Journal of Leukocyte Biology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Videos
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Grabher, C.
Right arrow Articles by Wittbrodt, J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Grabher, C.
Right arrow Articles by Wittbrodt, J.
(Journal of Leukocyte Biology. 2007;81:263-271.)
© 2007 by Society for Leukocyte Biology

Birth and life of tissue macrophages and their migration in embryogenesis and inflammation in medaka

Clemens Grabher*,1,2, Adam Cliffe*,1, Kota Miura{dagger},{ddagger}, Joel Hayflick{ddagger}, Rainer Pepperkok{dagger}, Pernille Rørth* and Joachim Wittbrodt*,2

* Developmental Biology Unit, and
{dagger} Cell Biology and Biophysics Unit, European Molecular Biology Laboratory, Heidelberg, Germany; and
{ddagger} ICOS Corporation, Bothell, Washington, USA

2Correspondence: C.G., Department of Pediatric Oncology, Dana-Farber Cancer Institute, 44 Binney St., Boston, MA 02115, USA. E-mail: clemens_grabher{at}dfci.harvard.edu; J.W., Developmental Biology Unit, European Molecular Biology Laboratory, Meyerhoftsr. 1, Heidelberg 69117, Germany. E-mail: wittbrodt{at}embl.de


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Macrophages detecting and migrating toward sites of injury and infection represent one of the first steps in an immune response. Here we directly image macrophage birth and migration in vivo in transgenic medaka fish. Macrophages are born as frequently dividing, immotile cells with spherical morphology that differentiate into flat, highly motile cells. They retain mitotic activity while spreading over the entire body. Cells follow restricted paths not only in directed migration, but also during patrolling. Along those paths the macrophages rapidly patrol the tissue and respond to wounding and bacterial infection from long distances. Upon injury they increase their speed and migratory persistence. Specifically targeting PI3-kinase isoforms efficiently blocks the wounding response and results in a distinct inhibition of cell motility and chemotaxis. Our study provides in situ insights into the properties of immature and migratory macrophages and presents a unique model to further test modulating compounds in vivo.

Key Words: Macrophage • cell migration • medaka • PI3K • inflammation


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Active cell movements are essential in many developmental processes including gastrulation, organogenesis and formation of the nervous and immune systems. Moreover, chemotaxis plays an important role during the entire life of an organism, with regeneration, wound healing and the function of the immune system depending on the ability of cells to migrate to specific locations in the body. To date, most of our knowledge about cell migration is based on studies in fast moving amoeboid cells and slow moving in vitro cultured cells using flat 2D substrates [1 2 3 4 ]. These in vitro studies elucidated details of the cytoskeletal machinery driving cell migration [5 , 6 ]. Monitoring migration through complex three-dimensional tissue in real-time in a living vertebrate is still in its infancy but will be essential to test, in vivo, the information gained from cell culture experiments [2 , 7 ]. Recently, synthetic 3D gel-matrix substrates have been used to mimic physiological conditions for cells in tissues [8 ] and differences in the migratory behavior and machinery in 3D are emerging [9 10 11 12 ], highlighting the importance of the three dimensional aspect of in vivo systems. For the study of inflammation, metastasis and autoimmune diseases, we need a better understanding of the migratory behavior of leukocytes (e. g. macrophages) in situ, penetrating complex tissues to reach sites of inflammation [13 ]. Current approaches toward in situ immune cell imaging in vertebrates utilize the invasive procedure of intra-vital microscopy. A vertebrate model system that allows direct, non-invasive observation of leukocyte migration in situ will thus be instrumental to study the 3D migratory machinery involved in developmentally and immunologically induced migration under physiological conditions or in the presence of immuno-modulating drugs.

In vertebrates, macrophages represent the earliest leukocytes in the embryo, providing the first line of defense against infection. In anamniotes, cells of the myeloid lineage originate from the rostral-most lateral plate mesoderm or rostral blood island (RBI) and early macrophages are first found in the yolk sac from where they disseminate through the embryo proper [14 15 16 17 18 19 20 21 22 23 ]. Previous studies in zebrafish characterized early macrophages observing migration on the yolk sack using Nomarski optics and epifluorescent imaging [15 , 16 , 24 ]. However, direct analysis of macrophage migration in more complex histological tissues during later stages of embryogenesis and in adult animals had not been possible so far.

Here we present a comprehensive in vivo analysis of morphological, migratory and functional characteristics of vertebrate embryonic macrophages from their first appearance to adult stages following the dynamics of an entire macrophage population using real-time in vivo confocal microscopy. We established transgenic medaka fish (Oryzias latipes) where macrophages are labeled with membrane-tethered YFP or nuclear monomeric RFP, expressed under the control of a myeloperoxidase (mpo) promoter. Early, immature, non-migratory macrophages are spherical, frequently dividing cells. In the course of cellular differentiation they flatten, display extensive filopodia formation and become highly motile, resulting in a 50-fold increase of the random walk coefficient (RWC). Interestingly, differentiated macrophages retain mitotic activity up to juvenile stages. Following a gain in motility, the macrophages actively patrol through the entire embryo. However, their migratory path is not random but follows well-defined routes and respects tissue barriers. We image macrophages rapidly responding to wounding and systemic infection. We specifically interfere with the migratory activity of macrophages by chemical inhibitors and show that Phosphoinositide-3-kinase (PI3K) signaling is required for their migration and chemotaxis in adult fish.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Generation of the transgenic medaka lines TG(FmpoP::memYFP) and TG(FmpoP::H2BmRFP)
The N-terminal sequence of Oryzias latipes mpo was blasted against the genome of Takifugu rupripes using the Ensembl Genome Browser (http://www.ensembl.org/) and 2kb upstream of the fugu mpo ORF were amplified by PCR (FmpoP). Primer sequences: Fmpop_up: 5'-CAT TCT CAT GTC ACA TTA AGT GCG ATT T-3'; Fmpop-low: 5'-ACA CGA GGA AGA CAG AGA GAA GCA TTT-3'. A 2kb FmpoP fragment was cloned to an expression vector to drive membraneYFP (Clontech) or H2BmRFP [25 ]. Two inverted I-SceI meganuclease recognition sites flanked the entire expression cassette. For transgenesis, one-cell stage embryos of the Cab inbred strain were injected [26 ]. Microinjection capillaries were backfilled with the injection solution [DNA (30ng/µl); Yamamoto buffer (0.5x); I-SceI buffer (0.5x, New England Biolabs); I-SceI meganuclease (0.35u/µl, New England Biolabs)]. DNA was prepared using a Qiagen Maxiprep kit (Qiagen, USA). Animals used in the study were kept according to national and international ethical provisions for animal husbandry as implemented at EMBL.

Cloning of partial sequences for Oryzias latipes orthologs of hematopoietic genes
Partial sequences of hematopoietic genes used for in situ hybridization were cloned by PCR upon in silico analysis of public ESTs (http://tigrblast.tigr.org/tgi/) or the publicly available preliminary medaka genome sequence (http://dolphin.lab.nig.ac.jp/medaka/index.php). Databases were searched by BLAST using zebrafish, fugu and human orthologs. PCR primer sequences for amplification of partial sequences of Ol-mpo, Ol-nephrosin, Ol-L-plastin, Ol-scl, Ol-lmo2, Ol-pu.1 and Ol-c/ebp{alpha} are available upon request.

Whole mount in situ hybridization, immunocytochemistry and histochemistry
Whole mount in situ hybridization and immunocytochemistry was performed as described [27 ]. Fast Red (Boehringer Mannheim) was used as fluorescent substrate for in situ probe detection. YFP was detected using a rabbit anti-GFP antibody (Molecular Probes, 1:500). An Alexa 488-conjugated anti-rabbit antibody used as a secondary antibody (Molecular Probes, 1:2000). Histochemistry to verify MPO activity in fluorescent cells was performed as described [28 ].

Imaging
For whole organism imaging, embryos/hatched fish were imaged in ERM with a Leica MZ16FA Stereo fluorescence microscope, using a GFP2 filter set. Pictures were obtained with a Leica DC500 digital camera using the Leica Firecam software. Imaging of ISH/ICH stained embryos was done on a Leica TCS SP confocal microscope. 2µm optical sections were recorded.

Live imaging of Hatched Fish
Imaging was performed on fish two days post hatching. Fish were anesthetized in ERM/T and imaged in MatTek glass bottom culture dishes. Prior to imaging fish were embedded, lateral side down in 1.5% low melting point agarose (dissolved ERM/T) then overlaid with ERM/T. Live imaging was performed using an inverted, Perkin Elmer Ultraspin RS spinning disk confocal microscope using a 488nm laser line with a 488 long pass filter cube to image YFP. YFP/RFP imaging was performed using 488nm and 568nm laser lines and a multi-pass filter cube. Image sequences were acquired using a 10x or 20x objective and optical sections were recorded 2.5µm apart or a 40x objective with 1µm between optical sections. Transmission images were generated by low-level illumination from above with a halogen lamp during imaging.

Wounding
For wounding experiments, a section of the tail of an anesthetized fish was cut using a razor blade (Fine Scientific Tools) prior to embedding. Care was taken during wounding not to cut any of the capillaries which vascularize the tail.

Image processing
Image sequences were processed using ImageJ (http://rsb.info.nih.gov/ij/) and a Perkin Elmer processing macro (http://www.embl.de/eamnet/html/pe_raw_macro.html).

Drug treatment
LY294002 (Calbiochem 440202) was dissolved in DMSO to give a 5mM stock. Fish were left to swim in ERM supplemented with 50µm drug for 45 min in the dark, then anesthetized and embedded in agarose supplemented with drug, and overlaid with fresh media containing tricaine and drug. IC87114 (Icos Corporation) was dissolved in DMSO to give a 10mM stock. Fish were pre-treated with 20µM drug for 30 min, embedded in IC87114 containing agarose. Media was exchanged every 15 min during the time course.

Single cell tracking
Algorithm and analysis methods
A macro for three-dimensional tracking was written in ImageJ (NIH Bethesda, MA, USA). The program consists of two main steps for the tracking. In the first step, we chose cells manually in z-projection sequence and recorded their initial ROI centering at (x1, y1). Second, z-position of the cell z1 was determined as a single z-slice with the highest average intensity for that ROI. In the third step, each cell was automatically tracked. The tracking algorithm was a combination of standard cross-correlation algorithm and center-of-mass calculation by thresholding [29 ]. For searching the position of the cell in a 3-D stack at time point t+1, neighboring voxels of the cell position at time point t was scored by cross-correlation value and the position with the highest cross-correlation was chosen. Size of the volume for this search was at the maximum 15 x 15 x 5 surrounding the position at the time point t. We used the following cross-correlation function.

Formula 1(1)
The centroid of the cell was then determined as its position. To visualize the tracks, we used two ways. First, xy-coordinates were plotted in the first image of the sequence and z-coordinates were color-coded by RGB color. Second, we plotted the three-dimensional coordinates in a surface plot in IgorPro (Wavemetrics, OR, USA) to present three-dimensional tracks.

Resulting three-dimensional coordinates were further analyzed by calculating their average instantaneous velocity and also by plotting mean-square-displacement value (MSD) vs. time-scale [30 ]. We call this plotting MSD plot. The smallest time scale was time interval of the sequence capturing which was 30 s. The smallest time scale was multiplied by integers step wise, which resulted in a series such as 60, 90, 120, and so on. The largest time scale was 1/3 of the track duration. Analysis of mobility using MSD plot is a standard method in protein mobility measurements [31 ] and cell movement analysis [7 , 32 ]. Fürth’s formula has been used to fit MSD plot of cell tracks to characterize cell movement [32 , 33 ]:

Formula 2(2)
where <d2(t)> is the mean square displacement for the time scale t, µ is the random walk coefficient, which corresponds to diffusion coefficient in case of molecules [34 ] and P is the persistence time reflecting the duration of directed movement. Note that when P is 0, Eq. (2) reduces to the MSD plot of well-known Brownian motion. When P is very small, MSD plot becomes closer to a linearly proportional straight line. Larger persistence time reflecting directed migration causes the upward curvature of MSD plot. MSD plotting in each of the x- (anterio-posterior), y-(dorso-ventral), and z-axis (tissue depth) enables determination of the biased direction of the movement. We fitted MSD plot with the Eq. (2) in IgorPro. Standard deviation of the mean of experimental results was used for calculating Chi-square values used for the optimization of curves.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
To track migrating macrophages from their first appearance to adult stages we generated transgenic (TG) medaka expressing membrane-tethered YFP (memYFP) or a nuclear H2BmRFP1 fusion gene [25 ] under control of a fugu mpo promoter fragment. In TG(FmpoP::memYFP) individual YFP positive cells can first be detected by in situ hybridization at stage 18 (~26hpf) in the rostral-most lateral mesoderm surrounding the cardiac field, in accordance with the origin of myeloid cells in the RBI [16 , 35 , 36 ]. Red fluorescent cells in TG(FmpoP::H2BmRFP) can first be detected several hours later due to the longer maturation time of mRFP compared to YFP. During the following four days of embryogenesis, labeled cells increase in number and gradually spread through the entire embryo, occasionally entering circulation (Fig. 1A 1B 1C 1D 1E ). At 5 days post fertilization memYFP+ cells accumulate in the region of the ventral venous plexus and in close proximity to the future kidney. This accumulation is due to both migration and localized cell division during the first 4 days of development (Fig. 1F and Video S1, S2) [37 ]. After young fish hatch from the protective chorion at 7dpf, fluorescent cells are spread over the entire body with higher density surrounding the digestive tract (mouth, gut) and fins (Fig. 1G 1H) . Macrophages in the skin are clearly visible using a dissecting microscope until ~30dpf when pigmentation in juvenile fish prevents visualization of deeper layers. Sectioning of adult fish (~6 months) revealed YFP labeled cells in the hematopoietic equivalent of mammalian bone marrow in fish, the marrow of the pronephros/kidney (Fig. 1I 1J) .


Figure 1
View larger version (58K):
[in this window]
[in a new window]

  		Figure 1. YFP expression in TG(FmpoP::memYFP). The Fmpo promoter drives YFP in macrophage-like cells during embryonic development. The earliest yfp transcripts are detectable by in situ hybridization at 26hpf, in macrophages around the rostral lateral plate mesoderm (A–C). Dissemination of cells continues during development (D, E). At 5dpf YFP labeled cells accumulate at the dorsal aorta (arrow) (F). After hatching, YFP positive macrophages cover the entire body (G, H). Cross-section of adult kidney stained with hematoxylin and eosin (I). Anti-GFP IHC on adjacent section reveals YFP positive cells in the kidney marrow (J). Anterior is to the left (A–H), embryo proper is outlined by dotted line (A–C), dorsal views (A–F), lateral views (G-H). Magnification: 20x (I, J). YFP+ macrophages are green. Yellow dots in D–G are autofluorescent pigment cells. Red fluorescence in G–H indicates autofluorescent remnants of digested food.

 
We verified the myeloid nature of fluorescent cells with medaka orthologs of the hematopoietic genes scl, lmo2, c/ebp{alpha}, pu.1, mpo, l-plastin and nephrosin in double ISH/IHC (Fig. 2 ). The stem cell specific genes scl and lmo2 are expressed in the lateral plate mesoderm, but are not co-expressed in memYFP+ cells (Fig. 2A 2B) . In contrast, genes expressed in hematopoietic progenitor cell populations (c/ebp{alpha}, pu.1) or more mature myeloid cells (l-plastin, mpo, and nephrosin) are co-expressed in memYFP+ cells, characterizing them as members of the myeloid lineage (Fig. 2C 2D 2E 2F 2G) [35 , 36 , 38 ]. In addition, labeled cells show strong MPO activity in enzymatic assays (Fig. 2H 2I) . MemYFP positive cells show a clear granular macrophage/monocyte-like morphology as revealed in cytospins of FACS sorted cells by May-Gruenwald-Giemsa staining (Fig. 2J) .


Figure 2
View larger version (38K):
[in this window]
[in a new window]

  		Figure 2. Identification of the myeloid origin of YFP positive cells. Fluorescent whole mount ISH/IHC analysis on embryos shortly after the onset of YFP expression (41hpf). Images show 3D projections of confocal sections. ISH for hematopoietic genes in red, anti-GFP IHC in green (A–G). Stem cell specific genes scl and lmo2 are not co-expressed in YFP positive macrophages (A, B). pu.1 (~80%; 30/37) and c/ebp-{alpha} (~70%; 26/38) are co-expressed with YFP in large fraction of cells (C, D). YFP co-expressed with myeloid leukocyte specific genes: l-plastin (~26%; 10/37), mpo (~73%; 14/19) and nephrosin (~43%; 20/46) (E–G). Histochemical staining for peroxidase activity reveals active peroxidase in YFP positive cells (H, I). May-Gruenwald-Giemsa staining of cytospun FACS sorted cells collected from hatched fish show macrophage/monocyte-like morphology (J). Magnification: x18.75 (A–G); x46 ((H, I); x50 (J). Images A–E are oriented as outlined in A, embryo in F is slightly shifted to the left. H and I show YFP and MPO staining in the tailfin. Note, only superficial YFP+ cells were enzymatically stained for MPO.

 
Taken together the cells labeled by YFP driven by the myeloperoxidase promoter in transgenic animals show all characteristics of early and late macrophages.

Real-time, in vivo confocal imaging was applied for morphological and functional analysis of memYFP labeled macrophages immediately following the onset of memYFP expression at developmental stage 20 (~31hpf) and in the tailfin of hatched fry 9-10dpf (Fig. 3 ) [39 ]. We observed two distinct types of macrophages, the former we will refer to as ‘early macrophage’ and the latter as ‘late macrophages’. Early macrophages show spherical morphology, extending thin filopodia and protruding blebs (Fig. 3A 3B , Video S3) [16 ]. They exhibit frequent mitotic activity (Fig. 3E 3F 3G 3H , Video S1). Cells first round up and cease all membrane ruffling. Cytokinesis takes place ~35 min after cessation of membrane ruffling. After cytokinesis, ruffling resumes immediately but daughter cells resume migration only after additional 20 min. Macrophages undergo considerable morphological changes during embryogenesis showing more pronounced shape changes and extensive motility at later stages (Fig. 3C 3D) . After day 3, macrophages become more flattened in morphology and display extensive filopodia and membrane protrusions. We will refer to this type of cell as ‘late macrophages’. Despite the obvious morphological changes accompanying further differentiation of the cells, we could observe occasional cell divisions occurring up to fry stages (9dpf). The process of cell division of late macrophages is similar to that of early macrophages including rounding up, ceasing motility, cell separation and remobilization within one hour. After cell division, macrophages immediately show late macrophage morphology and resume migration at similar speed to their mother cell (Video S4). Thus, some differentiated macrophages retain proliferation potential at least up to juvenile stages [21 , 23 ]. Macrophages show phagocytic activity in tissue, ventricles and blood vessels (Fig. 3I 3J , Video S5). We injected red fluorescent E. coli into ventricles of the brain at ~64hpf, into the bloodstream of hatched fry or subcutaneously into the tissue of the tailfin. Labeled macrophages engage bacteria in all situations, however, only macrophages in circulation and not macrophages residing in the tissue respond to bacteria injected into the bloodstream, while circulating macrophages occasionally leave the circulation to enter infected tissue. Video S5 shows an example of an YFP positive macrophage, already containing phagocytosed bacteria, engulfing additional bacteria in circulation as fast as 90 s. In the vasculature protein A coupled quantum dots tend to aggregate and clog capillaries at the inner margins. Interestingly we observed that this clogging could elicit a response from nearby tissue macrophages. Tissue macrophages slow down and accumulate at the outer surface of these vessels, indicating that the vessel endothelial cells trigger a macrophage response, however they never re-enter circulation at the site of clogging (data not shown).


Figure 3
View larger version (12K):
[in this window]
[in a new window]

  		Figure 3. Morphological, mitotic and phagocytic characterization of YFP positive myeloid leukocytes. In vivo confocal imaging was used for morphological and functional analysis of YFP labeled macrophages immediately following the onset of YFP expression (stage 21, 34hpf) (A, B, E–H) and in the tailfin of hatched fry 9-10dpf (C, D, I–L). The first macrophages show spherical morphology, extending thin filopodia (arrowheads) and performing distinct "blebbing" behavior (A, B, Video S3). Early macrophage-like cells (arrow) exhibit frequent mitotic activity (E-H, Video S1). Cells round up and cease any membrane ruffling (F, t=0’) until the two daughter cells separate from each other and reacquire motility (H, t=35’). Macrophages change morphology during embryogenesis showing more pronounced shape changes and high motility (C, D). We could observe occasional cell divisions occurring up to fry stages (9dpf) following the same course as at early stages including rounding up, ceasing motility and cell separation within 35 min. YFP labeled macrophages show phagocytic activity in tissue, ventricles and blood vessels (I–J). Red fluorescent E. coli were injected in to the bloodstream of hatched fry and imaged in the tail region. A YFP positive macrophage containing bacteria in the cytoplasm engulfs additional bacteria in 90 s (I–J, Video S5). Magnification: x40 (A–H); x20 (I–L).

 
To analyze the migratory behavior of individual macrophages, three-dimensional time series were recorded using a spinning-disc confocal microscope. We tracked macrophage movement using custom tracking software and determined several parameters of their motility. First, we calculated average velocity (AV), which is a conventional measure of cell migration activity. Second, we plotted their mean square displacement (MSD) as a function of different time scales. MSD is the squared value of net distance traveled by a cell during each time scale. We fitted Fürth’s formula (see Eq. 2 in Materials and Methods) and obtained random walk coefficient (RWC) and persistence time (PT). Random walk coefficient is similar to diffusion coefficient and is a measure of cell movement activity over time. While average velocity is affected largely by sampling frequency, RWC is a more reliable measure for cell migration activity. Persistence time reflects the duration of migration of a macrophage in a certain direction. When migration is random, PT becomes closer to the sampling interval. PT increases as direction of cell migration becomes more stable. (see Materials and Methods for further details) [32 , 33 ].

After the onset of YFP expression, macrophage motility is low for several hours. Only at ~39hpf, ~12 h after the first cells arise, do they start migrating faster and further away from their original location. This suggests the existence of an instructive or permissive signal within a specific developmental window [40 ]. (Fig. 4A , Table 1 2 and Video S2). Video S2 shows a 12 h time-lapse movie starting at ~33hpf. Individual macrophages gradually increase their velocity during this early 12 h period resulting in an increase of random walk coefficient of 2 µm2/min (average velocity = 1.4 µm/min) during the first 6 h to 9 µm2/min (average velocity = 2.3 µm/min) within the next 6 h (Fig. 4B 4C) . The random walk coefficient further increases to 107 µm2/min at juvenile stages (~9dpf). MSD plots show an increase in cell movement particularly distinct in the xy-plane (Fig. 4D 4E) .


Figure 4
View larger version (40K):
[in this window]
[in a new window]

  		Figure 4. Macrophage migration increases during the development of the embryo. Migrating macrophages were recorded continuously for 12h (33hpf–45hpf) and changes in their migration activity were examined. Average migration velocity during 15 min intervals shows that macrophages migrate faster in the later period (A). Time point 0 corresponds to 33hpf: 0 to 360 min corresponds to the ‘early’ phase and 360 to 720 min corresponds to the "late" phase. For every time point, 22 < n < 77. Bar = SEM. Images show three-dimensional cell migration tracks overlaid to the original video frames during the early phase (B) and the late phase (C). Z-positions (slice number) are color coded and a lookup-table is shown at the right-bottom corner of each image. White horizontal bar = 200µm. MSD plot of macrophage migration in early phase (red square), late phase (red circle) and 9dpf (blue circle) (D). Bar = SD. MSD plot in individual axes: x-axis; anterio-posterior (E), y-axis; dorso-ventral (F) and z-axis; tissue depth (G). Bar = SD. Note that the scale range for the z-axis MSD plot is much smaller than for the other two axes.

 

View this table:
[in this window]
[in a new window]

  		Table 1. Migration Parameters of Macrophages in Different Conditions: Persistence Time and Average Velocity

 

View this table:
[in this window]
[in a new window]

  		Table 2. Migration Parameters of Macrophages in Different Conditions: Random Walk Coefficient

 
During the first 6 h, persistence time is ~40 s. This means that macrophages are on average changing their direction every 40 s. Since the average of instantaneous velocity is 1.4 µm/min, cells do not move a full cell diameter (5 to 10 µm) before making a turn. These seemingly contradictory values simply indicate that cells protrude pseudopodia in different directions but these protrusions do not contribute to the effective displacement of the cell body. During the next 6 h of development, persistence time increases to ~2 min indicating that the turning frequency decreases. Average instantaneous velocity during this period is 2.3 µm/min. In this case, macrophages turn approximately every cell body length they move. The changes of cell shape, behavior and motility likely represent steps in differentiation during embryonic development.

We analyzed macrophage migration in detail, in intact tissue at juvenile stages (~9dpf). Macrophage movement was tracked by the same method described above and tracks were plotted on the first frame of the sequence (Fig. 5A ). To include the information of movement in z-axis, we color coded the z-position of macrophages. We show that macrophages migrate in random direction, exhibiting a patrolling behavior (Fig. 5A , Video S6). Surprisingly, when analyzed in the yz-plane, tracks follow a clear path (Fig. 5B , Video S7). Macrophages do not migrate through the muscle, nerve cord and cartilaginous tissue containing central tail, but patrol two planar regions: the left and right side of the fish tail. Some macrophages traverse from one side to the other through a passage close to the vertebrae (Fig. 5C 5D) . We plotted their three-dimensional MSD (Fig. 4D) and MSD in each axis (Fig. 4E 4F 4G) by fitting the MSD curve and obtained 107µm2/min as RWC and 60 s as PT (Tab. 1 and 2) . Both RWC and PT along z-axis were much lower than those along x- and y-axis (Fig. 4E 4F 4G , Tab. 1 and 2 ), reflecting limitations of macrophage movement in passing from one side to the other. Furthermore, the MSD plot for the x-axis is slightly steeper than that in the y-axis.


Figure 5
View larger version (69K):
[in this window]
[in a new window]

  		Figure 5. Macrophage migration within 9dpf fish tailfin. Macrophages migrating within 9dpf fish tailfin were recorded and their movements were tracked. Three-dimensional tracks were plotted on z-projected xy-plane of the first time point of the experiment. Movement in z-axis was coded in color (see right-bottom of the figure for the lookup-table). Bar = 100µm (A). The same tracks were plotted in 3D space, and viewed toward xy-plane (B, left panel) and yz-plane (B, right panel). Time points were color-coded, starting from dark brown and end in white (see lookup-table). Units are in µm in each axis (B). Migration track of one cell moving vertically through the entire tailfin (almost all z-planes) is highlighted (C). Note the highlighted tracks (green) in the 3D space plot. Some cells move from one side of the fin to the other following a certain path and never cross through the central region (D).

 
To characterize the response of tissue macrophages to wounding we cut a section of the tailfin of juvenile fish ~9 days post fertilization (Fig. 6A 6B 6E 6F 6G 6H , Video S8). To constrain the immunological response locally and to avoid bleeding only a small section, devoid of vascularization, was cut off. We observed a rapid response to the wound, with nearby macrophages moving toward the wound site. Over time more distant macrophages become activated, moving directly toward the wound, suggesting activation by a diffusible signal. Interestingly, macrophages are capable of further activation in response to wounding, resulting in a RWC of 188 µm2/min, approximately twofold higher than in unwounded fish (Tab. 2) . In addition to the changes in movement activity, macrophages also increase their migratory persistence during wound response. Upon challenge, persistence time of macrophages increases from 60 s to 128 s (Table 1) . In our experimental set up, the wound was always made in the right side of the image frame so that the bias in the direction of macrophage movement can be detected by comparing RWC and persistence time in individual axes. Both increases in the RWC and the persistence time upon wounding are due to an increased movement activity and a prolonged migration biased toward the wound, since only the RWC and the PT in x-axis increase whereas the values in y-axis remain unchanged. Migration activity in z-axis stays similar to that in intact tissue indicating that macrophages still respect the barriers in the center of the fin, even when responding to wounds.


Figure 6
View larger version (68K):
[in this window]
[in a new window]

  		Figure 6. Wounding activates and directs macrophage migration. Z-projections of three-dimensional sequences at different time points (A–D). Shortly after wounding macrophages are present at the wound site (bottom right corner). Bar = 100µm (A). After 60 min, many cells have reached the wound site. Three-dimensional tracks were plotted on the image frame. Movement in z-axis was color coded according to the lookup-table shown at the bottom right corner (B). We treated fish with the PI3K-{delta} inhibitor IC87114 and then wounded the tailfin. The image shows the first frame of the recording immediately after wounding (bottom right corner). Bar = 100µm (C). Cell migration tracks were overlaid to the image 60 min after wounding in (D). Note that although cells do migrate, cells barely reach the wound. MSD plots of macrophage migration in intact (blue), wounded (red), and wounded tailfin treated with PI3K-{delta} inhibitor IC87114 (green) are shown below the image frames (E–H), MSD in three-dimensional space (E), MSD in x-axis (F), y-axis (G) and in z-axis (H). Bar = SD. Results of curve fitting of these MSD plots are listed in Tables 1 and 2 .

 
Although cutting the tailfin did not damage capillaries, macrophages are occasionally seen to leave the circulation and enter the tissue to participate in the inflammatory response. This suggests that circulating macrophages can be activated by signals from cells within damaged tissue without involvement of capillary damage.

We next addressed whether the medaka macrophage migration model can be used to test the function and efficacy of drugs which block either the inflammatory response or cell migration. Current in vivo models for macrophage mobilization such as the mouse air pouch model do not allow direct observation of macrophage behavior, making it difficult to establish whether a drug inhibits chemotaxis or motility. We chose to test the role of the Phosphoinositide-3-kinase (PI3K) signaling pathway, which has been shown to be essential for cell migration in several systems (reviewed in [41 ]). The general inhibition of PI3K by treatment with the inhibitor LY294002 (50µM for 30 min) strongly inhibits macrophage migration. In treated, wounded fish, macrophages, even those close to the site of wounding, fail to respond (Video S9). In all cases cells round up and show membrane ruffling on all sides confirming a requirement of PI3K signaling for macrophage motility in vivo. At reduced doses of LY294002 (not shown) macrophage migration and chemotaxis appear normal, suggesting that the drug acts in a binary fashion, either completely inhibiting migration or having no effect. This indicates that residual PI3K activity is sufficient for in vivo migration.

To further test the role of PI3Ks in migration, we tested the drug IC87114, a specific inhibitor of the delta isoform of PI3K, which has been suggested to play a role in chemotaxis but not motility in vitro [42 , 43 ]. Treatment at 20µM IC87114 prior to wounding (Fig. 6C 6D 6E 6F 6G 6H , Video S10) leads to a greatly reduced motility, but macrophages are still capable of moving (in contrast to macrophages in LY294022 treated fish). Fitting of the MSD plot of these cells resulted in 14 µm2/min RWC and 31 s PT (Tab. 1 and 2) . Furthermore, macrophages show a reduced chemotactic response, with few macrophages reaching the wound site and many cells patrolling randomly (Fig. 6C 6D , Video S10). Those cells responding to the wound tended to be closest to the wound site. Cells reaching the wound site spend less time there than in untreated fish. Higher doses of IC87114 block all migration similar to LY294002. The effect of IC87114 is fully reversible, as cells regain normal activity after drug clearance or breakdown. These observations show that IC87114 (at high doses) interferes with general cell motility (similar to LY294022) and abolishes chemotaxis through specific inhibition of PI3K{delta} at intermediate doses.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
We have isolated a macrophage-specific promoter and used it to label macrophages in live fish to characterize the behavior of macrophages in complex tissues, during development and inflammation. We find that early macrophages initially show limited motility and develop into highly motile cells. This could occur after cells passing through a maturation-zone where the early cells mature and differentiate. Alternatively, macrophage motility in early cells could initially be suppressed or later be stimulated by a signal from the surrounding tissue (so initially favoring cell division over motility). A third possibility is that the physical characteristics of the surrounding tissue substrate changes as the embryo develops affording more efficient cell migration. We find that after cell division, macrophages resume migration at similar speeds to their mother cell, so with respect to its migratory properties, a late macrophage divides to give rise to two late macrophages, not two early, slow moving cells. This hints at a differentiation step, which is inherited by daughter cells. Interestingly we can still observe some relatively immotile early macrophages in juvenile fish (7dpf), around the developing kidney, the site of active hematopoiesis in fish. These cells likely differentiate from adult hematopoietic stem cells and represent adult progenitor cells that are responsible for long-term replenishment of the myeloid population.

A small proportion of the total population of macrophages is seen in circulation, and our observations indicate that although circulating macrophages may enter the surrounding tissue in response to infection, tissue macrophages cannot re-enter the blood vessels, although they can still receive inflammatory signals from there.

Using this system, we analyzed the behavior of macrophages in complex tissue. Despite the obvious space constraints, the cells are highly motile and capable of rapid directional migration. This migration is however not completely unconstrained. It is not clear at this point what keeps the cells constrained within the relatively flat, peripheral zones that they patrol. Macrophages may patrol along paths of least resistance. For example, adhesion between to different cell layers may be weaker than the adhesion between cells within a layer, making it easier for cells to pass between the two layers than to penetrate either layer. In our wounding model, macrophages still respect the tissue barriers even if this forces them to take a slightly longer path to a wound site. It will be interesting, in the future to test whether a wound within the center of the tissue could induce macrophages to efficiently cross these barriers and enter tissue from which they are otherwise excluded. However it is currently very difficult to generate such a wound without also damaging these barriers.

Our results indicate that despite the limitations that tightly packed tissue imposes on chemotactic signals, macrophages are capable of rapidly and precisely responding to wounds. Given sufficient time the signals that trigger this response attract macrophages over large distances (such as from the trunk into the fin; ~1mm). We find that these signals spread rapidly, eliciting a response from cells across the fin within 30-60 min. The identity of the signal or signals responsible has yet to be determined in fish. Generally, signals that lead to recruitment of leukocytes to wounds change over time with small molecules such as ATP, adenosine and bioactive lipids that leak from damaged cells representing the first attractants [44 ]. These are the likely candidates for the primary response in the described system as in our wounding assays no blood vessels have been damaged and thus release of growth factors from degranulating platelets or thrombocytes may not play a major role here. However, signals from fibroblasts and the first arriving macrophages such as TGFß or CSF1 are likely to enforce the inflammatory response involving tissue macrophages in fish similar to mammals [45 46 47 ].

Using inhibitors of PI3K we demonstrate that in vivo, PI3K is required for macrophage migration. In particular, PI3K{delta} is indispensable for chemotaxis and affects the amount of time a macrophage spends at a wound site. We have also clearly established medaka as a functional system to efficiently assay the effects of drugs on macrophage motility and chemotaxis. Using this system the effects of compounds can be specifically tested or screened for with respect to macrophage motility (change in random walk coefficient), chemotaxis (change in persistence time) or the chemotactic signal (fewer/closer cells respond).


   ACKNOWLEDGEMENTS
 
We thank E. Grzebisz and A. Nowicka for excellent fish husbandry and members of the Wittbrodt and Rørth labs for critical comments on the manuscript. We thank the advanced light microscope facility (ALMF) at EMBL for microscopy support. We thank J. H. for the kind gift of IC87114. A.C. received a fellowship from the Alexander von Humboldt Stiftung and a Marie Curie Intra-European fellowship. K.M. was supported by EAMNET (European Advanced Light Microscopy Network). This work was supported by the European Community through the NOE MAIN (targeting cell migration in chronic inflammation, P.R. and J.W.).


   FOOTNOTES
 
1 These authors contributed equally to this work. Back

Received August 8, 2006; revised September 19, 2006; accepted September 20, 2006.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Sahai, E., Marshall, C. J. (2003) Differing modes of tumour cell invasion have distinct requirements for Rho/ROCK signalling and extracellular proteolysis Nat. Cell Biol. 5,711-719[CrossRef][Medline]
  2. Condeelis, J., Segall, J. E. (2003) Intravital imaging of cell movement in tumours Nat. Rev. Cancer 3,921-930[CrossRef][Medline]
  3. Wolf, K., Mazo, I., Leung, H., Engelke, K., von Andrian, U. H., Deryugina, E. I., Strongin, A. Y., Brocker, E.-B., Friedl, P. (2003) Compensation mechanism in tumor cell migration: mesenchymal-amoeboid transition after blocking of pericellular proteolysis J. Cell Biol. 160,267-277[Abstract/Free Full Text]
  4. Van Haastert, P. J., Devreotes, P. N. (2004) Chemotaxis: signalling the way forward Nat. Rev. Mol. Cell Biol. 5,626-634[CrossRef][Medline]
  5. Borisy, G. G., Svitkina, T. M. (2000) Actin machinery: pushing the envelope Curr. Opin. Cell Biol. 12,104-112[CrossRef][Medline]
  6. Pollard, T. D., Borisy, G. G. (2003) Cellular motility driven by assembly and disassembly of actin filaments Cell 112,453-465[CrossRef][Medline]
  7. Witt, C. M., Raychaudhuri, S., Schaefer, B., Chakraborty, A. K., Robey, E. A. (2005) Directed migration of positively selected thymocytes visualized in real time PLoS Biol. 3,e160[CrossRef][Medline]
  8. Grinnell, F., Ho, C. H., Tamariz, E., Lee, D. J., Skuta, G. (2003) Dendritic fibroblasts in three-dimensional collagen matrices Mol. Biol. Cell 14,384-395[Abstract/Free Full Text]
  9. Friedl, P., Brocker, E. B. (2000) The biology of cell locomotion within three-dimensional extracellular matrix Cell. Mol. Life Sci. 57,41-64[CrossRef][Medline]
  10. Even-Ram, S., Yamada, K. M. (2005) Cell migration in 3D matrix Curr. Opin. Cell Biol. 17,524-532[CrossRef][Medline]
  11. Martins, G. G., Kolega, J. (2006) Endothelial cell protrusion and migration in three-dimensional collagen matrices Cell Motil. Cytoskeleton 63,101-115[CrossRef][Medline]
  12. Frey, M. T., Tsai, I. Y., Russell, T. P., Hanks, S. K., Wang, Y. L. (2006) Cellular Responses to Substrate Topography: Role of Myosin II and Focal Adhesion Kinase Biophys. J. 90,3774-3782
  13. Wood, W., Faria, C., Jacinto, A. (2006) Distinct mechanisms regulate hemocyte chemotaxis during development and wound healing in Drosophila melanogaster J. Cell Biol. 173,405-416[Abstract/Free Full Text]
  14. Ohinata, H., Tochinai, S., Katagiri, C. (1990) Occurrence of nonlymphoid leukocytes that are not derived from blood islands in Xenopus laevis larvae Dev. Biol. 141,123-129[CrossRef][Medline]
  15. Herbomel, P., Thisse, B., Thisse, C. (2001) Zebrafish early macrophages colonize cephalic mesenchyme and developing brain, retina, and epidermis through a M-CSF receptor-dependent invasive process Dev. Biol. 238,274-288[CrossRef][Medline]
  16. Herbomel, P., Thisse, B., Thisse, C. (1999) Ontogeny and behaviour of early macrophages in the zebrafish embryo Development 126,3735-3745[Abstract]
  17. Cuadros, M. A., Coltey, P., Carmen Nieto, M., Martin, C. (1992) Demonstration of a phagocytic cell system belonging to the hemopoietic lineage and originating from the yolk sac in the early avian embryo Development 115,157-168[Abstract]
  18. Cuadros, M. A., Martin, C., Coltey, P., Almendros, A., Navascues, J. (1993) First appearance, distribution, and origin of macrophages in the early development of the avian central nervous system J. Comp. Neurol. 330,113-129[CrossRef][Medline]
  19. Smith, S. J., Kotecha, S., Towers, N., Latinkic, B. V., Mohun, T. J. (2002) XPOX2-peroxidase expression and the XLURP-1 promoter reveal the site of embryonic myeloid cell development in Xenopus Mech. Dev. 117,173-186[CrossRef][Medline]
  20. Kurz, H., Christ, B. (1998) Embryonic CNS macrophages and microglia do not stem from circulating, but from extravascular precursors Glia 22,98-102[CrossRef][Medline]
  21. Sorokin, S. P., Hoyt, R. F., Jr, Blunt, D. G., McNelly, N. A. (1992) Macrophage development: II. Early ontogeny of macrophage populations in brain, liver, and lungs of rat embryos as revealed by a lectin marker Anat. Rec. 232,527-550[CrossRef][Medline]
  22. Takahashi, K., Naito, M., Takeya, M. (1996) Development and heterogeneity of macrophages and their related cells through their differentiation pathways Pathol. Int. 46,473-485[Medline]
  23. Takahashi, K., Yamamura, F., Naito, M. (1989) Differentiation, maturation, and proliferation of macrophages in the mouse yolk sac: a light-microscopic, enzyme-cytochemical, immunohistochemical, and ultrastructural study J. Leukoc. Biol. 45,87-96[Abstract]
  24. Redd, M. J., Kelly, G., Dunn, G., Way, M., Martin, P. (2006) Imaging macrophage chemotaxis in vivo: studies of microtubule function in zebrafish wound inflammation Cell Motil. Cytoskeleton 63,415-422[CrossRef][Medline]
  25. Megason, S. G., Fraser, S. E. (2003) Digitizing life at the level of the cell: high-performance laser-scanning microscopy and image analysis for in toto imaging of development Mech. Dev. 120,1407-1420[CrossRef][Medline]
  26. Loosli, F., Winkler, S., Wittbrodt, J. (1999) Six3 overexpression initiates the formation of ectopic retina Genes Dev. 13,649-654[Abstract/Free Full Text]
  27. Del Bene, F., Tessmar-Raible, K., Wittbrodt, J. (2004) Direct interaction of geminin and Six3 in eye development Nature 427,745-749[CrossRef][Medline]
  28. Kaplow, I. S. (1975) Letter: Substitute for benzidine in myeloperoxidase stains Am. J. Clin. Pathol. 63,451[Medline]
  29. Miura, K. (2005) Tracking movement in cell biology Adv. Biochem. Eng. Biotechnol. 95,267-295[Medline]
  30. Miura, K., Rietdorf, J. (2005) Image Quantitation and Analysis Stephens, D. eds. Cell Imaging Scion Publishing Ltd
  31. Saxton, M. J., Jacobson, K. (1997) Single-particle tracking: applications to membrane dynamics Annu. Rev. Biophys. Biomol. Struct. 26,373-399[CrossRef][Medline]
  32. Dunn, G. A., Brown, A. F. (1987) A unified approach to analysing cell motility J. Cell Sci. Suppl. 8,81-102[Medline]
  33. Bergman, A. J., Zygourakis, K. (1999) Migration of lymphocytes on fibronectin-coated surfaces: temporal evolution of migratory parameters Biomaterials 20,2235-2244[CrossRef][Medline]
  34. Berg, H. C. (1993) Random walks in biology Princeton University Press Cichester.
  35. Lieschke, G. J., Oates, A. C., Crowhurst, M. O., Ward, A. C., Layton, J. E. (2001) Morphologic and functional characterization of granulocytes and macrophages in embryonic and adult zebrafish Blood 98,3087-3096[Abstract/Free Full Text]
  36. Bennett, C. M., Kanki, J. P., Rhodes, J., Liu, T. X., Paw, B. H., Kieran, M. W., Langenau, D. M., Delahaye-Brown, A., Zon, L. I., Fleming, M. D., et al (2001) Myelopoiesis in the zebrafish, Danio rerio Blood 98,643-651[Abstract/Free Full Text]
  37. Rombout, J. H., Huttenhuis, H. B., Picchietti, S., Scapigliati, G. (2005) Phylogeny and ontogeny of fish leucocytes Fish Shellfish Immunol. 19,441-455[CrossRef][Medline]
  38. Crowhurst, M. O., Layton, J. E., Lieschke, G. J. (2002) Developmental biology of zebrafish myeloid cells Int. J. Dev. Biol. 46,483-492[Medline]
  39. Iwamatsu, T. (1994) Stages of normal development in the medaka Oryzias latipes Zoolog. Sci. 11,825-839
  40. Blaser, H., Eisenbeiss, S., Neumann, M., Reichman-Fried, M., Thisse, B., Thisse, C., Raz, E. (2005) Transition from non-motile behaviour to directed migration during early PGC development in zebrafish J. Cell Sci. 118,4027-4038[Abstract/Free Full Text]
  41. Procko, E., McColl, S. R. (2005) Leukocytes on the move with phosphoinositide 3-kinase and its downstream effectors Bioessays 27,153-163[CrossRef][Medline]
  42. Sadhu, C., Dick, K., Tino, W. T., Staunton, D. E. (2003) Selective role of PI3K delta in neutrophil inflammatory responses Biochem. Biophys. Res. Commun. 308,764-769[CrossRef][Medline]
  43. Sadhu, C., Masinovsky, B., Dick, K., Sowell, C. G., Staunton, D. E. (2003) Essential role of phosphoinositide 3-kinase delta in neutrophil directional movement J. Immunol. 170,2647-2654[Abstract/Free Full Text]
  44. Martin, P., Leibovich, S. J. (2005) Inflammatory cells during wound repair: the good, the bad and the ugly Trends Cell Biol. 15,599-607[CrossRef][Medline]
  45. Ridley, A. J. (2001) Rho GTPases and cell migration J. Cell Sci. 114,2713-2722[Abstract/Free Full Text]
  46. Frank, S., Madlener, M., Werner, S. (1996) Transforming growth factors beta1, beta2, and beta3 and their receptors are differentially regulated during normal and impaired wound healing J. Biol. Chem. 271,10188-10193[Abstract/Free Full Text]
  47. Jagadeeswaran, P., Sheehan, J. P., Craig, F. E., Troyer, D. (1999) Identification and characterization of zebrafish thrombocytes Br. J. Haematol. 107,731-738[CrossRef][Medline]



This article has been cited by other articles:


Home page
 J. Immunol.Home page
T. Wang, P. C. Hanington, M. Belosevic, and C. J. Secombes
Two Macrophage Colony-Stimulating Factor Genes Exist in Fish That Differ in Gene Organization and Are Differentially Expressed
J. Immunol., September 1, 2008; 181(5): 3310 - 3322.
[Abstract] [Full Text] [PDF]

Home page
 J. Leukoc. Biol.Home page
S. B. Brown, C. S. Tucker, C. Ford, Y. Lee, D. R. Dunbar, and J. J. Mullins
Class III antiarrhythmic methanesulfonanilides inhibit leukocyte recruitment in zebrafish
J. Leukoc. Biol., July 1, 2007; 82(1): 79 - 84.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Videos
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Grabher, C.
Right arrow Articles by Wittbrodt, J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Grabher, C.
Right arrow Articles by Wittbrodt, J.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS