye-swapped with reverse coloring. The mixed, labeled cRNA was hybridized onto Agilent 60-mer human 1A (V2) oligomicroarrays (G4110B). The hybridization of the cRNA-labeled targets to the glass slide was done using the Agilent 60-mer oligomicroarray processing protocol (version 4.1; Agilent Technologies), and the slides were scanned using the Agilent Microarray Scanner (version G2505B; Agilent Technologies). The raw data were loaded into the public domain MIAME (minimum information about a microarray experiment)-compliant database BASE (version 1.2.10) [14].

Experimental design and statistical analysis. A dye-swap case-control experimental design was chosen in which each of the 5 dye swaps consisted of cord blood samples from 1 patient and 1 control infant. Each patient and each control infant were included in exactly 1 dye swap. Feature extraction was done using Agilent G2567AA Feature Extraction software (version v.7.5; Agilent Technologies) with linear and Lowess normalization using default parameter settings. All control spots and all spots with intensities <300 in either channel were flagged.

Testing the genes one by one. For finding the differentially expressed genes, statistical analyses using Bayesian analysis of variance for microarrays (BAMarray; http:\www.bamarray .com) were done on the normalized data [15]. To use the BAMarray software, an expression matrix was made. To get rid of the dye effect, the mean of the 2 values obtained from each of the 2 experiments in the corresponding dye swap was computed for each case-control pair and gene. If 1 of the 2 values was missing, the mean value was set as equal to the nonmissing value in the pair. If both values were missing, the mean value was also missing. After the removal of genes with data from only 1 or 0 case-control pairs, missing values were imputed using the K-nearest neighbor imputation method in Significance Analysis of Microarrays statistical software (http:\www -stat.stanford.edu\'98tibs). The differentially expressed genes were then determined from the resulting expression matrix using BAMarray software.

Testing gene sets. The approach described above tests for each gene whether it is differentially expressed or not. A difficulty with this approach is that the power of the tests often is low, which implies that too few of the truly differentially expressed genes are identified. A solution to this problem is to test sets of genes instead of checking each gene separately. The gene sets might be, for example, the genes in an already known pathway, or it might be a set of coexpressing genes found in previous studies. The method, called “gene set enrichment analysis” (GSEA), is one method for checking whether an a priori- defined gene set S is differentially expressed between 2 groups (e.g., 2 different phenotypes) [16]. GSEA is performed by ranking the genes occurring in the expression matrix on the basis of their correlation with the phenotype. The goal is then to determine whether members of the gene set S tend to occur toward the top or bottom of the ranked list of genes. If this is the case, the gene set S is correlated with the phenotypic group distinction, and we conclude that the gene set S is up-regulated or down-regulated in group 1, compared with group 2. To decide whether S is correlated with the phenotypic class distinction, a statistic called an “enrichment score” (ES) is calculated, and the leading-edge subset can be interpreted as the core of a gene set that accounts for the enrichment signal. Examination of the leading-edge subset can reveal a biologically important subset within a gene set. The absolute value of the ES becomes high if the gene set S tends to occur toward the top or bottom of the ranked list of genes and becomes small otherwise. Whether the observed ES is large enough to conclude that S is differentially expressed between 2 groups is decided using a permutation-based test. Because we might want to test several different gene sets, we need to account for multiple testing. In the GSEA software, this is done by controlling the false discovery rate (FDR). The GSEA program was run, mainly with default parameter settings, using the same expression matrix as that used for the BAMarray analysis. The GSEA package includes an initial catalog of human gene sets. From this catalog, we chose the functional sets and included gene sets with at least 15 and at most 500 genes also occurring in the expression matrix. These functional sets contain genes whose products are involved in specific metabolic and signaling pathways, as well as genes coregulated in response to genetic and chemical perturbations. Because the GSEA software assumes that the expression matrix consists of some columns with measurements for group 1 and other columns with measurements for group 2, we had to modify the software such that the expression matrix could consist of columns where the measurements are log2 ratios of the measurements from group 1 and group 2.

Classification of genes. The gene list with individual Agilent probe identification numbers (IDs) was curated with tools from the gene database Genetools (http:\www.genetools.no), developed by the Norwegian Microarray Consortium (http:\www.mikromatrise.no), and annotated with gene ID, symbol, and name according to the National Center for Biotechnology Information database Entrez Gene (http:\www.ncbi.nlm.nih .gov.fcgi?dbpgene). Further classification of gene involvement in biological processes was done using the Gene Ontology Annotation database, developed by the European Bioinformatics Institute (http:\www.ebi.ac.uk).

QRT-PCR Studies

Patients and control infants. To evaluate the gene-expression results obtained by microarray analysis, QRT-PCR was performed on genes in cord blood samples from the 13 other hospitalized infants as well as on genes in cord blood samples from the 5 infants selected for microarray analysis. A pooled sample of cord blood RNA from the 5 male control infants in the microarray experiments was used as exogenous control in the QRT-PCR studies.

cDNA synthesis. Extracted total RNA was quantified with the ND-1000 spectrophotometer (NanoDrop Technologies) and qualified with the Bioanalyzer platform (Agilent Technologies).

Two micrograms of total RNA was cDNA synthesized by use of the High Capacity cDNA Archive kit (Applied Biosystems). The protocol was performed in accordance with the manufacturer’s instructions for the manual method, and the GeneAmp PCR System 9700 thermalcycler was used for performing the reverse transcription.

QRT-PCR with TaqMan low-density array (TLDA) cards. Quantitative real-time PCR with predesigned gene-expression assays was used in the study. TLDA (Applied Biosystems) is based on a microfluidic technology and allows gene-expression analysis for a number of selected genes [17]. The 7900HT real-time PCR software (SDS 2.2.2; Applied Biosystems) supports real-time relative quantification of nucleic acids using the comparative cycle threshold (C T method and relative quantification (RQ). The arithmetic formula for RQ is 2−ΔΔCT , where ΔΔC T is the normalized signal level in a sample relative to the normalized signal level in the corresponding calibrator sample [18].

TLDA technology allows the simultaneous measurement of expression of up to 384 genes in a single sample. Each array has 8 channels. Each channel services 48 wells. In this study, we loaded 4 samples per card. All cards were assayed for the different candidate genes in triplicate and for 3 housekeeping genes (the 18S rRNA gene, GUSB, and RPLOP ) per channel.

cDNA was diluted, and an appropriate amount was mixed with nuclease-free water (Ambion) and TaqMan 2× Universal Mastermix (Applied Biosystems) in accordance with protocols. The cDNA concentration corresponds to 700 ng of RNA per channel. The cards were then centrifuged at 331 g for 2 × 1 min, sealed, and loaded onto the 7900HT real-time PCR system (Applied Biosystems) in accordance with the manufacturer’s recommendations.

Statistical Analysis

QRT-PCR data were quantified using the SDS software package (version 2.2.2; Applied Biosystems), and results from each card were quantified using the RQ method. RQ results were analyzed by manual removal of poor amplifications, automatic determination of baseline and threshold, and automatic outlier removal. The RQ minimum and maximum confidence level was set at 95%.

Previous SectionNext Section

Results

Microarray Experiments

Feature Extraction software identified 21,073 unique gene sequences. After normalization and filtering of control spots and spots with signal intensities <300 in either channel, 15,239 genes were available for further evaluation.

Testing the genes one by one. When we performed statistical analysis of the data using the BAMarray software, we found 15 genes to be significantly differentially expressed between patients and control infants. Five of these genes (GZMA, ZAP70, TNFRSF25, IL7R, and TRA@) were down-regulated in all 5 microarray experiments and are associated with immune responses. Most of the remaining 10 genes were also down-regulated in the microarray experiments but are associated with several different other biological processes, according to the Entrez Gene database (figure 1 and table 3).

Table 3.
View larger version:
Table 3.

Genes highly differentially expressed by microarray analysis in cord blood from 5 infants hospitalized with respiratory syncytial virus bronchiolitis vs. control infants.

Testing gene sets. By using the GSEA method, we identified 30 gene sets to be down-regulated and 12 gene sets to be up-regulated, all with FDR q values <30%. Among these pathways, 4 had an FDR q value <10%; the pathway involved in regulation of the actin cytoskeleton was in a class of its own, with an FDR q value of 0.7% (table 4). In the microarray assays, 25 of 35 genes in this gene set were identified, of which 11 genes were identified within the leading-edge subset. Ten of these 11 genes are found in this leading-edge pathway (figure 2 and table 5), as seen in the Kyoto Encyclopedia of Genes and Genomes database (http:\www.genome.jp).

Table 4.
View larger version:
Table 4.

Gene sets down-regulated in cord blood of 5 infants hospitalized with respiratory syncytial virus (RSV) bronchiolitis vs. control infants, using Gene Set Enrichment Analysis (GSEA) software.

Table 5.
View larger version:
Table 5.

Genes involved in regulation of the actin cytoskeleton and down-regulated in cord blood.

Table 6.
View larger version:
Table 6.

Genes highly differentially expressed by microarray and analyzed with quantitative real-time polymerase chain reaction in cord blood of 18 infants hospitalized with respiratory syncytial virus bronchiolitis vs. control infants.

Figure 1.
View larger version:
Figure 1.

Genes highly differentially expressed by microarray analysis in cord blood of 5 infants hospitalized with respiratory syncytial virus bronchiolitis vs. control infants.

Figure 2.
View larger version:
Figure 2.

Genes involved in regulation of the actin cytoskeleton as given by the Kyoto Encyclopedia of Genes and Genomes database (http:\ www.genome.jp). Genes marked in red are down-regulated by microarray analysis in cord blood from 5 infants with respiratory syncytial virus (RSV) bronchiolitis vs. cord blood of 5 control infants exposed to RSV during infancy without bronchiolitis when analyzed with the Gene Set Enrichment Analysis tool (http:\www.broad.mit.edu). MAPK, MAP kinase.

QRT-PCR studies. Of the 15 genes differentially expressed in the cord blood microarray experiments, 11 were further evaluated by QRT-PCR using TLDA cards both in the 5 infants selected for microarray analysis and in 13 other infants with RSV infection. The 3 genes RPL11, RPL27A, and RPL32 were excluded from analysis with TLDA cards, and the gene SLC22A8 failed amplification. We found the genes FAM102A, TNFRSF25, and STMN3 to be down-regulated in both the microarray experiments and the QRT-PCR study, except for the results from one particular infant who showed results discordant from those of the other 17 infants for all 3 genes (figure 3 and table 6).

Figure 3.
View larger version:
Figure 3.

Quantitative real-time polymerase chain reaction analysis of genes highly differentially expressed by microarray analysis in cord blood of 18 infants hospitalized with respiratory syncytial virus bronchiolitis vs. control infants.

Previous SectionNext Section

Discussion

The present study used microarray analysis as a hypothesis-generating tool to identify the most differentially expressed genes and gene sets in cord blood from children hospitalized with RSV bronchiolitis during infancy. To further evaluate these genes, QRT-PCR analysis was performed. We found the genes FAM102A, TNFRSF25, and STMN3 to be the most differentially expressed single genes in our sample. When we analyzed the microarray gene-expression assays for gene sets with GSEA, we found a pathway regulating the actin cytoskeleton to be highly correlated with a predisposition toward subsequent RSV bronchiolitis.

The infants in the present study were all born in our hospital after uncomplicated deliveries, and all cord blood samples were collected immediately after birth by trained nursing personnel. All 18 hospitalized infants had acute bronchiolitis of moderate severity during infancy, meaning that they were hospitalized because of severe respiratory difficulties but were not in need of assisted ventilation, including treatment with continuous positive airway pressure or mechanical ventilation. Eight of the 18 hospitalized infants had a family history of atopic disease, which might influence the interpretation of the gene-expression results (tables 1 and 2). When we performed statistical analysis to compare these 2 subgroups, we found no significant differences for the genes analyzed. This was also true for the possible impact of mode of delivery on gene expression (data not shown).

The children in the control group had RSV infection during infancy with only mild symptoms and were chosen to exclude children from the control group who could have had a possible genomic predisposition for a more abnormal immune response characterizing the infants with overt bronchiolitis. We cannot give full evidence of prior RSV infection in the control infants. However, the complement fixation test has acceptable specificity and was the only routine analysis available for serological testing of respiratory viruses in Norway at the time of the study.

We found the 3 genes FAM102A, TNFRSF25, and STMN3 and a pathway involved in regulation of the actin cytoskeleton to be down-regulated in cord blood from infants with subsequent RSV bronchiolitis. However, to confirm that a down-regulation of these genes is of importance for susceptibility to RSV bronchiolitis in early life, a much larger study is needed. Furthermore, the specificity or lack of specificity of these cord blood gene expression to RSV must also be analyzed by comparison with that of other viruses causing bronchiolitis in early life [19, 20].

TNFRSF25, or death domain receptor 3 (DR3), is a tumor necrosis factor superfamily receptor whose expression can be induced on T cells by T cell receptor activation. Engagement of T cell DR3 does not, however, lead to apoptosis but rather to activation of NF-κB and to potentiation of interferon (IFN)-γ production [21]. Several studies have shown an association between an immunological cord blood imbalance that includes a delayed maturation of IFN-γ production and increased susceptibility to RSV bronchiolitis in early life. Lack of cord blood TNFRSF25 activation in the ill infants in our study further supports these observations [22-25].

A significant down-regulation of the gene set involved in the regulation of the actin cytoskeleton was also observed in our study. Other studies have shown that the actin cytoskeleton is important for T cell formation and signaling [26, 27]. In relation to the above findings of TNFRSF25 down-regulation, it is therefore tempting to propose that an immaturity in the formation of the T cell receptor may explain, at least in part, why TNFRSF25 is down-regulated and why some infants develop RSV bronchiolitis during early life. This is also interesting with respect to previous studies in our group indicating a delayed maturation of T cell subsets in newborns at high risk for allergy, compared with those at low risk [28, 29].

The 2 genes FAM102A and STMN3 were also down-regulated in our study. However, the possible role(s) of these genes has still to be investigated in relation to RSV infection. A significant down-regulation of the gene GZMA was also observed. This is another interesting observation, given its known association with cytotoxic T lymphocyte (CTL) function. CTLs kill their cellular targets by mechanisms that require direct contact between the effector and target cells. In one of these pathways, perforin and granzymes are secreted by exocytosis and together induce apoptosis of the target cell. The granule exocytosis pathway of cytotoxic lymphocytes is therefore crucial for immune surveillance and homeostasis [30]. In a mouse study by Rutigliano and Graham [31], a protective immune response to RSV generated by perforinmediated cytolysis and IFN production was postulated. Because of this and other reports, it is tempting to hypothesize that the down-regulation of granzymes in cord blood from infants with subsequent RSV bronchiolitis is associated with more-severe disease.

Microarray analysis can be used in cord blood studies as a hypothesis-generating tool to identify genes and gene sets of importance for susceptibility to and severity of RSV infection during early life; to our knowledge, this method has not been published before. The present study identified the genes FAM102A, TNFRSF25, and STMN3 and a pathway involved in regulation of the actin cytoskeleton to be down-regulated in cord blood from infants hospitalized with RSV bronchiolitis. However, more studies are needed to identify the specificity of these genes in relation to this disease.

Previous SectionNext Section

Footnotes

  • Potential conflicts of interest: none reported.

  • Financial support: Akershus University Hospital (internal research funds); Research Council of Norway (grant 154079/420, Statistical Analysis of Risk).

  • Received November 20, 2006.
  • Accepted February 21, 2007.
Previous Section
 

References

  1. ↵
    1. Law BJ,
    2. Carbonell-Estrany X,
    3. Simoes EA
    . An update on respiratory syncytial virus epidemiology: a developed country perspective. Respir Med 2002;96(Suppl B):S1-7.
  2. ↵
    1. Black CP
    . Systematic review of the biology and medical management of respiratory syncytial virus infection. Respir Care 2003;48:209-31.
    Medline
  3. ↵
    1. Ogra PL
    . Respiratory syncytial virus: the virus, the disease and the immune response. Paediatr Respir Rev 2004;5(Suppl A):S119-26.
  4. ↵
    1. Hacking D,
    2. Knight JC,
    3. Rockett K,
    4. et al
    . Increased in vivo transcription of an IL-8 haplotype associated with respiratory syncytial virus disease-susceptibility. Genes Immun 2004;5:274-82.
    CrossRefMedlineWeb of Science
  5. ↵
    1. Wilson J,
    2. Rowlands K,
    3. Rockett K,
    4. et al
    . Genetic variation at the IL10 gene locus is associated with severity of respiratory syncytial virus bronchiolitis. J Infect Dis 2005;191:1705-9.
    Abstract/FREE Full Text
  6. ↵
    1. Legg JP,
    2. Jones CA,
    3. Warner JA,
    4. Johnston SL,
    5. Warner JO
    . A hypothesis: antenatal sensitisation to respiratory syncytial virus in viral bronchiolitis. Arch Dis Child 2002;8:431-3.
  7. ↵
    1. Lockhart DJ,
    2. Winzeler EA
    . Genomics, gene expression and DNA arrays. Nature 2000;405:827-36.
    CrossRefMedline
  8. ↵
    1. Kubista M,
    2. Andrade JM,
    3. Bengtsson M,
    4. et al
    . The real-time polymerase chain reaction. Mol Aspects Med 2006;27:95-125.
    CrossRefMedline
  9. ↵
    1. Ebekian B,
    2. Carballal G,
    3. erqueiro C,
    4. Avila M,
    5. Salomon H,
    6. Weissenbacher M
    . [Serologic response measured by complement fixation in children with acute respiratory infection]. Rev Argent Microbiol 1990;22:62-7.
    Medline
  10. ↵
    1. Papadopoulos NG,
    2. Gourgiotis D,
    3. Javadyan A,
    4. et al
    . Does respiratory syncytial virus subtype influences the severity of acute bronchiolitis in hospitalized infants? Respir Med 2004;98:879-82.
  11. ↵
    1. Kehl SC,
    2. Henrickson KJ,
    3. Hua W,
    4. Fan J
    . Evaluation of the Hexaplex assay for detection of respiratory viruses in children. J Clin Microbiol 2001;39:1696-701.
    Abstract/FREE Full Text
  12. ↵
    1. Rainen L,
    2. Oelmueller U,
    3. Jurgensen S,
    4. et al
    . Stabilization of mRNA expression in whole blood samples. Clin Chem 2002;48:1883-90.
    Abstract/FREE Full Text
  13. ↵
    1. Schroeder A,
    2. Mueller O,
    3. tocker S,
    4. et al
    . The RIN: an RNA integrity number for assigning integrity values to RNA measurements. BMC Mol Biol 2006;7:3.
    CrossRefMedline
  14. ↵
    1. Brazma A,
    2. Hingamp P,
    3. Quackenbush J,
    4. et al
    . Minimum information about a microarray experiment (MIAME)-toward standards for microarray data. Nat Genet 2001;29:365-71.
    CrossRefMedlineWeb of Science
  15. ↵
    1. Ishwaran H,
    2. Rao JS
    . Detecting differentially expressed genes in microarrays using Bayesian model selection. J Am Stat Assoc 2003;98:438-55.
    CrossRefWeb of Science
  16. ↵
    1. Subramanian A,
    2. Tamayo P,
    3. Mootha VK,
    4. et al
    . Gene set enrichment analysis: A knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci USA 2005;102:15545-50.
    Abstract/FREE Full Text
  17. ↵
    1. Abruzzo LV,
    2. Lee KY,
    3. Fuller A,
    4. et al
    . Validation of oligonucleotide microarray data using microfluidic low-density arrays: a new statistical method to normalize real-time RT-PCR data. Biotechniques 2005;38:785-92.
    CrossRefMedlineWeb of Science
  18. ↵
    1. Livak KJ,
    2. Schmittgen TD
    . Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta CT) method. Methods 2001;25:402-8.
    CrossRefMedlineWeb of Science
  19. ↵
    1. Weissbrich B,
    2. Neske F,
    3. Schubert J,
    4. et al
    . Frequent detection of bocavirus DNA in German children with respiratory tract infections. BMC Infect Dis 2006;6:109.
    CrossRefMedline
  20. ↵
    1. Wolf DG,
    2. Greenberg D,
    3. Kalkstein D,
    4. et al
    . Comparison of human metapneumovirus, respiratory syncytial virus and influenza A virus lower respiratory tract infections in hospitalized young children. Pediatr Infect Dis J 2006;25:320-4.
    CrossRefMedlineWeb of Science
  21. ↵
    1. Papadakis KA,
    2. Zhu D,
    3. Prehn JL,
    4. et al
    . Dominant role for TL1Apathway in IL-12 plus IL-18-induced IFN-gamma production by peripheral blood and mucosal CCR9+ T lymphocytes. J Immunol 2005;174:4985-90.
    Abstract/FREE Full Text
  22. ↵
    1. Blanco-Quiros A,
    2. Gonzalez H,
    3. Arranz E,
    4. Lapena S
    . Decreased interleukin-12 levels in umbilical cord blood in children who developed acute bronchiolitis. Pediatr Pulmonol 1999;28:175-80.
    CrossRefMedlineWeb of Science
    1. Krishnan S,
    2. Craven M,
    3. Welliver RC,
    4. Ahmad N,
    5. Halonen M
    . Differences in participation of innate and adaptive immunity to respiratory syncytial virus in adults and neonates. J Infect Dis 2003;188:433-9.
    Abstract/FREE Full Text
    1. Bont L,
    2. Kimpen JL
    . Immunological mechanisms of severe respiratory syncytial virus bronchiolitis. Intensive Care Med 2002;28:616-21.
    CrossRefMedlineWeb of Science
  25. ↵
    1. Legg JP,
    2. Hussain IR,
    3. Warner JA,
    4. Johnston SL,
    5. Warner JO
    . Type 1 and type 2 cytokine imbalance in acute respiratory syncytial virus bronchiolitis. Am J Respir Crit Care Med 2003;168:633-9.
    Abstract/FREE Full Text
  26. ↵
    1. Fuller CL,
    2. Braciale VL,
    3. Samelson LE
    . All roads lead to actin: the intimate relationship between TCR signaling and the cytoskeleton. Immunol Rev 2003;191:220-36.
    CrossRefMedlineWeb of Science
  27. ↵
    1. Dustin ML,
    2. Cooper JA
    . The immunological synapse and the actin cytoskeleton: molecular hardware for T cell signaling. Nat Immunol 2000;1:23-9.
    CrossRefMedlineWeb of Science
  28. ↵
    1. Haddeland U,
    2. Karstensen AB,
    3. Farkas L,
    4. et al
    . Putative regulatory T cells are impaired in cord blood from neonates with hereditary allergy risk. Pediatr Allergy Immunol 2005;16:104-12.
    CrossRefMedlineWeb of Science
  29. ↵
    1. Haddeland U,
    2. Sletten GB,
    3. Brandtzaeg P,
    4. Nakstad B
    . Impaired interleukin (IL)-4-associated generation of CCR4-expressing T cells in neonates with hereditary allergy risk. Clin Exp Immunol 2005;139:314-22.
    CrossRefMedlineWeb of Science
  30. ↵
    1. Voskoboinik I,
    2. Smyth MJ,
    3. Trapani JA
    . Perforin-mediated target-cell death and immune homeostasis. Nat Rev Immunol 2006;6:940-52.
    CrossRefMedlineWeb of Science
  31. ↵
    1. Rutigliano JA,
    2. Graham BS
    . Prolonged production of TNF-alpha exacerbates illness during respiratory syncytial virus infection. J Immunol 2004;173:3408-17.
    Abstract/FREE Full Text
| Table of Contents

This Article

  1. J Infect Dis. (2007) 196 (3): 394-404. doi: 10.1086/519168
  1. AbstractFree
  2. » Full Text (HTML)Free
  3. Full Text (PDF)Free

Classifications

Share

  1. Email this article

Navigate This Article

Current Issue

  1. March 1, 2012 205 (5)
  1. Journal of Infectious Diseases
  1. Alert me to new issues

Most