Current Trends and Challenges in Natural Product Discovery
Current Trends and Challenges in Natural Product Discovery
Natural products - referring here to small molecules produced by living organisms,
particularly secondary metabolites - are an excellent yet currently underutilized
source of chemical diversity for drug discovery. Bioactive compounds isolated from
plants, fungi and bacteria have given rise to a wide range of human therapeutics,
and numerous others are effective tools in cell biology, biochemistry and pharmacology.
Despite this record of success, the field of natural product research has nonetheless
witnessed a significant decline in the past two decades, due in large measure to the
emergence of target validation, combinatorial chemistry and high-throughput screening
as a new paradigm for drug discovery [1]. Several challenges unique to natural products have contributed to this trend as
well, including the difficulties associated with isolating pure compounds from crude
extracts, identifying their mechanism of action, and synthesizing these oftentimes
highly complex structures [2], [3].
Conversely, the overall effectiveness of target-driven drug discovery has recently
been called into question [4], especially with respect to the limitations of combinatorial chemistry in terms
of providing sufficient structural diversity for the identification of bioactive,
drug-like compounds. Furthermore, there is a growing appreciation that natural products
are privileged structures that have been evolutionarily selected on the basis of their
ability to functionally interact with biological macromolecules [3]. Interest in natural products as a complementary, promising source of novel leads
for drug discovery will therefore likely continue to expand.
Another drawback to target-driven drug discovery is the reduced complexity of screening
assays - usually involving the ultra-high-throughput analysis of compound interactions
with single targets, but too often giving rise to lead compounds that do not perform
well in vivo [4]. The increasing awareness of this limitation is fueling interest in novel, high-content
bioassays which enable target-independent screening and function-based identification
of small molecules. Although cell-based screening assays using pathway-specific reporters
are already a step in the right direction, bioassays based on intact, living organisms
such as zebrafish provide an additional degree of physiological relevance that is
difficult to match in vitro.
Advantages of Zebrafish as a Model Organism
Advantages of Zebrafish as a Model Organism
Zebrafish have recently emerged as a powerful model for the rapid analysis of gene
function and small-molecule bioactivity in a variety of indications [5]. Because of their high genetic, physiological and pharmacological similarity to
humans, zebrafish are well-suited for the identification of bioactive natural products
with therapeutic potential. The primary advantages of zebrafish as a model organism
include the small size of embryos and larvae with which most tests can be performed
(1 - 5 mm, depending on developmental stage), the high fecundity of adult zebrafish
(hundreds of offspring per breeding pair per week), the optical transparency of embryos
and larvae (allowing easy visualization of internal organs and tissues), and the speed
at which these develop ex utero ([Fig. 1]). These features combine to create an ideal in vivo model suitable for medium-throughput phenotypic screening in microtiter plates. With
regard to testing the bioactivity of small molecules, a key advantage of using zebrafish
embryos and larvae is that compounds can simply be added to the (non-sterile) water
surrounding them, further increasing screening throughput (for many phenotypic assays,
hundreds of samples can be tested per day per investigator).
The zebrafish (Danio rerio) - a diploid minnow of the teleost family Cyprinidae, indigenous to the Ganges River
Basin - was first established as a genetic model in the 1970s by Streisinger and colleagues
[6]. This species was subsequently exploited for large-scale forward genetic screens
by the laboratories of Nüsslein-Volhard and Driever using ENU mutagenesis [7], [8], and by Hopkins and colleagues using retroviral insertional mutagenesis [9]. While the primary purpose of these screens was to elucidate the genetic mechanisms
of vertebrate development, several were designed to identify biomedically relevant
genes, such as those regulating the growth, differentiation and function of tissues
and organs. Examples include ENU mutagenesis screens for the isolation of zebrafish
mutants with defects in cartilage development [10], hematopoiesis [11], and cardiovascular development [12]. More recent forward genetic screens include assays for gastrointestinal function
[13], vascular development [14], and epilepsy [15].
Fig. 1 Zebrafish as a model organism for drug discovery. a Adult zebrafish; b microinjection
of one-cell stage embryos; c agarose-embedded larvae at 5 days post-fertilization
(dpf), revealing size in relation to paperclip; d, e visualization of developing vasculature
in embryo at 30 hours post-fertilization (hpf) by whole-mount in situ hybridization with flk-1 RNA probe, revealing outgrowing intersegmental vessels (ISVs); f, g embryo microinjected
with antisense morpholino oligomer (MO) targeting vascular endothelial growth factor
(VEGF), exhibiting lack of ISV outgrowth; h visualization of developing vasculature
in transgenic fli-1:eGFP embryo at 40 hpf by laser multiphoton confocal microscopy; i embryo treated
with PI3K inhibitor wortmannin, exhibiting reduced ISV outgrowth; j visualization
of bone formation in larva at 10 dpf by whole-mount Alizarin red staining; k larva
treated with glucocorticoid prednisolone, exhibiting reduced bone formation; l visualization
of cartilage formation in untreated embryo at 3 dpf by whole-mount Alcian blue staining;
m embryo microinjected with antisense MO targeting gene involved in craniofacial development,
exhibiting reduced staining of, and/or missing, cartilage elements; n visualization
of leukocyte migration by whole-mount myeloperoxidase staining in larva at 4 dpf with
tail fin injury; o larva with tail fin injury treated with the crude methanolic extract
of an anti-inflammatory East African medicinal plant (Fabaceae family), exhibiting
reduced leukocyte migration to site of injury.
Within the past decade, zebrafish have also become a versatile vertebrate model for
reverse genetics. Antisense morpholino phosphorodiamidate oligonucleotides [16] - commonly referred to as ”morpholinos” - are effective tools in zebrafish for the
specific inhibition of gene expression [17]. Dose-dependent ”knockdowns” of gene expression can be achieved by the microinjection
of morpholinos into single-cell-stage embryos ([Fig. 1]
b), wherein the antisense oligonucleotides can be designed to inhibit either the translational
initiation of mRNAs or the splicing of hnRNAs. Numerous genes have been functionally
analyzed in zebrafish in this manner [18], including several identified within the context of large-scale reverse genetic
screens [19], [20], [21]. In addition, it is now possible to use morpholinos for inducible gene knockdowns,
via the light-activated cleavage of photolabile complementary oligonucleotides that
block their ability to bind their target RNA [22]. Conversely, the efficacy of siRNA in zebrafish embryos remains a topic of debate
[23], [24], although it was recently shown that shRNA expressed from an integrated transgene
effectively inhibits gene expression in developing embryos [25].
Transgenesis methods in zebrafish are well-established, allowing the straightforward
generation of transgenic lines expressing fluorescent reporter proteins under the
control of tissue-specific promoters. Insertional mutagenesis screens using enhancer-trap
and gene-trap transposons and retroviral vectors are also providing many novel reporter
lines [26]. Collectively, this growing collection of transgenic zebrafish lines represents
a powerful resource for the systematic elucidation of organogenesis, and for many
aspects of cell biology to be performed in vivo [27]. In the area of angiogenesis, for example, transgenic lines have been used for the
detailed anatomic analysis of embryonic blood vessel formation [28], as well as to carry out both genetic [14] and compound screens [29] for the identification of novel genes and small molecules modulating these processes.
Recent innovations in zebrafish transgenesis include an elegant approach for targeted
ablations - e. g., for the generation of zebrafish larvae lacking pancreatic beta
cells, which can then be observed to regenerate - through the tissue-specific expression
of bacterial nitroreductase followed by incubation of the transgenic zebrafish embryos
in metronidazole [30]. In addition, a novel technique has been reported for the induction of transgene
expression in zebrafish using small-molecule agonists of the insect-specific ecdysone
receptor (EcR) [31], [32]. This approach allows for the dose-dependent regulation of transgene activity simply
through the addition of EcR agonists to the embryo medium - an elegant tool which
enables the precise temporal control of gene expression and which will therefore facilitate
the molecular dissection of signaling pathways during embryonic development.
As with any experimental system, there are clearly also limitations in working with
zebrafish as a model organism. First, given that most screens are carried out at early
life stages, it is important to consider the physiological differences between embryonic
or larval fish and adult mammals. In any case, bioactive small molecules identified
in zebrafish should ideally be validated in an appropriate rodent assay before being
prioritized as lead compounds. Second, depending on their molecular weight and hydrophobicity,
not all small molecules are readily absorbed by embryos and larvae, possibly leading
to false negatives during screening. Finally, given that zebrafish are vertebrates,
country-specific animal rights legislation and institutional bioethics regulations
need to be taken into account when planning experiments. It is worth noting that there
are several jurisdictions which do not consider unhatched fish embryos to be vertebrates
in the full legal sense, enabling certain large-scale screens to be carried out with
minimal administrative requirements.
Zebrafish-Based Analysis of Natural Products and other Small Molecules
Zebrafish-Based Analysis of Natural Products and other Small Molecules
Zebrafish were first proposed as an in vivo model for small-molecule drug discovery over fifty years ago [33]. This initial study investigated the utility of zebrafish embryos and larvae for
screening both natural products and synthetic compounds, especially with respect to
their effects on cell division and differentiation. In addition to numerous studies
over the following decades focusing on the detection or characterization of environmental
contaminants [34], [35], developing zebrafish embryos have been used to determine the embryotoxicity or
teratogenicity of several natural products consumed by humans, including various flavonoids
[36]; delta-9-tetrahydrocannabinol, the major psychoactive constituent of marijuana [37]; and arecoline, the major alkaloid in betel nuts [38]. In a related approach, adult zebrafish were recently utilized to confirm the piscicidal
properties of arylnaphthalide lignans isolated from Phyllanthus piscatorum, an ethnomedicinal plant used by the Amazonian Yanomami as a fish poison and antifungal
remedy [39], [40].
Despite an appreciable history of pharmacological investigation dating back over half
a century, it is only within the past decade that large-scale screens of small molecules
have been performed using zebrafish-based assays. Initial compound screens have focused
on the analysis of combinatorial chemistry libraries - encompassing up to 20,000 molecules
- for their ability to induce a variety of developmental phenotypes [41], [42], [43]. More recently, small-molecule screening in zebrafish has been employed to identify
synthetic compounds acting as suppressors of aortic coarctation, a congenital heart
defect [44], as modulators of pigmentation [45], [46], [47], as cell cycle inhibitors [48], as regulators of heart rate [49], and as modulators of tissue regeneration [50].
Not all large-scale screens performed in zebrafish have been designed to find potential
therapeutics, however. For example, bioactive small molecules can also be used to
induce index phenotypes - i. e., disease-like states - which can subsequently be exploited
in modifier screens to look for genetic or chemical suppressors of this phenotype.
Using this approach, compounds have been identified that induce reversible hemolytic
anemia [51]. In a related approach, compounds were screened for their ability to act as radiation
sensitizers [52]. Taken together, these studies have clearly established zebrafish as a facile model
for in vivo drug discovery [5], [53].
Zebrafish-Based Analysis of Plant Extracts
Zebrafish-Based Analysis of Plant Extracts
With respect to natural product discovery, no large-scale screens of compounds of
natural origin have yet been reported in zebrafish. In addition, beyond the natural
products mentioned above as having been analyzed in zebrafish, very few zebrafish-based
studies of plant extracts have been reported to date. Furthermore, the potential of
in vivo zebrafish assays for the bioactivity-guided fractionation of complex natural extracts
is only beginning to be explored.
One of first reports in this direction describes the use of transgenic zebrafish to
characterize the pro-angiogenic properties of the East Asian medicinal plant Angelica sinensis (dong quai), but does not mention the isolation of any bioactive constituents. In
this study, Lee and colleagues first demonstrate that an Angelica sinensis extract stimulates proliferation, migration, and tube formation in human umbilical
vein endothelial cells (HUVECs). They then substantiate these findings through the
observation of increased subintestinal vessel (SIV) growth in extract-treated fli-1:EGFP transgenic zebrafish, which have endothelial cell-specific expression of green
fluorescent protein [54].
Another report, presented at the 2006 American Society of Pharmacognosy Meeting, goes
one step further with respect to isolating bioactive constituents. Here, Xu and colleagues
report the bioactivity-guided purification of three anti-angiogenic compounds from
the anti-inflammatory Chinese medicinal plant Tripterygium wilfordii (lei gong teng), in this case using whole-mount histochemistry (alkaline phosphatase
staining) to visualize blood vessel development in zebrafish embryos [55]. These three molecules include the known angiogenesis inhibitors triptolide (a diterpene
triepoxide with immunosuppressive properties and inhibitory activity on NF-κB) and
celastrol (a triterpenoid with anti-inflammatory and antioxidant activity), as well
as cangoronine (a triterpenoid with no previously reported bioactivity, but with only
marginal anti-angiogenesis activity in this zebrafish assay).
At the 2007 European Zebrafish Genetics and Development Meeting, Patton and colleagues
describe the zebrafish bioassay-guided fractionation, using the induction of an early
embryonic phenotype as an assay, to partially isolate a cyclopamine-like activity
from a crude plant extract [56]. It is worth noting that cyclopamine, a steroidal jerveratrum alkaloid, was originally
isolated from Veratrum californicum (corn lilly) by virtue of its ability to induce cyclopia in newborn sheep, and that
more recently the mechanism of action of this teratogen was found to be the potent
inhibition of hedgehog signaling [57]. Because of this specific activity, cyclopamine is now an important reagent for
the control of embryonic stem cell differentation in vitro [58]. Given the important role of aberrant hedgehog signaling in several human cancers
(including basal cell carcinoma, medulloblastoma and rhabdomyosarcoma), these findings
qualify cyclopamine and similarly acting natural products as attractive lead compounds
for the development of novel anti-cancer therapeutics [59]. Indeed, the zebrafish-based screening of dietary alkaloids recently revealed several
of these - most of them similar in structure to cyclopamine - to have similar activities
as inhibitors of hedgehog signaling [60]. As such, using zebrafish embryos as an in vivo bioassay may be a promising strategy for the identification of novel natural products
that inhibit hedgehog signaling or other developmental pathways relevant for human
disease.
In order to further examine the potential of zebrafish as a platform for natural product
discovery, our laboratory recently initated an effort based on the zebrafish-mediated
screening and bioactivity-guided fractionation of crude plant extracts [61]. Using high-performance thin-layer chromatography (HP-TLC) in combination with high-resolution
electrospray ionization mass spectrometry (HR-ESI-MS) and chemical database analysis,
we were able to rapidly isolate emodin and an abietane diterpenoid, coleon A lactone,
from the crude methanolic extracts of two East African medicinal plants (Oxygonum sinuatum and Plectranthus barbatus, respectively). The zebrafish assay used for the activity-guided isolation of these
two compounds examined their ability to potentiate a sub-effective dose of the anti-angiogenic
compound SU5416, an indoline inhibitor of the vascular endothelial growth factor (VEGF)
receptor [62], in fli-1:EGFP transgenic embryos. In this context it is worth noting that emodin, an inhibitor
of the protein kinase CK2 [63], is known to restrict angiogenesis both in vitro [64] and in vivo [65], and that coleon A lactone was subsequently also found by our investigation to inhibit
endothelial cell proliferation and tube formation in vitro.
This study illustrates the utility of zebrafish, in combination with HP-TLC and HR-ESI-MS,
for the rapid and cost-effective bioactivity-guided isolation of natural products
from crude plant extracts. The primary advantages of this approach are (1) the requirement
for only sub-milligram quantities of crude extracts, fractions, and pure compounds
for an in vivo analysis of bioactivity, thereby avoiding the use of preparative column chromatography
during initial screening stages; and (2) the rapid identification of known compounds,
thereby achieving cost-effective dereplication early in the discovery process and
facilitating the prioritization of laboratory resources for the pursuit of novel molecules.
Furthermore, it highlights the potential of performing natural product discovery in
zebrafish using screening assays designed to identify novel modulators of specific
signaling pathways and/or novel potentiators of existing drugs.
Chemical Genetics in Zebrafish
Chemical Genetics in Zebrafish
A key challenge in natural product discovery is the identification of the mechanism
of action, and ultimately the biological target, of bioactive compounds. Chemical
genetics - the elucidation of biological function through the use of small molecules
- is a particularly powerful strategy in this regard, and is finding increasing application
in a number of model systems. Zebrafish represent a powerful system with which to
explore the mechanism of action of natural products and other small molecules, as
demonstrated in several recent reports described below. It is worth noting that such
studies usually rely on a combination of experiments in zebrafish (for phenotypic
data to ascertain pathway specificity) and mammalian cells (for molecular and biochemical
data to confirm these findings).
One of these studies identified the biological target of apratoxin A, a cyclodepsipeptide
cytotoxin isolated from the marine cyanobacterium Lyngbya majuscula [66]. Apratoxin A is one of several cyanobacterial metabolites with potent anti-tumor
activity, and although two of these (cryptophycin and dolastatin 10) have been shown
to bind microtubules, the other cyanobacterial cytotoxins remain elusive in terms
of their mode of action. In an elegant approach, Luesch and colleagues first combined
gene expression profiling and genome-wide, arrayed cDNA overexpression to reveal that
apratoxin A induces apoptosis in mammalian cells via the modulation of fibroblast
growth factor (FGF) signaling. After being shown to inhibit the phosphorylation-dependent
activation of STAT3, a downstream effector of FGF signaling, apratoxin A was then
found to affect FGF-dependent developmental processes in zebrafish embryos. These
phenotypic defects were similar to those induced by the FGF receptor inhibitor SU5402,
and included the downregulation of the FGF target gene mkp3 [67].
Another recent study using zebrafish focused on the mechanism of action of fumagillin,
a sesquiterpene antibiotic isolated from the airborne saprophytic fungus Aspergillus fumigatus [68] and currently used to treat microsporidiosis in humans and honeybees. Fumagillin
and its analogue O-(chloracetyl-carbamoyl)fumagillol (TNP-470) have been characterized as potent angiogenesis
inhibitors [69], and an N-(2-hydroxypropyl)methacrylamide (HPMA) copolymer conjugate of TNP-470 is currently
in late-stage preclinical development as an anti-angiogenesis drug for the treatment
of solid tumors [70]. Despite its known inhibition of the methionine aminopeptidase MetAP-2 [71], this interaction alone did not fully explain fumagillin’s biological properties.
To further explore how the restriction of MetAP-2 activity modulates angiogenesis,
Crews and colleagues treated developing zebrafish embryos with TNP-470, inducing a
truncated tail dysmorphology that through phenotype database analysis was found to
be highly reminiscent of the Wnt5a mutant pipetail. A similar phenotype was induced through antisense morpholino oligonucleotides directed
against MetAP-2, suggesting a functional link between MetAP-2 and Wnt5a signaling.
Further experiments in mammalian cells indicated that this interaction occurs in the
context of non-canonical Wnt signaling (i. e., via the receptor Frizzled2), and that
the point of action of TNP-470 is downstream of Frizzled2 and upstream of the non-canonical
Wnt effectors CamKII, c-Jun, and RhoA [72], [73].
Behavioral assays in zebrafish have also recently been used to examine the mechanism
of action of salvinorin A, a physchoactive trans-neoclerodane diterpenoid isolated from Salvia divinorum [74]. This Lamiaceae species has a long history of traditional use as an entheogen by
Mazatec shamans in Central America - an ethnomedicinal application supported by the
fact that salvinorin A is a potent kappa-opioid receptor agonist and currently the
most powerful known hallucinogen of natural origin. By examining the swimming behavior
and conditioned place preference of salvinorin A-treated adult zebrafish, it was possible
to ascertain reinforcing properties that were fully reversible by both nor-binaltorphimine
(a kappa-opioid receptor antagonist) and rimonabant (a CB1 cannabinoid receptor antagonist),
implicating both pathways in mediating the rewarding effects of salvinorin A and further
establishing zebrafish as a useful model for addiction research (see also [75]).
A further example of the power of chemical genetics in zebrafish is the recent description
of dorsomorphin, a novel small-molecule inhibitor of bone morphogenic protein (BMP)
signaling [76]. In this study, Peterson and colleagues screened 7,500 synthetic compounds, identifying
one that perturbed dorsoventral axis formation in early embryos - a phenotype highly
reminiscent of transgenic zebrafish overexpressing noggin (an extracellullar inhibitor of BMP) and lost-a-fin mutants (with a loss-of-function allele of the BMP receptor ALK8). Rescue experiments
in zebrafish embryos showed that dorsomorphin suppressed the ventralized phenotype
induced by antisense-mediated inhibition of chordin (another BMP inhibitor) - an elegant confirmation of dorsomorphin’s own ability to
block BMP signaling. Subsequent biochemical analysis of BMP pathway components in
dorsomorphin-treated mammalian cells confirmed that this compound specifically affects
SMAD 1/5/8-dependent signaling and not MAPK activity, both of which are stimulated
by BMP type I receptor activation. As expected from an inhibitor of BMP signaling,
dorsomorphin-treated zebrafish larvae exhibited reduced bone mineralization. Finally,
dorsomorphin proved its utility as a new pharmacological tool in mice, helping to
elucidate the role of BMP signaling in iron homeostasis.
Zebrafish Assays Useful for Natural Product Discovery
Zebrafish Assays Useful for Natural Product Discovery
Beyond the plethora of bioassays mentioned above and in [Table 1], we would like to highlight three areas of particular applicability for zebrafish-mediated
natural product discovery: immunity and inflammation, epilepsy, and cardiotoxicity.
Despite their clear in vivo relevance, rodent-based assays in these areas have inherent disadvantages in terms
of cost and throughput that render them less suitable as a primary screening platform
for the discovery of novel, bioactive small molecules. Zebrafish larvae, on the other
hand, are an ideal in vivo system for such applications, and are clearly appropriate for medium-throughput screening
of plant extracts and natural products.
Table 1 Overview of selected zebrafish assays with potential relevance for natural product
discovery
| Indication |
Zebrafish Assay |
Description |
Reference |
| Cancer |
Oncogene-induced tumors |
Heat-shock induction of Ras-dependent tumors in Cre/loxP transgenics |
[93], [94]
|
|
|
Cre RNA induction of Myc-dependent tumors in loxP transgenics |
[95]
|
|
|
T-cell leukemia induced by Myc overexpression |
[108]
|
|
Tumor suppressor-induced tumors |
Target-selected ENU mutagenesis screen for p53 mutations |
[96]
|
|
Mutagen-induced tumors |
Transplantation of isogeneic tumor cells between clonal zebrafish lines |
[97]
|
|
Tumor imaging |
Ultrasound imaging of liver tumors in adult zebrafish |
[98]
|
|
Tumor metastasis |
Transplantation of human tumor cells into zebrafish embryos |
[99], [100], [101], [102]
|
|
Radiation sensitivity |
Small-molecule screens for radiation sensitizers and protectants |
[52], [103], [109]
|
|
Genomic instability |
ENU mutagenesis screen for genomic instability mutants |
[104], [105]
|
|
Cell migration |
Antisense screen for genes affecting primordial germ cell migration |
[19]
|
|
Cell cycle |
Small-molecule screen for cell cycle inhibitors |
[48]
|
|
|
ENU mutagenesis screen for mitosis mutants |
[105]
|
|
|
Small-molecule screen for suppressors of mitosis mutant (crash and burn) |
[106]
|
| Skeletal disorders |
Cartilage formation |
ENU mutagenesis screen for cartilage mutants |
[10]
|
|
Osteoporosis |
Glucocorticoid-induced reduction of bone formation; characterization of bone anabolic
compounds |
[110], [111]
|
| Blood disorders |
Hematopoiesis |
ENU mutagenesis screen for hematopoiesis mutants |
[11]
|
|
|
Small-molecule screen for inducers of reversible hemolytic anemia |
[51]
|
|
Thrombosis |
Laser-induced thrombosis in larvae |
[112]
|
| Cardiovascular disorders |
Heart formation and function |
ENU mutagenesis screen for heart and cardiovascular mutants |
[12]
|
|
|
Small-molecule screen for suppressors of aortic coarctation mutant (gridlock) |
[44]
|
|
Angiogenesis |
ENU mutagenesis screen for angiogenesis mutants using transgenic reporter line |
[14]
|
|
|
Small-molecule screen for anti-angiogenic compounds using transgenic reporter line |
[28], [29]
|
|
|
Characterization of anti-angiogenic natural products in transgenic larvae |
[107]
|
|
Cardiotoxicity |
ENU mutagenesis screen for QT prolongation mutants |
[92]
|
|
|
Characterization of QT-prolonging drugs in zebrafish larvae |
[89], [90], [91]
|
|
Heart rate |
Small-molecule screen for modulators of heart rate |
[49]
|
| Neurological disorders |
Epilepsy |
Validation of anti-epileptic drugs in larvae with PTZ-induced seizures |
[86], [87]
|
|
|
ENU mutagenesis screen for seizure-resistant mutants |
[15]
|
|
Addiction |
Characterization of psychoactive drugs and drugs of abuse |
[74], [75], [113]
|
|
|
ENU mutagenesis screen for differences in cocaine sensitivity |
[114]
|
|
Behavior and learning |
ENU mutagenesis screen for embryos/larvae with behavioral defects |
[115]
|
|
|
Characterization of drug effects on non-associative learning in larvae |
[116]
|
|
|
Automated detection of behavioral defects |
[117]
|
|
Neurodevelopment |
ENU mutagenesis screen for enteric nervous system defects |
[118]
|
|
Axon myelination |
ENU mutagenesis screen for axon myelination defects |
[119], [120]
|
|
Neurodegeneration |
Transgenic line with overexpression of mutant human Tau protein |
[121]
|
|
ALS |
Transient transgenic model with overexpression of mutant human SOD protein |
[122]
|
| Immunity and inflammation |
Bacterial infection |
Characterization of immune response to bacterially infected larvae |
[77], [79], [80]
|
|
|
Small-molecule screening for antimicrobials in bacterially infected larvae |
[78]
|
|
Inflammation |
Histochemical visualization of leukocyte migration after injury |
[81]
|
|
|
Transgenic zebrafish lines for visualization of leukocyte migration |
[82], [83], [84]
|
| Metabolic disorders |
Lipid metabolism |
ENU mutagenesis screen for defects in digestive physiology |
[13]
|
|
Diabetes |
ENU mutagenesis screen for defects in pancreas formation using transgenic reporter
lines |
[123]
|
|
Liver disease |
ENU mutagenesis screens for liver formation and hepatomegaly |
[124], [125]
|
| Other |
Pigmentation |
Small-molecule screens for modulators of pigmentation and melanocyte differentatiation |
[45], [46], [47], [129]
|
|
Tissue regeneration |
Small-molecule screen for modulators of tissue regeneration |
[50]
|
|
Hearing loss |
Characterization of ototoxic and otoprotective agents on larval hair cells |
[126], [127]
|
|
Circadian rhythms |
ENU mutagenesis screen for circadian rhythm mutants |
[128]
|
| Note that several mutant/transgenic lines are described above in terms of how they
were generated (e. g., via ENU mutagenesis screens); the assays used for these screens
and/or the mutant/transgenic lines themselves may be useful for characterizing screening
natural products or other small molecules. |
Immunity and inflammation
With potential utility for the discovery of novel antimicrobial agents, several zebrafish
models for bacterial pathogenesis have been reported, most of which rely on the use
of Mycobacterium marinum [77], [78]. Other studies have examined the pathogenesis and inflammatory response of zebrafish
embryos to bacterial infection, for example, by Edwardsiella tarda, which results in the significant upregulation of interleukin-1β and TNFα as determined
by qRT-PCR [79]. Moving towards an easier and lower-risk model for the innate immune reponse to
bacterial infection in vivo, Watzke and colleagues have recently shown that bacterial lipopolysaccharides (LPS)
are able to induce a very similar upregulation of these two pro-inflammatory cytokines
in zebrafish embryos [80].
Zebrafish larvae are now also well-established as an in vivo model for inflammation. In 2001, Lieschke and colleagues first showed that leukocytes,
including both circulating granulocytes and macrophages, could be readily visualized
in zebrafish larvae though whole-mount histochemical staining for myeloperoxidase
activity within the first 48 hours after fertilization, and that transection of part
of the larval tail stimulated the migration and accumulation of leukocytes within
hours after injury [81]. More recently, two other groups have generated transgenic lines expressing EGFP
under the control of the myeloperoxidase promoter, enabling the direct visualization
of live leukocytes and therefore the observation of the inflammatory response in real
time [82], [83]. In combination with other possibilities to analyze the inflammatory response, for
example, using transgenic embryos with EGFP-expressing early macrophages [84], the zebrafish clearly has the potential to be an effective system for the identification
of anti-inflammatory natural products and other small molecules.
Epilepsy
Zebrafish have recently also emerged as a novel experimental model of epilepsy [85]. Baraban and colleagues first described the response of zebrafish larvae to pentylenetetrazol
(PTZ), a convulsant widely used to induce seizures in rodents [86]. This investigation confirmed the ability of PTZ to induce a dose-dependent series
of stereotypical behaviors, including clonus-type convulsions characteristic of epilepsy
in mammals. Electrophysiological recordings from the brains of PTZ-treated zebrafish
larvae indicated epileptiform discharges that were reduced through the application
of anti-epileptic drugs such as valproate and diazepam. Baraban and colleagues next
performed a forward genetic screen to identify zebrafish mutants insensitive to the
effects of PTZ, with the goal of elucidating the genetic basis of seizure resistance
[15].
In order to facilitate the discovery of novel anti-epileptic drugs, Goldsmith and
colleagues have recently developed an automated tracking system to monitor the movement
of zebrafish larvae in microtiter plates [87]. Using this methodology, 13 of 14 known anti-epileptic drugs were found to suppress
PTZ-induced behavioral abnormalities, thereby validating the suitability of zebrafish
for the high-throughput screening of potential anti-epileptic compounds. Intruigingly,
domoic acid (DA, a phycotoxin produced by Pseudo-nitzschia diatoms) was recently found to increase the susceptibility of zebrafish larvae to
PTZ, further confirming the role of DA in seizure induction in marine mammals exposed
to algal blooms [88]. Taken together, the progress in this area underscores the potential of zebrafish
as a promising in vivo bioassay for the discovery of anti-epileptic natural products.
Cardiotoxicity
Another noteworthy advantage of using zebrafish larvae for compound screening is that
various aspects of cardiotoxicity can be investigated at a very early stage in the
drug discovery process. Cardiotoxicity is one of the more common adverse effects seen
for many new chemical entities, and the clinical development of numerous drug candidates
has been discontinued because of their tendency to induce cardiac arrhythmias such
QT prolongation and torsades de pointes. One of the primary causes of cardiotoxicity
are unforeseen interactions of compounds with the HERG potassium channel (the protein
product of the KCNH2 gene, and also referred as the Kv11.1 potassium channel), responsible for the repolarizing IKr current in the cardiac action potential. Despite this appreciable risk, drugs are
normally only examined relatively late in pre-clinical testing using, for example,
isolated guinea pig hearts, which requires multi-milligram amounts of each compound
to be tested.
With their requirement for only microgram quantities of any given compound to be analyzed,
combined with the ability to readily observe heart function in transparent larvae,
zebrafish are well-suited as a front-line assay for cardiotoxicity. In 2003, two laboratories
validated zebrafish larvae as a reliable model to test compounds for their ability
to induce QT prolongation [89], [90], [91]. These studies showed that 22 out of 23 compounds known to cause QT prolongation
in humans induced bradycardia or atrioventricular block in zebrafish, and that similar
effects were observed by the antisense-mediated knockdown of Zerg (encoded by kcnh2, the zebrafish ortholog of human KCNH2, both of which share 99 % amino acid identity in the pore domain of their protein
products). In a related effort, one of the groups went on to establish a transgenic
line with myocardium-specific expression of EGFP, and used this model together with
automated fluorescence microscopy for the high-throughput screening of small molecules
that modulate heart rate in zebrafish embryos [49]. More recently, another group isolated the mutation responsible for an inherited
arrhythmia in zebrafish, determining the affected gene to be kcnh2 by genetic mapping and direct sequencing [92]. Together, these findings underscore the suitability of zebrafish to serve as a
useful model for prescreening natural products and other small molecules for their
potential risk to cause QT prolongation and other arrhythmias.
Outlook
Outlook
Zebrafish are now firmly established as a powerful research platform for many areas
of biology and drug discovery. Given the relative ease with which natural products
and other small molecules can be analyzed in zebrafish, it is conceivable that within
the next decade the large majority of currently available plant and fungal extracts
(and increasing numbers of microbial extracts) can be screened in a plethora of biomedically
relevant assays - an undertaking not currently feasible using other in vivo models. Zebrafish-based assays have the capacity to facilitate the bioassay-guided
fractionation of large numbers of bioactive extracts identified in such in vivo screens, and thereby to enable the isolation of numerous novel, bioactive natural
products - several of which are likely to be attractive lead compounds for the development
of new, effective drugs.
