Evolution and activation mechanism of the flavivirus class II membrane-fusion machinery

The flavivirus envelope glycoproteins prM and E drive the assembly of icosahedral, spiky immature particles that bud across the membrane of the endoplasmic reticulum. Maturation into infectious virions in the trans-Golgi network involves an acid-pH-driven rearrangement into smooth particles made of (prM/E)2 dimers exposing a furin site for prM cleavage into “pr” and “M”. Here we show that the prM “pr” moiety derives from an HSP40 cellular chaperonin. Furthermore, the X-ray structure of the tick-borne encephalitis virus (pr/E)2 dimer at acidic pH reveals the E 150-loop as a hinged-lid that opens at low pH to expose a positively-charged pr-binding pocket at the E dimer interface, inducing (prM/E)2 dimer formation to generate smooth particles in the Golgi. Furin cleavage is followed by lid-closure upon deprotonation in the neutral-pH extracellular environment, expelling pr while the 150-loop takes the relay in fusion loop protection, thus revealing the elusive flavivirus mechanism of fusion activation.


Reviewers' Comments:
Reviewer #1: Remarks to the Author: The manuscript by Vaney and colleagues sheds new and unique light on the role of the flavivirus prM protein and its protective role covering the envelope (E) protein fusion loop. The authors have examined the prM and E proteins of tick-borne encephalitis virus at acidic pH using X-ray crystallography. The structure reveals that there is a loop (150-loop) in the E protein that moves under differing pH conditions. Low pH favors the binding of the prM protein into a positively charged pocket at the E protein dimer interface. Following furin cleavage of prM and exposure to a neutral pH environment, pr is expelled and the 150-loop moves in to protect the fusion loop. The authors also show that pr moiety of prM is related to the HSP40 cellular chaperonin and has binding features that are similar. This is a significant and original piece of research that is very well described. The findings are relevant across the flavivirus genus.
Minor comments:  Page 9; second paragraph: The authors make the argument that the 150-loop plays a similar role in not only tick-borne flaviviruses but also the mosquito-borne flaviviruses. However, they do not discuss the observations that this loop has been shown to be dynamic with some flaviviruses having much greater movement (access) than others (for example, dengue virus versus Zika virus). This should be addressed. Again, on page 12 of the Discussion, they mention that "the 150-loop lid then snaps firmly into place" suggesting there is no further movement/dynamics. The data suggests that this is not true although one can understand that there is displacement of pr by the movement of the 150-loop.
Page 11; line 16: Can the authors explain the linkage between (prM/E)2 dimers that they are describing and the formation of the herringbone organization of E dimers?
Reviewer #2: Remarks to the Author: In this manuscript, the authors have identified, reconstituted, and determined the X-ray crystal structure of Tick-borne Encephalitis Virus envelope protein E in complex with pr protein. They found that the authentic N-terminal end of E is important in ordering the 150-loop that mediates a pHdependent shedding of the pr protein upon viral egress. They show that slightly acidic pH is necessary for 2-to-2 complex formation and that both charge repulsion and loop rearrangement modulate sequential maturation steps. The authors also describe the structural homology of pr with its pre-viral distant relative chaperonin protein.
Major comments: In all, this is a very nicely written and performed study with impressive mechanistic detail.
Minor comments: Page 4, line 1: dissociation implies release, whereas in this instance subunits in the trimer are handing-off to neighboring subunits in a conformational change. They don't really come apart.
Page 5, line 7: "did not permit to conclude" has a grammar error.
Page 5, lines 12-18: There is a distinct drop in both the k<on> and k<off> rates with increasing pH. The right-ward shift of the curves in Fig. 1E from pH 5.0 to 6.5 suggests modest conformational or charge changes to the affinity and yet they reach a common saturation level. Above the protonation pH of histidine, the k<on> is drastically reduced and therefore saturates at a much lower level, indicating the binding surface, or indeed the E dimer, might not exist in those conditions. This is glossed over in the text.
Page 4-5 "The SEC profile of the sE monomer is aberrant, eluting at a large volume corresponding to small molecules…" This statement is somewhat awkward and may only be understood by individuals who perform SEC regularly. Consider rewording.

Reviewer #1 (Remarks to the Author):
The manuscript by Vaney and colleagues sheds new and unique light on the role of the flavivirus prM protein and its protective role covering the envelope (E) protein fusion loop. The authors have examined the prM and E proteins of tick-borne encephalitis virus at acidic pH using X-ray crystallography. The structure reveals that there is a loop (150-loop) in the E protein that moves under differing pH conditions. Low pH favors the binding of the prM protein into a positively charged pocket at the E protein dimer interface. Following furin cleavage of prM and exposure to a neutral pH environment, pr is expelled and the 150-loop moves in to protect the fusion loop. The authors also show that pr moiety of prM is related to the HSP40 cellular chaperonin and has binding features that are similar. This is a significant and original piece of research that is very well described. The findings are relevant across the flavivirus genus.
We thank the reviewer for recognizing the originality of our research, and for her/his favorable assessment of our manuscript.
Minor comments:

Figure 1: provide the molar ratios for pr and sE in panel A.
The pr:sE stoichiometry was 2:1 monomer:monomer. This was added to the legend of Figure 1 (line 596)

2.Figure 2a: In the top view, the Calpha of pr is difficult to see. Could the authors improve the contrast or change transparency of the E protein surface representation?
In the revised version, we enhanced the color contrast in the cartoon representation of pr in Fig. 2a to make it stand out better.

Page 9; top paragraph: Can the authors discuss what role (if any) the glycan present on the 150-loop plays in the open lid versus closed lid conformations? Have they removed that glycan to determine whether the loop continues to function as predicted?
We have not removed the glycan to determine if the loop continues to function as predicted, and it would be beyond the scope of this manuscript to test this. Our collaborators in the Screaton lab in Oxford have shown that dengue viruses are viable when knocking out the corresponding glycan in the 150 loop (Personal communication). There are also data in the literature showing that knocking out the glycan in DENV2 results in a virus with a higher pH threshold for membrane fusion (0.65 pH units, https://doi.org/10. 1006/viro.1993.1252), indicating that the glycan does stabilize the pre-fusion dimer. The fact that we see an ordered glycan when the lid is in the "down" position indicates that the interprotomer interactions it displays are responsible for its ordering, as it is totally disordered when the lid is up. This is another indirect observation. The lack of carbohydrate attached to the 150 loop in several circulating flavivirus strains might be used as an indirect evidence for the function of the lid as not strictly depending on the attached glycan. We have modified the text to incorporate the above comments (see lines 262-266): "Our finding that the carbohydrate attached to Asn154 is ordered in the closed hinge and disordered in the open lid form implies both entropic and enthalpic contributions of the glycan in each state, suggesting that it plays a role in the interaction. Indeed, knocking out the glycan in DENV2 resulted in a virus with a higher pH threshold 34 . Yet, non-glycosylated variants of West Nile, Zika, and YF viruses are infectious and circulate in nature. These observations suggest that the lid is functional independent of its glycosylation status, which may however affect vector transmission, virulence and pathogenicity 35 ." 4.Page 9; second paragraph: The authors make the argument that the 150-loop plays a similar role in not only tick-borne flaviviruses but also the mosquito-borne flaviviruses. However, they do not discuss the observations that this loop has been shown to be dynamic with some flaviviruses having much greater movement (access) than others (for example, dengue virus versus Zika virus). This should be addressed. Again, on page 12 of the Discussion, they mention that "the 150-loop lid then snaps firmly into place" suggesting there is no further movement/dynamics. The data suggests that this is not true although one can understand that there is displacement of pr by the movement of the 150-loop. The reviewer appears to refer to the dynamic "breathing" of flavivirus particles, in which the buried fusion loop is transiently exposed and can be accessible to antibodies targeting the fusion loop. Such breathing will also affect the interaction of the 150 loop with the fusion loop, since it involves E dimer dissociation. But to comply with the reviewer's comments, we have eliminated the qualifier "firmly" in the sentence above, to say only "snaps into place" (line 277), as it is true that it might be more or less "firmly" depending on the flavivirus. We now discuss the observation that the extent of breathing can differ among flaviviruses, possibly reflecting differences in the strength of the snap-lock. We have modified the text at lines 280-282 to reflect our response to the reviewer. We write there: "The strength of the snap-lock may vary among flaviviruses, as suggested by differences in the phenomenon of breathing and the transient exposure of the FL in mature virions 36, 37 ."

Page 11; line 16: Can the authors explain the linkage between (prM/E)2 dimers that they are describing and the formation of the herringbone organization of E dimers?
We have shown for Zika virus and for dengue virus 2 (DENV2) that the E dimers have a natural affinity to pack laterally as in the rafts in the herringbone pattern. In the crystals of sE, sE dimers make extended rows of dimers packing in this way