What floats our boat Part I - Ferroxidases, Iron Permeases and APP




Iron is a trace element essential to all eukaryotes. In the trafficking of iron into fungi, like Baker's Yeast, Fe-uptake and cycling to and from the vacuole (to the left above, stained in red) requires a multi-copper oxidase; Fet3 localizes to the plasma membrane (stained in green) and there supports Fe-uptake. Iron uptake and vacuolar iron cycling in fungi is illustrated [to the right above, (A)]. Metazoa broadly express two ferroxidases, ceruloplasmin (Cp) and hephaestin (Hp) but these two enzymes are required not for iron uptake but for iron efflux from cells (B). While the kosmanlab often carries out experiments in yeast and on Fet3 given the flexibility of these systems, our major focus is on the role the mammalian ferroxidases Hp and Cp play in managing the delivery of iron to the brain across the blood-brain barrier (BBB) and between the cells of the neurovascular unit. This unit is shown to the right in the illustrations below of the brain's vasculature.
As illustrated above in (B), the key step in brain iron metabolism, whether into the brain via transcellular trafficking across the endothelial cells, or out of glial cells or neurons, is catalysis of iron efflux through ferroportin (Fpn) by a ferroxidase, either Hp or Cp. The level of Fpn in the membrane controls this iron efflux. Our work has demonstrated that sAPP increases the abundance of Fpn in the membrane as illustrated below. As a result, sAPP stimulates iron uptake into the brain. Our premise is that this represents a normal, physiologic sAPP function, one by which APP confers selective advantage. The fact that iron is associated with the amyloid plaques characteristic of late stage Alzheimer's disease appears to tie together both the physiologic and pathogenic outcomes of APP expression. Our goal is to deliniate the road-map of that connection.



To model the BBB, we use a transwell in which we plate capillary endothelial cells on a porous membrane on which they grow and form the tight-junction barrier characteristic of brain capillaries. An example of how we use this system to study how sAPP modulates iron efflux from this barrier into the "brain" is shown to the right. In cells that lack the ferroxidase, Hp, Fpn isn't present in the membrane. Fpn membrane localization and Fe-efflux is re-established upon addition of the mammalian ferroxidase, Cp, but not the fungal ferroxidase, Fet3.
sAPP stabilizes Fpn in the membrane but supports no Fe-efflux. However, by supporting Fpn in the membrane, sAPP along with a ferroxidase - Fet3 - re-establishes Fe-efflux. This experiment illustrates our use of various cell biology approaches as well as protein chemistry ones in that we express, purify and characterize the recombinant proteins that we use in many of our experiments.


"Designing" mammalian cells as vehicles for establishing protein function - which is the end point of most of our experiments - is not as easy as working in yeast or C. elegans. We use what works: sh- or si-RNA, CRISPR/Cas, transfection, infection and mice KOs. We maintain a colony of HEPHminus mice that provides us with primary endothelial cells, astrocytes and hippocampal neurons. For many experiments we use immortilized cells: HEK293 cells for simple tests of protein-protein interaction and function, a thoroughly validated human brain microvascular endothelial cell line for many of our transwell experiments, and retionic acid-differentiated SH5Y-SY "neurons." As an example of our use of HEK cells, Dr. Adrienne Dlouhy has been using FRET to examine the localization of and interaction between Fpn, Hp, and APP. She has determined that while sAPP binds to Fpn at the PM, endogenous APP does not co-localize. In contrast, Fpn-CFP and Hp-YFP exhibit substantial FRET indicating they do form a complex in the PM consistant with the partnership they have in support of Fe-efflux. Thus, her results support our premise that the ADAM10 cleavage of APP to yield sAPP is the physiologic processing event that leads to modulating Fe-efflux. As you'll see on the following page, Danielle Bailey is probing the connection between iron and the APP processing machinery.