The tyrosine kinase KDR is essential for the survival of HTLV-1-infected T cells by stabilizing the Tax oncoprotein

Human T-cell leukemia virus type 1 (HTLV-1) infection is linked to the development of adult T-cell leukemia/lymphoma (ATLL) and the neuroinflammatory disease, HTLV-1-associated myelopathy/tropical spastic paraparesis (HAM/TSP). The HTLV-1 Tax oncoprotein regulates viral gene expression and persistently activates NF-κB to maintain the viability of HTLV-1-infected T cells. Here, we utilize a kinome-wide shRNA screen to identify the tyrosine kinase KDR as an essential survival factor of HTLV-1-transformed cells. Inhibition of KDR specifically induces apoptosis of Tax expressing HTLV-1-transformed cell lines and CD4 + T cells from HAM/TSP patients. Furthermore, inhibition of KDR triggers the autophagic degradation of Tax resulting in impaired NF-κB activation and diminished viral transmission in co-culture assays. Tax induces the expression of KDR, forms a complex with KDR, and is phosphorylated by KDR. These findings suggest that Tax stability is dependent on KDR activity which could be exploited as a strategy to target Tax in HTLV-1-associated diseases.

We have addressed all the comments made by the reviewers as summarized below in the pointby-point responses.
Reviewer 1: 1. Despite the central role of KDR/VEGFR2 in the manuscript, the ligands of the VEGF family are not discussed at all.The authors mention VEGFR2 mutation in North-American ATL patients, do these mutations render the receptor inactive or do these mutations confer ligand-independent signaling?The authors rightfully state (line 88) "Vascular endothelial cell growth factors (VEGFs) critically regulate angiogenesis by binding to high-affinity receptor tyrosine kinases (RTKs), including VEGF receptors 1-3."However, a wealth of studies on angiogenesis and the role of VEGF in both HAM and ATL pathogenesis, as well as links to Tax have been published in HTLV infection and its associated pathologies (see references below).Can the authors link their compelling data on VEGFR2 to help resolve some of the conflicting data on VEGF/angiogenesis in the HTLV field?
We thank the reviewer for these important comments on the VEGF family in HTLV-1-associated diseases.In the revised manuscript, we have included a discussion (see lines 411-423) on the roles of VEGF in ATLL and HAM/TSP and how they could relate to our findings.We have also cited key references mentioned by the reviewer.Based on our data, we don't believe that VEGF will play a role in Tax stabilization since Tax sequesters KDR inside the cell in a perinuclear area where it is unable to interact with any ligands.The VEGFR2 mutations in North American patients have not been characterized but we speculate that the mutations may be gain-of-function that would confer ligand-independent signaling.

Ethical approval by the institutional Ethical
Committee for the use of patient cells seems to be missing in the manuscript or at least a waiver, e.g.use of leftover diagnostic samples.
We thank the reviewer for noticing the omission of this section in the manuscript.The revised manuscript has an Ethics Statement in the Methods section that describes the provision of the HAM/TSP blood samples.

3.
Can the authors explain why direct KDR interaction with Tax was not picked up in the several publications dedicated to the Tax interactome?Is the phenomenon cell type-specific?
The reviewer raises a good point.We have also performed a mass spectrometry screen to identify Tax interacting proteins (Gao et al. 2013 J. Virol.87: 13640-54) but did not identify KDR in this screen.We speculate that KDR was not previously identified as a Tax interacting protein because it is mainly associated with cell membranes and KDR is barely detectable unless membrane fractions are isolated.

Reviewer 2:
1.In Fig. 7 F. the figure is too small to read the specific description of the genes.This should be presented in a larger figure .Thank you for the comment.Fig. 7F has now been enlarged to make it easier for readers to read the protein names in the heat map.To make space for the enlarged figure, the previous Fig.7B and 7C have been moved to Supplementary Fig. 6. 9 A, authors showed effects of SU1498 on the PBMC from HAM patients and healthy controls.Authors described that HAM PBMC were cultured for 5 days before analysis.It is well known that HAM PBMCs will spontaneously proliferate in the culture without IL-2.However, it is not expected for the cases of healthy control PBMCs.Thus, additional description of the condition as to the control PBMCs is expected.

In Fig
The reviewers were both in agreement that we addressed the previous concerns and the manuscript is now acceptable for publication.
Reviewer 1: 1.The authors have addressed all pertinent questions and have adapted the manuscript accordingly.I agree with publication in its current form.
We are glad we have addressed all pertinent questions and thank the reviewer for recommending publication in its current form.
Reviewer 2: 1.The revised manuscript is considered to adequately address the reviewers' comments throughout and appears to have improved significantly.I believe that the information presented in the paper provides new insights into the interaction between the viral gene product and HTLV-1-infected cells.
We thank the reviewer for these comments.