Evaluating the Effects of Biotoxins on Immune Cell Functions in Zebrafish

Summary

This protocol describes flow cytometric assays that can evaluate the effects of toxin exposure on the endocytic functions of different subpopulations of zebrafish leukocytes. The use of specific functional inhibitors in the assay allows differentiation of the altered endocytic mechanisms.

Abstract

A variety of biological toxins can be present at harmful levels in the aquatic environment. Cyanobacteria are a diverse group of prokaryotic microorganisms that produce cyanotoxins in the aquatic environment. These biotoxins can be hepatotoxins, dermatoxins, or neurotoxins and can affect fish and mammals. At high levels, these compounds are fatal. At non-lethal levels, they act insidiously and affect immune cell functions. Algae-produced biotoxins include microcystin and anatoxin A. Aquatic animals can also ingest material contaminated with botulinum neurotoxin E (BoNT/E) produced by Clostridium botulinum, also resulting in death or decreased immune functions. Zebrafish can be used to study how toxins affect immune cell functions. In these studies, toxin exposures can be performed in vivo or in vitro. In vivo studies expose the zebrafish to the toxin, and then the cells are isolated. This method demonstrates how the tissue environment can influence leukocyte function. The in vitro studies isolate the cells first, and then expose them to the toxin in culture wells. The leukocytes are obtained by kidney marrow extraction, followed by density gradient centrifugation. How leukocytes internalize pathogens is determined by endocytic mechanisms. Flow cytometry phagocytosis assays demonstrate if endocytic mechanisms have been altered by toxin exposure. Studies using isolated leukocytes to determine how toxins cause immune dysfunction are lacking. The procedures described in this article will enable laboratories to use zebrafish to study the mechanisms that are impacted when an environmental toxin decreases endocytic functions of immune cells.

Introduction

There are many types of environmental biotoxins and immune suppressive agents. Algae blooms that contain bacterial toxins occur in inland waters and can also occur as biofilms¹. Cyanobacteria (blue-green algae) naturally occurs in all freshwater ecosystems. Cyanobacterial blooms have substantially increased in freshwater systems². At certain times, the Cyanobacteria can produce toxins that are harmful to aquatic and terrestrial animals. These toxins can affect the liver, skin, and mucous membranes, and/or the nervous system. Two compounds produced by Cyanobacteria are microcystin and anatoxin A. Microcystin is a cyclic heptapeptide³. Anatoxin A is an alkaloid⁴. Botulinum neurotoxin E (BoNT/E) is another toxin that occurs in aquatic systems. It is produced by Clostridium botulinum and can be ingested by aquatic animals⁵.

Exposure to environmental toxins affects fish and can also affect animal health and increase disease occurrences⁶. Understanding how these toxins affect immune cells is fundamental to determination of the risks associated with exposure to these substances. Zebrafish are an excellent model for studying the effects of environmental toxins on immune cells⁷. Developing a method that utilizes flow cytometry and zebrafish leukocytes is highly beneficial. Zebrafish have physiological relevance to humans, and this method can be applied to a wide range of research areas, from basic toxicology and immunology to drug discovery and developmental biology. Because they are aquatic organisms, zebrafish are particularly suitable for studying the effects of waterborne environmental toxins⁷. The use of zebrafish is less expensive than other vertebrate models, and their use raises fewer ethical concerns.

White blood cells, or leukocytes, are the first line of cellular defense against disease causing organisms. Endocytosis is the process of a cell taking up or internalizing a liquid or particle that is external to the cell. This is accomplished by the cell enclosing the compound in a vesicle⁸. Leukocytes use this process as the first step in killing pathogens and preparing a defense against disease. Phagocytosis is a type of endocytosis and was one of the first methods used to investigate the effects of environmental pollutants on fish health⁹. The Petrie-Hanson lab has developed methods using zebrafish leukocytes to screen biotoxins for their potential ability to interfere with leukocyte endocytic and phagocytic functions and impact immune defenses. The types of endocytosis included in these methods are pinocytosis, phagocytosis, calcium dependent receptor-mediated phagocytosis and mannose receptor mediated phagocytosis. Using flow cytometry methods with zebrafish were first described in the Petrie-Hanson lab⁹ and are used routinely to investigate aquatic toxins and pathogens. Rag1^-/- mutant zebrafish do not have T and B cells¹⁰ and can be used to specifically investigate innate immune cell mechanisms.

Flow cytometry is laser-based and can be used to determine the physical properties of cells. The forward scatter, or FSC value is plotted on the X axis and represents the size of the cell. The side scatter, or SSC, is plotted on the Y axis and represents the cytoplasmic granularity of the cell. The resulting plot demonstrates populations of cells with similar physical characteristics grouped together, with the different cell types appearing at various locations on a scatter plot. These populations can change location on the scatter plot as the physical characteristics of the cells change⁹. Using this technique with zebrafish leukocytes enables researchers to assess changes in cell populations in response to various stimuli, including environmental toxins.

The flow cytometer is multi-dimensional, and multiple types of fluorophores can be used in the evaluation to further characterize the cells and their activity. In the assays described in this protocol, endocytosis is characterized by measuring the amount of fluorescent material a cell has internalized. If and how toxin exposure affects endocytic mechanisms can be determined by comparing the ability of toxin exposed cells to take up the material compared to the ability of non-toxin exposed cells using flow cytometry. The endocytic processes that can be evaluated this way include pinocytosis, receptor mediated endocytosis and phagocytosis.

Pinocytosis is the uptake of soluble components, and it does not utilize cell receptors. Uptake involves cytoplasmic rearrangement by microfilaments and microtubules to form small vacuoles. Luciferase Yellow (LY) is a fluorescent dye used to measure liquid uptake by non-selective pinocytosis¹¹. Receptor mediated endocytosis involves the selective uptake of large molecules. Fluorescein (FITC) labeled dextran (DX) 40 can be used to evaluate this process. Phagocytosis is a form of endocytosis that ingests particles greater than 0.5 micrometers. This process is investigated by procedures using FITC-DX70 and FITC-Eschericia coli. DX40 and DX70 have molecular weights of 40,000 and 70,000, respectively. FITC-E. coli is the standard laboratory strain of E. coli bound to a fluor that can be measured by the flow cytometer. Many forms of the receptor mediated endocytosis require calcium as a signaling molecule and for cytoskeletal rearrangement⁹. Another type of receptor mediated endocytosis is mannose receptor (MR) mediated endocytosis. Mannose receptors are transmembrane proteins that recognize forms of mannan on microbial cell surfaces⁹. To optimize these procedures, a dose response curve should be created with each toxin to establish the doses to be used. A saturation curve should be performed for LY, FITC-DX40, FITC-DX70 and FITC-E. coli to assess the correct concentration to use.

The mechanisms used by leukocytes to internalize different particles may vary. To suggest which component of the process may be affected by toxin exposure, inhibitors can be added to block the phagocytic mechanisms. Cytochalasin D (CCD) will inhibit microtubule movement and therefore, pinocytosis. CCD does not influence receptor-mediated endocytosis¹¹. EDTA blocks calcium (Ca²⁺) dependent receptor-mediated endocytosis. Mannan is a natural ligand for the MR. Mannan is used as a mannose receptor inhibitor to assess if phagocytosis or pinocytosis is mannose receptor mediated⁹.

The purpose of this protocol is to demonstrate the procedures for determining whether toxin exposure has affected the ability of phagocytic leukocytes to uptake pathogens. These protocols may also discern if a specific endocytic mechanism is affected. Performing these assays on the flow cytometer allows further discrimination by selecting leukocyte populations based on size and cytoplasmic granularity to determine if leukocyte subpopulations have been differentially affected. This method relies on electronic gating of cell populations.

Protocol

This protocol has been approved by the Mississippi State University Institutional Animal Care and Use Committee (MSU-IACUC). All zebrafish used in this study were bred from a homozygous colony of rag1^-/- mutant zebrafish previously established in the specific pathogen-free hatchery in the College of Veterinary Medicine, Mississippi State University (MSU)¹⁰. Wild-type zebrafish were also propagated in this hatchery. In these studies, toxin exposures can be performed in vivo or in vitro. In vivo studies expose the zebrafish to the toxin, after which the leukocytes are isolated, indicating how the toxin and the tissue microenvironment may interact and influence leukocyte function. The in vitro studies isolate the leukocytes and then expose the cells to the toxin in culture wells.

1. Reagent and solution preparation

1. Fluorescent Activated Cells Sorting (FACS) Buffer: (Hanks Balanced Salt Solution without Calcium or Magnesium (HBSS) + 0.05% Bovine Serum Albumin (BSA)) (see Table of Materials). Prepare this fresh before cell isolation.
2. Tissue culture media (TCM): Use RPMI-1640 supplemented with glutamax and 10% fetal bovine serum (see Table of Materials).
3. Cytochalasin D (CCD): Prepare the stock solution by resuspending the commercially available product (see Table of Materials) in 1 mL of 200 proof absolute ethanol. Prepare the working solution by adding 20 μL of stock CCD to 980 μL of FACS buffer, to get the final concentration 2.5 μg/mL.
4. EDTA: The stock solution is 1 mg/mL. Prepare the working solution by adding 100 μL of the stock solution to 900 μL of FACS buffer, to get the final concentration 1 mM.
5. Mannan: Prepare a 1 mg/mL stock solution (see Table of Materials). Next, prepare the working solution by adding 100 μL of the stock solution to 900 μL of FACS buffer, to get the final concentration 500 μg/mL.
6. Prepare Lucifer Yellow (10 μg/mL) (LY, a fluorescent dye), Dextran-40 (DX-40, 500 μg/mL) and Dextran-70 (DX-70, 500 μg/mL) separately by making a 1 mg/mL solution of each by resuspending the commercially available reagents (see Table of Materials) in sterile water.
7. Fluorescein (FITC) bacteria: Grow Escherichia coli DH5α (see Table of Materials) with shaking in 100 mL Luria Bertani (LB) media supplemented with 50 µg/mL FITC overnight at 37 °C in a light protected environment get an optical density (OD) of 0.8 at 540 nm.
   1. Wash bacteria (3x) with phosphate buffered saline (PBS) by centrifugation for 10 min at 1000 x g followed by heating at 60 °C for 20 min. Wash 1 more time, then adjust bacterial concentrations to OD₅₄₀ 0.8. This will be the stock solution.
   2. Prepare the working solution by making a 1:100 dilution of the stock solution (1 mL stock: 99 mL FACS buffer). Adjust the final concentration to 1.8 x 10⁸ cells/mL.

2. Zebrafish care

1. Maintain the Zebrafish in a single-pass flow through system on dechlorinated municipal water and feed high protein fish meal and live Artemia to satiation.
2. At 6 months of age, remove mixed sex zebrafish from their tanks and transport them to the research laboratory for use in the experiment. This is the best age to use for the isolation of optimum number of leukocytes from the kidney marrow⁹.

3. Cell isolation

1. Euthanize 10 zebrafish in tricaine (~100 mL 4 mg/mL phosphate buffered Tricaine methane sulfonate/liter of fish water) (see Table of Materials) following established methods⁹.
2. Place a 40 µm cell strainer into a 50 mL conical centrifuge tube.
3. Remove the kidney marrow tissues⁹ and place them in a C-tube with 3 mL of tissue culture media and homogenize the tissue using a tissue dissociator. This is the preferred procedure.
4. Alternatively, disrupt the tissues with the rubber end of a 3 mL syringe in a cell strainer. After tissue is homogenized (or disrupted), pour the suspension through a 40 µm cell strainer placed on top of a 50 mL conical tube.
5. Centrifuge the filtered suspension at 500 x g for 5 min at 16 °C. Pour off the supernatant. Resuspend the cells in 3 mL of TCM. Repeat this step twice.
6. Carefully layer the cells onto 3 mL sterile-filtered, density gradient medium (1.077 g/mL) at room temperature.
7. Next, centrifuge at 800 x g for 20 min at 16 °C. Ensure that the brake is on the lowest setting, so the cells do not get resuspended when the centrifuge stops.
8. Remove the opaque buffy layer from the interface to a 14 mL round bottom tube.
9. Wash the cells by adding 5 mL of TCM and mixing with a Pasteur pipette. Centrifuge at 300 x g for 5 min at 16 °C.
10. Repeat the step, discarding the supernatant and resuspending the cell pellet in TCM to yield approximately 1 x 10⁶ cells/mL.

4. Cell viability assay

1. To monitor cell viability, evaluate cell death using propidium iodide (PI) staining.
   1. Add PI (200 µg/mL) at the concentration of 5 µL/mL of cells to be analyzed. Read on the PE/Texas Red channel.
      NOTE: Propidium iodide diffuses through the holes in the membranes of dead cells, thereby, staining them. The evaluated viability was 85% for the present study.

5. Toxin incubation

1. Aliquot 200 μL of cells from the isolated cell suspension into each well of a 6-well tissue culture plate (3 wells for control and 3 wells for toxin exposure). This allows for replicates to be run in triplicate.
2. Add 2 mL TCM to each well.
3. Add toxin (~2.5 μg/mL) to the three of the wells.
   NOTE: The toxin concentration should be optimized prior to beginning the experiment and will depend on the toxin used for the assay.
4. Incubate tissue culture plate for desired exposure time. The desired exposure time should be optimized prior to beginning the experiment and will depend on the toxin used for the assay. In the present study, the exposure time was ~1 h.
5. After exposure time, place tissue culture plate on ice for 10 min.
6. Pipette cells in each well up and down with care to remove any adherent cells from the plate.
7. Remove cells from the plate into a 14 mL round bottom centrifuge tube.
8. Centrifuge the cells at 300 x g for 5 min at 16 °C. Resuspend cells in 3 mL of FACS buffer. Repeat the step twice.
9. Resuspend the cells in 1.5 mL of FACS buffer.
   NOTE: For the in vivo studies, expose the zebrafish to the toxin (at a similar concentration as mentioned above), and then isolate the cells.

6. Endocytosis assay

1. Aliquot 100 μL of cells to 5 mL flow cytometry tubes. Label tubes as follows with four replicate tubes from each well: (1) LY, (2) DX40, (3) DX70, (4) FITC-E. coli.
2. Add fluorescence dye as stated in step 1.6 and step 1.7 to appropriate tubes: 50 μL of LY to tube (1), 50 μL DX40 to tube (2); 50 μL DX70 to tube (3); 50 μL FITC-E. coli to tube (4).
3. Incubate for 1 h. Add 200 μL of FACS buffer to each tube.
4. Centrifuge the cells at 400 x g for 5 min at 16 °C. Pour off the supernatant and resuspend the cells in 200 μL of FACS buffer. Repeat this step three times. Perform the flow cytometric analysis.

7. Flow cytometry

1. Perform flow cytometry to visualize the cell populations and collect data. Refer to Hohn et al.⁹ for details.
2. Identify the target cell populations at the first step. Use side scatter (SSC) (usually the vertical axis) to remove debris and small, pycnotic cells on the far-left side and forward scatter (FSC) (usually the horizontal axis) to remove the same debris on the bottom of the plot.
3. Eliminate the doublet reads (cells counted twice). This is performed by using a pulse geometry gate (FSC-H x FSC-A).
4. Determine the background signal. Run cells only fluorescence control and an isotype fluorescence control for each fluorochrome used, to identify background fluorescent levels.
   NOTE: Each fluorochrome has its own unique level of background fluorescence, influenced by cell autofluorescence and non-specific binding. This distinguishes between true positive signals and background fluorescence. When identifying and gating cell populations based on fluorescence, these control values are subtracted from the number of positive events, and the resulting number is the number used for analysis.
5. Run the samples and identify cell populations to be gated. Use FSC and SSC to morphologically characterize and identify different cell populations within a mixture of cells.
   NOTE: FSC is primarily related to the size of the cell. Larger cells will scatter more light in the forward direction, resulting in higher FSC values, while smaller cells will scatter less light and have lower FSC values. Different sizes of cells will group in clusters. For example, lymphocytes, being smaller, may appear as a separate cluster with lower FSC values compared to granulocytes. SSC is related to the granularity, complexity, and internal structure of the cell. Cells with more cytoplasmic granularity will have higher SSC values. Similar cells will cluster together. For example, neutrophils have cytoplasmic granules and will cluster at a higher SSC value than lymphocytes.
6. Run all the samples, and based on FSC/SSC, apply gates to cell populations to be further analyzed. Gated areas are applied based on cell density and populations of interest.
7. Plot gated areas onto histograms to analyze the Mean Fluorescence Intensity (MFI) of FITC in each gate for fluorescence at 495 nm / 519 nm with the blue laser.
8. Export cell counts as determined by the software to excel in the analysis program.
9. Analyze the data in a statistical software package (see Table of Materials).
10. Compare MFI of each population of control cells to MFI of cell populations exposed to toxin.

8. Effect of toxin on endocytic mechanisms and effects of inhibitors CCD, EDTA and mannan on endocytic mechanisms

NOTE: To uptake liquids or particles, the cell cytoplasm actively moves. This movement requires multiple structural elements and signaling pathways. Biotoxins can affect any of these elements or pathways. Comparing the effects of specific characterized inhibitors can be used to aid in the discernment of how the biotoxin may be acting¹¹.

1. Perform cell isolations as described in step 3.
2. Aliquot 100 μL of cells to 5 mL flow cytometry tubes (with four replicate tubes from each well).
   NOTE: Label tubes as follows: (5) CCD + LY (CCD inhibits pinocytosis by inhibiting microfilament and microtubule rearrangement; LY is used to assess liquid uptake by micropinocytosis)⁹. (6) EDTA + DX40 (EDTA blocks Ca²⁺ dependent receptor mediated phagocytosis and micropinocytosis; measuring if uptake involves a calcium dependent mechanism). (7) EDTA + DX70. (8) EDTA + FITC-E. coli. (9) Mannan + DX40 (mannan inhibits specific uptake by the mannose receptor, measuring the MR-dependent endocytosis⁹). (10) Mannan + DX70. (11) Mannan + FITC-E. coli.
3. Add inhibitors: 1 μL of CCD to tube (5); 10 μL EDTA to tubes (6-8); 100 μL Mannan to tubes (9-11). Follow the concentrations as mentioned in step 1. Incubate for 5 min.
4. Add fluorescent secondary to appropriate tubes: 50 μL of LY to tube (1) and (5); 50 μL of DX40 to tube (2), (6) and (9); 50 μL of DX70 to tube (3), (7) and (10); 50 μL of FITC-E. coli to tube (4), (8) and (11). Incubate for 30 min.
5. Add 200 μL of FACS buffer to each tube. Centrifuge at 400 x g at 16 °C for 5 min.
6. Pour off the supernatant and resuspend the cells in 200 μL of FACS buffer. Repeat steps 8.5-8.6 two times.
7. Run the samples on the flow cytometer following step 7.

Results

Endocytosis assays use mixed leukocytes isolated from gradients and gated for phagocytes and lymphocytes to determine if specific cellular mechanisms have been altered by toxin exposure. First, cells are gated based on size and granularity¹⁰. Dead, fragmented, or dying cells are visualized in the lower left corner of the scatter plot and are eliminated, not analyzed for phagocytosis. The phagocyte gate includes macrophages, Natural Killer (NK cells), and granulocytes, while the lymphocyte gate includes lymphocytes, phagocytic B cells, and nonspecific cytotoxic cells (NCCs) (Figure 1).

Phagocytosis is determined by the uptake of FITC-labeled particles, as shown by reading the fluorescence at 495 nm / 519 nm with the blue laser. Background autofluorescence and nonspecific binding fluorescence are eliminated by accounting for the mean fluorescence intensity (MFI) emitted by the control cells and the isotype control. A dose-response curve should be performed with each inhibitor to establish the concentrations to be used. If the flow cytometry results of cells exposed to the toxin being tested are the same as results of cells exposed to an inhibitor, then the toxin is causing a similar effect and may be inhibiting that endocytic mechanism. If the responses of the toxin and the inhibitor are additive, this suggests that the toxin may be inhibiting endocytosis by a different mechanism than the inhibitor.

Comparisons are made between cells not treated and cells treated with the toxin being tested (Figure 2), cells treated with toxin or inhibitor (Figure 3A), and cells treated with the inhibitor alone (Figure 3B). The representative results show that cells treated with microcystin were inhibited, similar to cells treated with CCD. These results demonstrate that microcystin inhibited liquid uptake by micropinocytosis. Gated cells are analyzed by one-way ANOVA, and results should be reported as the mean number of cells positive for FITC.

Figure 1: Representative flow cytometry scatterplot of leukocytes isolated from wild type zebrafish kidney tissue based on forward scatter (FSC) and side scatter (SSC) characteristics. The phagocytes gate includes macrophages/monocytes and granulocytes (23% of the gated cells. The lymphocytes gate includes lymphocytes, phagocytic B cells and nonspecific cytotoxic cells (NCCs) (26% of the gated cells). Please click here to view a larger version of this figure.

Figure 2: Mean fluorescence intensity (MFI) of phagocytic cells gated in Figure 1 after exposure to toxin. Tube (1) shows unstained cells while cells in tube (2) are emitting fluorescence through the uptake of LY. Please click here to view a larger version of this figure.

Figure 3: Mean fluorescence intensity (MFI) of phagocytes gated in Figure 1. Uptake of Lucifer yellow (LY) without the presence of Cytochalasin D (CCD) inhibitor (A); uptake of LY with the presence of CCD. Tube (1) shows unstained cells while cells in Tube (2) are emitting fluorescence through the uptake of LY. Please click here to view a larger version of this figure.

Figure 4: Overview of the steps involved in the protocol. *The duration and extent the fish are exposed to a biotoxin will vary depending on the specific biotoxin used and the animal care and biosafety regulations and protocols of different institutions. **The desired concentration used and the exposure time used should be optimized prior to the experiment and will depend on the toxin being used. Please click here to view a larger version of this figure.

Discussion

Utilizing flow cytometry with zebrafish leukocytes offers a powerful and versatile approach for studying the immune system in detail, assessing the impact of environmental toxins, and facilitating toxicological research. It provides a way to quickly and effectively evaluate the impact of toxins on immune cells and the immune response. The results reveal humoral factors involved and suggests how the fish's physiology and metabolism interact with environmental biotoxins. The overview of the protocol is depicted in Figure 4.

However, there are limitations to these methods. Preparing zebrafish leukocytes for flow cytometry can be more complex compared to other models. The small size of zebrafish can limit the amount of blood and the number of leukocytes available for analysis^9,10. These procedures require dissociating tissues to obtain single-cell suspensions, and cell viability can affect the accuracy of results. Compared to models like mice, the range of available immunological tools, such as antibodies specific to zebrafish cell markers, is more limited for zebrafish. This limitation can restrict the depth of analysis that can be conducted on zebrafish leukocytes. Moreover, the interpretation of flow cytometry data requires expertise, especially when dealing with complex immune responses. Misinterpretation of data can lead to incorrect conclusions, particularly in the context of environmental toxin impacts.

Results of cell samples with no inhibitors added can be compared to cell samples with the inhibitors. Results of cell samples with the toxin added should be compared to cell samples with the inhibitors. If cytoskeletal rearrangement has been affected, there will be limited LY uptake. If the toxin inhibits DX-40 or DX-70 uptake, different particulate sizes will not be phagocytosed. If the cells are not taking up the FITC-E. coli, then the toxin has inhibited phagocytosis.

In summary, zebrafish serve as an excellent model for studying the effects of environmental toxins on immune cells. This protocol outlines a method utilizing flow cytometry and zebrafish leukocytes to screen biotoxins, with a particular focus on endocytosis and phagocytosis processes¹¹. The goal of this study is to ascertain whether toxin exposure affects the ability of phagocytic leukocytes to uptake pathogens and to discern potential impacts on specific endocytic mechanisms.

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
10% fetal bovine serum	Gibco	A3160501
14 mL round bottom centrifuge tubes	BD Biosciences	352059
40 µm cell straininer	Corning	07-201-430
5 mL flow cytometry tubes	BD Biosciences	352235
50 mL conical centrifuge tube	Corning	14-959-49A
Absolute ethanol	Fisher	BP2818-500
Automated cell counter	Life Technologies Countess II FL		for studying cell viability
Bovine serum albumin (BSA)	Sigma	A3059
cytochalasin D (CCD)	Sigma	C8273-5MG
Dextran 40	Sigma	FD40-100MG
Dextran 70	Sigma	46945-100MG-F
Escherichia coli DH5α (or other lab bacterial strain)	New England Biolabs	C29871
Ethylenediaminetetraacetic acid (EDTA)	Sigma	ED4SS
Flow analysis software	Novoacea software
Flow cytometer	Novocyte 3000
Fluorescein	Fluka BioChemica	46950
hanks balanced salt solution without calcium or magnesium	Sigma	H4891
Histopaque 1077	Sigma	10771-100ML
Lucifer Yellow	Sigma	L0259-25MG
Mannan	Sigma	M7504-250MG
Phosphate buffered saline	Sigma	P3813
RPMI-1640 with GlutaMax	Gibco	61870036
Statistical software	SPSS
Toxin
Tricaine	Western Chemical Inc	NC0342409