Non-Toxic Chemical Interdiction of the Epithelial-to-Mesenchymal Transition by Targeting Cap-dependent Translation

doi:10.1021/cb9000475

ACS Chem Biol. Author manuscript; available in PMC 2010 May 15.

Published in final edited form as:

ACS Chem Biol. 2009 May 15; 4(5): 367–377.

doi: 10.1021/cb9000475

PMCID: PMC2796976

NIHMSID: NIHMS159016

Non-Toxic Chemical Interdiction of the Epithelial-to-Mesenchymal Transition by Targeting Cap-dependent Translation

Brahma Ghosh,^a¹ Alexey O. Benyumov,^b¹ Phalguni Ghosh,^a Yan Jia,^a Svetlana Avdulov,^b Peter S. Dahlberg,^b Mark Peterson,^b Karen Smith,^b Vitaly A. Polunovsky,^b Peter B. Bitterman,^b and Carston R. Wagner^a^*

^aDepartment of Medicinal Chemistry, University of Minnesota, Minneapolis, MN 55455, USA

^bDepartment of Medicine, University of Minnesota, Minneapolis, MN 55455, USA

^*Address correspondence to: C.R. Wagner, University of Minnesota, Department of Medicinal Chemistry, 8-174 Weaver Densford Hall, 308 Harvard St. S.E., Minneapolis, MN 55455. Phone: 612-624-2614. Email: wagne003/at/umn.edu

¹B. Ghosh and A.O.Benyumov contributed equally to this manuscript.

The publisher's final edited version of this article is available at ACS Chem Biol

See other articles in PMC that cite the published article.

Abstract

Normal growth and development depends upon high fidelity regulation of cap-dependent translation initiation; a process that is usurped and redirected in cancer to mediate acquisition of malignant properties. The epithelial-to-mesenchymal transition (EMT) is a key translationally-regulated step in the development of epithelial cancers as well as pathological tissue fibrosis (1–5). To date, no compounds targeting EMT have been developed. Here we report the synthesis of a novel class of Histidine Triad Nucleotide Binding Protein (HINT)-dependent pronucleotides that interdict EMT by negatively regulating the association of eIF4E with the mRNA cap. Compound eIF4E inhibitor-1 (4Ei-1) potently inhibited cap-dependent translation in a dose-dependent manner in zebrafish embryos without causing developmental abnormalities; and prevented eIF4E from triggering EMT in zebrafish ectoderm explants without toxicity. Metabolism studies with whole cell lysates demonstrated that the prodrug was rapidly converted into 7-Bn-GMP. Thus we have successfully developed the first non-toxic small molecule able to inhibit EMT, a key process in the development of epithelial cancer and tissue fibrosis by targeting the interaction of eIF4E with the mRNA cap; and demonstrate the tractability of zebrafish as a model organism for studying agents that modulate EMT. Our work provides strong motivation for the continued development of compounds designed to normalize cap-dependent translation as novel chemo-preventive agents and therapeutics for cancer and fibrosis.

INTRODUCTION

Recruitment of the small ribosome subunit to the 5’ end of mRNA is the rate-controlling step in the initiation of eukaryotic protein synthesis. For the majority of eukaryotic transcripts, this process requires assembly of the heterotrimeric translation initiation complex eIF4F and its association with the 7-MeGTP (m7GpppX) cap structure at the 5’ end of mRNA (6–9). The eIF4F complex consists of eIF4E, the mRNA cap-binding protein; eIF4A, a helicase; and eIF4G, a docking protein that associates with the 40S ribosomal subunit via contact with the multimeric adapter complex eIF3 (10). Under physiological conditions, eIF4E is the least abundant component of the initiation machinery and therefore controls the rate of translation initiation (11).

The translation initiation apparatus functions as a key regulatory hub in the flow of genetic information from the genome to the proteome (12–14). Normal growth and development features high fidelity regulation of transcript egress from the nucleus and ribosome recruitment (15,16). In contrast, cancer is characterized by unrestrained activation of the translation initiation machinery, altering the genome-wide pattern of translation initiation to favor cell autonomous function (17–20). This has generated considerable interest in the translation initiation machinery as a target for rational anticancer therapy (12,21).

The eIF4E–eIF4G binding step has been successfully targeted in cancer models (22); however, this approach may be limited by its lack of selectivity for cancer-related translation initiation. Targeting the interaction of eIF4E with the cap may afford a means to overcome this limitation. The efficiency of 2 key eIF4E-dependent steps in translation initiation, transcript egress from the nucleus and ribosome recruitment, varies widely among transcripts; with those encoding potentially oncogenic proteins showing a disproportionate dependence on the association of eIF4E with its cap. We propose to exploit this natural selectivity. One approach that has shown promise is indirect; using an RNA interference strategy to decrease the abundance of eIF4E (21,23). An alternative approach that we have taken is direct; using synthetic nucleotide derivatives such as 7-benzyl guanosine monophosphate (7-BnGMP) to block the binding of eIF4E to the mRNA cap (24,25). Unfortunately, while effective in cell free systems, its efficacy in cells is low. One approach to improving in vivo activity is to develop a stable pro-drug (pronucleotide) that can be converted into an active species by the target cell (26,27). Phosphoramidates are a promising class of compounds for this purpose. They are water soluble, non-toxic, and stable; and have been used in humans as potent antiviral and antitumor agents (28–32). They display significantly longer in vivo half-lives and greater volumes of distribution than their parent nucleotides (31,32). Nucleoside phosphoramidates can be converted efficiently inside the cell into their corresponding 5’-monophosphate nucleotide by a family of phosphoramidase enzymes called histidine triad nucleotide binding proteins (HINT) (33–35). Human HINT-1 (hHINT-1) binds preferentially to purine analogues (34) and is aberrantly expressed in cancer cells (36). The enzyme prefers phosphoramidates with unhindered primary amines and tolerates substitutions at the N-7 position of the purine base without loss of substrate specificity (34).

As a first step toward our goal, we report the successful implementation of a pronucleotide strategy for the intracellular release of 7-BnGMP to inhibit cap-dependent translation. By introducing a suitable alkyl or aryl-alkyl substituent at the N-7 position, we achieved dose-dependent inhibition of cap-dependent translation in zebrafish eggs/embryos without disturbing development; and using zebrafish ectoderm explants, effectively blocked the epithelial-to-mesenchymal transition (EMT) - a key step in the genesis of epithelial cancers and tissue fibrosis - with no apparent toxicity (1–5,37,38).

RESULTS and DISCUSSION

Affinity of cap-analogs for eIF4E

We synthesized a 12-member library of 7-alkylated 5’-aryl amine/amino acid phosphoramidates of guanosine and quantified affinity for three representative examples (Figure 1) by fluorescence quenching. Removal of the γ- and β-units of the phosphate chain led to a significant fall in K_d (Table 1) in accord with the published literature (39). This underscores the role played by the phosphate backbone in the recognition of the 5’-cap by the Lys and Arg residues of the eIF4E capbinding site (40).

Figure 1

Chemical structure of cap analogs

Table 1

Comparison of dissociation constants (K_d) for the phosphoramidates and cap analogs by fluorescence-quenching.

The strength of the interaction between eIF4E and the nucleotide monophosphates increased when the methyl substituent at N7 was replaced by a benzyl group. Thus, 7-BnGMP showed a 10-fold increase in binding affinity compared to 7-MeGMP (Table 1). Crystal structure determination and analysis of eIF4E complexed with 7-BnGMP and a 4E-BP1 peptide revealed that conformational changes in the cap-binding site induced by the presence of the benzyl substituent allows it to pack into a hydrophobic cavity dorsal to the π-stacked tryptophans (W56 and W102) of the eIF4E active site (41). This added interaction partly compensates for the decreased binding affinity incurred with the loss of two phosphate units.

The K_d for the cap-derived phosphoramidates (designated eIF4E-inhibitors, “4Ei”) followed the order 4Ei-1 < 4Ei-3 < 4Ei-2. Our in silico simulation studies of 4Ei-1 complexed with eIF4E revealed that the indole side-chain of the phosphoramidate resides within a concave hydrophobic pocket in the eIF4E active site (not shown). This may account for its greater affinity for the protein compared to the D-phenylalanine (4Ei-2) and D-alanine phosphoramidates (4Ei-3). The K_d value obtained for 4Ei-1 was, however, still 3200-fold higher than that for the natural substrate 7-MeGTP; indicating remarkably poor binding between 4Ei-1 and eIF4E.

Inhibition of cap-dependent translation by cap analogs in vitro

To directly evaluate how well our cap-analogs inhibited cap-dependent translation, we employed a cell-free translation assay that uses the dual-luciferase mRNA, RLUC-POLIRES-FLUC, as a reporter. In this system, translation of Renilla reniformis luciferase (RLUC) is strictly cap-dependent; whereas translation of firefly luciferase (FLUC) proceeds in a cap-independent manner via an IRES (42) (Figure 2A). In accord with prior reports for a different class of small molecule inhibitors of cap-dependent translation (22), all active compounds stimulated IRES-mediated translation concomitantly with inhibition of cap-dependent translation at concentrations ≥100 µM (Figure 2B). While most of the phosphoramidated 7-BnGMP derivatives revealed only marginal inhibitory potency compared to their parent compounds 7-BnGMP and 7-MeGTP; phosphoramidating 7-BnGMP with D-Phenylalanine (4Ei-2) or D-Alanine (4Ei-3) resulted in retention of more than 60% of the inhibitory activity - and tryptamine phosphoramidated 7-BnGMP (4Ei-1) retained more than 80% of the inhibitory activity of the parent compounds, with an IC₅₀ comparable to 7-BnGMP (16.7±3.2 vs. 15.9±2.0 µM) (Table 2).

Figure 2

Inhibition of cap-dependent translation by cap analogs in vitro

Table 2

Comparison of IC₅₀ values determined for our lead compounds 4Ei-1 and 7-BnGMP to authentic cap analogs 7-MeGTP and 7-MeGMP.

Since 4Ei-1 has been shown to be a substrate for human Hint1 and rabbit tissues have been previously shown to express the highly homologous (95%) rabbit Hint1, we examined the rabbit reticulocyte lysates for HINT1 activity. Based on endogenous phosphoramidase activity, the lysates were shown to posses 23 ng of active Hint per 20 µL of lysate. Thus, while 4Ei-1 has a low affinity for eIF4E, its ability to serve as a substrate for endogenous rabbit Hint1 enables it to be rapidly converted to the active species, 7-BnGMP.

One criterion for therapeutic potential is compound stability in a biological milieu. To assess this property, we pre-incubated compounds at 10 to 20 µM in the rabbit reticulocyte extract for 30 or 60 min. As evidenced by Renilla luciferase reporter activity, 7-MeGTP totally lost inhibitory activity within the first 30 min; whereas 7-BnGMP retained its level of activity for up to 60 min of pre-incubation (Supplementary Fig. 1).

Inhibition of cap-dependent translation by cap analogs in vivo

The teleost zebrafish (Danio rerio) is a promising model organism for drug discovery that enables the testing of efficacy, toxicity and biological activity in the same living system (43,44). In order to examine compound efficacy and toxicity, we microinjected the dual-luciferase reporter mRNA and a test compound into freshly fertilized zebrafish eggs. For calibration, we co-injected eggs with the poison cycloheximide (CHD), which inhibits both cap-dependent and IRES-mediated reporter translation (Supplementary Fig. 2). Consistent with our cell-free experiments, injections of 4Ei-1, -2 or -3 in doses ranging from 5 to 25 pmol per egg inhibited up to 30% of cap-dependent translation (Figure 3) without adverse effects on cell division or development (monitored for 96 h until completion of embryo development, data not shown). As seen in the cell-free system, introducing active compounds led to a dose-dependent increase in IRES-reporter translation, thereby excluding a non-specific toxic effect of the compounds on the translational machinery or the organism itself. These findings indicate that 7-BnGMP and its phosphoramidated derivatives can inhibit cap-dependent translation by up to 30% in the physiological context of living vertebrate cells without cytotoxicity.

Figure 3

Inhibition of cap-dependent translation by cap analogs in vivo

4Ei-1 inhibits a key step in the pathway to cancer and tissue fibrosis – the epithelial-to-mesenchymal transition

This result immediately raises the question of whether this level of translational inhibition is relevant to human disease. We sought to develop a novel system that not only featured a step in the genesis of cancer and tissue fibrosis that could be nullified, but one that would also allow us to distinguish a specific biological effect from the non-specific toxicity of inhibiting protein synthesis. For this purpose we chose a differentiation-dependent process in the cancer and fibrosis pathways, the epithelial-to-mesenchymal transition (EMT) (45). We generated a zebrafish explant model of EMT that is triggered by ectopic expression of eIF4E – an approach based on prior studies in Xenopus laevis (46). To validate the system, fertilized zebrafish eggs were co-injected with the dual luciferase reporter and mRNA encoding wild type murine eIF4E carrying a hemaglutinin tag, “HA”; noting that zebrafish and murine eIF4E-1A isoforms share 83% identity (Supplementary Fig. 3). A cap-binding mutant (eIF4E W56A) served as a negative control and Xenopus translation elongation factor-1α (xEF-1α) served as a neutral control. Expression and translational activity of exogenous wild type eIF4E was readily detected in cell lysates 5 h post injection (Figure 4).

Figure 4

Manipulating cap-dependent translation in zebrafish embryos

In our system, ectopic expression of eIF4E triggered EMT (Figure 5). After 24 h, explants from the embryos injected with eIF4E changed their shape from spherical to elliptical and began expressing the mesoderm-specific markers no tail (ntl) and myogenic differentiation (myoD). After 48h, motile cells emerged. When explants were excised from embryos co-injected with 4E WT mRNA and 4Ei-1 (16 pmol), EMT was completely abrogated with no pathological changes in explant morphology – indicating the absence of toxicity (experimental details provided in Supplementary Table 1 and Supplementary Fig. 4). These data show that 4Ei-1 specifically interdicts EMT, a key step in the genesis of epithelial cell cancers and tissue fibrosis; thus displaying an activity with promise for preventing the evolution of pre-malignant lesions and cancer in situ to invasive cancer – and offering an option for the interdiction of progressive fibrosis.

Figure 5

Inhibition of the epithelial-to-mesenchymal transition by 4Ei-1

Mechanism of 4Ei-1 inhibition: Activation by HINT

Despite displaying a low affinity for eIF4E, 4Ei-1 potently inhibited cap-dependent translation in vitro and in vivo (vide supra) and blocked EMT. To elucidate its mechanism of action, we examined the binding of a fluorescent peptide to the eIF4E binding pocket for eIF4G and found that 4Ei-1 did not interfere (data not shown). Thus it is unlikely that 4Ei-1 inhibits cap-dependent translation by blocking eIF4G binding to eIF4E (22). Prior studies indicate that 7-BnGMP can inhibit eIF4E binding to the cap analog, 7-MeGTP (24); and we show that 7-BnGMP binds to the cap-binding region in an orientation that compensates for loss of the beta phosphate observed for 7-MeGDP. We, therefore, hypothesized that 4Ei-1 is metabolized in cells releasing the active inhibitor – 7-BnGMP. By screening the zebrafish protein database (Entrez. PubMed), we found 2 sequences - AAH81526.1 and CAI29411.1 - that are 73% homologous to the human HINT-1 and HINT-3 enzymes. The presence of HINT activity in the early stages of embryogenesis has been recently demonstrated in bovine embryos as early as the 8-cell stage (47). Therefore, we examined zebrafish gastrula lysates for HINT activity and found, based on endogenous phosphoramidase activity, 1.95 ng per 50 µL. This indicates that both rabbit and zebrafish lysates can convert 4Ei-1 to the corresponding monophosphate; and suggests that 4Ei-1, our most potent cap-dependent translation inhibitor may be a better substrate for HINT hydrolase than either 4Ei-2 or 4Ei-3. These data are in accord with the idea that 4Ei-1 might be acting as a HINT-bioactive phosphoramidate prodrug of the eIF4E antagonist 7-BnGMP.

Enzymatic processing of 4Ei-1 to 7-BnGMP

To directly determine if 4Ei-1 acted as a HINT-bioactivatable prodrug, we quantified 4Ei-1 and 7-BnGMP as a function of time in both zebrafish embryo and rabbit reticulocyte lysates using liquid chromatography/mass spectrometry (METHODS). Within 1 h after addition of 4Ei-1 (750 µM), complete loss of 4Ei-1 with a concomitant increase in 7-BnGMP was observed. (Supplementary Fig. 5). As a control, we added 4Ei-1 to heat inactivated lysates and detected no conversion; instead observing values for 4Ei-1 concentration (1413±16 to 1583±167 ng/mL) close to the predicted value for the input quantity of 4Ei-1 at 150 µM (1395 ng/mL). These data indicate that in both rabbit and zebrafish cell lysates are fully capable of converting 4Ei-1 to 7-BnGMP, confirming the prodrug mechanism. Our data does not exclude the possibility that other cellular cap-binding proteins such as 4E homologous protein (4EHP) or heterodimeric nuclear cap-binding complex (CBC) might also be targets of 4Ei-1. However, we view this as unlikely because they have a much lower (1000-fold) affinity for the cap or mononucleotide cap analogs than eIF4E (48–50); and to date, neither has ever been implicated in EMT in any biological system.

CONCLUSIONS

The EMT is an essential differentiation program in early embryonic development when germ layers and organ topography are established (reviewed in (45)). The process involves loss of both cell polarity and tight inter-cellular junctions as ectodermal epithelial cells acquire a non-polarized, migratory mesenchymal phenotype. However, in some of the most prevalent and morbid human diseases, the EMT is usurped to mediate pathological changes. For example, in cancer the EMT enables epithelial cells to transit the cancer pathway acquiring the capacity to migrate, invade tissue planes and metastasize (reviewed in (51)); and in fibroproliferative diseases of the lung, liver and kidney, EMT is a source of pathological fibroblasts that deposit connective tissue distorting normal anatomy thus leading to organ failure (reviewed in (52)). Despite its importance in human disease, no therapies have been developed to interdict EMT. As a potentially dispensable process in adults that is central to the pathogenesis of cancer and fibrosis, compounds targeting EMT could in principle be robust therapeutic agents without significant toxicity.

METHODS

General Synthetic Procedures and Materials

All reagents were purchased from commercial vendors and used without further purification. Dowex 50WX8-200 cation exchange column was converted to its Na+ form by treatment with 1N NaOH, followed by washing with water to bring pH to neutrality. 1-Ethyl-3-(3-dimethyllaminopropyl)carbodiimide hydrochloride (EDC) was used from a previously unopened bottle. Analytical thin layer chromatorgraphy (TLC) was performed on 0.25 mm precoated Merck silica gel (SiO₂) 60 F₂₅₄. Column chromatography was performed on Purasil 60A silica gel, 230– 400 mesh (Whatman). ¹H and ³¹P NMR were recorded on a Varian Mercury-300 spectrometer. Chemical shifts are reported in ppm relative to residual deuterium oxide (D₂O) or external indicator 85% H₃PO₄ peaks for ¹H and ³¹P NMR, respectively. High-resolution mass spectrometry (HRMS) data were obtained on a Biotof II (Bruker) ESI-MS spectrometer. 7-Bn-guanosine monophosphate (7-BnGMP, disodium salt) was synthesized according to established methods (40).

Phosphoramidate Synthesis

Compound 4Ei-1 was synthesized as described earlier (34), and a slight modification of the same procedure, as outlined below, was used in the preparation of compounds 4Ei-2 and 4Ei-3:

7-Bn- guanosine monophosphate disodium salt (0.100 g, 0.200 mmol) and the appropriate amino acid methyl ester hydrochloride salt (D-phenylalanine methyl ester.HCl for 4Ei-2, and D-alanine methyl ester.HCl for 4Ei-3, 1.01 mmol, 5 equiv.) were dissolved in H₂O (5 mL) and the pH of the solution adjusted to ~7.20 by dropwise addition of dilute aq. NaOH. To the foregoing was added EDC (0.192 g, 1.01 mmol, 5 equiv.) dissolved in 2 mL of 2mM N-methylmorpholine (pH = 7.0) and the resultant solution allowed to stir at room temperature overnight. Upon complete consumption of the nucleotide starting material (TLC and ³¹P NMR), the product mixture was concentrated in vacuo and the residue chromatographed on Silica gel, eluting with CHCl₃/MeOH/H₂O (5: 2: 0.25, containing 0.5% NH₄OH). The solid obtained— EDC salt of the phosphoramidate— after evaporation of the solvent was passed through an ion-exchange column (Dowex-50WX8-200, Na+ form) and the relevant fractions pooled and lyophilized to give the desired phosphoramidate sodium salts as white amorphous solids in 35–40% yield.

Compound 4Ei-2: ¹H NMR (D₂O, 300 MHz) δ 7.22- 7.26 (m, 5H), 7.07- 7.06 (m, 3H), 6.86 (d, 2H), 5.81 (d, 1H), 5.38 (dd, 2H), 4. 54 (m, 1H), 4.23 (t, 1H), 4.12 (m, 1H), 3.81 (dd, 2H), 3.65 (t, 1H), 3.31 (s, 3H), 2.61 (m, 2H) ppm. ³¹P (D₂O, 121 MHz) δ 7.11 ppm. HRMS (ESI): m/z calcd for C₂₇H₃₁N₆NaO₉P⁺ (M)+ -- 637.1788, found 637.1785.

Compound 4Ei-3: ¹H NMR (D₂O, 300 MHz) δ 7.25- 7.26 (m, 5H), 5.90 (t, 1H), 5.52 (dd, 2H), 4. 53 (m, 1H), (4.19- 4.27, 2m, 2H), 3.91 (dd, 2H), 3.52 (t, 1H), 3.38 (s, 3H), 1.07 (d, 3H) ppm. ³¹P (D₂O, 121 MHz) δ 7.26 ppm. HRMS (ESI): m/z calcd for C₂₁H₂₇N₆NaO₉P⁺ (M)+ -- 561.1469, found 561.1474.

As is the case with most 7-benzylated guanosines, the signal due to the C-8 proton was not observed for either of the two compounds because of its rapid exchange with protons from the NMR solvent.

Fluorescence Titrations and Determination of K_d

The concentration of eIF4E was optimized to 200 nM and was used for all titration experiments, which was duplicated for each compound. Fluorescence spectra were recorded on a Cary Eclipse fluorescence spectrophotometer. Titration experiments were carried out at 22°C with freshly prepared HEPES buffer (50 mM Hepes, 100 mM KCl, 1 mM DTT, 0.5 mM EDTA) at pH 7.2. Nonlinear fitting was carried out using the statistics software JMP IN 4.0 (SAS Institute) in which the following equation was applied:

Fe = Fluorescence intensity of protein without ligand binding
Fel = Fluorescence intensity of protein with ligand binding
Et = Total concentration of the protein
Totalligand = Total concentration of the ligand
K_d = Dissociation constant

Each titration was duplicated and two parallel correction experiments were carried out—corresponding to the increase in fluorescence intensity by the intrinsic fluorescence of the cap analogs and decrease in fluorescence due to eIF4E degradation and dilution effect. A series of ligand stocks 20 µM, 50 µM, 100 µM, 250 µM, 500 µM, 1 mM, 2 mM, and 5 mM were prepared in order to obtain smooth titration curves by minimizing titration errors.

Cell-free Translation Assay

To directly assess the level of translation, we employed the dual-luciferase bicistronic reporter construct pcDNA3-rLuc-POLIRES-fLuc (42), which is designed so that the translation of Renilla reniformis luciferase (rLUC) is strictly cap-dependent, whereas the translation of firefly luciferase (fLUC) proceeds via an IRES in a cap-independent manner. The template encodes 2 forms of luciferase each producing a product emitting light at a unique wavelength. The plasmid was linearized with XmaI, purified from an agarose gel, and 5’-capped bicistronic luciferase reporter mRNA was generated by in vitro transcription (mMESSAGE mMACHINE kit, Ambion) using T7 polymerase, according to the manufacturer’s instructions. Reporter mRNA (0.02 µg) was added to the reaction mixture (17 µL Retic lysate, 1 µL 1.25 mM L-methionine, 1.25 µL translation buffer per reaction), as recommended (Retic Lysate IVT™, Ambion). Renilla luciferase translation was carried out at high ionic strength conditions (High salt mix: 150 mM potassium chloride); whereas firefly luciferase – at medium ionic strength conditions (Low salt mix: 25 mM potassium chloride). Test compounds or nuclease-free water (control) were introduced into the reaction mixture. In vitro translation was carried out at 30°C for 1h. The reaction was stopped by chilling (5 min on ice) and samples were diluted with 100 µL of nuclease-free water. Renilla and firefly luciferase activity/abundance was quantified by luminometry using the Dual-Luciferase Reporter Assay System (Promega) exactly as described in the technical manual, in a Lumat LB 9507 Luminometer (EG&G Berthold). Luminescence was measured in relative light units. Reporter translation in the samples was compared to the control, set at 100. Assays were performed in triplicate; the mean ± SEM values were determined for each compound concentration. To calculate IC₅₀, the compounds were tested at concentrations ranging from 2 to 2000̣ µM. Data in triplicate corresponding to each concentration were plotted and the IC₅₀ values were determined directly from the plot.

Fish and embryos

Adult Zebrafish (Danio rerio) maintenance, husbandry and embryo collection were performed at the Zebrafish Research Core Facility (Arnold and Mabel Beckman Center for Transposon Research) using standard procedures (53) with approval from the Institutional Animal Care and Use Committee at the University of Minnesota. Freshly fertilized eggs were obtained through natural spawning. Eggs were staged for microinjections according to earlier precedents (54). Embryos were kept at 28,5°C (standard temperature) and staged according to Kimmel et al (55).

Microinjections

Microinjections, embryo observation and image acquisition were performed on the stage of a Stemi-2000 stereomicroscope (Carl Zeiss MicroImaging, Inc.) equipped with a PowerShot A630 digital camera (Canon) and PLI 100 Pico-injector (Warner Instruments, LLC) at room temperature. Eggs in chorions were lined up and restrained in the agar grooves (53) with their animal poles upwards. Reporter mRNA solution (3 nL) was injected into the egg cytoplasm in the geometric center of the yolk depression (15 min post fertilization) through a micropipette with a splinted sharp tip of 2–3 µm in diameter. Test compounds were diluted to the final concentration with HBSS (Gibco) containing 0.03% phenol red (Sigma). Compound-containing medium (5 nL) was microinjected into the cytoplasm during furrow progression at the 2-cell stage in the second round of microinjections, performed through the opening in chorions made by the first injection. Sham-treated eggs (negative control) were injected with 5 nL of the carrier (HBSS). Eggs were kept at 28,5°C.

Zebrafish Translation Assay

Reporter mRNA was diluted with HBSS (Gibco) containing 0.03% phenol red (Sigma) to a concentration of 200 ng/µL. Aliquots were stored at −80°C until needed. Luciferase reporters were expressed ectopically after RLUC-POLIRES-FLUC mRNA injection into the single-cell fertilized eggs. We employed reporter in the form of mRNA, because transcripts are evenly distributed among dividing zebrafish embryonic cells after injection into fertilized eggs, and are translated into protein within the first day of development (56). Eggs were first injected with the bicistronic luciferase mRNA (0.5 ng) and then with test compound in doses ranging from 5 to 25 pmols, with 7-MeGTP (reference compound), or with the carrier (HBSS). To standardize the level of luciferase expression across the injected eggs, we used a stringent microinjection protocol (one micro-needle per series, with no recalibration or readjustment of injection parameters). To ensure stable results, we normalized luciferase readouts to the number of eggs and utilized greater than 20 embryos (n ≥ 20) per sample. In each series of experiments, eggs were injected continuously (groove after groove), and for drug administration - in groups comprising equal numbers of eggs from all grooves. The second injection was performed through the opening in egg’s chorion made by the first injection. In assay validation using CHD (single dose - 0.1 pmol), the estimated error range for precision and accuracy of the Zebrafish translation assay was determined to be within 15%. Normally developing embryos were harvested at the early gastrula stage (28.5° C, 4.5h after injection). Firefly and Renilla luciferase activity were analyzed using the Luciferase Reporter Assay system (Promega). Luciferase assays were performed in triplicate according to the manufacturer’s instruction in a Lumat LB 9507 (EG&G Berthold).

Ectoderm explant extraction and cultivation

Ectoderm explants (“animal caps”) were surgically removed from the apical region of late blastulae (stages Sphere-Dome, 4–4.3 hpf) at 28,5°C, in sterile Modified Barth’s saline (MBS) using established procedures (57) with the following modifications: square blocks of superficial apical blastoderm comprised approximately 70 cells in 3 to 4 tiers with a portion of the outer enveloping layer; and explants were cultivated separately in sterile MBS without antibiotics on agar with a 3% methyl cellulose cushion at 28,5°C for up to 48h. Under these carefully controlled conditions, ectoderm blastula explants retain a spherical shape and contain dividing cells destined only to an ectodermal fate (57,58). In unfavorable conditions, explants decompose and largely dissipate within 24h in culture. Explants from injected embryos or intact embryos (as a positive control) were harvested at 2 time points corresponding to 1h and 24h post excision and analyzed by RT-PCR for lineage and stage specific transcripts.

Immunoblot Analysis

Whole-embryo lysates were prepared from normally developing embryos (20 per sample). Eggs were rinsed with HBSS, dispersed in Lysis Buffer (50mM Tris-HCl (pH 7.5), 250mM NaCl, 50mM NaF, 5mM EDTA, 0.2% NP-40) with protease inhibitor cocktail (Complete MINI, Roche) on ice, kept on a shaker for 20 min at 4°C and centrifuged at 16,000g for 15 min with the supernatant retained. Protein concentration in the supernatant was measured using the BCA™ protein assay kit (Pierce). Twenty micrograms of protein was subjected to 8% SDS-PAGE under reducing conditions and transferred to nitrocellulose. Immuno-detection of proteins was carried out using primary antibodies (Rat anti-HA, 1:2000) followed by horseradish peroxidase-conjugated secondary antibodies (anti-rat IgG, 1:500), and incubation with chemiluminescence substrate (Pierce).

RT-PCR

RNA was extracted using TRI-reagent (Sigma). The RNA samples were treated with DNAase using Turbo DNA-free kit (Ambion) according to the manufacturer’s directions and converted to cDNA using the TaqMan reverse transriptase kit (Roche). Real-time PCR was performed using the Roche Light-Cycler with SYBR Green dye according to the manufacturer’s protocol (Roche). Amplified fragments were resolved on 1% agarose gels and sized according to standards. The sequences of the primers and the size of amplificate (bp) are provided below for each gene analyzed.

Oligonucleotide primers used for RT-PCR

Marker	Forward primer (5’-3’)	Reverse primer (3’-5’)	Size
cyt1	GGGGCGCAGGATTTGGTG	GTTCTTCTTGGTGCTCCTCCTCGA	499
gta3	CCTGGGCCTCTGTCCGTCTATCCT	CTCTTACACATTTGACGCCCCGGT	427
gsc	ACGGCAGCTGGCGATTTTA	CTTTATGGGACTGCAGCCGTG	404
ntl	GCCCGAGCTGCGTCTACATCC	GGAGTCAAGTTACCCCGGGGA	449
myoD	CTTAAACCCGACGAGCATCAC	TACTACCTAAAATACCCGGG	401

Amplified fragments were resolved on 1% acrylamide gels and visualized by BrdU staining. Gel images were acquired with a UVP BioDoc-it™ System.

HINT activity in the Rabbit reticulocyte and Zebrafish gastrula lysates

Activity of HINT was quantified by individually titrating 20 µL of Rabbit reticulocyte lysate and 50 µL of zebrafish gastrula lysate (both in 600 µL HEPES) against 30 µL of the fluorogenic substrate, AMP-Tryptamine Phosphoramidate (34). The observed increase in fluorescence intensity was plotted against time, and the slope of the rectilinear plots substituted in the following equation to obtain the value of the enzyme concentrations:

v = Slope/(F_pro − F_sub) = k_cat·[E]_t, where
v = velocity
F_pro = Fluorescence due to product (tryptamine)
F_sub = Fluorescence due to substrate (AMP-Tryptamine phosphoramidate)
[E]_t = Total enzyme concentration
k_cat for AMP-Tryptamine phosphoramidate = 2.1 ± 0.1 s⁻¹

The Rabbit reticulocyte and zebrafish embryo lysates contain 23 and 1.95 ng of the enzyme, respectively.

Liquid chromatography/Mass spectrometry (LC-MS/MS) of 4Ei-1 Metabolism

Rabbit reticulocyte lysate (Ambion) and zebrafish embryo lysates were used in studying 4Ei-1 conversion. In the zebrafish lysate preparation, 300 early pre-gastrula eggs were collected, washed with “embryo medium”, dried and mechanically homogenized. The crude homogenate was kept rotating for 30 min at 4°C, centrifuged (15 min, 12,000 rpm at 4 °C) with supernatant fluid retained and designated embryo lysate. All standard solutions and samples were kept on ice before analysis by LC-MS/MS.

Reagents

N, N-dimethylhexylamine (DMHA), tetrabutyl ammonium acetate (TBAA), and ammonium acetate were purchased from Sigma. Solvents used for LC analysis were HPLC grade. All solutions for instrument analyses were filtered through a 0.22 um membrane filter, and degassed prior to use. Microcon microcentrifuge filter device YM-10 (molecular weight cutoff - 10,000) was purchased from Millipore. Chromatographic separation was achieved using a narrow bore Eclipse XDB-C18 column (2.1 mm × 50 mm, 1.8 um, Agilent Technologies) eluted at a flow rate of 0.125 mL/min. An injection volume of 10 µL was used for standards and lysates. The mobile phase was composed of 50% Solvent A (10 mM ammonium acetate, pH 6.65) and 50% solvent B (methanol). The running time for each sample was 5 minutes. Sample temperature was maintained at 4 °C with a thermostat-controlled sample compartment. A TSQ quantum classic LC-ESI-MS/MS system (Agilent 1200 LC) was employed for all analyses. The mass spectrometer was operated in negative ion mode with nitrogen as a nebulizing and drying gas. 7-BnGMP and 4Ei-1 were directed to the detector for infusion. ESI source parameters and MS/MS parameters were optimized for maximum sensitivity. Negative ion ESI and selective multiple reaction monitoring (MRM) mode was used in all analyses. The [M-H]⁻ ion of 7-BnGMP (m/z 452.10) was isolated and subjected to collision-induced dissociation (CID) to give the product ion (m/z, 79.19, collision energy 52 V) for quantification. 4Ei-1 was analyzed analogously to give a parent ion (m/z, 594.00) and product ion (m/z, 148.84, collision energy 52 V). The standard curves were obtained in respective matrix with a known concentration of 7-BnGMP and 4Ei-1 ranging from 10 to10,000 ng/mL. Quantification of target compounds was carried out with XCalibur software (Thermo Scientific). Rabbit and zebrafish lysates were diluted 64-fold with 10 mM ammonium acetate (pH 6.65) and methanol in a ratio of 1:1, followed by heating at 50 °C for 30 minutes. Precipitates were filtered using microcon YM-10 (Millipore). The filtrate served as matrix for preparing standard solutions. 10 µL of a known concentration (100, 500, 1000, 5000, 10000, 50000, 100000 ng/mL) of both targets dissolved in methanol was added to a HPLC sample vial. Methanol was removed in a SC210A SpeedVac^® concentrator (Thermo Scientific). Lysate matrix (100 µL) was added to each concentrated vial to make the corresponding standard solutions (10, 50, 100, 500, 1000, 5000, 10000 ng/mL). For sample preparation, 4Ei-1 (5 µL) at 150 µM (final concentration) was incubated with lysates (17 µL) at 30 °C for 1 hour, followed by similar treatment (with or without heating at 50 °C for 30 minutes) as the standards. Controls were prepared in the same way except that dH₂O (Sigma) substituted for 4Ei-1.

Supplementary Material

1_si_001

Click here to view.^{(938K, pdf)}

ACKNOWLEDGEMENTS

We would like to thank C. Sagerström for advice regarding the explant experiments. This work was supported by the National Institute of Health grants: 1R01 HL076779 and 1R21RR024398 (P.B.B and A.O.B.), 5U01-CA091220 (V.A.P.), University of Minnesota AHC Faculty Development Grant (C.R.W.).

Footnotes

Supporting Information Available: This material is available free of charge via the Internet.

REFERENCES

1. Sahlgren C, Gustafsson MV, Jin S, Poellinger L, Lendahl U. Notch signaling mediates hypoxia-induced tumor cell migration and invasion. Proc Natl Acad Sci U S A. 2008;105:6392–6397. [PubMed]

2. Korpal M, Lee ES, Hu G, Kang Y. The miR-200 family inhibits epithelial-mesenchymal transition and cancer cell migration by direct targeting of E-cadherin transcriptional repressors ZEB1 and ZEB2. J Biol Chem. 2008;283:14910–14914. [PubMed]

3. Gavert N, Ben-Ze'ev A. Epithelial-mesenchymal transition and the invasive potential of tumors. Trends Mol Med. 2008;14:199–209. [PubMed]

4. Park SM, Gaur AB, Lengyel E, Peter ME. The miR-200 family determines the epithelial phenotype of cancer cells by targeting the E-cadherin repressors ZEB1 and ZEB2. Genes Dev. 2008;22:894–907. [PubMed]

5. Gregory PA, Bert AG, Paterson EL, Barry SC, Tsykin A, Farshid G, Vadas MA, Khew-Goodall Y, Goodall GJ. The miR-200 family and miR-205 regulate epithelial to mesenchymal transition by targeting ZEB1 and SIP1. Nat Cell Biol. 2008;10:593–601. [PubMed]

6. Gingras AC, Raught B, Sonenberg N. eIF4 initiation factors: effectors of mRNA recruitment to ribosomes and regulators of translation. Annu Rev Biochem. 1999;68:913–963. [PubMed]

7. Gallie DR. Protein-protein interactions required during translation. Plant Mol Biol. 2002;50:949–970. [PubMed]

8. Pestova TV, Hellen CU. Functions of eukaryotic factors in initiation of translation. Cold Spring Harb Symp Quant Biol. 2001;66:389–396. [PubMed]

9. Browing KS. The plant translational apparatus. Plant Mol Biol. 1996;32

10. Gallie DR. A tale of two termini: a functional interaction between the termini of an mRNA is a prerequisite for efficient translation initiation. Gene. 1998;216:1–11. [PubMed]

11. Gingras AC, Raught B, Sonenberg N. mTOR signaling to translation. Curr Top Microbiol Immunol. 2004;279:169–197. [PubMed]

12. Polunovsky VA, Bitterman PB. The cap-dependent translation apparatus integrates and amplifies cancer pathways. RNA Biol. 2006;3:10–17. [PubMed]

13. Raught B, Gingras AC. eIF4E activity is regulated at multiple levels. Int J Biochem Cell Biol. 1999;31:43–57. [PubMed]

14. Sonenberg N, Gingras AC. The mRNA 5' cap-binding protein eIF4E and control of cell growth. Curr Opin Cell Biol. 1998;10:268–275. [PubMed]

15. Richter JD, Theurkauf WE. Development. The message is in the translation. Science. 2001;293:60–62. [PubMed]

16. Vasudevan S, Seli E, Steitz JA. Metazoan oocyte and early embryo development program: a progression through translation regulatory cascades. Genes Dev. 2006;20:138–146. [PubMed]

17. Avdulov S, Li S, Michalek V, Burrichter D, Peterson M, Perlman DM, Manivel JC, Sonenberg N, Yee D, Bitterman PB, et al. Activation of translation complex eIF4F is essential for the genesis and maintenance of the malignant phenotype in human mammary epithelial cells. Cancer Cell. 2004;5:553–563. [PubMed]

18. Larsson O, Li S, Issaenko OA, Avdulov S, Peterson M, Smith K, Bitterman PB, Polunovsky VA. Eukaryotic translation initiation factor 4E induced progression of primary human mammary epithelial cells along the cancer pathway is associated with targeted translational deregulation of oncogenic drivers and inhibitors. Cancer Res. 2007;67:6814–6824. [PubMed]

19. Ruggero D, Montanaro L, Ma L, Xu W, Londei P, Cordon-Cardo C, Pandolfi PP. The translation factor eIF-4E promotes tumor formation and cooperates with c-Myc in lymphomagenesis. Nat Med. 2004;10:484–486. [PubMed]

20. Zimmer SG, DeBenedetti A, Graff JR. Translational control of malignancy: the mRNA cap-binding protein, eIF-4E, as a central regulator of tumor formation, growth, invasion and metastasis. Anticancer Res. 2000;20:1343–1351. [PubMed]

21. Graff JR, Konicek BW, Carter JH, Marcusson EG. Targeting the eukaryotic translation initiation factor 4E for cancer therapy. Cancer Res. 2008;68:631–634. [PubMed]

22. Moerke NJ, Aktas H, Chen H, Cantel S, Reibarkh MY, Fahmy A, Gross JD, Degterev A, Yuan J, Chorev M, et al. Small-molecule inhibition of the interaction between the translation initiation factors eIF4E and eIF4G. Cell. 2007;128:257–267. [PubMed]

23. Graff JR, Konicek BW, Vincent TM, Lynch RL, Monteith D, Weir SN, Schwier P, Capen A, Goode RL, Dowless MS, et al. Therapeutic suppression of translation initiation factor eIF4E expression reduces tumor growth without toxicity. J Clin Invest. 2007;117:2638–2648. [PMC free article] [PubMed]

24. Ghosh P, Park C, Peterson MS, Bitterman PB, Polunovsky VA, Wagner CR. Synthesis and evaluation of potential inhibitors of eIF4E cap binding to 7-methyl GTP. Bioorg Med Chem Lett. 2005;15:2177–2180. [PubMed]

25. Grudzien E, Stepinski J, Jankowska-Anyszka M, Stolarski R, Darzynkiewicz E, Rhoads RE. Novel cap analogs for in vitro synthesis of mRNAs with high translational efficiency. Rna. 2004;10:1479–1487. [PubMed]

26. Wagner CR, Iyer VV, McIntee EJ. Pronucleotides: toward the in vivo delivery of antiviral and anticancer nucleotides. Med Res Rev. 2000;20:417–451. [PubMed]

27. Meier C. Pronucleotides: Recent advances in the design of efficient tools for the delivery of biologically active nucleoside monophosphates. Syn Lett. 1998;XXX:233–242.

28. Wagner CR, Chang SL, Griesgraber GW, Song H, McIntee EJ, Zimmerman CL. Antiviral nucleoside drug delivery via amino acid phosphoramidates. Nucleosides Nucleotides. 1999;18:913–919. [PubMed]

29. Chang S, Griesgraber GW, Southern PJ, Wagner CR. Amino acid phosphoramidate monoesters of 3'-azido-3'-deoxythymidine: relationship between antiviral potency and intracellular metabolism. J Med Chem. 2001;44:223–231. [PubMed]

30. McGuigan C, Harris SA, Daluge SM, Gudmundsson KS, McLean EW, Burnette TC, Marr H, Hazen R, Condreay LD, Johnson L, et al. Application of phosphoramidate pronucleotide technology to abacavir leads to a significant enhancement of antiviral potency. J Med Chem. 2005;48:3504–3515. [PubMed]

31. Song H, Griesgraber GW, Wagner CR, Zimmerman CL. Pharmacokinetics of amino acid phosphoramidate monoesters of zidovudine in rats. Antimicrob Agents Chemother. 2002;46:1357–1363. [PMC free article] [PubMed]

32. Song H, Johns R, Griesgraber GW, Wagner CR, Zimmerman CL. Disposition and oral bioavailability in rats of an antiviral and antitumor amino acid phosphoramidate prodrug of AZT-monophosphate. Pharm Res. 2003;20:448–451. [PubMed]

33. Chou TF, Bieganowski P, Shilinski K, Cheng J, Brenner C, Wagner CR. 31P NMR and genetic analysis establish hinT as the only Escherchia coli purine nucleoside phosphoramidase and as essential for growth under high salt conditions. J Biol Chem. 2005;280:15356–15361. [PMC free article] [PubMed]

34. Chou TF, Baraniak J, Kaczmarek R, Zhou X, Cheng J, Ghosh B, Wagner CR. Phosphoramidate pronucleotides: a comparison of the phosphoramidase substrate specificity of human and Escherichia coli histidine triad nucleotide binding proteins. Mol Pharm. 2007;4:208–217. [PubMed]

35. Kim J, Chou TF, Griesgraber GW, Wagner CR. Direct measurement of nucleoside monophosphate delivery from a phosphoramidate pronucleotide by stable isotope labeling and LC-ESI(−)-MS/MS. Mol Pharm. 2004;1:102–111. [PubMed]

36. Yuan BZ, Jefferson AM, Popescu NC, Reynolds SH. Aberrant gene expression in human non small cell lung carcinoma cells exposed to demethylating agent 5-aza-2'-deoxycytidine. Neoplasia. 2004;6:412–419. [PMC free article] [PubMed]

37. Mani SA, Guo W, Liao MJ, Eaton EN, Ayyanan A, Zhou AY, Brooks M, Reinhard F, Zhang CC, Shipitsin M, et al. The epithelial-mesenchymal transition generates cells with properties of stem cells. Cell. 2008;133:704–715. [PMC free article] [PubMed]

38. Burk U, Schubert J, Wellner U, Schmalhofer O, Vincan E, Spaderna S, Brabletz T. A reciprocal repression between ZEB1 and members of the miR-200 family promotes EMT and invasion in cancer cells. EMBO Rep. 2008;9:582–589. [PMC free article] [PubMed]

39. Cai A, Jankowska-Anyszka M, Centers A, Chlebicka L, Stepinski J, Stolarski R, Darzynkiewicz E, Rhoads RE. Quantitative assessment of mRNA cap analogues as inhibitors of in vitro translation. Biochemistry. 1999;38:8538–8547. [PubMed]

40. Niedzwiecka A, Marcotrigiano J, Stepinski J, Jankowska-Anyszka M, Wyslouch-Cieszynska A, Dadlez M, Gingras AC, Mak P, Darzynkiewicz E, Sonenberg N, et al. Biophysical studies of eIF4E cap-binding protein: recognition of mRNA 5' cap structure and synthetic fragments of eIF4G and 4E-BP1 proteins. J Mol Biol. 2002;319:615–635. [PubMed]

41. Brown CJ, McNae I, Fischer PM, Walkinshaw MD. Crystallographic and mass spectrometric characterisation of eIF4E with N7-alkylated cap derivatives. J Mol Biol. 2007;372:7–15. [PubMed]

42. Poulin F, Gingras AC, Olsen H, Chevalier S, Sonenberg N. 4E-BP3, a new member of the eukaryotic initiation factor 4E-binding protein family. J Biol Chem. 1998;273:14002–14007. [PubMed]

43. Stern HM, Zon LI. Cancer genetics and drug discovery in the zebrafish. Nat Rev Cancer. 2003;3:533–539. [PubMed]

44. Zon LI, Peterson RT. In vivo drug discovery in the zebrafish. Nat Rev Drug Discov. 2005;4:35–44. [PubMed]

45. Thiery JP, Sleeman JP. Complex networks orchestrate epithelial-mesenchymal transitions. Nat Rev Mol Cell Biol. 2006;7:131–142. [PubMed]

46. Klein PS, Melton DA. Induction of mesoderm in Xenopus laevis embryos by translation initiation factor 4E. Science. 1994;265:803–806. [PubMed]

47. Goossens K, Van Soom A, Van Poucke M, Vandaele L, Vandesompele J, Van Zeveren A, Peelman LJ. Identification and expression analysis of genes associated with bovine blastocyst formation. BMC Dev Biol. 2007;7:64. [PMC free article] [PubMed]

48. Zuberek J, Kubacka D, Jablonowska A, Jemielity J, Stepinski J, Sonenberg N, Darzynkiewicz E. Weak binding affinity of human 4EHP for mRNA cap analogs. Rna. 2007;13:691–697. [PubMed]

49. Izaurralde E, Stepinski J, Darzynkiewicz E, Mattaj IW. A cap binding protein that may mediate nuclear export of RNA polymerase II-transcribed RNAs. J Cell Biol. 1992;118:1287–1295. [PMC free article] [PubMed]

50. Worch R, Niedzwiecka A, Stepinski J, Mazza C, Jankowska-Anyszka M, Darzynkiewicz E, Cusack S, Stolarski R. Specificity of recognition of mRNA 5' cap by human nuclear cap-binding complex. Rna. 2005;11:1355–1363. [PubMed]

51. Peinado H, Olmeda D, Cano A. Snail, Zeb and bHLH factors in tumour progression: an alliance against the epithelial phenotype? Nat Rev Cancer. 2007;7:415–428. [PubMed]

52. Radisky DC, Kenny PA, Bissell MJ. Fibrosis and cancer: do myofibroblasts come also from epithelial cells via EMT? J Cell Biochem. 2007;101:830–839. [PMC free article] [PubMed]

53. Westerfield M. The Zebrafish book. 3d edition 1995.

54. Benyumov AO, Dubrovskaya TA, Barmintsev VA, Shan L. Duration of the first mitotic cycles and staging of embryogenesis of danio rerio. Russian J. od Dev. Biol. 1995;26:132–138.

55. Kimmel CB, Ballard WW, Kimmel SR, Ullmann B, Schilling TF. Stages of embryonic development of the zebrafish. Dev Dyn. 1995;203:253–310. [PubMed]

56. Wang X, Wan H, Korzh V, Gong Z. Use of an IRES bicistronic construct to trace expression of exogenously introduced mRNA in zebrafish embryos. Biotechniques. 2000;29:814–816. 818, 820. [PubMed]

57. Grinblat Y, Lane ME, Sagerstrom C, Sive H. Analysis of zebrafish development using explant culture assays. Methods Cell Biol. 1999;59:127–156. [PubMed]

58. Sagerstrom CG, Grinbalt Y, Sive H. Anteroposterior patterning in the zebrafish, Danio rerio: an explant assay reveals inductive and suppressive cell interactions. Development. 1996;122:1873–1883. [PubMed]

59. Gritsman K, Zhang J, Cheng S, Heckscher E, Talbot WS, Schier AF. The EGF-CFC protein one-eyed pinhead is essential for nodal signaling. Cell. 1999;97:121–132. [PubMed]

60. Sagerstrom CG, Gammill LS, Veale R, Sive H. Specification of the enveloping layer and lack of autoneuralization in zebrafish embryonic explants. Dev Dyn. 2005;232:85–97. [PubMed]

61. Marcotrigiano J, Gingras AC, Sonenberg N, Burley SK. Cocrystal structure of the messenger RNA 5' cap-binding protein (eIF4E) bound to 7-methyl-GDP. Cell. 1997;89:951–961. [PubMed]

62. Joshi B, Lee K, Maeder DL, Jagus R. Phylogenetic analysis of eIF4E-family members. BMC Evol Biol. 2005;5:48. [PMC free article] [PubMed]