Mechanism of Ca²⁺ transport by ferroportin

Major concerns:

You have been able to robustly demonstrate Ca²⁺ binding. What is unclear is if Ca²⁺ is actually transported to a significant level? Is the Fpn protein just capable of one turnover in proteoliposomes? We highly recommend kinetics using 45CaCl2 (eg. PNAS E5354-E5362) to calculate actual Ca²⁺ turnover (kcat).

Uptake of Ca²⁺ is measured by changes of Fluo-4 fluorescence. The fluorescence intensity increases progressively as we increase the concentrations of the Ca²⁺ and reaches up to five-fold higher than the initial value. As shown in Author response 1, a five-fold increase of fluorescence intensity requires >20 µM of Ca²⁺. If we assume that a lipid headgroup has an area of ~60 Å2 and that liposomes have an average diameter of ~400 nm, we arrive at a rough estimate of >100 turnovers for each Fpn under 500 µM external Ca²⁺ as shown in Figure 1—figure supplement 4a.

Author response image 1

Download asset Open asset

Reviewer #1 (Recommendations for the authors):

The authors have been able to robustly demonstrate Ca²⁺ binding. What is unclear to me is how much Ca²⁺ is actually transported? Is the Fpn protein just capable of one turnover in proteoliposomes? The "cleanest" experiment would be to measure Ca²⁺ uptake using 45CaCl2 (eg. PNAS E5354-E5362) so one can calculate Vmax and kcat. From my understanding, metalloproteins do not necessarily use the optimum metal for catalysis, but the metal available in the appropriate concentration in where they function. I wonder if the Ca²⁺ site is just not selected against, because its not needed too, i.e., Ca²⁺ doesn't compete for Fe2+ uptake and the actual "turnover" for Ca²⁺ is very low?

See response to Major concerns.

Reviewer #2 (Recommendations for the authors):

The data reported here appear to be of high quality and are convincing; the experimental design is excellent and the necessary controls are appropriately employed. A couple of issues need clarification in the text. FPN is clearly an iron efflux pump and these studies make clear that Fpn can also import Ca²⁺, although it does not appear to function as an Fe2+-Ca²⁺ antiporter. What is less clear is whether Fpn will transport calcium bi-directionally. There are technical issues that would make this a bit hard to study, but if the authors think the answer is clear from their data, they should be explicit in explaining this. A further question that needs explaining is why bind and transport calcium? Cells have a high capacity for calcium flux independent of Fpn. Is there a physiological importance to this activity?

We added an explanation that the Fpn in liposomes is randomly oriented and made a schematic drawing (Figure 1—figure supplement 1c) to illustrate this point. We also measured Ca²⁺ transport when the Fab was added only to the external side, and we found that the rate of transport is higher than the rate when the Fab was added to both sides and lower than the rate when no Fab was added (Figure 1—figure supplement 1f–g). We take this result as further indication that Fpn is reconstituted in both orientations.

We are also intrigued by the calcium transport activity of Fpn, and more study is required to address the questions on biology Ca²⁺ uptake by Fpn. We speculate that loss of cellular Fe²⁺ will lead to Ca²⁺ influx through Fpn, which may trigger further cellular responses.

Although our study arrived at a different conclusion from that of Deshpande et al. 2018 with regard to Ca²⁺ transport, both studies share common grounds on the calcium binding site and transport of Fe²⁺ in the presence of Ca²⁺. We made a comment on this in Discussion.