Background

To understand the various functions of a protein, it is important to find out where it localizes within cells. Standard microscopic and biochemical methods only reveal the presence of a protein when its steady-state concentration is above the detection threshold. They do not rule out the possibility that it plays additional, important roles where it localizes only transiently (Gama-Carvalho and Carmo-Fonseca, 2001). For example, many RBPs perform functions in different cellular compartments where they accompany their bound mRNAs (often going undetected) and connect multiple steps in eukaryotic gene expression (Müller-McNicoll and Neugebauer, 2013). SR proteins (SRSF1 to SRSF12) are a family of RBPs that regulate transcription, pre-mRNA splicing, 3’end processing and mRNP packaging in the nucleus and appear exclusively nuclear at steady state (Howard and Sanford, 2015; Jeong, 2017). However, most family members shuttle continuously (but to different extents) between the nucleus and the cytoplasm, performing additional functions in mRNA export and translation (Caceres et al., 1998; Sapra et al., 2009; Maslon et al., 2014; Müller-McNicoll et al., 2016; Botti et al., 2017). Changes in RBP shuttling have been described in viral infections, early development, cellular differentiation and neurodegenerative diseases such as ALS and FTD, where pathological accumulation of prion-like RBPs such as FUS and TDP-43 in the cytoplasm forms solid neurotoxic aggregates (Ederle and Dormann, 2017; Liu et al., 2017). Thus, it is very important to know whether an RBP normally shuttles between the nucleus and the cytoplasm and if so, under which circumstances and how it is controlled.

With the tools currently available, it has been difficult to study the cytoplasmic functions of nuclear RBPs and to compare their shuttling abilities. An ingenious method was developed almost thirty years ago–the interspecies heterokaryon assay–in which donor and recipient cells from two different species (e.g., mouse and human) are fused and a protein present only in the donor nuclei gradually appears in the recipient nuclei if it shuttles (Borer et al., 1989). However, this assay provides only qualitative information. The fusion events are identified based on phase-contrast images and donor and recipient nuclei are identified based on distinct chromatin features, which makes the assay laborious, subjective and produces only small numbers of high-confidence heterokaryons. Moreover, fluorescently tagged RBPs are often expressed from transiently transfected plasmids, which results in very different RBP levels in different cells, ranging from barely detectable to non-physiologically high expression that may lead to partially aberrant cytoplasmic or subnuclear localization of RBPs [(Maharana et al., 2018) our unpublished observations]. Altogether, these limitations preclude any comparative analyses.

Here, we present a detailed experimental protocol to perform quantitative shuttling assays in cultured mammalian cells (Figure 1). Our assay is an improvement of the classical heterokaryon assay, with two novelties and a standardized imaging pipeline to perform quantitative measurements. The first novelty is the use of recipient cell lines expressing a fluorescently tagged membrane marker (CAAX-mCherry), which greatly facilitates the identification of heterokaryons containing both a donor and a recipient nucleus and thus allows the rapid and easy identification of a large number of high-confidence heterokaryons. The second novelty is the use of stable clonal donor cell lines, where a fluorescently (eGFP) tagged RBP of interest is expressed from a bacterial artificial chromosome (BAC), which has been integrated into the genome (Botti et al., 2017; Poser et al., 2008). Subsequent clonal selection of cells ensures equal and near-physiological levels of tagged RBPs in every donor cell to facilitate image acquisition, analysis and comparisons.

Our assay has been successfully applied to compare the shuttling activities of different SR proteins in the same cell line, between different cell lines and between differentiation states. Moreover, it allowed us to study the requirements for the shuttling of individual SR proteins using mutated proteins and knockdowns of nuclear export factors (Botti et al., 2017). We have used various cell lines as either donor or recipient cells: these comprise mouse (P19 and NIH3T3) and human (HeLa) cells. Although other cell lines remain to be tested, we are confident that any adherent cell line in which fluorescent RBPs can be expressed at physiological levels is suitable for our assay. This should include primary cells obtained from transgenic animals expressing a fluorescently tagged RBP of interest. We have successfully used P19 cells differentiated into neural cells as donors in our assays (Botti et al., 2017), and it should be possible to study and quantify shuttling of RBPs in other cellular models of differentiation, for example in mouse embryonic stem cells (ESCs) or induced pluripotent stem cells (iPSCs), or to compare shuttling of RBPs in distinct cellular differentiation fates (Hammarskjold and Rekosh, 2017). Moreover, our assay should allow to quantify changes in shuttling during viral infections and cellular stress, or to assess the impact of disease mutations in RBPs. In principle, our assay could even be adapted to visualize shuttling of long-noncoding RNAs (lncRNAs), for example through the insertion of binding sites for fluorescent MS2 binding protein (MS2-BP) or by inserting an aptamer sequence that binds a fluorescent dye (Ouellet, 2016).