Riemerella anatipestifer (ATCC11845) was purchased from the China Veterinary Drug Administration. The clinical strains were isolated from the diseased materials of various duck farms in Guangzhou and were identified and preserved by the Veterinary Pharmacology Laboratory of South China Agricultural University.

A small amount of bacterial solution was dipped in the glycerin meat solution suspension using an inoculation ring, scribed, and inoculated on the tryptone soy agar (TSA) [Qingdao Hope Bio-technology Co., Ltd.] plate containing 5% fetal bovine serum and placed on the incubator containing 5% CO₂ at 37°C for 24 h in sterile conditions. A single colony can be used for further tests. The single colony was isolated and inoculated in Tryptic-Soytone-Broth-Medium (TSB) [Qingdao Hope Bio-technology Co., Ltd.] medium containing 5% fetal bovine serum and cultured for 10 h to complete the logarithmic period for the test.

The MIC of doxycycline and chlorogenic acid against R. anatipestifer was determined by broth microdilution method according to the regulations of the (20). The drug stock solution was thawed at room temperature and thawed using 0.22 μm sterile membrane sterilization, diluted to the appropriate concentration with blank TSB medium. In a 96-hole plate, 100 μL of blank medium was added into 1–9 holes, and then 100 μL of diluted drug was added in the first hole. After mixing the blank medium and the diluted drug with a multi-channel pipette in the first hole, take 100 μL from the first hole add into the second hole,dilute to the ninth hole by double dilution, and discard 100 μL in the ninth hole. The final concentration range of doxycycline was 0.0625–16 μg/mL. The final concentration range of chlorogenic acid was 0.25–64 mg/mL. Then the bacterial suspension in the logarithmic phase was taken, diluted to 10⁶ CFU/mL, and 100 μL of bacterial solution was added into 1–9 wells, respectively, so that the final bacterial concentration is 5 × 10⁵ CFU/mL. Now, 100 μL of bacterial liquid and 100 μL of blank broth were added into hole 11, which acts as a positive control. Then, 200 μL of blank broth was added into hole 12, which was used as a negative control. The plate was put into the incubator containing 5% CO₂ at 37°C for 18 h and then taken out to observe the results. The minimum drug concentration that can inhibit the growth of bacteria as observed by the naked eye is recorded as the MIC of doxycycline or chlorogenic acid on R. anatipestifer. Three parallels were set, and the operation was repeated three times.

Riemerella anatipestifer was induced continuously with doxycycline of different concentrations. A total of 1–3 passages were co-cultured with doxycycline of 1/2MIC concentration, 4–5 passages were co-cultured with doxycycline of 1MIC concentration, 6–8 passages were co-cultured with doxycycline of 2MIC concentration, and 9–10 passages were co-cultured with doxycycline of 4MIC concentration. The single induction time was 12 h.

The standard strain (ATCC11845) and the induced strain were scraped from the plate into the TSB liquid medium, and the OD was adjusted to 1 and cultured in the TSB liquid medium for 2 h, and the bacteria were collected. The total RNA was extracted using the bacterial total RNA extraction kit [Tiangen Biochemical Technology (Beijing) Co., Ltd., China], and the RNA purity was determined using NanoDrop 2000, and then sent to Lianchuan Biological Company for quality inspection. After passing the quality inspection, the transcriptome was detected. The raw data of transcriptome sequencing were filtered to obtain high-quality data information; DESeq v1.20.0 software was used to study the differentially expressed genes among the three groups. The conditions were set as the difference multiple |log₂FC| > 1, with a significance p-value of <0.05. The mRNA of the three samples was screened for differentially expressed genes.

In order to verify the reliability of transcriptome data, the RNA of the standard strain (ATCC11845) and induced strain were extracted according to the same transcriptome sampling method (see 2.5). The RNA samples were subjected to a 20 μL system that was reverse transcribed into cDNA. The tetracycline-resistance genes tet (X), tet (O), and tet (A), which were characterized by standard strain (ATCC11845) and induced strain, were detected by RT-PCR.

Using the primer sets tet (X), tet (O), and tet (A) listed in Table 1, the tet (X), tet (O), and tet (A) genes were amplified from the R. anatipestifer induced strain by PCR. The purified amplicons were sequenced and digested with restriction enzymes and then connected to the same digestion shuttle vector pLMF03 [Ordered by Miaoling Biotechnology Co., Ltd.]. Then the corrected recombinant plasmid was introduced into E. coli S17-1 [Miaoling Biotechnology Co., Ltd.]. These sequences were analyzed by BLAST in NCBI.

As mentioned above, the correct recombinant plasmid was transferred to R. anatipestifer (ATCC 11845) by coupling transfer (21, 22). In short, Escherichia coli S17-1 containing the recombinant plasmid was used as the donor strain, while the reference strain ATCC 11845, which does not carry any tet gene and is sensitive to tetracycline, was used as the recipient strain (23). Meanwhile, the negative control empty vector pLMF03 was transferred to ATCC 11845 to obtain the transconjugant ATCC 11845 [pLMF03]. The MIC of doxycycline of the transconjugant was measured as described above.

The induced strain, R. anatipestifer, was continuously cultured with 1/2MIC chlorogenic acid for 12 h. After each generation of culture, the MIC of doxycycline was detected.

The induced strain R. anatipestifer was continuously cultured with 1/2MIC chlorogenic acid for 12 h. After each generation of culture, the transcription level of the tetracycline-resistance gene tet (X) was detected.

The sheldrakes utilized in this experiment were procured from the Guangxi Guiliu Poultry Nanning Hatching Branch, and the experiment has received approval from the Ethics Committee of Experimental Animals at South China Agricultural University (No. 2022A022). In total, 60 3-day-old male sheldrakes were randomly allocated into three groups, with 20 ducks per group. Following 5 days of acclimation feeding, each duck was intraperitoneally administered with 1 mL of solution containing 10⁹ CFU/mL of R. anatipestifer (induced resistant R. anatipestifer in 2.4). The doxycycline group received a dose of 20 mg/kg doxycycline via intramuscular injection to the thigh region, 12 h post-infection; the chlorogenic acid group was orally administered with a dose of 8 g/kg of chlorogenic acid by gavage, also 12 h after infection, while the control group received an equivalent volume of dilution solution (glyceraldehyde:normal saline =40:60, V/V) through intramuscular injection to the thigh region, 12 h post-infection. The drug administration occurred every 12 h for 10 consecutive doses.

All sheldrakes in the above groups were euthanized after 24 h of drug administration, and 4 ducks were randomly euthanized every 24 h. After euthanizing the ducks, the heart tissue was collected aseptically, and 1 g of the heart tissue was accurately weighed and added to 1 mL of normal saline for homogenization using a handheld homogenizer. The heart tissue was diluted 10 times to an appropriate gradient, and 20 μL of each series of diluent was dropped on the TSA plate containing 5% bovine serum. The strains were incubated at 37°C in a 5% CO₂ incubator for 20–24 h, and a single colony was selected for culture. The MIC of doxycycline and the expression of the tet (X) gene of R. anatipestifer were detected after different administration times.

GraphPad prism 9.4.1 software was used for data processing and plotting, and a t-test was used for statistical analysis. When the analysis result was a p-value of <0.05, there was a statistical difference.

The MIC of doxycycline and chlorogenic acid against R. anatipestifer was determined by the broth microdilution method according to the regulations of the CLSI (2013). The results showed that the MIC of doxycycline in the standard strain (ATCC11845) was 0.25 μg/mL. The MIC of chlorogenic acid in the standard strain (ATCC11845) was 4 mg/mL. The MIC results of clinical isolates are shown in Table 2.

After 10 generations of induction, we obtained a new drug-resistant strain, and the MIC of doxycycline to the induced drug-resistant strain increased from 0.25 μg/mL to 2 μg/mL. It is surprising to note that the MIC of chlorogenic acid to the induced drug-resistant strains did not change and was still 4 mg/mL.

In the established sequencing library, the proportion of base quality exceeding Q20 in the three groups of samples was more than 94.20%, and the proportion exceeding Q30 was more than 94.14% (Table 3), indicating that the sequencing results were good.

The gene data obtained by RNA SEQ were analyzed for differential expression using DESeq v1.20.0 software and screened under the conditions of |log₂FC| > 1, p < 0.05. The results showed that compared with the standard strain, there were 341 differentially expressed genes in the induced drug-resistant strain, of which 247 were upregulated and 154 were downregulated (Figure 1).

The results of RT-PCR showed that the content of drug-resistance genes of antibiotic-induced resistant strains was significantly higher than that of standard strains. The content of drug-resistance genes of clinical isolates was also significantly higher than that of standard strains. With the increase in MIC value, the content of drug-resistance genes also increased. The results are shown in Figure 2. Therefore, it can be preliminarily concluded that the resistance of R. anatipestifer to tetracycline antibiotics may be related to tet (A), tet (O), and tet (X) genes.

E. coli S17-1 pLMF03:: tet (X), E. coli S17-1 pLMF03:: tet (A), and E. coli S17-1 pLMF03:: tet (O) were used as the donor bacteria. RA ATCC 11845 was used as the recipient bacteria, the donor bacteria was mixed with the recipient bacteria at 1:4, and coated on the filter membrane. After being cultured on the blood plate, the bacteria were washed with magnesium chloride and screened on the blood plate containing kanamycin and cefoxitin resistance. The results as shown in Figure 3, tet (X), tet (A), and tet (O) fragments could be amplified by PCR, indicating that R. anatipestifer ATCC 11845 pLMF03:: tet (X), R. anatipestifer ATCC 11845 pLMF03:: tet (A), and R. anatipestifer ATCC 11845 pLMF03:: tet (O) transgenic strains were successfully constructed. The MIC of doxycycline on R. anatipestifer ATCC 11845 pLMF03:: tet (X), R. anatipestifer ATCC 11845 pLMF03:: tet (A), and R. anatipestifer ATCC 11845 pLMF03:: tet (O) transgenic strain was determined by the broth microdilution method. The results showed that the MIC of doxycycline on R. anatipestifer ATCC 11845 pLMF03:: tet (X) transgenic strain increased to 4 μg/mL, the MIC of doxycycline on R. anatipestifer ATCC 11845 pLMF03:: tet (A) transgenic strain increased to 1 μg/mL, and the MIC of doxycycline on R. anatipestifer ATCC 11845 pLMF03:: tet (O) transgenic strain increased to 2 μg/mL. However, the MIC of the negative control empty vector group was still 0.25 μg/mL. The results are shown in Figure 3 and Table 4.

The chlorogenic acid concentration of 1/2MIC was selected to act on the induced drug-resistant strains. After continuous passage, the MIC of doxycycline was detected again. We were surprised to find that the sensitivity of R. anatipestifer to doxycycline was improved after the selective passage of chlorogenic acid. Specific MIC values are shown in Table 5.

The expression of R. anatipestifer-resistance genes is shown in Figure 4. It can be seen from the above results that the expression trend of three differential genes, namely tet (A), tet (O), and tet (X) genes, in the induced drug-resistance group was reversed after the chlorogenic acid intervention, and the downregulation level of tet (X) gene was found to be the most significant.

The results demonstrated that the MIC of the control group, the doxycycline group, and the chlorogenic acid group remained at 2 μg/mL after 24 h, 48 h, and 72 h of administration. At 96 h and 120 h post-administration, the MIC of the control and doxycycline groups remained at 2 μg/mL, while the chlorogenic acid group exhibited a reduced MIC of 1 μg/mL. Figure 5 illustrates the expression of the tet (X) gene. It is evident that there was minimal variation in the tet (X) gene expression within the control group across different time points. In contrast, a slight increase in the tet (X) gene expression was observed in the doxycycline group with prolonged drug administration time. Conversely, a decrease in the tet (X) gene expression was noted in the chlorogenic acid group as drug administration time increased.