In situ molecular identification of the Influenza A (H1N1) 2009 Neuraminidase in patients with severe and fatal infections during a pandemic in Mexico City

Background In April 2009, public health surveillance detected an increased number of influenza-like illnesses in Mexico City’s hospitals. The etiological agent was subsequently determined to be a spread of a worldwide novel influenza A (H1N1) triple reassortant. The purpose of the present study was to demonstrate that molecular detection of pandemic influenza A (H1N1) 2009 strains is possible in archival material such as paraffin-embedded lung samples. Methods In order to detect A (H1N1) virus sequences in archived biological samples, eight paraffin-embedded lung samples from patients who died of pneumonia and respiratory failure were tested for influenza A (H1N1) Neuraminidase (NA) RNA using in situ RT-PCR. Results We detected NA transcripts in 100% of the previously diagnosed A (H1N1)-positive samples as a cytoplasmic signal. No expression was detected by in situ RT-PCR in two Influenza-like Illness A (H1N1)-negative patients using standard protocols nor in a non-related cervical cell line. In situ relative transcription levels correlated with those obtained when in vitro RT-PCR assays were performed. Partial sequences of the NA gene from A (H1N1)-positive patients were obtained by the in situ RT-PCR-sequencing method. Sequence analysis showed 98% similarity with influenza viruses reported previously in other places. Conclusions We have successfully amplified specific influenza A (H1N1) NA sequences using stored clinical material; results suggest that this strategy could be useful when clinical RNA samples are quantity limited, or when poor quality is obtained. Here, we provide a very sensitive method that specifically detects the neuraminidase viral RNA in lung samples from patients who died from pneumonia caused by Influenza A (H1N1) outbreak in Mexico City.


Background
During April 2009, the number of atypical pneumonia cases increased in Mexico City's hospitals and spread to almost all boroughs in the city; these cases were related to a new influenza A (H1N1) virus strain that was identified as the etiological agent [1][2][3][4]. In less than a month the virus spread worldwide and on June 11, 2009 the World Health Organization (WHO) declared the start of the first 21st century influenza pandemic.
Influenza A viruses belong to the Orthomyxoviridae family; they are characterized by a unique genome structure with a single-negative RNA strand, which codifies, among others, for two transmembrane proteins: hemagglutinin (HA) and neuraminidase (NA) [5][6][7]. HA plays an important role during the cell entry of influenza viruses. This protein is essential during the initial steps of infection because it is responsible for the attachment of the virus to sialic acid (SA) cellular receptor. This interaction explains, at least in part, the host range and tissue tropism of influenza viruses [5,8]. The NA of influenza viruses is a homotetrameric glycoprotein anchored by a fibrous stalk in the viral membrane. The protein possesses a globular head comprised of four monomers that constitute the active site composed of nine conserved residues. Its primary role in the infectious cycle is to liberate the viral progeny from infected cells. Its enzymatic activity catalyzes SA removal from its linkage to galactose, thereby destroying the receptor and allowing the virus to disseminate and infect other cells [8,9]. Furthermore, NA is also the main target of the antiviral drugs zanamivir and oseltamivir. These drugs closely resemble the structure of the natural substrate of the NA and thus prevent the removal of the SA residue from the glycopeptide receptor by the viral neuraminidase [5]. In addition to the increased transmissibility and the low or lack of immunity of the human population, the fact that new reassortment events may alter the pathogenicity of circulating strains, makes it crucial to monitor the progress of the pandemics at the molecular level [2,10,11].
Molecular methods are becoming more widely used for the detection of respiratory pathogens, in part because of their superior sensitivity, relatively rapid turnaround time, and ability to identify pathogens that are slow growing or difficult to culture. The recent novel H1N1 influenza A pandemic has been useful to underscore how quickly new molecular tests can become available for clinical use [12]. Previously, several groups have been used immunohistochemical or immunofluorescence detection for determine the localization of influenza virus antigens, as well as in situ hybridization for detection of viral sequences and ultrastructural examination to detect viral particles, in cases of fatal H1N1 influenza A virus infection during the period 2009-2010 [13][14][15][16][17][18][19]. These analyses were performed on sections from different tissues such as respiratory tissues (trachea, lung), heart, liver, and placenta [13][14][15][16][17][18][19].
Over the years, Reverse-transcriptase PCR is the recommended test for diagnosis and confirmation of infections due to pandemic 2009 influenza A(H1N1) virus [20]. Recently, modifications of this technology have emerged, some of which allow the rapid detection of multiple pathogens in a single test such as multiplex molecular technologies, reverse transcriptase-PCR, real-time PCR, microarrays and nucleic acid sequencing-based amplifications [12,21,22]. Other studies have also shown the usefulness of rapid immunoassays for seasonal influenza virus [23]. These methods have greatly enhanced the capability for surveillance and characterization of influenza viruses and their clinical utility for the detection of respiratory pathogens. However, these methods can not be easily applied for the analysis of paraffin-embedded tissues. In situ RT-PCR has some major strengths for the detection of specific nucleic acid sequences. First and foremost, one can detect particular sequences on archival material; second, this technique combines the extreme sensitivity of PCR with the cell localization ability similar to in situ hybridization [17]. A third strength of in situ RT-PCR relates to the issue of sample contamination in solutionphase PCR. Sample contamination, which can lead to false-positive results in PCR, limits its value as a diagnostic test for viral infections; this limitation is not encountered in in situ RT-PCR. Fourth, this technique is the only amplification technique that allows direct target-specific incorporation of a reporter nucleotide (such as DIGdigoxigenin-dUTP-labeled nucleotide) [24,25], thus eliminating the need for a hybridization step. Clearly, in situ RT-PCR has been useful for any target that is low copy and, thus, difficult to detect with standard in situ hybridization, which has a detection threshold of 10 to 20 copies per cell [17].
The purpose of the present study was to demonstrate that sensitive and specific molecular detection of pandemic influenza A (H1N1) 2009 strains is possible in archival material such as paraffin-embedded lung samples. This strategy would be useful to perform current and retrospective studies in a specific and reproducible manner.

Autopsy material
This project was approved after being checked by the "Institute Science and Bioethics Committee" (INER-Mexico). The Committee is responsible for evaluating the research projects to be performed, in order to monitor that every study meets the principles for research involving human subjects, established in the Declaration of Helsinki and its different revisions. The approval number is: B07-09.
From April 2009 to February 2010, eight lung samples were taken post mortem from patients who died of pneumonia and respiratory failure at Instituto Nacional de Enfermedades Respiratorias-Ismael Cosío Villegas (INER) and Clinica-32 of Instituto Mexicano del Seguro Social (IMSS), Mexico City. At least three autopsy samples were collected mainly from medium upper lobe of the lung ( Table 1). The size of lung autopsy samples ranged from 20 to 80 mm 2 area. Lung samples were paraffin-embedded, sectioned to obtain 5 μm thick sections and mounted on electrostatically charged slides. Molecular procedures (RT-PCR, in situ RT-PCR, sequencing and in vitro duplex amplification technique) were performed in all 8 patient lung samples, as well as in control samples. All specimens were studied by an expert pathologist (R.M.R.R.).

Molecular detection of A (H1N1) viruses
The detection of pandemic influenza A (H1N1) 2009 viruses was done using RT-PCR standard protocols as part of the Virology Department's surveillance routine and has been described elsewhere (http:// www.who.int/csr/resources/publications/swineflu/realtimeptpcr/en/index.html; [2,26]). All specimens were accompanied by a standard form with information about age, sex, date of illness onset, date of specimen collection, place of residence, clinical features of each patient, travel history, vaccination history and administration of antiviral treatment.

In situ RT-PCR
Direct in situ RT-PCR was performed as previously described with some modifications [24,25,[30][31][32] (Figure 1). Briefly, dried dewaxed sections on electrostatically charged slides were incubated with protein lysis buffer (0.1 M Tris-HCl pH 8.0, 50 mM EDTA pH 8.0) containing 0.5 μg/ml Proteinase K for 30 min at room temperature. After Proteinase K digestion, each tissue section was treated with 50 μl of a solution containing 1 U of DNase I, RNase-free (Roche, U.S.A.) during 48 h at room temperature. After thoroughly washing with DEPC-treated water, in situ reverse transcription was performed using the SuperScript II reverse transcriptase (Invitrogen, U.S.A.), following the manufacturer's specifications. In brief, 70 μl DEPC-treated water containing 3 μg of random primers oligonucleotides (mostly hexamers; Invitrogen, U.S.A.), 10 mM dNTP mix (10 mM each dATP, dGTP, dCTP and dTTP at neutral pH), 5X first-strand buffer, 10 mM DTT, recombinant ribonuclease inhibitor (40 U/μl) and reverse transcriptase (100 U/ tissue section) (Invitrogen, U.S.A.) were added to each section. Slides were incubated at 42°C for 1 h in a sealed humidified chamber. After thoroughly washing with ultrapure water, 50 μl of the PCR master mix solution containing 100mM digoxigenin-11-(2'-deoxy-uridine-5')triphosphate (DIG-11-dUTP; Roche, U.S.A.), 10X PCR buffer, 50 mM MgCl 2 and primer mix (10 μM each of forward and reverse primers) were added [31]. To reduce primer-dimer formation the PCR solution was heated to 70°C for 10 min before Taq DNA polymerase (5 U per reaction) was added. Negative controls included tissue obtained from two Influenza-like Illness (ILI) patients that were negative for the influenza A (H1N1) virus. These A (H1N1) negative patients showed: (1) severe acute respiratory illness defined as dyspnea plus bilateral infiltrates on the x-ray; (2) lack of 2009 H1N1 infection; and (3) a stay of at least 24 h in the intensive care units. Samples were taken after patients died. In addition, reactions without primers or reactions without Taq DNA polymerase were performed in consecutive tissue sections from all samples. In situ PCR was performed using the system provided by Perkin Elmer (U.S.A.). Fifty microliters of PCR master mix were added to each sample and the reaction was sealed using Ampli-Cover discs and clips (Perkin Elmer, U.S.A.). After assembly, slides were placed at 70°C in the GeneAmp In situ PCR system 1000 (Perkin Elmer, U.S.A.) until running was started (18 cycles). Amplifications for NA or β2-m were performed separately. Samples were first heated to 94°C (3 min) and then subjected to 18 cycles of: 94°C/ 1 min, 60°C/ 1.5 min and 72°C/ 1 min. After amplification a 10-min elongation step at 72°C was carried out. After cycling was complete, the temperature was kept at 4°C until disassembly. Clips were removed and AmpliCover discs were very carefully lifted from the slides without moving them sideways and slides were washed for 5 min in PBS followed by 5 min in 100% EtOH before they were air dried [30]. To ensure consistency and reproducibility and to eliminate PCR artifacts, all assays were performed on a minimum of three separate occasions. Slides were processed in order to detect in situ PCR products (see below).

Detection of in situ PCR products
An indirect immunolabeling method using a primary Anti-Digoxigenin antibody (Fab fragments; Roche, U.S.A.) conjugated to alkaline phosphatase was chosen to detect the PCR product. Briefly, blocking was carried out in 5% BSA (Sigma, U.S.A.) in PBS for 30 min. Slides were then drained and incubated with Anti-DIG antibody diluted 1:200 in 100 mM Tris-HCl pH 7.4 and 150 mM NaCl (100 μl per sample) for 2 h at room temperature. As negative control the primary antibody was omitted. Detection of alkaline phosphatase was carried out for 10-30 min using NBT/ BCIP kit (Zymed, U.S.A.). After detection, slides were rinsed in distilled water for 5 min and air dried before mounting in Permount histological mounting medium (Fisher Scientific, U.S.A.).

Digital Image Capture, analysis and quantification
All photomicrographs were obtained using a DFC290 HD digital camera (Leica Microsystems, USA). The following method of quantification was used for in situ RT-PCR analyses. After acquisition of the images using the digital camera, the experimental image files were opened in the PhotoImpact software (Ulead PhotoImpact SE version 3.02; Ulead Systems, U.S.A.). The images were digitally processed in order to obtain the better and homogeneous signal and then selected for analysis of relevant regions. The selected regions were then digitally analyzed using the Image-ProPlus Analysis Software (Ver 4.5.0.19, Media Cybernetics, Inc., U.S.A.). The amount of signal was quantified using a pixel matrix data (the color contained within a pixel inside an image at specific location). Therefore, chromogen quantity was determined by calculating the norm of the matrix file for that image. This allows pixels of similar "color" immediately adjacent to the index pixel to be included for analysis. All pixels falling within the selected threshold parameters were quantified, recorded and used to generate a graphic. The file for the control image was similarly generated: the control slide is acquired Figure 1 Experimental design used to detect the Neuraminidase-gene derived nucleic acid of A (H1N1) virus by in situ RT-PCR. Tissue fixed on electrostatically charged slides (1) was permeabilized and genomic DNA was eliminated using RNase-free DNase (2). After in situ reverse transcription (3), synthesized cDNA was amplified and detected on tissue (4). Reverse Transcription reaction product was recovered for amplification using in vitro PCR and subsequent sequencing (5) and quantification by Real Time quantitative PCR (RTqPCR) (6).
and treated identically as the experimental slide. Negative controls included: Two A (H1N1)-negative samples, reactions without reverse transcriptase and reactions without anti-DIG antibody. Beta2-microglobulin signal was used to normalize data in case of in situ RT-PCR quantification.

Sequencing of in situ RT-PCR products
After in situ reverse transcription, solution over the tissue section was recovered and the cDNA contained in this solution was amplified using standard liquidphase PCR. The nucleotide sequences of the PCR products were determined using the BigDye Terminator  (Tables 2 and 3).

In vitro Duplex amplification technique
In vitro duplex amplifications were performed using 2 μL of cDNA as template and reaction mixtures (25 μL) containing: 1x PCR buffer, 1.5 mM of MgCl 2 , 200 μM of each dNTP, 400 nM of each primer, 2.5 units of Taq DNA polymerase (Invitrogen, U.S.A.). Specific primers recognizing NA or β2-microglobulin sequences used for in situ amplification were included in the reaction (amplification size products of 729 and 100 base pairs, respectively). Amplifications were carried out using a GeneAmp PCR System 9700 thermal cycler (Applied Biosystems, U.S.A.). The duplex protocol included 3 minutes incubation at 95°C followed by 40 cycles of 1 min at 94°C, 1.5 min at 60°C and 1 min at 72°C. A final extension of 10 minutes at 72°C was performed. Finally, the duplex PCR products were separated by electrophoresis on a 2% agarose gel stained with ethidium bromide and visualized under UV light.

RNA isolation
As a control for in situ determinations, total RNA was isolated from autopsy samples obtained from patients with influenza illness using TRIzol reagent according to the manufacturer's instructions (Invitrogen). The RNA preparations were used for cDNA synthesis and Real-Time PCR.

Preparation of cDNA for Real-Time quantitative PCR (RTqPCR)
Three micrograms of total RNA were reverse transcribed in a 20 μl reaction using 100 U SuperScript II Reverse Transcriptase, following the manufacturer′s specifications (Invitrogen). In parallel, cDNAs synthesized during the in situ reverse transcription reaction were recovered and used for sequencing ( Figure 1).

Relative RNA Quantification by RTqPCR
The relative quantification of the A (H1N1) viral load was determined by RTqPCR using a 7300 Real Time PCR System (Applied Biosystems, USA). PCRs were processed through 35 cycles of a 3-step PCR, including 10 sec of denaturation at 95°C, a 10 sec primer dependent annealing phase (60°C), and a 10 sec template-dependent elongation at 72°C. The amplification of each template was performed in duplicate in one PCR run. The relative viral load was calculated as the ratio normalized to β2-microglobulin.

RTqPCR data analysis using 2 -ΔΔCT method
Real-time PCR was performed on the corresponding cDNA synthesized from each in situ sample. The data were analyzed using the equation described by Livak [32]. Validation of the method was performed as previously reported [32,33].

Expression levels of Neuraminidase A (H1N1) RNA in patients infected with Influenza
Specific primers recognizing the NA A (H1N1) influenza gene were used to establish expression levels in RNA samples. NA relative expression levels were determined in all eight samples using the β2-microglobulin housekeeping mRNA as internal control (Figure 2a). Primers and duplex amplification technique were validated analyzing a large number of nasopharyngeal swabs obtained from ILI patients (Manuscript in preparation). In addition, all samples were analyzed by quantitative RTqPCR (Figure 2b).  The expression pattern of the NA viral gene in lung samples obtained from patients with pneumonia was determined by in situ RT-PCR. We detected the expression of the NA gene in 100% of the samples (Figure 3; Table 2). As expected, no expression was detected by in situ RT-PCR in two A (H1N1)-negative patient samples nor in non-related cells (HeLa cervical cancer cell line; Figure 3). Relative in situ transcription levels correlated with those obtained when in vitro RT-PCR assays were performed. Interestingly, the in situ signal localization suggests that A (H1N1) influenza virus infects alveolarlike cells.

Discussion
In this study we describe a successful method for molecular detection of A (H1N1) influenza virus RNA in lung sections obtained from persons who died during the epidemic outbreak of influenza A (H1N1) 2009 in Mexico City. Although the current diagnostic technologies are reasonably effective for sporadic and epidemic influenza [34], new strategies including more sensitive and more specific tests have been developed enhancing not only diagnostic capability but also the need for retrospective testing [34]. The main goal of this work was the standardization of viral RNA genome detection and viral load determination in patients where sample collection such as good-quality RNA could not be performed.
In order to demonstrate that A (H1N1) Neuraminidase transcript was expressed in infected samples, we decided to use in situ RT-PCR. This technique combines the extreme sensitivity of PCR with the cell localizing ability similar to in situ hybridization and allows comparing epithelia with stroma. In situ RT-PCR results demonstrated that A (H1N1) influenza virus could be detected in the same tissue as that in which histopathological lesions had been observed (Figure 3). Furthermore, sequencing data confirmed that in situ technique was capable to specifically detect viral sequences. Although the primary goal of this study was to validate the in situ RT-PCR detection of NA influenza viral gene, we thought that these data may suggest that Mexico City could have played an important role for the dissemination of some variants throughout the world as indicated by the match between the Mexican samples included in this study and sequences described in other Countries such as New York, San Salvador, Texas, Wisconsin, Roma, Colombia District, Ontario, Auckland and Puerto Rico. In order to establish a phylogenetic analysis, the entire genome from the viruses should ideally be sequenced (Neuraminidase gene contains approximately 1,413 bp); the partial NA sequences generated by our group (729 bp) are of importance as they could provide a baseline of NA sequences from where the pandemic emerged. Additional phylogenetic analyses would be necessary to provide a better understanding of influenza NA lineages and their evolutionary dynamics, which may facilitate early detection of newly emerging influenza viruses and thus improve influenza surveillance and public health [35].
Recently, immuno-histopathological analyses of lung autopsy lung from patients with viral pneumonia revealed that the alveolar epitheliums as well as the alveolar macrophages are key target cells infected by influenza [36][37][38][39][40][41][42]. Similar conclusions have also been drawn from experiments where ex vivo lung tissue has been infected with the H5N1 virus [43][44][45]. Interestingly, morphological analyses performed in this study suggest that the in situ signal was detected in alveolar macrophage-like cells; however, the in situ detection of additional biomarkers would be necessary to provide evidence of the cellular type that is susceptible to A (H1N1) viral infection (i.e. bronchial and bronchiolar epithelial cells, pneumocytes and/or macrophages).
Molecular strategies such as in situ hybridization (ISH) have been used to identify cells infected with influenza virus in lung and other tissues [46][47][48][49]. One of the main features of the in situ RT-PCR is to take advantage of the Taq DNA polymerase to amplify a very low number of copies of a specific sequence in comparison with ISH in which the results depend on one-to-one strand recognition. To the best of our knowledge, the in situ RT-PCR strategy has been used in a limited number of cases for detecting influenza viral sequences in tissues such as myocardium and placenta [18,50]. Therefore, this is the first time that in situ detection of A H1N1 NA sequences is performed in lung sections. One novel advantage that we have developed performing in situ RT-PCR is the ability of recover and subsequently to amplify the cDNA synthesized during the in situ reverse transcription reaction, this cDNA can be used to perform nucleic acid sequence analyses as well as to detect novel biomarkers.

Conclusion
In summary, in situ retrospective detection of influenza A (H1N1) virus arising in Mexico City could provide important information to study the natural viral spread of influenza virus and, if it is possible, to correlate these data with the emergence of future new pathogenic influenza virus pandemic. Although the majority of pathology laboratories do not perform in situ RT-PCR techniques, the development of accurate tests for the detection of influenza in archival paraffin-embedded samples could be a complementary tool for the detection of the virus, enabling the laboratory to provide a prompt, definitive diagnosis, which would allow clinicians to initiate preventive methods, implement appropriate infection-control measures eventually decreasing the incidence of influenza cases, diminishing duration of hospitalization, reducing ancillary testing, and decreasing health care costs.