Development and Validation of PCR Assays for Improved Diagnosis of Infectious Coryza by Differentiating Pathogenic and Nonpathogenic Avibacterium paragallinarum Open Access

Despite its identification nearly a century ago, infectious coryza (IC) continues to pose a substantial challenge to the poultry industry both in the United States and globally (1). IC, caused by Avibacterium paragallinarum (AP), presents as an upper respiratory tract disease in chickens (2). The significant economic losses resulting from IC are not only due to marked reduction in egg production (10%–40%), mortality, and cost of vaccination but also because surviving flocks remain AP carriers for life (3). These carrier birds act as reservoirs, which complicate any attempts to control and eradicate the disease, particularly in multiage layer complexes (1,3,4). Multiple outbreaks of IC were reported in multiage layer complexes in the US (5,6), including the recent occurrences in Ohio, Indiana, and Iowa (7), placing IC in a status of an emerging disease within the U.S. poultry industry.

Despite its distinctive clinical presentation in naïve chicken flocks, characterized by severe infraorbital sinus swelling, a sharp decline in water and feed consumption, and subsequent egg production reduction, the definitive diagnosis of IC still requires laboratory confirmation by either bacterial isolation or real-time quantitative PCR (qPCR). However, the fastidious nature of AP and bacterial flora within the respiratory tract complicate the isolation process (4,8,9,10). Additionally, the isolation of AP requires the addition of nicotinamide adenine dinucleotide (NAD) as a growth factor or the use of nurse bacteria to support their growth (11). Therefore, qPCR became the diagnostic tool of choice, particularly when dealing with challenging-to-cultivate organisms such as AP (12). Moreover, PCR is recognized for its high specificity and sensitivity, coupled with reduced labor and turnaround time in comparison with traditional bacteriological methods (13,14,15).

Two species-specific probe-based qPCR assays for AP are currently available; the first targets the HPG-2 region and was developed by Corney et al. (15), then modified by Feberwee et al. (16), while the second assay targets the recN gene (DNA repair protein gene) and was developed by Kuchipudi et al. (8). Furthermore, a conventional PCR (17) targeting the HPG region is also available for diagnosing AP.

AP is a primary pathogen for chickens (2), which means its presence should lead to clinical disease in naïve flocks. However, multiple layer flocks in several states in the Midwest reported positive qPCR results without any history of clinical signs with complete absence of any respiratory signs, normal egg production, normal feed, and water consumption (18). These results are contradictory and cast strong doubts on the validity of IC diagnosis in these cases based on qPCR-positive results. The specificity of the qPCR assay was confirmed by our team, yielding negative results for various avian respiratory pathogens and other species within the genus Avibacterium, excluding any nonspecific positives. Additionally, isolates were acquired from these flocks and characterized using whole genome sequencing (WGS) with an average nucleotide identity (ANI) of around 96.5% to the AP-type strain NCTC 11296, confirming their taxonomic classification as AP at the genomic level according to the established ANI score for speciation (19,20,21,22). However, these novel isolates were distinct from the classical pathogenic AP, as reported recently (23). Therefore, the explanation of the positive qPCR from normal flocks was attributed to the existence of variant AP strains that are not pathogenic to chickens, which will be referred to as nonpathogenic AP (npAP).

The identification of these new npAP strains in naïve layer flocks complicates the lab diagnosis of IC. All currently published PCR assays (8,15,17), as well as matrix‐assisted laser desorption ionization time‐of‐flight (MALDI‐TOF) mass spectrometry (24), cannot differentiate between the pathogenic AP and the npAP. Recent data show that npAP population is widely spread among the healthy layer flocks in the top table egg–producing states in the United States (25). The wide spread of npAP and the lack of differential qPCR represent a significant challenge for IC diagnosis and in turn a significant challenge to any disease prevention, control, and eradication effort in the face of this emerging outbreak.

Therefore, the purpose of this study is to develop and validate two new qPCR assays capable of distinguishing between the pathogenic AP (typically associated with clinical manifestations) and the newly discovered npAP strains (isolated from normal naïve layer flocks) directly from clinical samples. This paper provides a comprehensive account of our assays’ design and validation procedures, as well as their performance against clinical isolates and clinical samples.

Twelve oropharyngeal swab pools (five birds/pool) from six completely normal layer flocks known to be positive for npAP were submitted for testing using AP-specific PCRs, one conventional (17) and two qPCRs (8,15), to test their capacity in differentiating between pathogenic and npAP. Additionally, two sets of 15 sentinel birds that initially tested negative for AP were placed in two of these layer flocks at two different time points and then retested again after 15 days. Moreover, the presence of the HPG and recN targets in the npAP strains was also verified in silico after WGS analysis confirming that acquired isolates belong to AP species.

To locate specific target regions for the new qPCR assays that can differentiate between the pathogenic AP and the new npAP strains, comparative genomic analysis was performed. This includes bacterial genomes representing 43 pathogenic AP strains (Table 1), the newly announced npAP genomes of GenBank accessions CP104914.1 and CP104917.1 (23), and four more recently sequenced npAP isolates. The comparative genomics analysis was done using the BV-BRC Proteome Comparison service at https://www.bv-brc.org/app/SeqComparison (26,27,28) and BV-BRC Comparative Systems service at https://www.bv-brc.org/app/Comparative Systems (28,29,30,31,32).

The analysis yielded multiple genomic differences between AP and npAP genomes; however, two of these differences are key findings. Our first finding pertains to the capsular biosynthetic locus, which includes three functional regions (33), with two of these regions being the export system and one region as the biosynthesis locus with a total of 11–13 genes (34,35,36). Our comparison identified six genes that were present only in the pathogenic AP and absent in npAP strains. We selected the hctA gene to be the qPCR target for the pathogenic AP. This gene is a capsular polysaccharide transporter gene within the capsular biosynthetic locus (35). The hctA gene is a gene of 648 bp in length located from position (181,411 to 182,058) in the reference genome sequence of AP strain ESV-135 with GenBank accession number NZ_CP050316.1. An additional reason for selecting hctA is that it has been described in the literature as a virulence gene for AP, with its knockout resulting in decreased virulence (34). Our second key finding was a large unique insertion found only in the HMTp210 gene of the npAP genomes. This insertion makes the gene approximately three times longer than the HMTp210 of pathogenic AP and disrupts the hyper-variable region (HVR) of this gene leading to negative results with the HVR-specific conventional PCR (37). The HMTp210 gene spans from nucleotide position 871,682 to 888,325 in isolate AG21-0333 (CP104914.1) and from 595,422 to 614,906 in isolate AP-2 (CP104917.1). These insertion sequences are unique across all NCBI databases, so we selected them to be the qPCR target for the npAP.

Therefore, we selected these two targets to be our targets for the differential qPCR, one target specific for the pathogenic AP (hctA) and the other target specific for the npAP (HMTp210). To further confirm the specificity of the two newly identified targets, sequences of the hctA gene and the HMTp210 insertion underwent two rounds of queries using NCBI BLASTn against the nr/nt database (38) as well as against the comprehensive NCBI whole-genome shotgun contigs (wgs) database (39). Sequences for the hctA gene (n = 16) and HMTp210 insertions (n = 2) were downloaded from GenBank in FASTA format. Additionally, four HMTp210 insertion sequences from WGS of additional npAP isolates were generated at the time of target selection. Then the hctA gene and HMTp210 insertion sequences were subjected to multiple sequence alignment using Clustal W (40) in BioEdit version 7.0.5.3 (41) independently. This alignment aimed to identify a highly conserved region among the sequences of each target. A conserved 420 bp segment of the hctA gene and 1087 bp within the np-HMTp210 insertion was identified to serve as the targets for designing primers and probes for our assays. Further verification of the specificity of the 420 bp to pathogenic AP only and 1087 bp to npAP only was done with one additional BLAST search against both the NCBI nr/nt and wgs databases.

The primers and TaqMan® probes were designed within the selected regions using the Eurofins Genomics qPCR Primer and Probe Design tool (https://eurofinsgenomics.eu/en/ecom/ tools/qpcr-assay-design/). Following this, a manual review and adjustments were conducted in accordance with the established design principles (42,43). The in silico specificity check for the designed primers and probes was done using NCBI BLASTn against the nr/nt database with standard presets. Analysis of the oligonucleotides for the presence of primer dimers and secondary structures was done using the Oligo Analysis tool (http://www.operon.com/tools/oligo-analysis-tool.aspx). Additionally, the online Integrated DNA Technologies (IDT, Coralville, IA) oligo Analyzer 3.1 tool (https://www.idtdna.com/calc/ analyzer) was utilized to calculate the stability of any secondary structure or primer-dimers using the Gibbs free energy (ΔG) (44,45).

All primers and probes were synthesized by IDT. The TaqMan probes were labeled with the reporter dye FAM on the 5′ end and the quencher IABkFQ (Iowa Black® FQ, IDT, Coralville, IA) on the 3′ end, in addition to the internal ZEN quencher. The sequences of the primers and probes for each assay are presented in Table 2.

Similar conditions were adopted for both assays using Real-Time PCR System 7500 (Applied Biosystems, Carlsbad, CA) with standard 7500 run mode. A total of 20 µl reaction volume was utilized, including TaqMan Fast Virus 1-step Master Mix (5 µl; Applied Biosystems, Carlsbad, CA), primers (0.4 µmol final concentration), probe (0.2 µmol final concentration), 8.240 µl of water, and 5 µl of DNA template. The following amplification conditions were adopted: 50 C for 5 min; 95 C for 20 sec with optics off; 40 cycles of 95 C for 15 sec followed by 60 C for 60 sec with optics on.

Each run included a negative control (PCR-grade H₂O) and a positive control (extracted DNA from pathogenic and npAP isolates confirmed by WGS). The SDS 1.5.1 software (Applied Biosystems, Carlsbad, CA) was used for results analysis with manual cycle threshold (C_T) and auto baseline analysis settings. The C_T was identified by multiplying the delta Rn of the peak fluorescence by 0.05 (46).

The primers and probe of the current recN assay (8) were ordered from IDT and used as a standard screening assay for universal detection of AP (both pathogenic and npAP).

Given the recent discovery of the npAP, and lack of any differential tools to distinguish between the two populations, no golden standard diagnostic assays are available to be used to verify and validate the results of the developed differential qPCR. To overcome this gap, we used conventional PCR and Sanger sequencing of the hypervariable region (HVR) and the HMTp210 insertion to verify the differential qPCR results. The HVR will be successfully amplified and sequenced only in case of pathogenic AP. On the other hand, the HMTp210 insertion will be successfully amplified and sequenced only in case of npAP. We targeted the HVR of 1.6 kbp in the HMTp210 for the pathogenic AP using ⊿ 5-1 forward (5′-GATGGCACAATTACATTTACA-3′) and ⊿ 5-1 reverse (5′-ACCTTGAGTGCTAGATGCTGTAGGTGC-3′) primers designed by Sakamoto et al. (37). The following cycling conditions were used: 95 C for 20 sec (one cycle) followed by 40 cycles of 95 C for 15 sec, 55 C for 10 sec, and 68 C for 2 min, then a cycle of final extension at 68 C for 5 min, while for the HMTp210 insertion, we designed a conventional PCR assay targeting a 438 bp segment within the 1087 bp as unique and conserved within the HMTp210 insertions in the npAP, as described above. These primers were designed using Primer-Blast (https://www.ncbi.nlm. nih.gov/tools/primer-blast/) and subsequently refined through manual adjustments. The sequences of the newly designed primers are F: 5′-GTATTGAGTTAGGAACGGATG-3′ and R: 5′-CAACGGTAAAGGTTTGAGAG-3′. To ensure specificity and prevent potential primer dimer formation, the primers underwent in silico verification using the same tools mentioned earlier. The following thermal profile was adopted based on primers melting temperature: 95 C for 20 sec (one cycle) followed by 40 cycles of 95 C for 15 sec, 57 C for 10 sec, and 68 C for 30 sec, and then a final extension at 68 C for 5 min (one cycle).

For both assays, a total of 50 µl reaction mix using Superscript III One-Step RT-PCR System with Platinum Taq DNA Polymerase (Thermo Fisher Scientific, Waltham, MA), forward primer 2 µl (0.4 µmol final concentration), reverse primer 2 µl (0.4 µmol final concentration), and 14 µl of PCR-grade water was applied. Positive PCR products were purified using the ExoSAP-IT kit (Applied Biosystems, Carlsbad, CA) (47) and submitted for Sanger sequencing (DNA facility, Iowa State University).

Extraction of nucleic acids from different bacterial and viral isolates as well as clinical samples was carried out with a Kingfisher-Flex instrument (Thermo Fisher Scientific, Waltham, MA) using a MagMAX™ Pathogen RNA/DNA Kit (Thermo Fisher Scientific, Waltham, MA) following the manufacturer’s instructions. The nucleic acids were extracted from 100 µl of each sample/isolate and eluted into 90 µl of elution buffer.

The process of qPCR assay validation and evaluation was done following the generally published guidelines (48,49,50,51,52).

Inclusivity was tested to determine the ability of each assay to detect different isolates and clinical samples representing the corresponding target agent (pathogenic AP for hctA assay and npAP for the np-HMTp210 assay).

Regarding the pathogenic AP (pathogenic AP isolates in this paper are defined as isolates obtained from clinically diseased flocks showing clinical signs suggestive of IC as well as containing both the capsular biosynthetic locus [involving hctA] and hypervariable region [HVR] of the typical HMTp210 gene), we obtained 28 isolates (the AP ATCC 29545 reference strain + 27 field isolates) from the bacteriology section at the Iowa State University–Veterinary Diagnostic Laboratory (ISU-VDL) and 10 clinical oropharyngeal swab pools (five swabs/pool) from clinically positive cases. Confirmation of pathogenic AP isolates and clinical samples was done based on Sanger sequence analysis of the HVR in the HMTp210 gene. The hctA assay’s inclusivity was evaluated using pathogenic isolates and clinical samples shown in Supplemental Table 1.

For npAP (npAP isolates in this paper are defined as isolates obtained from normal layer flock as well as containing the unique insertions in the HMTp210 gene), our initial validation included six npAP isolates and the two initial isolates (AG21-0333 and AP-2), along with four more isolates that were retrieved from normal flocks (53). Additionally, 86 oropharyngeal swab pools, with each pool representing five birds, from normal healthy flocks with positive PCR results using the recN assay, were also included in the investigation. The confirmation of the nonpathogenic status of the six isolates and the 86 swab pools was done based on the outcomes of our conventional PCR assay and Sanger sequencing analysis. The np-HMTp210 assay’s inclusivity was evaluated using the nonpathogenic isolates and samples in Supplemental Table 1.

To determine if the qPCR assays are not reacting to non-target agents (i.e., causing false positive results), exclusivity was assessed. To rigorously investigate this, we tested both assays against a panel of 63 non-AP bacterial and viral isolates representing microorganisms commonly found coexisting with AP in oropharyngeal samples (Supplemental Table 1). The bacterial isolates included non-AP Avibacterium species (n = 27), Gallibacterium anatis (n = 12), Pasteurella multocida (n = 5), Mycoplasma gallisepticum (n = 1), Mycoplasma iowae (n = 1), Mycoplasma synoviae (n = 1), Bordetella avium (n = 2), Ornithobacterium rhinotracheale (n = 2), Escherichia coli (n = 1), Erysipelothrix rhusiopathiae (n = 1), and Staphylococcus aureus (n = 1). The viral isolates were infectious bronchitis virus (IBV) (n = 3), infectious laryngotracheitis (ILT) (n = 2), Newcastle disease virus (NDV) (n = 2), and avian reovirus (n = 2).

Moreover, to investigate the assays’ interaction with the normal microbial flora present in clinical samples, we included 101 oropharyngeal swab pools (each containing five swabs) that tested negative using the recN qPCR assay. We also tested the six npAP isolates and 86 npAP samples with the hctA assay, while the 28 pathogenic AP isolates and the 10 clinical samples were tested with the np-HMTp210 assay to determine cross-reactivity of the assays.

The analytical sensitivity was tested to determine the limit of detection (LOD), C_T cutoff value, and the limit of quantification (LOQ) of the two assays.

The LOD is the lowest copy number of the target that each assay can detect in more than 95% of the tested replicates (54). Once the LOD was determined, the equivalent C_T value was chosen as the C_T cutoff value of the assay (55). LOQ is the lowest number of copies that can be quantified where the linearity is maintained on the standard curve. To calculate the previous parameters, the following steps were conducted.

Generation of the calibration curve (standard curve): The gBlock was diluted using a tenfold dilution series (1×10⁹ to 1×10⁰ copies/µl) followed by three twofold serial dilutions from the last positive dilution to generate the standard curve. The average C_T value for each dilution was calculated based on the results from three independent runs, and each contains four replicates. To create linear equations with R² values for the two targets, the average C_T values were plotted against log₁₀ of tenfold serial dilutions of the construct (copy number/µl).

Using the generated standard curve, the following parameters were also estimated:

• Coefficient of determination (R²): to evaluate the linearity of the data and pipetting/dilution errors

• qPCR efficiency (%): it is calculated from the slope of the standard curve according to the equation of efficiency = (10^[−1/slope] – 1) × 100, and we used the ThermoFisher Scientific qPCR Efficiency Calculator tool (qpcr-efficiency-calculator.html)

• Dynamic range: the range of the standard curve’s maximum and minimum detectable copy numbers when acceptable linearity (R² ≥ 0.98) and efficiency (between 90% and 110%) are seen.

Repeatability (intraassay variation) and reproducibility (interassay variation). To evaluate the robustness and precision of each assay, the standard deviation (SD) of the average C_T value was calculated. We tested each dilution four times in each run (repeatability) in three independent runs (reproducibility), and the SD was estimated for each run independently to evaluate the repeatability’s precision, and then was calculated for the three runs to evaluate the reproducibility of each assay.

The samples obtained from the six npAP-positive flocks showed specific positive results in all three PCR assays used, as did the sentinel birds introduced afterward. Furthermore, the in silico check confirmed the presence of the target segments in the npAP strains, confirming the assays’ lack of differential capacity between pathogenic and npAP strains.

A 420-bp segment that is present only in pathogenic AP was selected within the hctA gene. BLAST analysis against NCBI BLAST nr/nt and whole-genome shotgun contigs databases confirmed its specificity. The hctA assay’s forward and reverse primers were designed to amplify a 124 bp segment within the 420 bp segment while the probe-annealing site was five nucleotides away from the 3′ end of the forward primer.

Primers were designed to amplify a 132-bp segment in the 1087 bp segment that was selected and BLASTed as a unique segment in npAP. The np-HMTp210 assay’s probe was designed to anneal four nucleotides away from the 3′ end of the forward primer.

The primers’ melting temperature (Tm) difference was within 0.5 C, and the Tm of probes was higher than the primers’ Tm by 5 C and 9 C in the hctA and np-HMTp210 assays, respectively. The data of oligonucleotide sequences and locations concerning the reference strains are presented in Table 2.

The BLAST analysis revealed that the primers and probes of the hctA and np-HMTp210 assays exhibited a higher specificity toward the pathogenic AP and the npAP, respectively. Concerning the formation of primer dimers and primer/probe dimers, no significant secondary structures were observed with a ΔG value for each structure falling below the acceptable threshold of −9 kcal/mol.

The conventional PCR result’s analysis relied on the product size visualization using Qiaxcel (1.6 kb for the pathogenic AP HMTp210 HVR and 438 bp for the HMTp210 insertion of the nonpathogenic AP) as well as Sanger sequencing.

We generated 28 HVR Sanger sequences (1.6 kb) for the 28 pathogenic AP isolates as well as 10 Sanger sequences (1.6 kb) for the 10 clinical samples. The analysis showed that all of them are true positives to pathogenic AP. At the same time, we generated 6 and 86 Sanger sequences of the HMTp210 insertion (438 bp) for the npAP isolates and samples, respectively. Analysis of these sequences confirmed all of them as true-positive npAP, which agrees with the developed qPCR results from these samples.

The hctA qPCR assay showed high inclusivity against all 28 tested pathogenic AP isolates and 10 clinical samples, with no cross-reactivity against any other samples included in the validation process. The np-HMTp210 assay showed positive results only with the npAP 6 isolates and 86 samples, but gave negative results across all other tested isolates and samples (Table 3). No cross-reactivity is demonstrated in the qPCR amplification plot shown in Fig. 2.

After completing the validation using the samples described above and during wider surveillance efforts, samples were collected from 80 healthy layer sites. Out of 80 sites, 28 sites were positive with the recN PCR assay and 24/28 were positive for the np-HMTp210 assay and negative for the hctA assay, indicating that these are indeed npAP. In four sites (4/28) across two different states we acquired 21 OP swab pools that tested positive for both np-HMTp210 and hctA assays. From these sites, we successfully isolated nine isolates, out of which seven isolates were positive for both targets and the two remaining isolates were negative for both targets.

The LOD of the hctA and np-HMTp210 assays is 1 copy/µl (1 × 10³ copies/ml) and 2.5 copies/µl (2.5 × 10³ copies/ml), respectively. These values correspond to C_T cutoff value of 37.17 and 35.08. As for the LOQ, both assays have been limited to 10 copies/µl (1 × 10⁴ copies/ml) because they could not maintain the linearity all the way to their LOD.

Depending on the standard curve generated as shown in Fig. 3, both assays had R² of 0.9999. The overall efficiency of the hctA and np-HMTp210 assays was 94.62% and 92.99%, respectively. Both assays showed a wide dynamic range from C_T 16.23 to C_T 33.55 for the hctA assay and C_T 9.01 to C_T 33.59 for the np-HMTp210 assay (Table 4).

Based on the number of the true positives and true negatives tested by each assay, both assays showed a diagnostic specificity and diagnostic sensitivity of 100%. The hctA assay’s results were 38 true positives (28 isolates and 10 samples of pathogenic AP), 256 true negatives (101 known negative swab pools, 86 npAP swab pools, 63 non-AP isolates, and 6 npAP isolates), zero false positives, and zero false negatives. Although the np-HMTp210 assay’s results were 92 true positives (86 nonpathogenic swab pools and six nonpathogenic isolates), 202 true negatives (101 known negative swab pools, 10 pathogenic AP clinical samples, 63 non-AP isolates and 28 pathogenic AP isolates), zero false positives, and zero false negatives.

The intraassay variation all the way to their LOQ was evaluated in terms of C_T standard deviation (SD), which ranged from 0.017 to 0.43 (first run), 0.02 to 0.48 (second run) and 0.07 to 0.18 (third run) for the hctA assay. On the other hand, the SD of the nt-HMTp210 assay’s C_T values ranged from 0.03 to 0.67 (first run), 0.005 to 0.36 (second run), and 0.04 to 0.39 (third run). The SD representing the interassay variation ranged from 0.03 to 0.55 and 0.03 to 0.24 for the hctA and nt-HMTp210 assays, respectively (Supplemental Table 2).

One of the foremost challenges confronting the poultry industry in the United States and across the world pertains to bacterial respiratory diseases (56). A major one among these diseases is IC, which poses a significant threat to the poultry industry worldwide. In the United States, the recognition of IC as an emerging poultry disease dates back to 2018, during which Pennsylvania has witnessed a series of outbreaks (5). Concurrently, numerous states situated in the northeastern and the Midwestern regions of the United States have grappled with multiple IC outbreaks causing significant economic losses (7,8).

Confirmation of IC diagnosis hinges on the laboratory verification of the presence of AP in support of the characteristic clinical symptoms and post-mortem lesions (1). Bacterial isolation is the traditionally preferred method for detecting, identifying, and diagnosing the bacterial diseases (57,58). However, isolating AP in cultures is a challenging and time-consuming endeavor (9). On the other hand, utilizing qPCR offers an efficient confirmatory diagnostic tool due to its advantages in terms of time efficiency, specificity, and sensitivity.

Recently, a distinct phenomenon concerning AP has come to our attention, where completely normal layer flocks, without prior AP exposure or vaccination, have shown positive PCR results due to the presence of a newly discovered npAP population (25). This phenomenon was confirmed with the isolation and characterization of npAP isolates from layer flocks (23). The npAP isolates appear to lack pathogenic characteristics, evidenced by the frequent isolation of npAP from completely healthy layer flocks with no previous history of IC or vaccination as well as from naïve sentinel birds placed within these flocks. However, verification of the lack of pathogenicity of npAP through experimental studies is necessary. All currently available PCR assays cannot differentiate between these two AP populations, which represents an IC diagnostic dilemma (8,15,17). Our approach was guided by a comparative genomic analysis of pathogenic and npAP strains, leading us to select two targets for our PCR assay design: the hctA gene for the hctA assay (targeting the pathogenic AP) and the HMTp210 insertion for the np-HMTp210 assay (targeting the nonpathogenic AP).

The newly developed PCR assays demonstrated a very high level of specificity. The two assays correctly identified AP in a wide range of samples, including pathogenic AP using the hctA assay and npAP through the np-HMTp210 assay. Notably, these assays exhibited no cross-reactivity with any other tested microorganisms, even those that are taxonomically related, such as Pasteurella multocida and Gallibacterium anatis. Both the hctA and np-HMTp210 qPCR assays exhibit a high analytical sensitivity. Following the guideline that dictates 95% of tested replicates should yield positive results (54), we estimated the LOD for the hctA assay at 1 copy/µl (equivalent to 1 × 10³ copies/ml), while for the np-HMTp210 assay, it was 2.5 copies/µl (equivalent to 2.5 × 10³ copies/ml).

When interpreting qPCR results, it is imperative to address vital quality control parameters, including R² and efficiency (51). The R² value serves to assess data linearity (59) and to demonstrate the variability present in the assay replicates. A threshold of R² surpassing 0.98 is pivotal to deem the qPCR assay as stable (48), and ideally, this value should span across five or six log₁₀ concentrations (51). In our newly developed hctA and np-HMTp210 assays, the linearity, reflected by R², was maintained (R² = 0.9999) over six and eight logs of dynamic range, respectively. However, it is important to note that both assays did not maintain linearity all the way to the LOD; therefore, the LOQ for both assays remained at 10 copies/µl.

The efficiency of PCR holds significant importance as it serves as a key metric for evaluating the effectiveness of a qPCR assay (60). In an ideal scenario, the qPCR efficiency would stand at 100%, signifying a doubling of the PCR product with each successive amplification cycle (61), However, achieving this level of efficiency consistently over time is a challenging endeavor. In practical terms, this efficiency parameter is typically observed within the range of 90% to 110% (52,62) or, alternatively, between 90% and 105% as suggested in (48). The overall efficiencies of the newly designed hctA and np-HMTp210 qPCR assays were estimated to be 94.62% and 92.99%, respectively, both of which fall within the acceptable limit. For the qPCR assay to be established as a reliable diagnostic test, it is essential that it exhibits both repeatability and reproducibility. Remarkably, the two developed qPCR assays demonstrated a high level of consistency not only with the replicates of a single run but also across the independent conducted runs. This underscores their robustness and establishes their potential as a dependable diagnostic tool in the foreseeable future.

When evaluating the diagnostic performance of the designed qPCR assays, the estimation of diagnostic specificity and diagnostic sensitivity stands out as the primary performance indicators established during the assay validation process (50). These calculations are primarily based on the identification of true positive and true negative samples. In the case of the newly developed assays, both diagnostic specificity and diagnostic sensitivity were determined to be 100%, indicating the absence of false positives and false negatives with no cross-reaction between the two assays and corresponding AP strains. However, after successfully completing the validation process, which included a substantial number of representative samples and confirming the effectiveness of the newly designed qPCRs as a reliable tool for distinguishing between pathogenic and npAP, our diagnostic assays faced a new challenge. While conducting a surveillance study looking for prevalence of npAP (25), we identified four sites of completely normal layer flocks with no clinical signs and no prior history of IC vaccination or infection. Despite this, both sites yielded positive results for both the hctA and np-HMTp210 assays. We initially thought that these results represent a mixed infection in those four sites. As a follow-up, nine AP isolates were obtained from these sites, and six representative isolates were subsequently subjected to WGS. Sequence analysis of four out of six isolates unveiled that these isolates harbored both the unique insertion in HMTp210 and the capsular polysaccharide locus including hctA gene, while the remaining two isolates revealed an even more surprising finding. Similar to other npAP, these two isolates still have large insertions in the HMTp210 gene, which renders the gene much longer than the pathogenic AP (16725 bp in npAP compared to approximately 6250 bp in AP). However, the sequence of these insertions did not contain the conserved sequence that we had used as a target for the developed differential qPCR. The genomes of these two isolates were still devoid of hctA. Hence, they were negative for both PCR assays.

These new findings suggest this newly discovered npAP population is less homogenous than we had initially anticipated. Even within the same site, a mixed population of npAP isolates can exist. This is challenging when attempting to develop a new universally differential diagnostic tool. Although this assay is capable of differentiating between pathogenic AP and npAP in the great majority of cases, these new findings render the differential capacity of the developed qPCR incomplete. However, the differential qPCR assays developed in this paper are still valuable assays. At the time of writing, the U.S. poultry industry is concurrently facing multiple emerging diseases, including IC. The incomplete differential capability of these qPCR assays still represents a useful tool to combat this emerging disease. Meanwhile, we will continue to study this npAP population and fine-tune our target selection, aiming to find new targets that can either complement the two selected targets or completely replace them. One or more new targets could be added to capture the parts of the population that are initially missed by the two targets. Alternatively, one or more targets could replace one or both of the current targets to achieve the complete differentiation between the two AP populations.

Although this finding may limit the effectiveness of the developed differential qPCR assay in distinguishing pathogenic AP from npAP, it underscores the importance of investigating and identifying the primary virulence factors for AP. Previous studies have established the role of the capsular polysaccharide as a crucial virulence determinant for AP, evidenced by challenges with nonencapsulated AP resulting in either no clinical signs (63) or reduced virulence (34). However, our results uncovered an npAP population having the capsular polysaccharide with extra-long insertions in the HMTp210 gene, which draws the attention to the HMTp210 gene as the main potential virulence factor, with its function possibly disrupted by these insertions. Given our access to the WGS data of the most recent isolates, we have multiple putative virulent targets currently under investigation. Therefore, we aim to investigate the primary virulence genes to enhance our understanding of AP epidemiology and to refine prevention, control, and eradication strategies. Our future objectives entail identifying other specific qPCR targets for the development of a complete differential qPCR assay.

Currently, the only diagnostic work that can reliably differentiate between pathogenic AP and npAP is isolation, conducting WGS and comparative genomics on the isolates. Although the newly developed qPCR assays represent a significant advancement in differentiating between pathogenic and npAP in most cases, the complete reliance on them may be compromised because of the continual emergence of new populations of npAP. Addressing this challenge necessitates further efforts to identify target genes for the consistent differentiation between pathogenic and npAP, thereby alleviating the current diagnostic uncertainty surrounding IC.

Supplemental data associated with this article can be found at https://doi.org/10.1637/aviandiseases-D-24-00041.s1.

The authors acknowledge the Ministry of Higher Education and Scientific Research of the Arab Republic of Egypt for funding Mostafa Shelkamy’s Ph.D. program at College of Veterinary Medicine, Iowa State University. The study was supported in part by funding from the Egg Industry Center (Grant No. SG2706645).

ANI =

average nucleotide identity;

AP =

Avibacterium paragallinarum;

C_T =

cycle of threshold;

D-SN =

diagnostic sensitivity;

D-SP =

diagnostic specificity;

FN =

false negative;

FP =

false positive;

IBV =

infectious bronchitis virus;

IC =

infectious coryza;

IDT =

Integrated DNA Technologies;

ILT =

infectious laryngotracheitis;

LOD =

limit of detection;

LOQ =

limit of quantification;

MALDI‐TOF =

matrix‐assisted laser desorption ionization time‐of‐flight;

NAD =

nicotinamide adenine dinucleotide;

NDV =

Newcastle disease virus;

npAP =

nonpathogenic AP strains;

OP =

oropharyngeal;

RT-PCR =

reverse transcription polymerase chain reaction;

SD =

standard deviation;

Tm =

melting temperature;

TN =

true negative;

TP =

true positive;

WGS =

whole genome sequencing;

qPCR =

real-time quantitative PCR