Dynamic organization of cortical actin filaments during the ooplasmic segregation of ascidian Ciona eggs.

Spatial reorganization of cytoplasm in zygotic cells is critically important for establishing the body plans of many animal species. In ascidian zygotes, maternal determinants (mRNAs) are first transported to the vegetal pole a few minutes after fertilization and then to the future posterior side of the zygotes in a later phase of cytoplasmic reorganization, before the first cell division. Here, by using a novel fluorescence polarization microscope that reports the position and the orientation of fluorescently labeled proteins in living cells, we mapped the local alignments and the time-dependent changes of cortical actin networks in Ciona eggs. The initial cytoplasmic reorganization started with the contraction of vegetal hemisphere approximately 20 s after the fertilization-induced [Ca2+] increase. Timing of the vegetal contraction was consistent with the emergence of highly aligned actin filaments at the cell cortex of the vegetal hemisphere, which ran perpendicular to the animal-vegetal axis. We propose that the cytoplasmic reorganization is initiated by the local contraction of laterally aligned cortical actomyosin in the vegetal hemisphere, which in turn generates the directional movement of cytoplasm within the whole egg.

INTRODUCTION

During early development of many animal species, including vertebrates, redistributions of cytoplasmic maternal substances in zygotes play important roles in the determination of body axes in the embryos (Lyczak ; Pelegri, 2003; Weaver, 2004; Ma ; Nishikata ). The ascidian zygotes show drastic reorganization of cytoplasm from the time before the first cell division (Sardet ; Makabe and Nishida, 2012), which has been traditionally called ooplasmic segregation. The initial cytoplasmic reorganization occurs immediately after the fertilization caused by the contraction of cortical actomyosin network (Jeffery and Meier, 1983; Jeffery, 1984; Sawada, 1983; Sawada and Osanai, 1984, 1985). The following reorganization of cytoplasm is driven by multiple microtubule systems that include sperm asters (Sawada and Schatten, 1988; Chiba ; Ishii ; Goto ). Importantly, these cytoplasmic movements are critically important for the establishment of the body axes of the ascidian embryo during development (Nishida, 2005; Makabe and Nishida, 2012). The interplay of these two cytoskeletal dynamics helps redistribution of the maternal mRNAs called postplasmic/PEM RNAs, which include the muscle determinant, macho-1 (Nishida, 2005; Makabe and Nishida, 2012). These maternal mRNAs regulate gene expressions for cell-specific differentiations, cell polarity, and cell division planes in the blastomeres (Sardet, 2003; Prodon ).

Ooplasmic segregation in ascidian eggs is triggered by fertilization. The first phase is initiated immediately after the fusion of the sperm to the egg, in which the myoplasm, a mitochondria-rich cytoplasmic domain containing postplasmic/PEM RNAs, begins to move toward the vegetal pole. Electron microscopic studies have suggested that the contraction of the cortical domain (plasma membrane lamina; PML) drives the first phase of ooplasmic segregation (Jeffery and Meier, 1983; Jeffery, 1984). Since this domain was stained by fluorescent phalloidin (Jeffery and Meier, 1983; Sawada and Osanai, 1985; Chiba ), the actin cytoskeleton was likely to be the major component of this domain. Inhibitor studies had suggested the involvement of the actin cytoskeleton in the movements (Zalokar, 1974; Reverberi, 1975; Sawada and Osanai, 1985; Chiba ; Takatori ). Jeffery and Meier (1983; also see Jeffery, 1984; and Sawada, 1983) had proposed similar models independently for the mechanism of ooplasmic segregation in ascidian eggs. These models are based on the contraction of actomyosin networks that spread over the egg cortex. Their models proposed that the actin-rich cell cortex is connected to the egg plasma membrane and to inner cytoplasm. The initial event of ooplasmic segregation starts with the contraction of the basket-shaped, actin-rich cell cortex toward the vegetal pole. The contraction of the actin-rich cortex to the vegetal pole produces the force that displaces the vegetal cytoplasm to the animal hemisphere. The updated models also support the models of Jeffery and Meier (1983) and Sawada (1983) (Prodon ; Sardet ; Makabe and Nishida, 2012) and add more precise explanations for the reorganization patterns of maternal factors such as mitochondria, the cortical endoplasmic reticulum (cER), and postplasmic/PEM RNAs during ooplasmic segregation.

The mechanism of how actomyosin contractility leads to the directional movement of maternal substances remains an unsolved problem. In this study, we have tackled this problem by using two advanced light microscopy techniques. One is selective plane illumination microscopy (SPIM) (Huisken and Stainier, 2009) to visualize the morphological change and cytoplasmic movement initiated by the fertilization-induced Ca2+ wave (Dumollard ). SPIM allowed us to observe Ca2+ dynamics and the following cytoplasmic movements in the whole Ciona egg as large as 140 µm in diameter with a high signal-to-noise ratio, revealing the local differences of Ca2+ dynamics and cytoplasmic movement of the fertilized Ciona egg in the vegetal hemisphere and those in the animal hemispheres in live eggs. Another is the polarized fluorescence microscope (instantaneous FluoPolScope) that we have developed for observing the orientation of fluorescently labeled biomolecules in live cells (Mehta ; Nordenfelt ; Swaminathan ; Altartouri ; Nakai ). The alignment of cortical actomyosin filaments is the key event for determining the orientation of force production in a variety of cell movements such as cell division and cell migration. The observation of cortical F-actin alignments is a powerful approach to explore the mechanisms of ooplasmic segregation. However, this trial has been hampered by the spatial resolution of light microscopy in the context of live cell imaging. The instantaneous FluoPolScope reports alignment of molecular assemblies with single-molecule sensitivity (Mehta ). On the basis of correlative imaging of cytoplasmic movement, intracellular [Ca2+], and the spatial organization of actin filaments in fertilized Ciona eggs, we found distinct spatial organizations of the actin cytoskeleton and their dynamics at the vegetal hemisphere and the animal hemisphere of Ciona eggs during the cytoplasmic reorganization. We will update the mechanisms of the spatial organization of cortical actomyosin for the first ooplasmic segregation in ascidian eggs.

RESULTS

Optical dissection of cytoplasmic reorganization in Ciona eggs by using SPIM

We examined the dynamics of cytoplasmic reorganization in fertilized Ciona eggs by using a new imaging tool, the SPIM. As shown in Figure 1A, fertilization induced an increase in intracellular [Ca2+] and the overall morphological change of the egg in parallel with the translocation of the autofluorescent myoplasm (Deno, 1987; Nishikata ) toward the vegetal pole. The myoplasm, which was observed with 488 nm laser light sheet illumination and a green emission filter (529/24 nm), was distributed over the cell cortex of the vegetal hemisphere of Ciona eggs. The domain is composed of a mitochondria-rich inner layer and a cER domain with postplasmic/PEM RNAs (Sardet ). The movement of the myoplasm resulted in the formation of a dense layer of myoplasm at the vegetal pole (Sardet ). Together with the myoplasmic contraction, we observed a series of deformations of the fertilized egg. These morphological changes were associated with the elevation of intracellular [Ca2+] (Speksnijder ; Roegiers ), as is shown in the top panel of Figure 1B. After the elevation of [Ca2+], the deformation of the egg started with a transient extension of the egg along the animal–vegetal (A–V) axis for less than 2 min (aspect ratio >1), followed by a long-lasting compression to the equatorial plane (aspect ratio <1) (bottom panel of Figure 1B). The onset of the egg extension was observed ∼20 s after the onset the [Ca2+] increase. The compression of the egg started in the middle of the slow decay phase of the fertilization-induced Ca2+ wave.

FIGURE 1:

Fertilization-induced intracellular Ca2+ wave and the following cytoplasmic reorganization in the C. intestinalis egg. (A) Time series fluorescence images of Ca2+ indicator Rhod-dextran (top lane) and autofluorescence (bottom lane) in fertilized Ciona eggs observed with SPIM. A white arrow in the first autofluorescence image indicates the position of the meiotic spindle as a landmark of the animal pole. A and V indicate the positions of the animal pole and the vegetal pole, respectively. The numbers at the top right of selected images indicate the time (in minutes) after the start of [Ca2+] elevation. Bar, 50 µm. (B) Fertilization-induced [Ca2+] changes (top panel) and the morphology changes (bottom panel) observed in nine preparations are plotted against time. Fertilization-induced [Ca2+] changes are shown as the normalized fluorescence intensity changes of Rhod-dextran. The morphological changes are expressed as a time series of change in the aspect ratio, the value derived from the length between the animal and the vegetal poles (LA-V) divided by the equatorial radius (L). Time zero for each data set is adjusted by the initial onset of Ca2+ elevation. (C, D) Kymographs of autofluorescent granules (mitochondria) in the myoplasm (C) and 100 nm fluorescent beads microinjected into the cytoplasm (D), both moving along the cortex of the Ciona egg after fertilization. Averaged fluorescence intensity profiles along the curved areas enclosed by white broken lines (inset) were aligned from left to right to create the kymographs. The top of the kymograph represents the animal pole (A in the insets), and the bottom represents the vegetal pole (V in the insets). Vertical white broken lines in C and D represent the time when the movement of autofluorescent granules in the myoplasm started. (E, F) Trajectories of autofluorescent granules in myoplasm (E) and 100 nm fluorescent beads (F) that moved along the cell cortex of the Ciona egg after fertilization. Vertical black lines in the graphs represent the time when the movement of autofluorescent granules starts. The times (abscissa in minutes) are measured after the onset of [Ca2+] elevation.

As shown in Figure 1, C and D, and Supplemental Movie S1, we monitored fertilization-triggered cortical movements in both vegetal and animal hemispheres by tracking the autofluorescent granules in the myoplasm and 100-nm-diameter fluorescent polystyrene beads (microsphere; ex: 580 nm/em: 605 nm; Molecular Probes) that had been microinjected into the cytoplasm of unfertilized Ciona eggs. We confirmed in separate observations that the autofluorescent granules were specifically stained with a fluorescent mitochondria marker, MitoTracker Red CMXRos (see Supplemental Figure S1). In two kymographs along the circumference of the egg cell cortex (Figure 1, C and D), the autofluorescent granules at the vegetal hemisphere and the microinjected beads at the animal hemisphere all moved toward the vegetal pole along the cortex. We also observed a short, directional movement of cytoplasm with polystyrene beads in the middle of the eggs, from the vegetal pole to the animal pole direction (unpublished data). We found that the contraction of myoplasm at the vegetal hemisphere preceded the cortical movements at the animal hemisphere. In Figure 1D, there was a short latency at the start of polystyrene bead movement along the cortex at the animal hemisphere. Tracking analysis of autofluorescent granules at the vegetal hemisphere (Figure 1E) and the polystyrene beads at the animal hemisphere (Figure 1F) revealed that the initiation of cortical movement at the animal hemisphere started approximately 5–20 s after the onset of the contraction of myoplasm at the vegetal hemisphere. The total distance of translocation of microinjected beads along the cortex of the animal hemisphere was approximately 40–60 µm (Figure 1F), whereas the myoplasmic movement from the equator to the vegetal pole was approximately 30 µm (Figure 1E). The speeds at the onset of their movements were similar (58.5 µm min–1 for myoplasm and 49.3 µm min–1 for beads at the animal hemisphere) and gradually slowed down within 4–5 min. These time correlations of cytoplasmic movements suggested the idea that the initiation of myoplasmic contraction at the vegetal hemisphere leads the first phase of ooplasmic segregation in ascidian zygotes, as proposed in the models of Jeffery and Meier (1983) and Sawada (1983).

Movie S1

Cytoplasmic movements in . Movement of microinjected fluorescent polystyrene particles (100 nm in diameter, upper panel) and that of fluorescent granules (lower panel) were simultaneously observed in fertilized Ciona egg with SPIM. Each frame was taken every 5 s. Exposure time, 400 ms. Transmittance of ND filters, 10% for 488 nm 20 mW laser (Sapphire, Coherent) and 5% for 561 nm 35 mW laser (85-YCA, Melles Griot). Replay speed, 6 frames/s.

The direction of intracellular Ca 2+ waves correlates with the cell polarity of fertilized Ciona eggs

We observed the correlation of Ca2+ waves with actomyosin-driven ooplasmic segregation and egg morphology changes (Supplemental Movie S2). Multiple Ca2+ waves with different origins occurred after fertilization. The first, fertilization induced, long-lasting large Ca2+ wave originated from a small area (Figure 2A, top lane) that had been reported as the point of sperm–egg fusion (Speksnijder ; Roegiers ). To enhance the wavefront images of the traveling Ca2+ waves observed with SPIM, we processed the series of time-derivative Rhod-dextran fluorescence images by subtracting the consecutive images (taken every 2 s) one after another (Figure 2A, middle lane). Consistent with previous reports (Speksnijder ; McDougall and Sardet, 1995), the speed of propagation of the fertilization-induced Ca2+ wave was approximately 7.5 µm s–1. It is worth noting that in the SPIM images in Figure 2A, the speed of Ca2+ wave propagation along the cell cortex was faster than that of the wave that ran in the middle of the cytoplasm; the shape of the wavefront was convex in the beginning, then it was straight in the middle, and then it changed to concave in the end of the traveling wave. Similar observations were reported in ascidian eggs by using confocal microscopy and simulation models (Dupont and Dumollard, 2004).

FIGURE 2:

The origins and propagation directions of intracellular Ca2+ waves in the fertilized C. intestinalis egg. (A) Top lane: Time series of Rhod-dextran fluorescence images showing the propagation of the fertilization-induced Ca2+ wave. The images were taken every 2 s. Times after the beginning of initial [Ca2+] elevation are shown at the top left of selected images. Middle lane: Time derivative images of Rhod-dextran fluorescence obtained by subtracting two consecutive images in the top lane, one after the other. Bottom lane: Time series of autofluorescence images during the first fertilization-induced Ca2+ wave. A and V indicate the positions of the animal pole and the vegetal pole, respectively. Bars, 50 µm. (B) Top lane: Time series of Rhod-dextran fluorescence images showing the propagation of the Ca2+-sequestering wave. Times after the beginning of initial fertilization-induced Ca2+ elevation are shown at the top left of selected images. Middle lane: Time derivative images of Rhod-dextran fluorescence that enhance the wavefront of the Ca2+-sequestering wave. Bottom lane: Autofluorescence images during a passage of the Ca2+-sequestering wave. A and V indicate the positions of the animal pole and the vegetal pole, respectively. Bars, 50 µm. (C) Top lane: Time series of Rhod-dextran fluorescence images showing a propagation of the pacemaker Ca2+ wave. Times after the beginning of initial fertilization-induced [Ca2+] elevation are shown at the top left of selected images. Middle lane: Time derivative images of Rhod-dextran fluorescence during the propagation of the pacemaker Ca2+ wave. Bottom lane: Autofluorescence images during the propagation of the pacemaker Ca2+ wave. A and V indicate the positions of the animal pole and the vegetal pole, respectively. Bars, 50 µm. (D) Histograms showing the directions of wave propagations in the first fertilization-induced Ca2+ wave (blue), the pacemaker waves (red), and the sequestering waves (green). Ninety degrees (90°) represents the wave direction toward the animal pole and –90° represents the direction toward the vegetal pole. Numbers of eggs examined: 23 for the fertilization-induced wave, 13 for pacemaker waves, and 13 for sequestering waves. (E) Oscillation of intracellular [Ca2+] monitored by Rhod-dextran fluorescence (top trace) and that of autofluorescence (bottom trace). Time is measured after the beginning of initial fertilization-induced [Ca2+] elevation. Timings indicated by a, b, and c correspond to the timings of the first frames of panels A, B, and C, respectively.

Movie S2

Propagation of intracellular Ca2+ waves in fertilized . Propagation of Ca2+ waves from the fertilization induced Ca2+ wave, the sequestering Ca2+ wave to the first pacemaker Ca2+ wave. Each frame was taken every 2 s. Exposure time, 200 ms. Transmittance of ND filter used, 5% for 561 nm 35 mW laser (85-YCA, Melles Griot). Replay speed, 10 frames/s.

A series of short repetitive Ca2+ waves followed the initial fertilization-induced Ca2+ wave (Figure 2C), which have been called Ca2+ pacemaker waves (Speksnijder ; Roegiers ; Dumollard and Sardet, 2001). In our observations, these repetitive waves originated from the vegetal hemisphere, but the initial wave was often observed not as a small point, but as a broad cup-shape area as shown in the time-derivative images (Figure 2C, middle lane, t = 2′46″). The location of this cup-shape Ca2+ distribution was consistent with that of the autofluorescent myoplasm (Figure 2C, bottom lane), indicating that pacemaker Ca2+ waves originated from the subcellular structures that associated with myoplasm. The propagation rate of the pacemaker Ca2+ waves was approximately 25–50 µm s–1, which was much faster than that of the fertilization-induced Ca2+ wave.

SPIM imaging showed not only the wave of [Ca2+] increase but also a wave of decreasing [Ca2+], which we called the “Ca2+-sequestering wave.” As far as we know, this is the first time a propagating wave of [Ca2+] decrease is reported. The Ca2+-sequestering waves followed the initial fertilization Ca2+ wave as is clearly observed in Figure 2B and after the following Ca2+ pacemaker waves (unpublished data). In other words, we observed Ca2+-sequestering waves every time after [Ca2+] increases. Interestingly, the origins of these Ca2+-sequestering waves were consistently from the animal pole. The Ca2+-sequestering wave was not a back of the propagating Ca2+ wave when the wave passes by, as in many cases we observed that the propagating direction of the Ca2+ wave was different from that of the Ca2+-sequestering wave. The velocity of the Ca2+-sequestering waves was 5–6 µm s–1, which was slightly faster than that of the first Ca2+ wave and much slower than that of the Ca2+ pacemaker waves. We did not detect any difference between the velocities of Ca2+-­sequestering waves observed after the fertilization-induced Ca2+ wave and those observed after pacemaker Ca2+ waves. In Figure 2D, we plotted the propagating directions of the first Ca2+ waves, Ca2+ pacemaker waves, and Ca2+-sequestering waves with respect to the A–V axis. The origins of the fertilization-induced Ca2+ waves spread all over the surface of the Ciona egg, which was not consistent with the observations using a different species of ascidian, Phallusia, in which the origin of the first Ca2+ waves was around the animal pole (Speksnijder ; Roegiers ). In the following Ca2+ pacemaker waves, approximately 70% (nine of 13 waves) traveled to the animal pole side. In the Ca2+-­sequestering waves, all waves we observed (13 waves) propagated along the A–V axis from the animal pole to the vegetal pole, regardless of the origins and the propagation directions of the preceding [Ca2+] raises. On the basis of these observations, in later experiments we used the direction of Ca2+-sequestering waves as a reference to identify the A–V axis of fertilized Ciona eggs.

It is worth noting that the autofluorescence of Ciona eggs, observed through the 529/24 nm emission filter, oscillated along with the Ca2+ waves (Figure 2E). These changes could be observed in the autofluorescence signal at the myoplasm as well as at the cytosol of the fertilized eggs. Interestingly, the changes in the autofluorescence correlated only with the Ca2+ pacemaker waves, but not with the first fertilization-induced Ca2+ wave. At the onset of the fertilization-induced increase in [Ca2+] (Figure 2E, vertical broken line a), we did not observe the increase of autofluorescence intensity, but we observed the oscillation of autofluorescence that was correlated with the Ca2+ oscillation. Similar autofluorescence changes that correlated with changes in [Ca2+] had been reported in fertilized mouse eggs (Dumollard ).

Correlative imaging of intracellular [Ca 2+], cortical actin dynamics, and membrane contraction of fertilized Ciona eggs observed with total internal reflection fluorescence microscopy

Previous studies (Jeffery and Meier, 1983; Jeffery, 1984; Sawada and Osanai, 1984, 1985) had demonstrated that the first phase of ooplasmic segregation and egg morphology changes were closely related to the contractility of actomyosin at the cell cortex of ascidian eggs. Based on fluorescence imaging of fixed cells by fluorescent phalloidin labeling (Jeffery and Meier, 1983; Sawada, 1983; Chiba ) or by isolated cortex (Sardet ; Prodon ), the actin cytoskeleton seemed to exist very close to the cortical membrane. But the mechanisms of the actomyosin system that translocate the maternal substances to the vegetal pole were not clear in the context of live cell imaging. We have studied actin cytoskeleton dynamics in fertilized Ciona eggs by using total internal reflection fluorescence microscopy (TIRFM). We microinjected Alexa Fluor 488 phalloidin (AF488-phalloidin; final concentration, ∼10 nM; see Materials and Methods for details) and Rhod-dextran (final concentration, 50–500 nM; see Materials and Methods for details) into unfertilized Ciona eggs. The eggs with AF488-phalloidin and Rhod-dextran were placed on cleaned glass coverslips and were observed with TIRFM. The observed area was approximately 40–50 µm in diameter, which was one third of the average diameter of Ciona eggs (Figure 3A). As we have described for SPIM imaging, autofluorescence images could be observed with the fluorescence imaging setting for observing AF488-phalloidin. We compared the fluorescence intensities of Ciona eggs before and after microinjecting AF488-phalloidin. Autofluorescence accounted for roughly 1/5 of the observed intensity at 529 nm.

FIGURE 3:

Correlative imaging of [Ca2+], F-actin dynamics, and membrane changes in the fertilized C. intestinalis egg. (A) Schematic drawing showing the imaging area of the Ciona egg placed ona glass coverslip for the TIRFM. (B) TIRFM images of Rhod-dextran (top panels) and AF488-phalloidin (bottom panels) before (left) and 30 s after (right) the onset of the first [Ca2+] elevation. Bars, 10 µm. (C) Time series of fluorescence intensity changes in Rhod-dextran for Ca2+ concentration (top), AF488-phalloidin for F-actin (middle), and the membrane tension index derived from the equations of Hiramoto (1976) for estimating the membrane tension of round cells (bottom). Time 0 is the onset of the first [Ca2+] elevation. (D) The normalized and averaged [Ca2+] changes and the following changes of F-actin. The values 5 s before the onset of the [Ca2+] increase and the peak values of the Ca2+ wave were used for normalizing the absolute values. The center of each circle represents the mean value, and the length of the vertical line represents the SD. Six waveforms of the pacemaker wave were used for averaging the normalized values.

Upon fertilization, we observed increases in the fluorescence intensities of both Rhod-dextran and AF488-phalloidin in the Ciona egg cortex adjacent to the glass coverslip (Figure 3B). Monitoring these fluorescence changes allowed us to correlate [Ca2+] and F-actin assembly that was labeled with AF488-phalloidin. Among 14 correlative observations, one set of [Ca2+] and F-actin changes as well as the membrane area change are shown in Figure 3C and Supplemental Movie S3 as a typical example. Consistently with a previous report (Dumollard ) and with our SPIM observations (Figures 1 and 2), the fertilized egg showed an initial, long-lasting (∼2 min) Ca2+ wave, which was followed by multiple pacemaker waves that were observed within 5–6 min from the first [Ca2+] elevation. In AF488-phalloidin fluorescence signals, we observed a large transient increase followed by small fluctuations after fertilization. During the period of pacemaker wave propagation, we found that there was a delay between the increase of [Ca2+] and that of the following AF488-phalloidin fluorescence (Figure 3C; vertical lines indicate the onset times for the increases in [Ca2+]). The normalized and averaged changes of six pacemaker Ca2+ waves and those of the corresponding AF488-phalloidin fluorescence revealed that the AF488-phalloidin fluorescence intensity reached its maximum value approximately 30 s after the maximum increase of [Ca2+] (Figure 3D). Thus, the transient changes of AF488-phalloidin fluorescence followed the transient [Ca2+] changes, with a delay of approximately 30 s for each period after the [Ca2+] change.

Movie S3

Changes in cortical Ca. Three correlative movies showing the changes in Ca2+ (left), AF488-phalloidin distribution (middle) and the edge morphology (right) of the contact surface of Ciona egg with glass coverslip observed before and after fertilization. Frame interval, 5 s. Replay speed, 20 frames/s. Bar, 10 μm.

We observed periodical changes in the contact surface of Ciona eggs with the glass coverslip. These changes might reflect the cortical tension of the eggs. To estimate the change in cortical tension, weemployed an equation by Hiramoto (1976) for estimating the cortical tension of sea urchin eggs placed on a glass surface. In Hiramoto’s estimation, tension at the membrane (T) of a spherical egg placed on a flat substrate is proportional to the inverse of the area attached to the substrate,

where R is the radius of the egg at the largest optical section, R is the radius of membrane area attached to the substrate, r is the density of the egg, s is the density of seawater, and g is the gravity constant. At the bottom of Figure 3C, we plotted the time-dependent change of the inverse of the membrane area attached to the glass coverslip as membrane tension index T to show the relative changes in tension at the membrane while [Ca2+] and the amount of F-actin were changing. T changed simultaneously with the increase of Rhod-dextran and AF488-phalloidin fluorescence, suggesting that Ca2+-induced actomyosin contraction induced the membrane contraction at the cell surface. Similarly to the changes in AF488-phalloidin, the onset of the membrane tension increase was approximately 30 s after the increase of [Ca2+], which was consistent with the delay in the morphological change after the [Ca2+] increase, as we observed in SPIM imaging (Figure 1B).

Approximately 8–10 min after the onset of the first Ca2+ increase, the egg cortex entered a relaxation phase, observed as slacking of the membrane on the surface of the glass coverslips. The AF488-phalloidin fluorescence intensity decreased to a level lower than those we had observed before the fertilization. Following this phase, we observed the second oscillation phase of [Ca2+] (Dumollard and Sardet, 2001) and AF488-phalloidin–labeled F-actin (starting at around 11 min in Figure 3C) with intervals of 1–2 min. Similar to the first oscillation phase, [Ca2+] and AF488-phalloidin oscillations were observed in a correlated manner. For each elevation of [Ca2+], the increase of AF488-phalloidin fluorescence followed with a latent time of approximately 30 s (Figure 3D). We observed that the increase of AF488-phalloidin fluorescence and the changes in the attached membrane area occurred almost synchronously without any detectable time differences.

Fluorescence polarization imaging of AF488-phalloidin–labeled F-actin revealed the transient alignment of F-actin at the cortex of fertilized Ciona eggs during the first ooplasmic segregation.

We used TIRFM-based instantaneous FluoPolScope (Mehta ) for exploring the spatial/temporal dynamics of F-actin alignments at the cell cortex of Ciona eggs, especially during the first phase of ooplasmic segregation. The instantaneous FluoPolScope detects the polarization orientation of fluorescence by acquiring four different polarized fluorescence images of the same object as shown in Supplemental Movie S4. In Figure 4A and Supplemental Movie S5 (left), we show a series of typical TIRFM images of Ciona eggs with AF488-phalloidin before (t = –1′) and after (t = 3′ –12′) fertilization, until the time of the first cell division (t = 44′). In the fluorescence images (top photographs in Figure 4A), we observed speckle images of AF488-phalloidin bound to F-actin. We used speckles of AF488-phalloidin staining to compute the anisotropy of fluorescence (polarization factor) and the absolute polarization orientation of AF488 that report the degree of alignment and the net orientation of F-actin. We used the polarization factor of AF488 to remove the nonspecific staining signal of AF488 from further analysis, as the fluorescence from AF488-phalloidin was highly polarized only when it was bound to actin filaments. AF488-phalloidin bound to F-actin shows the polarized fluorescence orientation that is parallel to the length of F-actin (Mehta ). By observing the polarized fluorescence of AF488-phalloidin, we mapped the alignments of actin filaments at the cell cortex of Ciona eggs (bottom panel of Figure 4A and Supplemental Movie S5 [right]). As shown in the map of the fluorescence polarization orientation of AF488-phalloidin, we found highly aligned F-actin at the cell cortex at t = 3′ after fertilization. In this particular case, the orientation of F-actin alignment was around 150° in the laboratory coordinate system, as was observed in the peak of orientation histogram of the fluorescence polarization of AF488-phalloidin at t = 3′ (Figure 4B). It is worth noting that there was partially aligned F-actin in the same orientation before fertilization, as we see in the peak of the top histogram of fluorescence polarization orientation of AF488-phalloidin in Figure 4B. These observations suggested that there were preexisting, weakly aligned F-actin networks at the egg cortex before fertilization. The fertilization-induced signaling, which includes the elevation of [Ca2+], triggered the increasing density of polymerized actin (as a fluorescence intensity increase of AF488-phalloidin in Figure 4A, top trace) and the facilitation of F-actin alignment (as an increase of polarization factor in Figure 4A, bottom trace) to the same orientation that had existed before the fertilization. The alignment of F-actin was observed for 9–10 min (Figure 4, A and C) after the fertilization and then disappeared along with the disassembly of F-actin (Figure 4A, t = 12′). Thus, the emergence of F-actin to the cell cortex was always associated with highly ordered alignment of F-actin, as we see in the fluorescence intensity and the polarization factor of AF488-phalloidin (Figure 4C). The polarization factor (anisotropy of polarized fluorescence, see Materials and Methods for further details) of F-actin alignment reached the highest value when the contractile ring for the first cell division was observed at around 44–45 min after fertilization (Figure 4A, t = 44′).

FIGURE 4:

Transient F-actin alignment at the vegetal hemisphere during the first ooplasmic segregation of the C. intestinalis egg. (A) Time series of TIRFM images (top panel) and florescence polarization orientation maps (bottom panel) of a Ciona egg microinjected with AF488-phalloidin, before and after fertilization observed at the vegetal hemisphere, as we identified from the location of the sperm aster as shown in D. Positions of AF488-phalloidin particles bound to F-actin are shown as yellow dots. Orientation of the yellow bars represents the fluorescence polarization orientation of AF488, and the length of the bars represents the polarization factor. Bars, 10 µm. The numbers at the top right indicate the time (in minutes) before and after the onset of initial cortical contraction of the fertilized egg. (B) Histograms of the fluorescence polarization orientation of AF488-phalloidin in the Ciona egg cortex before (top) and after (bottom; 3 min after the onset of the first cortical contraction) fertilization. The frequency value represents the numbers of particles divided by the total number of particles analyzed. Orientations are shown in a laboratory coordinate system. (C) Time changes of the fluorescence intensity (top) and the fluorescence polarization of AF488-phalloidin observed at the cell cortex of the Ciona egg during fertilization. Time 0 represents the onset of initial cortical contraction of the fertilized egg. The relative intensity is derived from the averaged fluorescence intensity of measured area divided by the averaged background fluorescence intensity of the area outside the cell. For calculating the polarization factor, see Materials and Methods. (D) The change in the fluorescence intensity of AF488-phalloidin (Phal, left) and the movement of the sperm aster labeled with CiEB1-mCherry (EB1, right). The numbers at the top right indicate the time (in minutes) after the onset of initial cortical contraction. Bar, 10 µm.

Movie S4

Changes of F-actin alignments at the vegetal hemisphere of . A. A movie of the quadrant polarized fluorescence images of Ciona egg with AF488-phalloidin before and after fertilization. Degrees embedded in each quadrant indicate the transmission axis of polarization. Frame intervals, 10 s. Replay speed, 15 frames/s. B. Transient changes of F-actin alignments are shown from the time before fertilization to the first cell division of Ciona egg. (Left) Movie of the distribution of AF488-phalloidin obtained from the average of four polarized fluorescence images. (Right) Movie of fluorescence polarization map computed from four polarized fluorescence images of AF488-phalloidin. Positions of AF488-phalloidin particles bound to F-actin are shown as yellow dots, and the fluorescence polarization orientations of the particle are shown as yellow bars. The length of the bars represents the polarization factor of the fluorescent particles. Frame intervals, 10 s. Replay speed, 15 frames/s.

Movie S5

Changes of F-actin alignments at the vegetal hemisphere of . A. A movie of the quadrant polarized fluorescence images of Ciona egg with AF488-phalloidin before and after fertilization. Degrees embedded in each quadrant indicate the transmission axis of polarization. Frame intervals, 10 s. Replay speed, 15 frames/s. B. Transient changes of F-actin alignments are shown from the time before fertilization to the first cell division of Ciona egg. (Left) Movie of the distribution of AF488-phalloidin obtained from the average of four polarized fluorescence images. (Right) Movie of fluorescence polarization map computed from four polarized fluorescence images of AF488-phalloidin. Positions of AF488-phalloidin particles bound to F-actin are shown as yellow dots, and the fluorescence polarization orientations of the particle are shown as yellow bars. The length of the bars represents the polarization factor of the fluorescent particles. Frame intervals, 10 s. Replay speed, 15 frames/s.

We tried to determine the orientation of observed F-actin alignments with respect to the A–V axis. Observation of microtubules by using CiEB1-mCherry (see Materials and Methods for details) allowed us to correlate the movement of sperm aster with F-actin dynamics. As shown in Figure 4D and Supplemental Movie S6, the sperm aster, which was detected as a radially aligned CiEB1-mCherry bound to the vicinity of the plus end of microtubules, moved across the egg cortex during the cytoplasmic reorganization. Previous observations (Sardet ; Roegiers ; Ishii ) have revealed that after the initial phase of ooplasmic segregation, the sperm aster moves to the posterior-vegetal hemisphere along the A–V axis. The movement of the sperm aster in Figure 4D suggested that the alignments of F-actin that we observed in the vegetal hemisphere were normal to the A–V axis. To confirm whether the F-actin is consistently aligned perpendicular to the A–V axis in Ciona eggs, in the following section, we observed the alignment of F-actin at various locations where we could determine the A–V axis and the hemispheres of the observed area.

Movie S6

Changes of F-actin alignments at the vegetal hemisphere of . A. A movie of the quadrant polarized fluorescence images of Ciona egg with AF488-phalloidin before and after fertilization. Degrees embedded in each quadrant indicate the transmission axis of polarization. Frame intervals, 10 s. Replay speed, 15 frames/s. B. Transient changes of F-actin alignments are shown from the time before fertilization to the first cell division of Ciona egg. (Left) Movie of the distribution of AF488-phalloidin obtained from the average of four polarized fluorescence images. (Right) Movie of fluorescence polarization map computed from four polarized fluorescence images of AF488-phalloidin. Positions of AF488-phalloidin particles bound to F-actin are shown as yellow dots, and the fluorescence polarization orientations of the particle are shown as yellow bars. The length of the bars represents the polarization factor of the fluorescent particles. Frame intervals, 10 s. Replay speed, 15 frames/s.

Locally distinct cortical F-actin alignments in fertilized Ciona eggs during the first ooplasmic segregation

On the basis of the correlative observation of the meiotic spindle (marker of animal pole), mitochondria (marker of vegetal hemisphere), and the Ca2+ wave propagations using SPIM (Figures 1 and 2), we found that the A–V axis could be estimated from the orientation of the sequestering waves. In 14 data sets among 22 trials of the instantaneous FluoPolScope imaging of Ciona eggs, we correlatively observed F-actin orientations, distribution of mitochondria and yolk granules, and Ca2+ waves before and after fertilization. We found there were two distinct F-actin alignment orientations at the cortex of fertilized Ciona eggs during the first phase of ooplasmic segregation. F-actin aligned normal to the A–V axis (Figure 5, C and D, and Supplemental Movie S7) at the vegetal hemisphere where we observed a dense distribution of submicron-diameter mitochondria (Figure 5A) with the transmitted light imaging. The time series of cortical F-actin alignment changes (Figure 5B) and the histograms of the polarization orientation of AF488-phalloidin–stained F-actin before and after the fertilization showed that in this particular case, the orientations of F-actin were around 126° (before fertilization) and 130° (after fertilization) in the laboratory coordinate system (Figure 5D). Based on the direction of the Ca2+-sequestering wave, the estimated A–V axis was 47°. In contrast, at the equatorial region where we observed the boundary of the region with submicron scale mitochondria and that with large (2–3-µm-diameter) yolk granules (Figure 5E), F-actin alignments before and after fertilization were near parallel to the A–V axis (Figure 5, F–H, and Supplemental Movie S8). We summarize the observed orientations of F-actin alignments with respect to the A–V axis at different locations of Ciona eggs in Figure 5I and Table 1.

FIGURE 5:

Locally distinct F-actin alignments in the C. intestinalis egg during the first ooplasmic segregation. (A) Bright field image of the vegetal hemisphere of the Ciona egg surface showing a uniform distribution of mitochondria. The A–V axis was determined by the propagation orientation of Ca2+-sequestering waves. Bar, 10 µm. (B) Time series of TIRFM images of AF488-phalloidin at the cell cortex of Ciona observed in the same area as in A before (t = –1′) and after (t = 5′, 7′ 30′′) the onset of fertilization-induced [Ca2+] elevation. (C) Time series of the fluorescence polarization maps of AF488-phalloidin at the cell cortex of Ciona in the same area as observed in A and B before (t = –1′) and after (t = 5′, 7′ 30′′) the onset of fertilization-induced Ca2+ elevation. (D) Histograms of the fluorescence polarization orientation of AF488-phalloidin in the Ciona egg before (orange) and 7 min 30 s after (yellowish green) the onset of fertilization-induced [Ca2+] elevation. Orientations are shown in a laboratory coordinate system. (E) Bright field image of the Ciona egg surface around the equator showing a boundary between the mitochondria-rich area and that rich in yolk granules. The A–V axis was determined by the propagation orientation of Ca2+-sequestering waves. Bar, 10 µm. (F) Time series of TIRFM images of AF488-phalloidin at the cell cortex of Ciona in the same area as observed in E before (t = –1′) and after (t = 2′, 4′) the onset of fertilization-induced [Ca2+] elevation. (G) Time series of fluorescence polarization maps of AF488-phalloidin at the cell cortex of Ciona in the same area as observed in E and F before (t = –1′) and after (t = 2′, 4′) the onset of fertilization-induced [Ca2+] elevation. (H) Histograms of the fluorescence polarization orientation of AF488-phalloidin in the Ciona egg before (orange) and 4 min after (yellowish green) the onset of fertilization-induced [Ca2+] elevation. Orientations are shown in a laboratory coordinate system. (I) Panel that summarizes the distribution of ensemble orientation of F-actin (estimated from the fluorescence polarization orientations of AF488-phalloidin) with respect to the estimated orientation of the A–V axis in Ciona eggs observed in 14 embryos in Table 1. The orientations of F-actin alignments observed at the vegetal hemisphere, around the equator, and at the animal hemisphere are shown in red, green, and yellow, respectively.

TABLE 1:

F-actin alignments observed at various locations of cell cortex in C. intestinalis eggs.

Analyses of F-actin alignments and A–V axes observed in the vegetal hemispheres (No. 1–No. 9, red columns), around the equator (No. 10–No.11, green columns), and in the animal hemisphere (No. 12–No. 14, yellow columns). Net F-actin alignments were determined from the orientation histograms of ensemble fluorescence polarization of AF488-phalloidin fluorescence. The orientation of the A–V axis for each egg was identified by the propagation direction of Ca2+-sequestering waves observed at the same area. See Materials and Methods for further details.

Movie S7

Correlative imaging of Ca. Correlative movie of time derivative images of Rhod-dextran (left) and that of the polarized fluorescence orientation maps of AF488-phalloidin (right) observed at the cell cortex of Ciona egg. A. Movie taken in the vegetal hemisphere. Frame intervals, 5 s. Replay speed, 10 frames/s. B. Movie taken around the equator. Frame intervals, 1 s. Replay speed, 50 frames/s.

Movie S8

Correlative imaging of Ca. Correlative movie of time derivative images of Rhod-dextran (left) and that of the polarized fluorescence orientation maps of AF488-phalloidin (right) observed at the cell cortex of Ciona egg. A. Movie taken in the vegetal hemisphere. Frame intervals, 5 s. Replay speed, 10 frames/s. B. Movie taken around the equator. Frame intervals, 1 s. Replay speed, 50 frames/s.

Thus, on the basis of our observations of F-actin orientation, distributions of mitochondria and yolk granules, and the Ca2+-sequestering wave, we found that there was a local difference in the spatial organization of F-actin alignments with respect to the A–V axis at different locations of Ciona eggs during the initial phase of ooplasmic segregation. In 9 cases of 14, at the vegetal hemisphere, the observed F-actin alignments were close to the orientation perpendicular to the A–V axis. In the regions that we determined to be around the equator or animal hemisphere, F-actin alignments were mostly parallel along the A–V axis. We were not successful in imaging cortical F-actin orientation near the animal pole and vegetal pole, presumably because the attachment of egg membrane of these areas to the glass coverslip might interfere with the normal process of ooplasmic segregation.

DISCUSSION

Generation of cytoplasmic movements in fertilized Ciona eggs during actomyosin-driven ooplasmic segregation

By using SPIM, we imaged the dynamics of cytoplasmic reorganization in Ciona eggs after fertilization. The primary object of this imaging was to understand the mechanisms that cause cytoplasmic movements in fertilized Ciona eggs. Tracking of the movements of autofluorescent granules (mitochondria) at the vegetal hemisphere and microinjected fluorescent beads that spread over the whole cytoplasm of the eggs revealed that the onset of the first cytoplasmic reorganization was initiated by the cortical contraction at the myoplasm (Figure 1, C and E). The movement of myoplasm was followed by large cortical movements along the surface of the animal hemisphere, from the animal pole to the equator along the cell cortex (Figure 1, D and F). Our observations were fundamentally consistent with the hypothetical models proposed in previous studies (Jeffery and Meier, 1983; Jeffery, 1984; Sawada, 1983); the idea being that the first ooplasmic segregation is based on the contraction of the cell cortex at the vegetal hemisphere and the expansion of the cell cortex at the animal hemisphere.

Simultaneous monitoring of intracellular [Ca2+] and cytoplasmic movement revealed that, in fertilized Ciona eggs, cytoplasmic reorganization started after the completion of fertilization-induced [Ca2+] increase in the entire egg (Figures 1 and 3). In other words, the transient local difference of [Ca2+] during the propagation of the Ca2+ wave was not likely to be the main cause of local difference of actomyosin-driven cytoplasmic movements during the first ooplasmic segregation. These results might be inconsistent with the previous report using Phallusia eggs, that there was a spatial correlation between the sperm entry site and the location of a small protrusion near the vegetal pole (contraction pole) after the contraction of the vegetal hemisphere (Speksnijder ; Roegiers ). The previous observations of these authors suggested that the local difference of [Ca2+] during Ca2+ wave propagation might affect the actomyosin-driven deformation of the eggs in the vicinity of the vegetal pole. Rather, our observations indicated that the spatial dynamics of the first cytoplasmic reorganization in Ciona eggs after fertilization is controlled by the organization of the cytoskeletal architectures that have already been programmed during the spatial organization of oocyte during oogenesis.

Roles of Ca 2+ waves for the establishment of cell polarity in fertilized Ciona eggs

We refined the Ca2+ waves in fertilized Ciona eggs by using SPIM, which allowed us to analyze the propagation of Ca2+ waves in three-dimensional (3D) space. By tracing the wavefront of these spatial [Ca2+] changes, we located the origins for each wave in 3D space. On the basis of our analysis, we found that in Ciona, the origins of the first Ca2+ wave (fertilization-induced Ca2+ wave) were not limited in the vicinity of the animal pole (Figure 2D). A series of pacemaker waves that follow the first Ca2+ wave originated from a broad area within the vegetal hemisphere that corresponded to the location of the myoplasm. The vegetal origins of the pacemaker waves was generally consistent with previous reports that the ER and mitochondria-rich domains around the contraction pole were the origins of Ca2+ pacemaker waves (Speksnijder ; Roegiers ; Dumollard and Sardet, 2001).

We found a series of novel Ca2+-sequestering waves that followed each rise of [Ca2+] (Figure 2B). Interestingly, based on our observations, the sequestering waves consistently originated from the animal pole (Figure 2D). The possible mechanisms for the propagating Ca2+ sequestration are unclear. Dumollard found a Ca2+-sequestering function of ascidian mitochondria, which was blocked by an application of inhibitors for the mitochondrial respiration function (Dumollard ). Similar to the simultaneous observations of [Ca2+] and mitochondrial flavoproteins (FAD++) fluorescence changes in fertilized mouse eggs (Dumollard ), we found synchronous changes of [Ca2+] and autofluorescence observed through a 529/24 nm emission filter (Figure 2E), suggesting the involvement of mitochondria for the regulation of Ca2+ sequestration, in addition to the cytoplasmic [Ca2+] increase. At this point, we have no evidence to suggest the intracellular signals from the animal pole that mediates Ca2+ sequestration waves. Since the animal pole is the site for the egg nucleus, some signaling molecules of nuclear origin such as cdk1-cyclinB (Bischof ) might be involved in Ca2+ sequestration.

Local control of F-actin alignment as a possible mechanism for generating cytoplasmic movement during the first ooplasmic segregation of fertilized Ciona eggs

By using our own fluorescence polarization imaging method, we have found transient emergence and dissolution of highly aligned F-actin at the cortex of Ciona eggs right after fertilization (Figure 4A). The onset of the cortical F-actin alignments was consistent with that of the morphological changes of the eggs, as we observed with SPIM (Figure 1). On the basis of the correlative imaging of the propagation of Ca2+-sequestering waves that propagated along the A–V axis, we found that the orientations of F-actin alignments at the vegetal hemisphere were almost normal to the A–V axis, whereas the orientations observed at the equator and animal hemisphere were closer to the A–V axis. To our knowledge, this is the first observation of F-actin alignments and their dynamics at the cortical domain of living eggs during fertilization. Interestingly, we observed weak alignments of cortical F-actin in unfertilized eggs, in which the orientations were similar to those observed after fertilization with respect to the A–V axis (Figures 4 and 5). These observations suggest that the locally distinct spatial organizations of the actin cytoskeleton at the cell cortex of the Ciona egg have been built during oogenesis. Thus, the basic alignments of actin filaments with respect to the A–V axis were established before fertilization and were kept unchanged even after fertilization. It is unclear how actin filaments are tangentially aligned with respect to the A–V axis during oogenesis.

We found that the presence of sperm aster changed the alignments of F-actin as we observed a less-dense and less-organized F-actin network at the location where the sperm aster was observed (Figure 5, A and D). The sperm aster and associated microtubule system alter the morphological change of the Ciona egg after fertilization (Sardet ; Roegiers ) and the following cytoplasmic rearrangement that reallocates myoplasm and associated postplasmic/PEM RNAs to the future posterior side of the embryo (Sardet ; Makabe and Nishida, 2012). The roles of the sperm aster for organizing the architectures of cortical F-actin have been left as unanswered important questions for future studies.

On the basis of previously proposed models (Jeffery and Meier, 1983; Jeffery, 1984; Sawada, 1983) and our observations of F-actin alignments in Ciona eggs before and after fertilization, we propose a model including locally distinct F-actin alignments that might generate the force for cytoplasmic reorganization in fertilized egg (Figure 6). In this model we speculate that there are two orthogonal alignments of F-actin at the vegetal hemisphere and around the equatorial domain. At the vegetal hemisphere, actin filaments align latitudinal orientations that organize the constriction of the cup-shaped myoplasm along the latitudinal orientation. Like a contractile ring during cytokinesis, the contraction of laterally aligned F-actin bundles at the vegetal hemisphere squeezes the vegetal cytoplasm to the animal hemisphere. The movement of cytoplasm in the middle of the egg was somehow rectified near the animal pole and was converted to the stream that moves the cortex around the animal pole toward the equator along the longitudinal orientation. This cortical flow is associated with either passive or active alignment of actin filaments along the A–V axis. We postulate that the first cytoplasmic reorganization is thus accomplished through these cortical movements associated with the locally organized contractility of actomyosin at the cortex of ascidian eggs.

FIGURE 6:

Local control of F-actin alignments for the first ooplasmic segregation. (A) Schematic drawings of cortical F-actin alignment changes before (left) and after (right) fertilization. The contraction of actomyosin at the vegetal hemisphere and resulting compression of vegetal cytoplasm leads to the cortical flow at the animal hemisphere that reallocates maternal factors in the cell cortex of the fertilized C. intestinalis egg (right). (B) Schematic drawings of intracellular Ca2+ waves in the fertilized C. intestinalis egg. Left: Scheme of the cross-section of a Ciona unfertilized egg showing the locations of the meiotic spindle and myoplasm. Right: Three patterns of the propagation of the Ca2+ wavefront across the egg. From left to right, fertilization-induced wave, sequestering wave, and pacemaker wave.

One of the challenges throughout our fluorescence polarization imaging of Ciona eggs was the moderate success rate when using the TIRFM system. The tight attachment of Ciona eggs to the glass coverslip (this is required for TIRFM imaging) interfered with the dynamic movement of the egg surface toward the vegetal pole after fertilization (Roegiers ), which caused abnormal deformation of the eggs before the first cell division. Instead of TIRFM, SPIM is a more favorable imaging system for Ciona eggs as it allows us to image suspended cells without tightly holding the cell surface. It is our future plan to develop SPIM with fluorescence polarization analysis, which will allow us to analyze the 3D orientations of F-actin and other protein assemblies in living cells/tissue and whole animals/plants in 3D space.

Bement, von Dassow, and their collaborators have reported an interesting observation of F-actin waves in the cell cortex of Xenopus and echinoderm embryonic cells in anaphase (Bement ). We have observed traveling waves of F-actin in Ciona eggs that were similar to those observed in Xenopus and echinoderms in size and speed, after the release of second polar body, which corresponds to anaphase before the cytokinesis of fertilized eggs (see Supplemental Movie S4 and S5; from 7 to 10 s in the movie). Our analysis of fluorescence polarization orientation revealed that there were alignments of F-actin in the waves. Bement proposed a model of F-actin and Rho as a reaction-diffusion system that represents excitable dynamics. Our unique molecular orientation approach will help in exploring the architectural dynamics of traveling F-actin waves in Xenopus eggs and eggs of other animal species.

MATERIALS AND METHODS

Request a protocol through Bio-protocol.

Collection of animals and the preparation of eggs

The ascidians Ciona intestinalis were collected from local yacht harbors near the Marine Biological Laboratory, Woods Hole, Massachusetts. The eggs collected from the oviduct of dissected Ciona were dechorionated with 1% sodium thioglycolate (Sigma Aldrich) and 0.05% protease (from Streptomyces griseus; Sigma Aldrich) dissolved in filtered natural seawater (FNSW) (Ishii ).

SPIM imaging

We used a home-built a SPIM for monitoring the Ca2+ waves and the cytoplasmic movements of Ciona eggs after fertilization. Laser light sheet illumination was created by scanning beams from laser heads (488 nm, 20 mW, Sapphire, Coherent; 561 nm, 35 mW; 85-YCA; Melles Griot) with electromagnetic scanning mirrors (GVS202; Thorlabs). The scanned laser beams formed a light sheet at the specimen plane through an Olympus 4 × 0.25NA dry objective lens. The amplitude and the scan rate of the laser beams were controlled by a dual-channel function generator (Model 4087; BK precisions). Fluorescence emission was collected through a NA0.95, XLUMPLANFL 20× Olympus water immersion lens. A green filter (529/24 nm; Semrock) was used for autofluorescence imaging, and a red filter (624/40 nm; Semrock) was used for Rhod-dextran or fluorescent bead imaging. Fluorescence images were taken with an EMCCD camera (iXon Ultra; Andor Technology). Collected images were fed into the computer (Precision 390; Dell) for further image analyses.

Instantaneous fluorescence polarization imaging

The instantaneous FluoPolScope (Mehta ) was used to observe the orientation of local F-actin. Fluorescence polarization of AF488-phalloidin speckles reports the local orientation of F-actin (Mehta ). The laser (DPSS laser BCD; 488 nm, 20 mW; Melles Griot) for excitation was circularly polarized by a quarter-wave plate (WPQ-4880–4M; Sigma-koki) and focused at the back focal plane (BFP) of the objective lens (PlanApo TIRFM 60 × 1.49 N.A. oil; Nikon) of an inverted microscope (TE2000-E; Nikon). The circularly polarized laser beam was rotated in the BFP of the objective lens, with rotational frequency higher than 300 Hz to achieve isotropic illumination at the specimen plane. The beam was scanned along a circular annulus in the BFP using a pair of galvo-scanned mirrors (GVS202; Thorlabs). Fluorescence images were taken with an EMCCD camera (iXon+; Andor Technology). The pixel size of the EMCCD camera at the specimen plane was measured to be 180 nm. Emission filters we used were as follows: 529/24 nm transmission filter for AF488-phalloidin (Molecular Probes), 609/54 nm for Rhod-dextran (Molecular Probes) and for mCherry.

Preparation of eggs for fluorescence imaging

A solution of 5 µM AF488-phalloidin in distilled water (DW) was microinjected into the cytoplasm of dechorionated Ciona eggs for staining F-actin. The final concentration of AF488-phalloidin was estimated to be as low as 5–10 nM, based on the volume of microinjected AF488-phalloidin solution that was approximately from 1/1000 to 1/500 of the total cell volume of Ciona eggs. In general, fluorescent phalloidin as low as 10 nM has little effect on the assembly/disassembly dynamics of actin in live cells (Tani ; Yam ; Mehta ). For imaging spatial/temporal changes of intracellular Ca2+ concentration, we used Rhod-dextran. Rhod-dextran (50–500 µM) in distilled water was microinjected into the eggs. The final concentration of Rhod-dextran was estimated to be 50–500 nM, based on the volume of Rhod-dextran solution injected into the eggs. For imaging dynamics of microtubules in Ciona eggs, we constructed fluorescent protein fused microtubule plus-end binding protein EB1 from the Ciona EST clone (clone ID; cieg011b01) provided by Noriyuki Satoh, OIST, Japan, through the RIKEN BioResource Center (Satou ). The open reading frame was amplified by PCR and inserted into the BmtI/BspEI site of the pRSETB-mCherry vector (Addgene plasmid 32382). The construct was transformed into Escherichia coli strain BL21(DE3). The CiEB1-mCherry proteins expressed with histidine tag were purified using a Ni-NTA column (GE Healthcare), and then the solvent was exchanged with DW by using a NAP-5 column (GE Healthcare). Purified CiEB1-mCherry in DW was microinjected into the dechorionated Ciona eggs together with AF488-phalloidin solution. The concentration of CiEB1-mCherry was estimated to be 20–40 nM, based on the volume of microinjected CiEB1-mCherry solution and the concentration of CiEB1-mCherry (18.2 µM). For tracking the movement of cytoplasmic reorganization processes, we microinjected 100-nm-diameter fluorescent polystyrene beads (Molecular Probes) to the eggs. Diluted solutions (1/100; in DW) of the original 2% suspension of the beads were microinjected to the eggs. The volume of microinjection was 1/1000 to 1/500 of the total volume of Ciona eggs.

Ciona eggs with microinjected fluorescence labels/indicators were mounted on washed glass coverslips (No. 1; Matsunami Glass). After letting the eggs settle on the glass, we observed the eggs with transmitted light to check the distribution of mitochondria and yolk granules. One 106th dilution of sperm suspensions was applied for fertilizing dechorionated eggs. For TIRFM imaging, dechorionated eggs were tightly attached to the surface of the glass coverslips. However, in ascidian development, movement of the egg surface is essential for the normal development of the embryo (Roegiers ). To minimize the interference of the tightly attached glass surface to the membrane dynamics of fertilized eggs, we added bovine serum albumin (BSA; Sigma Aldrich) to FNSW. We confirmed that in the seawater with 50 µg/ml BSA, approximately 67% of fertilized Ciona eggs showed normal first cell division. We set the first cell division as the benchmark to judge whether the normal development processes take place. The temperature of the imaging chamber was kept at 18°C throughout image acquisitions.

Image analyses

The polarization of a fluorescent particle was characterized by its degree of polarization, called the polarization factor (p), and the orientation (φ) of its maximum polarization. The particle intensity recorded in each quadrant is related to the average intensity across the four quadrants (I), the polarization factor, the orientation (φ), and the background intensity (Ibg) as follows:

where n= 0°, 45°, 90°, 135°.

Retrieval of the polarization orientation and the polarization factor of fluorescent objects are efficiently expressed in terms of the Stokes parameters of the polarization-resolved emission:

The above relationship can be written in matrix form as S = M(I−Ibg), where I = [I I I I] is the column vector of input intensities, S = [S] is the column vector of Stokes parameters, and M, called the instrument matrix, is the matrix of coefficients that relate these two vectors according to Eq. 2. The matrix M in the above equation is replaced by the instrument matrix that is generated using the tangential polarizer as described previously (Mehta ).

The total particle intensity, polarization factor, and ensemble orientation were obtained from Stokes parameters as follows:

For other image analyses such as for quantitatively analyzing the propagation of Ca2+ waves, cytoplasmic flow, and the contraction of cortical membrane areas, we used Fiji, an open-source image processing package based on ImageJ (Schindelin ). For analyzing the propagation orientation of Ca2+-sequestering waves at the narrow egg surface observed by TIRFM, the wavefront was manually determined and the line normal to the wavefront was defined as the orientation. The movement of cytoplasmic particles was tracked by using the MTrackJ plug-in (Meijering ).

Click here for additional data file.

Click here for additional data file.