(C) PLOS One
This story was originally published by PLOS One and is unaltered.
. . . . . . . . . .
Activation of actin-depolymerizing factor by CDPK16-mediated phosphorylation promotes actin turnover in Arabidopsis pollen tubes [1]
['Qiannan Wang', 'Center For Plant Biology', 'School Of Life Sciences', 'Tsinghua University', 'Beijing', 'Yanan Xu', 'Shuangshuang Zhao', 'State Key Laboratory Of Plant Physiology', 'Biochemistry', 'College Of Biological Sciences']
Date: 2023-04
As the stimulus-responsive mediator of actin dynamics, actin-depolymerizing factor (ADF)/cofilin is subject to tight regulation. It is well known that kinase-mediated phosphorylation inactivates ADF/cofilin. Here, however, we found that the activity of Arabidopsis ADF7 is enhanced by CDPK16-mediated phosphorylation. We found that CDPK16 interacts with ADF7 both in vitro and in vivo, and it enhances ADF7-mediated actin depolymerization and severing in vitro in a calcium-dependent manner. Accordingly, the rate of actin turnover is reduced in cdpk16 pollen and the amount of actin filaments increases significantly at the tip of cdpk16 pollen tubes. CDPK16 phosphorylates ADF7 at Serine128 both in vitro and in vivo, and the phospho-mimetic mutant ADF7 S128D has enhanced actin-depolymerizing activity compared to ADF7. Strikingly, we found that failure in the phosphorylation of ADF7 at Ser128 impairs its function in promoting actin turnover in vivo, which suggests that this phospho-regulation mechanism is biologically significant. Thus, we reveal that CDPK16-mediated phosphorylation up-regulates ADF7 to promote actin turnover in pollen.
Here, we report that CDPK16 up-regulates the activity of pollen-specific Arabidopsis ADF7 by phosphorylating it at Ser128. Our finding differs from the consensus view that kinase-mediated phosphorylation inactivates ADF/cofilin. Our study thus significantly enhances our understanding of the regulation of ADF/cofilin. In addition, our findings provide the direct evidence linking Ca 2+ /CDPK signaling to actin dynamics during pollen tube growth and suggest a scenario in which high Ca 2+ -CDPK activity can be unified with ADF7 activation to maintain the high dynamics of actin filaments at pollen tube tips. Our study thus sheds light on the functional adaptation of ADF/cofilin and the regulation of actin dynamics during polarized pollen tube growth.
Actin-depolymerizing factor (ADF)/cofilin, a central regulator of actin dynamics, is a stimulus-responsive protein [ 8 ] which has been implicated in the regulation of actin dynamics in pollen tubes [ 9 – 13 ]. The activity of ADF/cofilin is subject to tight regulation by kinase-mediated phosphorylation [ 14 – 17 ], pH [ 18 – 22 ], and interactions with other binding partners, such as CAP1 [ 23 , 24 ], AIP1 [ 25 – 27 ], and Coronin [ 28 ], etc. Kinase-mediated phosphorylation represents a well-known inactivation mechanism of ADF/cofilin, with the phosphorylation occurring at Ser3 or Ser6 of ADFs/cofilins from different organisms [ 8 , 29 ]. This type of phospho-regulation has been reported for plant ADFs [ 30 , 31 ]. Specifically, plant ADFs were demonstrated or proposed to be inactivated by calcium (Ca 2+ )-dependent protein kinase (CDPK)-mediated phosphorylation [ 29 , 31 , 32 ], but the biological significance of this regulatory mechanism remains to be explored. In particular, it remains a mystery how ADF contributes to the enhancement of actin dynamics at pollen tube tips where CDPK(s) is supposed to be active.
The actin cytoskeleton is a dynamic and signaling-responsive structure that has been implicated in numerous physiological cellular processes including cytokinesis, cell migration, polarized cell growth, and various intracellular trafficking events [ 1 ]. A dynamic actin cytoskeleton is absolutely required for polarized growth of pollen tubes [ 2 , 3 ], which provide passage for 2 non-motile sperm cells to ensure double fertilization in flowering plants. Previous studies revealed that there exists a distinct population of extremely dynamic actin filaments in pollen tubes [ 2 – 5 ]. These filaments assume a distinct spatial distribution and form a unique structure called the “apical actin structure” at pollen tube tips [ 6 , 7 ]. It is fascinating to consider how pollen tubes control the dynamics and spatial distribution of apical actin filaments in response to various signals during the rapid extension of pollen tubes.
( A ) Intracellular localization of CDPK16-eGFP in pollen grains during germination. Yellow triangles indicate the germination aperture, and white circles indicate the growth direction of the germinating pollen tube. Bar = 10 μm. ( B ) Intracellular localization of CDPK16-eGFP in pollen tubes. The upper panel is the time-lapse images showing the intracellular localization of CDPK16-eGFP in growing pollen tubes. Bar = 10 μm. The lower panel shows transverse sections at 240 s. Bar = 5 μm. Asterisks indicate the growth direction of the pollen tube, and red brackets indicate the apical region with less CDPK16-eGFP signals. The dashed yellow lines indicate the base of the subapical region. The red triangles indicate the membrane of the vegetative nucleus and the yellow triangles indicate the PM. The distance of transverse sections from the tip is indicated above the images. ( C ) Covisualization of CDPK16-eGFP with FM4-64 in the pollen tube. FM4-64 labels the PM but not internal membranes. The right panel shows transverse sections with their distance from the tip indicated in the images. White triangles indicate the PM and blue triangles indicate the membrane of the vegetative nucleus. Bars = 10 μm. CDPK, calcium-dependent protein kinase; PM, plasma membrane.
To determine the intracellular localization of CDPK16 in pollen tubes, we generated a CDPK16-eGFP fusion construct with its expression under the control of the CDPK16 native promoter. CDPK16-eGFP can rescue the LatB-resistant pollen germination phenotype of cdpk16 mutants ( S13 Fig ), which suggests that CDPK16-eGFP is functional. We found that CDPK16 is mainly localized in an inner circle close to the border of ungerminated pollen grains ( Fig 6A ). During the onset of pollen germination, the CDPK16-eGFP signal is reduced at the germination aperture and is subsequently enriched in the region near the PM of the emerging pollen tube ( Fig 6A and S9 Movie). We also examined the intracellular localization of CDPK16-eGFP in late-stage pollen tubes and found that it mainly localized to the PM, with the strongest signal in the subapical region ( Fig 6B and S10 Movie). CDPK16-eGFP also forms small dots within the cytoplasm of pollen tubes ( Fig 6B ). We do not currently know what those structures are. The PM localization of CDPK16-eGFP was also confirmed by covisualization of the fluorescent lipophilic dye FM4-64. CDPK16-eGFP colocalized with FM4-64, and the eGFP signal was obviously stronger at the subapical region of the pollen tube ( Fig 6C ). We also found that CDPK16-eGFP is localized to the nucleus ( Fig 6B and 6C ). Our intracellular localization data are actually consistent with a previous finding that CDPK localizes to membranes via myristoylation and palmitoylation [ 35 ].
( A ) Images of actin filaments in pollen tubes. Pollen tubes derived from adf10, proADF7::gADF7; adf7 adf10, proADF7::gADF7 S128A ; adf7 adf10 and proADF7::gADF7 S128D ; adf7 adf10 were subjected to actin staining with Alexa-488 phalloidin. The projection images and the associated optical sections are presented. Several transverse sections derived from the same pollen tube are shown in the right panels; the distance of these sections from the pollen tube tip is indicated in the images. The dashed red lines indicate the base of the apical region (3 μm from tube tip), and the dashed yellow lines indicate the base of the subapical region (10 μm from tube tip). Bars = 5 μm. ( B ) The 3D distribution of fluorescence intensity of actin filaments within the apical and subapical region (0–10 μm from tube tip), which was generated by ImageJ software with a 3D interactive “Surface Plot” function. Warm and cold colors indicate higher and lower fluorescence, respectively. Numerical data underlying this panel are available in S5 Data . ( C ) Quantification of the fluorescence intensity of actin filaments stained with Alexa-488 phalloidin in pollen tubes. The fluorescence intensity (mean ± SEM) was plotted from the tip to the base along the growth axis of pollen tubes. The dashed red lines indicate the base of the apical region (3 μm from tube tip), and the dashed yellow lines indicate the base of the subapical region (10 μm from tube tip). More than 40 pollen tubes were measured. Data are presented as mean ± SEM. Numerical data underlying this panel are available in S5 Data . ( D–F ) Quantification of the fluorescence intensity of Alexa-488 phalloidin within 3 different regions of pollen tubes. Data are presented as mean ± SEM, n = 3, ns, no significant difference, *P < 0.05, **P < 0.01 by Student’s t test. Numerical data underlying this panel are available in S5 Data . ADF, actin-depolymerizing factor.
To directly visualize the effect of phosphorylation of ADF7 at its Ser128 on the actin cytoskeleton in pollen tubes, we directly visualized the actin cytoskeleton in pollen tubes after staining with Alexa-488 phalloidin. We found that actin filaments were overall brighter in proADF7::gADF7 S128A ; adf7 adf10 and proADF7::gADF7 S128D ; adf7 adf10 pollen tubes than in adf10 and proADF7::gADF7; adf7 adf10 pollen tubes ( Fig 5A ), and the increase in the amount of actin filaments is much more obvious within the apical and subapical regions of pollen tubes ( Fig 5A and 5B ). Accordingly, the average fluorescence intensity of phalloidin staining within the apical and subapical regions of proADF7::gADF7 S128A ; adf7 adf10 and proADF7::gADF7 S128D ; adf7 adf10 pollen tubes is significantly higher than in adf10 and proADF7::gADF7; adf7 adf10 pollen tubes ( Fig 5C–5E ), whereas no significant difference was detected in the shank region ( Fig 5C and 5F ). The actin cytoskeleton exhibits severe turnover defects in the apical and subapical regions but not in the shank region of pollen tubes harboring ADF7 S128A and ADF7 S128D , which suggests that phosphorylation of ADF7 at Ser128 has biologically meaningful consequence within apical and subapical regions that have high concentration of cytosolic Ca 2+ . This, on the other hand, suggests that CDPK16-mediated regulation of ADF7 is biological significant in pollen tubes.
Next, we determined the biological significance of this phospho-regulation mechanism by introducing the non-phosphorylatable ADF7 S128A and phospho-mimetic ADF7 S128D into pollen. Considering that ADF7 and ADF10 redundantly regulate actin turnover in pollen [ 11 ], we assumed that the functional difference between ADF7 S128A and ADF7 or ADF7 S128D and ADF7 might be amplified in the adf10 mutant background. We selected transgenic lines containing comparable amounts of ADF7 S128A , ADF7 S128D , and ADF7 both transcriptionally ( S11A Fig ) and translationally ( S11B and S11C Fig ) for subsequent analyses. We initially found that ADF7 S128A functions almost the same as ADF7 in supporting pollen tube growth, whereas ADF7 S128D has slightly but significantly higher activity than ADF7 and ADF7 S128A in this regard ( S11D and S11E Fig ). However, in the presence of LatB, we found that pollen tubes harboring both ADF7 S128A and ADF7 S128D grew significantly faster than pollen tubes harboring ADF7 ( S11D and S11F Fig ), which suggests that ADF7 S128A and ADF7 S128D have reduced activity in promoting actin turnover in pollen tubes compared to ADF7. It is quite puzzling that ADF7 S128D has reduced capability in promoting actin turnover in pollen compared to ADF7, as ADF7 S128D has higher activity in depolymerizing and severing actin filaments than ADF7 in vitro ( Fig 4F–4J ). The reasonable explanation here is that ADF7 S128D cannot fully mimic the function of phosphorylated ADF7 in pollen. Strikingly, we found that the germination of pollen harboring ADF7 S128A is resistant to LatB compared to pollen harboring WT ADF7 when CDPK16 is overexpressed ( S12 Fig ). This suggests that phosphorylation of ADF7 at its Ser128 mainly accounts for the role of CDPK16 in promoting actin turnover in pollen.
We next examined the actin-depolymerizing activity of Ser128 mutants of ADF7. Compared to WT ADF7, ADF7 S128D had enhanced actin-depolymerizing activity and ADF7 S128A had roughly similar activity, as determined by the high-speed F-actin co-sedimentation assay ( Fig 4F and 4G ) and the kinetic actin-depolymerizing assay ( Fig 4H ). This was further confirmed by direct visualization of actin filaments ( Fig 4I and 4J ). Importantly, we found that CDPK16 failed to enhance the actin-depolymerizing activity of ADF7 S128A in vitro ( S10 Fig ), which suggests that Ser128 of ADF7 is the major residue targeted by CDPK16. These data together suggest that Ser128 in ADF7 is phosphorylated by CDPK16, and phosphorylation of this residue increases ADF7 activity.
( A ) Identification of a phosphorylated peptide with the phosphate group conjugated to Ser128. 8His-ADF7 protein was isolated from pollen grains derived from proADF7::8His-gADF7; adf7 and subjected to LC-MS analysis. The original LC-MS data underlying this panel are available in S4 Data . The original pictures are available in S1 Raw Images. ( B ) CDPK16 can phosphorylate Ser128 in ADF7 in vitro. After incubation of 20 μm ADF7 with 5 μm CDPK16 in kinase buffer for 30 min, the sample was separated by SDS-PAGE. The ADF7 band was cut out and subjected to mass spectrometry analysis. A phosphorylated ADF peptide with the phosphate group conjugated to Ser128 was identified. The original LC-MS data underlying this panel are available in S4 Data . The original pictures are available in S1 Raw Images. ( C ) CDPK16 fails to phosphorylate ADF7 S128D -6×His. Purified ADF7-6×His and ADF7 S128D -6×His were incubated with CDPK16 in a kinase reaction buffer and the phosphorylation signals were detected by [γ- 32 P] ATP autoradiography. The original pictures are available in S1 Raw Images. ( D ) Quantification of the relative phosphorylation level of ADF7 shown in ( C ). The data are presented as mean ± SE, n = 3, **P < 0.01 (Student’s t test). Numerical data underlying this panel are available in S4 Data . ( E ) 2D electrophoresis assay. Total proteins from mature pollen of WT, proADF7::gADF7 S128A ; adf7 and proADF7::gADF7 S128D ; adf7 were subjected to 2D electrophoresis analysis. Immunoblotting was performed and probed with anti-ADF7 antibody. (a) Indicates the presumed phosphorylated ADF7 or ADF7 S128D and (b) indicates unphosphorylated ADF7 or ADF7 S128A . The original pictures are available in S1 Raw Images. ( F ) SDS-PAGE analysis of protein samples from a high-speed F-actin co-sedimentation experiment. F-actin, 3 μm; ADF7, 20 μm; ADF7 S128A , 20 μm; ADF7 S128D , 20 μm. The original pictures are available in S1 Raw Images. ( G ) Quantification of the amount of actin in the supernatant fractions in ( F ). Data are presented as mean ± SE, n = 3, ns, no significant difference, *P < 0.05 by Student’s t test. Numerical data underlying this panel are available in S4 Data . ( H ) Kinetic actin filament depolymerization assay. Actin filaments were depolymerized more rapidly by ADF7 S128D than by ADF7 and ADF7 S128A . Briefly, 5 μm pre-clarified ADF7, ADF7 S128A , or ADF7 S128D were incubated with 5 μm preassembled actin filaments (50% NBD-labeled) for 2 min. The mixtures were subsequently diluted 25-fold into buffer G and actin depolymerization was monitored by measuring the changes in fluorescence. Numerical data underlying this panel are available in S4 Data . ( I ) Images of actin filaments stained with Rhodamine-Phalloidin. The concentration of preassembled actin filaments is 2 μm. Bar = 10 μm. ( J ) Quantification of the average length of actin filaments shown in ( I ). The average length of actin filaments was plotted, and the data are presented as mean ± SE, n = 3, ns, no significant difference, **P < 0.01 (Student’s t test). Numerical data underlying this panel are available in S4 Data . ADF, actin-depolymerizing factor; CDPK, calcium-dependent protein kinase; LC-MS, liquid chromatography–mass spectrometry; WT, wild type.
To uncover the phosphorylation site(s) of ADF7 in vivo, we performed mass spectrometry analysis on 8His-ADF7 pulled down from total proteins extracted from pollen. An ADF7 phospho-peptide was repeatedly identified with the phosphate group conjugated to Ser128 ( Fig 4A ). Strikingly, we found that CDPK16 adds phosphate to the same serine of ADF7 in vitro ( Fig 4B ). This suggests that Ser128 in ADF7 might be the major site of phosphorylation by CDPK16. In support of this notion, we found that the extent of CDPK16-mediated ADF7 phosphorylation in vitro was reduced significantly after Ser128 was replaced with aspartic acid (ADF7 S128D ) ( Fig 4C and 4D ). Furthermore, we found that the charge behavior of ADF7 S128D and non-phosphorylatable ADF7 S128A (Ser128 replaced with Alanine) is similar to that of the predicted phosphorylated ADF7 and non-phosphorylated ADF7 ( Fig 4E ), respectively. Ser128 is conserved in all the class II ADFs in Arabidopsis, i.e., ADF7, ADF8, ADF10, and ADF11 ( S7 Fig ) and in class II ADFs from other plant species ( S8 Fig ). We next generated a poly-clonal antibody against this phospho-peptide, designated as anti-phospho-ADF7(Ser128), and found that it specifically recognizes CDPK16-phosphorylated ADF7 ( S9A Fig ). Interestingly, we found that CDPK16 phosphorylates ADF7 in a Ca 2+ -dependent manner ( S9B and S9C Fig ). Importantly, we found that it recognizes 8His-ADF7 pulled down from total pollen extract, and treatment with λ-phosphatase reduced the signal ( S9D Fig ), which further suggests that phosphorylation of ADF7 at Ser128 does occur in vivo.
To determine whether CDPK16 promotes actin turnover through activating ADF7, we tested whether gain of function of ADF7 can alleviate the actin turnover defects in cdpk16 mutant pollen. Indeed, we found that overexpression of ADF7 suppressed the actin turnover phenotype in cdpk16 mutant pollen ( S6A and S6B Fig ). ADF7 and ADF10 act redundantly to control actin turnover in pollen [ 11 ], and our results above suggest that ADF7 might be the more relevant substrate of CDPK16 in vivo ( Fig 2F and 2G ). Therefore, we wondered whether loss of function of CDPK16 will cause an additive effect on actin turnover induced by loss of function of ADF10 in pollen. As expected, we found that the LatB-resistant pollen germination phenotype is more severe in adf10 cdpk16-1 double mutants compared to cdpk16-1 or adf10 single mutants ( S6C and S6D Fig ). These data together suggest that CDPK16 promotes actin turnover at least partly through up-regulating ADF7 activity in pollen.
( A ) SDS-PAGE analysis of the protein samples from a high-speed F-actin co-sedimentation experiment in the presence of Ca 2+ . F-actin, 3 μm; ADF7, 20 μm; CDPK16 (+), 1.0 μm; CDPK16 (++), 2.5 μm; CDPK16 (+++), 5.0 μm. The supernatant fractions (S) and pellets (P) were separated on SDS-PAGE gels, and proteins were detected by Coomassie Brilliant blue R 250 staining. The original pictures are available in S1 Raw Images. ( B ) Quantification of the amount of actin in the supernatant fractions shown in ( A ). Data are presented as mean ± SE, n = 3, *P < 0.05 and **P < 0.01 by Student’s t test. Numerical data underlying this panel are available in S3 Data . ( C ) SDS-PAGE analysis of the protein samples from a high-speed F-actin co-sedimentation experiment in the absence of Ca 2+ . The conditions were exactly the same as in ( A ) except for the presence of 0.5 mM EGTA instead of 0.5 mM CaCl 2 in the kinase reaction buffer. F-actin, 3 μm; ADF7, 20 μm; CDPK16 (+), 1.0 μm; CDPK16 (++), 2.5 μm; CDPK16 (+++), 5.0 μm. The original pictures are available in S1 Raw Images. ( D ) Quantification of the amount of actin in the supernatant fractions shown in ( C ). Data are presented as mean ± SE, n = 3, ns, no significant difference by Student’s t test. Numerical data underlying this panel are available in S3 Data . ( E ) Images of actin filaments stained with equimolar Rhodamine-Phalloidin. The concentration of actin filaments is 2 μm. Bar = 10 μm. ( F ) Quantification of the length of actin filaments. Data are presented as mean ± SE, n = 3, ns, no significant difference, **P < 0.01 by Student’s t test. Numerical data underlying this panel are available in S3 Data . ( G ) Time-lapse images of actin filaments. F-actin, 150 nM (50% Oregon green-labeled); ADF7, 500 nM; CDPK16, 125 nM. The red arrows indicate actin filament severing events. Bar = 10 μm. ( H ) Quantification of the average severing frequency of actin filaments. Data are presented as mean ± SE, ns, no significant difference, *P < 0.05 by Student’s t test. Numerical data underlying this panel are available in S3 Data . ADF, actin-depolymerizing factor; CDPK, calcium-dependent protein kinase.
Next, using a high-speed F-actin co-sedimentation assay, we found that CDPK16 increased the activity of ADF7 in depolymerizing actin filaments in vitro ( Fig 3A and 3B ), and this enhancement by CDPK16 was reduced in the absence of Ca 2+ ( Fig 3C and 3D ). These results suggest that CDPK16 promotes the actin-depolymerizing activity of ADF7 in a Ca 2+ -dependent manner. We previously showed that the actin disassembly activity of ADF4 is regulated by phosphorylation in stomata [ 30 ]. Interestingly, we found that CDPK16 only weakly, albeit significantly, enhanced the actin-depolymerizing activity of ADF4 ( S5 Fig ), which suggests that CDPK16 promotes the actin-depolymerizing activity of ADF7 in a somewhat specific manner. In addition, we found that CDPK16 enhanced the activity of ADF7 in shortening actin filaments in vitro ( Fig 3E and 3F ). As ADF/cofilin also contributes to actin depolymerization via fragmentation of actin filaments [ 34 ] and ADF7 can sever actin filaments [ 10 ], we determined whether CDPK16 can promote the severing activity of ADF7. We used total internal reflection fluorescence microscopy (TIRFM) to show that CDPK16 enhanced the activity of ADF7 in severing actin filaments ( Fig 3G and 3H and S5 – S8 Movies ). Thus, our study suggests that CDPK16 enhances ADF7-mediated actin depolymerization and severing in vitro.
( A ) Identification of proteins in CDPK16-6×His pull-down fractions by LC-MS/MS. ADF7 is one of the proteins that appears in the pull-down fraction. ( B , C ) Two representative peptides derived from ADF7 are presented. A full list of the peptides is presented in S2 Data . ( D ) CDPK16 interacts with ADF7. The interaction between ADF7 and CDPK16 was determined by qualitative analysis of luciferase (LUC) activity using the LCI assay. ( E ) CDPK16 phosphorylates ADF7 in vitro. The left panel shows the CDPK16-6×His and ADF7 input proteins, which were detected by Coomassie Brilliant blue R 250 staining. The right panel shows phosphorylated ADF7, which was detected by [γ- 32 P] ATP autoradiography. The original pictures are available in S1 Raw Images. ( F ) Detection of ADF7 and ADF10 protein spots by 2D gel-electrophoresis. Total proteins from WT, adf7 and adf10 pollen were separated by 2D gel-electrophoresis. Protein spots were revealed by western blot analysis probed with anti-ADF7 antibody, which also detects ADF10. The original pictures are available in S1 Raw Images. ( G ) Detection of ADF7 and ADF10 in total proteins extracted from WT and cdpk16 pollen. Total proteins from WT pollen in the presence and absence of λpp or from pollen of cdpk16 mutants and CDPK16 overexpressors were separated by 2D gel-electrophoresis and subjected to western blot analysis probed with anti-ADF7 antibody. (a), (b), and (c) in ( F ) and ( G ) represent phosphorylated ADF7, ADF7, and ADF10, respectively. The original pictures are available in S1 Raw Images. ADF, actin-depolymerizing factor; LCI, luciferase complementation imaging; LC-MS/MS, liquid chromatography–mass spectrometry/mass spectrometry; WT, wild type.
To determine whether CDPK16 regulates actin turnover via interaction with certain actin-binding proteins (ABPs) in vivo, we performed pull-down experiments followed by mass spectrometry to search for candidate interacting proteins of CDPK16. Interestingly, we found that ADF7 is enriched in the CDPK16 pull-down fraction ( Fig 2A–2C ). The direct interaction between CDPK16 and ADF7 was confirmed by the luciferase complementation imaging (LCI) assay ( Fig 2D ) and further validated by showing that CDPK16 can phosphorylate ADF7 in vitro ( Fig 2E ). To determine whether CDPK16 can phosphorylate ADF7 in vivo, we decided to treat total pollen proteins with phosphatase in the presence and absence of CDPK16, followed by 2D electrophoresis and antibody detection of ADF7. However, the currently available anti-ADF7 antibody cross-reacts with the highly similar ADF10, so we initially analyzed adf7 and adf10 mutants to distinguish ADF7 from ADF10 after electrophoresis ( Fig 2F ). Comparing the results from WT, adf7 and adf10, it appears that ADF7 might be subject to posttranslational modification, as there are several protein spots corresponding to ADF7 ( Fig 2F ). Protein spot (a) is the presumed phosphorylated form of ADF7, based on its mobility after 2D electrophoresis ( Fig 2F ). In support of this speculation, we found that treatment with λ-phosphatase reduced the intensity of spot (a) in total proteins extracted from WT pollen ( Fig 2G ). Furthermore, we found that the intensity of protein spot (a) is reduced in cdpk16 mutant pollen total extract whereas is increased in pollen total extract from CDPK16 overexpressors compared to WT ( Fig 2G ), which suggests that CDPK16 can phosphorylate ADF7 in pollen.
( A ) Time-lapse images of actin filaments decorated with Lifeact-eGFP in growing WT and cdpk16-1 pollen tubes. Yellow brackets indicate the region occupied by membrane-originated actin filaments. The right panels are Kymograph analyses of the growing WT and cdpk16-1 pollen tubes shown in the left panel. The region occupied by the membrane-originated actin filaments is marked by 2 red lines. Bar = 10 μm. ( B ) Quantification of the width of the region occupied by membrane-originated actin filaments in WT and cdpk16-1 pollen tubes. Data are presented as mean ± SE, **P < 0.01 (Student’s t test). More than 400 time points from 10 pollen tubes were measured. Numerical data underlying this panel are available in S1 Data . ( C ) Time-lapse images of actin filaments at the apical region in WT and cdpk16-1 pollen tubes. Apical actin filaments are indicated by different colored dots. The severing events of actin filaments are indicated by red arrows. Bar = 5 μm. ( D ) Dynamic parameters of apical actin filaments in WT and cdpk16-1 pollen tubes. Data are presented as mean ± SE, with the number of filaments in parentheses. NS, no significant difference, **P < 0.01, (Student’s t test). WT, wild type.
Next, we performed real-time visualization of the dynamics of actin filaments decorated with Lifeact-eGFP and found that, consistent with previous observations [ 5 ], actin filaments continuously polymerized from the plasma membrane (PM) within the apical region of pollen tubes ( Fig 1A and S1 Movie ). We found that PM-originated actin filaments are obviously brighter within the apical region of cdpk16 pollen tubes than in WT pollen tubes ( Fig 1A and S2 Movie ). Kymograph analysis showed that the region occupied by membrane-originated actin filaments was enlarged in cdpk16-1 pollen tubes compared to WT ( Fig 1A , right panel). This is supported by physical measurements showing that the distance from the base of the apical actin structure to the pollen tube tip (mean ± SE) increased significantly in cdpk16-1 pollen tubes (6.20 ± 0.047 μm) compared to WT pollen tubes (4.27 ± 0.037 μm) ( Fig 1B ). Through monitoring the dynamics of individual PM-originated actin filaments ( Fig 1C and S3 and S4 Movies ), we found that the average severing frequency of actin filaments was significantly reduced in cdpk16-1 pollen tubes compared to WT ( Fig 1D ). Accordingly, the maximal filament length and lifetime increased significantly in cdpk16-1 pollen tubes compared to WT ( Fig 1D ). However, we did not notice obvious differences in the elongation and depolymerization rates of PM-originated apical actin filaments in cdpk16-1 pollen tubes compared to WT ( Fig 1D ). Thus, our results suggest that the amount of apical actin filaments is increased and their dynamics are reduced in cdpk16-1 pollen tubes.
We next determined the role of CDPK16 in regulating pollen germination and pollen tube growth. We initially compared the time course of pollen germination in WT and the 2 cdpk16 mutants, and found that pollen germination is accelerated early on in the cdpk16 mutants, but the overall pollen germination rate in cdpk16 mutants does not differ from that in WT ( S4A Fig ). This suggests that loss of function of CDPK16 promotes pollen germination. In addition, we found that the rate of pollen tube growth is significantly reduced in cdpk16 mutants compared to WT ( S4B and S4C Fig ), which suggests that CDPK16 promotes normal pollen tube growth. These data together suggest that CDPK16 maintains the normal rate of pollen germination and promotes pollen tube growth.
To understand the regulation of actin turnover in pollen, we performed forward chemical genetic screening to uncover mutations that alter the sensitivity of pollen germination to latrunculin B (LatB). Our attention was attracted by 1 T-DNA insertion knockout mutant allele of CDPK16, designated as cdpk16-1 ( S1A and S1B Fig ). Germination of cdpk16-1 pollen is resistant to LatB (see below). To demonstrate that the LatB-resistant pollen germination phenotype is indeed caused by the mutation in CDPK16, we also generated another cdpk16 mutant allele by the CRISPR/cas9 approach [ 33 ], designated as cdpk16-2 ( S1C Fig ). We found that cdpk16-1 and cdpk16-2 mutant pollen germinates better than wild-type (WT) pollen in the presence of LatB ( S1D and S1E Fig ), which suggests that the germination of cdpk16 mutant pollen is resistant to LatB. Accordingly, we found that the relative growth rate of pollen tubes was increased significantly in cdpk16 mutants compared to WT in the presence of LatB ( S1F and S1G Fig ), which suggests that loss of function of CDPK16 renders pollen tube growth resistant to LatB. We next found that there is no overt difference in terms of the overall brightness and organization of actin filaments in cdpk16 mutant pollen grains compared to WT in the absence of LatB ( S2A–S2C Fig ). We noticed that actin filaments became fragmented in both WT and cdpk16 mutant pollen grains after treatment with 150 nM LatB, but the extent of fragmentation is less obvious in cdpk16 mutant pollen grains than in WT ( S2A Fig ). In addition, we found that LatB-triggered actin depolymerization is inhibited in cdpk16 mutant pollen grains compared to WT, as evidenced by the significantly higher relative amount of actin filaments in cdpk16 mutant pollen grains compared to WT ( S2B and S2C Fig ). In line with the above observations, we found that overexpression of CDPK16 ( S3A Fig ) renders pollen germination sensitive to LatB ( S3B–S3D Fig ). Thus, these data suggest that CDPK16 promotes actin turnover in pollen.
Discussion
We here demonstrate that CDPK16 promotes actin turnover at least partly through the up-regulation of ADF7 activity in pollen. Our in vitro biochemical data show that CDPK16 phosphorylates Ser128 in ADF7 and up-regulates its actin-depolymerizing and severing activities. These findings differ from previous demonstrations or assumptions that kinase-mediated phosphorylation of plant ADFs occurs at Ser6 and down-regulates ADF activity [29–32,36,37]. Our study is the first to show that kinase-mediated phosphorylation up-regulates ADF/cofilin activity. Our results suggest that ADF7 is well suited to the regulation of actin dynamics within the apical region of pollen tubes, which harbor a tip-high Ca2+-gradient.
ADF/cofilin is an essential regulator of actin dynamics, and therefore, there is great interest in the mechanisms that control its activity. It is well known that kinase-mediated phosphorylation is one of the major mechanisms for inactivating ADF/cofilin [38]. This inactivating phosphorylation occurs at Ser3 in animals and has been demonstrated both in vitro [39] and in vivo [40]. A similar mechanism has been demonstrated for plant ADFs, with the phosphorylation occurring at Ser6 [29]. The plant community takes it for granted that phosphorylation-mediated regulation of plant ADFs is achieved by adding a phosphate group to Ser6. Therefore, researchers normally manipulate Ser6 in order to abolish phosphorylation-mediated regulation of plant ADFs in vivo [9,37]. We found that Ser6 is also highly conserved among class II ADFs (S8 Fig), but we did not detect any ADF7 peptides containing phosphorylated Ser6 during our mass spectrometry analyses. This suggests that the frequency of Ser6 phosphorylation in ADF7 is comparatively low in pollen, even if this phospho-regulation mechanism does apply to ADF7. However, we repeatedly identified the ADF7 peptide containing phosphorylated Ser128 (Fig 4A), which suggests that the phosphorylation of ADF7 at Ser128 frequently occurs in vivo. In support of this notion, we found that treatment with phosphatase reduced the amount of phosphorylated ADF7 in total pollen extract probed with anti-phospho-ADF7(Ser128) antibody (S9D Fig). In support of our observations, a recent report showed that CPK3-mediated phosphorylation of the C-terminus of ADF4 links actin remodeling to pattern-triggered immunity and effector-triggered immunity [41], but the biochemical mechanism underlying this regulation remains unknown. The biological significance of ADF7S128 phosphorylation is supported by data showing that ADF7S128A retains roughly the same actin-depolymerizing activity as ADF7 in vitro (Fig 4F–4J), but has reduced activity in promoting actin turnover within the apical and subapical regions of pollen tubes (Figs 5 and S11). Surprisingly, we found that ADF7S128D also has reduced activity in promoting actin turnover in pollen (S11 Fig), suggesting that ADF7S128D cannot fully mimic the function of phosphorylated ADF7 with the phosphorylation occurring at its Ser128. In support of this speculation, we found that ADF7S128D only has slightly enhanced activity in depolymerizing and severing actin filaments compared to ADF7 in vitro (Fig 4), whereas incubation of ADF7 with CDPK16 dramatically enhanced the actin-depolymerizing activity of ADF7 (Fig 3), albeit only a part of ADF7 protein is supposed to be phosphorylated by CDPK16 under the same condition based on the western blot results probed with anti-phospho-ADF7(Ser128) (S9A Fig). ADF7S128D cannot fully represent phosphorylated ADF7 in depolymerizing actin filaments, which could be due to the possibility that Ser128 in ADF7 is not the only residue targeted by CDPK16. In the future, identification of other potential residue(s) in ADF7 that might be targeted by CDPK16 will help to clarify whether this possibility does exist. Up-regulation of the activity of class II ADFs by phosphorylating Ser128 might be a universal mechanism in plants, as Ser128 is highly conserved among class II ADFs (S8 Fig). Surprisingly, based on the results from 2D gels (Fig 2F and 2G), ADF7 seems to be preferentially subjected to phosphorylation in pollen when compared to ADF10. We do not know currently how this selective phosphorylation is achieved. Nonetheless, we report a completely new mechanism for regulating the activity of ADF/cofilin in which kinase-mediated phosphorylation enhances the activity of ADF7 in Arabidopsis. Interestingly, we found that CKL2 also enhances the actin-depolymerizing activity of ADF7 (S14 Fig). We do not know why CKL2 has opposing effects on the actin-depolymerizing activity of ADF4 and ADF7, as CKL2 inhibits the actin-depolymerizing activity of ADF4 [30]. Nonetheless, this observation suggests that regulation of ADF7 by different kinases might allow actin dynamics to be integrated into different signal transduction pathways.
Since the discovery that ADF is phosphorylated and inactivated by CDPK(s) in plants [32], researchers in this field have been confused about how ADF(s) enhance actin dynamics at the tip of pollen tubes and root hairs, where the Ca2+ concentration is high and CDPKs are supposed to be constitutively active. One of the explanations proposed by researchers is that dephosphorylated ADF is redistributed to the tip of root hairs via an unknown mechanism to promote actin turnover [29]. Our findings that CDPK16-mediated phosphorylation enhances the activity of ADF7 (Fig 3) suggest that ADF7 is well suited to enhancing actin turnover locally at pollen tube tips. In support of this notion, we found that cdpk16 mutant pollen tubes exhibit more severe defects in actin turnover at pollen tube tips (Fig 1) and ADF7S128A has reduced activity in promoting actin turnover at pollen tube tips (Fig 5), although ADF7S128A has roughly the same actin-depolymerizing activity as ADF7 in vitro (Fig 4F–4J). We found that CDPK16 is concentrated on the membrane at the subapex but is comparatively less on the membrane at the apex (Fig 6), suggesting that the majority of CDPK16-mediated phosphorylation of ADF7 mainly occurs at the subapex. However, the cytoplasmic phosphorylated ADF7 should be able to rapidly diffuse to the apex to promote actin turnover, explaining why the rate of actin turnover was reduced at pollen tip including apical and subapical regions (Fig 1). Nonetheless, our findings completely change our view of how ADF contributes to the enhancement of actin dynamics within the apical region of pollen tubes, where there is a high Ca2+ concentration that will in turn activate CDPK(s). Based on our in vitro and in vivo data, we propose that CDPK16 phosphorylates ADF7 and enhances its actin severing and depolymerizing activity to promote actin turnover within the apical region of pollen tubes where the free Ca2+ concentration can reach the micromolar range [42,43]. As such, CDPK16 is involved in controlling the length and spatial distribution of apical actin filaments generated by membrane-anchored formins (Fig 7). Our findings thus suggest a scenario that unifies the high Ca2+-CDPK activity with ADF activation to maintain the high dynamics of actin filaments at pollen tube tips.
PPT PowerPoint slide
PNG larger image
TIFF original image Download: Fig 7. Schematic model illustrating the role of CDPK16 in regulating the activity of ADF7 and actin dynamics within the growth domain of pollen tubes. The left panel shows the organization of actin filaments within the apical and subapical regions of the pollen tube, where actin filaments are mainly polymerized from the PM by membrane-anchored formins utilizing profilin-actin complexes in the cytoplasm [60,74,75]. ADF has been implicated in the regulation of the turnover of apical actin filaments via severing and depolymerization [11]. This limits the length of filaments and shapes their organization to form the unique “apical actin structure” [6,7]. We found here that the PM-localized CDPK16 is also involved in promoting the turnover of those PM-originated actin filaments via phosphorylation of ADF7 at Ser128 to enhance its actin severing and depolymerizing activity. The boxed region in the left panel, where CDPK16 is comparatively concentrated, is enlarged in the upper right panel. The schematic diagram in the lower right panel shows that phosphorylation of Ser128 in ADF7 enhances its actin severing and depolymerizing activity. ADF, actin-depolymerizing factor; CDPK, calcium-dependent protein kinase; PM, plasma membrane.
https://doi.org/10.1371/journal.pbio.3002073.g007
As a universal second messenger, Ca2+ is involved in numerous physiological cellular processes in eukaryotes. The cytosolic concentration of Ca2+ ([Ca2+] cyt ) undergoes rapid changes in response to various internal and external stimuli, which are normally accompanied by the rapid rearrangement of actin filaments [44]. The Ca2+ levels are sensed by different Ca2+-binding proteins, such as the conserved eukaryotic protein calmodulin (CaM) or proteins carrying a CaM-like domain. Among them, CDPKs represent the largest subfamily of Ca2+ sensors in plants and have been implicated in different aspects of plant physiology [45–51]. Although several lines of evidence in the literature link CDPK signaling to the actin cytoskeleton [41,49,52], the underlying molecular mechanisms remain to be established. Our finding that CDPK16 promotes actin turnover by up-regulating the activity of ADF7 provides a mechanistic link between Ca2+ signaling and rapid actin rearrangement in pollen tubes or in other plant cells in general. In this way, actin can be reorganized in response to environmental and developmental signals mediated by fluxes in [Ca2+] cyt . Our study thus enhances our understanding of the link between Ca2+ signaling and actin dynamics in plants.
In summary, we report for the first time that kinase-mediated phosphorylation up-regulates the activity of ADF/cofilin. Our findings significantly enhance our understanding of the regulation of ADF/cofilin and suggest a scenario in which high Ca2+/CDPK activity can be unified with ADF activation to promote actin dynamics at pollen tube tips or root hair tips. Our results shed light on the molecular mechanisms underlying the regulation of actin dynamics during polarized growth of tip-growing cells and have general implications for understanding the dynamic interplay between Ca2+ signaling and actin dynamics in plants.
[END]
---
[1] Url:
https://journals.plos.org/plosbiology/article?id=10.1371/journal.pbio.3002073
Published and (C) by PLOS One
Content appears here under this condition or license: Creative Commons - Attribution BY 4.0.
via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/
