|Home | About | Journals | Submit | Contact Us | Français|
Outbreaks of human adenovirus (HAdV) acute respiratory illness (ARI) have been well documented among civilians and unvaccinated military recruits. Among the 7 recognized HAdV species (A to G), species B (particularly serotypes 3, 7, 11, 14, and 21) and E (serotype 4) have more often been associated with epidemic ARI. Rapid detection and type-specific identification of these viruses would enhance outbreak response and help guide prevention and control measures. To this end, we developed type-specific real-time quantitative PCR (qPCR) assays for HAdV types 3, 4, 7, 11, 14, 16, and 21 targeting the HAdV hexon gene. All type-specific qPCR assays reproducibly detected as few as 10 copies/reaction of quantified hexon recombinant plasmids with a linear dynamic range of 8 log units (101 to 108 copies); in contrast, a generic qPCR assay that detects all HAdV types run concurrently detected between 10 and 100 copies/reaction, depending on the virus type. No nonspecific amplifications were observed with concentrated nucleic acid from 51 HAdV prototype strains or other common respiratory pathogens. All members of a panel of 137 previously typed HAdV field isolates and positive clinical specimens were correctly characterized by the type-specific qPCR assays; two different HAdV types were detected in three of the clinical specimens and confirmed by amplicon sequencing. The qPCR assays permit sensitive, specific, and quantitative detection and identification of seven clinically important respiratory HAdVs and should provide a convenient adjunct to classical typing methods for a rapid response to HAdV outbreaks.
Human adenoviruses (HAdVs) frequently infect children, mainly causing diseases of the respiratory, ocular, and gastrointestinal tracts. Although most HAdV acute respiratory illnesses (ARIs) are mild and self-limited, severe and occasionally fatal infections have been reported in newborns and infants (1), the immunodeficient (2), those with underlying comorbidities (3), and occasionally, otherwise healthy adults (4–7).
HAdVs are classified within the family Adenoviridae, genus Mastadenovirus, and are further divided into seven species (A to G) and over 50 recognized types (8). Clinical symptoms, disease severity, and epidemiological patterns of infection are often determined by virus species or type. HAdV ARI is most often caused by species B, C, and E viruses. Species C viruses are ubiquitous and endemic in pediatric populations, where infections are often subclinical or lead to sporadic ARIs (9). Species B (primarily types 3, 7, 11, 14, and 21) and the sole species E (type 4) viruses have been associated with epidemics of often severe ARIs affecting children and adults. Community HAdV ARI clusters, as recently occurred with the emergence of HAdV strain 14p1 in the United States and Europe (10, 11, 12), including discrete institutional outbreaks among civilians (13), outbreaks covering wide geographic areas (6, 14, 15, 16), and continuous outbreaks among unvaccinated military personnel undergoing basic training (17, 18, 19), have been reported.
Laboratory diagnosis of HAdV infection can be achieved by virus isolation, antigen detection, PCR, and serology. Serotype identification is classically performed by serum neutralization and/or hemagglutination inhibition with type-specific hyperimmune animal antisera. Immunotyping methods provide definitive identification of serotype but are excessively time-consuming, labor-intensive, and restricted to a few laboratories that possess reference immune reagents. A variety of molecular methods have been advanced to augment immunotyping, including genome restriction analysis (20), PCR-coupled microarrays (21, 22), PCR-fragment length analysis (23), electrospray ionization mass spectrometry (24), and more commonly, partial sequencing of specific target genes (25, 26). Although an advance over immunotyping methods, these methods require specialized technical skills and utilize technologies that may be of limited availability outside the research laboratory setting.
Single and multiplex conventional and real-time quantitative PCR (qPCR) assays using species- and type-specific primers/probes have been described for some HAdVs that cause ARIs (27–30). qPCR is particularly well suited for this purpose, offering (i) a closed system with a low risk from amplicon contamination, (ii) rapid, quantitative, and high-throughput specimen testing, (iii) a wide network of technical expertise and available instrumentation, and (iv) easy implementation and integration into existing qPCR assay panels for other respiratory pathogens. For these reasons, we developed and validated sensitive and type-specific qPCR assays for detection and identification of the epidemic-associated respiratory HAdVs, types 3, 4, 7, 11, 14, 16, and 21, to facilitate rapid outbreak response.
(Data from this study were presented at the 27th Clinical Virology Symposium, Daytona, FL, 8 to 11 May 2011.)
HAdV prototype strains 1 to 33 were originally obtained from the American Type Culture Collection (Manassas, VA) or the Research Resources Branch of the National Institute of Allergy and Infectious Diseases (Bethesda, MD); prototype strains 34 to 51 were kindly provided by the originating laboratories. An additional 104 geographically and temporally diverse respiratory specimens (nasopharyngeal and oropharyngeal swabs, sputa, and lung tissue) and 33 field isolates collected during outbreak investigations and routine surveillance and previously determined to be positive for HAdV types 3 (n = 27), 4 (n = 22), 7 (n = 10), 11 (n = 7), 14 (n = 47), 16 (n = 13), and 21(n = 11) by PCR and sequencing of hypervariable regions (HVRs) 1 through 6 of the hexon gene (25) (see below) were available for testing. To evaluate assay specificity, we also tested for other respiratory pathogens, including respiratory syncytial virus, human metapneumovirus, parainfluenza viruses 1 to 4, rhinovirus 1A, coronaviruses 229E and OC43, influenza viruses A and B, cytomegalovirus, herpes simplex virus, human bocavirus, Streptococcus pneumoniae, and Haemophilus influenzae, and pooled nasal wash samples from 20 consenting healthy new military recruits predicted to contain diverse human microbiological flora.
Total nucleic acid (TNA) extracts were prepared from 100 μl of virus isolates or 200 μl of clinical specimens using a NucliSens miniMAG or easyMAG extraction system following the manufacturer's instructions (bioMérieux, Durham, NC). Extracts were retained at −70°C until use.
Hexon gene sequences of HAdV prototype and field strains available from GenBank (NCBI/NLM) were aligned, and type-specific conserved regions were identified. Multiple primer/probe sets were designed using Primer Express software, version 3.0 (Applied Biosystems, Foster City, CA), to give predicted type-specific amplification and show no major nonspecific homologies on BLAST analysis. Hydrolysis probes were labeled at the 5′ end with 6-carboxyfluorescein (FAM) and internally or at the 3′ end with black hole quencher 1 (Biosearch Technologies, Inc., Novato, CA). Optimal primer/probe concentrations for each assay were determined by checkerboard titrations, and primer/probe combinations giving the best overall performance with a limited panel of HAdVs were further evaluated. Primer/probe sets that performed best under the conditions described below and with no identifiable cross-reactions were chosen for further study (see Table S1 in the supplemental material).
qPCR assays specific for HAdV types 3, 4, 7, 11, 14, 16, and 21 were run under the conditions used for a generic HAdV qPCR (HAdV-pan) assay that we previously developed on the basis of the primer/probe sequences described by Heim et al. in 2003 (31) that detect all HAdV types. Briefly, 25-μl reaction mixtures were prepared by adding 5 μl of sample nucleic acid extract to 20 μl of iQ Supermix (Bio-Rad, Hercules, CA) containing optimal concentrations of primers/probes (see Table S1 in the supplemental material). Thermocycling was performed on a Stratagene Mx3000P qPCR system (Agilent Technologies, Santa Clara, CA) or Applied Biosystems 7500 Fast Dx real-time PCR system (Life Technologies) programmed for 3 min at 95°C to activate the iTaq DNA polymerase and 45 cycles of 15 s at 95°C and 1 min at 60°C. Threshold cycle (CT) values were determined by manually adjusting the fluorescence baseline to fall within the exponential phase of the amplification curves and above any background signal. A positive test result was considered a well-defined curve that crossed the threshold cycle within 40 cycles. Positive and negative template controls were included in all runs to qualify assay performance.
Recombinant DNA plasmids containing sequence-confirmed full hexon genes of HAdV type 3, 4, 7, 11, 14, 16, and 21 prototype strains were prepared for analytical sensitivity studies by DNA Technologies Inc. (Gaithersburg, MD). The hexon recombinant plasmids (HRPs) were quantified using a NanoDrop 1000 spectrophotometer (NanoDrop Technologies, Wilmington, DE). To generate standard curves for quantitative assessment, replicate serial 10-fold dilutions (101 to 108 copies/reaction) of the HRPs were prepared in 10 mM TE (Tris-EDTA) buffer containing 100 μg/ml herring sperm DNA (Promega, Madison, WI) and stored at −20°C until use.
HAdV hexon HVRs 1 to 6 were amplified using nested primer sets described previously (25). HVRs 1 to 6 and qPCR amplification products were sequenced bidirectionally using the internal primers and qPCR assay forward and reverse primers, respectively, and an ABI Prism BigDye Terminator cycle sequencing ready reaction kit (version 1.1) on an ABI 3130XL DNA sequencer (Applied Biosystems). Sequencher (version 4.7) software (Gene Codes, Ann Arbor, MI) was used for sequence assembly and editing.
The optimized qPCR assays were first evaluated with serial dilutions of the quantified HRPs. The type-specific and pan-qPCR assays gave linear dynamic ranges of 8 log units (101 to 108 copies) (see Fig. S1 in the supplemental material) and amplification efficiencies exceeding 95%. To estimate the limit of detection of each qPCR assay, serial dilutions of the respective HRPs containing 1,000, 100, 10, and 1 copies/reaction were each tested in 16 replicates. All type-specific qPCR assays could reproducibly detect ~10 HRP copies/reaction (Table 1). The HAdV-pan qPCR assay showed sensitivity comparable to that of the type-specific qPCR assays for HAdV types 4, 14, and 21 but was less sensitive for HAdV types 3, 7, 11, and 16, detecting between 10 and 100 HRP copies/reaction.
During the design phase, the primers/probes of each qPCR assay were evaluated in silico with HAdV hexon gene sequences published in GenBank; no significant sequence homologies with other HAdV types were observed. BLAST analyses found no significant homologies with human genome or other human microbial flora that would possibly lead to false-positive results. Each qPCR assay was then tested against concentrated genomic DNA (CT values, <20 by HAdV pan-qPCR) from HAdV prototype strains 1 to 51 to assess type specificity; no cross-reactions were detected (Table 2). The specificity of the qPCR panel was further assessed by testing laboratory cultures or positive clinical specimens known to contain human viral and bacterial pathogens that may be present in the respiratory tract, as described in Materials and Methods. No positive results were obtained with any of the samples with the exception of the human nasal wash pool, where a weak positive result (CT, 38.9) was obtained with the HAdV-pan assay; subsequent PCR and hexon sequencing (25) identified low-level HAdV type 1 (HAdV-1) in the pool, suggesting that one or more of the recruits were subclinically infected with this virus.
To assess the performance of the qPCR assays with diverse field isolates and HAdV-positive clinical specimens, DNA from 134 previously typed HAdV strains was tested by all assays (Table 3). HAdV was detected in all samples and correctly identified to type by the respective qPCR assay. However, three clinical specimens had two HAdV types identified by qPCR (HAdV-3 [CT, 33.24] and HAdV-4 [CT, 36.61]; HAdV-3 [CT, 18.89] and HAdV-14 [CT, 36.56]; HAdV-7 [CT, 20.96] and HAdV-3 [CT, 34.92]), and these were confirmed by sequencing the respective qPCR amplicons. In contrast, only the predominant type (the type with the lower CT value) was identified by our standard method of PCR and partial hexon gene sequencing (25).
In response to a community-wide outbreak of severe ARI caused by a newly emergent strain of HAdV-14 (7), we developed an HAdV-14 type-specific qPCR assay for rapid case identification. This assay was subsequently used to investigate several HAdV-14 outbreaks among civilians and military personnel (5, 10, 32). To complement this assay, in this study we developed additional type-specific qPCR assays that run under conditions identical to those for other HAdVs documented to cause ARI outbreaks, including HAdV types 3, 4, 7, 11, and 21. We also developed an assay for the rarely identified species B HAdV-16, found to be associated with cases of acute febrile respiratory illness in Egypt (33) and Kenya (unpublished data) that may pose a future risk of spread outside these regions.
In response to the need for rapid identification of HAdVs that pose a risk to unvaccinated new military recruits, Washington et al. (34) developed a multiplexed PCR assay for a similar panel of five HAdVs (HAdV types 3, 4, 7, 14, and 21) based on Luminex xMAP technology (Luminex, Austin, TX) and Metzgar et al. (28) developed fluorescence resonance energy transfer (FRET) qPCR assays for the same viruses using the Joint Biological Agent Identification and Diagnostic system (BioFire Diagnostics, Salt Lake City, UT) and LightCycler (version 2.0; Roche, Indianapolis, IN) platforms. Though reported to be sensitive and specific, the FRET qPCR assays were predicted to be 5 to 50 times less sensitive than our corresponding qPCR assays. Moreover, both assay panels were developed and validated on platforms that are not commonly used outside U.S. Department of Defense-affiliated laboratories.
As shown by others (33, 34, 35), species- and type-specific PCR assays are capable of revealing coinfections with different HAdVs that otherwise might go undetected. In this study, we identified several specimens by our qPCR assays that contained two HAdV types, one predominant type and one subordinate type. In contrast, our routine typing method of generic PCR amplification and sequencing of HVRs 1 to 6 (25) identified only the predominant HAdV type in each case, demonstrating that virus present at a higher concentration can mask virus present at a lower concentration when using direct sequencing methods for typing. Although the clinical value of detecting mixed infections is unclear, our type-specific qPCR assays will prove useful for outbreak investigations and may facilitate future studies to better define the full complexity of HAdV infection. The quantitative capability of these assays will help better define the role of viral load in determining clinical outcomes and monitoring the efficacy of antiviral therapies.
Despite these advantages, qPCR assays have several potential limitations. Our assay panel is limited to those HAdV types most often associated with epidemic ARIs; other types more rarely associated with ARI outbreaks would not be identified (36). Although all HAdV strains evaluated were successfully identified, their full genetic diversity may not be represented in our study. Unrecognized genetic heterogeneity in the primer/probe region could compromise assay performance. Moreover, because the qPCR assays target sequences in the HVRs that encode epitopes that define HAdV serotype (37), intertypic recombinant strains occurring between the hexon and fiber genes would not be distinguished. For example, our qPCR assay amplifies both HAdV genome types 11p (strain Slobitski) and 11a (also designated HAdV-11/H14  and HAdV-55 ) with equal efficiency (data not shown). The former is rarely isolated and more often associated with acute hemorrhagic cystitis and immunocompromised host disease (40), whereas the latter, which emerged as a respiratory pathogen causing outbreaks of ARI (16, 41, 42, 43), was found to possess the HAdV-14 fiber gene, likely acquired through recombination (12). In these cases, partial hexon and fiber genes or full-genome sequencing would be recommended to more fully characterize these viruses (26).
In conclusion, our panel of sensitive and specific qPCR assays will permit rapid detection and type-specific identification of several important HAdVs associated with epidemic ARIs. These assays will provide a valuable epidemiologic tool for investigation of future HAdV outbreaks.
The contents of this report are solely the responsibility of the authors and do not necessarily represent the official views of the U.S. Centers for Disease Control and Prevention (CDC) or the U.S. Department of Health and Human Services (DHHS). Names of specific vendors, manufacturers, or products are included for public health and informational purposes; inclusion does not imply endorsement of the vendors, manufacturers, or products by the CDC or DHHS.
Published ahead of print 16 January 2013
Supplemental material for this article may be found at http://dx.doi.org/10.1128/JCM.03297-12.