Effect of GFP tags on the localization of EB1 and EB1 fragments in vivo

• Published in: Cytoskeleton (Hoboken, N.J.) N.J.), 2010-01, Vol.67 (1), p.1-12
• Main Authors: Skube, Susan B., Chaverri, José M., Goodson, Holly V.
• Format: Article
• Language: English
• Subjects: Animals; Cercopithecus aethiops; CLIP-170; COS Cells; EB1; GFP; Green Fluorescent Proteins - genetics; Green Fluorescent Proteins - metabolism; Humans; Microscopy, Fluorescence; Microtubule-Associated Proteins - genetics; Microtubule-Associated Proteins - metabolism; Microtubules - metabolism; Neoplasm Proteins - genetics; Neoplasm Proteins - metabolism; Recombinant Fusion Proteins - genetics; Recombinant Fusion Proteins - metabolism; TIPs
• ISSN: 1949-3584; 1949-3592
• DOI: 10.1002/cm.20409

EB1 is a microtubule plus‐end tracking protein that plays a central role in the regulation of microtubule (MT) dynamics. GFP‐tagged EB1 constructs are commonly used to study EB1 itself and also as markers of dynamic MT plus ends. To properly interpret these studies, it is important to understand the impact of tags on the behavior of EB1 and other proteins in vivo. To address this problem and improve understanding of EB1 function, we surveyed the localization of expressed EB1 fragments and investigated whether GFP tags alter these localizations. We found that neither N‐terminal nor C‐terminal tags are benign: tagged EB1 and EB1 fragments generally behave differently from their untagged counterparts. N‐terminal tags significantly compromise the ability of expressed EB1 proteins to bind MTs and/or track MT plus ends, although they leave some MT‐binding ability intact. C‐terminally tagged EB1 constructs have localizations similar to the untagged constructs, initially suggesting that they are benign. However, most constructs tagged at either end cause CLIP‐170 to disappear from MT plus ends. This effect is opposite to that of untagged full‐length EB1, which recruits CLIP‐170 to MTs. These observations demonstrate that although EB1‐GFP can be a powerful tool for studying microtubule dynamics, it should be used carefully because it may alter the system that it is being used to study. In addition, some untagged fragments had unexpected localizations. In particular, an EB1 construct lacking the coiled‐coil tracks MT plus ends, though weakly, providing evidence against the idea that EB1 +TIP behavior requires dimerization. © 2009 Wiley‐Liss, Inc.