The methodology for generating IBV attenuated strains using a versatile backbone applicable to variants

The IBV variant presents a continuous challenge for disease prevention and control. We strategically employed attenuated vaccine strains as a robust backbone. By incorporating SAVE, we successfully mitigated the pathogenicity of these viral variants. Our comprehensive screening process identifies an attenuated backbone that is optimally suited for the variant strains of IBV. The attenuated backbone serves as a versatile platform capable of efficiently integrating S genes from various IBV strains through the TAR cloning technique. This targeted approach ensures the development of vaccines with enhanced safety and efficacy profiles tailored to address the unique challenges posed by the ever-evolving IBV landscape. A flowchart summarizing the selection and testing of the attenuated backbone is shown in Fig. 1.

Fig. 1: Schematic overview of the attenuated backbone selection and validation process.

This chart provides a streamlined visual summary of the steps involved in identifying and evaluating the attenuated backbone. For comprehensive insights, refer to the accompanying text. The CPB (purple) of a coding sequence is scored as the mean of each codon pair score. The codon pair score (CPS; blue) is determined by calculating the natural log of the ratio of the total number of times a given codon pair is observed in an organism's coding genome to the number of times the codon pair is expected to appear. CpG and TpA dinucleotides are highlighted in red and green, respectively.

The attenuation of the rH-QX(S) recombinant strain was incomplete

To manipulate the IBV genome rapidly and effectively, we developed an IBV reverse genetic system based on TAR cloning in yeast (Fig. 2A). The rH-QX(S) recombinant strain was rapidly rescued via TAR cloning (Fig. 2B), following a method described in the literature that can attenuate QX-type virulent strains41,42,43. The growth kinetics demonstrated that rH120 and rH-QX(S) replicated with comparable efficiency in chicken embryos (Fig. 2C).

Fig. 2: Rescue and pathogenicity analysis of the rH-QX(S) recombinant strain.

A Schematic workflow of yeast-based TAR cloning for the IBV reverse genetics system. Step 1: PCR amplification of subgenome fragments from IBV; step 2: co-transformation of subgenome DNA fragments and a linearized pYES1L vector into yeast; step 3: screening of single colonies; step 4: sequencing analysis of the recombinant plasmid; and step 5: virus rescue. B Schematic representation of the IBV H120 genome structure and the construction process for the recombinant strains rH120 and rH-QX(S). C Growth kinetic analysis of the rH120 and rH-QX(S) recombinant strains, with viral RNA copies quantified by RT-qPCR. The data are presented as the means ± SDs (n = 3). Statistical significance was determined by Student's t test. D Survival rate of chickens infected with QX (wild-type), rH120, rH-QX(S), or PBS within 14 dpi. E Viral load assessment in the trachea, lungs, and kidneys of chickens from the QX (WT), rH120, and rH-QX(S) groups at 3, 6, 9, and 14 dpi. The dashed line indicates the PBS control group. The data are presented as means ± SDs (n = 2). F Viral shedding analysis of the throat and cloaca of chickens from the QX (WT), rH120, and rH-QX(S) groups at 6 and 12 dpi. The dashed line represents the PBS control group. The data are presented as means ± SDs (n = 10). One-way ANOVA was applied to assess the significance of differences in viral load and shedding among groups (E, F), where ns indicates not significant, *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001. (G) Gross lesion evaluation, (H) histopathological examination, and (I) IHC of trachea, lungs, and kidneys from chickens in the QX (WT), rH120, and rH-QX(S) groups.

The pathogenicity of the rH-QX(S) strain was subsequently assessed. We closely monitored the patients' clinical symptoms and recorded mortality rates from 1 to 14 d post-infection (dpi). The QX (WT) group presented with symptoms such as tracheal rales and sneezing (data not shown), resulting in a mortality rate of 80%. In contrast, the rH-QX(S) group presented milder symptoms and a mortality rate of 10% (Fig. 2D). RT-qPCR analysis of the viral load in tissues revealed that compared with the QX(WT) group, the rH-QX(S) group presented a significant reduction in the trachea at 14 dpi (P < 0.01) and in the kidneys at 3, 6, and 14 dpi (P < 0.05 or P < 0.01) (Fig. 2E). Furthermore, viral shedding in the throat and cloaca was significantly lower in the rH-QX(S) group than in the QX(WT) group at 6 and 12 dpi (P < 0.01 or P < 0.0001) (Fig. 2F). The rH120 group showed lower viral load and shedding than the QX(WT) and r-HQX(S) groups.

Upon necropsy, the QX(WT) group presented significant tissue damage, characterized by pronounced tracheal and lung hemorrhages, along with urate deposition in the kidneys. In stark contrast, the rH-QX(S) group demonstrated notable alleviation of the severity of these lesions, with only minor tracheal hemorrhages, localized lung congestion, and no discernible kidney lesions (Fig. 2G). The histopathology results revealed notable exfoliation of the ciliated epithelium in the trachea, hemorrhage and inflammatory cell infiltration in the lungs, and glomerular atrophy coupled with necrosis of the renal tubule epithelial cells in the QX(WT) group. In comparison, the rH-QX(S) group presented a reduction in histopathological damage across all tissues, with shedding of mucosal epithelial cells and infiltration of inflammatory cells in tracheal tissues, hemorrhage and inflammatory cell infiltration in the lungs, and no significant kidney lesions (Fig. 2H). However, in the rH120 and PBS groups, no significant lesions or histopathological changes were observed in the trachea, lungs, and kidneys (Fig. 2G, H). Immunohistochemistry revealed a notable decrease in IBV antigen-positive cells within the trachea and lung tissues of the rH-QX(S) group. In contrast, no IBV antigen was detected in the kidney tissues (Fig. 2I). These findings indicated that the rH-QX(S) recombinant strain presented reduced virulence.

In conclusion, although the rH-QX(S) recombinant strains have demonstrated a certain level of attenuation, the attenuation of the rH-QX(S) recombinant strain was incomplete, as evidenced by the observed chicken mortality, tracheal hemorrhage, ciliary loss, and IBV antigen-positive cells within the trachea and lung tissues of the rH-QX(S) group. However, this research underscores the potential of S gene swapping with attenuated strains as a precise and effective strategy for attenuating IBV.

The recoded subgenomic fragments of H120 exhibit lower codon pair bias (CPB)

The subgenomic fragments of the H120 strain were recoded to incorporate lower CPB, a strategy aimed at further attenuating the rH-QX(S) recombinant strain. To achieve this, the HF7 (including NSP14, NSP15, and NSP16) nucleotide sequence within the H120 genome was recoded using codon pairs that are strongly underrepresented in Gallus gallus. This approach was applied to the potential virulence regulatory genes NSP14, NSP15, NSP16, and HF7 in H120, resulting in the creation of CPD14, CPD15, CPD16, and CPDF7. A critical consideration in this process was the preservation of the transcription regulatory sequence (TRS) of the S gene, which is located at the 3' end of ORF1b in the H120 genome. To maintain the S gene's functionality downstream of HF7, we retained approximately 100 base pairs (bp) within the TRS region (specifically, the sequence from 20214 to 20313 bp, containing the CUGAACAA motif) during the coding process. Compared with the original parental sequences, the sequences of the recoded subgenomes presented a notable increase in the frequency of CpG and TpA dinucleotides coupled with a reduced CPB (Table 1). This re-coding is expected to contribute to a further reduction in the virulence of the rH-QX(S) strain.

Table 1 Characteristics of the parent sequence and CPD sequence of H120 subgenome fragmentsAssessment of pathogenicity of the recombinant strains rH-CPD14-QX(S), rH-CPD15-QX(S), rH-CPD16-QX(S), and rH-CPDF7-QX(S) in 1-day-old SPF chickens

To further diminish the virulence of the rH-QX(S) recombinant strains, this study employed the CPD strategy to recode specific segments of the H120 genome (Fig. 3A). The growth kinetics revealed that at 48 hours postinoculation (hpi), the number of RNA copies of each recombinant strain was significantly lower than that of the rH120 strain, with the following statistical significance: rH-CPD14-QX(S) (P < 0.001), rH-CPD15-QX(S) (P < 0.001), rH-CPD16-QX(S) (P < 0.01), and rH-CPDF7-QX(S) (P < 0.001). These findings suggested that the CPD encoded subgenomic segments influenced the replication efficiency of the recombinant strains (Fig. 3B).

Fig. 3: Pathogenicity assessment of the recombinant strains rH-CPD14-QX(S), rH-CPD15-QX(S), rH-CPD16-QX(S), and rH-CPDF7-QX(S) in 1-day-old SPF chickens.

A Schematic representation of the construction process for the recombinant strains rH120, rH-CPD14-QX(S), rH-CPD15-QX(S), rH-CPD16-QX(S), and rH-CPDF7-QX(S). B Growth kinetics analysis of the recombinant strains rH120, rH-CPD14-QX(S), rH-CPD15-QX(S), rH-CPD16-QX(S), and rH-CPDF7-QX(S). Viral RNA copies were quantified via RT-qPCR. The data are expressed as means ± SDs (n = 3). Statistical significance was determined via Student's t test, *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001. C Survival rates of chickens infected with QX (WT), rH120, rH-CPD14-QX(S), rH-CPD15-QX(S), rH-CPD16-QX(S), rH-CPDF7-QX(S), and PBS within 14 dpi. D Viral load assessment in the trachea, lungs, and kidneys of chickens from the QX (WT), rH120, rH-CPD14-QX(S), rH-CPD15-QX(S), rH-CPD16-QX(S), and rH-CPDF7-QX(S) groups at 3, 6, 9, and 14 dpi. The dashed line represents the PBS control group. The data are presented as means ± SDs (n = 2). E Viral shedding analysis of the throat and cloaca of chickens from the QX (WT), rH120, rH-CPD14-QX(S), rH-CPD15-QX(S), rH-CPD16-QX(S), and rH-CPDF7-QX(S) groups at 6 and 12 dpi. The dashed line indicates the PBS control group. The data are shown as means ± SDs (n = 10). One-way ANOVA was used to evaluate the significance of differences in viral load and shedding among groups (D, E), where ns indicates not significant. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001. F Gross lesion assessment and (G) histopathological examination of trachea, lung, and kidney tissues from the QX (WT), rH120, rH-CPD14-QX(S), rH-CPD15-QX(S), rH-CPD16-QX(S), and rH-CPDF7-QX(S) groups.

The pathogenicity of these recombinant strains was subsequently evaluated in 1-day-old SPF chickens. Clinical symptoms and mortality rates were closely monitored from 1 to 14 dpi. The QX (WT) group presented symptoms such as tracheal rales and sneezing, resulting in an 80% mortality rate. In contrast, no significant clinical symptoms or mortality were observed in the PBS, rH120, rH-CPD14-QX(S), rH-CPD15-QX(S), rH-CPD16-QX(S), or rH-CPDF7-QX(S) groups (Fig. 3C). Viral loads in tracheal, lung, and kidney tissues at 3, 6, 9, and 14 dpi, as well as viral shedding at 6 and 12 dpi, were assessed via RT-qPCR. Compared with those in the QX(WT) group, the viral loads in all tissues of the recombinant groups rH-CPD14-QX(S), rH-CPD15-QX(S), rH-CPD16-QX(S), and rH-CPDF7-QX(S) were significantly lower at 3, 6, 9, and 14 dpi (P < 0.05 or P < 0.0001). However, the viral load in the lung tissue of the rH-CPD15-QX(S) group at 6 dpi and the rH-CPD14-QX(S) group at 9 dpi did not differ significantly from that of the QX(WT) group (Fig. 3D). Furthermore, viral shedding in the throat and cloaca was significantly lower in the recombinant groups than in the QX(WT) group at 6 and 12 dpi (P < 0.05 or P < 0.0001) (Fig. 3E). Among all the groups, the rH-CPDF7-QX(S) group presented the least amount of viral shedding in the throat and cloaca at 6 and 12 dpi (P < 0.0001), with no significant difference compared with the rH120 group (Fig. 3E).

Upon necropsy, the QX(WT) group presented with severe tissue damage, which included pronounced tracheal and lung hemorrhages, along with urate deposition in the kidneys. In stark contrast, the PBS, rH120, rH-CPD14-QX(S), rH-CPD15-QX(S), rH-CPD16-QX(S), and rH-CPDF7-QX(S) groups presented no significant tissue lesions. Notably, the rH-CPD14-QX(S) group displayed minor punctate hemorrhages in the trachea (Fig. 3F). Histopathological examination further revealed the severity of the lesions in the QX(WT) group, which included the detachment of mucosal epithelial cells and the infiltration of inflammatory cells in the tracheal tissues, hemorrhage and inflammatory cell infiltration in the lungs, and glomerular atrophy along with necrosis in the epithelial cells of the renal tubules. Conversely, the rH120, PBS, rH-CPD14-QX(S), rH-CPD15-QX(S), rH-CPD16-QX(S), and rH-CPDF7-QX(S) groups presented no significant histopathological alterations in the trachea, lungs, or kidneys (Fig. 3G). Immunohistochemistry revealed that IBV antigen was absent in the trachea, lung, or kidney tissues of the PBS, rH120, rH-CPD14-QX(S), rH-CPD15-QX(S), rH-CPD16-QX(S), and rH-CPDF7-QX(S) groups (Supplementary Fig. 5A).

These results indicate that attenuated QX-type recombinant strains can be rapidly developed using the recoded genomes of H120 as a backbone while integrating the S gene from the QX strain. Among these recoded recombinant strains, the rH-CPDF7-QX(S) strain demonstrated the most significant attenuation, with both the viral load in tissues and viral shedding from the throat and cloaca being significantly reduced, indicating its potential as a promising candidate for further vaccine development.

The CPDF7 subgenome plays a critical role in virus attenuation

To evaluate the impact of the CPDF7 subgenome on virus attenuation, the QF7 fragments within the QX strains were replaced with HF7 from the H120 genome, or the CPDF7 subgenome fragments were recoded via TAR cloning, yielding the recombinant strains rQX, rQX-HF7, and rQX-CPDF7 (Fig. 4A). Growth kinetics indicated that the rQX strain achieved peak replication at 36 hpi. Notably, the rQX-CPDF7 and rQX-HF7 recombinant strains presented significantly fewer RNA copies at 36 hpi than the rQX strain did (P < 0.01), indicating that the insertion of the HF7 and CPDF7 gene fragments significantly influenced the growth characteristics of the rQX-HF7 and rQX-CPDF7 strains (Fig. 4B).

Fig. 4: The CPDF7 subgenome plays a critical role in virus attenuation.

A Schematic representation of the construction process for the recombinant strains rQX, rQX-HF7, and rQX-CPDF7. B Growth kinetics analysis of the recombinant strains rQX, rQX-HF7, and rQX-CPDF7. Viral RNA copies were quantified via RT-qPCR. The data are presented as means ± SDs (n = 3). Statistical significance was determined via Student's t test, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. C Survival rate analysis of chickens infected with rQX, rQX-HF7, or rQX-CPDF7 within 14 dpi. D Viral load assessment in the trachea, lungs, and kidneys of chickens from the rQX, rQX-HF7, and rQX-CPDF7 groups at 3, 6, 9, and 14 dpi. The dashed line indicates the PBS control group. The data are shown as means ± SDs (n = 2). E Viral shedding assessment in the throat and cloaca of chickens from the rQX, rQX-HF7, and rQX-CPDF7 groups at 6 and 12 dpi. The dashed line represents the PBS control group. The data are presented as means ± SDs (n = 10). One-way ANOVA was applied to evaluate the significance of differences in viral load and shedding among groups (D, E), with ns indicating not significant, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. F Gross lesion evaluation, (G) histopathological examination, and (H) IHC of trachea, lung, and kidney tissues from the rQX, rQX-HF7, and rQX-CPDF7 groups.

The pathogenicity of these recombinant strains was subsequently evaluated in 1-day-old SPF chickens, and their clinical symptoms and mortality rates were closely monitored from 1 to 14 dpi. The rQX group displayed symptoms such as tracheal rales and sneezing, with a mortality rate reaching 70%. The rQX-HF7 group presented with signs of dyspnea and tracheal rales, with a mortality rate of 30%. In stark contrast, the rQX-CPDF7 strain induced only mild clinical symptoms, and no mortality was recorded (Fig. 4C). RT-qPCR analysis of the viral load in trachea, lung, and kidney tissues revealed that, compared with those in the rQX and rQX-HF7 groups, the viral load in the rQX-CPDF7 group was significantly lower at 3, 6, 9, and 14 dpi (P < 0.05 or P < 0.001) (Fig. 4D). Furthermore, the rQX-CPDF7 group presented significantly less viral shedding in the throat and cloaca at 6 and 12 dpi than the rQX group did (P < 0.001 or P < 0.0001) (Fig. 4E).

Upon necropsy, the rQX and rQX-HF7 groups presented with severe pathological changes, characterized by disseminated hemorrhage in the trachea, lung congestion and necrosis, and the kidneys displayed a "flower-spotted" appearance due to white urate deposits. In contrast, the rQX-CPDF7 group presented only minor tracheal hemorrhages, with no significant damage observed in the lungs or kidneys (Fig. 4F). Histopathological examination revealed that the rQX and rQX-HF7 groups presented severe tissue damage, including detachment of mucosal epithelial cells in the trachea, hemorrhage, inflammatory cell infiltration in the lungs, atrophy of the glomeruli, and necrosis of the epithelial cells of the renal tubules. The rQX-CPDF7 group, however, presented a marked reduction in histopathological lesions across all the tissues examined, with only minor tracheal epithelial cell shedding, slight lung hemorrhage, and no significant kidney damage (Fig. 4G). Compared with the rQX and rQX-HF7 groups, the rQX-CPDF7 group presented a significant reduction in IBV antigen-positive cells within the trachea and lung tissues, whereas IBV antigen was absent in the kidney tissues (Fig. 4H).

Collectively, these findings suggest that the rQX-CPDF7 group demonstrated reduced pathogenicity and that the CPDF7 subgenome exerted a potent attenuating effect compared with the QF7 and HF7 subgenomes. Notably, the CPDF7 subgenome undergoes many nucleotide alterations, all of which do not result in any amino acid changes. This extensive genetic modification significantly reduces the likelihood of the rQX-CPDF7 strain reverting to a wild-type phenotype, thereby effectively minimizing the risk of reversion to virulence. Furthermore, these results suggest that NSP14, NSP15, and NSP16 are promising candidates for the attenuation of IBV.

The rH-CPDF7 backbone is applicable to the rapid attenuation of IBV variants

To explore the applicability of the rH-CPDF7 backbone in attenuating IBV, the spike (S) genes from both TW-type and GVI-type IBV strains were successfully integrated into the rH-CPDF7 backbone via TAR cloning (Fig. 5A). The growth kinetics indicated that the RNA copy numbers of the engineered rH-CPDF7-TW(S) and rH-CPDF7-GVI(S) strains were notably lower at 36 hpi than those of the rH120 strain, with statistical significance at P < 0.05 and P < 0.01, respectively (Fig. 5B).

Fig. 5: The rH-CPDF7 backbone is applicable for the rapid attenuation of IBV variants.

A Schematic detailing the assembly process for the recombinant strains rH-CPDF7-TW(S) and rH-CPDF7-GVI(S). B Growth kinetics analysis of the recombinant strains rH-CPDF7-TW(S) and rH-CPDF7-GVI(S). Viral RNA copies were quantified via RT-qPCR. The data are presented as means ± SDs (n = 3). Statistical significance was assessed via Student's t test, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. C Survival rate analysis of chickens infected with rH-CPDF7-TW(S) and rH-CPDF7-GVI(S) within 14 dpi. D Viral load measurements in the trachea, lungs, and kidneys of chickens from the rH-CPDF7-TW(S) and rH-CPDF7-GVI(S) groups at 3, 6, 9, and 14 dpi. The dashed line represents the PBS control group. The data are shown as means ± SDs (n = 2). E Viral shedding assessment in the throat and cloaca of chickens from the rH-CPDF7-TW(S) and rH-CPDF7-GVI(S) groups at 6 and 12 dpi. The dashed line indicates the PBS control group. Data are presented as means ± SDs (n = 10). One-way ANOVA was used to evaluate the significance of differences in viral load and shedding among groups (D, E), with ns indicating not significant; *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. F Gross lesion evaluation and (G) histopathological examination of trachea, lung, and kidney tissues from the rH-CPDF7-TW(S) and rH-CPDF7-GVI(S) groups.

The pathogenic potential of the rH-CPDF7-TW(S) and rH-CPDF7-GVI(S) strains was subsequently investigated in 1-day-old specific-pathogen-free (SPF) chickens. Clinical signs and mortality rates were meticulously tracked from 1 to 14 dpi. The TW(WT) group presented with clinical manifestations such as tracheal rales and sneezing, resulting in a 30% mortality rate. The GVI(WT) group presented with signs of dyspnea and tracheal rales, yet without any mortality. In stark contrast, neither clinical symptoms nor mortality was detected in the PBS, rH120, rH-CPDF7-TW(S), and rH-CPDF7-GVI(S) groups (Fig. 5C). The viral loads in the trachea, lung, and kidney tissues, as well as viral shedding at 6 and 12 dpi, were quantitatively analyzed via RT-qPCR. Compared with the TW(WT) group, the rH-CPDF7-TW(S) group presented a marked reduction in the viral load within the trachea at 6 dpi and 9 dpi (P < 0.05) and in the kidneys at 3 dpi (P < 0.05) (Fig. 5D). However, the rH-CPDF7-GVI(S) group did not show a significant difference in viral load within tissues compared with the GVI(WT) group (Fig. 5D). Moreover, the rH-CPDF7-TW(S) group experienced a substantial and highly significant decrease in viral shedding from the throat and cloaca at both 6 dpi and 12 dpi (P < 0.001 or P < 0.0001) relative to the TW(WT) group (Fig. 5E). In alignment with these findings, the rH-CPDF7-GVI(S) group also exhibited notable reductions in viral shedding in the throat and cloaca at 6 and 12 dpi (P < 0.05 or P < 0.0001) compared with the GVI(WT) group (Fig. 5E).

An autopsy revealed that the TW(WT) group presented severe tissue damage, characterized by extensive tracheal hemorrhage and lung congestion, along with a distinctive "flower-spotted kidney" appearance caused by white urate deposits in the kidneys. The GVI(WT) group presented with tracheal hemorrhage and mucous discharge, yet no significant pathological changes were noted in the lungs or kidneys. Notably, the PBS, rH120, rH-CPDF7-TW(S), and rH-CPDF7-GVI(S) groups presented no notable lesions in the trachea, lungs, or kidneys (Fig. 5F). Histopathological examination further revealed significant damage to the trachea, lungs, and kidneys of the QX(WT) group, including the shedding of tracheal mucosal epithelial cells, pulmonary hemorrhage, and renal damage characterized by glomerular atrophy and necrosis of the renal tubular epithelium. The GVI(WT) group also presented tracheal lesions with similar characteristics, but no significant kidney pathology was observed. Conversely, the PBS, rH120, rH-CPDF7-TW(S), and rH-CPDF7-GVI(S) groups were free from significant pathological changes in the trachea, lungs, and kidneys (Fig. 5G).

In conclusion, compared with the homologous wild-type strains, the rH-CPDF7-TW(S) and rH-CPDF7-GVI(S) groups presented significant reductions in mortality, tissue viral load, viral shedding, and tissue lesions. These results indicate that the rH-CPDF7 backbone is applicable to the rapid attenuation of IBV variants, offering a promising approach for live-attenuated vaccines against IBV variants.

Assessment of the immune response and protective effect of vaccine candidate strains rH-CPDF7-QX(S), rH-CPDF7-TW(S), and rH-CPDF7-GVI(S)

Finally, to thoroughly analyze the immune response and protective effects of vaccine candidate strains rH-CPDF7-QX(S), rH-CPDF7-TW(S), and rH-CPDF7-GVI(S), a cohort of 1-day-old SPF chickens was immunized (Fig. 6A). The IBV-specific serum antibody titers of the vaccinated groups increased progressively over the 21 d observation period. Compared with those in the rH120-vaccinated group, the serum antibody levels in the rH-CPDF7-QX(S), rH-CPDF7-TW(S), and rH-CPDF7-GVI(S) groups did not significantly differ at 7, 14, and 21 d postvaccination (dpv) (Fig. 6B). Additionally, at 21 dpv, the levels of cytokines, including IL-2, IL-4, IL-6, and IFN-γ, in the rH-CPDF7-QX(S), rH-CPDF7-TW(S), and rH-CPDF7-GVI(S) groups were comparable to those in the rH120-vaccinated group, yet both sets of vaccinated groups presented significantly higher cytokine levels than did the PBS group (P < 0.0001) (Fig. 6C). Importantly, the rH-CPDF7-QX(S), rH-CPDF7-TW(S), and rH-CPDF7-GVI(S) recombinant strains induced higher levels of neutralizing antibodies than did the rH120 group (P < 0.001 or P < 0.0001) (Fig. 6D).

Fig. 6: Evaluation of the immune response and protective effect of vaccine candidate strains rH-CPDF7-QX(S), rH-CPDF7-TW(S), and rH-CPDF7-GVI(S) in chickens.

A Schematic representation of the immunization protocol for vaccine candidate strains rH-CPDF7-QX(S), rH-CPDF7-TW(S), and rH-CPDF7-GVI(S). B Serum-specific antibody titers of the rH-CPDF7-QX(S), rH-CPDF7-TW(S), and rH-CPDF7-GVI(S) vaccinated groups were assessed via commercial ELISA at 7, 14, and 21 dpv. The dashed line indicates the S/P value of the lowest positive sample (0.2). C Detection of the cytokines IL-2, IL-4, IL-6, and IFN-γ at 21 dpv. D Determination of virus neutralization titers at 21 dpv by serum titration in chicken embryos. The virus neutralization titers are expressed as the log2 of the reciprocal of the highest serum dilution that inhibited 50% of the chicken embryo lesions. The data are presented as means ± SDs (n = 5). E Survival rate analysis of chickens following homologous challenge with the QX, TW, and GVI strains at 21 dpv. F Viral load measurements in the trachea, lungs, and kidneys of chickens from the vaccinated groups at 3, 6, 9, and 14 dpc. The horizontal dashed line represents the PBS group. The data are presented as means ± SDs (n = 2). G Viral shedding assessment in the throat and cloaca of chickens from the vaccinated groups at 6 and 12 dpc. The horizontal dashed lines indicate the PBS group. The data are presented as means ± SDs (n = 10). One-way ANOVA was used to assess the significance of differences in viral load and shedding among the groups (B–D and F–G). ns, not significant; *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. (H) Gross lesion evaluation and (I) histopathological examination of trachea, lung, and kidney tissues from the vaccinated groups.

The vaccinated groups were challenged with 105 EID50 of the homologous virulent strains QX(WT), TW(WT), and GVI(WT) via the ocular and intranasal routes at 21 dpv (recorded as 0 days post-challenge, dpc). Clinical signs were closely monitored, and mortality rates were recorded daily from 1 to 14 dpc (Fig. 6A). Viral loads in the trachea, lung, and kidney tissues were measured at 3, 6, 9, and 14 dpc, and viral shedding was assessed at 6 and 12 dpc via RT-qPCR. The PBS/QX group displayed mild tracheal rales and sneezing, with a 20% mortality rate, whereas the rH120/QX group presented similar symptoms but a lower mortality rate of 10%. The PBS/TW, rH120/TW, PBS/GVI, and rH120/GVI groups presented mild tracheal rales and sneezing, but no mortality. In contrast, no clinical signs or mortality were observed in the rH-CPDF7-QX(S)/QX, rH-CPDF7-TW(S)/TW, rH-CPDF7-GVI(S)/GVI, and negative control (NC, PBS group not challenged) groups (Fig. 6E). Comparative analysis revealed that the rH-CPDF7-QX(S)/QX group had a significantly lower viral load in the trachea and kidneys than the PBS/QX and rH120/QX groups did (P < 0.01 or P < 0.0001) (Fig. 6F). However, no significant differences in viral load were noted among the PBS/TW, rH120/TW, and rH-CPDF7-TW(S)/TW groups or the PBS/GVI, rH120/GVI, and rH-CPDF7-GVI(S)/GVI groups (Fig. 6F). The lung tissue had a lower viral load (data not shown). Compared with those in the PBS/QX and rH120/QX groups, viral shedding in the throat and cloaca of the rH-CPDF7-QX(S)/QX group was significantly reduced (P < 0.01 or 0.0001). While there was no significant difference in the tissue viral load among the PBS/TW, rH120/TW, and rH-CPDF7-TW(S)/TW groups, the rH-CPDF7-TW(S)/TW group presented a significant reduction in viral shedding in the throat and cloaca at 6 and 12 dpc compared with the PBS/TW and rH120/TW groups (P < 0.05 or 0.0001) (Fig. 6G). Similarly, no significant differences in tissue viral load were detected among the PBS/GVI, rH120/GVI, and rH-CPDF7-GVI(S)/GVI groups, but the rH-CPDF7-GVI(S)/GVI group exhibited a significant reduction in viral shedding in the throat and cloaca at 6 dpc and 12 dpc compared with the PBS/GVI and rH120/GVI groups (P < 0.001 or P < 0.0001) (Fig. 6G).

The autopsy findings indicated that the PBS/QX and rH120/QX groups suffered from severe tissue damage, characterized by widespread tracheal hemorrhage, lung congestion, and renal swelling with urate deposition. In the PBS/TW and rH120/TW groups, there was evidence of tracheal larynx hemorrhage accompanied by mucous exudation, whereas the lungs and kidneys showed no significant pathological changes. Similarly, the PBS/GVI and rH120/GVI groups presented tracheal punctate hemorrhages with mucous exudation but no apparent lesions in the lungs or kidneys. In contrast, the rH-CPDF7-QX(S)/QX, rH-CPDF7-TW(S)/TW, rH-CPDF7-GVI(S)/GVI, and negative control (NC) groups presented no significant lesions in the trachea, lungs, or kidneys (Fig. 6H). The histopathological examination further revealed severe tracheal lesions in the PBS/QX and rH120/QX groups, including the loss of tracheal cilia, thickening of mucosal epithelial cells, and inflammatory cell infiltration.

Additionally, there was pulmonary hemorrhage and inflammatory cell infiltration, along with renal damage marked by glomerular atrophy and necrosis of the renal tubular epithelium. In the PBS/TW and rH120/TW groups, similar tracheal lesions were observed, as were pulmonary hemorrhage and inflammation. The PBS/TW group also presented renal damage with glomerular atrophy and necrosis, whereas the kidneys of the rH120/TW group appeared unaffected. In the PBS/GVI and rH120/GVI groups, tracheal cilia shedding, mucosal epithelial cell thickening, and inflammatory cell infiltration were noted, along with mild pulmonary hemorrhage and inflammation, but no significant renal lesions were detected. In stark contrast, the rH-CPDF7-QX(S)/QX, rH-CPDF7-TW(S)/TW, rH-CPDF7-GVI(S)/GVI, and NC groups were free from significant pathological changes in the trachea, lungs, and kidneys (Fig. 6I).

In conclusion, the recombinant strains rH-CPDF7-QX(S), rH-CPDF7-TW(S), and rH-CPDF7-GVI(S) effectively triggered robust humoral and cellular immune responses in chickens. Notably, these strains produced higher levels of neutralizing antibodies against homologous virulent strains than did the rH120 strain. Vaccination with the rH-CPDF7-QX(S), rH-CPDF7-TW(S), and rH-CPDF7-GVI(S) strains effectively mitigated clinical symptoms, reduced mortality, and decreased viral loads and shedding in chickens, thereby protecting against homologous strains. In summary, vaccine candidate strains rH-CPDF7-QX(S), rH-CPDF7-TW(S), and rH-CPDF7-GVI(S) elicited robust immune responses and protective effects, suggesting their potential as effective vaccines against IBV variants.