Comparative Biochemistry and Physiology, Part C 155 (2012) 18–25
Contents lists available at ScienceDirect
Comparative Biochemistry and Physiology, Part C j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / c b p c
A high-throughput histoarray for quantitative molecular profiling of multiple, uniformly oriented medaka (Oryzias latipes) embryos☆ Napo K.M. Cheung a, David E. Hinton b, Doris W.T. Au a,⁎ a b
State Key Laboratory in Marine Pollution and Department of Biology and Chemistry, City University of Hong Kong, Kowloon, Hong Kong Special Administrative Region Division of Environmental Sciences and Policy, Nicholas School of the Environment, Duke University, Durham, North Carolina, USA
a r t i c l e
i n f o
Article history: Received 13 February 2011 Received in revised form 18 May 2011 Accepted 20 May 2011 Available online 1 June 2011 Keywords: Early life stage toxicity Medaka Molecular toxicology Multiplex embryos histoarray Hypoxia
a b s t r a c t Embryos of aquatic animal model fish have proven to be useful organisms for developmental biology and for early life stage toxicity tests. By virtue of their transparent chorions, imaging of normal and abnormal development can be detected and related to exposure or to alterations due to environmental factors. However, the detection of changes at sub-individual levels of organization is hampered by time required to detect important events within cells and tissues of affected organisms. We describe herein development of a highly cost effective embryo chip enabling stringent inter-individual comparisons and multiplex detection in embryos and eleutheroembryos. As a proof of principle we examine cell proliferation and controlled cell death in normoxic and hypoxic conditions and relate these to tissue turnover in individual organisms. Coupled with a recently developed whole adult animal platform, we can now move beyond the common approach focusing on single target organ to the detection and characterization of systemic phenomena (syndromes) affecting the organism. Taken together, we can now determine adult consequences of early life stage exposure and assess ability of exposed individuals to respond to stresses superimposed along the axis of time. © 2011 Elsevier Inc. All rights reserved.
1. Introduction Model fish embryo (Schmale et al., 2007; Hinton et al., 2009) is an emerging tool for various scientific fields such as developmental biology (Wittbrodt et al., 2002) and early life stage toxicity tests with compounds/factors of potential biomedical and environmental significance (Lammer et al., 2009; Howarth et al., 2010). Detailed atlases on sequential developmental processes of medaka (Iwamatsu, 2004) and zebrafish (Kimmel et al., 1995) have provided precise information on timing needed for normal developmental stages to be achieved; and, when coupled with controlled environmental conditions (light, temperature), the time for- and quality of development provide useful endpoints in early life stage toxicity tests (Hardman et al., 2008; Smith et al., 2009). In addition, economy of these assays is realized by the large number of embryos and fresh hatchlings that can be used to provide precise dose–response relationships (Villalobos et al., 2000).
☆ This paper is based on a presentation given at the 5th Aquatic Annual Models of Human Disease conference: hosted by Oregon State University and Texas State University-San Marcos, and convened at Corvallis, OR, USA September 20–22, 2010. ⁎ Corresponding author at: Department of Biology and Chemistry, City University of Hong Kong, Tat Chee Avenue, Kowloon, Hong Kong SAR. E-mail address:
[email protected] (D.W.T. Au). 1532-0456/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.cbpc.2011.05.010
While the above advantages are obvious, the small size of these organisms makes isolation of specific organs and subsequent localization of gene expression, and proteins difficult within the intact organism. In addition, mechanistic toxicological investigations require an understanding of molecular events within specific cells and tissues (Bozinovic et al., 2011). Immunohisto- and cytochemistry in sections of control and treated embryos require precise orientation among and between organisms of experimental groups at the cell, tissue and organ levels to localize biological responses. Recently, a platform using intact adult medaka was developed enabling high resolution detail of organs, tissues and cells in individuals simultaneously (Kong et al., 2008). This platform has been applied to various studies illustrating the value of analysis of syndromes involving multiple organ systems (i.e., endocrine disruption with involvement of brain, pituitary, liver, and ovary) (Park et al., 2008; Park et al., 2009; Tompsett et al., 2009). While this has worked well with adult fish, what is needed is a method specifically for early life stages that would provide a reliable way to process multiple organisms simultaneously while ensuring uniformity and precision of orientation among individuals. A technique for processing and sectioning multiple 7-day-old zebrafish (Danio rerio) larvae (hatchlings) (Tsao-Wu et al., 1998) employed agarose-embedded tissue arrays and yielded very satisfactory results. Briefly, an engineered acrylic mold with 128 teeth of 5 × 0.8 × 1 mm dimensions was specially prepared (not available
N.K.M. Cheung et al. / Comparative Biochemistry and Physiology, Part C 155 (2012) 18–25
commercially). Agarose was poured over the mold and allowed to set producing a flat platform with 128 wells. Two previously fixed zebrafish larvae were placed in each well, oriented to specific conditions, and processed as a single, multi-specimen block and then sectioned. Our attempts to repeat this process proved unsuccessful. Despite efforts with various fabrication firms, we were not able to obtain similarly tooled acrylic molds that would accommodate eleutheroembryos (newly hatched free swimming larvae with yolk sacs attached; see Embry et al., 2010) nor the sphere-like shape of embryonated eggs of medaka. Moreover, the chorions, i.e., the surrounding membranes of embryonated eggs have proved formidable barriers to embedment and subsequent processing steps. We report herein a highly cost effective procedure that has enabled us to uniformly and precisely orient chorionated embryos and eleutheroembryos (EEs) in agarose molds. This high throughput platform (embryo chip) allows synchronical processing and sectioning of multiple chorionated embryos and EEs, providing high resolution evaluation of multiple, similarly-oriented individuals simultaneously. As a proof of principle, we applied the method to multiple Japanese ricefish (medaka; Oryzias latipes) embryos that were part of a larger investigation of chronic effects of aquatic hypoxia on growth of medaka embryos and their livers. Endpoints reported in this paper were hepatic biometry and tissue turnover (as measured by PCNA cell proliferation and TUNEL apoptosis).
2. Materials and methods 2.1. Fish source Japanese rice fish (medaka; Oryzias latipes), of orange–red outbred lines, originated from the Duke University, Molecular Aquatic Toxicology Laboratory. All fish were maintained under conditions approved by the Institutional Animal Care and Use Committee (IACUC), Duke University. In 2008, a cohort of medaka was transferred to City University of Hong Kong and maintained as a colony since then.
2.2. Culture and care The City University of Hong Kong breeding stock was maintained in aquaria of the following dimensions (in cm): 39.5 L × 23.5 W × 27.5 D. Spawning conditions (26 ± 1 °C; pH 7.3 ± 0.1; 7.2 ± 0.2 mg O2 L − 1; 14 h — light:10 h — dark) were maintained in the static aquaria equipped with activated carbon filters (ClearView 100, Aqua One, Australia) suspended from chamber wall. Each day, 50% of the tank water was replaced with dechlorinated tap water. Breeders were fed twice daily with Otohime β1 (Nisshin Co, Japan) and supplemented with hatched brine shrimp (Artemia nauplia) (Lucky Brand, O.S.I. Marine Lab, USA) once daily for 3 days per week. Under the care and culture conditions described above, medaka spawned daily and masses of embryonated eggs (20–30/mass) were collected at 1 h after the onset of light-period. Collected eggs were filtered through- and gently rubbed against a fish net to detach individual eggs from egg clusters. The eggs were thoroughly rinsed to remove adhered fecal matter and incubated in an embryo rearing medium (ERM; Hardman et al., 2007) with 0.001% methylene blue to prevent fungal infection. The developmental stages were identified and addressed with reference to the morphological characteristics established by Iwamatsu (2004). Viable, normal Stage 9 (late morula) embryos were chosen for this study and groups of individual embryos were monitored daily until hatching (Kirchen and West, 1976), usually 9 days post-fertilization. This regime provided embryos and EEs for analyses as described below. In addition to the Duke University animal care and use protocols, strict adherence to the European Union Directive (EC, 1986) was maintained throughout the studies.
19
2.3. Anesthesia Chorionated embryos and EEs were sedated by placement in icecold ERM, and were immediately fixed as below. 2.4. Fixation and processing Individuals were fixed by submersion in 10 vol. of fresh 4% paraformaldehyde (Electron Microscopy Sciences, USA) in 0.1 M sodium phosphate buffer (pH 7.2, ~200 Osm) and stored overnight at 4 °C before subsequent processing. 2.5. Agarose gels for embedment and orientation of multiple embryos (Embryo chip) An agarose bed was prepared by boiling 1.5% agarose (molecular grade, Sigma-Aldrich, USA) in double-distilled water, cooling to ~55 °C and casting in a polystyrene Petri dish (60 mm) at ca. 3 mm thickness. The resultant cast was then chilled at 4 °C for 5 min (rapid gelling). A glass Pasteur pipette (mouth outer diameter ~1.5 mm; Brand GmbH, Wertheim, Germany) was used to punch individual wells in the gel, all the way down to the Petri dish bottom. Commonly, we prepared 5 columns of individual wells in 5 rows at intervals of ~1 mm. Each well was filled with 0.1 M sodium phosphate buffer (pH 7.2) and air bubbles were removed by fine needle aspiration. Under dissecting microscope, a previously fixed embryo was inserted into the well, carefully pushed down with a glass Pasteur pipette until it touched the Petri dish bottom. This procedure ensured alignment of organisms on the same horizontal plane. Next, organisms in wells were gently moved with fine forceps (Dumont no. 4/no.5, Fine Science Tools, USA) to the desired orientation. Once desired orientation was achieved, fine forceps tips were used to carefully puncture the chorions, but not the individual embryos. This practice provided pathways for entry of embedding medium in subsequent procedures described below. Excess buffer on gel was blotted using lintless lens paper (KimWipes, Kimberley-Clark, USA) and the area containing the embryos was excised with a scalpel. If necessary, orientation of each embryo on the excised gel was verified and reoriented to original position under a dissecting microscope. After this, the excised gel was briefly dipped into molten agarose (at 55 °C) and cooled for 1 min at 4 °C completely sealing wells and their embryo contents. Procedure for hatched/dechorionated EEs was slightly different due to their elongated shape. Agarose gel was prepared as above. Next, a series of three or more overlapping wells was made, forming a single canal. Once EEs were transferred into the appropriate canals, any air bubbles and excess liquid were removed as above. Individual EEs was gently pushed with forceps to one side of the canal where it remained attached to the agarose wall by capillary action. A drop of molten agarose was then added to fill up space in the canal, and, whenever desired, orientation was fine-tuned with forceps until the agarose fully solidified. Subsequent steps were followed as above. The embryo chip was immersed into 50% ethanol in a glass Petri dish (90 mm). Observable bubbles in gel matrix were removed by piercing through with fine forceps, or fine diameter needle and aspirated or dislodged by gentle pressing on the chip. Bubble-free chips with specimens were treated, as a single tissue specimen, and dehydrated through a series of ethanol solutions [70% (1 h), 95% (1 h), and three changes of 100% ethanol (1 h each)]. Next, they were cleared in xylene until transparent and mildly agitated until individual gels sink (usually 1–2 h). Cleared blocks were infiltrated with four changes of paraffin (Paraplast X-tra, Kendall, USA; 1 h each) at 55 °C. Cured paraffin blocks were serially sectioned using a rotary microtome at 5 μm thickness. Each section was mounted on individual Menzel-Glaser Superfrost Plus glass slide (Glasbearbeitungswerk GmbH & Co., Germany) and air-dried overnight at room temperature.
20
N.K.M. Cheung et al. / Comparative Biochemistry and Physiology, Part C 155 (2012) 18–25
2.6. Hematoxylin and eosin (H&E) staining
2.9. Terminal UTP nick end labeling (TUNEL)
Paraffinized embryo chip sections were pre-melted at 60 °C, dewaxed in two-changes of xylene (5 min each) and rehydrated through 100% (3 min), 100% (3 min), 95% (1 min), 70% (1 min) ethanol and two-changes of double-distilled water (1 min each). Sections were submerged in Mayer's hematoxylin (Dako, Denmark) for 1 min and rinsed under running tap water. Excessive stain was removed by three quick dips in acid ethanol (1% concentrated HCl in 70% ethanol) and rinsed as above. Bluing was done in 0.1% sodium bicarbonate solution for 1 min. Rinsed sections were equilibrated with 70% ethanol for 1 min and immersed in eosin solution (0.2% eosin Y and 0.5% glacial acetic acid in 80% ethanol) for 30 s, then thoroughly rinsed. Stained sections were dehydrated through 70% (1 min), 95% (1 min), 100% (two changes, 1 min each) ethanol and two changes of xylene (3 min each). Finally, sections were mounted in Permount (FisherScientific, USA).
De-waxing, pseudoperoxidase activity quenching and rehydration were done as described. Nuclei acids were unmasked by brief heating in 0.01 M citrate buffer, pH 6.0, for 1 min in microwave oven (900 W). Nuclei with DNA nicks were labeled using in situ Cell Death Kit, POD (Roche, Switzerland) according to the procedures recommended by the manufacturer. In brief, TUNEL reaction mixture (vial1, vial2) was incubated on the sections for 60 min at 37 °C, under cover of parafilm to minimize evaporation, and washed as above. Fluorescencechromogen signal conversion was carried out by incubation with anti-fluorescein antibody (Converter-POD, vial 3) at 37 °C for 30 min (under parafilm cover). Sections were washed, visualized, counterstained and mounted as previously described.
2.7. Hypoxia exposure In order to establish proof of principle, we examined partial findings from an extensive analysis of the effects of chronic hypoxia on medaka embryo development, growth, and time to hatch. An equal number of medaka embryos (15 replicates; 30 embryos per replicate) was assigned to either hypoxia treatment group (1.5 ±0.1 mg O2 L − 1) or to normoxia control group (7.2 ± 0.2 mg O2 L − 1) from Stage 9 until 39 (hatching — Iwamatsu, 2004). Construction of exposure systems followed procedures of Yu et al. (2006). Individual systems were dedicated to normoxia or to hypoxia conditions. In each system, ERM was pumped through filter units (ClearView 100) at the maximum flow rate (~ 160 L/h). Nitrogen and room air were mixed in the medium to reach desired oxygen level for each system. These levels were maintained and continuously monitored throughout the experiment. Replicate net cages were used to suspend embryos in each system. Other physical parameters were maintained identical to conditions of breeding tanks (e.g. 26 ± 1 °C; 14 h — light:10 h — dark, etc.). Survivors were verified by activity and phenotypes indicating normal or abnormal development (Hardman et al., 2007). Remaining normoxic and hypoxic embryos as well as EEs were incorporated into embryo chips, processed and sectioned as above. The sections were used for H&E staining, proliferating cell nuclear antigen (PCNA) immunocytochemistry and Terminal UTP nick end labeling (TUNEL) assays.
2.8. Proliferating cell nuclear antigen (PCNA) immunohistochemistry Sections were de-waxed and rehydrated as above except that endogenous pseudoperoxidase activity was quenched by 20 min immersion in 3% H2O2 (in 100% methanol) after de-waxing. Antigen retrieval was mediated by three rounds of heating and cooling (5 min each) in 0.01 M citrate buffer, pH 6.0, in microwave oven (900 W). Sections were allowed to cool to room temperature and blocked for 30 min with 1% bovine serum albumin (GE Healthcare, United Kingdom) in phosphate buffered saline (pH 7.4). Mouse monoclonal anti-PCNA antibody (1:2000; PC10, Dako) was incubated on the sections at room temperature for 2 h and washed off by three changes of tris-buffered saline, pH 7.4 (5 min each with mild agitation). Horseradish peroxidase (HRP) conjugated, anti-mouse secondary antibody (Labeled Polymer, Dako) was incubated on the sections at room temperature for 30 min and washed off as above. Targets were visualized through Dako DAB Envision Detection System and counterstained with hematoxylin. Slides were dehydrated and mounted as above.
2.10. Imaging Cured slides were imaged under bright-field microscope (Zeiss Axioplan 2) coupled with CCD camera (ColorView II) through analySIS Professional software (Soft Imaging System, Germany). Multiple overlapping fields (400×) completely spanning each of the embryos (~100 fields for the whole embryo including 1–2 fields covering the liver) were taken, stored lossless as TIFF and assembled into panorama using Adobe Photoshop CS5. 2.11. Stereological quantification Quantification of PCNA and TUNEL signals followed that of Kong et al. (2008). Briefly, multiple sections of the same reaction containing the tissue-of-interest (whole embryos or their livers) were randomly selected. Volume density of positive signals within the tissue-of-interest was estimated by point counting following stereological principles of Weibel (1979) and Reed and Howard (1998). The estimation is unbiased provided that adequate number of sections is taken into account. As determined by preliminary measurements, 5 random sections (panoramas) were found to be optimal in minimizing variance regardless of staining and target tissue (data not shown), hence designated for all subsequent quantification procedures. The same principle was also applied to estimate the volume ratio between the liver and the corresponding embryo, but expressed as hepatosomatic index (HSI) instead. 2.12. Statistics Normality and homoscedascity were checked with Shapiro–Wilk test and Levene's test respectively. Student t-test was replaced by Mann–Whitney test whenever the data failed normality assumption. All statistical tests were conducted using SPSS 19 (IBM, USA) at α = 0.05. Effect sizes (d) were calculated at 1-β = 0.8. Unless otherwise specified, error range was reported as 95% confidence interval. 3. Results 3.1. Overview of the embryo chips Multiple rows and columns of individual chorionated medaka embryos were embedded in a single paraffin block (Fig. 1A). Higher magnification (Fig. 1B) reveals that 25 individual embryos are contained in 1 cm 2 of block face. In addition, the uniform alignment across individuals was revealed as denoted by the similar position of dark structure (left eye) in each individual. When sectioned, stained and mounted on glass histological slides, low magnification views of embryos included the embryo proper and the attached yolk sac (Fig. 1D). In the same figure the rostrum was oriented toward the left in each individual. With slight modification, this design was also adaptable to orient eleutheroembryos at any desired angle (Fig. 2).
N.K.M. Cheung et al. / Comparative Biochemistry and Physiology, Part C 155 (2012) 18–25
21
Fig. 1. Elements of the embryo chip. A: Paraffin block is supported in cassette (yellow). View of block face shows agarose chip embedded near surface. B: Trimmed block face with magnified view showing 25 embryos in similar orientation. C: Image of living late stage medaka embryo in chorion prior to fixation and processing. D: Photomicrographs of neighboring embryos following sectioning and staining (H&E) for survey morphology. Large pink areas in each are yolk in yolk sac. Note similar orientation of embryos for comparative morphological assessment.
Higher magnification views of internal organs revealed the components of brain, differentiating gill arches and kidney tubules using survey morphology, PCNA immunohistochemistry and TUNEL assay (Fig. 3). As is shown, selection of fields from different regions of individual embryos permits stringent comparison.
3.2. Proliferation and apoptosis after hypoxia exposure In the whole embryos, PCNA-positive nuclei were more abundant in normoxia control (Fig. 4A) than those that were exposed to hypoxia (Fig. 4B). In liver, the number of hepatocytes positive for PCNA (proliferative) in the normoxic liver (Fig. 5A) far exceeded that of the hypoxia (Fig. 5B). PCNA proliferation in both the whole embryos and
livers were significantly suppressed under hypoxia (p b 0.05)(Fig. 6A and C). By contrast, similar abundance of TUNEL-positive (apoptotic) cells was found between normoxic and hypoxic embryos (Fig. 4C and D) and livers (Fig. 5C and D). No significant difference in spatial abundance of TUNEL-positive nuclei was detected between treatments (Fig. 6C and D). The liver PCNA:TUNEL ratio in normoxia control was 4.64 (well above 1, i.e. PCNA ≫ TUNEL). However, the livers from hypoxiaexposed embryos had a PCNA:TUNEL ratio (0.59) lower than 1 (i.e. PCNA b TUNEL) indicating proliferation in liver was over-suppressed and out-rated by the apoptosis after exposure. Contrastingly, the whole body remained proliferative (PCNA:TUNEL = 1.46) under hypoxia, despite in a lesser extent than the normoxia control (PCNA:TUNEL = 2.11). 3.3. Hepatosomatic index Marked differences in liver area were seen when sections of embryos from normoxic and from hypoxic groups were compared (Fig. 5: A vs. B; C vs. D). In contrast to the HSI of normoxic controls (0.991 ± 0.291%), the liver–body volume ratio of hypoxia-exposed embryos (0.358 ± 0.066%) showed severe reduction (d = 2.887; p = 0.007). This suggests that liver growth was uncoupled from the body growth under hypoxic conditions of this study. This could be accounted for by the more severe suppression of proliferation in hypoxic liver as reported above. 4. Discussion
Fig. 2. Eleutheroembryos embedded in embryo chip with rostrum pointed to the left. Preparation of this chip, as described in methods, is only slightly different from the one for chorionated embryos. A: Parasagittal section obtained by orienting the sagittal plane of an embryo parallel to the surface of the chip, showing the cranial portion, wellsegmented somites and yolk sac. B: Oblique section of a neighbor embryo oriented at different angle. The position of this embryo was adjusted to have its sagittal plane tilted from the chip surface by ~ 45°. Sectioning resulted in diagonal cutting from the left eye and inner ear through the right branchial arches along the body axis. This illustrates the flexibility of definable orientation within the embryo chip. H&E stain. Scale bar = 300 μm.
4.1. Embryo chip Inexpensive, cost effective and high throughput medaka embryo chips have been produced. The construction process requires materials that are readily available in most biological laboratories: agarose, glass Pasteur pipette, pointed forceps and a dissecting microscope. This design is not only cost-effective but also streamlines the processing of immense number of embryos. Each chip holds 25 organisms
22
N.K.M. Cheung et al. / Comparative Biochemistry and Physiology, Part C 155 (2012) 18–25
Fig. 3. Regions of individual embryos within a given chip section were assembled and arranged to create each horizontal array (same reaction). In this assemblage, vertical arrays compared serial 5 μm sections using different reactions while horizontal array illustrated similarly reacted parasagittal sections of embryos. Top array (A–C) shows H&E staining. A: Showing retina (lower left hand corner) and brain (mesencephalon — left, cerebellum — center and rhombencephalon — right). Pink line at upper left and right corners is chorion. B: This parasagittal section plane transects portions of developing branchial arches and provides longitudinal view of branchial/pharyngeal mucosa. C: Parasagittal plane of section provides view of kidney showing tubules. Second horizontal array (D–F) was reacted for proliferating cell nuclear antigen and yellow-brown stain marks proliferating cells of structures described in A–C above. Bottom array (G–I) shows location of cells undergoing programmed cell death (apoptosis — TUNEL reaction). Vertical arrays (A, D, G), (B, E, H) and (C, F, I) are serial sections of identical regions reacted as described above. Positive (yellow-brown) PCNA and TUNEL signals are indicated by arrowheads and full arrows respectively. Scale bar = (A, D, G) 160-, (B, E, H) 108- and (C, F, I) 100 μm.
per cm 2 and groups of control/treated embryos can be incorporated into the same chip. Once embedded as routine paraffin block, under uniform orientation, every cutting at the chip provides histological section of multiple individuals at almost identical sectional plane showing comparable organs and structures. The higher resolution and maintenance of desired orientation in all individuals have made
localization and quantification of molecular and cellular events feasible and effortless. The embryo chip facilitates parallel histological analyses and gene/protein expression profiling within the same individual. Paraffin sections from the chip are presumably compatible with the entire array of routine and specialized histological techniques (H&E for
Fig. 4. Embryo 5 μm paraffin sections reacted for PCNA immunohistochemistry (A–C) and TUNEL assay (D–F) are from normoxia (A & D) or hypoxia (B & E) groups. Positive PCNA and TUNEL signals (brown patches) are denoted (arrowheads and full arrows) respectively. Note that there were less PCNA-positive signals in hypoxia-exposed embryo. C: Negative control for PCNA immunohistochemistry. Primary antibody was excluded from staining procedure. F: Negative control for TUNEL assay. Terminal deoxynucleotidyl transferase was omitted in the assay. Scale bar = 300 μm.
N.K.M. Cheung et al. / Comparative Biochemistry and Physiology, Part C 155 (2012) 18–25
23
Fig. 5. Paraffin section (5 μm thickness) of embryo chip illustrating the morphology of liver in embryos from normoxia (A, C) and hypoxia (B, D) groups. Sections were reacted for PCNA immunohistochemistry (A, B) or for TUNEL assay (C, D). The abundance of positive cells (arrowheads) in A versus B corresponds with less growth (suggested by relative abundance in identically enlarged fields). The relative abundance of cells positive for apoptosis (full arrows) in C & D suggests that controlled cell death (apoptosis) was not responsible for different growth status in hypoxic versus normoxic individuals. Scale bar = 30 μm.
Fig. 6. Stereological quantification — estimated volume density of proliferating and apoptotic nuclei in individual embryos (upper panel) and of liver only (lower panel) that was positively labeled by PCNA immunohistochemistry (left panel) or TUNEL assay (right panel) respectively. Black columns represent embryos raised under normoxic vs. those raised under hypoxic (gray columns) conditions. See methods for stereological analysis. * = statistically significant difference between exposure conditions, p b 0.05; n.s. = not significant.
24
N.K.M. Cheung et al. / Comparative Biochemistry and Physiology, Part C 155 (2012) 18–25
general histology; Sudan stain as lysochrome; immunohistochemistry for protein localization; in situ hybridization labels nuclei acid; to name but a few) for formaldehyde-fixed, paraffin-embedded (FFPE) samples, since the chip is basically an agarose matrix encapsulating FFPE embryos. With the exploit of these histological techniques, each of the ca. 200 sections (5 μm in thickness) from a single chip can be stained for a horizon of histological and molecular endpoints. As demonstrated by one part of our hypoxia study, we employed parallel H&E staining, PCNA immunohistochemistry and TUNEL assay to survey gross histological and biometry alterations, protein expression (PCNA, a.k.a. processivity factor for DNA polymerase δ during S-phase of cell cycle) and nuclei acid modification (chromosomal DNA nicking that is intrinsic to apoptosis), respectively. This parallel detection has enabled the integration between biometric changes (reduction of HSI) and underlying cellular event alterations (suppressed proliferation in excess to basal apoptosis). As proof of principle, we have also successfully applied fluorescent in situ hybridization (FISH) on the embryo chip sections to label actb mRNA (data not shown) manifesting the practicability of gene expression analysis. This multiplex nature renders the embryo chip a powerful tool for toxicological studies involving spectrum of interested cellular and molecular responses. We anticipate our embryo chip can be used in combination with the whole adult medaka histoarray (Kong et al., 2008) for parallel histology and molecular profiling in mechanistic toxicological studies revealing cross-generation (from exposed adult to progeny embryo) or life-long (from exposed embryo to adulthood) effects. We appreciate recent endeavor, despite confined to whole-organism level, to achieve high-throughput toxicity testing by harnessing the small size of medaka embryos. Oxendine et al. (2006) have exploited the arrangement of single embryo per well in a 96-well microtiter plate. Their design facilitates the exposure of individual embryos to toxic compounds within their respective wells and allows continuous monitoring on the gross developmental process of the same embryo, attaining rapid and cost-effective developmental toxicity screening. Yet, miniature teleost embryos underpin practical challenges at suborganismal studies since isolation of organs and tissues is arduous, if not impossible. In theory, this hurdle can be circumvented by aforesaid histological techniques performing on multiple embryo tissue sections. Coupled with stereological tools, Kong et al. (2008) remarkably demonstrated the practical use of these methods on whole adult medaka section for high resolution, quantitative molecular and histological profiling in multiple organs and tissues without involving any isolation procedure. Despite being promising, the use of histological techniques on teleost embryos has an elusive prerequisite of well-defined orientation. As described earlier on, Tsao-Wu et al. (1998) pioneered the use of agarose matrix to maintain the orientation of and concurrently section multiple young zebrafish larvae. Coincidentally, a similar design is being promoted by Megason (2009 and 2010) as “MegaMounts” for live imaging of developing embryos at designated orientation. Although both approaches can satisfactorily maintain the orientation at desired position, their success relies on the accessibility to high-end milling facilities crafting finely detailed molds to generate agarose matrixes that affix the embryos properly. In addition, dechorionation appears to be obligatory in using these designs since the orientation of sphere-like chorionated embryos is unlikely to be maintained with ease. And, the impermeability of chorion to embedding media complicates routine embedding, and thus sectioning, procedures. Despite the existence of standard mechanical and enzymatic dechorionation procedures for medaka (Villalobos et al., 2000; Kinoshita et al., 2009; Padilla et al., 2009), they are often time consuming and may be destructive to early stage embryos presenting an ill-defined additional factor that must be considered in determining biological responses that were due to specific test compound(s) or experimental manipulation or both.
Unlike the above designs, producing the agarose matrix used in our chip does not require pre-crafted mold. The agarose wells, punched with a glass Pasteur pipette, have comparable size (diameter ~ 1.5 mm) with medaka's yolk sac, thus chorionated embryos and freshly hatched eleutheroembryos with yolk sac attached can be locked-in perfectly. Auto-rotation of spherical chorionated embryo inside the well is restricted due to friction between chorion and the agarose wall; unless gentle force is applied then the embryo can be oriented in any desired position. Devastating dechorionation is also expendable in our chip as direct puncture of the chorion makes satisfactory penetration of embedding medium, hence serial paraffin sections can be obtained for multiple molecular–histopathological analyses. We do acknowledge our mold-free embryo chip has its limitations. Each chip is limited to single use only. In addition, the chip is prepared manually, and interchip variation must be avoided as much as possible. Imprecision in spacing of embryos and additional labor cost are expected. Construction time also varies depending on the scale of the chip. In our hands, constructing a chip with 100 embryos consumes about 30 min. Notwithstanding these limitations, we regard this embryo chip as a highly efficient, yet easily adaptable and cost-effective, system-wide approach to detect biological phenomena in embryos. 4.2. Restricted hepatic growth under hypoxia This study is, to the best of our knowledge, the first report on cell turnover and biometry in teleost embryonic livers. As discussed above, due to the microscopic nature of the rudimentary livers, these measurements would be inconceivable without the embryo chip. With the aid of stereological tools, we evidenced the estimated HSI was reduced through hypoxia exposure. Since HSI is a normalized measure of the liver size (i.e. per unit body size), the reduction of HSI under hypoxic condition manifested liver growth was restricted. Alarmingly, the restricted liver mass was probably diminishing, as indicated by PCNA:TUNEL b 1, at the time of hatching. These indicated that the embryonic liver was severely spared under hypoxia. Since the liver provides a vast array of vital functions, such as synthesis and secretion of hormones, uptake and metabolism of nutrients, and biotransformation and excretion of exogenous/endogenous toxic substances (Hinton et al., 2001), perturbation to normal growth and development of liver might compromise normal homeostasis and biotransformation capability, hence hampering fitness of the survived individuals. Much work has been done to certify whether the livers of the exposed and post-recovery individuals were functionally compromised, but this is beyond the scope of this paper. 5. Summary and perspective We now have, with the embryo chips and the adult medaka histoarray, a means to achieve multi-dimensional assessment of adverse effects of environmental factors and toxicants. This approach will take us beyond the classical paradigm of restriction to single target organs to one of a system-wide detection and characterization of phenomena (syndromes) affecting the organism. Because adverse effects on the individual may arise at any point in the life span and may be the result of prior exposure and concurrent challenge; we must consider superimposition of environmental and chemical factors. With the advances herein, we have the basis for effective risk communication to decision makers, regulatory authorities and the public. Acknowledgment The work described in this paper was supported by grants from the Research Grants Council of the Hong Kong Special Administrative Region, China (project no. 9041468; CityU 160009) and the University Grants Committee, Area of Excellence Grant (AoE/P-04/04).
N.K.M. Cheung et al. / Comparative Biochemistry and Physiology, Part C 155 (2012) 18–25
References Bozinovic, G., Sit, T., Hinton, D., Oleksiak, M., 2011. Gene expression throughout a vertebrate's embryogenesis. BMC Genomics 12, 132. EC, 1986. Directive 86/609/EEC of 24 November 1986 on the approximation of laws, regulations and administrative provisions of the Member States regarding the protection of animals used for experimental and other scientific purposes. Official J. Eur. Commun. L358, 1–29. Embry, M.R., Belanger, S.E., Braunbeck, T.A., Galay-Burgos, M., Halder, M., Hinton, D.E., Leonard, M.A., Lillicrap, A., Norberg-King, T., Whale, G., 2010. The fish embryo toxicity test as an animal alternative method in hazard and risk assessment and scientific research. Aquat. Toxicol. 97, 79–87. Hardman, R., Kullman, S.W., Hinton, D.E., 2007. Application of in vivo methodologies to investigation of biological structure, function and xenobiotic response in seethrough medaka (Oryzias latipes). Fish Biol. J. MEDAKA 11, 43–65. Hardman, R., Kullman, S.W., Yuen, B., Hinton, D.E., 2008. Non invasive high resolution in vivo imaging of alpha-naphthisothiocyanate (ANIT) induced biliary toxicity in STII medaka (Oryzias latipes). Aquat. Toxicol. 86, 20–37. Hinton, D.E., Segner, H., Braunbeck, T., 2001. Chapter 4. Toxic responses of the liver. In: Schlenk, D., Benson, W.H. (Eds.), Target Organ Toxicity in Marine and Freshwater Teleosts. Volume 1 Organs. Taylor and Francis, London, pp. 224–268. Hinton, D.E., Hardman, R.C., Kullman, S.W., Law, J.M., Schmale, M.C., Walter, R.B., Winn, R.N., Yoder, J.A., 2009. Aquatic animal models of human disease: selected papers and recommendations from the 4th conference. Comp. Biochem. Physiol. C Toxicol. Pharmacol. 149, 121–128. Howarth, D.L., Law, S.H.W., Law, J.M., Mondon, J.A., Kullman, S.W., Hinton, D.E., 2010. Exposure to the synthetic FXR agonist GW4064 causes alterations in gene expression and sublethal hepatotoxicity in eleutheroembryo medaka (Oryzias latipes). Toxicol. Appl. Pharmacol. 243, 111–121. Iwamatsu, T., 2004. Stages of normal development in the medaka Oryzias latipes. Mech. Dev. 121, 605–618. Kimmel, C.B., Ballard, W.W., Kimmel, S.R., Ullmann, B., Schilling, T.F., 1995. Stages of embryonic development of the zebrafish. Am. J. Anat. 203, 253–310. Kinoshita, M., Murata, K., Naruse, K., Tanaka, M., 2009. Medaka: biology, management and experimental protocols, First ed. Wiley–Blackwell Ames, Iowa. 419. Kirchen, R.V., West, W.R., 1976. The Japanese medaka. Its care and development. Carolina Biological Supply, Burlington, NC. Kong, R.Y.C., Giesy, J.P., Wu, R.S.S., Chen, E.X.H., Chiang, M.W.L., Lim, P.L., Yuen, B.B.H., Yip, B.W.P., Mok, H.O.L., Au, D.W.T., 2008. Development of a marine fish model for studying in vivo molecular responses in ecotoxicology. Aquat. Toxicol. 86, 131–141. Lammer, E., Carr, G.J., Wendler, K., Rawlings, J.M., Belanger, S.E., Braunbeck, T., 2009. Is the fish embryo toxicity test (FET) with the zebrafish (Danio rerio) a potential
25
alternative for the fish acute toxicity test? Comp. Biochem. Physiol. C Toxicol. Pharmacol. 149, 196–209. Megason, S.G., 2009. In toto imaging of embryogenesis with confocal time-lapse microscopy. Methods Mol. Biol. 546, 317–332. Megason, S.G., 2010. Protocol: Mounting embryos in megamounts. https://wiki.med. harvard.edu/SysBio/Megason/MegaMounts2010Last access: 08-Feb-2011. Oxendine, S.L., Cowden, J., Hinton, D.E., Padilla, S., 2006. Adapting the medaka embryo assay to a high-throughput approach for developmental toxicity testing. Neurotoxicology 27, 840–845. Padilla, S., Cowden, J., Hinton, D.E., Johnson, R., Flynn, K., Hardman, R.C., Yuen, B., Law, S., Kullman, S.W., Au, D.W.T., 2009. Use of Medaka in Toxicity Testing. Current Protocols in Toxicology. John Wiley and Sons, New York, pp. 1–36. Park, J.W., Tompsett, A., Zhang, X.W., Newsted, J.L., Jones, P.D., Au, D., Kong, R., Wu, R.S.S., Giesy, J.P., Hecker, M., 2008. Fluorescence in situ hybridization techniques (FISH) to detect changes in CYP19a gene expression of Japanese medaka (Oryzias latipes). Toxicol. Appl. Pharmacol. 232, 226–235. Park, J.W., Tompsett, A.R., Zhang, X.W., Newsted, J.L., Jones, P.D., Au, D.W.T., Kong, R., Wu, R.S.S., Giesy, J.P., Hecker, M., 2009. Advanced fluorescence in situ hybridization to localize and quantify gene expression in Japanese medaka (Oryzias latipes) exposed to endocrine-disrupting compounds. Environ. Toxicol. Chem. 28, 1951–1962. Reed, M.G., Howard, C.V., 1998. Surface-weighted star volume: concept and estimation. J. Microsc. 190, 350–356. Schmale, M.C., Nairn, R.S., Winn, R.N., 2007. Aquatic animal models of human disease. Comp. Biochem. Physiol. C Toxicol. Pharmacol. 145, 1–4. Smith, L.E., Carvan Iii, M.J., Dellinger, J.A., Ghorai, J.K., White, D.B., Williams, F.E., Weber, D.N., 2009. Developmental selenomethionine and methylmercury exposures affect zebrafish learning. Neurotoxicol. Teratol. 32, 246–255. Tompsett, A.R., Park, J.W., Zhang, X., Jones, P.D., Newsted, J.L., Au, D.W.T., Chen, E.X.H., Yu, R., Wu, R.S.S., Kong, R.Y.C., Giesy, J.P., Hecker, M., 2009. In situ hybridization to detect spatial gene expression in medaka. Ecotoxicol. Environ. Saf. 72, 1257–1264. Tsao-Wu, G.S., Weber, C.H., Budgeon, L.R., Cheng, K.C., 1998. Agarose-embedded tissue arrays for histologic and genetic analysis. Biotechniques 25, 614–618. Villalobos, S.A., Hamm, J.T., Teh, S.J., Hinton, D.E., 2000. Thiobencarb-induced embryotoxicity in medaka (Oryzias latipes): stage-specific toxicity and the protective role of chorion. Aquat. Toxicol. 48, 309–326. Weibel, E.R., 1979. Stereological Methods. Academic, NY, USA. Wittbrodt, J., Shima, A., Schartl, M., 2002. Medaka — a model organism from the Far East. Nat. Rev. Genet. 3, 53–64. Yu, R.M., Chen, E.X., Kong, R.Y., Ng, P.K., Mok, H.O., Au, D.W., 2006. Hypoxia induces telomerase reverse transcriptase (TERT) gene expression in non-tumor fish tissues in vivo: the marine medaka (Oryzias melastigma) model. BMC Mol. Biol. 7, 27.
