used to transfer information between cells [1, 2]. The gamut of
transferred RNAs includes microRNAs (miRNAs), tRNAs, small
noncoding RNAs (sncRNAs), Y RNAs, long noncoding RNAs
(lncRNAs), fragments of mRNAs, and full-length mRNAs [3–
14]. RNAs have been shown to undergo transfer by both contactindependent (i.e., via extracellular vesicles, such as exosomes) and
contact-dependent routes (i.e., via long thin cytoplasmic projections called membrane nanotubes; mNTs) [13–15]. The transfer of
RNAs has been shown to affect the transcriptome of downstream
acceptor cells by either miRNA or lncRNA regulation or direct
transfer of mRNA [8, 10, 13, 16, 17]. Although several studies
profiled the RNAs contained in the exosomes by employing DNA
microarrays or RNA sequencing (RNA-seq), no systematic study
has been done to identify the transferred RNAs present in downstream acceptor cells after transfer [18–20]. Hence, the scope of
RNA transfer has not been fully understood due to the lack of
unbiased and quantitative approaches to study transferred RNAs.
In our work, we have demonstrated the direct intercellular
transfer of full-length mRNAs in a contact-dependent manner, via
mNTs [13]. mNT-mediated mRNA transfer was observed between
adherent mammalian cells in culture, including between heterologous cell types [e.g., mouse embryonic fibroblasts (MEFs),
HEK293, U2OS, and Hela cells] and cell states (i.e., primaryprimary, primary-immortalized, and immortalized-immortalized).
mRNA transfer largely correlated with gene expression and, while a
number of mRNAs were shown to undergo transfer (e.g., mouse
β-actin and human cyclin D1, BRCA1, MT2A), we employed MS2
aptamer-tagged mouse β-actin (β-actin-MBS) to show that transfer
is mNT (and not EV) mediated, regulated by stress, and can be
visualized using either single-molecule fluorescent in situ hybridization (smFISH) or live imaging using the MS2 coat protein fused
to GFP. While numerous questions abound, the intercellular transfer of mRNA appears to be a common phenomenon of adherent
cells in in vitro culture. However, before determining whether this
phenomenon also occurs in vivo it is important to first examine the
extent of transfer, i.e., which species of mRNAs undergo transfer,
and at what levels, and whether transferred mRNAs have common
cis-acting determinants that allow for their selection for transfer to,
and translation in, acceptor cells.
In this chapter, we put forth a simple method to elucidate the
entire spectrum of transferred RNAs after the co-culture of two
heterologous cell populations (i.e., “donor” and “acceptor” cells)
in a non-biased and quantitative fashion. Briefly, donor and acceptor cell lines are cultured together in the same dish, following which
the different cell types are sorted into their component cell populations using cell surface antigen-based magnetic sorting. Total RNA
from the sorted populations is collected and checked for integrity
and quality, and the RNA samples are subjected to RNA-seq to
196
Sandipan Dasgupta and Jeffrey E. Gerst
transferred RNAs includes microRNAs (miRNAs), tRNAs, small
noncoding RNAs (sncRNAs), Y RNAs, long noncoding RNAs
(lncRNAs), fragments of mRNAs, and full-length mRNAs [3–
14]. RNAs have been shown to undergo transfer by both contactindependent (i.e., via extracellular vesicles, such as exosomes) and
contact-dependent routes (i.e., via long thin cytoplasmic projections called membrane nanotubes; mNTs) [13–15]. The transfer of
RNAs has been shown to affect the transcriptome of downstream
acceptor cells by either miRNA or lncRNA regulation or direct
transfer of mRNA [8, 10, 13, 16, 17]. Although several studies
profiled the RNAs contained in the exosomes by employing DNA
microarrays or RNA sequencing (RNA-seq), no systematic study
has been done to identify the transferred RNAs present in downstream acceptor cells after transfer [18–20]. Hence, the scope of
RNA transfer has not been fully understood due to the lack of
unbiased and quantitative approaches to study transferred RNAs.
In our work, we have demonstrated the direct intercellular
transfer of full-length mRNAs in a contact-dependent manner, via
mNTs [13]. mNT-mediated mRNA transfer was observed between
adherent mammalian cells in culture, including between heterologous cell types [e.g., mouse embryonic fibroblasts (MEFs),
HEK293, U2OS, and Hela cells] and cell states (i.e., primaryprimary, primary-immortalized, and immortalized-immortalized).
mRNA transfer largely correlated with gene expression and, while a
number of mRNAs were shown to undergo transfer (e.g., mouse
β-actin and human cyclin D1, BRCA1, MT2A), we employed MS2
aptamer-tagged mouse β-actin (β-actin-MBS) to show that transfer
is mNT (and not EV) mediated, regulated by stress, and can be
visualized using either single-molecule fluorescent in situ hybridization (smFISH) or live imaging using the MS2 coat protein fused
to GFP. While numerous questions abound, the intercellular transfer of mRNA appears to be a common phenomenon of adherent
cells in in vitro culture. However, before determining whether this
phenomenon also occurs in vivo it is important to first examine the
extent of transfer, i.e., which species of mRNAs undergo transfer,
and at what levels, and whether transferred mRNAs have common
cis-acting determinants that allow for their selection for transfer to,
and translation in, acceptor cells.
In this chapter, we put forth a simple method to elucidate the
entire spectrum of transferred RNAs after the co-culture of two
heterologous cell populations (i.e., “donor” and “acceptor” cells)
in a non-biased and quantitative fashion. Briefly, donor and acceptor cell lines are cultured together in the same dish, following which
the different cell types are sorted into their component cell populations using cell surface antigen-based magnetic sorting. Total RNA
from the sorted populations is collected and checked for integrity
and quality, and the RNA samples are subjected to RNA-seq to
196
Sandipan Dasgupta and Jeffrey E. Gerst