A risk marker of tribasic hemagglutinin cleavage site in influenza A (H9N2) virus

Low pathogenic avian influenza A(H9N2) virus is endemic worldwide and continually recruit internal genes to generate human-infecting H5N1, H5N6, H7N9, and H10N8 influenza variants. Here we show that hemagglutinin cleavage sites (HACS) of H9N2 viruses tended to mutate towards hydrophilic via evolutionary transition, and the tribasic HACS were found at high prevalence in Asia and the Middle East. Our finding suggested that the tribasic H9N2 viruses increased the viral replication, stability, pathogenicity and transmission in chickens and the virulence of mice compared to the monobasic H9N2 viruses. Notably, the enlarged stem-loop structures of HACS in the RNA region were found in the increasing tribasic H9N2 viruses. The enlarged HACS RNA secondary structures of H9N2 viruses did not influence the viral replication but accelerated the frequency of nucleotide insertion in HACS. With the prevailing tendency of the tribasic H9N2 viruses, the tribasic HACS in H9N2 viruses should be paid more attention.

affect the cleavability of HA0, resulting in the increased viral titers. In other words, the difference in growth titers might be simply due to the difference in susceptibility of different HACS to host proteases. Thus, it is interesting to see the growth ability of the tribasic H9N2 virus in the absence of trypsin.
2. Figures 3d and 3e. The authors determined the effect of trypsin on the HA0 cleavage and concluded that there was no significant difference in the cleavability between rV08-PSRSSR and rV08-PAKSKR viruses. However, it should be important to see the cleavability at different concentrations of trypsin. Additional experiments are needed for this conclusion.
3. Figure 5. In this reporter assay, only luciferase activity was shown. It would also be important to show the sequence changes (i.e., actual nucleotide insertions) of the expressed RNAs in the cells or virions. Although the authors mention, "using Sanger sequencing, the single-and double-nucleotide insertions or double-nucleotide deletion occurred in the synthetize of mRNA or vRNA in the enlarged stem-loop structure of HACS ( Fig. S5)", detailed methods of this sequencing analysis are missing. Explanation of Fig. S5 is also unclear. For example, what do the numbers in the parenthesis represent (e.g., 10/10)? How many percent of the viral RNAs contained the insertion(s)? 4. In Figure 5, the authors showed the stem-loop structure with 27nt sequences. However, it is not appropriate to generate secondary structures using just one type of software. The authors should use several different types of software to confirm the prediction of the RNA structures. In addition, the analyses using longer sequences (e.g., 40 and 50nt) should also be carried out.
Minor comments.
1. Some terms repeatedly used in the text, such as monobasic and tribasic, should be explained briefly. 6. Figure 1b-e. Additionally, the authors could insert titles such as "China", "All except China" so that the readers can read the figures easily. 7. Figure 2 legend symbol. These motif sequences should be consistent throughout the manuscript. For example, replace "ATSGR/GLF" with "PASTGR/G", etc., in Figure 2.
8. Figure 3c. The procedure for determining the thermal stability of the viruses is not clear.
According to the figure legend, the viruses were inoculated for 1, 2, 3, 4, and 5 h at 50°C.
11. The title of Table S1. "Virus titers in chicken cloacal and oropharyngeal swabs" should be changed. Actually, this table does not show virus titers, it only shows infection rate, i.e., infected/total chickens. 12. The order of supplementary figures is strange in the text. Figure S1 is cited on line 312 after Figure S2-S6. It should be re-arranged.
13. Figures S2 and S3. How did the authors select these low pathogenic H5 and H7 strains?
Are these limited to low pathogenic avian influenza viruses isolated from ducks or chickens?
16. There are many grammatical errors throughout the manuscript.

Reviewer #2 (Remarks to the Author):
Review report A novel marker of triabasic hemagglutinin cleavage site in influenza A (H9N2) virus Influenza A viruses (subtype H9N2) are prevalent in poultry in many countries and there are reports of occasional transmission to humans which is indicative of their pandemic potential.
Cleavage of the surface glycoprotein hemagglutinin (HA) of H9N2 by host cell proteases at a distinct site is the central to H9N2 infections.
The authors analysed variations observed in the hemagglutinin cleavage site (HACS) of H9N2 using an array of computational & experimental approaches to evaluate replicate advantages, if any. The authors report that tribasic HACS in H9N2 are evolving to be hydrophilic and therefore provide distinct advantage to the virus for replication, stability, pathogenicity as well as transmission. The authors have made a case for how enlarged structure of RNA facilitate insertion without having any impact replication. It is definitely a well-structured study reported as a concise story.
I recommend that the authors come up with an info graphic summary this work to pictorially summarise various cleavage sites along side of assessment criteria (pH, temperature etc) summarising advantages/limitations. It would certainly add value to the manuscript. They found that the tribasic H9N2 virus (i.e., a virus with the PAKSKR/G motif) had higher replication capacity in vitro and in vivo. Higher pH stability and thermostability were also observed in the tribasic H9N2 virus. In the tribasic H9N2 virus but not in the other viruses tested, the stem-loop structure of the RNA region that encodes the tribasic motif (PAKSKR), had an enlarged loop which has been shown to be important for nucleotide insertions at this region. This manuscript has interesting data and would provide important insights into the field. However, there are several weaknesses in its current state. Following are points that may improve the manuscript.
Major comments. Figures 3a and 3b. The growth curve experiments were conducted in the presence of trypsin. It could be assumed that the addition of basic amino acid residues at HACS might affect the cleavability of HA0, resulting in the increased viral titers. In other words, the difference in growth titers might be simply due to the difference in susceptibility of different HACS to host proteases. Thus, it is interesting to see the growth ability of the tribasic H9N2 virus in the absence of trypsin.

1.
Response: Thank the reviewers for their constructive comments and valuable recommendations. We have assessed the growth ability of the dibasic H9N2 viruses (PARSSR/G-motif) and tribasic H9N2 viruses (PAKSKR/G-motif) in the absence of trypsin; however, we found that H9N2 viruses could not be efficiently cleaved and replicated. For low pathogenic H9N2 viruses, the HA0 precursor must be cleaved into the HA1 and HA2 subunits with the help of host protease, thereby revealing a fusion peptide for membrane fusion and replicate in the cells. In addition, many other studies proved that low pathogenic H9N2 viruses could not cleave and replicate in the avian and mammalian cells in the absence of the host protease.  Figure 5, the authors showed the stem-loop structure with 27nt sequences.

In
However, it is not appropriate to generate secondary structures using just one type of software. The authors should use several different types of software to confirm the prediction of the RNA structures. In addition, the analyses using longer sequences (e.g., 40 and 50nt) should also be carried out.
Response: Authors agree with the reviewer. We use several different types of software (RNAfold, mfold, and RNAstructure) and long RNA sequences (40 nt) to confirm the prediction of the RNA structures; please see supplementary Figures 5, 6, and 7. Minor comments.
1. Some terms repeatedly used in the text, such as monobasic and tribasic, should be explained briefly.  8. Figure 3c. The procedure for determining the thermal stability of the viruses is not clear. According to the figure legend, the viruses were inoculated for 1, 2, 3, 4, and 5 h at 50°C. Is this true? Should this be "incubated for 1, 2, 3, 4, and 5 h at 50°C"? If so, the x axis needs to be corrected.

Response
Response: Authors agree with the reviewer. The word "inoculated" have been changed to "incubated"; please see page 48 line 755 and Figure 3c.
Response: Authors agree with the reviewer. We have corrected "RAKSKR" to "PAKSKR"; please see Figure 7.
11. The title of Table S1. "Virus titers in chicken cloacal and oropharyngeal swabs" should be changed. Actually, this table does not show virus titers, it only shows infection rate, i.e., infected/total chickens.

Response:
Authors agree with the reviewer. We have changed the titles to "The infection rate in chicken cloacal and oropharyngeal swabs"; please see Table S1.
12. The order of supplementary figures is strange in the text. Figure S1 is cited on line 312 after Figure S2-S6. It should be re-arranged.
Response: Authors agree with the reviewer. We have re-arranged the orders of supplementary figures.
13. Figures S2 and S3. How did the authors select these low pathogenic H5 and H7 strains? Are these limited to low pathogenic avian influenza viruses isolated from ducks or chickens?
Response: Authors agree with the reviewer. The HACS sequences of low pathogenic H5, H7, and H9 precursors were all avian species (chickens, ducks, and wild birds) available in the GISAID database (http://www.gisaid.org/). Identical HACS sequences were removed. We have described more detail in the Methods section; please see page 37 lines 578-580.