Regional adaptations and parallel mutations in Feline panleukopenia virus strains from China revealed by nearly-full length genome analysis

Protoparvoviruses, widespread among cats and wild animals, are responsible for leukopenia. Feline panleukopenia virus (FPLV) in domestic cats is genetically diverse and some strains may differ from those used for vaccination. The presence of FPLV in two domestic cats from Hebei Province in China was identified by polymerase chain reaction. Samples from these animals were used to isolate FPLV strains in CRFK cells for genome sequencing. Phylogenetic analysis was performed to compare our isolates with available sequences of FPLV, mink parvovirus (MEV) and canine parvovirus (CPV). The isolated strains were closely related to strains of FPLV/MEV isolated in the 1960s. Our analysis also revealed that the evolutionary history of FPLV and MEV is characterized by local adaptations in the Vp2 gene. Thus, it is likely that new FPLV strains are emerging to evade the anti-FPLV immune response.

Major comments to the authors. 1. The authors compared the isolated two FPLV strains with historical strains. But few information was introduced about these historical strains. More background information about historical strains should be included. Resp: Historical samples are those isolated in cats prior the spread of the FPLV to the dogs which occurred before 1980s (molecular clock estimates that the host transmission occurred between [1970][1971][1972][1973][1974][1975][1976][1977][1978]. After 1980 the virus adapted to dogs and was named CPV-2 and soon a new variant (named CPV-2a) replaced CPV-2. In the original manuscript a brief discussion about these strains was included (Lines 316-320) we also included some explanation in the lines 22-224.
2. In the Ancestral reconstruction section, P13, line 271-308, the authors claimed "the conflicting location of LSJ-2014, CSJ-2015 in the maximum likelihood trees", please explain more about this conflicting location of LSJ-2014, CSJ-2015 in the maximum likelihood trees.VP1 only contains 59 more amino acids (177nt) than VP2. Resp: Since phylogenetic trees can rotate vertically and this has no meaning to the topology then Vp1, Vp2 and Ns2 trees are not distinct in their topology because clade #1, #2 and #3 are present in all trees. Please note that in in all trees clades #2 and #3 are related (they share a common ancestor node). Also note that strains LSJ-2014 and CSL-2015 are closely related with the clade #3 in the Ns1 and Vp1 tree while in the Vp2 tree these strains are closely related with the clade #2.
Why the phylogenetic tree of VP1 and VP2 is so different (conflicting)? the authors showed some mutations in the roots of #1, #2, #3, #4, and #5 in figure 6, does that mean all viruses in this branch contain the same mutations?
Resp: No necessarily all strains in the leafs of a tree share the same residue that was inferred in a certain node because the ancestral reconstruction relies on phylogenies and it is based on a codon model that follows an evolutionary approach. So the probability of a residue in a hypothetical ancestral protein depends not only on the frequency of all amino acids in the alignment, it also depends on the likelihood of having that residue at a certain internal node of the inferred tree.
3. The authors claimed that the isolated FPLV may emerged as evasion strains due to selective pressures of immune surveillance. Do you have any information about the immune situation of the cats which the two FPLV isolated from? Resp: Besides depression, fever, intense haemorrhagic vomiting, diarrhea, and severe leukopenia no other information was available Minor comments to the authors. 1. P8, line 47, when you introduce amino acids, it is better use VP2 protein but not Vp2 gene. Resp: we did change in the manuscript 2. P8, line 49-50, typos, "presente", "posiions" Resp: we did change in the manuscript 3. P9, line 92, not sure why include (Table S1) here. Resp: "Table S1" was removed from this part of the text 4. P9, line 112-115, the buffer is PBST but not PBS? What is the primary and secondary antibodies used for IFA? What fluorescent dye was conjugated in the second antibody? Resp: PBST is PBS plus Tween used as standard washing buffers. In immunolabeling assays primary antibodies bind directly to the antigens and secondary antobodies bind to the Fc domains of primary antibodies. The Fc domain is constant within the same animal class. The Fc domain of secondary antibodies are labeled (usually with FITC) only one type of secondary antibody is required to bind to many types of primary antibodies, this reduces the cost by labeling. Typo, "aded" in line 113. "wells" in line 114 is better use "cells". Resp: This was changed in the manuscript.