Higher organisms such as mammals are composed of a number of organs that all fulfill specific tasks for which they need to communicate. To enable targeted, controlled communication, these functional units have to be separated, both spatially and biochemically. All animal organs are therefore surrounded by epithelia which, besides protecting them from harmful environmental influences, ensure a controlled exchange of substances with their surroundings. The smallest unit is the polarized epithelial cell. These cells form a barrier exactly the thickness of a cell layer consisting of an uninterrupted row of cells. The tight junctions between the cells create a hermetically sealed surface through which controlled transport processes take place.
Epithelial cells typically have a polar structure. The most striking feature is the difference in the cytoplasmic membrane. There are two different domains: the apical membrane facing the exterior of the organ and the basolateral membrane that faces the interior of the organ or neighboring cells inside the epithelium. The two membrane domains exhibit clearly different protein and lipid compositions. For example, various digestion enzymes such as LPH (lactase-phlorizin-hydrolase) or SI (sucrose-isomaltase) are only transported into the apical cytoplasm membrane of the intestinal epithelium, where they reach their site of action – the intestinal lumen. These enzymes are not found in the basolateral membrane. There would be no point in transporting them there – it would be a waste of energy.
So what are the basic mechanisms of this specific protein sorting? The answer is provided by the proteins themselves, which all carry specific topogenous signals. These can be short amino acid sequences (YXXφ, NPXY, LL, L) like those found in basolaterally transported proteins. The signal effect may also be based on the glycosylation of the proteins as is the case with various apical proteins. GPI anchors (Glycosyl-Phosphatidyl-Inositol) also serve as apical sorting signals. All topogenous signals are an identifying code for other proteins or lipid structures, which then act as a sorting receptor or sorting platform. For example, some basolateral proteins are sorted by adapter proteins (AP1A) of the clathrin-dependent transport in the trans-Golgi network and transported to the target membrane  (Gonzalez and Rodriguez-Boulan, 2009). Proteins with GPI anchors, on the other hand, show a high affinity with specific cholesterin-rich lipid structures – lipid rafts – which are sorting platforms for the apical membrane. There are yet other apical proteins like P75-GFP, that are identified and bound due to their special glycosylation of the sugar-binding protein Galectin-3, which ultimately leads to transport into the correct membrane domain (Figure 1) .
In order to study the processes by which proteins are sorted and transported into the apical membrane of polarized epithelial cells in more detail, the obvious step is to take a closer look at this cell pole under the microscope. TIRF microscopy (Total Internal Reflection Fluorescence) is particularly useful for such studies, as it delivers highly detailed images of structures on, in and immediately under the apical cytoplasm membrane. This applies to both fixed and living cells.
Fig. 1: Protein sorting in polarized epithelial cells – Polarized epithelial cells bear two different plasma membrane domains – an apical and a basolateral – which differ in their protein and lipid constitution and are separated by Tight-Junctions. Several apical (X,Y) and basolateral protein sorting mechanisms have been discovered. One apical pathway is controlled by the carbohydrate binding protein Galectin-3 which is attaching to some apically sorted proteins like P75-GFP, LPH or GP-114. This sorting process takes place in a post Golgi compartment which most probably is a Recycling Endosome or in close conjunction to it. ER: Endoplasmatic Reticulum, TGN: Trans-Golgi-Network, PGC: Post Golgi Compartment.
Proteins being transported to the apical membrane or involved in this transport process can be visualized in the microscope by marking them with a fluorescing protein tag (e.g. GFP) or by immunofluorescence staining. Apical TIRFM is a suitable method for this, as it is capable of resolving structures at the apical cell pole that go undetected in other microscopy techniques . This means that apical proteins (e.g. P75-GFP) can be imaged together with their sorting receptor Galectin-3-DsRed. Co-localization events suggest an interaction of the two proteins (Figure 2). So TIRFM enables in-vivo visualization of protein-protein interactions in apical protein trafficking that have already been proven by biochemical methods. As the evanescent field of the TIRF microscope penetrates no further than 300 nm into the cell, the area close to the membrane is clearly resolved, allowing the protein interaction site to be defined to an area in or immediately beneath the apical membrane.
TIRFM shows up tubular structures at the apical cytoplasm membrane which are very probably microvilli and intracellular endosomal structures. Certain endosomes are thought to be protein sorting centers. It was thus possible to detect both the apical sorting receptor Galectin-3 and its target protein P75-GFP in a Rab-11- positive recycling endosome directly under the apical surface .
TIRFM can therefore be used to show that the apical protein P75-GFP and its sorting receptor Galectin-3 are present in a sub-apical recycling endosome. This indicates that it could be the protein sorting site.
Interesting approaches for future studies are live observations of protein sorting processes with the aid of TIRFM. Due to the strong membrane staining of most apical proteins, it will be necessary to resort to inducible protein systems that allow the pathway of an apical protein from the endoplasmic reticulum to the cytoplasm membrane to be tracked.
The already mentioned carbohydrate-binding Galectin-3 plays a role in the apical protein sorting process. Galectin-3 is seen as the proven sorting receptor for the apical proteins P75-GFP and GP-114. The knock-down of this lectin results in the basolateral transport of P75-GFP and GP-114 . Besides the above-mentioned search for the protein sorting process site, there is also the question of the fate of Galectin-3 once it has performed its task as sorting receptor. Biochemically, it has already been shown that some of the lectin passes from the cell to the apical medium. Galectin-3 is hardly ever found in the basolateral medium. Unfortunately, it has not yet been possible to record this exocytosis process with TIRFM.
On the other hand, Galectin-3 is re-endocyted by polarized epithelial cells such as MDCK cells. This process can be observed by applying recombinant Alexa633-coupled Galectin-3 to the apical side of the cells. In the TIRFM image, point-shaped red Galectin-3 signals can be seen on the cell surface at the beginning of the experiment. These spots become weaker and more diffuse as recording continues, because the absorbed Galectin-3-Alexa633 moves out of the evanescent field of the TIRFM after its uptake. If the same experiment is observed in the 3D stack, recorded with an epifluorescence microscope, it becomes clear that Galectin-3-Alexa633 is ingested into intracellular compartments (probably endosomes) during the course of the experiment (Figure 3).
Fig. 3: Recombinant Galectin-3-A633 is endocytosed at the apical cell pole – The apical TIRFM signal of Galectin-3-A633 is vanishing gradually after starting endocytosis at 37 °C. 3D stacks of epifluorescence data – seen from the apical side – show a moving Galectin-3-A633 signal from the apical surface of MDCK cells into intracellular endosomal structures. The location of these structures can be compared with the plasma membrane staining of P75-GFP. Time after starting endocytosis is mentioned on the left. Penetration depth 90 nm, Scale bar 10 µm.
As already mentioned, TIRFM improves the detail rendering of individual structures on the apical surface of polarized epithelial cells. It images not only intracellular compartments but also the cytoplasm membrane and thus associated proteins. In this way, the actin cytoskeleton at the apical cell pole can be shown by staining with phalloidin-coupled fluorochromes. The result is a plastic image of many tubular shapes showing, among other things, the microvilli prominent in MDCK cells. The apical protein P75-GFP is found in their membrane (Figure 4.1A). However, not all P75-GFP-tagged areas are situated in microvilli (arrow heads). Very probably, these are the above-mentioned recycling endosomes that have a similar shape.
If MDCK cells stained with a certain form of tubulin (acetylated tubulin) are observed in the apical TIRFM, a single long, tubular structure can be seen. This is the primary cilium (Figure 4.1B), which serves as a mechanosensory tool of the cell. The apical protein P75-GFP can be found here, too. Vesicular transport of this protein together with its sorting receptor Galectin-3 can be observed along or in the primary cilia of living MDCK cells (Figure 4.2). This too may be a process that leads to lectin secretion.
Fig. 4.1: P75-GFP associates with cytosceletal structures in TIRFM – A) The apical protein P75-GFP can be found in microvilli structures which are stained by actin binding Phalloidin-Rhodamin (arrows). Furthermore there are P75-GFP signals detectable outside microvilli which are most probably recycling endosomes (arrow heads). B) P75-GFP does not only show up in microvilli (arrow heads) but also in primary cilia (arrow), stained by primary cilium specific acetylated Tubulin Penetration depth 90 nm, Scale bar 10 µm.
Fig. 4.2 and Video: P75-GFP and Galectin-3-DsRed at the primary cilium – Life cell imaging of apical TIRFM deciphers vesicular movements of Galectin-3-DsRed together with P75-GFP (arrows) along the primary cilium of MDCK cells. Absolute time 3:55 min, Penetration depth 90 nm, Scale bar 10 µm.
In conclusion, it can be said that TIRFM enables a dramatically improved view of the interesting area of the apical cell pole of polarized epithelial cells, where apically sorted proteins such as P75-GFP can be observed together with their sorting receptor galectin-3. It is also possible to see the potential sorting compartment – a recycling endosome. TIRFM not only visualizes protein transport processes on the secretory pathway, but also endocytosis events in which Galectin-3 is re-absorbed into the cell. These processes can be viewed in both fixed and living cells. In the latter case, it is also possible to record transport events of Galectin-3 and P75-GFP in the primary cilium.
- Delacour D, Cramm-Behrens CI, Drobecq H, Le BA, Naim HY, Jacob R: Requirement for Galectin-3 in apical protein sorting. Curr Biol. 16 (2006) 408–414.
- Gonzalez A, Rodriguez-Boulan E: Clathrin and AP1B: key roles in basolateral trafficking through trans-endosomal routes. FEBS Lett 583 (2009) 3784-3795.
- Schneider D, Greb C, Koch A, Straube T, Elli A, Delacour D, Jacob R: Trafficking of Galectin-3 through endosomal organelles of polarized and non-polarized cells. Eur J CellBiol (2010).http://www.ncbi.nlm.nih.gov/pubmed/20705359
- Greb C, Jacob R: Fluoreszenzmikroskopie an der apikalen Membran polarer Epithelzellen – TIRF-Mikroskopie auf den Kopf gestellt. BIOspektrum 17:3 (2011) 314–316.
- Greb C, Jacob R: Fluorescence Microscopy of the Apical Membrane of Polarized Epithelial Cells – TIRF Microscopy turned upside Down. Science Lab (2011) http://www.leica-microsystems.com/science-lab/tirf-microscopy-of-the-apical-membrane-of-polarized-epithelial-cells/