Edited by:
Reviewed by:
*Correspondence:
This article was submitted to Plant Cell Biology, a section of the journal Frontiers in Plant Science
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
In plants, the actin cytoskeleton, which is composed of filamentous actin (F-actin), plays key roles in organelle transport (
Excellent techniques are available to visualize F-actin organization in fixed plant cells either based on electron microscopy, or on fluorescence microscopy after staining with fluorescently labeled actin antibodies or derivatives of phalloidin, a membrane-permeable fungal metabolite that specifically associates with actin filaments (
Although other interesting markers for live-cell imaging of plant F-actin are available (e.g.,
Here, we present a direct comparison of F-actin labeling, as well as of effects on F-actin-dependent cell expansion, in transiently transformed tobacco pollen tubes expressing YFP-mTn, YFP-FABD2 or Lifeact-YFP fusion proteins at different levels. All fusion proteins tested contained the same yellow fluorescent protein (YFP), which was attached via a flexible 5× Glycine-Alanine linker to one of the three different F-actin binding domains introduced above. cDNAs coding for the different fusion proteins were cloned into the same plasmid vector and expressed under the control of the pollen specific Lat52 promoter (
For several reasons, transient transformation of tobacco pollen tubes is ideally suited to directly compare the performance of different fluorescent
Irrespective of the technique applied, longitudinally oriented F-actin fibers are invariably observed in the pollen tube shank, which appear to mediate myosin-dependent organelle transport (cytoplasmic streaming;
With the study described here, we were pursuing the following key aims:
(1) Rank the performance of the three fluorescent fusion proteins tested (YFP-mTn, YFP-FABD2 and Lifeact-YFP) as non-invasive F-actin marker based on the following two criteria: (a) ability to reveal as much information as possible about F-actin organization in normally growing tobacco pollen tubes, and (b) potential to affect tobacco pollen tube F-actin organization and growth rate at high expression levels.
(2) Enhance our understanding of the organization of the pollen tube actin cytoskeleton through the identification of F-actin structures, which (a) can be non-invasively visualized in normally growing tobacco pollen tubes based on YFP-mTn, YFP-FABD2 or Lifeact-YFP expression, or (b) are only observed when one of these markers is expressed at levels that affect F-action organization and cell expansion.
Results obtained not only help deciding, which of the three markers tested should be preferentially employed for non-invasive F-actin labeling in future experiments, they are also important for the assessment of a large amount of data generated using these markers, which has already been reported in the literature (e.g., the paper by
Standard techniques (
cDNA sequences were constructed coding for Lifeact-YFP, YFP-mTn and YFP-FABD2 fusion proteins composed of eYFP (AAX97736; BD Biosciences-Clontech; San Jose, CA, United States) fused via a flexible 5× Gly-Ala linker to different F-actin binding domains identical to those reported in the original publications introducing the three F-actin markers: (1)
The Lifeact-YFP, YFP-mTn and YFP-FABD2 cDNAs were inserted between the Lat52 promoter (
Tobacco (
Fresh tobacco pollen was plated on solid culture medium (
To acquire low magnification epifluorescence images, 6 h after gene transfer transiently transformed pollen tubes growing on solid culture medium were transferred onto a coverslip by cutting a section of the medium and flipping in upside-down directly on the glass surface as previously described (
To broadly classify the level of YFP or YFP-fusion protein expression in brightly fluorescent pollen tubes, and to determine the length of these cells, micrographs were imported into IMAGEJ (National Institute of Health; Bethesda, MD, United States). Subsequently, the “Image/Adjust/Threshold” function was used to highlight in red color pixels displaying gray levels above an adjustable threshold. Moving this threshold stepwise down from the maximal gray level (256), the brightest pixels (the first ones to turn red) in the image of each pollen tube excluding the pollen grain was identified. It was necessary to exclude the pollen grain from this analysis, because pollen grains often contained parts of the living, transformed protoplast and in this case emitted very bright fluorescence as a consequence of their large diameter. Based on this procedure, transformed pollen tubes were broadly classified into four groups emitting fluorescence at four different levels (
Effects of transient F-actin marker expression at different levels on tobacco pollen tube growth.
To record confocal images of F-actin organization, 6–8 h after gene transfer transiently transformed pollen tubes were transferred onto a coverslip as described above for epifluorescence imaging. Serial confocal sections were imaged in the 6× line averaging mode at a step size of 0.5 μm using a TCS SP5 II laser scanning confocal microscope (Leica Microsystems; Wetzlar, Germany) and an HCX PL APO 63×/1.20 water immersion lens. Fluorescence was excited by a 514 nm argon laser and imaged in the range from 530 to 600 nm. Line-by-line sequential imaging was employed to quasi-simultaneously record confocal images of YFP-FABD2 labeled F-actin organization and transmitted light reference images (Differential Interference Contrast, DIC). Growth rates of individual pollen tubes were determined after confocal imaging of F-actin organization by measuring the distance between the positions of the apex on two images taken in the same focal plane at a 2 min interval.
Microsoft Excel (Microsoft Corporation; Redmond, WA, United States) was employed to compute the average and the 95% confidence interval of individual sets of pollen tube length or growth rate measurements; as well as to assess the significance of differences between such data sets based on one-way ANOVA
A direct comparison was performed of effects of high-level expression of either free YFP, or of one of the F-actin markers Lifeact-YFP, YFP-mTn or YFP-FABD2, on the growth of tobacco pollen tubes transiently transformed by particle bombardment. In these experiments, the Lat52 promoter was used to drive transgene expression, which unlike the 35S promoter is highly active in pollen tubes (
Analysis of the average length of pollen tubes emitting fluorescence at different levels as defined above demonstrated that high-level expression of all marker proteins tested inhibited pollen tube growth in a dose-dependent manner (
In addition, the percentage of all YFP, Lifeact-YFP or YFP-mTn expressing pollen tubes analyzed to generate the graph shown in
Together, data shown in
Six to eight hours after gene transfer, F-actin organization was visualized using confocal microscopy in morphologically unaffected pollen tubes (regular diameter, normal cytoplasmic organization), which expressed the different F-actin markers at the lowest detectable levels and emitted the weakest fluorescence among all transformed pollen tubes observed (
F-actin organization as visualized by confocal microscopy in normally growing tobacco pollen tubes expressing Lifeact-YFP at low level. Serial confocal sections taken at a step size of 0.5 μm (6× line averaging) are displayed. The pollen tube shown was growing at a rate of 6.2 μm/min after confocal imaging. 0.0 μm: first image of the series showing the cell cortex closest to the microscope lens. Arrows: helical arrangement of cortical F-actin; asterisks: subapical F-actin fringe. Scale bar: 10 μm.
F-actin organization as visualized by confocal microscopy in normally growing tobacco pollen tubes expressing YFP-mTn at low level. Serial confocal sections taken at a step size of 0.5 μm (6× line averaging) are displayed. The pollen tube shown was growing at a rate of 5.5 μm/min after confocal imaging. 0.0 μm: first image of the series showing the cell cortex closest to the microscope lens. Arrows: helical arrangement of cortical F-actin; asterisks: subapical F-actin fringe; arrowheads (open and closed): actin fibers apparently extending from the longitudinally oriented F-actin network in the shank to the subapical F-actin fringe, which form a “fork-like subapical structure” connected to the fringe either directly (–3.5 μm; closed arrowheads) or indirectly (–1.0 μm; open arrowheads) via extensions in the extreme cortex (–1.5 μm; open arrowheads). Scale bar: 10 μm.
F-actin organization as visualized by confocal microscopy in tobacco pollen tubes expressing YFP-FABD2 at low level and growing at reduced rates. Serial confocal sections taken at a step size of 0.5 μm (6× line averaging) are displayed, each along with a simultaneously recorded transmitted light reference image (Differential Interference Contrast, DIC; displayed underneath the corresponding confocal section) to illustrate cell outline at the time of imaging. The pollen tube shown was growing at a rate of 3.66 μm/min after confocal imaging. 0.0 μm: first image of the series showing the cell cortex closest to the microscope lens. Brackets in reference images: region largely devoid of F-actin labeling covering the first ca. 15 μm at the tip. Scale bar: 10 μm.
Serial confocal optical sectioning showed that YFP was diffusely distributed throughout the pollen tube cytoplasm (Supplementary Figure
F-actin organization as observed by confocal imaging of pollen tubes expressing YFP-FABD2 at lowest detectable levels (
Confocal imaging of brightly fluorescent pollen tubes displaying morphological aberrations (irregular diameter, disturbed cytoplasmic organization), which expressed Lifeact-YFP or YFP-mTn at high levels that inhibited cell expansion (
F-actin organization as visualized by confocal microscopy in tobacco pollen tubes expressing the indicated F-actin markers at high levels. Single sections
As discussed above, YFP-FABD2 interferes with the growth of pollen tubes already at minimal detectable expression levels (
When expressed at low levels, Lifeact-YFP and YFP-mTn consistently label the same F-actin structures in tobacco pollen tubes: (a) a network of primarily longitudinally oriented fine actin fibers in the shank, which are helically arranged in the cell cortex, and (b) a subapical F-actin fringe (
In addition to the F-actin structures described in the previous paragraph, which are commonly observed in pollen tubes of different species using
More recently, the dynamic behavior of Lifeact-GFP labeled actin fibers was investigated at the extreme apex of growing Arabidopsis pollen tubes (
As discussed above, at low expression levels Lifeact-YFP and YFP-mTn non-invasively and detectably label the same F-actin structures in normally growing tobacco pollen tubes, which can also be observed using other techniques in pollen tubes of different species. This establishes that both markers are excellent tools to investigate pollen tube F-actin structures and dynamics
To visualize the actin cytoskeleton,
Our results demonstrate that Lifeact-YFP clearly is the marker of choice for F-actin visualization in tobacco pollen tubes, as it has a somewhat lower tendency than YFP-mTn to affect actin organization, and strictly F-actin-dependent cell expansion, when expressed in these cells. However, YFP-mTn works almost equally well and results obtained with this marker are certainly perfectly valid, if care is taken to ensure that only normally growing pollen tubes are analyzed. Interestingly, YFP-mTn appears to label the subapical F-actin fringe slightly more effectively than Lifeact-YFP. Stronger interaction with this F-actin structure, which appears to be absolutely essential for cell expansion (
As it is the case for most, if not all, GFP-based makers, when expressed above a marker specific threshold level Lifeact-YFP and YFP-mTn affect their target structure and therefore stop functioning as non-invasive markers. Because pollen tube growth is extremely sensitive to perturbation of the actin cytoskeleton, the growth rate of these cells can be seen as a suitable indicator of the threshold expression level, above which
In most cells types other that pollen tubes, or when structures with less critical functions than the pollen tube actin cytoskeleton are observed, it clearly is more difficult to define an expression level threshold for non-invasive labeling by GFP-based markers. In such cases, it appears advisable to image cells that express GFP-based markers for F-actin or other structures at the lowest detectable expression level, possibly under the control of promoters whose activities have been selected for best performance (
At high expression levels, at which pollen tube growth rates are reduced, both Lifeact-YFP and YFP-mTn induced actin bundling and reorganization. Whereas a subapical actin fringe was often not detected under these conditions, cytoplasmic actin fibers generally appeared to be thicker, more abundant as well as more randomly oriented, and frequently extended from the shank all the way to the extreme apex. As discussed above, we believe F-actin structures exclusively visible in tobacco pollen tubes expressing Lifeact-YFP or YFP-mTn at levels at which tip growth is inhibited have to be considered structural artifacts. Interestingly, tobacco pollen tubes growing at reduced rates because of high-level Lifeact-YFP or YFP-mTn expression often contained small cytoplasmic F-actin rings. Similar F-actin rings have previously been observed in different types of plants cells, which expressed GFP-based F-actin markers (including Lifeact-YFP and YFP-mTn) or were labeled with phalloidin-derived fluorescent probes (
By contrast to Lifeact-YFP and YFP-mTn, YFP-FABD2 is not suitable as an
Interestingly, GFP-FABD2 fusion proteins were proposed to perform better than GFP-mTn as F-actin markers in transgenic Arabidopsis plants, based on comparing F-actin structures in different cell types labeled by stable expression of the two markers under the control of the constitutively active CaMV 35S promoter (
(1) At low expression levels, Lifeact-YFP and YFP-mTn non-invasively label the same key components of the actin cytoskeleton in normally growing tobacco pollen tubes: (a) a network of primarily longitudinally oriented actin fibers in the shank, (b) a subapical F-actin fringe and, less effectively, (c) actin fibers connecting these two structures.
(2) At high expression levels, Lifeact-YFP, YFP-mTn and YFP-FABD2 inhibit tobacco pollen tube growth, stimulate actin filament bundling and induce F-actin reorganization. High-level Lifeact-YFP or YFP-mTn expression disrupts F-actin structures observed in normally growing pollen tubes (i.e., the subapical F-actin fringe) and induces the formation of structural artifacts including actin fibers in the apical cytoplasm and small F-actin rings.
(3) Lifeact-YFP is the marker of choice for non-invasive F-actin visualization in tobacco pollen tubes, as it has a somewhat lower potential than YFP-mTn to generate diffuse cytoplasmic background, to alter F-actin organization and to interfere with tip growth. YFP-FABD2 is not suitable as an F-actin marker in tobacco pollen tubes, because it affects F-actin organization and tip growth already at the lowest detectable expression levels, and fails to label parts of the actin cytoskeleton.
(4) With caution, key results of our analysis of the performance of Lifeact-YFP, YFP-mTn and YFP-FABD2 in tobacco pollen tubes can probably be extrapolated to other cell types. When expressed at high levels, these three markers, as well as presumably all other available GFP-based
AM-R acquired all data and contributed to the design of the work, to data analysis as well to the writing of the MS. BK conceived the work, was responsible for data analysis and interpretation, and wrote the final version of the MS.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
The Supplementary Material for this article can be found online at: