Original Article

Immunology and Cell Biology (2012) 90, 955–965; doi:10.1038/icb.2012.37; published online 17 July 2012

Starfish ApDOCK protein essentially functions in larval defense system operated by mesenchyme cells

Ryohei Furukawa1,2, Hiromi Funabashi2, Midori Matsumoto1,2 and Hiroyuki Kaneko1

  1. 1Department of Biology, Research and Education Center for Natural Sciences, Keio University, Yokohama, Japan
  2. 2Center for Chemical Biology, School of Fundamental Science and Technology, Graduate School of Science and Technology, Keio University, Yokohama, Kanagawa, Japan

Correspondence: Professor H Kaneko, Department of Biology, Research and Education Center for Natural Sciences, Keio University, Kohoku-ku, Hiyoshi 4-1-1, Yokohama, Kanagawa 223-8521, Japan. E-mail: theband@z3.keio.jp

Received 6 June 2012; Accepted 19 June 2012
Advance online publication 17 July 2012



In larvae of the starfish, Asterina pectinifera, mesenchyme cells operate in the defense system through various behaviors. We have investigated mesenchyme cell dynamics during the immune response by identifying ApDOCK, a new member of the DOCK180 superfamily protein. In 4-day-old bipinnaria larvae processed for morpholino oligonucleotide-mediated knockdown of ApDOCK, injection of inorganic foreign substances revealed that (1) mesenchyme cells fail to undergo either directed migration toward a large oil-droplet or persistent spreading on the oil-droplet after contact; (2) neither uptake of micro-beads nor cell-to-cell fusion on the large oil-droplet differed from that of mesenchyme cells from control larvae. Similar behaviors were also recorded in experiments where bacteria were injected. Under culture conditions, the expression level of ApDOCK mRNA was significantly associated with the immunological behavior of mesenchyme cells. Apparently, the mesenchyme cells from ApDOCK loss-of-function larvae exhibited insufficient lamellipodium formation via lack of fibrous form of actin organization at the leading edge. These results suggest that the migratory congregation and persistence of encapsulation of larval mesenchyme cells are intracellularly regulated by ApDOCK protein, and this regulation is associated with organization of cytoskeletal actin.


Cell migration; DOCK180 superfamily; encapsulation; lamellipodium formation; larval mesenchyme cells; starfish

The starfish Asterina pectinifera is an echinoderm that is phylogenetically close to the origin of the deuterostome. We have previously reported that the larva of A. pectinifera is equipped with a simple defense system in which a single type of mesenchyme cell operates as an immunocompetent cell.1 The structural and functional aspects of the defense system are as follows. At the 4-day-old bipinnaria larval stage, the number of mesenchyme cells ranges from 150 to 190 per larva, corresponding to ~1% of the total cell number. Most of these cells are distributed beneath the ectodermal and endodermal walls, which are composed of an epithelial monolayer. Each mesenchyme cell develops cellular processes to contact other mesenchyme cells. The location of cells fluctuates in time in the immunological naive situation. Thereby, the overall appearance of mesenchyme cells is viewed as a three-dimensional and dynamic network similar to a basket-like structure (see also Figure 2). In the pathological and physiological situations of larvae, the mesenchyme cells phagocytose cell debris effluxed from the epithelial layers and foreign substances taken into the blastocoel through the body wall, respectively.

With regard to defense cell behaviors, the functions of larval mesenchyme cells have been uncovered through injection experiments with various foreign substances.1 When a relatively large amount of bacteria is injected into the blastocoel of the larval anterior portion where mesenchyme cells are scarcely distributed, multiple mesenchyme cells undergo migratory congregation from several directions. Then, they activate concomitantly to form an aggregate and phagocytose the bacteria. During this process, the mesenchyme cells frequently undergo cell-to-cell fusion with one another to form multinucleated giant cells by an ‘encapsulation’ strategy that completely seals off the foreign body. They are finally dispersed from the region injected with bacteria, thereby cleaning the blastocoel. In another injection experiment where the bacteria are replaced by an oil-droplet with a size similar to that of the injected bacteria, the encapsulation behavior of the mesenchyme cell is more clearly observed. Encapsulation of the oil-droplet is accomplished by an extraordinary extension of the lamellipodium of each mesenchyme cell, and is not cancelled thereafter. In this way, the defense system of larva is underpinned by a variety of different mesenchyme cell dynamics to achieve migratory congregation, encapsulation, phagocytosis and multinucleated giant cell formation. Therefore, the larval mesenchyme cell is functionally similar to the vertebrate macrophage.

To improve our understanding of the larval defense system of A. pectinifera, we have undertaken studies to elucidate the molecular mechanism(s) that regulates mesenchyme cell dynamics during the immune response to foreign bodies. We recently reported that the starfish ApSRCR1 protein, the orthologue of a vertebrate scavenger receptor cysteine-rich-domain-containing protein, serves as an opsonin against bacteria in the larval defense system of this animal.2 ApSRCR1 protein is extracellularly released by mesenchyme cells and promotes their phagocytosis and aggregate formation. On the other hand, little is known about how the mesenchyme cell is regulated by intracellular molecule(s) to undergo its many defense cell behaviors.

On the basis of previous studies, DOCK180 superfamily proteins are expected to be the candidates for critical intracellular regulators in the defense cell behaviors of larval mesenchyme cells. DOCK180-related proteins are known to be guanine nucleotide exchange factors and have a shared intramolecular structure, wherein two protein domains, termed DOCK homology region-1 (DHR-1) and DHR-2, are arranged.3 These domains are involved in the recruitment of DOCK180 superfamily proteins toward the plasma membrane to rearrange cytoskeletal actin in concert with other types of molecules, such as ELMO, Crk, and Rac proteins.3, 4 They have also been shown to regulate several important cell behaviors, such as the spreading and migration of mammalian epithelial cells,5 phagocytosis of apoptotic cells and cell migration of the distal tip cells in C. elegans,6 cell migration of epithelial cells and cell-to-cell fusion of myoblast cells in D. melanogaster,7, 8 cell-to-cell fusion of myoblast cells in zebrafish9 and axon growth of hippocampal neurons in rat,10 among others. Thus, DOCK180-related proteins have been structurally conserved in a wide variety of animals during evolution, and function as intracellular regulators of cell dynamics. Interestingly, the DOCK2 protein, which is a member of the DOCK180 superfamily proteins, is exclusively expressed in hematopoietic cells including B- and T-lymphocytes,11 neutrophils,12 natural killer cells13 and macrophages.11 Significant efforts have been made to identify the regulatory mechanism of DOCK2 in these hematopoietic cells, with a focus on chemotactic migration.4, 12, 14

In the present study, we report the first identification of a DOCK180 superfamily protein in the starfish, which we have called ApDOCK protein after the species name of A. pectinifera. We show that ApDOCK is a new member of the DOCK-A subfamily of DOCK180-related proteins. Following the molecular identification of ApDOCK, we attempted to prepare ApDOCK loss-of-function bipinnaria larva by using antisense morpholino oligonucleotides (MOs). By means of foreign body injection into the ApDOCK loss-of-function larvae, we show that ApDOCK regulates larval mesenchyme cells only during the immune response, and especially in migratory congregation and persistence of encapsulation. Some of these results have also been captured in live specimens under time-lapse microscopy (Supplementary Movies 1–7).



Molecular structure of ApDOCK

A DOCK180 orthologue was identified in A. pectinifera from the complementary DNA (cDNA) library of cultured mesenchyme cells and termed ApDOCK (GenBank Accession No. AB669902). The length of the whole cDNA sequence was ~5800bp, and the open reading frame encoded a polypeptide of 1866 amino-acid residues. The deduced amino-acid sequence showed the characteristic domain architecture of an ApDOCK protein. As represented in Figure 1a, ApDOCK protein has an intramolecular arrangement comprising two regions of high sequence homology, named DHR-1 and DHR-2; these regions are functionally important domains of the DOCK180 superfamily protein (see the earlier section), and are evolutionarily conserved throughout this superfamily (Figures 1b and c). ApDOCK protein was also characterized by the presence of an N-terminal Src Homology 3 domain (yellow) and by a C-terminal proline-rich region (black) containing one Crk-binding consensus motif (PPxLPxK motif), which has been previously reported in mammalian DOCK180.15

Figure 1.
Figure 1 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Sequence analysis of ApDOCK protein. (a) Schematic diagram of the structure of ApDOCK protein. ApDOCK protein contains an Src Homology 3 domain domain (yellow) at the N-terminus and proline-rich region (black) at the C-terminal flanking region in addition to DHR-1 (blue) and DHR-2 (red) domains. (b, c) Multiple alignment of the DHR-1 (b) and DHR-2 (c) domain. Each domain in ApDOCK protein, DOCK180 (BAA09454), DOCK2 (BAA13200), MBC (NP_477144) and CED-5 (AAC38973) was aligned with ClustalW. Amino-acid residues that are conserved in more than 50% of the sequences are colored blue. (d) Phylogenetic tree of the DOCK180 superfamily. The scale bar corresponds to 0.1 changes per amino acid. The four subfamilies of the DOCK180 superfamily were previously denoted by Cote and Vouri.15 National Center for Biotechnology Information protein accession numbers are provided except for ApDOCK protein.

Full figure and legend (642K)

Figure 1d shows a phylogenetic tree of the DOCK180 superfamily including ApDOCK protein. ApDOCK protein was classified into the DOCK-A subfamily (green group in Figure 1d). In the DOCK180 subfamily group, ApDOCK protein exhibited high similarity to the congeneric proteins of other animals; this similarity is shown to not only human DOCK180 (~70% similarity), DOCK2 (67% similarity) and DOCK5 (66% similarity) in vertebrates, but also to sea urchin DOCK (84% similarity) and Drosophila myoblast city (MBC) (55% similarity) in invertebrates. In particular, the Src Homology 3 domain, DHR-1 and DHR-2 domains of ApDOCK protein showed 55%, 59% and 49% similarity to the respective domains of human DOCK180.

ApDOCK-MO does not affect the spatio-temporal arrangement of mesenchyme cells formed by normal developmental migration

To determine the effect of loss-of-function of ApDOCK, three different concentrations (1.2, 12, 120μM) of ApDOCK-MO or control-MO were injected into discharged follicle eggs and fertilized with sperm. At the respective concentrations, both types of experimental egg showed a similar time course of normal development, during which mesenchyme cells started to ingress into the blastocoel from the tip of the archenteron at the mid-gastrula stage, and subsequently migrated to the ectodermal and mesendodermal walls (n>100, respectively). By using MC5 monoclonal antibody (mAb) (see Methods), we examined the spatial distribution of mesenchyme cells from thirty 4-day-old ApDOCK-MO and control-MO larvae; these larvae were obtained from eggs that had been injected with MO at a concentration of 120μM in each of the three experiments. In both the samples all mesenchyme cells were highlighted, indicating that ApDOCK-MO did not disrupt the synthesis of MC5 antigens by mesenchyme cells (Figures 2a and b). The majority of mesenchyme cells in both samples were identically distributed along the ciliary band in the ectodermal wall and in the internal organs derived from the mesendodermal walls, such as the coelomic pouches and the digestive tract. Close-up views of individual mesenchyme cells revealed that the morphology was indistinguishable between the ApDOCK-MO and control-MO samples (Figures 2c and d): each cell had developed branched cellular processes in multiple directions. Thus, there were no marked differences in spatio-temporal distribution or morphology of mesenchyme cells between the two samples, and these characteristics were also comparable to those of mesenchyme cells during normal development, as reported previously.1, 16

Figure 2.
Figure 2 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Spatial distribution and morphology of mesenchyme cells from 4-day-old bipinnaria larva obtained from eggs injected with 120μM control- or ApDOCK-MO. (a, b) Spatial distribution of mesenchyme cells in the control-MO (a) and ApDOCK-MO (b) larvae. (c, d) A single mesenchyme cell from the control-MO (c) and ApDOCK-MO (d) larvae. Both cell samples were immunofluorescently stained with MC5 mAb, and are shown by means of the stacking function of the confocal microscope. Scale bars: 50μM (a and b), 5μM (c and d).

Full figure and legend (175K)

ApDOCK loss-of-function suppresses the directed migration of larval mesenchyme cells toward a large oil-droplet in a dose-dependent manner

As described in the earlier section, multiple mesenchyme cells vigorously undergo migratory congregation against foreign bodies. To examine the regulatory function of ApDOCK protein in mesenchyme cell dynamics, a large oil-droplet (diameter, 20μm) was injected into the blastocoel of 4-day-old control-MO or ApDOCK-MO larvae (MO concentration, 120μM). Figure 3 shows an experiment in which the injection site was defined at the central region of the blastocoel in the larval anterior portion (Figure 3a, red circle), wherein few mesenchyme cells were distributed. In the control-MO larvae, the mesenchyme cells showed a typical response to the oil-droplet at 2h: approximately six to eight mesenchyme cells had congregated around the oil-droplet from a number of different directions, and some of them had spread over the surface of the oil-droplet while extending lamellipodium (‘encapsulation’, control-MO in Figure 3b and Supplementary Movie 1). The migratory congregation of mesenchyme cells was activated after approximately 30min. Subsequently, the ‘encapsulation’ behavior of the cells became prominent at and after 60min, and persisted all of the time. In contrast, in the ApDOCK-MO larvae, the blastocoelic region around the oil-droplet remained the same in appearance at 2h as it did before 2h, although a few of the mesenchyme cells were occasionally sensitive to the oil-droplet (ApDOCK-MO in Figure 3b). In many cases, no mesenchyme cells had migrated toward the oil-droplet until 2h (Supplementary Movie 2).

Figure 3.
Figure 3 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Effect of ApDOCK-MO on immunological reactions of larval mesenchyme cells against a large oil-droplet. (a) The two blastocoelic regions used in the injection experiments. Shown are paired Nomarski and confocal microscope images of normal 4-day-old bipinnaria larva that were immunofluorescently stained with MC5 mAb. The fluorescent photo is a stacked image consisting of seven optical sections at 1-μm intervals. Red circle: the anterior blastocoelic region where the mesenchyme cells are sparsely distributed. Yellow circle: the blastocoelic region near the ectodermal wall where multiple mesenchyme cells are densely located. (b) Immunological reactions of mesenchyme cells in 4-day-old bipinnaria larvae obtained from eggs injected with 120μM MO. An oil-droplet was injected into the blastocoelic region as shown by red circle in (a). Shown are paired confocal microscope and Nomarski images of mesenchyme cells and their nuclei that have been immunofluorescently stained with MC5 mAb (green) and PI (red), respectively. The images are shown by means of the stacking and optical sectioning functions of the confocal microscope. (c) Dose-dependent effect of ApDOCK-MO and control-MO on the immune response of larval mesenchyme cells. Boxplots show the number of the mesenchyme cells in contact with the oil-droplet 2h after injection (n=35, respectively). The number of the cells was calculated by the nucleus stained by PI. The concentrations of the injected MO are indicated. ***P<0.001; **P<0.01; NS, not significant (t-test). (d) Immunological reactions of mesenchyme cells against a large oil-droplet injected into the blastocoel as shown by yellow circle in (a). Sequential frames were captured from the time-lapse movies under a light microscope (see also Supplementary Movie 3). Times indicate minutes after injection. Note that two mesenchyme cells separated from the oil-droplet after contacting it (black arrow and arrowhead). See text for further explanation. (e) Merge of Nomarski and confocal microscope images shows insufficient encapsulation of the oil-droplet by mesenchyme cells in ApDOCK-MO larvae. The image is from an ApDOCK-MO larva that has been immunofluorescently stained with MC5 mAb at 2h after injection. Scale bars: 100μM (a); 20μM (b, d and e).

Full figure and legend (300K)

To confirm the specificity of ApDOCK-MO for the immune behavioral suppression of the mesenchyme cells, we examined the dose-dependent effects of ApDOCK-MO on the numbers of mesenchymal cells that congregate to encapsulate the oil-droplet at 2h after injection (Figure 3c). Ten-fold serial increases in ApDOCK-MO caused a significant decrease in the number of mesenchyme cells that encapsulated the oil-droplet (Figure 3c). On the other hand, the number of encapsulating mesenchyme cells in control-MO larvae remained constant in the range of six–eight cells (Figure 3c). These analyses were performed on 35 oil-droplet-injected samples for each of the three concentrations of ApDOCK-MO and control-MO in each of two experiments. We did not examine the effects of concentrations of ApDOCK-MO stronger than 120μM.

ApDOCK loss-of-function inhibits larval mesenchyme cells from persistent spreading on the large oil-droplet even when the cells can contact the oil

In the 4-day-old ApDOCK-MO bipinnaria larvae, an oil-droplet with a size equal to that in Figure 3b was injected in the vicinity of the ectodermal wall where the mesenchyme cells are densely located (Figure 3a, yellow circle). Figure 3d shows the time-sequence of the immunological response of four mesenchyme cells to the injected oil-droplet (see also Supplementary Movie 3). During the first 10min immediately after injection, two mesenchyme cells (white arrowhead and black arrow) came close to the oil-droplet and made contact with it. In the next 10min, a third mesenchyme cell (black arrowhead) got in line with the first two mesenchyme cells. By 75min, however, two mesenchyme cells had left the oil-droplet (Figure 3d; black arrow and arrowhead), and only one mesenchyme cell (white arrowhead) remained in contact with the oil-droplet (Figure 3d; white arrowhead). A fourth mesenchyme cell (Figure 3d; white arrow) disappeared from this view without coming close to the oil-droplet, even though the distance from this cell to the oil-droplet was similar to that of the three mesenchyme cells that made contact with the oil-droplet. As compared with control-MO larvae (Supplementary Movie 1), the mesenchyme cells in ApDOCK-MO larvae tended to retract their lamellipodia once they were on the surface of the oil-droplet (Supplementary Movie 3). In this way, the oil-droplet was not well encapsulated by multiple mesenchyme cells in ApDOCK-MO larvae (Figure 3e). This failure of encapsulation was observed in all 35 ApDOCK-MO larvae examined, whereas the mesenchyme cells of all of the control-MO larvae (n=20) completely encapsulated the injected oil-droplet.

We investigated whether the ApDOCK-mediated immunological behaviors of mesenchyme cells occur at an earlier developmental stage when mesenchyme cells vigorously migrate to the ectodermal and mesendodermal walls in the blastocoel. When the oil-droplet was injected into control-MO gastrula (30h after fertilization) as shown in Figure 3, no mesenchyme cells underwent encapsulation even after they contacted the oil-droplet (Supplementary Movie 4). Rather, this dynamic aspect was similar to those of the ApDOCK-MO bipinnaria larva shown in Figure 3d and Supplementary Movie 3. The mesenchyme cells did not cancel their developmental migration under this experimental condition. These observations were made from two experiments that each used 15 gastrula embryos.

Mesenchyme cells undergo normal cell-to-cell fusion on a large oil-droplet and engulfment of micro-beads in ApDOCK loss-of-function larvae

Next, we examined mesenchyme cell dynamics with regard to multinucleated cell formation and phagocytosis in 4-day-old ApDOCK loss-of-function larvae within 2h of injecting an oil-droplet or micro-beads. As we wished to observe each type of mesenchyme cell dynamics when the larval mesenchyme cells come into contact with the large oil-droplet and micro-beads (see the earlier section), we performed two experiments in which each of these foreign substances was injected in the vicinity of the ectodermal wall, as in Figure 3d. Concerning cell-to-cell fusion, we attempted to locate the nucleus of the mesenchyme cells as they were seen partially spreading over the oil-droplet. By double fluorescence staining with MC5 mAb and propidium iodide (PI), the mesenchyme cells were found to contain two nuclei in their cytoplasm and thus showed normal cell-to-cell fusion (Figure 4a). Similar results were obtained from 17 out of 20 ApDOCK-MO samples. Likewise, the mesenchyme cells revealed active engulfment of the micro-beads in all ApDOCK-MO larvae (Figure 4b). There was no marked difference in the engulfment activity of mesenchyme cells around the micro-beads between ApDOCK-MO and control-MO samples in either of two experiments (n=15 larvae).

Figure 4.
Figure 4 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Immunological reactions of mesenchyme cells against a large oil-droplet and micro-beads in ApDOCK loss-of-function larvae. In 4-day-old ApDOCK-MO bipinnaria larvae, an oil-droplet (a) and micro-beads (b) were injected into the blastocoel where multiple mesenchyme cells are located, and then fixed with PFA 2h later. (a) Paired confocal microscope and Nomarski images that were processed for staining with MC5 mAb and PI are shown by means of the stacking and optical sectioning functions of the confocal microscope, respectively. Note that two nuclei (red) are contained in the mesenchyme cell spreading on the oil-droplet. (b) Paired confocal microscope and Nomarski images. In the three pairs of images, three mesenchyme cells are seen to engulf four, six and one micro-bead(s), respectively. See text for further explanation. Scale bars: 20μM.

Full figure and legend (81K)

Mesenchyme cells in ApDOCK loss-of-function larvae display similar defensive behaviors against the injected bacteria

To generate a more realistic immune response, we conducted the injection experiment with bacteria that had been previously labeled with rhodamine. Two hours after injection of bacteria into the central region of the blastocoel of 4-day-old control-MO larva, multiple mesenchyme cells formed an aggregate that encapsulated bacteria within (Figure 5a: control-MO). On the other hand, in ApDOCK-MO larva, approximately the same numbers of mesenchyme cells were dispersed without forming an aggregate(s) and the individual cells phagocytosed the bacteria (Figure 5a: ApDOCK-MO). Leftover bacteria were widely observed in the area of injection in ApDOCK-MO larva. These effects were observed in all of the 30 control- MO and 23 ApDOCK-MO larvae examined.

Figure 5.
Figure 5 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Immunological reactions of mesenchyme cells against bacteria in ApDOCK loss-of-function larvae. Four-day-old control- or ApDOCK-MO bipinnaria larvae were injected with rhodamine-labeled bacteria (red) into the arbitrary region of the blastocoel in the anterior portion of the larva, and immunohistochemically stained by MC5 mAb (green) at 2h after injection. (a) Typical images of mesenchyme cells after the bacteria had been injected at the central region of the blastocoel, as performed in Figure 3. Note that the mesenchyme cells form an aggregate in the control-MO, but not in the ApDOCK-MO samples. The green and red merged image shows that the bacteria were phagocytosed by mesenchyme cells. (b) Another image of ApDOCK-MO mesenchyme cells. Here, injection of bacteria was carried out in the vicinity of the ectodermal wall wherein the mesenchyme cells were closely distributed with one other. Arrow indicates that cell-to-cell fusion occured between two mesenchyme cells. Confocal microscope images were obtained by the stacking and optical sectioning functions for the immunofluorescent and Nomarsky photographs, respectively. Scale bars: 20μM.

Full figure and legend (92K)

Figure 5b shows an experiment in which the rhodamine-labeled bacteria were injected in the immediate vicinity of the ectodermal wall of ApDOCK-MO larva, wherein the mesenchyme cells were relatively abundant, as shown in Figure 3d. In this case, most of the individual mesenchyme cells phagocytosed bacteria, but they failed to form any easily detected aggregates. Occasionally, however, there was an indication of cell-to-cell fusion, as confirmed by more than two mesenchyme cells that were closely connected to each other by the bulk of their cell bodies (Figure 5b; arrow). Even in this situation, part of bacteria remained unphagocytosed. Similar images were obtained from all 7 ApDOCK-MO larvae.

The expression level of ApDOCK mRNA is significantly altered with the progress of the immune reaction of mesenchyme cells in culture

We profiled ApDOCK regulation at the mRNA level with regard to the behavior of the mesenchyme cells during the immune response. To this end, we used mesenchyme cells isolated on a culture dish from 4-day-old bipinnaria larvae (see Methods). After being allowed to spread out for 4h, the mesenchyme cells exhibited a network pattern in which each cell was associated with each other by cellular processes (Figure 6a). They fidgeted at their position, while preserving the structural network pattern under the naive situation (see Supplementary Movie 5). Figure 6b shows the time course of the immunological reaction of mesenchyme cells when they were challenged with bacteria in culture. After 30min, the mesenchyme cells continued to rapidly move about the field, thereby distorting the original network pattern. Throughout this period, the mesenchyme cells continued to phagocytose bacteria (see Supplementary Movie 5, arrows). The same results were obtained in each of the two experiments.

Figure 6.
Figure 6 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Time dependency of ApDOCK mRNA expression in cultured mesenchyme cells at 30, 60, 120min after challenge with bacteria. (a) Phase-contrast microscopy of cultured mesenchyme cells prepared from 4-day-old bipinnaria larvae. The mesenchyme cells crawled out from the extracellular matrix over a period of 4h. The cells enclosed by the rectangle are also shown in (b) and Supplementary Movie 6. (b) Phase-contrast micrographs of the mesenchyme cells after the bacteria were applied in culture. Each photograph was captured from Supplementary Movie 5. (c) Real-time PCR of ApDOCK mRNA in the bacteria application experiment. Data are shown as the level of ApDOCK mRNA expression relative to that in naive mesenchyme cells (0min). Vertical bars represent the mean±s.e. (n=3). Scale bars: 20μM.

Full figure and legend (169K)

Next, we examined the expression profile of ApDOCK mRNA in the bacteria application experiment (Figure 6c). At 30min, ApDOCK mRNA expression had increased 1.7-fold as compared with naive mesenchyme cells at 0min. After 60min, the expression of ApDOCK mRNA began to decrease to the original level when the bacteria challenge was started.

ApDOCK loss-of-function results in not only imperfect lamellipodium formation but also deficient membrane ruffles at the leading edge of larval mesenchyme cells in the culture

Under the same culture conditions used in Figure 6, we examined the effect of ApDOCK loss-of-function on the detailed morphology and cell dynamics of mesenchyme cells. The mesenchyme cells from control-MO larvae highly extended the lamellipodium over a large area (Figure 7a), while dynamically changing their shape by expansion and contraction of the lamellipodium with the membrane ruffles (Supplementary Movie 6). In contrast, mesenchyme cells from ApDOCK-MO sample were apparently deprived of part of the lamellipodium (Figure 7a). In addition, their shape change was relatively slow as compared with that of cells from control-MO larvae (Supplementary Movie 7). In a comparison of the area occupied by individual single mesenchyme cells (see Methods), the area occupied by mesenchyme cells from control-MO larvae was an ~1.4-fold greater than that from ApDOCK-MO larvae (P<0.001, t-test, three experiments) (Figure 7b).

Figure 7.
Figure 7 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Appearance of lamellipodium of cultured mesenchyme cells from ApDOCK loss-of-function larvae. (a) Phase-contrasted image of mesenchyme cells prepared from 4-day-old control-MO and ApDOCK-MO bipinnaria larvae (4h after inoculation). Note that the lamellipodium are markedly underdeveloped in mesenchyme cells from ApDOCK-MO larvae, thereby appearing as a refractile morphology. See text for further explanation. (b) Quantitative analysis of the expanding area of a single mesenchyme cell (4h after inoculation). The mesenchyme cells of ApDOCK-MO larvae occupy 434.4±29.2μM2 on average by extension of their lamellipodia (n=70), while the experimental counter mesenchyme cells (control-MO larvae, n=72) occupy 625.3±39.5μM2 (means±s.e.). Dots represent the outliers, ***represents P<0.001 in t-test. (c) Effects of ApDOCK loss-of-function on F-actin distribution in mesenchyme cells in in vitro. Shown are confocal microscope and Nomarski (differential interference contrast microscope; DIC) images of an optical section that was immunohistochemically double-stained with MC5 mAb (MC5) and rhodamine-phalloidin (F-actin). Mesenchyme cells obtained from control-MO and ApDOCK-MO larvae were fixed with PFA 4h after inoculation. Note that some F-actin was concentrated at the membrane ruffles at the edge of the lamellipodia of mesenchyme cells from control-MO larvae (arrowheads). n: nuclei. Scale bars: 50μM (a); 20μM (c).

Full figure and legend (289K)

Assembly of the fibrous form of actin (F-actin) has been known to be closely involved in the formation of membrane ruffles.17 We performed double fluorescence analysis with rhodamine-phalloidin (F-actin marker) and MC5 mAb (mesenchyme cell surface marker) to observe the distribution of F-actin in larval mesenchyme cells under the same culture conditions as in Figure 7a. In mesenchyme cells from control-MO larvae, the fluorescent signal of rhodamine-phalloidin was concentrated at the leading edge of lamellipodia, where the fluorescent signal of MC5 mAb was also condensed (Figure 7c; arrowheads of upper panel). This co-localization image of F-actin and MC5 antigen was obtained for 3 of 28 mesenchyme cells in two experiments. In contrast, in a mesenchyme cell from ApDOCK-MO larvae, the MC5 mAb exhibited no condensed pattern of fluorescent signals (Figure 7c; lower panel) (32 cells examined in two experiments). The fluorescent signal of phalloidin could be also detected throughout the cytoplasm of the mesenchyme cells. On the other hand, between the control-MO and ApDOCK-MO larvae, there was no obvious difference in the intensity of the fluorescent signals of rhodamine-phalloidin at the filopodia of the mesenchyme cells (Figure 7c; arrows). Except for these regions, the intensity of fluorescent signals of rhodamine-phalloidin was weaker in the cytoplasm of mesenchyme cells from ApDOCK-MO larvae than in that of mesenchyme cells from control-MO larvae.



In the present study, we identified a cDNA clone encoding a protein that belongs to the DOCK180 superfamily in the starfish Asterina pectinifera, and have termed it ApDOCK. We prepared ApDOCK-MO-injected 4-day-old bipinnaria larvae and their experimental counterparts (control-MO samples), and principally examined the effect of loss-of-function of ApDOCK on several types of mesenchyme cell dynamics in the larval defense system (see the earlier section). A combined experiment with oil-droplet injection and dose-dependent studies of ApDOCK-MO confirmed the specificity of ApDOCK-MO as a reliable tool to examine mesenchyme cell dynamics in both a series of injection experiments (Figures 3, 4, 5) and analyses of the detailed morphology of mesenchyme cells under culture conditions (Figure 7). Furthermore, the real-time PCR approach using a culture system of mesenchyme cells underpins the actual involvement of ApDOCK in the innate immune system of the starfish larvae (Figure 6). Below, our findings on ApDOCK protein are discussed with regard to its structural features, immunological roles and the possible regulatory mechanism of mesenchyme cell dynamics during the defense processes.

Phylogenetic tree analysis disclosed that ApDOCK is an ancestral protein of the deuterostome DOCK-A subfamily (Figure 1d). It has been well documented that proteins of the DOCK-A subfamily share an intramolecular structure comprising Src Homology 3 domain, DHR-1 and DHR-2 domains, which are involved in recruitment of the proteins to the plasma membrane3 and directive organization of the actin cytoskeleton.18 In this way, in many cell types, proteins of the DOCK-A subfamily has essential and also common roles in cell dynamics, such as cell migration, phagocytosis, cell-to-cell fusion and shape determination (see the earlier section). It should be noted, however, that DOCK2 protein in the DOCK-A subfamily is restricted to expression in hematopoietic cell lines.4, 11, 14 Intriguingly, this defined expression of DOCK2 indicates a functional identity with ApDOCK protein, because both proteins has an essential role(s) in immunological situations (Figures 2, 3, 4, 5, 6). The Crk-binding consensus motif (Figure 1a: PPxLPxK) raises concern about a structural difference between the DOCK2 and ApDOCK proteins. This consensus motif is absent from DOCK2, but present in ApDOCK protein. Crk is a well-known adaptor protein that mediates the recruitment of cytoplasmic proteins to tyrosine kinase19, 20 and that interacts with proteins of the DOCK-A subfamily.9, 21, 22 It might be possible that, although the role of ApDOCK protein in the immune response is conserved in evolution, the Crk-binding consensus motif of ApDOCK protein has been replaced by another sequence in DOCK2.

Our results indicate that ApDOCK protein regulates mesenchyme cell dynamics exclusively in the larval defense system. The injection experiments, in which large oil-droplet and micro-beads were injected at arbitrary blastocoelic regions, facilitated direct visualization of distinguishable overlapping mesenchyme cell dynamics; this regulation is essential for migratory congregation and persistence of encapsulation. In addition to these inorganic foreign bodies, the injection of bacteria also confirmed that the mesenchyme cell is regulated by ApDOCK, and its regulation defines the response regarding congregated migration but not phagocytosis (Figure 5a). In ApDOCK-MO samples, demonstration of the failure of mesenchyme cells to encapsulate the bacteria (Figure 5b), even though some mesenchyme cells were closely located to one other, is considerably similar to that seen in Figure 3d. The regulation of mesenchyme cells dynamics by ApDOCK protein was furthermore demonstrated under culture conditions with applied bacteria (Figure 6). As shown in Figure 6, expression of ApDOCK was altered at the mRNA level, including both up- and downregulation, in a time-dependent manner in response to the application of bacteria. Interestingly, the mesenchyme cell seems to increase its migration activity in concert with timing of upregulation in ApDOCK mRNA, regardless of whether phagocytosis occurs or not (Supplementary Movie 5). For this culture system, details are currently unknown regarding other cell dynamics, encapsulation and cell-to-cell fusion, or regarding the relationship between the migration activity of mesenchyme cells and downregulation of ApDOCK at the mRNA level.

Actin organization under control of DOCK2 provides the crucial mechanism for chemotactic migration of several types of immune cells.12, 14, 23 This would be also the case when the larval mesenchyme cells undergo migratory congregation, which is ostensibly a similar process to chemotactic migration, although no chemokine molecules have been identified as yet in the larval defense system of A. pectinifera. In addition, actin organization under control of ApDOCK protein is also essential for persistence of encapsulation on the foreign body (Figures 3d and e), because in culture conditions larval mesenchyme cells from ApDOCK loss-of-function larvae failed to organize sufficient F-actin at the leading edge of lamellipodium (Figure 7c), which could only exhibit a vigorless ruffling movement (Supplementary Movie 7). Furthermore, attention should be paid to lamellipodium formation, which is significantly inhibited in mesenchyme cells from ApDOCK loss-of-function larvae (Figures 7a and b). We therefore consider that ApDOCK protein regulates lamellipodium formation via actin organization, which contributes to generation of a physical force that permits the persistence of encapsulation. This possibility would explain why larval mesenchyme cells from ApDOCK loss-of-function larvae can only undergo encapsulation transiently (Figure 3d and e, and Supplementary Movie 3). Similarly, larval mesenchyme cells are certain to require a physical force generated by actin organization under control of ApDOCK to undergo congregated migration (Figure 3b).

On the other hand, it should be noted that some types of mesenchyme cell dynamics, other than migratory congregation and persistence of encapsulation, does not appear to be regulated by ApDOCK protein. Remarkably, the mesenchyme cells from ApDOCK loss-of-function larvae underwent cell-to-cell fusion on the large oil-droplet and uptake of micro-beads. Although mesenchyme cells numbers are considerably few to evaluate multinucleated giant cell formation, cell-to-cell fusion was evidenced by the near location of two nuclei (Figure 4a). Regarding the uptake of micro-beads, although quantitative analysis was not conducted, the numbers of micro-beads (Figure 4b) were comparable to those seen in normal 4-day-old larvae.1 Previous studies have shown that the vertebrate macrophage shares a lot of proteins for cell-to-cell fusion and phagocytosis,24 and also both types of cell dynamics are regulated by DOCK180 protein not but DOCK2 protein.21, 25 Thus, in the larval mesenchyme cell, DOCK180-related proteins other than ApDOCK protein might regulate multinucleated giant cell formation and phagocytosis via cell-to-cell fusion and uptake dynamics.

Our study is not only a functional demonstration of a new member of the DOCK180 superfamily protein but also the first report of the immunological function of a DOCK180 orthologue in a marine invertebrate. In addition, ApDOCK protein is the first protein to be identified as a regulator of actin organization involved in cell behaviors in echinoderms. In order to understand the evolution of mechanisms that regulate the behaviors of immune cells, further studies are needed to explore the target(s) of ApDOCK protein, such as the Rho GTPase family member Rac, which are well known to be positioned downstream of DOCK180 superfamily protein.3 At present, a member of the Rho GTPase family is yet to be identified in starfish species. The present study raises two significant issues that should be addressed in future studies. The first issue pertains to the other roles of ApDOCK protein when the mesenchyme cell operates in the larval defense system. Except for chemotactic migration of hematopoietic cells (see the earlier section), DOCK2 protein has been shown to regulate the production of immune modulators, such as interleukin-2 in T cells,25 type 1 Interferon in dendritic cells26 and the inflammasome adaptor ASC (apoptosis-associated speck-like protein containing a caspase recruitment domain) in dendritic cells and lymphocytes.27 The production of these immune modulators has been also reported in the macrophage.28, 29 Likewise, the larval mesenchyme cell might utilize orthologues of these immune modulators through ApDOCK regulation. The second issue concerns the developmental expression of ApDOCK protein. Our finding that the embryonic mesenchyme cell is indifferent to the oil-droplet, similar to mesenchyme cells in ApDOCK loss-of-function larva (Supplementary Movie 4) indicates the possibility that ApDOCK is not expressed until the embryo reaches the bipinnaria stage. A specific antibody against ApDOCK protein would enable us to elucidate the developmental process of the larval defense system in A. pectinifera.



Discharged follicle-free eggs

Immature eggs were obtained from the starfish Asterina pectinifera by cutting ovaries into small pieces in ASW (Artificial Seawater; MarineArt SF-1, Tomita Pharmaceutical, Tokushima, Japan). Then, they were gently suspended for five to ten times in Ca2+-free seawater (Jamarin, Osaka, Japan) to remove the follicle cells, and then placed in ASW. The composition of ASW has been previously described.2

cDNA cloning, sequencing and sequence analyses

Degenerate PCR cloning was used to explore the partial cDNA of an orthologous gene of the DOCK180 superfamily from the mesenchyme cell cDNA library that has been reported previously.2 degenerate primers, DOCK1F (5′- GCGGGAGAAGATCTTCTTCgtntgycarat-3′) and DOCK4R (5′- GGTCGGCGATCAGGccdatdatraa-3′), were designed by the CODEHOP program (http://dbmi-icode-01.dbmi.pitt.edu/i-codehop-context) using the following vertebrate and invertebrate DOCK180 superfamily members: Homo sapiens, BAA09454; Mus musculus, NP_001028592; Rattus norvegicus, XP_219424; Canis lupus familiaris, XP_544064; Equus caballus, XP_001489814; Strongylocentrouts purpuratus, XP_785550. The PCR program was performed using AmpliTaq Gold (Applied Biosystems, Tokyo, Japan) with the following cycling conditions: 5min at 94°C; 35 cycles of 94°C for 30s, 49°C for 60s, 74°C for 105s; and a final extension step at 74°C for 5min. The amplification product was sequenced and termed the ApDOCK clone. To obtain the full-length ApDOCK cDNA, total RNA was isolated from cultured mesenchyme cells as described previously.2 Then, 5′- and 3′-rapid amplification of cDNA ends (RACE) were performed using a SMART RACE cDNA Amplification Kit (Clontech, Shiga, Japan) according to the manual with the RACE primers DOCK-5′race (5′-GGGCAGCACGGGTAGCCTTTTCCGCAGC-3′) and DOCK-3′race (5′-CTTCAGGCCCACCCAGATCCCTTGGCCC-3′). All of the DNA sequencing was performed on a Fluorescence DNA sequencer (ABI PRISM 3100, Applied Biosystems) using BigDye Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems).

The ApDOCK cDNA and protein sequences were analyzed by the BLAST algorithm (http://www.ncbi.nlm.nih.gov/BLAST), InterProScan (http://www.ebi.ac.uk/Tools/InterProScan) and MOTIF (http://motif.genome.jp). The phylogenetic tree analysis was performed as described previously.15 Amino-acid sequences of human, Drosophila melanogaster, Caenorhabditis elegans, S. purpuratus, Dictyosterium discoideum and A. pectinifera family proteins were aligned with ClustalW (http://www.ebi.ac.uk/Tools/msa/clustalw2/). To generate the unrooted phylogenetic tree, neighboring-joining analysis was applied to pairwise sequence distances calculated with the PHYLIP package using the Kimura two-parameter method.15 The final output was generated with FigTree v1.1.2 (http://tree.bio.ed.ac.uk/software/figtree/).


Antisense MO (ApDOCK-MO) and control-MO were obtained from Gene Tools LLC (Philomath, OR, USA). The sequence of ApDOCK-MO was 5′-ACCCAACCAGTCATGTTTCACAGTC-3′, which is complementary to the sequence containing the translation start site of ApDOCK mRNA (underlined). The control-MO differed from the ApDOCK-MO in the sequence of five base pairs (5′-ACTCAACTAGTCACGTCTCACAATC-3′, mis-paired bases underlined). Each MO was dissolved in sterile water at 1mM to generate a stock solution, which was diluted to the appropriate concentration before injection. The discharged follicle-free eggs were injected with ApDOCK-MO or control-MO. They were allowed to develop until the 4-day-old bipinnaria stage in ASW at 20°C.16 The detailed injection procedure has been described previously.30

Silicon oil (Shin-Etsu Chemical co, Ltd, Tokyo, Japan), fluorescently labeled polystyrene latex beads (diameter 2μm: Fluoresbrite carboxy YG microsphere; Polysciences Inc., Warrington, PA, USA) and Escherichia coli (DH5α strain) were used for injection experiments in ApDOCK-MO and control-MO samples (4-day-old bipinnaria larvae) as described previously.1 E. coli were labeled by rhodamine B isothiocyanate (Sigma-Aldrich, Tokyo, Japan) using the following procedure: the bacteria were incubated for 1h at room temperature with ASW containing 0.6mM rhodamine B isothiocyanate, washed three times with ASW, and adjusted to a concentration corresponding to OD600=0.5 in ASW.

Preparation of larval mesenchyme cells in a culture dish and bacteria challenge

Mesenchyme cells from ApDOCK-MO or control-MO bipinnaria larvae were transferred to the culture dishes. The basic principle of this method has been previously reported.31 Briefly, ten bipinnaria larvae of each MO type were treated with 1ml of the dissociate medium (DM; 1.2M glycine in distilled water containing 1% ASW and 6% newborn calf serum). After 5min, the medium was replaced with 1ml of fresh DM, and the larvae were then gently suspended in a glass mouth pipet (tube diameter, ~150μm; Drummond Scientific Company, Broomall, PA, USA) to remove the ectodermal epithelial cells. These peeled larvae were dropped onto a glass coverslip in a plastic culture dish Falcon 1007 (diameter, 3.5 cm; BD, Tokyo, Japan) containing 3ml of ASW supplemented with 4% newborn calf serum. Samples were incubated at 20°C for 4h to allow the mesenchyme cells to crawl out of the extracellular matrix, and then the culture medium was refreshed. For bacteria challenge, the concentration of E. coli was adjusted to an OD600 of 0.5 in ASW, and 500μl of E. coli-ASW was added to the cultured mesenchyme cells.

Fluorescent microscopy and measurement of the extension areas of cultured mesenchyme cells

To determine the localization, morphological shape and immune response of mesenchyme cells from 4-day-old ApDOCK-MO and control-MO bipinnaria larvae, indirect immunofluorescent microscopy was performed using MC5 mAb, as described previously.1 MC5 mAb has been previously reported to react with a membranous type of metalloproteinase of the astacin family.30 Its antigen is intensely synthesized in embryonic and larval mesenchyme cells,16 and therefore it used as a ‘mesenchyme cell marker’. In some samples, the nuclei were fluorescently viewed with PI (0.1μgml−1, Wako, Osaka, Japan).1 To detect the intracellular localization of the F-actin, cultured mesenchyme cells from the ApDOCK-MO and control-MO larvae were processed on a glass coverslip for indirect immunofluorescence with MC5 mAb and then treated with rhodamine-phalloidin (1/1000 dilution, Invitrogen, Tokyo, Japan). All specimens were examined with a laser confocal microscope (Fluoview, Olympus, Tokyo, Japan).

To measure the extension area of cultured mesenchyme cells prepared from each MO sample, the cells were processed for double fluorescent staining with PI and MC5 mAb as described above. Next, the extension area was measured for each single cell of cultured mesenchyme cells by Image J software (National Institutes of Health, Bethesda, MD, USA).

Real-time PCR

A cDNA mix corresponding to cultured mesenchyme cells at each time point of the bacteria challenge experiment was synthesized by using a ReverTra Ace qPCR RT kit (TOYOBO, Osaka, Japan) as a template. A 127-bp fragment of the ApDOCK gene (nucleotides: 415–541) was amplified by a pair of primers, DOCK-qPCR-F (5′-AGAATGGGGAGTGCTGTGGA-3′) and DOCK-qPCR-R (5′-TACAGGGAGGGTGCCTGAGA-3′). As a reference gene, a 114-bp fragment of the A. pectinifera 18S rRNA gene was amplified by 18S-3-f and 18S-3-r primers as described previously.2 A five-fold serial dilution of the cDNA mix was used to generate a standard curve. Real-Time reverse transcriptase-PCR was performed by using a StepOne Real-Time PCR System (Applied Biosystems) with the KAPA SYBR FAST qPCR master mix (KAPA BIOSYSTEMS, Woburn, MA, USA). Each sample was analyzed in triplicate wells, and the PCR products were resolved by electrophoresis to confirm that the expected PCR products were obtained.

Time-lapse video microscopy

For in vivo time-lapse imaging, silicon oil-injected ApDOCK-MO or control-MO larvae or embryos were dropped between the spacer of two double-stick tapes (NM-10, Nichiban, Tokyo, Japan) that were attached to a glass bottom dish (35 mm dish, D110400, Matsunami, Osaka, Japan). They were then immobilized with a nylon mesh (pore size; 200μm), and further covered with a glass coverslip lain on the spacers. Lastly, the glass bottom dish was filled with ASW. Time-lapse observation of mesenchyme cell behavior was performed by a light microscope (IX71, Olympus) with DP controller software (Olympus). The images were captured every 5min for 2h. To investigate the lamellipodium formations and phagocytosis, time-lapse images of larval mesenchyme cells in culture were captured every 3min in the culture condition described above.



  1. Furukawa R, Takahashi Y, Nakajima Y, Dan-Sohkawa M, Kaneko H. Defense system by mesenchyme cells in bipinnaria larvae of the starfish, Asterina pectinifera. Dev Comp Immunol 2009; 33: 205–15. | Article | PubMed |
  2. Furukawa R, Matsumoto M, Kaneko H. Characterization of a scavenger receptor cysteine-rich-domain-containing protein of the starfish, Asterina pectinifera: ApSRCR1 acts as an opsonin in the larval and adult innate immune systems. Dev Comp Immunol 2012; 36: 51–61. | Article | PubMed |
  3. Cote JF, Vuori K. GEF what? Dock180 and related proteins help Rac to polarize cells in new ways. Trends Cell Biol 2007; 17 (8): 383–93. | Article | PubMed | CAS |
  4. Reif K, Cyster J. The CDM protein DOCK2 in lymphocyte migration. Trends Cell Biol 2002; 12: 368–73. | Article | PubMed | CAS |
  5. Sanders MA, Ampasala D, Basson MD. DOCK5 and DOCK1 regulate Caco-2 intestinal epithelial cell spreading and migration on collagen IV. J Biol Chem 2009; 284: 27–35. | Article | PubMed |
  6. Wu YC, Horvitz HR. C. elegans phagocytosis and cell-migration protein CED-5 is similar to human DOCK180. Nature 1998; 392: 501–504. | Article | PubMed | ISI | CAS |
  7. Rushton E, Drysdale R, Abmayr SM, Michelson AM, Bate M. Mutations in a novel gene, myoblast city, provide evidence in support of the founder cell hypothesis for Drosophila muscle development. Development 1995; 121: 1979–88. | PubMed | ISI | CAS |
  8. Erickson MRS, Galletta BJ, Abmayr SM. Drosophila myoblast city encodes a conserved protein that is essential for myoblast fusion, dorsal closure, and cytoskeletal organization. J Cell Biol 1997; 138: 589–603. | Article | PubMed | ISI | CAS |
  9. Moore CA, Parkin CA, Bidet Y, Ingham PW. A role for the Myoblast city homologues Dock1 and Dock5 and the adaptor proteins Crk and Crk-like in zebrafish myoblast fusion. Development 2007; 134: 3145–53. | Article | PubMed | CAS |
  10. Watabe-Uchida M, Govek EE, Van Aelst L. Regulators of Rho GTPases in neuronal development. J Neurosci 2006; 26: 10633–5. | Article | PubMed | CAS |
  11. Nishihara H, Kobayashi S, Hashimoto Y, Ohba F, Mochizuki N, Kurata T et al. Non-adherent cell-specific expression of DOCK2, a member of the human CDM-family proteins. Biochim Biophys Acta 1999; 1452: 179–87. | Article | PubMed | ISI | CAS |
  12. Kunisaki Y, Nishikimi A, Tanaka Y, Takii R, Noda M, Inayoshi A et al. DOCK2 is a Rac activator that regulates motility and polarity during neutrophil chemotaxis. J Cell Biol 2006; 174: 647–652. | Article | PubMed | CAS |
  13. Kunisaki Y, Tanaka Y, Sanui T, Inayoshi A, Noda M, Nakayama T et al. DOCK2 is required in T cell precursors for development of Valpha14 NK T cells. J immunol 2006; 176: 4640–4645. | PubMed | CAS |
  14. Fukui Y, Hashimoto O, Sanui T, Oono T, Koga H, Abe M et al. Haematopoietic cell-specific CDM family protein DOCK2 is essential for lymphocyte migration. Nature 2001; 412: 826–831. | Article | PubMed | ISI | CAS |
  15. Cote JF, Vuori K. Identification of an evolutionarily conserved superfamily of DOCK180-related proteins with guanine nucleotide exchange activity. J Cell Sci 2002; 115 (Part 24): 4901–13. | Article | PubMed | ISI | CAS |
  16. Hamanaka G, Hosaka E, Kuraishi R, Hosoya N, Matsumoto M, Kaneko H. Uneven distribution pattern and increasing numbers of mesenchyme cells during development in the starfish, Asterina pectinifera. Dev Growth Differ 2011; 53: 440–9. | Article | PubMed |
  17. Weed SA, Parsons JT. Cortactin: coupling membrane dynamics to cortical actin assembly. Oncogene 2001; 20: 6418–34. | Article | PubMed | ISI | CAS |
  18. Meller N, Merlot S, Guda C. CZH proteins: a new family of Rho-GEFs. J Cell Sci 2005; 118 (Part 21): 4937–46. | Article | PubMed | CAS |
  19. Kiyokawa E, Hashimoto Y, Kobayashi S, Sugimura H, Kurata T, Matsuda M. Activation of Rac1 by a Crk SH3-binding protein, DOCK180. Genes Dev 1998; 12: 3331–6. | Article | PubMed | ISI | CAS |
  20. Feller SM. Crk family adaptors-signalling complex formation and biological roles. Oncogene 2001; 20: 6348–71. | Article | PubMed | ISI | CAS |
  21. Lee WL, Cosio G, Ireton K, Grinstein S. Role of CrkII in Fcgamma receptor-mediated phagocytosis. J Biol Chem 2007; 282: 11135–43. | Article | PubMed | CAS |
  22. Smith HW, Marra P, Marshall CJ. uPAR promotes formation of the p130Cas-Crk complex to activate Rac through DOCK180. J Cell Biol 2008; 182: 777–90. | Article | PubMed | CAS |
  23. Sanui T, Inayoshi M, Noda M, Iwata E, Stein JV, Sasazuki T et al. DOCK2 regulates Rac activation and cytoskeletal reorganization through interaction with ELMO1. Blood 2003; 102: 2948–2950. | Article | PubMed | ISI | CAS |
  24. McNally AK, Anderson JM. Multinucleated giant cell formation exhibits features of phagocytosis with participation of the endoplasmic reticulum. Exp Mol Pathol 2005; 79: 126–35. | Article | PubMed |
  25. Pajcini KV, Pomerantz JH, Alkan O, Doyonnas R, Blau HM. Myoblasts and macrophages share molecular components that contribute to cell-cell fusion. J Cell Biol 2008; 180: 1005–19. | Article | PubMed | CAS |
  26. Gotoh K, Tanaka Y, Nishikimi A, Nakamura R, Yamada H, Maeda N et al. Selective control of type I IFN induction by the Rac activator DOCK2 during TLR-mediated plasmacytoid dendritic cell activation. J Exp Med 2010; 207: 721–730. | Article | PubMed | CAS |
  27. Ippagunta SK, Malireddi RK, Shaw PJ, Neale GA, Walle LV, Green DR et al. The inflammasome adaptor ASC regulates the function of adaptive immune cells by controlling Dock2-mediated Rac activation and actin polymerization. Nat Immunol 2011; 12: 1010–6. | Article | PubMed | CAS |
  28. Nathan CF. Secretory products of macrophages. J Clin Invest 1987; 79: 319–26. | Article | PubMed | ISI | CAS |
  29. Mariathasan S, Newton K, Monack DM, Vucic D, French DM, Lee WP et al. Differential activation of the inflammasome by caspase-1 adaptors ASC and Ipaf. Nature 2004; 430: 213–8. | Article | PubMed | ISI | CAS |
  30. Hamanaka G, Matsumoto M, Hoshi M, Kaneko H. Studies on function of MC5 molecule that is a novel membrane-type metalloproteinase of astacin family during morphogenesis of the starfish, Asterina pectinifera. In:Harris LG, Bottger SA, Walker CW, Lesser MP, (eds). Echinoderms: Durham. Taylor & Francis Group: London. 2010 pp 457–462.
  31. Kaneko H, Takaichi S, Yamamoto M, Dan-Sohkawa M. Acellularity of starfish embryonic mesenchyme cells as shown in vitro. Development 1990; 109: 129–138.


We thank members of the Asamushi Marine Biological Station of Tohoku University and Dr Osamu Kawase for supplying the starfish. We also thank members of Tateyama Marine Laboratory, Marine Coastal Research Center, Ochanomizu University for keeping the adult starfish. We are particularly grateful to Dr Motonori Hoshi for critical discussion. This work was supported by Keio Gijuku Fukuzawa Memorial Fund for the Advancement of Education and Research (to HK) and Keio Gijyuku Academic Development Funds (to RF).

Supplementary Information that accompanies this paper is available on the Immunology and Cell Biology website