Design and use of fluorescent fusion proteins in cell biology

Abstract

This unit describes strategies for designing functional fluorescent fusion protein constructs. Such constructs can be exploited as probes of cellular environments, protein dynamics, protein life histories, protein binding partners, and markers in living cells. The properties and uses of many currently available fluorescent proteins are discussed. In addition, alternative approaches and troubleshooting guidelines are provided.

Figure 21.4.1

Appropriate positioning of a fluorescent protein (FP) in a fluorescent fusion protein construct. Preferred sites of FP fusion in the primary sequence of the protein of interest are indicated by happy face icons, and domains to be avoided are indicated by sad face icons. Each tertiary structure shows the folding of the sample construct with the FP (represented as a cylinder) fused at an optimal site. (A). A hypothetical globular protein expressed in the cytoplasm can have the FP fused at either the NH₂₋ or the COOH-terminus. Typically, one end of the protein of interest will contain a functional domain that may be sterically hindered by an FP, and so it is useful to make both of the possible constructs. (B) A hypothetical lumenal protein contains an NH₂₋terminal signal sequence (SS), a mature domain, and a COOH-retention sequence (KDEL). An FP placed immediately after the SS or immediately before the retention sequence is less likely to interfere with the functioning of either sequence. (C) A single membrane–spanning protein has the additional constraint that the FP cannot be placed within or near the transmembrane domain (TMD), as this will disrupt the domain and cause problems with membrane integration. (D) A membrane multispanning protein has the same constraint as the example in panel C, but in multiple locations. The loops between the transmembrane domains are also poor choices, because the exact spacing between transmembrane domains is often important for protein folding, and because these loops often contain functional domains. Abbreviation: cyt, cytoplasm.

Figure 21.4.2

Relative sizes of (A) immunoglobulin G (IgG; reference for comparison with panels B to D), (B) green fluorescent protein (GFP), (C) the Discosoma red fluorescent protein (DsRed) tetramer, and (D) biarsenical tetracysteine.

Figure 21.4.3

The fusion of a fluorescent protein (FP) to a native protein may change the protein's normal localization pattern or may lead to the formation of aggregates or oligomers. (A) The fusion of nonmonomerized enhanced green fluorescent protein (EGFP) to a resident endoplasmic reticulum (ER) membrane protein induces the formation of an organized smooth ER structure. (B) The fusion of monomerized EGFP to the same protein does not grossly alter the structure of the ER. (C) A Cos-7 cell expressing two fluorescent fusion proteins (FFPs), one containing monomerized green fluorescent protein (mGFP; left-hand image) and the other containing monomerized red fluorescent protein (mRFP; right-hand image). The mGFP-containing FFP localizes to the ER network, and similarly, the mRFP-containing FFP colocalizes to the ER membranes. The image yielded by the mRFP-tagged protein shows bright puncta, which are probably mRFP aggregates.