Identification of EGFR and RAS Inhibitors using Caenorhabditis elegans

Summary

The genetically tractable nematode Caenorhabditis elegans can be used as a simple and inexpensive model for drug discovery. Described here is a protocol to identify anticancer therapeutics that inhibit the downstream signaling of RAS and EGFR proteins.

Abstract

The changes in the plasma membrane localization of the epidermal growth factor receptor (EGFR) and its downstream effector RAS have been implicated in several diseases including cancer. The free-living nematode C. elegans possesses an evolutionary and functionally conserved EGFR-RAS-ERK MAP signal cascade which is central for the development of the vulva. Gain of function mutations in RAS homolog LET-60 and EGFR homolog LET-23 induce the generation of visible nonfunctional ectopic pseudovulva along the ventral body wall of these worms. Previously, the multivulval (Muv) phenotype in these worms has been shown to be inhibited by small chemical molecules. Here we describe a protocol for using the worm in a liquid-based assay to identify inhibitors that abolish the activities of EGFR and RAS proteins. Using this assay, we show R-fendiline, an indirect inhibitor of K-RAS, suppresses the Muv phenotype expressed in the let-60(n1046) and let-23(sa62) mutant worms. The assay is simple, inexpensive, is not time consuming to setup, and can be used as an initial platform for the discovery of anticancer therapeutics.

Introduction

The cellular pathways that regulate developmental events within organisms are highly conserved among all metazoans. One such pathway is the EGFR-RAS-ERK mitogen activated protein kinase (MAPK) signaling cascade which is a critical pathway that governs cell proliferation, differentiation, migration and survival^1,2. Defects in this signaling pathway can lead to pathological or disease states such as cancer. The epidermal growth factor receptor (EGFR) has shown to be highly expressed in human tumors, including 50% of oral squamous cell carcinomas, and contributes to the development of malignant tumors^3,4,5. Whereas mutations in the three RAS isoforms H-, K- and N-RAS are major drivers for malignant transformation in multiple human cancers. Amongst these three RAS isoforms, oncogenic mutations in K-RAS are most prevalent^6,7,8. For EGFR and RAS to function, they must localize to the plasma membrane (PM). Preventing the localization of these molecules to the PM can completely abrogate the biological activity of this signal pathway^9,10. Hence the inhibition of the localization of these proteins to the PM is a therapeutic strategy to block the downstream signaling and the resulting adverse outcomes. Using a high-content screening assay, fendiline, an L-type calcium channel blocker, was identified as an inhibitor of K-RAS activity¹¹. Nanoclustering of K-RAS to the inner leaflet of the PM is significantly reduced in the presence of fendiline. Furthermore, K-RAS is redistributed from the plasma membrane to the endoplasmic reticulum (ER), Golgi apparatus, endosomes, and cytosol. More importantly, the proliferation of pancreatic, colon, lung, and endometrial cancer cell lines expressing oncogenic mutant K-RAS is blocked by the inhibition of downstream signaling by fendiline¹¹. These data suggest fendiline functions as a specific K-RAS anticancer therapeutic that causes the mis-localization of the RAS protein to the PM.

The nematode Caenorhabditis elegans has been extensively studied in the context of development. Many of the signal pathways that govern development in the worm are evolutionary and functionally conserved. For example, the EGFR mediated activation of RAS and the subsequent activation of the ERK MAPK signal cascade is conserved in the worm¹². The cascade is represented by the following proteins: LET-23 > LET-60 > LIN-45 > MEK-2 > MPK-1. LET-60 is homologous to RAS, while LET-23 is homologous to EGFR. In the worm, this pathway regulates the development of the vulva¹³. The vulva is an epithelial aperture on the ventral body wall of the worm that allows fertilized eggs to be laid. The formation of the vulva in the worm is dependent on the exposure of the vulval precursor cells (VPC) to a gradient of activation of the EGFR-RAS-MAPK signal cascade. During the normal development, the proximal VPCs receive strong signals from the gonadal anchor cells to differentiate into 1° and 2° cell fates which give rise to a functional vulva¹². Whereas distal VPCs differentiate into 3° cell fates that fuse to the hypodermal syncytium and do not form vulva due to depleted signaling. In the absence of signaling, all VPCs differentiate into 3° cell fates resulting in the formation of no vulva. However, constitutive signaling leads to the formation one or more non-functional vulva due to the induction of all VPCs to assume 1° and 2° cell fates.

Mutations that cause defective or excessive vulval induction have been identified for many of the genes that encode for proteins representing this pathway. Defective vulval induction results in a vulvaless (Vul) phenotype, while excessive vulval induction results in a multivulva (Muv) phenotype that is represented by the development of numerous nonfunctional ectopic pseudovulvae throughout the ventral body wall. The Muv phenotype expressed by the let-60(n1046) strain is due to a gain of function mutation in RAS, while in the let-23(sa62) strain it is due an activating mutation in EGFR^14,15. The strong Muv phenotype in these mutant strains has been shown to be perturbed by pharmacological interventions as demonstrated by the treatment of let-60(n1046) worms with the MEK-1 inhibitor U0126^16,17. Interestingly, we have shown that R-fendiline and inhibitors that affect sphingomyelin metabolism suppress the Muv phenotype in the worm¹⁸. To demonstrate these inhibitors block let-60 signaling at the level of RAS, the lin-1 null strain has been utilized¹⁷. Lin-1 is an Ets-like inhibitory transcription factor that functions as a repressor in the development of the vulva¹⁹. Strong reversion of the Muv phenotype in let-60(n1046) worms and no effect on lin-1 null worms suggest that these inhibitions occur at the level of RAS.

In this protocol, we demonstrate the use of C. elegans as a model to identify inhibitors of RAS and EGFR proteins. Using a liquid-based assay, we demonstrate the inhibitory effects of R-fendiline by suppressing the Muv phenotypes in the let-60(n1046) and let-23(sa62) mutant strains of C. elegans. This assay validates the use of C. elegans as a tool in the initial phase of drug discovery for anticancer therapeutics.

Protocol

1. Nematode growth medium (NGM) plate preparation

1. Add 2.5 g of peptone and 3 g of NaCl to 970 mL of deionized water contained in a 2 L Erlenmeyer flask. Stir contents using a magnetic stir bar. Thereafter, add 20 g of agar to the flask. Autoclave the contents of the flask at 121 °C and a pressure of 15 lb/in² for 30 min. After sterilization, place the flask on a stir plate and allow the medium to cool until the temperature reaches 50 °C.
2. To prepare the NGM plates add the following reagents to the cooled medium: 25 mL of 1 M potassium phosphate buffer (pH = 6.0), 1 mL of 1 M MgSO₄, 1 mL of 1 M CaCl₂, 1 mL of (5 mg/mL in 95% ethanol) cholesterol, 1 mL of (10% v/w in ethanol) nystatin, and 1 mL of 25 mg/mL streptomycin.
3. Under a laminar flow, pour the cooled medium into 60 mm x 15 mm sterile Petri dishes. Let the plates solidify for 2 h. These plates can be kept for 1 month at 4 °C.

2. Propagation of C. elegans

1. Spot 100 μL of overnight grown E. coli OP50 onto the center of each NGM plate and allow the plates to dry for 24 h in a laminar hood. Subsequently, the plates can be stored in polystyrene container.
   NOTE: The E. coli OP50 culture is grown in Luria-Bertani (LB) media at 37 °C media prior to seeding of the plates. The culture can be stored at 4 °C for 1–2 months and used for periodic seeding of plates.
2. Using a sterile worm pick, gather 10–12 gravid adult worms from a previously grown plate under a dissecting microscope. Transfer these worms to a fresh NGM plate seeded with E. coli OP50 and incubate the plate for 24 h at 20 °C.
3. After 24 h, using a sterile worm-pick remove adult worms from the plates. Incubate the plates at 20 °C for ~ 3 days. Embryos will develop into gravid adult worms.

3. Preparation of a synchronous C. elegans culture

1. Collect gravid adult worms into a 15 mL conical tube by washing 2–4 plates with M9W.
   1. Preparation of M9W: Dissolve 5 g of NaCl, 6 g of Na₂HPO₄, and 3 g of KH₂PO₄ in deionized water to a final volume of 1 L. Autoclave the solution at 121 °C at a pressure of 15 lb/in² for 30 min. Add 1 mL of 1 M MgSO₄ to the cooled solution.
2. Pellet the worms by centrifuging the tube at 450 x g for 1 min. Decant the supernatant without disturbing the worm pellet.
3. Prepare worm lysis solution by combining 400 μL of 8.25% sodium hypochlorite (household bleach) and 100 μL of 5 N NaOH. Add this solution to the worm pellet and flick the tube to mix contents of the tube. Prevent overbleaching of the embryos in the tube by observing the lysis of the worms under a dissecting microscope.
4. Add 10 mL of M9W to the conical tube to dilute the lysis mix when 70% of the adult worms have lysed.
5. Centrifuge the tube at 450 x g for 1 min. Replace the supernatant with 10 mL of M9W.
6. Repeat step 3.5 two times.
7. After completing the washing steps, add 3–5 mL of M9W to resuspend the egg pellet. Rotate the tube overnight at a speed of 18 rpm on a tube rotator at room temperature (RT).
8. After overnight incubation, remove the tube from the rotator, pellet the L1 larvae by centrifuging the tube at 450 x g for 1 min. Aspirate the M9W until 250 μL is left in the tube.
9. Shake the tube to resuspend the larvae. Add three 5 μL drops containing the larvae onto a Petri dish lid. Using a dissecting microscope count the number of worms in each drop and determine the number of worms in 1 μL.

4. Preparation of drug assays

NOTE: The steps in this assay are shown in Figure 1.

1. Grow 30 mL of E. coli OP50 in a 50 mL conical tube at 37 °C overnight in an orbital shaker at 150 rpm.
2. Spin overnight grown E. coli OP50 culture at 4,000 x g for 10 min to pellet the cells. Remove the supernatant and resuspend the pellet in 3 mL of M9W to concentrate the culture.
3. Prior to preparing the working solutions for the drug assay, add 0.1 mL of (5 mg/mL in 95% ethanol) cholesterol into 100 mL of M9W.
4. Prepare working solutions of each experimental drug by diluting each drug plus vehicle (dimethyl sulfoxide; DMSO) in 4.8 mL M9W supplemented with cholesterol to obtain the appropriate concentrations. Dissolve DMSO in 4.8 mL of M9W supplemented with cholesterol to prepare the vehicle control.
5. Add 200 μL of concentrated E. coli OP50 culture to each tube containing the vehicle control or drugs and vortex tubes to ensure they are mixed.
6. To perform the experiment, add 2 mL of each working drug solution or vehicle control to each well in a 12-well tissue culture plate. Test each drug concentration and vehicle control in duplicate.
7. Add ~100 L1 larvae per well (see synchronous C. elegans culture steps) using a sterile micropipette. Limit the volume of worms to 10 μL.
8. Incubate the plates at 20 °C.
9. Supplement wells with 50 μL of 10x concentrated E. coli OP50 on day 3 of the assay if needed.

5. Agarose pad preparation for microscopy

1. Prepare a 2% w/v solution of agarose by dissolving 0.1 g of agarose in 5 mL of deionized water. Heat the contents in a microwave to dissolve the agarose. 20 slides can be prepared from 5 mL of agarose solution.
2. Place strips of lab tape along two glass slides. The tape will act as spacers limiting the thickness of the agarose pads. Thereafter, place a third clean glass slide between the taped slides.
3. To make an agarose pad, spot 100 μL of molten agarose onto the center of the clean slide. Place another clean glass slide across and on top of the agarose and gently press down to form a pad. Remove the top slide when the pad has solidified.

6. Observation of the Muv phenotype in the let-60, let-23 and lin-1 strains

NOTE: Only candidate drugs that suppress the Muv phenotypes in let-23 and let-60 strains will be assayed using the lin-1 strain to determine if the inhibition occurs at the level of RAS or EGFR.

1. When the appropriate stage of the life cycle is reached (3 days for the let-60 and let-23 strains, 5 days for the lin-1 strain), remove the plates from the incubator. and collect the worms in 15 mL conical tubes using a pipette.
2. Centrifuge the tubes at 450 x g for 1 min. Remove M9W without disturbing the worm pellet and add 5 mL of fresh M9W.
3. Repeat step 6.2 two times.
4. Without disturbing the worm pellet, remove all remaining M9W by aspiration. Thereafter, add 500 μL of 2 mM sodium azide or 2 mM tetramisole hydrochloride. Allow tubes to incubate at RT for 15 min. 2 mM sodium azide or 2 mM tetramisole hydrochloride will anesthetize the worms for imaging.
   CAUTION: Make sure to wear personal protective equipment (PPE) when handling sodium azide. The solution must be prepared under a chemical hood.
5. Add 10 μL of the anesthetized worm suspension onto the center of an agarose pad. Place a #1.5 coverslip gently over the worm suspension. If needed, fix the coverslip with some nail polish to prevent drying out.
6. Use a DIC microscope to observe the let-60(n1046), let-23(sa62) and lin-1(sy254) strains. Image the worms at the 10x and 20x magnifications.
7. For let-60(n1046) and let-23(sa62), score the adult worms based on the presence or absence of the Muv phenotype. While for the lin-1 strain, count the number VPCs that adopted 1° or 2° cell fates on the ventral side of L4 larvae using a high resolution DIC microscope.
8. After scoring the micrographs for each strain and drug treatment, perform Student’s t-test to determine the statistical differences between the treatments.

Results

We first demonstrate that R-fendiline is able to suppress the Muv phenotype in the let-60(n1046) mutant strain compared to the DMSO treated worms. Our data shows that R-fendiline is able to block the Muv phenotype in the let-60(n1046) in a dose-dependent manner (Figure 2A,B). However, non-reversal of the Muv phenotype was observed in the lin-1 null mutant strain in response to increasing concentrations of R-fendiline (Figure 2B). The data suggests that R-fendiline blocks activated let-60 signaling at the level of RAS in C. elegans. Similarly, we observed the Muv phenotype was significantly reduced in the let-23(sa62) strain in response to 3, 10 and 30 µM R-fendiline treatment relative to the DMSO treated worms (Figure 2C,D). In all experiments, Students t-test was used to determine the statistical significance.

Figure 1: Flowchart representing the steps involved in preparing the drug assays using let-60(n1046), let-23(sa62) and lin-1(sy254) strains. Please click here to view a larger version of this figure.

Figure 2: R-fendiline alters let-60 and let-23 function in C. elegans in a dose-dependent manner. (A) Representative images of let-60(n1046) worms in the presence of vehicle (DMSO) or 30 μM R-fendiline. (B) Quantification of Muv phenotype in let-60(n1046) worms treated in the presence of DMSO, 3, 10 and 30 µM R-fendiline, or 30 μM U0126. (C) Representative images of let-23(sa62) worms in the presence of vehicle (DMSO) or 30 µM R-fendiline. (D) Quantification of Muv phenotype in let-23(sa62) worms treated in the presence of DMSO, 3, 10 and 30 µM R- fendiline, or 30 μM U0126. In all images the pseudovulva are indicated by white arrows and normal vulva by white asterisks. A total of 60 worms were imaged for each treatment. The experiment was repeated 3 times. (*** P<0.001 and ** P<0.01 were considered significant) Please click here to view a larger version of this figure.

Discussion

The assays we describe using the worm are simple and inexpensive to identify inhibitors of EGFR and RAS function. C. elegans is an attractive model for drug discovery because it is easy to grow in the lab due to the short life cycle (3 days at 20 °C) and the ability to generate large numbers of larvae. More importantly, the EGFR-RAS-ERK MAPK pathway is evolutionarily and functionally conserved with mammals providing a genetically tractable system to analyze the effects of EGFR and RAS inhibitors. Further, the transparent nature of the worms enables an investigator to visualize distinct structures and the localization of Green fluorescent Protein (GFP) or other fluorophore fused to proteins of interest by DIC and fluorescent microscopy.

The protocols we used to propagation and maintain the various C. elegans used in this study were previously established^20,21. However, in the preparation of NG plates we incorporated streptomycin and nystatin to prevent bacterial and fungal contamination. The addition of these antimicrobial agents did not impede the development of the worms and the induction of the Muv phenotype.

There are several advantages for obtaining L1 larvae by the worm synchronization protocol. Many larvae can be obtained from 2 or more plates containing gravid adults and the larvae collected are all age synchronized. This ensures development of the worms is consistent within the population. Some mutant strains displaying the Muv phenotype are poor egg layers resulting in low yields of larvae as seen in the null mutation harbored in the Ets family transcription factor lin-1¹⁹. Bleach treatment of the lin-1 gravid adults will significantly increase the number of larvae required for the assay.

It is vital to observe the lysis of the worms during the preparation of a synchronous C. elegans culture. The thick eggshell partially protects the embryos from the action of the bleach-sodium hydroxide mix even as the cuticle of the larvae and adults dissolve. However, prolonged exposure to the lysis solution will penetrate this protective casing leading to the death of the embryos. Hence it is important to stop the action of the bleach-sodium hydroxide mix when 70% of the adult worms have lysed. This is achieved by diluting the lysis solution with M9W. The maintenance of the arrested L1 larvae is another step to consider in the synchronous C. elegans culture protocol. Prolonged incubation of the L1 arrested larvae in M9W can cause them to transform into the dauer stage due to accumulation of the dauer pheromone. To avoid the formation of the dauer stage, it is suggested to use the L1 arrested larvae within 1–2 days of collecting the embryos.

The basal Muv phenotype is 60%–90% and 90% for the let-60(n1046) and let-23(sa62) worms, respectively. This suggests the let-60(n1046) worms are subject to phenotypic drift and, therefore, it is important to report the basal levels of expression of the Muv phenotype. Treatment of let-60(n1046) and let-23(sa62) worms with R-fendiline and other K-RAS inhibitors reduce the percentage of worms expressing the Muv phenotype. However, it is also reported that some inhibitors of the EGFR-RAS-ERK MAPK pathway may reduce the number of pseudovulvae per worm alone or may affect both the expression and the number of pseudovulvae¹⁷. Hence it is important to count the number of pseudovulvae in the Muv worms in both drug and DMSO treated worms. The Muv phenotype expressed in let-60(n1046) and let-23(sa62) adult worms is clearly visible under a dissecting microscope. However, the lin-1 (null) strain is relatively unhealthy, developmentally impaired and the vulval protrusions are poorly distinguished in the adult worms. Therefore, instead of the ectopic vulval protrusions, in lin-1 (null) worms, VPCs that adopt a adopted 1° or 2° cell fates on the ventral side in the L4 stage, can be counted using a high resolution DIC microscope.

The assay is inexpensive, easy, and not time consuming to setup. To further improve the processing time, the worms can be anesthetized to a final concentration of 2 mM sodium azide within the wells of tissue culture plate and imaged using a dissecting microscope equipped with a camera. Another modification of the assay would be to use heat killed E. coli OP50 instead of live bacteria²³. It has been shown that bacteria can metabolize certain small molecules leading to reduced bioactivities²⁴.

Previous studies have shown that the induction of the vulva in the worm is dependent on certain environmental cues. The vulvaless phenotypes of lin-3(n378) and let-23(n1045) have shown to be partially suppressed by starvation and exiting the dauer stage¹⁴. Furthermore, a study by Moghal et al., showed that vulvaless lin-3(n378), let-23(sy1) and let-60(sy95dn) mutants grown in M9 buffer had a higher number of VPCs assuming vulval cell fates compared with animals grown on standard NG plates²⁴. The data suggests worms grown in a liquid environment influences vulval induction. However, in our studies we did not observe the suppression of the Muv phenotype in a liquid environment.

In this protocol we demonstrate the use of C. elegans to evaluate the anti-RAS properties of fendiline. In a previous study, we have shown that multiple acid sphingomyelinase inhibitors, including tricyclic antidepressants such as desipramine, imipramine, and amitriptyline inhibit the Muv phenotype¹⁸. Furthermore, inhibitors of the sphingomyelin and ceramide biosynthetic pathways suppress the Muv phenotype expressed in the let-60(n1046) worms. These findings using the worm were validated in mammalian cell lines.

In conclusion, we demonstrate the use of C. elegans to identify inhibitors of EGFR and RAS activity in a liquid-based assay. Furthermore, the worm provides another system to identify and characterize the mechanisms of action of anti-RAS and EGFR therapeutics.

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
Media and chemicals
Agarose	Millipore Sigma	A9539-50G
Bacto Peptone	Fisher Scientific	DF0118-17-0
BD Difco Agar	Fisher Scientific	DF0145-17-0
BD Difco LB Broth	Fisher Scientific	DF0446-17-3
Calcium Chloride	Fisher Scientific	BP510-500
Cholesterol	Fisher Scientific	ICN10138201
Magnesium Sulfate	Fisher Scientific	BP213-1
Nystatin	Acros organics	AC455500050
Potassium Phosphate Dibasic	Fisher Scientific	BP363-500
Potassium pPhosphate Monobasic	Fisher Scientific	BP362-500
R-Fendiline			Commercially Synthesized (Pharmaceutical grade)
Sodium Azide	Millipore Sigma	S2002-25G
Sodium chloride	Fisher Scientific	BP358-1
Sodium Hydroxide	Fisher Scientific	SS266-1
8.25% Sodium Hypochlorite			Bleach
Sodium Phosphate Dibasic	Fisher Scientific	BP332-500
Streptomycin Sulfate	Fisher Scientific	BP910-50
(−)-Tetramisole Hydrochloride	Millipore Sigma	L9756
UO126 (MEK inhibitor)	Millipore Sigma	19-147
Consumables
15mL Conical Sterile Polypropylene Centrifuge Tubes	Fisher Scientific	12-565-269
50mL Conical Sterile Polypropylene Centrifuge Tubes	Fisher Scientific	12-565-271
Disposable Polystyrene Serological Pipettes 10mL	Fisher Scientific	07-200-574
Disposable Polystyrene Serological Pipettes 25mL	Fisher Scientific	07-200-575
No. 1.5  18 mm X 18 mm Cover Slips	Fisher Scientific	12-541A
Petri Dish with Clear Lid (60 x 15 mm)	Fisher Scientific	FB0875713A
Petri Dishes with Clear Lid (100X15mm)	Fisher Scientific	FB0875712
Plain Glass Microscope Slides (75 x 25 mm)	Fisher Scientific	12-544-4
12- Well Tissue Culture Plates	Fisher Scientific	50-197-4804
Software
Prism	Graphpad
Bacterial Strains
E. coli OP50
Worm Strains
Strain	Genotype	Transgene	Source
MT2124	let-60(n1046) IV.		CGC
MT7567	lin-1(sy254) IV.		CGC
PS1839	let-23(sa62) II.		CGC