Observation and characterisation of macrophages in zebrafish liver

Micron 132 (2020) 102851

Contents lists available at ScienceDirect

Micron journal homepage: www.elsevier.com/locate/micron

Observation and characterisation of macrophages in zebrafish liver a,

b

a

b

T

a,c,d

Delfine Cheng *, Marco Morsch , Gerald J. Shami , Roger S. Chung , Filip Braet School of Medical Sciences (Discipline of Anatomy and Histology) — The Bosch Institute, The University of Sydney, NSW 2006, Australia Faculty of Medicine and Health Sciences, Macquarie University, Sydney, NSW 2109, Australia Australian Centre for Microscopy & Microanalysis, The University of Sydney, NSW 2006, Australia d Charles Perkins Centre (Cellular Imaging Facility), The University of Sydney, NSW 2006, Australia a

b c

A R T I C LE I N FO

A B S T R A C T

Keywords: Correlative microscopy Hepatic immune system Hepatic sinusoidal cells Liver Macrophages Multi-modal imaging Ultrastructure

Kupffer cells are liver-resident macrophages that play an important role in mediating immune-related functions in mammals and humans. They are well-known for their capacity to phagocytose large amounts of waste complexes, cell debris, microbial particles and even malignant cells. Location, appearance and functional aspects are important features used to identify these characteristic cells of the liver sinusoid. To-date, there is limited information on the occurrence of macrophages in zebrafish liver. Therefore, we aimed to characterise the ultrastructural and functional aspects of liver-associated macrophages in the zebrafish model by taking advantage of the latest advances in zebrafish genetics and multimodal correlative imaging. Herein, we report on the occurrence of macrophages within the zebrafish liver exhibiting conventional ultrastructural features (e.g. presence of pseudopodia, extensive lysosomal apparatus, a phagolysosome and making up ∼3% of the liver volume). Intriguingly, these cells were not located within the sinusoidal vascular bed of hepatic tissue but instead resided between hepatocytes and lacked phagocytic function. While our results demonstrated the presence and structural similarities with liver macrophages from other experimental models, their functional characteristics were distinctly different from Kupffer cells that have been described in rodents and humans. These findings illustrate that the innate immune system of the zebrafish liver has some distinctly different characteristics compared to other animal experimental models. This conclusion underpins our call for future studies in order to have a better understanding of the physiological role of macrophages residing between the parenchymal cells of the zebrafish liver.

1. Introduction The presence of liver macrophages within mammalian species, including human and animal models, has been well documented over the years (Dixon et al., 2013; Jones and Summerfield, 1988). The initial discovery by von Kupffer in 1876 concerned phagocytic “stellate cells” in the liver (von Kupffer, 1876). Later named after its discoverer and currently known as Kupffer cells, these cells are defined as liver-resident macrophages (Alterman, 1977). In rodents and humans, liver -resident macrophages play a central role in liver homeostasis (Naito et al., 2004) by removing foreign particulates that enter the liver blood circulation (Wisse, 1974). By doing so, they tightly control the balance between liver health and disease (Bilzer et al., 2006; Brunt et al., 2014). Under physiological conditions, they are confined to the liver vasculature, anchored to the sinusoidal endothelium by worm-like invaginations of the plasma membrane, termed vermiform processes

(Wisse, 1977), in a peculiar close apposition. Their position is strategic as they act as cellular guardians and upon stimulation, may become motile and actively migrate to their target and/or injury site (Frevert et al., 2006; P. J. MacPhee et al., 1992). The cells are known to secrete a cocktail of cytokines upon activation that result in the recruitment of immune cells such as natural killer cells, neutrophils—including blood platelets—and as such contribute to the local healing process (Wake et al., 1989). Kupffer cells play an important role in the activation and proliferation in alcoholic liver disease (Zhang et al., 2016), liver transplantation (Fahrner et al., 2016) and in scavenging large macromolecular waste complexes—including ageing cells—from the blood (Wardle, 1987; Wisse et al., 1996). Additionally, their role in removing cancerous cells and infectious organisms (e.g. viruses, bacteria and parasites) has been studied in much detail (Anthony et al., 2012; Boltjes et al., 2014; Seki et al., 2011; Tsutsui and Nishiguchi, 2014). Liver-resident macrophages represent the largest population of

⁎

Corresponding author at: The Victor Chang Cardiac Research Institute, 405 Liverpool St, Darlinghurst, 2010 NSW, Australia. E-mail addresses: delfi[email protected] (D. Cheng), [email protected] (M. Morsch), [email protected] (G.J. Shami), [email protected] (R.S. Chung), fi[email protected] (F. Braet). https://doi.org/10.1016/j.micron.2020.102851 Received 31 October 2019; Received in revised form 12 February 2020; Accepted 12 February 2020 Available online 13 February 2020 0968-4328/ © 2020 Elsevier Ltd. All rights reserved.

Micron 132 (2020) 102851

D. Cheng, et al.

experiments were conducted under Macquarie University Animal Ethics and Biosafety approvals (2012/050 and 2015/033; 5201401007). The fish were maintained at 28 °C in a 13 h light and 11 h dark cycle and embryos were collected by natural spawning and raised at 28.5 °C in E3 solution according to standard protocols (Westerfield, 2000). Zebrafish larvae of 12 days post-fertilisation (dpf) (n = 6), juveniles of 1-month post-fertilisation (mpf) (n = 2) and adults of 3 mpf (n = 2) were used in this study. This sample provides a representation of 3 developmental stages. The fish were used following an overnight starvation in order to reduce the amount of food in the gut, previously shown to be autofluorescent (Hedrera et al., 2013).

tissue-resident macrophages in the body (Sasse et al., 1992) and belong to the family of liver sinusoidal cells that make up 35 % of the total liver cell number and approximately 17 % of the total liver volume, depending on the species (Blouin, 1977; Weibel et al., 1969). The ultrastructural characteristics of liver-resident macrophages include an extensive lysosomal apparatus with numerous membrane-bound vesicles that make up the majority of their cytoplasmic content. Their overall cell shape is dynamic and resembles their restlessness in a continuing search to engulf waste complexes (Wake et al., 1989). Their uptake capacity is reported to be endless and the cells are able to continuously renew themselves, including the recruitment and differentiation into a new population of liver-resident phagocytes (Naito et al., 2004). Kupffer cells can also be identified by their characteristic functional properties—i.e. phagocytosis and, as a result, production of peroxidase enzyme— (Bouwens, 1988; Warnock et al., 1987; Wisse, 1974). In fact, nanoparticles such as latex beads or carbon particles are well-established markers for the functional identification of mammalian Kupffer cells in vitro and in vivo. Within minutes following injection into the circulation, beads and carbon particles (ranging from 8 to 20 μm) can be found aggregating within rat Kupffer cells in vitro and in vivo experimental conditions (Akilbekova et al., 2015; Bautista et al., 1992; Kamimoto et al., 2005). Transgenic animal models expressing fluorescent proteins under specific promoters have been extensively used to study organ development, cellular interactions and drug responses in different fields of biomedical sciences (Wei, 1997). With the ease of genetic manipulations in the zebrafish came a library of transgenic zebrafish, available publicly through the Zebrafish International Resource Center (University of Oregon), and amongst them is the macrophage-specific zebrafish line mpeg1, which possess red fluorescent macrophages (Ellett et al., 2011). This transgenic fish has been utilised to study the dynamics of innate immune system in vertebrates as well as its interactions with a number of pathogens, and particularly human pathogens (Kanther and Rawls, 2010; Svahn et al., 2018). In previous studies, we demonstrated that the zebrafish liver displays similar functions and microanatomy to its human and rodent counterparts (Cheng et al., 2016), including the ability to filter, uptake and process large protein complexes within the hepatic sinusoids (Cheng et al., 2019, 2017). However, there is limiting literature available on the existence and function of the macrophages residing within the zebrafish liver. In this work, we aimed to characterise the ultrastructure, occurrence and functional aspects of liver-associated macrophages present in the mpeg1 transgenic zebrafish. Structural and functional similarities were observed and compared to Kupffer cells in the human and rodent livers (for reviews, see Jones and Summerfield, 1988; Wisse et al., 1996). We applied correlative light and electron microscopy (CLEM), a concept effective in cell image analysis by localising, identifying and studying the functional and structural properties of the cells' of interest across different imaging platforms (Kobayashi et al., 2012; Müller-Reichert and Verkade, 2012). Noteworthy, previous correlated fluorescence-, and scanning electron microscopy (SEM) studies in zebrafish have already demonstrated the advantageous nature of this approach in disclosing cellular transport and distribution of albumin in the zebrafish liver (Cheng et al., 2019) including the localisation of cellular targets within large tissue volumes when combined with serial sectioning techniques (Wacker et al., 2016).

2.2. Live confocal imaging We followed the zebrafish agarose embedding and injection methods previously described (Cheng et al., 2017; Morsch et al., 2017). Once the fish were stabilised in a glass-bottomed Petri dish, the liver area was imaged using an upright confocal microscope (SP5, Leica Microsystems). 2.3. Intrahepatic latex beads injection for labelling phagocytic activity With the fish stabilised in agarose in a glass-bottomed Petri dish, a benchtop micromanipulator was used to deliver 2.8–5.2 nl depending on the size of the fish of a mixture of 1000-, 100- and 55 nm sized latex beads 1:1:1 ratio, 28 % solid concentration, Sulphate-modified polystyrene, Cat numbers L1528, L1148 and LB11, Sigma Aldrich, Australia) directly into the liver. Refer to the Discussion part for more details on injection volumes. For the juveniles and adult fish, care must be taken during the needle insertion not to cause substantial damage to the liver as the fish skin is thicker and needle insertion can be more resistant. For the adult fish, the skin opacity prevented direct visualisation of the liver under the dissecting microscope, but the location of the liver can be easily guessed. Following injection, half of the fish were processed for correlative fluorescence and electron microscopy studies while the other half were processed for ultrastructural studies. 2.4. Sample preparation for correlative fluorescence and electron microscopy studies Following injection, the bead mixture was left circulating for 45 min. before the zebrafish were fixed in a solution of 4% paraformaldehyde and 0.5 % glutaraldehyde in buffer solution (0.1 M sodium cacodylate buffer supplemented with 4% sucrose and 0.15 mM CaCl2, pH 7.3). To ensure adequate fixation penetration and optimal sample preservation, the tail was cut off for the juveniles (1 mpf) and the liver was dissected out about 1 h after initial fixation for the adults (3 mpf). All the samples were then further fixed overnight at 4 °C and subsequently processed for electron microscopy using a protocol, which has been shown to provide optimum combination of in situ fluorescence retention and ultrastructural preservation in the zebrafish (Cheng et al., 2017), particularly the liver. The samples were gradually dehydrated to 50 % ethanol, en bloc stained with uranyl acetate solution (saturated with 50 % ethanol) overnight, further dehydrated to 80 % ethanol and finally embedded in LR White resin. Following polymerisation, the resin blocks were sectioned using an ultramicrotome (Ultracut 7, Leica Microsystems, Switzerland). For the larvae fish, consecutive cross-sections were generated throughout the entire liver by means of serial sectioning (Blumer et al., 2002; Micheva and Smith, 2007). Ribbons of 12 sections of 500 nm thick, followed by one section of 90 nm thick, were collected on squared coverslips. Coverslips were previously coated with poly-Llysine and glow discharged just preceding collection to improve section adherence and reduce sections wrinkles (Cheng et al., 2017). Routinely, one 500 nm-thick section was collected onto a glass slide and either stained with 0.5 % toluidine blue to assess the general histology, preservation quality and orientation of the liver with light microscopy or

2. Materials and methods 2.1. Animal model We utilised a previously characterised transgenic zebrafish line, Tg (mpeg1:GAL4,UAS:mCherry) (gl22Tg), here referred to as mpeg1, which expresses a red fluorescence protein reporter (mCherry) under a macrophages specific promoter. The zebrafish were housed in a purposebuilt zebrafish facility at Macquarie University (Australia) and 2

Micron 132 (2020) 102851

D. Cheng, et al.

uranyl acetate and Reynold’s lead citrate for 10 min each. In between each staining step, distilled water was used to carefully rinse the coverslips to remove excess stain. Once air-dried, the coverslips were coated with a thin layer of evaporated carbon (approx. 15 nm), mounted onto aluminium stubs with carbon tape and grounded with silver paste. The imaging was performed using back-scattered detection SEM (Sigma HD VP, Carl ZEISS, Australia). The 90 nm sections previously imaged under fluorescence microscopy were automatically relocated in the SEM, also calibrated with the ZEN 2 software, and imaged in backscattered electron detector mode at an accelerating voltage of 15 kV and working distance of 6 mm. The SEM image contrast was inverted to ease image interpretation. For correlative analysis, the fluorescent and SEM images were overlayed at matching pixel resolution (Cheng et al., 2017) and the correlative image data presented in the figures were adjusted for brightness and contrast to ease visualisation (Sedgewick, 2008).

stained with 3,3ʼ-Diaminobenzidine Peroxidase Substrate (DAB) to label peroxidase activity (see section below on Diaminobenzidine staining for labelling macrophages peroxidase activity). Finally, the ribbons of sections on the coverslips were imaged using fluorescence microscopy followed by scanning electron microscopy. 2.5. Sample preparation for ultrastructural studies by electron microscopy Following injection, one fish from each developmental stage (larvae, juvenile and adult) was fixed in a solution of 4% paraformaldehyde and 2.5 % glutaraldehyde in buffer solution (0.1 M sodium cacodylate buffer supplemented with 4% sucrose and 0.15 mM CaCl2, pH 7.3). To ensure adequate fixation penetration and optimal sample preservation the tail was cut off for the juveniles (1 mpf), and the liver was dissected out about 1 h after initial fixation for the adults (3 mpf). All samples were then further fixed overnight at 4 °C and subsequently processed for electron microscopy using a protocol previously described (Cheng et al., 2016). Briefly, the protocol involves post-fixation with 1% osmium tetroxide + 1.5 % potassium ferrocyanide for 2 h, followed by incubation in 1% thiocarbohydrazide for 20 min, 2% OsO4 for 20 min, 1% uranyl acetate over night at 4 °C and Walton’s lead aspartate for 30 min at 60 °C. the samples were finally dehydrated in a series of ethanol, gradually infiltrated in EPON (hard grade) over 5 days and polymerised in pure resin over 2 days. Following polymerisation, 70 nm sections from the liver were generated using an ultramicrotome (Ultracut 7, Leica Microsystems, Switzerland), collected on 200 mesh copper grids and post-stained with 2% uranyl acetate and Reynold’s lead citrate for 10 min each. Finally, the samples were imaged under a transmission electron microscope (TEM) (JEM1400, JEOL, Japan), operating at 120 kV.

2.8. Diaminobenzidine staining for labelling macrophages peroxidase activity To determine the presence of peroxidase activity in the macrophages, 500 nm sections across the liver were generated using an ultramicrotome (Ultracut7, Leica Microsystems, Switzerland). The sections were collected on glass slide previously coated with poly-L-lysine and glow-discharged. The sections were first imaged under fluorescent microscope to determine the location of the mpeg1-labelled cells within the liver and to record their position within the section. The sections were then incubated for 1 h with 0.1 % of DAB (SIGMAFAST™ DAB with Metal Enhancer, Lot # SLBF4765 V) then further enhanced with 1% osmium tetroxide for 10 min. Sections were washed thoroughly with phosphate buffer in between staining steps. The same areas previously imaged under fluorescence microscopy were retrieved and imaged under brightfield light microscopy to locate DAB stained structures (darker in appearance). Finally, images from both imaging platforms were overlayed by matching pixel, using Adobe Photoshop CS6.

2.6. Fluorescence microscopy The coverslip containing the serial sections was loaded onto a ZEISS sample relocation holder (Zeiss Life Science Cover Glass 22 × 22 holder) and calibrated with the ZEN 2 software (Blue edition, the Shuttle & Find plugin) to allow for subsequent automated relocation of the section. For quantitative analysis and to detect the overall presence and distribution of red fluorescent cells, one 500 nm-thick section per ribbon (equivalent to approximately every 6 μm) was observed under fluorescence microscopy (40x/0.95 objective, Axio Observer Z1). Anatomical compartments and organs were easily discernible due the slight background fluorescence originating from the sample itself (Cheng et al., 2017). The red fluorescence originating from the fluorescent cells were easily distinguishable from the background fluorescence. Red fluorescent cells with a visibly rounded cytoplasm – containing a nucleus - were manually counted throughout the fish liver (n = 3) to ensure the same cell would not be counted twice. The ImageJ plugin 3D Object counter was used to analyse the volumes and surface areas (Bolte and Cordelières, 2006) of macrophages within and outside the liver. The total liver volumes were also measured using ImageJ: the liver on each section was manually traced and the surface area measured. The total volume corresponds to the total surface area times the section thickness (500 nm) times the image interval (6 μm). For ultrastructural characterisation of the red fluorescent cells by correlative microscopy, every 90 nm section containing a red fluorescent cell within the liver was also imaged using fluorescence microscopy and the red fluorescent cells coordinates were recorded accordingly. Manual recording of sections coordinates was also possible by noting the ribbon number, if the Shuttle & Find plugin was not available, as the 90 nm section was always the last section of each ribbons.

3. Results 3.1. Fluorescence microscopy observations of liver macrophages Live confocal imaging of mpeg1 zebrafish at 12 dpf revealed a large population of red fluorescent cells present in the fish, including in the liver (Fig. 1. In general, they all presented a rounded nuclear area measuring approx. 10 μm surrounded with multiple long cytoplasmic extensions of up to 25 μm in length. Following morphometric analysis on these cells, we determined no significant differences in size and shape between the macrophages present in the liver and those from the surrounding. In fact, the average volume of liver-associated macrophages was 2294 ± 875 μm3 with a surface area averaging 2677 ± 820 μm2 n = 29 macrophages from 4 different fish. On the other hand, the average volume of surrounding macrophages was 3862 ± 1876 μm3 with a surface area averaging 4028 ± 1834 μm2 (n = 39 macrophages from 4 different fish). The sample preparation protocol applied herein was successful in retaining the in situ red fluorescence from the mpeg1 cells for on-section SEM imaging. The red fluorescence was easily distinguishable from the background fluorescence (likely originating from the aldehyde present in the initial fixative or due to the composition of the tissue itself), even on sections as thin as 90 nm (Fig. 2). From the 500 nm thick sections, we counted an average of 31 macrophages in a total liver volume of 1,249,233 μm3 (equivalent to 1 macrophage in every 41,088 μm3 of liver tissue or 3% of the liver volume) (Supplementary information Table S1). Based on the wide-field microscopy observations, the cells were generally sparsely dispersed, and we did not observe any preferential distribution within the organ.

2.7. Scanning electron microscopy To improve conductivity and contrast for SEM imaging, the tissue sections mounted on coverslips were post-stained with 2% aqueous 3

Micron 132 (2020) 102851

D. Cheng, et al.

Fig. 1. Overview of macrophages within the zebrafish larvae at 12 dpf. (A) Images show a fish lying sideways with the head on the left and the tail on the right. Brightfield overview image of an mpeg1 zebrafish shows the orientation of the fish. The blue dotted line represents the approximate limits of the liver. (B) Confocal microscopy maximum projection image overlayed with brightfield image of a 12 dpf mpeg1 zebrafish larvae, around the liver. The mpeg1 is a transgenic line of zebrafish which has red fluorescence specifically targeted to macrophages. Macrophages often appear to adopt a stellate shape (C) with multiple processes called pseudopodia. The total thickness of the confocal dataset was 83 μm. Scale bars = 40 μm (B) and 10 μm (C). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article).

macrophages. Compared to hepatocytes, the electron density of their cytoplasm was lower, the lysosomes were numerous and of various sizes, and the pseudopodia were rich in cytoskeleton matrix. We have not observed structural differences between macrophages from a larva compared to a juvenile or adult fish. The macrophages location (e.g. interhepatic, perisinusoidal and periductal) was also consistent throughout all 3 development stages.

3.2. Ultrastructural characterisation of liver macrophages To determine the ultrastructural characteristics and sub-organ localisation of macrophages in the zebrafish liver, we located the mpeg1 cells by fluorescence imaging then relocated them in back-scattered SEM using the ZEISS Shuttle & Find automated system. We revealed that the red fluorescent mpeg1-labelled cells were located in between hepatocytes and at the periphery of sinusoids and bile ducts (Fig. 2 and 3). Higher magnification SEM images showed that these cells possessed an irregular shape, a large nucleus, numerous small dense cytoplasmic vesicles and a large perinuclear dense body. The irregular shape of the cells was particularly pronounced at the proximity of the sinusoids where long cytoplasmic extensions prominently project towards the sinusoid, sometimes occupying the space of Disse and nearly touching the endothelium (Fig. 2). We did not encounter mpeg1-labelled cells located within the liver sinusoids neither circulating within the blood vessels nor anchored to the endothelium. Since ultrastructural preservation was not optimal using this protocol (but rather aimed at maintaining fluorescence), we further investigated the ultrastructure of liver macrophages with TEM using samples which were processed using an optimised protocol for ultrastructure preservation. The ultrastructural information we obtained from our correlative fluorescence and electron microscopy study, along with previous work (Herbomel et al., 2001; Keightley et al., 2014; Lieschke et al., 2001), provided us with key ultrastructural features on liver macrophages (e.g. size, presence of pseudopodia, numerous lysosomes). We then used these keys features as a basis to identify macrophages in the zebrafish liver at highresolution, without fluorescent markers (Fig. 4). Large cells in between hepatocytes and at proximity of sinusoids and bile ducts which possessed an irregular shape, pseudopodia and contained numerous lysosomes and a large perinuclear phagolysosome were identified as

3.3. Characterisation of liver macrophages phagocytic function To evaluate the ability of liver macrophages to phagocytose foreign material, we injected a mixture of 3 sizes of latex beads directly into the hepatic system and monitored the possible uptake, concentration and behaviour in 7 fish (3 larvae, 2 juveniles and 2 adults). For the larvae, live confocal imaging alone could have been used to determine the latex beads uptake as they were visible under a confocal microscope (Fig. 5). In fact, in the 3 experiments using larvae, live confocal microscopy showed beads present in the circulation and sometimes accumulating in the vicinity of the liver, but we could not conclusively determine whether the beads were within the liver, in the surrounding cavity or circulating, as liver boundaries were difficult to determine by eye (Fig. 5). For the juveniles and the adults, the fish were placed directly into fixative 45 min post-injection as confocal microscopy was not a viable option due to the thicker and pigmented skin of those fish. We therefore fixed these fish and processed them using a protocol which preserves the fluorescence for subsequent on-resin imaging. Light microscopy observation of toluidine blue stained sections revealed good ultrastructural preservation of the liver organ (Fig. 6). Inherent to the injection method, a site of injury is present within the liver of variable sizes within which signs of autolysis and beads can be observed. A healthy zone was drawn in the liver which was 3 cells away from the injury site: in that zone, cells did not exhibit signs of damage 4

Micron 132 (2020) 102851

D. Cheng, et al.

Fig. 2. CLEM visualisation of a mpeg1 cell (red) within the liver of a zebrafish larvae (12 dpf). Insert (top left): Wide-field fluorescent micrograph (mosaic) of a 90 nm resin embedded mpeg1 zebrafish larvae of 12 dpf, fixed and processed using a protocol whereby the in situ fluorescence was retained. The cross-section was generated across the liver organ whereby the oesophagus (Oes) and the liver (Liv) can be easily identified. Red fluorescent cells are sparsely present but are clearly distinguishable from the background fluorescence. The boxed area encloses a red fluorescent cell which was relocated in SEM. Large image: Overlayed wide-field fluorescence and back-scattered SEM images from the white boxed in the insert. The SEM image depicts cellular and intracellular details at high-resolution within the white boxed area. The overlayed image allows to confidently locate the cell which exhibits red fluorescence between hepatocytes (Hep) and next to the sinusoid (Sin). The mpeg1 cell presents characteristic features of macrophages such as a large nucleus (N), numerous intracellular vesicles of various sizes (conceivably endosomes, lysosomes; indicated by the arrows), a perinuclear phagolysosome (P) and an irregularly-shaped elongated cell membrane (arrowheads). The pseudopodia of the macrophage are directed towards the sinusoid and are in close proximity to the sinusoidal endothelium (blue highlight). These extensions are thin, branched and occupy the space of Disse, a characteristic of macrophages that has been described in other teleost fish (Rocha et al., 1994). Scale bar = 2 μm. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article).

observed a nucleus, multiple vesicles and a cytoskeleton matrix present within a cytoplasmic matter enclosed by very elongated plasma membrane (Fig. 8). These are suggested to be cytoplasmic extensions of macrophages residing in the kidney and the ventral venux plexus, both being known sites of haematopoiesis (Crowhurst et al., 2004). Overall, while our injection method into the zebrafish liver lead to successful dispersal into the circulation and uptake by a population of cells, our observations could not conclude the active ingestion of these beads by liver macrophages.

or autolysis and was used for our analysis. In the undamaged area, toluidine blue stained sections did not reveal cells or subcellular compartments within the liver that incorporated latex beads. Instead, we observed the larger 1000 nm and the mid-size 100 nm latex beads present within the vessels around the kidney Fig. 6A and B) and the hepatic sinusoids (Fig. 6C and D). The smallest ones of 55 nm are below the resolving power of light microscopy. Imaging the 90-nm sections under fluorescence microscopy and back-scattered SEM within the liver from fish at any of the 3 developmental stages (locating the 55 nm beads) did not reveal any areas within the liver that incorporated the latex beads (even the 55 nm size beads). Although it was not possible to screen for the beads within other organs from the juvenile and adult fish (as they were dissected out), we screened the surrounding organs in the larvae, which were kept as whole during the processing. In these juvenile fish, we located some latex beads within the circulation but more interestingly, we observed aggregations of beads in the vicinity of both the kidney and the basal viscera vein (Fig. 7). In both areas, the beads were present within a compartment of elongated shape. By imaging the same area on subsequent sections, we eventually revealed that compartment to be the long cytoplasmic extension of cells: on subsequent sections, we

3.4. Diaminobenzidine staining for labelling macrophages peroxidase activity To determine the presence and localisation of the characteristic peroxidase enzyme within the liver macrophages, we stained zebrafish liver sections with DAB. Using brightfield light microscopy, we located the presence of DAB-positive cells sparsely distributed within the liver (dark brown appearance). Fluorescence microscopy image before DAB staining overlayed with brightfield image of the same section after DAB staining revealed that the DAB-positive cells did not correspond to the mpeg1 cells and reciprocally, mpeg1-labelled cells did not stain with 5

Micron 132 (2020) 102851

D. Cheng, et al.

Fig. 3. Examples of the various shapes of macrophages in the zebrafish liver sections, as observed by CLEM. Overlays of fluorescence and back-scattered SEM images of mpeg1 cells in the zebrafish liver highlight the intrahepatic localisation instead of within the sinusoids (Sin), thin branch-like structure of the pseudopodia and the inclusion of numerous small electron dense cytoplasmic vesicles, interpreted as lysosomes (arrows) as seen in (A). Hep: hepatocyte; Sin: sinusoid. Scale bars = 2 μm.

dense bodies of various sizes (lysosomes), and a large perinuclear dense vesicle, interpreted as a phagolysosome (Douglas and Tuluc, 2010; Hampton et al., 1987; Rocha et al., 1994). When present in fish, macrophages were often reported to be difficult to find due to their low occurrence (Hampton et al., 1987; Sakano and Fujita, 1982). In the brown bullhead fish for example, another teleost fish, resident sinusoidal macrophages have been described based on ultrastructural and functional studies using latex beads injection and peroxidase staining (Hampton et al., 1987). In other teleost fish livers, such as catfish (Ferri and Sesso, 1981), rainbow trout (Anderson and Mitchum, 1974) and brown trout (Rocha et al., 1997), liver macrophages have been described only based on morphological similarities. Although macrophages have also been previously reported in the zebrafish liver, it was mainly using histological staining and light microscopy techniques (Schwendinger-Schreck, 2013), which lack a degree of accuracy and resolution. Specific molecular markers encoding macrophage-restricted cell surface molecules and enzymes have been identified in the zebrafish and their mammalian counterparts, including l-plastin, draculin, fms and lysozyme (Bennett et al., 2001; Herbomel et al., 1999; Kalev-Zylinska et al., 2002; Parichy et al., 2000). Here, we took particularly advantage of the mpeg1 transgenic zebrafish line, which has a red fluorescent protein (mCherry) expressed under a macrophage specific promoter (Ellett et al., 2011). This transgenic line has already proven to be very useful in the study of various macrophage-related studies such as innate immune system development, angiogenesis, and immune system responses: particularly for its specificity and segregated labelling from neutrophils (Gerri et al., 2017; Morsch et al., 2015). Fluorescent

DAB (Fig. 9). This further supports that macrophages present in the zebrafish liver at the larvae stage did not display their typical phagocytic ability as they did not exhibit peroxidase enzyme activity. 4. Discussion In this study, we used combinatorial labelling techniques with correlative microscopy (Jahn et al., 2012; Su et al., 2010) to determine the presence, localisation and cellular ultrastructure of macrophages in the zebrafish liver (Fig. 2). The protocol successfully preserved endogenous fluorescence from a mpeg1 zebrafish line, and allowed accurate quantification and ultrastructural characterisation of the macrophages in the liver. We determined the macrophages present in the zebrafish liver to occur 5 times less (3% of total liver volume) compared to those found in rodents (15 % of total liver volume) (Blouin, 1977; Weibel et al., 1969) while their ultrastructural characteristics are highly similar. The macrophages identified in our transgenic line however lack a characteristic functional aspect as found in rodent macrophages, namely phagocytosis which could impair their use in hepatic immunerelated diseases. Zebrafish macrophages originate from the anterior lateral plate mesoderm at around 12 h post-fertilisation (Onnebo et al., 2004). In the adult zebrafish, resident macrophages are observed in the kidney (site of haematopoiesis from 4 dpf) and the spleen while other macrophages have been reported to circulate in the blood, the brain, the spinal cord and the heart, among other organs (Menke et al., 2011). Liver macrophages in fish and rodents are described to be of a large irregularly shape with a well-developed Golgi apparatus, multiple cytoplasmic 6

Micron 132 (2020) 102851

D. Cheng, et al.

Fig. 4. Overview of zebrafish liver macrophages in larvae (A), juvenile (B-C) and adult (D) stages observed under TEM. The large electron-dense phagolysosomes (P) and numerous lysosomes (Ly) are characteristic features of macrophages making them easily identifiable. These cells cytoplasm are less electron-dense compared to hepatocytes. Their location is in between hepatocytes (Hep) and in close proximity to bile ducts (BD) in the different developmental stages. Scale bars = 2 μm.

structure-function evidence that could not be otherwise obtained with one single microscopy tool alone (Loussert Fonta and Humbel, 2015). Of particular note, others used similar imaging approaches in zebrafish to study transient blood vessel fusion events (Armer et al., 2009), autophagy (Hosseini et al., 2014), and to identify rare cell populations (Goudarzi et al., 2015). Our cross-correlative observations provided a strong degree of confidence to determine the cells within the zebrafish liver that are mpeg1-positive and qualified as macrophages (Fig. 2 and 3). Although we reported the occurrence of zebrafish liver macrophages to be 5-fold lower (3% total liver volume) compared to that found in rodents (15 % of total liver volume), they exhibited similar morphology, such as a large irregular shape with pseudopodia and numerous lysosomes. Their location, a key aspect to identify Kupffer cells in rodents, was found to be different: in the zebrafish, macrophages were situated in between hepatocytes and at close proximity to sinusoids and bile ducts. This implicated that their role in clearing the blood may be altered or be less effective compared to Kupffer cells that are

markers facilitate the detection of small regions of interest within large volumes and high-throughput, using fluorescent microscopy (Cheng et al., 2017) (Fig. 1). For quantitative analysis purposes, confocal microscopy alone on a whole animal was not sufficient as liver boundaries were difficult to draw precisely. Hence, in this study, we instead fixed the animals and processed them using a protocol previously described (Cheng et al., 2017) to preserve in situ fluorescence and quantified the red fluorescent cells using wide-field fluorescence microscopy on resin sections. This method allowed imaging of regions of interest that were located deep within the tissue and, when combined with array tomography technique, we were able to image larger volumes at improved resolution (spatial resolution equivalent to the thickness of the resin sections) (Micheva et al., 2010). Our analysis on the distribution of liver macrophages was performed on 500-nm thick sections imaged every 6 μm that covered the entire liver (the approximate size of macrophage nucleus was 5 μm; see Fig. 2). This analytical bio-imaging approach provided accumulated

7

Micron 132 (2020) 102851

D. Cheng, et al.

Fig. 5. Overview confocal microscopy images showing the distribution of latex bead in zebrafish. Shown are the maximum projection image (A) and corresponding 3D stack rotated to 30° (B) that were captured 10 min post injection. The blue dotted lines mark the approximate limits of the liver. Beads can be seen as single or aggregates within the liver area. However, through this imaging approach, it is difficult to accurately determine whether they are within the liver organ or in the surrounding area (i.e., in the body cavity, the skin and/or the nearby circulation). Scale bar = 40 μm. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article).

the successful application of our direct injection approach but no phagocytic uptake through liver macrophages. We did find beads within the sinusoids, in the circulation, of both the adult and juvenile fish. We conclude that the latex beads accumulated in the cytoplasmic extensions of macrophages present in the kidney and the ventral venous plexus, at least in the larvae stage. This is in line with descriptions by Lieschke et al. (2001), who described these two areas as macrophages preferential areas of residence in the zebrafish. These observations were also in line with descriptions by Crowhurst et al. (2004) who administered carbon particles in zebrafish embryos of 2–3 dpf and located them 1 h later within the ventral vein region. This region was then suggested to be a possible site of production (or preferred site of residence for macrophages). On the other hand, Van Wettere et al. (2013) and Schwendinger-Schreck (2013) suggested the liver macrophages commonly found in fish might play a similar role to Kupffer cells, in terms of phagocytosis of cellular debris due to their shared immunoreactivity responses. However, in our hands, none of the liverassociated macrophages took up any of the injected latex beads 1000-, 100- nor 55 nm size even in the adult stage. It can be proposed that due to their localisation not within the sinusoids the macrophages do not get exposed to the beads to trigger a phagocytic response and the sinusoidal endothelial cells act like a barrier between the blood and the hepatocytes. To rule out the potential wrong choice of beads, we applied a molecular marker to validate to determine the phagocytic property of those macrophages, namely peroxidase staining. This method is widely used to label allows the labelling of peroxidase enzyme, the product of

located within the liver sinusoids. In order to assess the macrophages functional properties, we applied a latex beads injection method and peroxidase staining which are both well-accepted markers of phagocytotic activity in mammalian (Yamashita et al., 1985; Bouwens et al., 1986) and marine animal (Corsaro et al., 2000) hepatic sinusoidal research, including other teleost fish (Hampton et al., 1987). A direct injection into the liver will inevitably cause tissue damage. However, damaged cells can be clearly differentiated from the healthy ones even at the level of light microscopy. This method has also been successfully applied previously to study macromolecular uptake in the liver (Cheng et al., 2019). The results from both studies have been nonconforming: the beads were left circulating for 45 min. before fixation, a time frame chosen to ensure resident macrophages had time to take up these beads. Mathias et al. (2009) determined that inflammatory macrophages responded 2 h after injury while Lieschke et al. (2001) showed phagocytosis uptake in macrophages to occur in as short as 10 min. We used latex beads of 3 different sizes 1000-, 100- and 55 nm to cover possible preferential particle size uptake. Fish from 3 different development stages larvae, juvenile and adult were examined to cover for development differences and distinctive maturation stages. In rodents, the beads would be taken up by liver macrophages in as short as 20 min Timmers et al., 2002). However, in our studies, despite our capacity to detect ‘rare’ cellular events or individual cells within large volumes, we were unable to detect liver macrophages that incorporated latex beads within the liver of our sampled animals (n = 6 animals). Instead, we identified the beads within the circulation, the kidney and the ventral area, indicating

Fig. 6. Brightfield image of toluidene blue stained sections showing the localisation of injected latex beads in the zebrafish. The large latex beads 1000 nm, black arrow and the midsize latex beads 100 nm, red arrow) were seen both predominantly within vessels around the kidney (A,B) and within the circulation of the liver organ (C,D). Latex beads were not observed within any cell type or other anatomical compartments within the liver. Scale bars = 10 μm (A,C) and 2 μm (B,D). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article).

8

Micron 132 (2020) 102851

D. Cheng, et al.

Fig. 7. Back-scattered SEM micrographs from 90 nm cross sections through a 12 dpf zebrafish injected with a mixture of 1000-, 100- and 55 nm sized latex beads. The beads were left circulating for 45 min prior to fixation. High-magnification SEM imaging around the renal area A, B and the ventral venous plexus C of the fish revealed the presence of latex beads. These underpin the succesful administration of beads into the circulation. The areas where latex beads accumulated were false coloured green for ease of visualisation: these structures have elongated shapes but no particular arrangement. Scale bars = 5 μm. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article).

Fig. 8. Higher magnification of kidney-associated macrophages that have accumulated the injected latex beads. A: overview of a mpeg1 zebrafish in cross-section using fluorescence microscopy. Bright red fluorescence can be observed in specific areas. B: higher magnification of the boxed area in (A). Green line delineates the kidney and blue lines delineate the pronephric tubules within the kidney. Intense fluorescence is present within the pronephric tubules but those correspond to lipid vacuoles (as reported previously by Noonan et al., 2016). C: high-magnification SEM image of a cell within the kidney area which has accumulated numerous latex beads (arrows) of approx. 1 μm in size. This structural compartment can be defined as a cell since a nucleus can be seen within the cytoplasm, surrounded by a plasma membrane. Scale bars = 50 μm (A), 25 μm (B), 2 μm (C). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article).

(2004) findings that indicated macrophages in zebrafish were myeloperoxidase activity negative and a defined population of adult and embryonic zebrafish macrophages do not express this enzyme.

phagocytosis, which was confirmed in rodents and human (Deimann, 1984), as well as in other teleost species in which Kupffer cells have been reported (Hampton et al., 1987), In our studies, peroxidase staining also did not label the mpeg1 cells but instead labelled other cells, conceivably granulocytes as suggested by Lieschke et al. (2001). The lack of peroxidase activity is further supported by Onnebo et al. 9

Micron 132 (2020) 102851

D. Cheng, et al.

Fig. 9. Cross-sections of mpeg1 zebrafish across the liver were imaged using fluorescence microscopy followed by peroxidase staining and brightfield imaging to determine the localisation and identification of peroxidase positivecells within the liver. A, fluorescent image showing an overview of macrophages (red) present in the mpeg1 zebrafish. Macrophages were found predominantly in the kidney area (B, D) and some in the liver (C, E). B and C are brightfield images of the boxed areas in A, after DAB peroxidase staining. D and E are the fluorescent images of B and C respectively. F is an overlayed image of C and E, and G of B and D, respectively and at higher magnification. Although some cells have reacted to the peroxidase staining and appear dark brown (arrows), they did not correspond to the macrophages as they did not correlate with the red fluorescent cells. Although further examination would be necessary, the literature suggest they could be granulocytes (Lieschke et al., 2001). Scale bars = 5 μm. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article).

5. Conclusion

from the Australian Microscopy & Microanalysis Research Facilities (AMMRF) at the Australian Centre for Microscopy & Microanalysis (ACMM), The University of Sydney and the Zebrafish facility at Macquarie University. DAB reagent was kindly provided by the Murphy lab from Anatomy and Histology, The University of Sydney. F.B. is indebted to the Australian Research Council (ARC) for funding some of the work reported herein (ARC LE110100203). We wish to thank the generous donors contributing towards the research at Macquarie University and the Snow Foundation for their support towards M.M.

In this study, we have identified and characterised the presence of resident liver macrophages in the zebrafish. Interestingly, while we found that these macrophages express all the typical structural elements, their different location (i.e. not within the liver sinusoidal space) was an indication that their role in clearing the sinusoidal blood might not be maintained. Most importantly, we were unable to demonstrate the characteristic phagocytic function as observed in mammals and other teleost fish. This may suggest that the zebrafish liver macrophages are in an immature stage of genetic or phylogeny differentiation. This all has important implications for future zebrafish studies that aim to increase our understanding of different physiological and pathophysiological processes that are situated within the close proximity of the liver sinusoidal vascular bed, and in particular studies in which liver macrophages are known to play a pivotal role. Hence, additional studies are needed to build a complete picture on the function of zebrafish liver macrophages and particularly how it compares to rodents and humans. For example, outstanding questions are how zebrafish macrophages deal with infectious matter, inflammation, malignant growth or how prone zebrafish are to liver diseases in general. Based on their localisation and structural properties we speculate that macrophages residing between liver parenchymal cells might be actively recruited to the liver sinusoidal space via transmigration upon injury or other adverse events.

Appendix A. Supplementary data Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.micron.2020.102851. References Akilbekova, D., Philiph, R., Graham, A., Bratlie, K.M., 2015. Macrophage reprogramming: influence of latex beads with various functional groups on macrophage phenotype and phagocytic uptake in vitro. J. Biomed. Mater. Res. A. 103, 262–268. https://doi. org/10.1002/jbm.a.35169. Alterman, K., 1977. The kupffer cell, a short historical review. In: Wisse, E., Knook, D.L. (Eds.), Kupffer Cells and Other Liver Sinusoidal Cells. Elsevier / North-Holland Biomedical Press, Amsterdam, pp. 3–20. Anderson, B.G., Mitchum, D.L., 1974. Atlas of Trout Histology. Wyoming Game & Fish Dept., Cheyenne. Anthony, B.J., Ramm, G.A., McManus, D.P., 2012. Role of resident liver cells in the pathogenesis of schistosomiasis. Trends Parasitol. 28, 572–579. https://doi.org/10. 1016/j.pt.2012.09.005. Armer, H.E.J., Mariggi, G., Png, K.M.Y., Genoud, C., Monteith, A.G., Bushby, A.J., Gerhardt, H., Collinson, L.M., 2009. Imaging transient blood vessel fusion events in zebrafish by correlative volume Electron microscopy. PLoS One 4, e7716. https://doi. org/10.1371/journal.pone.0007716. Bautista, A.P., Schuler, A., Spolarics, Z., Spitzer, J.J., 1992. In vivo latex phagocytosis primes the Kupffer cells and hepatic neutrophils to generate superoxide anion. J. Leukoc. Biol. 51, 39–45. https://doi.org/10.1002/jlb.51.1.39. 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., Look, A.T., 2001. Myelopoiesis in the zebrafish, Danio rerio. Blood 98, 643–651. Bilzer, M., Roggel, F., Gerbes, A.L., 2006. Role of Kupffer cells in host defense and liver disease. Liver Int. Off. J. Int. Assoc. Study Liver 26, 1175–1186. https://doi.org/10.

Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Acknowledgements The authors are grateful for the facilities and technical assistance 10

Micron 132 (2020) 102851

D. Cheng, et al.

004. Jones, E.A., Summerfield, J.A., 1988. Kupffer cells. In: Arias, I., Jakoby, W., Popper, H., Schachter, D., Shafritz, D.A. (Eds.), The Liver: Biology and Pathobiology. Raven Press, New York, pp. 683–704. Kalev-Zylinska, M.L., Horsfield, J.A., Flores, M.V.C., Postlethwait, J.H., Vitas, M.R., Baas, A.M., Crosier, P.S., Crosier, K.E., 2002. Runx1 is required for zebrafish blood and vessel development and expression of a human RUNX1-CBF2T1 transgene advances a model for studies of leukemogenesis. Dev. Suppl. 129, 2015–2030. Kamimoto, M., Rung-Ruangkijkrai, T., Iwanaga, T., 2005. Uptake ability of hepatic sinusoidal endothelial cells and enhancement by lipopolysaccharide. Biomed. Res. 26, 99–107. https://doi.org/10.2220/biomedres.26.99. Kanther, M., Rawls, J.F., 2010. Host–microbe interactions in the developing zebrafish. Curr. Opin. Immunol. 22, 10–19. https://doi.org/10.1016/j.coi.2010.01.006. Innate immunity, Antigen processing. Keightley, M.-C., Wang, C.-H., Pazhakh, V., Lieschke, G.J., 2014. Delineating the roles of neutrophils and macrophages in zebrafish regeneration models. Int. J. Biochem. Cell Biol. 56, 92–106. https://doi.org/10.1016/j.biocel.2014.07.010. Kobayashi, K., Cheng, D., Huynh, M., Ratinac, K.R., Thordarson, P., Braet, F., 2012. Imaging fluorescently labeled complexes by means of multidimensional correlative light and transmission electron microscopy: practical considerations. Methods Cell Biol. 111, 1–20. https://doi.org/10.1016/B978-0-12-416026-2.00001-7. 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. https://doi.org/10.1182/blood.V98.10.3087. Loussert Fonta, C., Humbel, B.M., 2015. Correlative microscopy. Arch. Biochem. Biophys. 581, 98–110. https://doi.org/10.1016/j.abb.2015.05.017. MacPhee, P.J., Schmidt, E.E., Groom, A.C., 1992. Evidence for Kupffer cell migration along liver sinusoids, from high-resolution in vivo microscopy. Am. J. Physiol.Gastrointest. Liver Physiol. 263, G17–G23. https://doi.org/10.1152/ajpgi.1992.263. 1.G17. Mathias, J.R., Dodd, M.E., Walters, K.B., Yoo, S.K., Ranheim, E.A., Huttenlocher, A., 2009. Characterization of zebrafish larval inflammatory macrophages. Dev. Comp. Immunol. 33, 1212–1217. https://doi.org/10.1016/j.dci.2009.07.003. Menke, A.L., Spitsbergen, J.M., Wolterbeek, A.P.M., Woutersen, R.A., 2011. Normal anatomy and histology of the adult zebrafish. Toxicol. Pathol. 39, 759–775. https:// doi.org/10.1177/0192623311409597. Micheva, K.D., Smith, S.J., 2007. Array tomography. Neuron 55, 25–36. https://doi.org/ 10.1016/j.neuron.2007.06.014. Micheva, K.D., O’Rourke, N., Busse, B., Smith, S.J., 2010. Array tomography: high-resolution three-dimensional immunofluorescence. Cold Spring Harb. Protoc. 2010https://doi.org/10.1101/pdb.top89. pdb.top89. Morsch, M., Radford, R., Lee, A., Don, E.K., Badrock, A.P., Hall, T.E., Cole, N.J., Chung, R., 2015. In vivo characterization of microglial engulfment of dying neurons in the zebrafish spinal cord. Front. Cell. Neurosci. 9, 321. https://doi.org/10.3389/fncel. 2015.00321. Morsch, M., Radford, R.A.W., Don, E.K., Lee, A., Hortle, E., Cole, N.J., Chung, R.S., 2017. Triggering cell stress and death using conventional UV laser confocal microscopy. JoVE J. Vis. Exp. https://doi.org/10.3791/54983. e54983–e54983. Müller-Reichert, T., Verkade, P., 2012. Introduction to correlative light and electron microscopy. Methods Cell Biol. 111, xvii–xix. https://doi.org/10.1016/B978-0-12416026-2.03001-6. Naito, M., Hasegawa, G., Ebe, Y., Yamamoto, T., 2004. Differentiation and function of Kupffer cells. Med. Electron Microsc. Off. J. Clin. Electron Microsc. Soc. Jpn. 37, 16–28. https://doi.org/10.1007/s00795-003-0228-x. Noonan, H.R., Metelo, A.M., Kamei, C.N., Peterson, R.T., Drummond, I.A., Iliopoulos, O., 2016. Loss of vhl in the zebrafish pronephros recapitulates early stages of human clear cell renal cell carcinoma. Dis. Model. Mech. 9, 873–884. https://doi.org/10. 1242/dmm.024380. Onnebo, S.M.N., Yoong, S.H.S., Ward, A.C., 2004. Harnessing zebrafish for the study of white blood cell development and its perturbation. Exp. Hematol. 32, 789–796. https://doi.org/10.1016/j.exphem.2004.04.012. Parichy, D.M., Ransom, D.G., Paw, B., Zon, L.I., Johnson, S.L., 2000. An orthologue of the kit-related gene fms is required for development of neural crest-derived xanthophores and a subpopulation of adult melanocytes in the zebrafish, Danio rerio. Dev. Suppl. 127, 3031–3044. Rocha, E., Monteiro, R.A., Pereira, C.A., 1994. The liver of the brown trout, Salmo trutta fario: a light and electron microscope study. J. Anat. 185, 241–249. Rocha, E., Monteiro, R.A.F., Pereira, C.A., 1997. Liver of the brown trout, Salmo trutta (Teleostei, Salmonidae): a stereological study at light and electron microscopic levels. Anat. Rec. 247, 317–328. https://doi.org/10.1002/(SICI)1097-0185(199703) 247:3<317::AID-AR3>3.0.CO;2-R. Sakano, E., Fujita, H., 1982. Comparative aspects on the fine structure of the teleost liver. Okajimas Folia Anat. 58, 501–519. https://doi.org/10.2535/ofaj1936.58.4-6_501. Sasse, D., Spornitz, U.M., Maly, I.P., 1992. Liver architecture. Enzyme 46, 8–32. Schwendinger-Schreck, J., 2013. The zebrafish: atlas of macroscopic and microscopic anatomy. Yale J. Biol. Med. 86, 435. Sedgewick, J., 2008. Scientific Imaging With Photoshop: Methods, Measurement, and Output, 1 edition ed. New Riders, Berkeley, CA. Seki, S., Nakashima, H., Nakashima, M., Kinoshita, M., 2011. Antitumor immunity produced by the liver Kupffer cells, NK cells, NKT cells, and CD8 CD122 T cells. Clin. Dev. Immunol. 2011, 868345. https://doi.org/10.1155/2011/868345. Su, Y., Nykanen, M., Jahn, K.A., Whan, R., Cantrill, L., Soon, L.L., Ratinac, K.R., Braet, F., 2010. Multi-dimensional correlative imaging of subcellular events: combining the strengths of light and electron microscopy. Biophys. Rev. 2, 121–135. https://doi. org/10.1007/s12551-010-0035-2. Svahn, A.J., Don, E.K., Badrock, A.P., Cole, N.J., Graeber, M.B., Yerbury, J.J., Chung, R.,

1111/j.1478-3231.2006.01342.x. Blouin, A., 1977. Morphometry of liver sinusoidal cells. In: Wisse, E., Knook, D.L. (Eds.), Kupffer Cells and Other Liver Sinusoidal Cells. Elsevier / North-Holland Biomedical Press, Amsterdam, pp. 61–71. Blumer, M.J.F., Gahleitner, P., Narzt, T., Handl, C., Ruthensteiner, B., 2002. Ribbons of semithin sections: an advanced method with a new type of diamond knife. J. Neurosci. Methods 120, 11–16. https://doi.org/10.1016/S0165-0270(02)00166-8. Bolte, S., Cordelières, F.P., 2006. A guided tour into subcellular colocalization analysis in light microscopy. J. Microsc. 224, 213–232. https://doi.org/10.1111/j.1365-2818. 2006.01706.x. Boltjes, A., Movita, D., Boonstra, A., Woltman, A.M., 2014. The role of Kupffer cells in hepatitis B and hepatitis C virus infections. J. Hepatol. 61, 660–671. https://doi.org/ 10.1016/j.jhep.2014.04.026. Bouwens, L., 1988. Structural and functional aspects of Kupffer cells. Revis. Sobre Biol. Cel. RBC 16, 69–94. Bouwens, L., Baekeland, M., De Zanger, R., Wisse, E., 1986. Quantitation, tissue distribution and proliferation kinetics of Kupffer cells in normal rat liver. Hepatol. Baltim. Md 6, 718–722. Brunt, E.M., Gouw, A.S.H., Hubscher, S.G., Tiniakos, D.G., Bedossa, P., Burt, A.D., Callea, F., Clouston, A.D., Dienes, H.P., Goodman, Z.D., Roberts, E.A., Roskams, T., Terracciano, L., Torbenson, M.S., Wanless, I.R., 2014. Pathology of the liver sinusoids. Histopathology 64, 907–920. https://doi.org/10.1111/his.12364. Cheng, D., Shami, G.J., Morsch, M., Chung, R.S., Braet, F., 2016. Ultrastructural mapping of the zebrafish gastrointestinal system as a basis for experimental drug studies. Biomed Res. Int. 2016, 8758460. https://doi.org/10.1155/2016/8758460. Cheng, D., Shami, G., Morsch, M., Huynh, M., Trimby, P., Braet, F., 2017. Relocation is the key to successful correlative fluorescence and scanning electron microscopy. In: Müller-Reichert, T., Verkade, P. (Eds.), Methods in Cell Biology, Correlative Light and Electron Microscopy III. Academic Press, pp. 215–244. https://doi.org/10.1016/bs. mcb.2017.03.013. Cheng, D., Morsch, M., Shami, G.J., Chung, R.S., Braet, F., 2019. Albumin uptake and distribution in the zebrafish liver as observed via correlative imaging. Exp. Cell Res. 374, 162–171. https://doi.org/10.1016/j.yexcr.2018.11.020. Corsaro, C., Scalia, M., Leotta, N., Mondio, F., Sichel, G., 2000. Characterisation of Kupffer cells in some Amphibia. J. Anat. 196, 249–261. https://doi.org/10.1046/j. 1469-7580.2000.19620249.x. Crowhurst, M.O., Layton, J.E., Lieschke, G.J., 2004. Developmental biology of zebrafish myeloid cells. Int. J. Dev. Biol. 46, 483–492. https://doi.org/10.1387/ijdb. 12141435. Deimann, W., 1984. Endogenous peroxidase activity in mononuclear phagocytes. Prog. Histochem. Cytochem. 15, 1–58. Dixon, L.J., Barnes, M., Tang, H., Pritchard, M.T., Nagy, L.E., 2013. Kupffer cells in the liver. Compr. Physiol. 3, 785–797. https://doi.org/10.1002/cphy.c120026. Douglas, S.D., Tuluc, F., 2010. Chapter 67. Morphology of monocytes and macrophages. In: Lichtman, M.A., Kipps, T.J., Seligsohn, U., Kaushansky, K., Prchal, J.T. (Eds.), Williams Hematology. The McGraw-Hill Companies, New York, NY. Ellett, F., Pase, L., Hayman, J.W., Andrianopoulos, A., Lieschke, G.J., 2011. mpeg1 promoter transgenes direct macrophage-lineage expression in zebrafish. Blood 117, e49–e56. https://doi.org/10.1182/blood-2010-10-314120. Fahrner, R., Dondorf, F., Ardelt, M., Settmacher, U., Rauchfuss, F., 2016. Role of NK, NKT cells and macrophages in liver transplantation. World J. Gastroenterol. 22, 6135–6144. https://doi.org/10.3748/wjg.v22.i27.6135. Ferri, S., Sesso, A., 1981. Ultrastructural study of Kupffer cells in teleost liver under normal and experimental conditions. Cell Tissue Res. 220, 387–391. https://doi.org/ 10.1007/BF00210516. Frevert, U., Usynin, I., Baer, K., Klotz, C., 2006. Nomadic or sessile: can Kupffer cells function as portals for malaria sporozoites to the liver? Cell. Microbiol. 8, 1537–1546. https://doi.org/10.1111/j.1462-5822.2006.00777.x. Gerri, C., Marin-Juez, R., Marass, M., Marks, A., Maischein, H.-M., Stainier, D.Y.R., 2017. Hif-1 alpha regulates macrophage-endothelial interactions during blood vessel development in zebrafish. Nat. Commun. 8, 15492. https://doi.org/10.1038/ ncomms15492. Goudarzi, M., Mildner, K., Babatz, F., Riedel, D., Klämbt, C., Zeuschner, D., Raz, E., 2015. Correlative light and Electron microscopy of rare cell populations in zebrafish embryos using laser marks. Zebrafish 12, 470–473. https://doi.org/10.1089/zeb.2015. 1148. Hampton, J.A., Klaunig, J.E., Goldblatt, P.J., 1987. Resident sinusoidal macrophages in the liver of the brown bullhead (Ictalurus nebulosus): an ultrastructural, functional and cytochemical study. Anat. Rec. 219, 338–346. https://doi.org/10.1002/ar. 1092190403. Hedrera, M.I., Galdames, J.A., Jimenez-Reyes, M.F., Reyes, A.E., Avendaño-Herrera, R., Romero, J., Feijóo, C.G., 2013. Soybean meal induces intestinal inflammation in zebrafish larvae. PLoS One 8, e69983. https://doi.org/10.1371/journal.pone. 0069983. Herbomel, P., Thisse, B., Thisse, C., 1999. Ontogeny and behaviour of early macrophages in the zebrafish embryo. Dev. Suppl. 126, 3735–3745. Herbomel, P., Thisse, B., Thisse, C., 2001. Zebrafish early macrophages colonize cephalic mesenchyme and developing brain, Retina, and epidermis through a M-CSF receptordependent invasive process. Dev. Biol. 238, 274–288. https://doi.org/10.1006/dbio. 2001.0393. Hosseini, R., Lamers, G.E., Hodzic, Z., Meijer, A.H., Schaaf, M.J., Spaink, H.P., 2014. Correlative light and electron microscopy imaging of autophagy in a zebrafish infection model. Autophagy 10, 1844–1857. https://doi.org/10.4161/auto.29992. Jahn, K.A., Barton, D.A., Kobayashi, K., Ratinac, K.R., Overall, R.L., Braet, F., 2012. Correlative microscopy: providing new understanding in the biomedical and plant sciences. Micron 1993 (43), 565–582. https://doi.org/10.1016/j.micron.2011.12.

11

Micron 132 (2020) 102851

D. Cheng, et al.

marker of macrophage renewal during BCG-induced inflammation in the rat lung. Am. J. Pathol. 128, 171–180. Wei, L.-N., 1997. Transgenic animals as new approaches in pharmacological studies. Annu. Rev. Pharmacol. Toxicol. 37, 119–141. https://doi.org/10.1146/annurev. pharmtox.37.1.119. Weibel, E.R., Stäubli, W., Gnägi, H.R., Hess, F.A., 1969. Correlated morphometric and biochemical studies on the liver cell. I. Morphometric model, stereologic methods, and normal morphometric data for rat liver. J. Cell Biol. 42, 68–91. Westerfield, M., 2000. The Zebrafish Book a Guide for the Laboratory Use of Zebrafish Danio (Brachydanio) Rerio. ZFIN [Eugene, Or.]. Wisse, E., 1974. Observations on the fine structure and peroxidase cytochemistry of normal rat liver Kupffer cells. J. Ultrastruct. Res. 46, 393–426. Wisse, E., 1977. Ultrastructure and function of Kupffer cells and other sinusoidal cells in the liver. Med. Chir. Dig. 6, 409–418. Wisse, E., Braet, F., Luo, D., De Zanger, R., Jans, D., Crabbé, E., Vermoesen, A., 1996. Structure and function of sinusoidal lining cells in the liver. Toxicol. Pathol. 24, 100–111. https://doi.org/10.1177/019262339602400114. Yamashita, K., Fujita, H., Kawamata, S., 1985. Fine structural and cytochemical aspects of granuloma formation derived from Kupffer cells in mice injected with latex particles. Arch. Histol. Jpn. Nihon Soshikigaku Kiroku 48, 315–326. Zhang, P., Jiang, X., Nie, X., Huang, Y., Zeng, F., Xia, X., Wu, S., 2016. A two-photon fluorescent sensor revealing drug-induced liver injury via tracking γ-glutamyltranspeptidase (GGT) level in vivo. Biomaterials 80, 46–56. https://doi.org/10.1016/ j.biomaterials.2015.11.047.

Morsch, M., 2018. Nucleo-cytoplasmic transport of TDP-43 studied in real time: impaired microglia function leads to axonal spreading of TDP-43 in degenerating motor neurons. Acta Neuropathol. (Berl.) 136, 445–459. https://doi.org/10.1007/ s00401-018-1875-2. Timmers, M., Vermijlen, D., Vekemans, K., Zanger, R.D., Wisse, E., Braet, F., 2002. Tracing DiO-labelled tumour cells in liver sections by confocal laser scanning microscopy. J. Microsc. 208, 65–74. https://doi.org/10.1046/j.1365-2818.2002. 01063.x. Tsutsui, H., Nishiguchi, S., 2014. Importance of Kupffer cells in the development of acute liver injuries in mice. Int. J. Mol. Sci. 15, 7711–7730. https://doi.org/10.3390/ ijms15057711. Van Wettere, A.J., Law, J.M., Hinton, D.E., Kullman, S.W., 2013. Anchoring hepatic gene expression with development of fibrosis and neoplasia in a toxicant-induced fish model of liver injury. Toxicol. Pathol. 41, 744–760. https://doi.org/10.1177/ 0192623312464308. von Kupffer, C., 1876. Uber sternzellen der leber. Briefliche mitteilung an Prof. Waldeyer. Arch Mikrosc Anat 12, 353–358. Wacker, I., Spomer, W., Hofmann, A., Thaler, M., Hillmer, S., Gengenbach, U., Schröder, R.R., 2016. Hierarchical imaging: a new concept for targeted imaging of large volumes from cells to tissues. BMC Cell Biol. 17. https://doi.org/10.1186/s12860-0160122-8. Wake, K., Decker, K., Kirn, A., Knook, D.L., McCuskey, R.S., Bouwens, L., Wisse, E., 1989. Cell biology and kinetics of Kupffer cells in the liver. Int. Rev. Cytol. 118, 173–229. Wardle, E.N., 1987. Kupffer cells and their function. Liver 7, 63–75. Warnock, M.L., Sniezek, M., Shellito, J., 1987. Endogenous peroxidase activity as a

12