If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Department of Chemistry, UF Scripps Biomedical Research, Jupiter, FL, USANatural Products Discovery Center, UF Scripps Biomedical Research, Jupiter, FL, USA
Department of Chemistry, UF Scripps Biomedical Research, Jupiter, FL, USANatural Products Discovery Center, UF Scripps Biomedical Research, Jupiter, FL, USA
Non-small cell lung cancer (NSCLC) is the most common type of lung cancer and accounts for ∼84% of all lung cancer cases. NSCLC remains one of the leading causes of cancer-associated death, with a 5-year survival rate less than 25%. This type of cancer begins with healthy cells that change and start growing out of control, leading to the formation of lesions or tumors. Understanding the dynamics of how the tumor microenvironment promotes cancer initiation and progression that leads to cancer metastasis is crucial to help identify new molecular therapies. 3D primary cell tumor models have received renewed recognition due to their ability to better mimic the complexity of in vivo tumors and as a potential bridge between traditional 2D culture and in vivo studies. Vast improvements in 3D cell culture technologies make them much more cost effective and efficient largely because of the use of a cell-repellent surfaces and a novel angle plate adaptor technology. To exploit this technology, we accessed the Natural Products Library (NPL) at UF Scripps, which consists of crude extracts, partially purified fractions, and pure natural products (NPs). NPs generally are not very well represented in most drug discovery libraries and thus provide new insights to discover leads that could potentially emerge as novel molecular therapies. Herein we describe how we combined these technologies for 3D screening in 1536 well format using a panel of ten NSCLC cells lines (5 wild type and 5 mutant) against ∼1280 selected members of the NPL. After further evaluation, the selected active hits were prioritized to be screened against all 10 NSCLC cell lines as concentration response curves to determine the efficacy and selectivity of the compounds between wild type and mutant 3D cell models. Here, we demonstrate the methods needed for automated 3D screening using microbial NPs, exemplified by crude extracts, partially purified fractions, and pure NPs, that may lead to future use targeting human cancer.
3D primary cell tumor models have received renewed recognition due to their ability to better mimic the complexity of in vivo tumors and as a potential bridge between traditional 2D culture and in vivo studies. 3-Dimensional (3D) cancer models have become an essential tool for the evaluation of new chemotherapeutics to fight cancer, yet automated drug screening assays using these models remains challenging for drug discovery. Advances by our group and others have enabled the screening of cancer 3D models in a rapid, highly miniaturized, and cost-effective fashion that permits direct compound response profiling to be generated in a phenotypic manner [
]. 3D ex vivo tumor models are believed to better recapitulate the features of the disease such as cell interactions, hypoxia, and phenotypic heterogeneity of the cancer tissue and drug response [
]. Patient tumor derived organoids models can be generated from minimal amounts of primary tissue and can be used as models for 3D high-throughput screening (HTS). To effectively deploy 3D models for HTS there are generally two essential requirements: to produce the required amount of spheroids, and to yield acceptable assay performance [
] or microcavity geometry plates that are surface repellent that can be quickly adapted for HTS. Although, we have extensively explored both methods, there are limitations and associated costs to these technologies that still play a key factor in the decision to pursue 3D research into high throughput environments. The ability to move rapidly into affordable, scaffold free systems for HTS is a top priority for our current research efforts [
]. Vast improvements in 3D cell culture technologies make them much more cost effective and efficient largely because of the use of a cell-repellent surfaces and a novel angle plate adaptor technology. To explore other methods, we developed a reliable cost effective and efficient method to form spheroids. The design was created in collaboration with Greiner BioOne using an in-house 3D printer to create an angle plate adapter. By combining the use of a cell-repellent surface and this angle plate adaptor technology, we were able to facilitate the formation of 3D cultures in a scaffold free system that is uHTS compatible. This new method was well validated, side by side, with other available methods which include the micro cavity ultra-low attachment surface plates from Corning and the magnetic bioprinting technology from Greiner BioOne, both of which we have used extensively before [
As part of this effort, we focused on a panel of ten non-small cell lung cancer cell lines (NSCLC). Non-small cell lung cancer (NSCLC) is the most common type of lung cancer and accounts for ∼84% of all lung cancer cases. NSCLC remains one of the leading causes of cancer-associated death, with a 5-year survival rate less than 25%. This type of cancer begins with healthy cells that change and start growing out of control, leading to the formation of lesions or tumors. Understanding the dynamics of how the tumor microenvironment promotes cancer initiation and progression that leads to cancer metastasis is crucial to help identify new molecular therapies. Previous research demonstrated KRAS dependency in the context of KRAS G12C mutant cell lines and the confirmation of culture-dependent effects of KRAS across monolayer vs. 3D spheroids [
]. To that end, the purpose of this study was to identify NPs that inhibit KRAS in a focused subset of NSCLC organoids in 3D format via HTS. We provide evidence that while spheroid cultures traditionally rely on artificial extracellular matrices (ECMs) to facilitate aggregation into 3D models, we were able to conduct HTS screening in a completely scaffold-free system.
To exploit this technology, we accessed the Natural Products Library (NPL) at UF Scripps, which consists of crude extracts, partially purified fractions, and pure natural products (NPs). NPs have significantly contributed to the discovery of novel chemistry, drug leads, and tool molecules to probe and address complex challenges in biology and medicine [
]. NPs generally are not very well represented in most drug discovery libraries and thus provide new insights to discover leads that could potentially emerge as novel molecular therapies. NP discovery and development is still emerging and remains as one of the best sources of new drug leads. It will continue to play a highly significant role in the drug discovery and development process for the foreseeable future [
]. In this study, we demonstrate the methods for fully automated 3D screening that identified active NPs from the microbial NPL at UF Scripps that are now being investigated further for NP dereplication and molecular determination and target engagement within NSCLCs, that may lead to future use targeting human cancer. This technology proved to be robust and reliable across platforms with a nice overlap in the activity of NPs when compared to each other[
]. The use of the plate adapter technology reduced the cost in comparison with the current 3D platforms which allows us to broaden into other areas of cancer research and other diseases all while in a 3D in vitro microenvironment [
Cell Culture medium for all the cell lines except Calu-1
Prepare RPMI 1640, 10% (vol/vol) FBS and 1% (vol/vol) antibiotic-antimycotic solution. The media is filtered through a 0.22um PES filter system. This can be stored at 4 °C for 2-3 weeks.
Calu-1 culture medium:
Prepare McCoy, 10% (vol/vol) FBS and 1% (vol/vol) antibiotic-antimycotic solution. The media is filtered through a 0.22um PES filter system. This can be stored at 4 °C for 2-3 weeks.
Procedure
Cell line thawing preparation
CRITICAL Thaw cell lines following ATCC guidelines depending on the cell line ● Timing 1 h (cell thawing), 3-4 d cell growth)
1. Prepare cell culture medium as described in Materials.
2. Warm up cell culture medium to 37 °C in a water bath.
3. Sterilize the biosafety cabinet by using UV light and wipe down all working surfaces with 70% (vol/vol) ethanol.
4. Thoroughly spray 70% (vol/vol) ethanol onto the outer surface of the bottles containing growth medium and bring them inside the biosafety cabinet.
5. Transfer 9 ml of cell culture medium into a 15-ml conical tube
6. Thaw a cryovial of NSCLC cancer cells by placing it into a 37 °C water bath (<1 min).
7. Take out the cryovial from the water bath when half the content is thawed, spray 70% (vol/vol) ethanol and quickly place it into the biosafety cabinet.
8. Add the cryovial content slowly dropwise to the 15mL conical tube containing warmed medium
9. If need it, add cell culture medium to recover additional cells from the cryovial. Add it to the same tube from Step 8.
10 Centrifuge at 125xg for 5-7 min.
11. Aspirate the growth medium by using vacuum suction inside the biosafety cabinet.
12 Gently resuspend the cells in 1 ml of warm cell culture medium using a P1000 pipette.
13. Add 4mL of cell growth media and mix the cells gently.
14. Take out 10 μl of cells and mix with 10 μl of trypan blue stain (1:1 ratio). Transfer the solution to cell-counting chamber slides to calculate the concentration of cells.
15. Seed cells according to the ATCC specifications in a T75 flask depending on the concentration of cells into the respective flask size.
16. Take out the culture flask from the biosafety cabinet and observe it under a brightfield microscope.
17. Cells should be round and floating in the medium.
18. Place the culture flask into the 37 °C CO2 incubator.
CRITICAL STEP Check cells under a microscope the next day. Healthy cells should be adherent to the bottom of the culture flask.
19. After 2–3 d of culture, observe the cells under the brightfield microscope. The cells are read to be expanded when they cover ∼80% of the bottom surface, depending on the doubling time and how confluent the cells were seeded.
Expand cell line and cell culture maintenance
● Timing 1 h (tumor cell expansion), 3-4 d cell growth)
CRITICAL Expand cells according to their doubling time and ATCC guidelines which can be cell line dependent.
20. Warm up cell growth medium to 37 °C in a water bath. Warm up the dissociation reagent TrypLE to room temperature (20–24 °C).
21. Sterilize the biosafety cabinet. Details are provided in Step 3.
22. Using 70% (vol/vol) ethanol, wipe the outer surface of the flask containing healthy and adherent cell lines that grew in the 37 °C CO2 incubator (from Step 19) and bring them inside the biosafety cabinet.
23. Thoroughly spray 70% (vol/vol) ethanol on to the outer surface of the bottles containing cell growth medium and bring them inside the biosafety cabinet.
24. Aspirate the growth medium by using vacuum suction inside the biosafety cabinet. Tilt the flask to avoid touching cells directly.
25. Add an appropriate volume of 1× DPBS at room temperature onto the cells depending on the surface area of the flask or plate. Make sure that the DPBS covers the entire surface of adherent cells by slightly tilting the flask.
26. Aspirate the 1× DPBS by using vacuum suction.
27. Add the appropriate amount of TrypLE to the flask. Make sure that TrypLE
covers the entire surface of adherent cells.
28. Place the flask back in the 37 °C CO2 incubator and wait for 3–5 min.
CRITICAL: The incubation with TrypLE will vary depending on the cell line
29. Take the flask out of the CO2 incubator. Gently tap the side of the flask with the palm of one hand while holding the flask securely with the other hand.
a. This will allow most cells to dissociate from the surface. Observe for cell dissociation from the surface under a microscope.
CRITICAL STEP The necessary incubation time may vary. If most of the cells are still attached, repeat Step 29 for an additional 2–3 min. Be careful not to over-trypsinize the cells (cells start to form visible clumps or the cells start getting lysed).
30. Take the flask back inside the biosafety cabinet. Add an equal volume of tumor cell culture medium. Tilt the flask and flush the bottom surface where the cells attached with culture medium. Repeat the flushing step four to five times to cover different parts of the flask surface.
CRITICAL STEP Minimize creating air bubbles while pipetting.
31. Transfer the cell-containing solution to a 15-ml or a 50 conical tube depending on the volume.
32. Centrifuge the tube at 300xg for 5 min at room temperature.
33. Repeat Steps 11–14 to count cell numbers and determine if there are enough cells for generating the number of spheroids for the amount of plates that will be seeded. Roughly 1 × 106 cells are required to generate spheroids in a 1536-well plate.
34. If cells need to be expanded, repeat Steps 15–33. Otherwise, skip to Step 35.
Seed NSCLC cells for the formation of spheroids, compound addition and 3D cell viability assay
● Timing 1.5 h (NSCLC seeding), 1h (Compound addition after 24hrs post cell seeding and angle adaptor incubation) and 4 d (spheroid formation and duration of the assay before cell titer glo addition)
This procedure gives details for harvesting the cells to seed them in cell repellent plates. See Table 1 for summarized steps for the seeding and formation of the spheroids/organoids. This method utilizes the cell repellent plates using a novel angle adaptor technology and a 3D viability assay that uses a detection reagent more adapted to spheroids that features a tailored lysis buffer (Cell Titer-Glo 3D, Part# G9683, Promega, Madison, WI).
35. Prepare cell culture growth media for the assay as described in Materials.
36. Warm up cell culture growth medium to 37 °C in a water bath. Warm up TrypLE to room temperature.
37. Cells were grown to ∼85% confluence in their respective complete growth media in T75 or T175 flasks.
38. Prepare cells by trypsinizing and resuspending the cells. Repeat Steps 22–33
39. Gently resuspend the cells in 1 ml of warm cell growth culture medium using a P1000 pipette.
40. Add the appropriate volume of cell culture growth media to the cells and pipette up and down to mix well and filter the cells using a 70um cell strainer.
40. Take out 10 μl of cells and mix with 10 μl of trypan blue stain (1:1 ratio). Transfer the solution to cell-counting chamber slides to calculate the concentration of cells.
41. To seed cells in a 1536-well plate, prepare cells at the appropriate cell concentration 0.5× 105 cells/mL in the appropriate volume of cell culture growth media pre-filtered through a cell strainer needed depending on the number of plates to be seeded it.
a.
This cell concentration is suitable for 250 cells/well. Please adjust as need it depending on the number of cells per well depending on the cell line.
b.
For the three of the cell lines h522 wild type, h2039 mutant and HCC1171 mutant cell line, 500 cells per well were seeded for the screening based on the assay implementation results.
c.
Prepare the cell suspension into a Nalgene Wide Mouth Bottles HDPE that are compatible for the automated dispenser taking in account enough volume for priming the lines and dead volume.
42. Using a automated flying reagent dispenser (FRD), dispense 5uL of culture media to Col1 thru Col2 and 5uL of cells suspension in each well to Col3 thru Col 48 into a Greiner Bio-One cell repellent surface plate
Table 1Protocol for angle adaptor for 1536-well format.
Table #1. 3D Format Angle Adaptor Protocol
Step
Parameter
Value
Description
Equipment
1
Harvesting and dispensing cells
5 µL/well
Trypsinize cells and prepare a cell suspension in growth media*
Biosafety A2 Cabinet/ Flying Reagent Dispenser
2
Angle Adaptor Incubation
24 hrs
Put the plate in the multi-stack angle adaptor and incubate for 24hrs at 37°C, 5%CO2 and 95% humidity
Angle Adaptor
3
Addition of DMSO or compounds
10nL/well
Compounds were pinned using the most concentrated source of the Natural Product library
GNF/Kalypsis Transfer Station
4
Incubation
72 hrs
Incubate at 37°C, 5%CO2 and 95% humidity
Steri-Cult Incubator
5
Equilibration
10min
Equilibrate assay plates at room temperature
6
3D Cell Titer Glo
5 µL/well
Spin plates
Centrifuge
7
Incubation
60 min
Room temperature, dark
8
Read
Lumi
Plates were read for 10sec under Lumi mode
Microplate image reader
Step Notes.
1. Cells are plated in their respective growth media. h1975 WT, h522, h226, h1838, h838, h2030, h1075 G12D and h358 use RPMI1640, 10% Fetal Bovine Serum and 1X Antibiotic-Antimycotic. CALU1’s growth media consists of McCoy instead of RPMI1640 and HCC1171 use RPMI1640 with 10% Heat Inactivated Fetal Bovine Serum.
2. h226, CALU1, h358, h1975 WT, h1975 G12D, and h838 use 250 cells/well. h2030, h1838, HCC1171 and h522 use 500 cells/well.
CRITICAL: Make sure to dispense the cells at a low speed to avoid bubbles and ensure a more accurate dispensing across the plate.
a.
During the dispensing, swirling the bottle containing the cells to keep the cells in solution while dispensing.
43. Centrifuge the 1536-well plate at 100g for 5 min at room temperature.
a.
After cell seeding, a brief centrifugation was performed to ensure the cells were at the bottom of the plate
b.
Gasketed heavy assay lids (GNF) are used to help minimize evaporation issues due to the duration of the assay
CRITICAL: If the plate is not centrifuged, the cells might not settle properly into these plates.
44. After centrifugation, check to see the cells at the bottom of the 1536-well plate.
CRITICAL STEP Handle gently to avoid disturbing the cells.
45. Place the assay plate containing the assay plates seeded with cells into the angle adaptor and incubate at 37 °C CO2 incubator for 24hrs. (see Fig. 1)
Fig. 1Greiner 45 Angle Adaptor Technology Development. Greiner cell repellent plates readily form spheroids using cells by combining the use of a cell-repellent surface and a novel angle plate adaptor technology. The angle plate adaptor technology was improved by modifying the angle and the number of plates across the different versions that were developed. Note that the shape of the bottom of the wells is compatible with automated confocal microscopy.
All assay plates were placed on the angle adaptor for 24 hours to allow the cells to aggregate in one corner of the well to facilitate the formation of the 3D spheroids.
46. Check cell formation into spheroids under a brightfield microscope.
47. After 24 hours, cells were treated with NPs library or vehicle (10 nL, 0.2% DMSO) via the GNF/Kalypsis robotic pintool
a.
Check cell formation into spheroids under a brightfield microscope before addition of compounds.
b.
The 10nL pintool was used for screening based on a preliminary assessment that was performed.
48. After pinning, plates were placed in the normal orientation in the incubator at 37C, 95% Relative Humidity and 5% CO2 for 3 days.
49. Cell viability was then assessed after 72-hour incubation using Cell-Titer Glo 3D reagent according to manufacturer's instructions.
50. The ViewLux microplate reader (Perkin Elmer, Waltham, MA) was used to quantify luminescence signal.
51. Concentration response curves (CRC) and IC50 values of 2 pharmacological control compounds (ARS-1620 and Doxorubicin) were used as the guide for assay optimization and drug screening.
Detailed methods and procedures
Development of the novel angle adaptor technology
This project utilized custom 3D printed plate holders which were designed to allow the plates to be incubated at a 45-degree angle relative to a typical horizontal incubator shelf. An initial single-plate holder design was provided by Greiner Bio-One for testing. This design was iterated upon using Autodesk Fusion 360 (San Rafael, CA), to enable multiple plates to be stacked in a single adapter. All variants of the angled plate holder were designed to provide sufficient room for lidded or unlidded microplates. 1-plate, 2-plate and 5-plate variants of the stack were designed, but the CAD model can be easily adapted to other configurations in order to support desired plate quantities. Prototype components were fabricated with PETG thermoplastic on a Prusa i3 MK3S 3D printer with the following slicer parameters: 100% infill, 0.3mm layer height. Note, while we used these adapters for 1536 well format, they are agnostic to the plate density and should work equally as well in use for 3D modelling in 96 or 384 well plates.
Cell preparation
Primary human non-small lung cancer cells derived from the lung adenocarcinoma disease were purchased from ATCC unless indicated otherwise. The cryovials were thawed and expanded according to the ATCC recommendations (see Methods/Procedures). Four KRAS wild type cell lines were used, h522 (Part# CRL-5810), h226 (Part# CRL-5826), h1838 (Part# CRL-5899) and h838 (Part# CRL-5844). Four mutants with the G12C KRAS mutation were used, h2030 (Part# CRL-5914), h358 (Part# CRL-5807), Calu1 (Part# HTB-54) and HCC1171 was obtained from Korean Cell Line Bank (Part# KCLB 71171). In addition, a cell line with the mutation KRAS G12D along with the isogenic pair, h1975 wild type was utilized as part of the study. The cells were cultured in flasks and expanded as 2 dimensional monolayers in RPMI 1640 (Part# 10-040-CV, Corning Inc, N.Y.) with 10% fetal bovine serum (Part# 97068-085, VWR, Radnor, PA), and 1× Anti-Anti (Part# 15240-062, Gibco Life Technologies, Carlsbad, CA) at 37oC, 5% CO2 and 95% relative humidity, unless a different media was indicated as per the ATCC or KCLB. The growth media for Calu1 used McCoy's 5A Modified Medium (Part# 16600082, GIBCO Life Technologies, Carlsbad, CA) with 10% serum (Part# 97068-085, VWR, Radnor, PA), and 1× Anti-Anti (Part# 15240-062, Gibco Life Technologies, Carlsbad, CA). The growth media for HCC1171 used RPMI 1640 (Part# 10-040-CV, Corning Inc, N.Y.) with 10% heat inactivated fetal bovine serum (Part# 35-011-CV, Corning, Inc, N.Y.), and 1× Anti-Anti (Part# 15240-062, Gibco Life Technologies, Carlsbad, CA). Cells were harvested and utilized for the purpose of 3D testing.
The natural products library at UF Scripps
The microbial NPL at UF Scripps consists of 46,031 crude extracts, 28,739 partially purified fractions, and 650 pure NPs, which were made and isolated from 14,635 actinobacteria that were cultured, on average, in three different media [
]. These actinobacterial strains at the Natural Products Discovery Center (NPDC), UF Scripps, totaling 125,127 specimens, were isolated over the last eight decades, many of which can no longer be accessed in their natural environment, making this collection extremely important to human health. Selected members of the NPL utilized for this effort consists of 360 crude extracts, 800 partially purified fractions, and 120 pure NPs that were reformatted in 384-well format and subsequently in 1536-well format that was used for the screening described herein.
3D culture and 3D viability assay optimization
Assays were optimized by testing different variables including cell number, aggregation time on the angle adaptor, 3D confirmation by Z-stack analysis, and incubation time after drug addition. A detailed stepwise protocol of the final assay conditions is presented in Table 1. The design and the creation of the 3D printer files used to make the angle adaptors is described in the Materials section. As a point of comparison, we also tested these same cells using Corning spheroid plate technology (Part# 4527, Corning Inc., NY) [
]. The Corning spheroid-based assay follows the same protocol including the brief centrifugation after seeding cells into the Corning Spheroid plates to facilitate spheroid formation except that (1) cells are not incubated on the angle adaptor, and (2) and cells are seeded into the U-bottom ultra-low attachment plates to allows the 3D formation. The formation of 3D structure for each of the cell lines was confirmed by Hoechst staining followed by confocal imaging using an INCell Analyzer 6000 high content reader to collect images which were then used for Z-stack analysis as previously described [
Use of this technique requires some adaption based on the cells or organoids under study. The time that takes cells to form organoids and spheroids will depend on their source and the type of cells used. Some cells can form spheroids within 24hrs after the incubation in the angle adaptor while others may require just a few hours and sometimes even days. We noticed that some cells that do not form nice compact spheroids and are more “organotypic”, meaning their appearance in 3D is more amorphous and less sphere shaped. In some case these lines may benefit form a longer incubation time in the 3D plates or on the angle adaptor. In this case, we will recommend keeping the plates in the angle adaptor during the entire duration of the experiment. Other issues that can be observed are some cells don't aggregate uniformly in the well and can cause the cells to be more spread out. Typically, this doesn't present with variability in phenotypic whole well readouts but can display differently when imaging which should be determined for each cell type/line.
Data analysis
HTS campaign and data processing
A 1,280 subset of the NPL, consisting of the selected crude extracts, partially purified fractions, and pure NPs, was screened a focused panel of 10 cell non-small lung cancer spheroid at a pace of 10 plates per day. Data files were uploaded into the UF Scripps’ institutional database for individual plate quality control and hit selection. Assay plates were determined acceptable only if their Z’ was > 0.5 [
Application of parallel multiparametric cell-based FLIPR detection assays for the identification of modulators of the muscarinic acetylcholine receptor 4 (M4).
Test Well refers to those wells with cells treated with test compounds. High Control is defined as wells containing medium only (100% inhibition), and Low Control wells contain cells treated with DMSO only (0% inhibition).
The high and low controls were also applied for assay quality evaluation in terms of Z’ [
]. Day to day assay response and stability was assessed using 2 pharmacological control compounds (ARS1620 and Doxorubicin) that we required to be within 3-fold of the expected IC50, on an experiment to experiment basis and across all experiments, for each cell model. A low control-based hit cut-off was used to define active NPs for each assay. This cutoff is calculated as the average percent inhibition of the low control wells plus three times their standard deviation [
A rapid phenotypic whole-cell screening approach for the identification of small-molecule inhibitors that counter β-lactamase resistance in Pseudomonas aeruginosa.
], Any sample exhibiting an activity higher than the cutoff calculated was declared active.
Concentration response assays
The selected hits were prepared as 10-point, 3-fold serial dilutions and tested against 10 non-small lung cancer cell lines described above in 3D format in triplicate using the 50nL pintool and the most concentrated samples of the NPL (estimated at ∼10 mM at most for the active NPs). For each test of the active NPs, % inhibition was plotted against NP concentration. A four-parameter equation describing a sigmoidal dose-response curve was then fitted with adjustable baseline using Assay Explorer software (Symyx Technologies, Santa Clara, CA). The reported CC50 values were generated from fitted curves by solving for the X-intercept value at the 50% inhibition level of the Y-intercept. The highest concentration tested is unknown due to the nature of these natural products. The pharmacological controls were prepared as 20-point, 3-fold serial dilutions and tested against the panel of NSCLC cells in 3D format in triplicate starting from 10 µM nominal concentration. For these control compounds, % inhibition was plotted against compound concentration. CC50 values of the control compounds were determined by fitting the concentration response curve data (CRC) with a four-parameter variable slope method in GraphPad Prism (GraphPad software, San Diego, CA).
Results
Assay implementation of 3D cell culture in 1536wpf using a novel angle adaptor technology
This novel angle adaptor technology was tested in 1536-well format across a panel of 10 different non-small lung cancer cell lines. The original design was only for one plate, but the design can be modified in order to enable multiple plates stacked into the same angle adaptor unit (Fig. 1). Using this method allowed us to implement the same phenotypic 3D cell viability approach that was developed and miniaturized in a 1536-well format using the ultra-low attachment spheroid microplates [
]. This novel technology was tested side by side to the ultra-low attachment method available in 1536 format from Corning and it demonstrated similar results. Bright field images at 24hrs and 96hrs post seeding clearly show the formation of the cells into spheroids (Fig. 2A). Assay statistics was comparable and % activity was decent between both methods, despite of the variability that these cultures can show when they are grown in 3D (Fig. 2B). The pharmacological controls were also compared between the two technologies and the CC50 values were within 3-fold, which indicates that the angle adaptor provides comparable results to the microcavity geometry plates. (Fig. 2C)
Fig. 23D Spheroid Technologies Test Comparison. (A) Bright field images were taken at different time points after seeding the cells in the Greiner ultra-low attachment plates vs. Corning microcavity geometry plates. (B). A correlation of the outcomes between both methods was performed. (C). Concentration response curve for the positive controls was plotted. Each curve represents the mean and the standard deviation of 16 replicates. Error bars are included and shown in SD.
In order to implement the assay, different cell densities were tested to determine the cell linearity of the assay for each of the cell lines. Different cell densities were seeded into a Greiner 1536 well cell repellent plate (Fig. 3A). Once seeded, the plates were incubated on a 45° angle device inside the cell culture incubator for 24hrs. This initial device was designed by Greiner and then multi-stack angle devices were created in house with a 3D printer. The angle plate method will aggregate the cells towards the corner of the well to help with the formation of the 3D spheroids in a cell repellent surface plate. Assay statistics were calculated for the different conditions tested and the optimal cell density was selected for the screening based on the linearity and the Z’ values (Fig. 3B). Images were taken at intervals of 0hrs, 24hrs and 96hrs post seeding to monitor the aggregation and 3D formation. After 96hrs of incubation, cell viability was measured using a simple 3D Cell Titer Glo luminescence assay. In a separate test to confirm the 3D formation, cells were stained with Hoechst and images were taken using a confocal microscope, which demonstrated the successful formation of 3D spheroids. (Fig. 3C)
Fig. 33D Cell Density Linearity test. (A) Cell number was assessed to determine the linearity of the assay. Z’ values were calculated with the different cell number tested following addition of the CellTiter-Glo® (B). Across all the assay conditions tested, the Z’ values ranged from 0.76 for the lowest cell number tested and about 0.8 for the additional cell numbers tested. (C). Z-stack analysis for 3D confirmation was performed for each of the cell lines used for this study. In this example, we are showing the Z-stack for h358 mutant cell line.
Primary screening of non-small lung cancer spheroid/organoid models using novel angle adaptor technology
In this study, we used a novel angle adaptor technology (Fig. 1) for testing primary non-small lung cancer cells in a 1536 well microplate formatted in 3D for high throughput screening. Using this 3D angle adaptor technology, disaggregated cells were seeded in cell repellent plates and pre-incubated on the angle adaptor that facilitates the formation of spheroids in the absence of artificial extracellular matrices (ECM). A panel of 10 non-small lung cell lines mostly derived from lung adenocarcinoma disease were used for the screening. To accomplish this, we employed a simple phenotypic assay to measure the cell viability in response to cytotoxic NPs [
], a process previously implemented for pancreatic organoid performed by Hou et al.
Initially, the first version of the angle adaptor was designed by Greiner and in collaboration with UF Scripps to test the technology. We were able to create different versions using a simple 3D printer, all of which proved to be compatible for use with these methods (Fig. 1). Another important consideration was the HTS concentration of the NPs. Due to their inherent nature, there is no concentration associated. As part of the assay implementation, an assessment was conducted to measure the effect of these NPs by testing different pintool sizes (data not shown). Based on the different parameters tested, we tested each molecule in triplicate using the 10nL pintool which demonstrated the sensitivity needed to identify cytotoxic NPs at a rate that is HTS amenable while not ending up with all NPs showing cytotoxicity. (Fig. 4). The assay performance was robust across the HTS, yielding an average Z’ of 0.79 ± 0.06 and a signal to background of 119.28 ± 9.86 for all ten assays. A summary of the primary screening results is presented in Table 2 in which we showed the assay statistics, the hit rate and the activity across the different type of samples of the natural products library tested.
Fig. 4Scatter plot for NSCLC G12C Mutant KRAS Cell Line Natural Product Library. The primary data from the screening for one of the mutant cell lines, h358, including high and low controls, are displayed. The S:B ratio = 122.65 ± 1.25 and Z’ = 0.80 ± 0.01 between high and low controls wells, as well as broad distribution of hits indicate that the assay is robust. (●) Represent low control wells containing cells in the presence of DMSO only, (●) represent data wells containing compounds and (●) represent high control wells containing media + DMSO. Pharmacological controls are also shown in the scatter plot, (●) represent wells containing cells in the presence of Doxorubicin and (●) represent wells containing cells in the presence of ARS-1620.
Identification of potent and selective inhibitors of the Plasmodium falciparum M18 Aspartyl Aminopeptidase (PfM18AAP) of human malaria via high-throughput screening.
] to calculate the cutoff for to determine which hits are considered active for each of the cell lines, a cumulative total of over 1000 hits were found across all 10 cell lines. Most of the hits overlapped between the cell lines. Overall, the extracts and pure compounds showed comparable and higher hit rates (10%), whereas fractions showed lower hit rates (4%) across all 10 cell lines. In addition, two control compounds were used to verify the sensitivity of the assay. Concentration response curves (CRC) of ARS-1620 and Doxorubicin were accessed using the same 3D format for each of the cell lines (Fig. 2C). ARS-1620 is a small molecule inhibitor that is highly potent and selective for KRASG12C mutation in vitro and in vivo [
]. Utilizing, ARS-1620 as a pharmacological control allowed us to confirm the sensitivity against the different cell lines in this study. In order to proceed to dose response assays, a multicomponent score data analysis approach was performed to analyze the data in which both wild type and KRASG12C mutant average % response was compared. A score assessment was applied based on the activity across all cell lines tested to determine which NPs to select for dose response studies. Conducting this deeper analysis between the wild type and mutant hit correlation between the cell lines allowed us to identify 128 NPs which were selected for dose response assays.
Dose response assays
The selected 128 top hits were tested in triplicate across the same panel of non-small lung cancer organoids in 3D format using the same angle adapter technology. The assay statistics Z’ average data for the 3D dose response data are 0.74 ± 0.10 with a signal to background of 115.52 ± 10.43 for all ten assays. Again, the uncertainty of the highest concentration for the active NPs in the crude extracts and fractions, estimated at 10 mM, make it difficult to report an accurate CC50 values generated from fitted curves. The approach used was based on the percent response and the effect of each NPL hit based on the pintool size. The analysis reported was based on the effect of these NPs and how they behave between the wild type and mutant cell lines. Using this approach, about five hits showed some selectivity towards the cancer cell lines and less activity some of the wild type cells tested which are presented in Table 3. Current efforts now consist of selectivity assays to determine the efficacy of these compounds towards the G12C mutant KRAS cell lines and dereplication of the NPL hits to structurally characterized the active NPs. In addition, proliferative/cytotoxicity assays are being performed using normal human peripheral blood mononuclear cells to determine cytotoxicity associated to normal human cells other than lung.
Table 3Most promising compounds identified of the Natural Product Screening Library.
This technique is completely adaptable to other multi-well formats which will allow researchers to scale to their needs. The limitations for the usage will be the cost of the plates and the space required to place the angle adaptors in the incubators. Another limitation is not the technique itself but, rather will be to generate the amount of cells needed for either 384 well format or 1536-well format depending on the organoids. This technique is a completely scaffold free system that will not require the usage of Matrigel or any other ECM components. For us, it affords a method to screen large libraries using 3D cell culture models with a rapid and reliable approach that facilitates early determination of drug efficacy in cancer models. This technique is cost effective and does not require expensive tools that can be associated with other platforms to create 3D cell cultures models. This protocol is easily accessible to any laboratory and opens the opportunity to be applied to other models of disease in a 3D environment
Author contributions
Virneliz Fernández Vega: Assay scientist performing the project and the screening
Dong Yang: Natural products chemist who participated in data analysis and selection of the hits
Luis Ortiz Jordán: Assisted with the cells and the experiments
Louis Conway: Scientist in assisting with the data analysis
Li Yun Chen: assisted with the cells in the Parker lab
Justin Shumate: High throughput engineer supporting the equipment
Pierre Baillargeon: High throughput engineer supporting the equipment and the database tools
Louis Scampavia: directed the engineering at the HTS facility
Christopher Parker: Conceptualized the agenda and provided the cells and help with the data analysis
Ben Shen: Conceptualized the project and supplied the natural products used in this study.
Timothy P. Spicer: Directed the design, execution, as well as funded the overall project.
Declaration of Competing Interest
The authors declared no potential conflict of interest with respect to the research, authorship and or publication of this article.
Acknowledgements
We thank Lina DeLuca (Scripps, Florida) for her assistance with NP and compound management. Natural products drug discovery in Shen lab is supported in part by NIH grants GM134954. This work was also supported in part by the Office of the Assistant Secretary of Defense for Health Affairs through the (Lung Cancer Research Program) under award no. W81XWH-20-1-0431 (C.G.P.).
References
Hou S.
Tiriac H.
Sridharan B.P.
Scampavia L.
Madoux F.
Seldin J.
Souza G.R.
Watson D.
Tuveson D.
Spicer T.P.
Advanced development of primary pancreatic organoid tumor models for high-throughput phenotypic drug screening.
Application of parallel multiparametric cell-based FLIPR detection assays for the identification of modulators of the muscarinic acetylcholine receptor 4 (M4).
A rapid phenotypic whole-cell screening approach for the identification of small-molecule inhibitors that counter β-lactamase resistance in Pseudomonas aeruginosa.
Identification of potent and selective inhibitors of the Plasmodium falciparum M18 Aspartyl Aminopeptidase (PfM18AAP) of human malaria via high-throughput screening.
00005-9&id=gr1.jpg){kind=link}
00005-9&id=gr2.jpg){kind=link}
00005-9&id=gr3.jpg){kind=link}
00005-9&id=gr4.jpg){kind=link}