News from the caves: update on the structure and function of caveolae Elena Shvets 1 , Alexander Ludwig 2 and Benjamin James Nichols 1 Recent data from the study of the cell biology of caveolae have provided insights both into how these flask-shaped invaginations of the plasma membrane are formed and how they may function in different contexts. This review discusses experiments that analyse the composition and ultrastructural distribution of protein complexes responsible for generating caveolae, that suggest functions for caveolae in response to mechanical stress or damage to the plasma membrane, that show that caveolae may have an important role during the signalling events for regulation of metabolism, and that imply that caveolae can act as endocytic vesicles at the plasma membrane. We also highlight unexpected roles for caveolar proteins in regulating circadian rhythms and new insights into the way in which caveolae may be involved in fatty acid uptake in the intestine. Current outstanding questions in the field are emphasised. Addresses 1 MRC-LMB, United Kingdom 2 School of Biological Sciences, Nanyang Technological University, 60 Nanyang Drive, Singapore 637551, Singapore Corresponding author: Nichols, Benjamin James (ben@mrc-lmb.cam.ac.uk) Current Opinion in Cell Biology 2014, 29:99–106 This review comes from a themed issue on Cell organelles Edited by William A Prinz and David K Banfield http://dx.doi.org/10.1016/j.ceb.2014.04.011 0955-0674/# 2013 Elsevier Ltd. All rights reserved. Introduction Caveolae are invaginations of the plasma membrane with a highly characteristic bulb-like morphology, and a defined size. Although they were identified in the early days of cellular electron microscopy [1,2], many questions remain unanswered as to how caveolae form and, particularly, how they function. The purpose of this review is to very briefly summarise the basic protein components of caveolae, to outline major questions in the field, and to provide more detail on selected discoveries published over the past year or so. These discoveries provide answers to some out- standing issues, and raise interesting new possibilities. The protein caveolin 1, identified by the lab of Richard Anderson in 1992 [3], is a key component of caveolae. It is inserted into the inner leaflet of the plasma membrane, and held in place by multiple acylation [4]. More recently, proteins now termed cavins have been identified as additional structural elements [5–8,9  ,10  ,11  ]. There are 3 caveolins and 4 cavins in vertebrates; their tissue distribution, alternative nomenclature and basic proper- ties are summarised in Box 1. Over the past couple of years another two caveolar proteins have been identified. Pacsin 2 (also called syndapin 2) contains a membrane curvature binding or sensing BAR domain, co-localises with at least a fraction of caveolae, and participates in their morphogenesis [12,13  ,14  ]. EHD2, an ATPase, is also present in caveolae [14  ,15  ,16  ]. How are caveolae formed? Given the recently much expanded parts list for caveolae briefly summarised above and in Box 1, one obvious question is how do these components fit together around the caveolar bulb? Cavin proteins bind to each other in vitro, and can be co-immunoprecipitated from cells [7,8,10  ,17]. They form large complexes that can be resolved in sucrose velocity gradients [18  ]. It has, how- ever, been unclear how these complexes associate with caveolins, and how the constituent proteins are distributed around the caveolar membrane. Similarly, electron micro- scopy provides good evidence for a striated protein coat on caveolae, but the molecular composition of this coat has not been clear [3,19]. Data from our laboratory sheds light on these issues. All (or most) of the cavin and caveolin protein expressed in HeLa cells can be isolated in a single type of 80S complex, the caveolar coat complex [20  ]. The stoichiometry of this complex is likely to be 12 caveolin:3 cavin 1:1 cavin 2 or cavin 3 molecules (Figure 1c). Cavins 2 and 3 compete for binding sites within the overall complex, and can be found in separate sub-complexes with cavin 1. Use of immuno-EM and a novel genetically encoded tag for electron microscopy (miniSOG) allowed us to show that the 80S complex is distributed all around the caveolar bulb, while EHD2 has a different distribution at the neck of caveolae [15  ,16  ,21]. The stoichiometry and likely distri- bution of caveolar protein complexes is summarised in Figure 1. Whether the 80S complex represents the entire protein coat found in a single caveola, or an intermediate oligomeric form, is currently unclear. Structural infor- mation of the isolated 80S complex and of caveolar sub- complexes may help answer this question. We also still lack a firm understanding of how EHD2 and Pacsin 2 recognize the caveolar coat and/or some other determinant within caveolar membranes (such as specific lipids or membrane curvature). Moreover, the existence of a single species of Available online at www.sciencedirect.com ScienceDirect www.sciencedirect.com Current Opinion in Cell Biology 2014, 29:99–106