emission of the various FPs across the different fluorescence detectors. This overlap is eliminated by compensation, which involves
the reciprocal subtraction of signals from the detectors in proportion to their spectral overlap. If compensation is excessive, it
reduces sensitivity and dynamic range. Consequently, the spectral
properties of different FPs used in combination must not be narrowly separated across the spectrum. For example, GFP and YFP
are generally not employed together for flow cytometry in any
organism, since the high degree of spectral overlap between the
fluorescence emission of these two proteins, when excited at
488 nm and using a conventional cytometer, requires excessive
compensation. This issue can be handled by clever involvement of
other laser lines (for example, 450 nm (blue) for selective GFP
excitation [50], and even combined with 532 nm (yellow) for
selective YFP excitation). The recent emergence of spectral analyzers has also greatly increased the ability of cytometers to distinguish different FPs, including GFP and YFP [59].
The inherent ability of current instruments to distinguish six
FPs is only one part of the problem; it does not imply that this
number of FPs can be stably expressed in transgenic organisms,
given issues of adverse pleiotropic effects. On the other hand,
Hawley et al. [59] have demonstrated that it is possible to detect
simultaneous expression of five FPs (ECFP, EGFP, EYFP, DsRed,
and HcRed) in lentivirus-transfected mammalian Sp2/0-Ag14
cells, perhaps since this line is capable of tolerating a high level of
overall protein synthesis. Culture beyond a period of several
months has not been tested (T. Hawley, pers. commun.). Recently,
simultaneous expression of five different FPs (CFP, EGFP, YFP,
mOrange2, and mApple fused to various organellar targeting signals) has been reported in transfected Cos7 cells, but observations
were done only for brief time periods following transfection, and
permanent transgenic cell lines were not produced [60]. Other
workers have reported successful co-expression of BFP, Cerulean3,
GFP, mCherry, and iRFP in large, highly productive mammalian
cell lines such as HeLa and K562 cells (M. Paulsen, EMBL Heidelberg, pers. commun.).
In terms of pleiotropic effects, Costantini et al. [61] have
described how to handle the tendencies of native FPs to oligomerize and to form inappropriate disulfide linkages. It is clear that
quantitative measurement of FP fluorescence within different subcellular compartments requires that the FPs remain in equivalent
monomeric states within these compartments, that they remain
unaffected by the different biochemical environments encountered,
and that their expression does not perturb the native structure of
the organelle. Focusing on the endomembrane-secretory pathway,
they designed variants of the Blue (EBFP2), Cyan (Cerulean),
Yellow (Venus), and Green (sfGFP) Fluorescent Proteins that
avoid forming inappropriate disulfide linkages, and that, as
262
David W. Galbraith and Guiling Sun
the reciprocal subtraction of signals from the detectors in proportion to their spectral overlap. If compensation is excessive, it
reduces sensitivity and dynamic range. Consequently, the spectral
properties of different FPs used in combination must not be narrowly separated across the spectrum. For example, GFP and YFP
are generally not employed together for flow cytometry in any
organism, since the high degree of spectral overlap between the
fluorescence emission of these two proteins, when excited at
488 nm and using a conventional cytometer, requires excessive
compensation. This issue can be handled by clever involvement of
other laser lines (for example, 450 nm (blue) for selective GFP
excitation [50], and even combined with 532 nm (yellow) for
selective YFP excitation). The recent emergence of spectral analyzers has also greatly increased the ability of cytometers to distinguish different FPs, including GFP and YFP [59].
The inherent ability of current instruments to distinguish six
FPs is only one part of the problem; it does not imply that this
number of FPs can be stably expressed in transgenic organisms,
given issues of adverse pleiotropic effects. On the other hand,
Hawley et al. [59] have demonstrated that it is possible to detect
simultaneous expression of five FPs (ECFP, EGFP, EYFP, DsRed,
and HcRed) in lentivirus-transfected mammalian Sp2/0-Ag14
cells, perhaps since this line is capable of tolerating a high level of
overall protein synthesis. Culture beyond a period of several
months has not been tested (T. Hawley, pers. commun.). Recently,
simultaneous expression of five different FPs (CFP, EGFP, YFP,
mOrange2, and mApple fused to various organellar targeting signals) has been reported in transfected Cos7 cells, but observations
were done only for brief time periods following transfection, and
permanent transgenic cell lines were not produced [60]. Other
workers have reported successful co-expression of BFP, Cerulean3,
GFP, mCherry, and iRFP in large, highly productive mammalian
cell lines such as HeLa and K562 cells (M. Paulsen, EMBL Heidelberg, pers. commun.).
In terms of pleiotropic effects, Costantini et al. [61] have
described how to handle the tendencies of native FPs to oligomerize and to form inappropriate disulfide linkages. It is clear that
quantitative measurement of FP fluorescence within different subcellular compartments requires that the FPs remain in equivalent
monomeric states within these compartments, that they remain
unaffected by the different biochemical environments encountered,
and that their expression does not perturb the native structure of
the organelle. Focusing on the endomembrane-secretory pathway,
they designed variants of the Blue (EBFP2), Cyan (Cerulean),
Yellow (Venus), and Green (sfGFP) Fluorescent Proteins that
avoid forming inappropriate disulfide linkages, and that, as
262
David W. Galbraith and Guiling Sun