ToxSci Advance Access originally published online on September 6, 2007
Toxicological Sciences 2007 100(2):486-494; doi:10.1093/toxsci/kfm235
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Aristolochic Acid Induces Heart Failure in Zebrafish Embryos That is Mediated by Inflammation
* Institute of Cellular and Organismic Biology, Academia Sinica, Taipei 115, Taiwan
Institute of Bioscience and Biotechnology, National Taiwan Ocean University, Keelung 202, Taiwan
1 To whom correspondence should be addressed at Institute of Cellular and Organismic Biology, Academia Sinica, 128 Academia Road, Section 2, Nankang, Taipei 115, Taiwan. Fax: 886-2-2789-9503. E-mail: huangcc{at}gate.sinica.edu.tw.
Correspondence may also be addressed to John Yu. E-mail: johnyu{at}gate.sinica.edu.tw.
Received July 9, 2007; accepted August 29, 2007
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUPPLEMENTARY DATAFUNDINGREFERENCES |
|---|
Aristolochic Acid (AA) is a component of Chinese herbs that has been found to be toxic to multiple organs in adults. Its toxicity to developing embryos has not been reported. Here, we describe that AA specifically causes heart defects in developing zebrafish embryos in a dosage-dependent manner. The treated embryos are able to develop their hearts normally up to 24 h postfertilization, when cardiac contraction initiates, but begin to show deformation and reduction of the hearts followed by gradual contractility loss and eventually lethality, suggesting that AA is primarily affecting cardiac physiology rather than cardiogenesis. Histological analyses reveal that the AA-treated hearts develop hypertrophy and disorganization of cardiomyocytes and loss of endocardium. By transmission electron microscopy, we observed broken and disorganized cardiac fibers in the AA-treated hearts. AA induces the expression of proinflammation genes, including cox-2, IL-1ß, and others. The AA-induced cardiac defects can be attenuated by the cox-2 antagonist NS398 via reducing the expression of the inflammatory genes. This attenuation could be further enhanced by known heart failure drugs, such as angiotensin-converting enzyme inhibitor and ß-adrenergic receptor antagonist. In contrast, the heart defects are enhanced by a ß-adrenergic receptor agonist. In summary, AA causes profound toxicity to zebrafish embryos that exhibit pathophysiological and pharmacological features resembling those of heart failure in humans and other model organisms, and thus, zebrafish could be a new model for studies on heart failure.
Key Words: aristolochic acid; zebrafish; heart failure; inflammation; cox-2.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUPPLEMENTARY DATAFUNDINGREFERENCES |
|---|
Aristolochic acid (AA) is a natural product found in Chinese herbs and known to cause nephropathy in humans and other experimental organisms (Vanhaelen et al., 1994
; Vanherweghem et al., 1993). A number of studies revealed its toxicity on multiple organs, including stomach, intestine, and others (Cosyns et al., 1998; Mengs, 1987), suggesting that AA imposes toxicity to diverse, likely more, organs and its mechanism might be far more complicated than our current understanding from the studies in kidney. It is not known whether AA also causes toxicity in developing embryos.
During development, the heart is the one of the first organs to function. While great effort has been put in recent years to uncovering the genetic mechanisms of heart development using model organisms, including Drosophila, zebrafish, and mouse, in hopes of combating inherited cardiac diseases, studies on noninherited heart diseases during and beyond cardiogenesis are limited. Even though a good number of studies have shown genetic conservation of heart development in zebrafish (Stainier, 2001), our understanding of the similarity in cardiac physiology between fish and humans is just emerging (Baker et al., 1997; Bendig et al., 2006). To gain the pharmacological perspective of the zebrafish cardiology studies and build a better link to human health, it is important to compare the regulation of cardiac function and/or diseases involving other physiological systems, e.g., the immune system, in fish and humans.
One of the common cardiac diseases is heart failure, which could result from injury, infection or toxin-induced damage on the myocardium, and genetic disorders (Dyer and Fifer, 2003). By simple definition, heart failure is the progressive decrease in cardiac output. Heart failure is also found to associate with other cardiovascular or hemodynamic diseases, such as hypertension, and renal defects. More recently, inflammation was found to contribute to the progression of heart failure as well with murine models (Delgado et al., 2004; Höcherl et al., 2002; Wong et al., 1998) and thus could be another therapeutic target for heart failure (Lisman et al., 2002; Willerson et al., 1998). While these models might provide a tool for studying the complicated interaction between heart failure and other physiological systems, their use in understanding the genetic and pharmacological mechanisms of heart failure is limited due to the long reproduction cycle, high cost of animal husbandry, and operational difficulties. It is therefore important to develop another tractable animal model for basic research and pharmacological tests for heart failure and/or other cardiac diseases.
Here, we report that AA induces specific cardiac defects beyond cardiogenesis in zebrafish. Our phenotypic characterizations show that the pathological development in the AA-treated hearts resembles that described for the heart failure in humans and other experimental model animals. We provide evidence that the progression of AA-induced cardiac defects is mediated by inflammation as proinflammation genes are induced by AA but an anti-inflammation drug can suppress the induction and, more important, attenuates the cardiac defects. We further demonstrate that doxorubicin, a drug that is known to cause cardiac cytotoxicity in humans and mouse, also induces inflammation-mediated heart defect in zebrafish. Finally, the AA-treated zebrafish respond appropriately to drugs that are related to human heart failure. These results demonstrate that the zebrafish embryos could be a model for studying the pathological and pharmacological mechanisms of heart failure and/or other noninherited heart diseases.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUPPLEMENTARY DATAFUNDINGREFERENCES |
|---|
Zebrafish husbandry and breeding.
The zebrafish stocks used in this study are maintained following standard procedures (Westerfield, 2000
) and bred by in vitro fertilization. In brief, mature male and female zebrafish were set up in a breeding tank separated by a mesh screen the night before breeding. Soon after the light is turned on the next morning, fish were anesthetized in 
0.16% tricaine for 1–2 min. The female fish was placed on a paper towel to dry the body surface briefly before being transferred into a 6-cm dish. The eggs were expelled onto the dish by gently pressing the lateral and ventral side of the belly with fingers. The male fish were also dried briefly and then put ventral side up on a sponge well. Under a dissecting microscope, the lateral sides of the male belly were milked with a pair of blunt forceps while holding a capillary tube with the other hand close to the anus to collect sperm. Immediately, sperm were mixed with the eggs, and 1 ml of egg water was added (more details in Westerfield, 2000).
Chemical treatment on zebrafish embryos.
The zebrafish embryos of 6-h postfertilization (hpf) stage were collected and arrayed into 96-well plates in 200 µl of egg water (distilled water containing 60 µg/ml sea salt from Coralife, Carson, CA) which, depending on the experimental design, were replaced with the same volume of egg water containing desired concentration of chemicals at different times of zebrafish development (Peterson et al., 2000). Chemical stocks are described in online supplementary data. After treatment, embryos were collected into 6-cm petri dish and dechorionated before fixation or RNA extraction.
Immunohistochemistry and in situ hybridization.
Embryos were first fixed in 4% paraformaldehyde overnight in 4°C for immunohistochemistry and in situ hybridization (see online supplementary data for details), and whole-mount in situ hybridization was done with the InsituPro VS robot by Intavis Bioanalytical Instruments AG Inc. (Koln, Germany) following the standard manufacturer procedure. Apoptotic cells were detected with the TUNEL kit (Roche Diagnostic, Mannheim, Switzerland).
Transmission electron microscopy.
Embryos were fixed in 2% glutaraldehyde, 2% paraformaldehyde, and 0.1M sodium phosphate pH 7.4 overnight in 4°C followed by standard infiltration, embedding, and sectioning procedures (see online supplementary data for details). The sections were analyzed under the Hitachi H-7000 transmission electron microscope.
Quantitative PCR.
Wild-type embryos were treated with chemicals and then dechorionated manually at the end of treatment. Fifty embryos from each treatment were subjected to RNA extraction with Trizol following commercial instruction. Around 5 µg of total RNA were then used to generate cDNAs with Superscript II (Invitrogen, Carlsbad, CA). The cDNA then was used for quantitative PCR with Power Cybergreen labeling kit, and the PCR was performed with ABI7000 thermocycler (both from Applied Biosystems, Foster City, CA). The data were analyzed with ABI7000 software provided by the manufacturer.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUPPLEMENTARY DATAFUNDINGREFERENCES |
|---|
AA Causes Cardiac Defects in Zebrafish Embryos
AA has been found to cause toxicity in multiple organs in humans and mouse (Cosyns et al., 1998
; Mengs, 1987; Vanhaelen et al., 1994; Vanherweghem et al., 1993). In this project, we utilize zebrafish to study whether or not AA causes toxicity to developing embryos and its toxicological mechanisms. In the pilot experiments, the embryos were treated with 10 or 20µM of AA starting at 6 hpf and examined daily. While there were no apparent changes in morphological phenotypes in most tissues at this time (Fig. 1B), we had observed cardiac phenotypes, which appeared as early as 48 hpf embryonic stage. These changes include the distortion of heart shape and decrease of heart size (Figs. 1C vs. 1D) and gradual decrease of heart rate (see Supplementary Movies 1 and 2), and within a few days the circulation ceased and the animals died (not shown). To gain a better understanding of the AA toxicity on developing hearts, we measured the heart size of control and treated embryos. In this experiment, we labeled the zebrafish hearts with the monoclonal antibody MF20 which recognizes sarcomeres (Bader et al., 1982) and allows us to visualize the heart more clearly (Fig. 1E and Supplementary Figs. S1C and D). We measured the distance between the sinus venosus and the bulbus arteriosus (SV-BA) (Fig. 1E, also Lin et al., 2007) as well as the diameter of atrium-ventricle (AV) valve (Supplementary Figs. S1E and F) of 48 and 70 hpf embryos treated with different concentrations of AA. The results showed that the SV-BA distance decreases significantly in the 20µM-AA (p < 0.01), but not in the 10 (p = 0.10) or 5µM (p = 0.39), treated 48 hpf embryos (Fig. 1F, black bars). At 70 hpf, however, the treated embryos showed dosage-dependent decreases of the heart size (Fig. 1F, white bars, p = 0.3 for 5µM, p = 0.001 for 10µM, and p < 0.0001 for 20µM group). AA-treated embryos also showed decrease in AV diameter (Supplementary Fig. S1H). In addition, we found that the skeletal muscle length which is also marked by MF20 (Supplementary Figs. S1A and B) decreased in most of the 70 hpf but not the 48 hpf AA-treated embryos (Supplementary Fig. S1G), suggesting that AA cardiac toxicity might secondarily cause growth retardation and apoptosis (below) after 48 dpf. These findings suggest that the developing heart is the primary target of AA during zebrafish embryogenesis.
|
10. (G) The kinetics of AA toxicity is concentration dependent. Each data point is based on 40 embryos.In order to understand the kinetics of AA toxicity, experiments with various concentrations of AA and time course examinations of the cardiac phenotype were carried out. The results show that at a given time, 48 hpf for example, the incidence of cardiac phenotype is dependent on AA dosage, 71% with 20µM; 34%, 10µM;
0%, 5µM (Fig. 1G, n = 30), and the EC50 is approximately 15µM. As the treatment prolonged, both the incidence and severity increased. At 77 hpf, while 5µM AA induced the identical cardiac defects in 100% embryos (Fig. 1G), those in 10 and 20µM groups died already. Thus, the kinetic of AA toxicity toward the developing heart is dosage dependent.
Latency of AA-Induced Cardiac Toxicity
To determine which morphogenetic events during heart development might be affected by AA, we treated the embryos at various stages with 10µM of AA for various periods of incubation time. First, when the embryos were treated with AA at early stages (before hearts began to contract), i.e., from 6 to 12 or from 6 to 24 hpf, and examined after 72 hpf, it was showed that the treated hearts developed with grossly normal morphology beyond the initiation of contraction (not shown) but continued to develop the small and deformed heart phenotypes later (Fig. 2A). When AA treatment begins at 24, 48, or even as late as 72 hpf when cardiogenesis is complete (Stainier, 2001), the treated hearts also develop the same phenotypes (Fig. 2A). These results suggest that AA poses toxicity latency during and beyond cardiogenesis. To explore when during the early development AA begins its toxicity effect, we treated 6 or 14 hpf embryos with 10µM AA for 2, 4, 6, or 8 h, washed out the chemical, and examined their heart development at later times. The results show that the 4-, 6-, and 8-h–treated but not the 2-h–treated 6 hpf embryos develop the small and deformed heart phenotypes (Fig. 2B). Therefore, AA toxicity could begin as early as 10 hpf (4-h treatment on 6 hpf embryos) during the development of cardiac precursor cells.
|
Abnormal Distribution of Cardiac Sarcomeres in AA-Treated Hearts
To better understand the cardiac defects, we next examined the AA-treated hearts with immunostaining and in situ hybridization. First, we used the MF20 monoclonal antibody to examine the distribution of sarcomeres in cardiac muscles (Bader et al., 1982
). In control embryos, MF20 stained uniformly in the heart of control embryos suggesting the even distribution of cardiac sarcomeres (Fig. 3A). In contrast, the MF20 staining in AA-treated embryos showed a more intense pattern and highlighted the small size of the heart (Fig. 3B), suggesting an aggregation of the cardiac sarcomeres. The immunostaining of skeletal sarcomeres in AA-treated embryos appears normal (Supplementary Fig. S1). On the other hand, whole-mount in situ hybridization showed that the expression level of myosin heavy chain, which is expressed in both ventricle and atrium (Supplementary Figs. S1I and J), and cardiomyosin light chain, which is expressed in the ventricle (Supplementary Figs. S1K and L), are indistinguishable between AA-treated and control embryos, indicating that the AA toxicity does not alter the expression level of the genes of major cardiac fiber components.
|
Myocardium Abnormalities and Endocardium Loss in AA-Treated Hearts
To explore the causes to the small and deformed heart phenotypes caused by AA, we tested the following hypotheses: could it be due to (1) excess apoptosis, (2) loss of cardiac cell/tissue layer, or (3) disorganization of cardiac cells? First, TUNEL apoptosis assay showed indistinguishable level of apoptotic cells in the regions for DMSO (Figs. 3C and 3D) and the AA-treated hearts with mild (Figs. 3E and 3F) or severe (Figs. 3G and 3H) phenotypes. These findings suggest that the small heart phenotype is probably not due to excess cell death. High level of apoptotic cell death, however, was observed in AA-treated embryos starting from the tail (white arrows in Figs. 3E and 3G) and spreading toward the anterior part of the animal (Fig. 3G), presumably due to the gradual loss of circulation as the cardiac defects worsened.
Next, the cellular organization of the AA-treated hearts was examined in detail. In 48 hpf embryos of control, the zebrafish heart was found to be composed of an outermost one-cell layer of cardiomyocytes followed by the cardiac jelly and the innermost thin endocardium layer (Fig. 3I) (Stainier, 2001). In AA-treated embryos, the cardiomyocytes were often stacked up to 2–3 layers and the endocardium was discontinuous (Fig. 3J) or absent in severe embryos (Fig. 3K). Closer examination revealed the enlargement of the cardiomyocytes in AA- treated zebrafish hearts (insets in Figs. 3J and 3K).
To confirm the endocardium loss in AA-treated hearts, we utilized the TG(fli1:enhanced green fluorescent protein (EGFP)y1 embryos which express EGFP in the vascular endothelial and endocardial cells (Lawson and Weinstein, 2002). In control embryos, endocardial cells reside at the innermost lining of the heart (Fig. 3L, also see Fig. 3A). In AA-treated embryos, the number of green fluorescent protein (GFP) cells inside the heart is dramatically reduced (Fig. 3M, also see Fig. 3B). At this point, the hearts were still able to contract, albeit slower, weaker, and unevenly (Supplementary Movie 2). Thus, it is concluded that AA causes distortion of myocardium, endocardium loss, and likely cardiac hypertrophy leading to the small and deformed heart phenotypes.
Breakdown of Cardiac Muscle Fibers in the AA-Treated Hearts
The above intense immunostaining using MF20 suggests two possibilities, (1) there is structural defect in sarcomeres leading to cardiac fiber aggregation or (2) the sarcomere structure is normal and the intense staining is simply due to noncontracting cardiac sarcomeres. To distinguish these two possibilities and also to explore further the causes to the contractility loss in AA-treated hearts, we analyzed the cardiac muscle fibers by transmission electron microscope (TEM). In the control ventricle, the cardiac fibers are distributed widely in the entire ventricular myocardium (Fig. 4A) and the sarcomeres have regular length (Fig. 4E). In AA-treated hearts, however, the ventricular muscle fibers are fewer (Fig. 4B), and the sarcomeres are thinner and have irregular length (Fig. 4F). In the atrium, while the muscle fibers are nicely organized along the apical side of the control myocardium (Figs. 4C and 4G), the atrial muscle fibers in the AA-treated hearts are disrupted (Fig. 4D) and appear severed when observed at high magnification (Fig. 4H).
|
Association between Inflammation and AA Cardiac Cytotoxicity
Growing evidence suggests the relation between inflammatory response and the progression of human heart failure (Gullestad and Aukrust, 2005
). The findings of the AA-induced toxicology led us to postulate that the AA-induced cardiac phenotypes were also mediated by inflammation. The expression of proinflammatory genes was examined with quantitative PCR (QPCR). In order to see the temporal induction, embryos were collected after treatment with AA for 16, 24, or 32 h starting at 24 hpf (Fig. 5). The results showed that 32 h of AA treatment induced more than four- and eightfold increases of cox-2 and IL-1ß expression, respectively (Figs. 5A and 5B), comparing with the control embryos. AA also induces higher level expression of SAA (serum amyloid 
), C/EBPB (CCAAT/enhancer-binding protein B), and C/EBPG genes (Supplementary Fig. S2) at 24 h of treatment. AA, however, does not induce significant expression of complement C3 gene (Supplementary Fig. S2). These results indicate that AA induces inflammatory response in the embryos.
|
We then tested whether the AA-induced inflammation is involved in the cardiac phenotype progression by treating the embryos with AA and the anti-inflammatory drug NS398 (N-[2-(Cyclohexyloxy)-4-nitrophenyl] methanesulfonamide), which is a selective cox-2 inhibitor (Futaki et al., 1994
). We found that NS398 can significantly attenuate the heart phenotype progression (Fig. 5C). In order to see consistent and greater attenuation by NS398, we set up two sets of experiments, one with 10µM (white lines) and the other 5µM (black lines) of AA. For example, while near 100% of the embryos incubated with 10µM AA alone (Fig. 5C, white diamond) already develop the heart phenotype at 56 hpf, those in 10µM NS398 and 10µM AA (Fig. 5C, white squares) show only 20% penetrance of the cardiac defects. N398 alone does not cause any developmental defects in the embryos (not shown). More importantly, the treatment of NS398 inhibits the induction of cox-2, IL-1ß, and SAA by AA (Fig. 5A and 5B and not shown). These results strongly suggest that the AA-induced cardiac defect is mediated by inflammatory response. To rule out the possibility that NS398 might directly bind AA to abolish its toxicity, we treated the embryos with AA first from 6 to 24 hpf and replaced with egg water (Fig. 5C, white and black circles) or NS398 containing water (Fig. 5C, white and black triangles). We found that NS398 can still reduce the heart phenotypes after AA is removed. Finally, we tested whether the AA toxicity is reversible by adding NS398 to the AA-treated embryos that already developed weak or strong heart failure. The results show that NS398 is not able to rescue the AA-treated embryos once the phenotypes have become detectable (not shown).
Inflammatory System During Early Embryogenesis
Above we showed that AA can induce significant expression of proinflammation genes within 24 h of treatment in 24 hpf embryos, and we wondered why AA treatment in the early embryos, e.g., 6 hpf does not cause any phenotype before 48 hpf. We then tested the idea that perhaps the inflammatory system is not responding to AA during early development. We measured the expression levels of cox-2, IL-1ß, CEBP/B, CEBP/G, and SAA in the embryos that are treated with AA starting from 6 hpf for 15, 26, or 48 h. The results showed that in AA-treated embryos (Supplementary Fig. S3, white bars), the expression of these genes remained the same as in control embryos (black bars) up to 26 h of treatment which is longer than the time needed for AA to induce the expression of these genes in 24 hpf embryos (Figs. 5A and 5B). The expression levels of the proinflammation genes were then significantly increased after 48 h of AA treatment on the 6 hpf embryos (Supplementary Fig. S3). These results indicate that the inflammatory system is not induced by AA in the early embryos even though the proinflammation genes are being expressed.
Heart Defects Induced by Doxorubicin in Zebrafish Embryos
Doxorubicin is a widely used anticancer drug with a side effect of heart damage. In mice, doxorubicin causes heart damage and induces heart failure (van Acker et al., 1996) which is later found to be mediated by inflammation (Delgado et al., 2004). We asked whether zebrafish heart will respond to doxorubicin in a similar way as in mouse. Upon doxorubicin treatment, the zebrafish hearts develop edema in the pericardiac chamber followed by cardiac distortion (Fig. 5D), eventually leading to contractility loss, severe body edema, and lethality (not shown). More importantly, our QPCR results show that the expression of cox-2 and IL-1ß is elevated by doxorubicin in the treated embryos (Supplementary Fig. S4, gray bars). These results demonstrate that doxorubicin induces similar cardiac defects in zebrafish as in mammals. In addition, both AA and doxorubicin-induced cardiac defects involve inflammation.
Pharmacological Response of the AA-Treated Hearts
Due to the described resemblance in phenotypes and physiology between AA-induced cardiac defects and human heart failure disease, we sought for evidence of the similarities in pharmacological response between the AA-induced cardiac defects and the human heart failure. We wondered whether or not the AA-treated hearts will respond to the heart failure–related drugs. Current medications for heart failure target hemodynamic or cardiac contractility, such as angiotensin-converting enzyme inhibitor (ACE-I), angiotensin receptor blocker, ß-adrenergic receptor antagonist (ß-blocker), and others. Here, we report the tests with the following compounds: NS398 for anti-inflammation, ACE-I, metoprolol (a ß-adrenergic receptor antagonist), and isopreternol (a ß-adrenergic receptor agonist). In order to see the subtle difference of drug effect, we divide the cardiac severity into two categories, the severe phenotype where the ventricle contraction stopped completely (black portion of each bar in Fig. 6) and weak phenotype where the hearts appeared small and distorted but both ventricle and atrium are still contracting (white portion of each bar in Fig. 6). As a control, AA alone can induce severe phenotype in > 90% of the treated embryos. First, we tried single treatment and found that the percentage of embryos with severe phenotype is 30% lower by 200µM ß-adrenergic receptor antagonist, metoprolol (Fig. 6A, MET), but not changed by 10µM of the compound. In contrast, the ß-adrenergic receptor agonist isopreterenol was found to enhance the AA-induced heart defect (Fig. 6B). The ACE-I alone at 50µM can also reduce the AA-induced cardiac defect (not shown) but has no effect at 10µM (Fig. 6C, lane 2). NS398 exhibits the best ability in attenuating AA toxicity among the tested compounds by reducing severe phenotype down to 60% (Fig. 6C, lane 4). Interestingly, while double combination of 10µM ACE-I and NS398 showed slightly lower percentage of embryos with severe phenotype (Fig. 6C, lane 5), combination of 10µM MET, 10µM ACE-I, and NS398 show synergistic attenuation by reducing the severe phenotype down to less than 40% of the treated embryos (Fig. 6C, lane 6). These results indicate that the AA-treated zebrafish embryos adequately respond to heart failure–related compounds.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUPPLEMENTARY DATAFUNDINGREFERENCES |
|---|
Although AA is known to cause nephropathy in adult humans (Vanhaelen et al., 1994
; Vanherweghem et al., 1993), our results show that AA might have specific toxicity toward the developing hearts in zebrafish. Our attempt to test whether AA can cause similar nephropathy in zebrafish by treating the developing embryos with 10 or 20µM of AA at 24, 48, or, 72 hpf when zebrafish kidney has begun to function failed to detect any cyst formation (Drummond et al., 1998). Immunostaining using the Na+/K+ ATPase monoclonal antibody 6F also showed normal morphology of AA-treated pronephric ducts (data not shown). One possibility is that the heart phenotypes develop much earlier than the kidney in zebrafish so that there is little opportunity to observe any obvious nephropathic changes. Alternatively, the discrepancy could be explained by the fundamental difference between zebrafish embryos and mammals in the uptake of small molecules. AA could enter the zebrafish embryos by passive diffusion and be able to reach every cell at high concentration directly. In particular, zebrafish hearts might be accessible to AA from outside due to the fact that they are only protected by a thin pericardiac chamber which is permeable easily for small molecules like AA. The AA toxicity toward developing zebrafish embryos seems to specifically target the physiological and structural maintenance of the heart after cardiac contraction is initiated. First, the AA-treated embryos are able to develop a normal contracting heart initially. Second, the AA treatment beyond 72 hpf can also induce the identical cardiac failure. Third, we found that while the heart size in control embryos (presented by the SV-BA distance) shows little growth between 48 and 70 hpf, AA causes decreases of the heart size in a dosage- and time-dependent manner. Finally, we also showed that the shrinkage of AA-treated hearts is associated with several cellular abnormalities, including sarcomeres aggregation, endocardium loss, and cardiac fiber breakdown but not with apoptosis. All these cellular defects together can well cause loss of cardiac contraction and circulation. Our attempt to distinguish which of these defects might be the primary root for the others by searching for the earliest cellular defects turned out to be inconclusive as the association of them was so tight. Furthermore, AA does not seem to affect the transcription of cardiac myosin genes. Together, our data suggest a cellular mechanism for the AA-induced cardiac failure which results from an inflammation-mediated destruction of cardiac fibers induced by AA after cardiac contraction begins. However, further studies are needed in order to understand the molecular mechanisms of how the cardiac toxicity of AA is initiated.
Our studies reveal that AA can induce an inflammatory response associated with the progression of the cardiac defects in zebrafish embryos. A number of proinflammation genes are increasingly induced along with the progression of the cardiac phenotype caused by AA, including cox-2, IL-1ß, C/EBPB, C/EBPG, and SAA, which were postulated to be a predictor of cardiac severity in heart failure patients (Katayama et al., 2006). Note that cox-2 is induced relatively late and thus might not be the primary cause of the heart defects, which is also postulated by other studies (Delgado et al., 2004). In other words, cox-2 might contribute to the heart defect progression, likely via amplifying the inflammatory response, but not the initiation. Interestingly, the inflammatory system is not responsive to AA during early embryogenesis even though the proinflammation genes are expressed, suggesting the AA-induced inflammation might be mediated by temporal or tissue-specific factors.
The AA-induced cardiac defects in zebrafish display significant similarities with human failing hearts. Pathologically, AA-treated hearts show contractility loss and signs of cardiac hypertrophy. Physiologically, we show that the AA-induced heart defects are mediated by inflammation. Both features have been shown or suggested in humans and in murine models (Delgado et al., 2004). Furthermore, we found that doxorubicin can also induce similar heart phenotypes in zebrafish embryo. Finally and more importantly, AA-treated hearts respond to heart failure–related drugs appropriately. However, understanding of the relationships between the physiological systems that are regulated by these drugs will require further studies. Nevertheless, all these results strongly suggest that the cardiophysiology in zebrafish is significantly relevant to mammals and that zebrafish can be a suitable model for cardiophysiology studies.
The zebrafish embryo provides unique advantages for studying the relationship between inflammation, hemodynamics, and heart diseases. First, the transparency of zebrafish embryos offers an easy and quick examination of the heart morphology and some aspects of cardiac physiology. Second, the speed of disease progression can be easily manipulated as we have demonstrated that low concentrations of AA can induce the defect at much slower rate than high concentrations, which allows us to examine the progression of molecular and cellular aspects of the cardiac defect. Third, the small zebrafish embryos make it possible to perform cheap and yet complicated drug experiments, which offer the opportunity to develop novel drug treatments to the disease. We demonstrated that a combination of drugs targeting different aspects of cardiovascular physiology can have synergistic benefits on the AA-induced cardiac defect (Fig. 6). Finally, using zebrafish genetics, we should be able to isolate mutations that are resistant to AA-induced cardiac defects, which will allow us to uncover the genetic factors involved in the progression of the heart disease. We will expect to isolate mutations involved in inflammation and hemodynamics and cardiac remodeling, which might be missed in the traditional morphology-based mutant screening due to the lack of prominent morphological phenotypes in these mutations.
|
SUPPLEMENTARY DATA |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUPPLEMENTARY DATAFUNDINGREFERENCES |
|---|
Supplementary data are available online at http://toxsci.oxfordjournals.org/.
|
FUNDING |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUPPLEMENTARY DATAFUNDINGREFERENCES |
|---|
Distinguished Postdoctoral Fellowship from Academia Sinica (to C.C.H.); Genomic Center of Academia Sinica, Taiwan (94M011-1).
|
ACKNOWLEDGMENTS |
|---|
We thank Dr Ling-Huei Yih for providing the aristolochic acid for initial tests, Dr Sheng-Ping L. Hwang for providing the cardiomyosin light chain and myosin heavy chain probes and some cardiovascular drugs, and Dr Yu-Ching Peng, M.D., for helpful discussions on heart failure disease. We are grateful to Dr Li-Tzu Li and the Core Facilities group of Institute of Cellular and Organismic Biology at Academia Sinica for their technical assistance.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUPPLEMENTARY DATAFUNDINGREFERENCES |
|---|
Bader D, Masaki T, Fischman D. Immunochemical analysis of myosin heavy chain during avian myogenesis in vivo and in vitro. J. Cell Biol. (1982) 95:763–770.
Baker K, Warren K, Yellen G, Fishman M. Defective "pacemaker" current (Ih) in a zebrafish mutant with a slow heart rate. Proc. Natl. Acad. Sci. U.S.A. (1997) 94:4554–4559.
Bendig G, Grimmler M, Huttner I, Wessels G, Dahme T, Just S, Trano N, Katus H, Fishman M, Rottbauer W. Integrin-linked kinase, a novel component of the cardiac mechanical stretch sensor, controls contractility in the zebrafish heart. Genes Dev. (2006) 20:2361–2372.
Cosyns J, Goebbels R, Liberton V, Schmeiser H, Bieler C, Bernard A. Chinese herbs nephropathy-associated slimming regimen induces tumours in the forestomach but no interstitial nephropathy in rats. Arch. Toxicol. (1998) 72:738–743.[CrossRef][Web of Science][Medline]
Delgado R III, Nawar M, Zewail A, Kar B, Vaughn W, Wu K, Aleksic N, Sivasubramanian N, McKay K, Mann D, et al. Cyclooxygenase-2 inhibitor treatment improves left ventricular function and mortality in a murine model of doxorubicin-induced heart failure. Circulation (2004) 109:1428–1433.
Drummond I, Majumdar A, Hentschel H, Elger M, Solnica-Krezel L, Schier A, Neuhauss S, Stemple D, Zwartkruis F, Rangini Z, et al. Early development of the zebrafish pronephros and analysis of mutations affecting pronephric function. Development (1998) 125:4655–4667.[Abstract]
Dyer G, Fifer M. Pathology of Heart Disease. A Collaborative Project of Medical Students and Faculty (2003) Baltimore, MD: Lippincott Williams and Wilkins.
Futaki N, Takahashi S, Yokoyama M, Arai I, Higuchi S, Otomo S. NS-398, a new anti-inflammatory agent, selectively inhibits prostaglandin G/H synthase/cyclooxygenase (COX-2) activity in vitro. Prostaglandins (1994) 47:55–59.[CrossRef][Web of Science][Medline]
Gullestad L, Aukrust P. Review of trials in chronic heart failure showing broad-spectrum anti-inflammatory approaches. Am. J. Cardiol. (2005) 95:17C–23C. discussion 38C–40C.[CrossRef][Web of Science][Medline]
Höcherl K, Dreher F, Kurtz A, Bucher M. Cyclooxygenase-2 inhibition attenuates lipopolysaccharide-induced cardiovascular failure. Hypertension (2002) 40:947–953.
Katayama T, Nakashima H, Takagi C, Honda Y, Suzuki S, Iwasaki Y, Yamamoto T, Yoshioka M, Yano K. Serum amyloid a protein as a predictor of cardiac rupture in acute myocardial infarction patients following primary coronary angioplasty. Circ. J. (2006) 70:530–535.[CrossRef][Web of Science][Medline]
Lawson N, Weinstein B. In vivo imaging of embryonic vascular development using transgenic zebrafish. Dev. Biol. (2002) 248:307–318.[CrossRef][Web of Science][Medline]
Lin C, Hui M, Cheng S. Toxicity and cardiac effects of carbaryl in early developing zebrafish (Danio rerio) embryos. Toxicol. Appl. Pharmacol. (2007) 222:159–168.[CrossRef][Web of Science][Medline]
Lisman K, Stetson S, Koerner M, Farmer J, Torre-Amione G. The role of tumor necrosis factor alpha blockade in the treatment of congestive heart failure. Congest. Heart Fail. (2002) 8:275–279.[Medline]
Mengs U. Acute toxicity of aristolochic acid in rodents. Arch. Toxicol. (1987) 59:328–331.[CrossRef][Web of Science][Medline]
Peterson R, Link B, Dowling J, Schreiber S. Small molecule developmental screens reveal the logic and timing of vertebrate development. Proc. Natl. Acad. Sci. U.S.A. (2000) 97:12965–12969.
Stainier D. Zebrafish genetics and vertebrate heart formation. Nat. Rev. Genet. (2001) 2:39–48.[CrossRef][Web of Science][Medline]
van Acker S, Kramer K, Voest E, Grimbergen J, Zhang J, van der Vijgh W, Bast A. Doxorubicin-induced cardiotoxicity monitored by ECG in freely moving mice. A new model to test potential protectors. Cancer Chemother. Pharmacol. (1996) 38:95–101.[CrossRef][Web of Science][Medline]
Vanhaelen M, Vanhaelen-Fastre R, But P, Vanherweghem J. Identification of aristolochic acid in Chinese herbs. Lancet (1994) 343:174.[Web of Science][Medline]
Vanherweghem J, Depierreux M, Tielemans C, Abramowicz D, Dratwa M, Jadoul M, Richard C, Vandervelde D, Verbeelen D, Vanhaelen-Fastre R. Rapidly progressive interstitial renal fibrosis in young women: Association with slimming regimen including Chinese herbs. Lancet (1993) 341:387–391.[CrossRef][Web of Science][Medline]
Westerfield M. The Zebrafish Book. A Guide for the Laboratory Use of Zebrafish (Danio rerio) (2000) Eugene, OR: University of Oregon Press.
Willerson J, Delgado R III, Mann D. Treating relentlessly progressive congestive heart failure: What next? Tex. Heart Inst. J. (1998) 25:235–237.[Web of Science][Medline]
Wong S, Fukuchi M, Melnyk P, Rodger I, Giaid A. Induction of cyclooxygenase-2 and activation of nuclear factor-kappaB in myocardium of patients with congestive heart failure. Circulation (1998) 98:100–103.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||