Near-infrared imaging in fission yeast using a genetically encoded phycocyanobilin biosynthesis system
This article has been Reviewed by the following groups
Listed in
- Evaluated articles (Review Commons)
Abstract
Near-infrared fluorescent protein (iRFP) is a bright and stable fluorescent protein with near-infrared excitation and emission maxima. Unlike the other conventional fluorescent proteins, iRFP requires biliverdin (BV) as a chromophore. Here, we report that phycocyanobilin (PCB) functions as a brighter chromophore for iRFP than BV, and that biosynthesis of PCB allows live-cell imaging with iRFP in the fission yeast Schizosaccharomyces pombe. We initially found that fission yeast cells did not produce BV and therefore did not show any iRFP fluorescence. The brightness of iRFP–PCB was higher than that of iRFP–BV both in vitro and in fission yeast. We introduced SynPCB2.1, a PCB biosynthesis system, into fission yeast, resulting in the brightest iRFP fluorescence. To make iRFP readily available in fission yeast, we developed an endogenous gene tagging system with iRFP and all-in-one integration plasmids carrying the iRFP-fused marker proteins together with SynPCB2.1. These tools not only enable the easy use of multiplexed live-cell imaging in fission yeast with a broader color palette, but also open the door to new opportunities for near-infrared fluorescence imaging in a wider range of living organisms.
This article has an associated First Person interview with the first author of the paper.
Article activity feed
-
-
-
Note: This rebuttal was posted by the corresponding author to Review Commons. Content has not been altered except for formatting.
Learn more at Review Commons
Reply to the reviewers
- General Statements We are grateful to all reviewers for the critical comments and valuable suggestions that have helped improve our paper. We provide point-to-point answers to the comments and added detailed explanations in the preliminary revised manuscript. We put the comments made by the reviewer in* italics* with our responses below.
Description of the planned revisions
____Reviewer #1____
**Major comments:**
- The authors thought almost all iRFP forms a holo-complex with BV when HO1 is expressed in fission yeast. This should be proved by calculating the percentage of BV-iRFP in yeast. It is meaningful to compare the percentage of BV-iRFP and PCB-iRFP …
Note: This rebuttal was posted by the corresponding author to Review Commons. Content has not been altered except for formatting.
Learn more at Review Commons
Reply to the reviewers
- General Statements We are grateful to all reviewers for the critical comments and valuable suggestions that have helped improve our paper. We provide point-to-point answers to the comments and added detailed explanations in the preliminary revised manuscript. We put the comments made by the reviewer in* italics* with our responses below.
Description of the planned revisions
____Reviewer #1____
**Major comments:**
- The authors thought almost all iRFP forms a holo-complex with BV when HO1 is expressed in fission yeast. This should be proved by calculating the percentage of BV-iRFP in yeast. It is meaningful to compare the percentage of BV-iRFP and PCB-iRFP in vitro and in yeast.
We would like to thank the reviewer for this valuable comment. We plan to quantify the percentage of iRFP holo-complex in vitro and in yeast by using fluorescence correlation spectroscopy (FCS). In FCS, the fluctuation of fluorescence emitted from a very tiny space (i.e., confocal volume, ~ 1 fL) in solution is measured and statistically analyzed by autocorrelation, providing the average number of fluorescent molecules in the confocal volume (Krichevsky and Bonnet, 2002, Rep. Prog. Phys.). The fusion proteins of iRFP with mNeonGreen (mNG) are purified or expressed in yeast, and thus the stoichiometry of iRFP to mNG must be 1:1. The numbers of iRFP and mNG are measured by FCS in vitro or in yeast. If iRFP forms 100% holo complex with BV or PCB, the number of fluorescent iRFP is equal to that of mNG. We have used FCS to measure the concentration of fluorescent protein in mammalian cells (Sadaie et al., 2014, MCB; Komatsubara et al., 2020, JBC), and therefore have no technical difficulties. One possible limitation is the maturation efficiency of mNG in vitro and in yeast, although it would not disturb at least the qualitative comparison of holo-complex formation between iRFP-BV and iRFP-PCB. In this preliminary version of the revised manuscript, we added the data of FCS measurement in yeast, but not yet in vitro, in Supplementary Figure S5A. The numbers of fluorescent iRFP molecules relative to the numbers of mNG were comparable between iRFP-BV (HO1 expressing cell) and iRFP-PCB (SynPCB2.1 expressing cell). Of note, the number of fluorescent iRFP molecules in cells treated with external BV was significantly lower than that in cells expressing HO1, suggesting the low permeability of BV.
In addition to the FCS measurement, we prepare the following backup experiments; recombinant iRFP proteins are mixed with sufficient amounts of BV or PCB, and used them as a reference. Then, the iRFP proteins are expressed and purified from yeast, and subjected to Zinc blot with the reference iRFP proteins to compare the percentage of iRFP holo-complex in vitro and in yeast. This method is feasible, but we are not able to quantitatively determine the percentage of iRFP holo-complex with BV or PCB.
-The brightness of fluorescent proteins in organisms often depends on the molecular brightness (fluorescence quantum yield and extinction coefficient) and the amounts of fluorescent proteins. The authors indicated that iRFP-PCB is brighter than iRFP-BV at the molecular level. To calculate the amounts of iRFP-PCB and iRFP-BV when different proteins are expressed in yeast, it is better to explain the results that phycocyanobilin was better than biliverdin for the imaging in fission yeast.
We agree with the reviewer’s comment. We will quantify the amounts of iRFP in the fission yeast cells expressing iRFP, iRFP/HO1, and iRFP/SynPCB2.1 by western blot analysis and fluorescent imaging. With regard to western blot analysis, the iRFP proteins are normalized by Tubulin in western blot analysis, which has been used as the reference protein.
During this preliminary revision period, we assessed the expression levels by fluorescence microscopy. The fusion proteins of iRFP fused with mNG were expressed in yeast and the expression levels were quantified by the fluorescence signals of mNG. In this preliminary version of the revised manuscript, we only added fluorescence imaging data in Supplementary Figure S4. The fluorescence intensities of mNG among cells expressing iRFP, iRFP/HO1, and iRFP/SynPCB2.1 were comparable as well as cells treated with BV or PCB. We will further confirm this result by western blotting analysis.
To test the application of PCB as chromophore in mammalian cells, a HO1 gene knock out mammalian cells should be used.
We would like to thank the reviewer for this excellent suggestion. It is indeed interesting to examine the effect of an *HO1 *gene knock-out on the iRFP fluorescence and the application of PCB as a chromophore in mammalian cells. We plan to knock out the *HO1 *gene in HeLa and HeLa/BVRA KO cells by conventional CRISPR/Cas9 genome editing techniques. After the establishment of HO1-KO HeLa cell lines, iRFP fluorescence is assessed in the same way as we did in Figure S9. In addition, we will carefully investigate the effect of BV in serum on iRFP fluorescence.
The authors may use the all-in-one plasmids carrying SynPCB2.1 and iRFP fusion protein genes to image the target proteins in mammalian cells.
According to the reviewer’s suggestion, we will construct an all-in-one plasmid carrying SynPCB2.1 and an iRFP fusion protein gene for iRFP imaging in mammalian cells.
To this end, the IRES and iRFP genes are inserted downstream of the SynPCB2.1 gene cassette, and iRFP fluorescence is confirmed in mammalian cells.
Description of the revisions that have already been incorporated in the transferred manuscript
Reviewer #1
**Minor comments:**
- In line 35. iRFP is derived from bacteriophytochromes.
We have replaced “phytochrome” with “bacteriophytochrome”.
- In line 62. The reference of Rodriguez et al. deals with allophycocyanin instead of bacteriophytochromes.
We have included the following note in the revised manuscript.
(Page 3, line 56)
“Near-infrared fluorescent proteins have been developed through the engineering of phytochromes, which are photosensory proteins of plants, bacteria, and fungi (Chernov et al., 2017), or allophycocyanin, which is a light-harvesting phycobiliprotein of cyanobacteria (Rodriguez et al., 2016). “
- In lines 65-66. Bacteriophytochromes bind BV, phycocyanin, allophycocyanin and cyanobacterial phytochromes bind PCB, and plantal phytochromes bind PΦB.
Thank you for the notification. We have added the explanation in the revised manuscript as follows.
(Page 3, line 65)
“Unlike the canonical fluorescent proteins derived from jellyfish or coral, phytochromes and allophycocyanin require a linear tetrapyrrole as a chromophore such as biliverdin IXα (BV), phycocyanobilin (PCB), or phytochromobilin (PΦB); bacteriophytochromes bind to BV, allophycocyanin and cyanobacterial phytochromes bind to PCB, and plantal phytochromes bind to PΦB.”
- In lines 67-71. The authors described the biosynthesis of BV, PCB and PΦB. But it missed many references.
We agree with this reviewer’s comment, and have included the references in the revised manuscript as follows.
(Page 3, line 70)
“These linear tetrapyrroles are produced from heme (Terry and Lagarias 1991; Beale 1993). Heme-oxygenase (HO) catalyzes oxidative cleavage of heme to generate BV with the help of ferredoxin (Fd), an electron donor, and ferredoxin-NADP+ reductase (Fnr) (Cornejo, Willows, and Beale 1998). In cyanobacteria, PCB is produced from BV through PcyA, Fd, and Fnr, while in plants PΦB is synthesized from BV using HY2, Fd, and Fnr (Muramoto et al. 1999; Frankenberg et al. 2001; Kohchi et al. 2001).“
- In line 270. One full stop should be deleted in "cells.. Fission".
We have corrected this mistake.
- In line 272. How did the authors add the high concentration of PCB (625 microM) into the culture? PCB is insoluble.
As the reviewer pointed out, 625 microM PCB seem to be insoluble in PBS solution and fission yeast culture medium, because insoluble PCB debris was observed in the medium. We have included the following note in Materials and Methods of the revised manuscript.
(Page 20, line 505)
“Of note, 625 μM PCB or 625 μM BV is insoluble in PBS solution and fission yeast culture medium, and 25 μM PCB or 25 μM BV is insoluble in mammalian cell culture medium, because insoluble PCB or BV debris is observed.”
- In line 401. smURFP is derived from allophycocyanin instead of cyanobacteriochromes.
We have corrected this mistake.
- In line 474. As BV and PCB are insoluble, how do the authors add the pigments into DMEM?
We directly added BV or PCB dissolved in DMSO into the culture medium of HeLa cells, i.e., DMEM, 10% FBS, and maintained cells for 3 hours. We have included the following note in the figure legend of the revised Supplementary Information.
Fig. S9 legend
“Of note, 25 μM BV or PCB remained unsolved in the cultured medium for mammalian cells, and therefore chromophores in the medium could be saturated under these conditions.”
- In line 509. In which solvent are BV and PCB dissolved?
BV or PCB dissolved in DMSO was added into His-iRFP PBS solution. We have added the explanations as follows.
(Page 22, line 555)
“Purified His-iRFP in PBS solution was mixed with an excess amount (1:5 molar ratio) of BV or PCB dissolved in DMSO, followed by size exclusion chromatography with NAP-5 Columns (Cytiva, 17085301) to remove free BV or PCB.”
- In lines 546, 547. What is ROI?
We apologize for the reviewer’s confusion. ROI is the abbreviation of “region of interest.” We have included the full expression in the revised manuscript.
- The authors should indicate whether codon optimization was necessary when HO1, PcyA, Fd and Fnr were expressed in yeast or in mammalian cells.
We would like to thank the reviewer for pointing out this issue. The codon optimization for expressing SynPCB genes was needed in mammalian cells (Uda et al., PNAS, 2017). In fission yeast, we used the same SynPCB gene optimized for human codon usage, but not for fission yeast codon usage. As far as we used, there have been no problems with PCB synthesis in fission yeast. We have included this important note in the revised manuscript as follows.
(Page 20, line 483)
“The nucleotide sequence of these genes and SynPCB were optimized for human codon usage (see Benchling link; Table S1). ”
- Figure S1 should show the software and algorithm for the construction of phylogenetic.
Thank you for pointing out the ambiguity in our statement. We have not constructed the species tree on our own; instead, we have manually drawn the tree by Adobe Illustrator, based on the latest genome-scale phylogeny of fungi (Yuanning Li et al. 2021), where results of multiple algorithms were compared to evaluate the reliability of phylogenies. To make it explicit, we revised the manuscript as follows.
(Page 24, line 627)
“We have manually drawn the evolutionary relationship among representative species (Nguyen et al. 2017) based on a recent genome-scale phylogeny (Yuanning Li et al. 2021), which is consistent with the current consensus view of the fungal tree of life (James et al. 2020).”
Here, to validate that the tree in Figure S1 is consistent with the current consensus phylogeny, we have added a reference to a recent review on the fungal tree of life (Timothy Y. James et al., Ann Rev Microbiol, 2020). Additionally, we modified the tree in Figure S1 to represent the remaining ambiguity among subdivisions in Basidiomycota (Yuanning Li et al. 2021, Figure 4G).
- In Figure 1C and Figure 2C. The authors should explain the reasons why the fluorescence of iRFP was lower when yeast was treated with excessive BV and PCB.
The decrease in iRFP intensity under 625 μM PCB or 625 μM BV could be due to cell death and/or toxicity by the excess DMSO. We have included the explanation in the revised manuscript.
Figure 1 legend
“The decrease in iRFP intensity under 625 μM BV could be due to cell death and/or toxicity by the excess DMSO”
Figure 2 legend
“The decrease in iRFP intensity under 625 μM PCB or BV could be due to cell death and/or toxicity by the excess DMSO”
- In Supplementary Information, line 51. What is "teh"?
We have corrected this mistake.
Reviewer #2
**Minor comments:**
- In the paragraph headed "iRFP brightens iRFP more efficiently..." there is mention of "NLS-iRFP-NLS". Please introduce the abbreviation (nuclear localisation sequence?) and, if possible, why the sequence is present N- and C-terminally.
We would like to thank the reviewer for this comment. The abbreviation of NLS (nuclear localization signal) has been already included in the previous version of manuscript (page 5, line 113). Two NLSs are fused with iRFP because the addition of a single NLS does not sufficiently localize the protein at the nucleus. We have included this note in the revised manuscript (page 5, line 114).
- The fluorescence intensity increase upon PCB compared to BV treatment of S. pombe cultures expressing iRFP (Fig. 2C,D) appears about 7-fold, which is far more than the factor of 2 measured on the level of fluorescence quantum yield, and the protein levels were determined to be comparable. Are there any factors conceivable that further enhances the signal of PCB-bound iRFP in S. pombe?
As we have discussed in the previous version of the manuscript, we presume the three factors enhancing fluorescence of iRFP-PCB in fission yeast; (1) the increase in quantum yield (~ 1.61-fold, Fig. 3F), (2) blue-shifted excitation and emission spectrum of iRFP-PCB, which are beneficial for our microscopic setup, and (3) the efficiency of chromophore formation (~ 1.75-fold) (Rumyantsev et al. 2015). During this revision, we calculated to what extent the second factor potentially enhances iRFP-PCB fluorescence compared to iRFP-BV. Based on the emission spectrum (Fig. S3C, left), iRFP-PCB is approximately 1.3-fold more effectively excited by 640 nm of excitation laser than iRFP-BV. Similarly, the detection of iRFP-PCB fluorescence is about 2.0-fold more efficient than that of iRFP-BV with our emission filter (665-705 nm emission filter). Based on these data, the rough estimation yields 1.61*1.3*2*1.75 = 7.3-fold increase, which is comparable with the experimental results showing 5~10-fold increase in iRFP-PCB (Figs. 2C, 2D, 3G). We have included this additional discussion in the revised manuscript as follows:
(page 17, line 429)
“Based on the emission and excitation spectrum (Fig. S3C), iRFP-PCB is approximately 1.3-fold more effectively excited by 640 nm of the excitation laser, and detected about 2.0-fold more efficiently with our emission filter (665-705 nm emission filter) in comparison to iRFP-BV.”
(page 17, line 433)
“Based on these data, the rough estimation yields 1.61*1.3*2*1.75 = 7.3-fold increase, which is comparable with the experimental results showing the 5~10-fold increase in iRFP-PCB fluorescence compared to iRFP-BV (Figs. 2C, 2D, 3G).”
- In the spectral data of Fig. 3D,E, there appears to be a spectrometer artifact at around 450 nm in all spectra, which is not commented on. It is certainly not part of the cofactor / holoprotein spectra.
The peaks around 450 nm in spectra are the artifact of our spectrometer for unknown reasons. We have added an explanation of this artifact in the revised manuscript.
(page 10, line 287,290)
“Of note, there is a spectrometer artifact at around 450 nm in all spectra.”
- Lines 292/296: "Fig. 3H" should read "Fig. 2G".
We have corrected these mistakes.
- The reader may wonder whether it is a common or rather rare phenomenon that the exchange of BV by PCB or some other bilin as (usually covalently bound) cofactor in iRFPs (or, more general, also in the parental phytochromes from different branches of the tree of Life) can be performed. Are there comparable studies available in the literature in addition to the work of Rumyantsev et al. (2015)?
According to the reviewer’s suggestion, we examined the fluorescence of miRFP670 and miRFP703 in fission yeast cells under the conditions of DMSO, BV, or PCB treatment, or SynPCB2.1 expression. We found that the fluorescence intensities of miRFP670 and miRFP703 with PCB treatment or SynPCB2.1 expression showed much higher values than those with BV treatment as observed in iRFP fluorescence in this study (Figure S7). These data indicate that the replacement of BV with PCB is beneficial for enhancing the fluorescence intensity of other iRFPs. We have included the data in the revised manuscript as follows:
(page 12, line 340)
“To explore the generality of the application of PCB and SynPCB2.1 system to other near-infrared fluorescent proteins, we measured fluorescence intensities of miRFP670 and miRFP703, which are derived from a different branch of bacteriophytochrome RpBphP1 (Shcherbakova et al. 2016), in fission yeast treated with BV or PCB or expressing SynPCB2.1 (Fig. S7). The fluorescence intensities of both miRFP670 and miRFP703 were enhanced by the addition of PCB and the expression of SynPCB2.1 compared to the addition of BV in a similar manner iRFP (Fig. S7). From these data, we concluded that PCB biosynthesis by SynPCB2.1 is suitable for imaging with near-infrared fluorescent proteins in fission yeast.”
Reviewer #3
(1) In the Fig. 1D, authors tried to measure the BV incorporation into fission yeast cells. How about the time after 180 min, is it still a increase for the fluorescence?
According to the reviewer’s suggestion, we did the same experiments in Figure 1D for up to 24 hours. As the reviewer expected, iRFP fluorescence still increased gradually for up to 24 hours. We replaced Figure 1D with the new data. We do not have any good idea why iRFP fluorescence gradually increases in the presence of BV.
(2) The authors used the Fig. 3H in the manuscript, but I did not find the H in the Fig. 3.
We apologize for the reviewer’s confusion. We have corrected these mistakes.
(3) The information for the BVRA KO HeLa cells need to be provided.
We have included the information of the BVRA KO HeLa cells in the materials and methods.
(page 21, line 519)
“BVRA KO HeLa cells have been established previously (Uda et al. 2017)”
Description of analyses that authors prefer not to carry out
Reviewer #1
- What is the affinity (KD) of BV and PCB to iRFP? Whether is this related to the fluorescence intensity of iRFP in yeast?
The attachment of BV or PCB to iRFP is an irreversible reaction, i.e., BV or PCB is covalently attached to iRFP. For this reason, it is technically impossible to measure equilibrium dissociation constant (Kd) values to evaluate the affinity of BV or PCB to iRFP.
-
Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.
Learn more at Review Commons
Referee #3
Evidence, reproducibility and clarity
In this manuscript, entitled "Near-infrared imaging in fission yeast by genetically encoded biosynthesis of phycocyanobilin", Sakai and co-workers report that phycocyanobilin (PCB) could function as a brighter chromophore for iRFP than BV, and biosynthesis of PCB allows live-cell imaging with iRFP in the fission yeast. They first found that fission yeast cells did not produce BV and therefore did not show any iRFP fluorescence due to the lack of BV and HO gene. Upon the addition of external BV, the iRFP fluorescence signal could be recovered. In addition, expression of HO1 in fission yeast cells also resulting in the iRFP …
Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.
Learn more at Review Commons
Referee #3
Evidence, reproducibility and clarity
In this manuscript, entitled "Near-infrared imaging in fission yeast by genetically encoded biosynthesis of phycocyanobilin", Sakai and co-workers report that phycocyanobilin (PCB) could function as a brighter chromophore for iRFP than BV, and biosynthesis of PCB allows live-cell imaging with iRFP in the fission yeast. They first found that fission yeast cells did not produce BV and therefore did not show any iRFP fluorescence due to the lack of BV and HO gene. Upon the addition of external BV, the iRFP fluorescence signal could be recovered. In addition, expression of HO1 in fission yeast cells also resulting in the iRFP fluorescence. Next, they found that PCB brightens iRFP more efficiently than BV in fission yeast and in vitro, while the fluorescence excitation and emission spectra were 10 nm blue-shifted in iRFP-PCB compared to iRFP-BV. Finally, they introduced a system named SynPCB2.1 for efficient PCB biosynthesis in fission yeast and concluded that PCB biosynthesis by SynPCB2.1 is ideal for iRFP imaging in fission yeast based on the experiments data. They also developed all-in-one plasmids carrying SynPCB2.1 and iRFP fusion protein genes to image the target proteins in fission yeast. In the final part of this manuscript, PCB was tested as an iRFP chromophore in HeLa cells, however it does not offer significant advantage over BV. Overall, this work provides a system for efficient iRFP imaging in fission yeast. Yet, several issues have been identified and need to be addressed in order to strengthen or even validate some of the conclusions made by the authors.
(1) In the Fig. 1D, authors tried to measure the BV incorporation into fission yeast cells. How about the time after 180 min, is it still a increase for the fluorescence?
(2) The authors used the Fig. 3H in the manuscript, but I did not find the H in the Fig. 3.
(3) The information for the BVRA KO HeLa cells need to be provided. To test the application of PCB as chromophore in mammalian cells, a HO1 gene knock out mammalian cells should be used. The authors may use the all-in-one plasmids carrying SynPCB2.1 and iRFP fusion protein genes to image the target proteins in mammalian cells.
Significance
This work provides new information regarding the chromophore of iRFP. It shows that PCB brightens iRFP more efficiently than BV in fission yeast and in vitro. The all-in-one plasmids system developed in this work supplies a tool for iRFP imaging in the organisms without BV production. Although the iRFP-PCB produces a brighter fluorescence compared with iRFP-BV, the fluorescence excitation and emission spectra were ~15 nm blue-shifted compared to that of iRFP-BV, which might restrict its applications in tissues imaging. As the abundant of BV exist in the mammalian cells, iRFP-PCB does not offer significant advantage over iRFP-BV.
Referee Cross-commenting
All the reviewers gave reasonable suggestions.
-
Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.
Learn more at Review Commons
Referee #2
Evidence, reproducibility and clarity
Summary:
iRFPs have been engineered from the chromophore-binding domains of bacterial phytochromes to expand the wavelength range of genetically-encoded markers for fluorescence imaging and sensing applications into the far-red or near-infrared. In contrast to fluorescent proteins from jellyfish or corals, in which the chromophore is formed autocatalytically, iRFPs require a bilin chromophore for fluorescence such as biliverdin IX alpha (BV), phycocyanobilin (PCB) or phytochromobilin (P[phi]B). While available in some, these molecules may be not available in sufficient amounts in the particular cell type under study, and, therefore, …
Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.
Learn more at Review Commons
Referee #2
Evidence, reproducibility and clarity
Summary:
iRFPs have been engineered from the chromophore-binding domains of bacterial phytochromes to expand the wavelength range of genetically-encoded markers for fluorescence imaging and sensing applications into the far-red or near-infrared. In contrast to fluorescent proteins from jellyfish or corals, in which the chromophore is formed autocatalytically, iRFPs require a bilin chromophore for fluorescence such as biliverdin IX alpha (BV), phycocyanobilin (PCB) or phytochromobilin (P[phi]B). While available in some, these molecules may be not available in sufficient amounts in the particular cell type under study, and, therefore, frequently requires the introduction of at least one further gene such as heme oxygenase (HO, to generate BV from heme) or in addition PcyA (which produces PCB from BV). In this study, the authors show that one of the previously described iRFPs (denoted as iRFP713 in the original studies), which is non-fluorescent upon expression in the fission yeast Schizosaccharomyces pombe due to the lack of an endogenous HO gene, is able to complement with PCB in S. pombe when added externally (to the culture medium) or if produced intracellularly with the help of a plasmid developed for PCB synthesis (termed SynPCB2.1). It is shown that iRFP with PCB bound as cofactor exhibits brighter fluorescence than the BV-bound original iRFP713, which is due to a nearly two-fold increased fluorescence quantum yield as shown by in vitro reconstituted and in vivo generated iRFP with bound PCB. S. pombe cells harbouring the SynPCB2.1 system for PCB synthesis even release ("leak") PCB into the surrounding medium to be taken up by cells not producing PCB. Furthermore, the authors report the generation of a plasmid for C-terminal tagging of a protein of interest by iRFP, novel genome integration vectors and all-in-one plasmids harbouring the genes for PCB synthesis and iRFP-fused marker proteins. The genome integration vectors ensure that stable one-copy integration into the genome occurs at different uncritical loci on each of the chromosomes (different from the common Z-locus), which are in gene-free regions and distant enough for crossing strains, and at sites that do not affect the auxotrophy characteristics of cells. The plasmid system, termed pSKI, harbours elements for propagation in E. coli, constitutive or inducible promoters, a multiple cloning site, a selection marker cassette and homology arms separated by a unique restriction enzyme cutting site for plasmid linearization. The viability of the C-terminal fluorescence tagging system relying on PCB-bound iRFP in S. pombe is shown with 10 different proteins of variant and highly specific intracellular location, which all show the expected cellular distribution pattern by fluorescence imaging. Multi-spectral fluorescence labelling schemes (manifold of five) were successfully tested in S. pombe with four specifically localized proteins harbouring different conventional FP tags, and, in addition, one labeled with iRFP(PCB). Finally, it was tested whether PCB can be exploited to yield a brighter iRFP fluorophore also in a mammalian model cell line, HeLa cells. Here, treatment of cells with externally added BV or PCB produced a similar increase in iRFP fluorescence and also knockout cells devoid of the BV breakdown enzyme BVRA did not show different fluorescence levels upon BV or PCB treatment. Therefore, it is concluded that PCB is applicable to iRFP imaging in mammalian cells though offering no advantage, which to some extent limits the advantages to cells which naturally do not produce BV from heme.
Major comments:
The claim that the developed plasmid system for multi-spectral fluorescence imaging applications in fission yeast S. pombe, which relies on PCB synthesis and incorporation of PCB into iRFP, is fully justified by the data shown in the manuscript. Interpretation and the derived conclusions are put forward in a decent and balanced way. No additional experiments are required to justify the claims.
Minor comments:
- In the paragraph headed "iRFP brightens iRFP more efficiently..." there is mention of "NLS-iRFP-NLS". Please introduce the abbreviation (nuclear localisation sequence?) and, if possible, why the sequence is present N- and C-terminally.
- The fluorescence intensity increase upon PCB compared to BV treatment of S. pombe cultures expressing iRFP (Fig. 2C,D) appears about 7-fold, which is far more than the factor of 2 measured on the level of fluorescence quantum yield, and the protein levels were determined to be comparable. Are there any factors conceivable that further enhances the signal of PCB-bound iRFP in S. pombe?
- In the spectral data of Fig. 3D,E, there appears to be a spectrometer artifact at around 450 nm in all spectra, which is not commented on. It is certainly not part of the cofactor / holoprotein spectra.
- Lines 292/296: "Fig. 3H" should read "Fig. 2G".
- The reader may wonder whether it is a common or rather rare phenomenon that the exchange of BV by PCB or some other bilin as (usually covalently bound) cofactor in iRFPs (or, more general, also in the parental phytochromes from different branches of the tree of Life) can be performed. Are there comparable studies available in the literature in addition to the work of Rumyantsev et al. (2015)?
Significance
The observation of an increased fluorescence quantum yield upon PCB insertion into iRFP is a technical advance with value as such, which will stimulate more detailed studies by spectroscopists (fluorescence, IR, Raman in particular) to clarify the principles underlying the increased quantum yield. While the findings and the developments of this study present a clear advance for imaging applications in fission yeast (available for cell biologists), at least the data on HeLa cells show that the method may not present an advance in terms of delivering a better (i.e. brighter) near-infrared chromophore for mammalian cells. It seems that the ability of a cell line to produce BV (from heme with the help of a HO) limits the potential of PCB addition to iRFP-expressing cells, since BV is readily taken up by the protein shortly after protein biosynthesis. Thus, more elaborate interventions in heme metabolism may be required to fully exploit the potential of the findings in mammalian cell systems and to fulfil the claim that also the optogenetics toolbox may benefit from the findings in the future.
My own expertise relates to spectroscopic analyses (fluorescence, IR, Raman) of iRFP and mutant variants thereof in search of the determinants for increased fluorescence quantum yield, fluorescence lifetime analyses and photoreceptor research.
-
Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.
Learn more at Review Commons
Referee #1
Evidence, reproducibility and clarity
- This paper provided a method for near-infrared imaging using iRFP and proved that phycocyanobilin was better than biliverdin for the imaging in fission yeast.
Major comments:
- What is the affinity (KD) of BV and PCB to iRFP? Whether is this related to the fluorescence intensity of iRFP in yeast?
- The authors thought almost all iRFP forms a holo-complex with BV when HO1 is expressed in fission yeast. This should be proved by calculating the percentage of BV-iRFP in yeast. It is meaningful to compare the percentage of BV-iRFP and PCB-iRFP in vitro and in yeast. -The brightness of fluorescent proteins in organisms often depends on the …
Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.
Learn more at Review Commons
Referee #1
Evidence, reproducibility and clarity
- This paper provided a method for near-infrared imaging using iRFP and proved that phycocyanobilin was better than biliverdin for the imaging in fission yeast.
Major comments:
- What is the affinity (KD) of BV and PCB to iRFP? Whether is this related to the fluorescence intensity of iRFP in yeast?
- The authors thought almost all iRFP forms a holo-complex with BV when HO1 is expressed in fission yeast. This should be proved by calculating the percentage of BV-iRFP in yeast. It is meaningful to compare the percentage of BV-iRFP and PCB-iRFP in vitro and in yeast. -The brightness of fluorescent proteins in organisms often depends on the molecular brightness (fluorescence quantum yield and extinction coefficient) and the amounts of fluorescent proteins. The authors indicated that iRFP-PCB is brighter than iRFP-BV at the molecular level. To calculate the amounts of iRFP-PCB and iRFP-BV when different proteins are expressed in yeast, it is better to explain the results that phycocyanobilin was better than biliverdin for the imaging in fission yeast.
Minor comments:
- In line 35. iRFP is derived from bacteriophytochromes.
- In line 62. The reference of Rodriguez et al. deals with allophycocyanin instead of bacteriophytochromes.
- In lines 65-66. Bacteriophytochromes bind BV, phycocyanin, allophycocyanin and cyanobacterial phytochromes bind PCB, and plantal phytochromes bind PΦB.
- In lines 67-71. The authors described the biosynthesis of BV, PCB and PΦB. But it missed many references.
- In line 270. One full stop should be deleted in "cells.. Fission".
- In line 272. How did the authors add the high concentration of PCB (625 microM) into the culture? PCB is insoluble.
- In line 401. smURFP is derived from allophycocyanin instead of cyanobacteriochromes.
- In line 474. As BV and PCB are insoluble, how do the authors add the pigments into DMEM?
- In line 509. In which solvent are BV and PCB dissolved?
- In lines 546, 547. What is ROI?
- The authors should indicate whether codon optimization was necessary when HO1, PcyA, Fd and Fnr were expressed in yeast or in mammalian cells.
- Figure S1 should show the software and algorithm for the construction of phylogenetic.
- In Figure 1C and Figure 2C. The authors should explain the reasons why the fluorescence of iRFP was lower when yeast was treated with excessive BV and PCB.
- In Supplementary Information, line 51. What is "teh"?
Significance
- near-infrared fluorescent protein broadens the spectrum of fluorescent protein and facilitates deep tissue imaging. Based on iRFP, this work successfully constructed the near-infrared fluorescent protein with PCB as chromophore in fission yeast by using the method of synthesizing PCB, which has been realized in mammalian cells by this group. The method of synthesizing PCB in yeast not only facilitates fluorescence imaging, but also meaningful for optogenetics experiments in yeast. This work is useful for the study of fluorescence imaging, optogenetics and biosynthesis when using yeast as a model organism.
-
-