Flow Cytometry Protocols for Surface and Intracellular Antigen Analyses of Neural Cell Types

Summary

We provide a detailed description of a protocol for flow cytometric analysis of surface antigens and/or intracellular antigens in neural cell types. Critical aspects of experimental planning, step-by-step methodological procedures, and fundamental principles of flow cytometry are explained in order to enable neurobiologists to exploit this powerful technology.

Abstract

Flow cytometry has been extensively used to define cell populations in immunology, hematology and oncology. Here, we provide a detailed description of protocols for flow cytometric analysis of the cluster of differentiation (CD) surface antigens and intracellular antigens in neural cell types. Our step-by-step description of the methodological procedures include: the harvesting of neural in vitro cultures, an optional carboxyfluorescein succinimidyl ester (CFSE)-labeling step, followed by surface antigen staining with conjugated CD antibodies (e.g., CD24, CD54), and subsequent intracellar antigen detection via primary/secondary antibodies or fluorescently labeled Fab fragments (Zenon labeling). The video demonstrates the most critical steps. Moreover, principles of experimental planning, the inclusion of critical controls, and fundamentals of flow cytometric analysis (identification of target population and exclusion of debris; gating strategy; compensation for spectral overlap) are briefly explained in order to enable neurobiologists with limited prior knowledge or specific training in flow cytometry to assess its utility and to better exploit this powerful methodology.

Introduction

Flow cytometry has been extensively exploited in immunology, hematology and oncology to define cell populations via intrinsic scatter properties, cell surface antigen expression, and other fluorescence parameters^1-3. Our insights into blood lineage development and disease are a result to a significant degree of the continuous refinement of this methodology after its initial implementation^4,5. Increased awareness of the quantitative and overall analytic potential of flow cytometry has recently encouraged its more widespread use in stem cell research and may enable similarly profound progress in a shorter time frame⁶. However, the application of flow cytometry to specifically analyze and isolate neural populations has long been perceived as challenging. In contrast to hematopoietic cells that naturally exist in suspension, neural cell types are typically harvested from excessively complex sources that may include glia and various other surrounding cells as well as an intricate network of process-bearing neurons. Consequently, neurobiology has yet to implement the versatility of flow cytometry to its complete potential in daily research routines. However, as long as a viable single cell suspension can be generated (and protocols have been devised and optimized for that purpose⁷), flow cytometry and fluorescence-activated cell sorting (FACS) can be considered a valuable element of the analytical repertoire in neurobiology^8-11.

Figure 1. Principle of flow cytometric analysis and components of a flow cytometer. Flow cytometers comprise three main systems: fluidics, optics and electronics. A streamlined flow of cells in suspension (prepared from primary tissue or in vitro culture) is accomplished by the sheath fluid via hydrodynamic focusing, restricting the sample to its center core. The optical parts are composed of lasers that illuminate the stream of cells and optical filters that direct the signal to the appropriate detectors. The light signals detected are converted to electronic signals, subsequently processed by a computer and visualized on a monitor for data analysis and gating. Please click here to view a larger version of this figure.

Users of flow cytometric methods profit from at least a basic understanding of the underlying fundamentals including a cytometer’s building blocks (for review see^12,13; also see Figure 1). A laser beam intersects with a hydrodynamically focused fluidic stream that contains the cells in suspension, which in turn pass through the laser beam in ‘single file’ one after another. The interception of a cell (or any other particle, for that matter) with the laser results in the scattering of light from this interrogation point. Scattered light can be detected in continuation of the laser direction (forward scatter, associated with the size of the particle), as well as perpendicular to its direction (side scatter; reflecting the granulosity of the particle/cell). These aforementioned scatter properties do not require specific labeling, which is why an unlabeled sample (or also cellular debris, air bubbles, etc.) will generate a signal (event) on the bivariate forward scatter versus side scatter plot commonly used for initial gating. By using the appropriate lasers and filters specific for the corresponding excitation and emission spectra, a cell can be analyzed for its positivity, level of intensity, or absence of fluorescent markers. The majority of flow cytometric applications have focused on characterization via cell surface antigens. Unlike the hematopoietic lineage, the neural lineage has remained less extensively defined according to surface epitope expression patterns⁵. One advantage of exploiting surface antigens is that live cells can be subjected to cell sorting paradigms such as FACS. In contrast, intracellular antigen staining requires fixation and permeabilization steps to mediate the epitope-antibody interaction, precluding downstream applications that require viable cells. Of note, such approaches still allow for numerous quantitative assays¹⁴ as well as downstream analyses for RNA and protein expression¹⁵. Hematology, immunology and oncology have often used more than a dozen markers in conjunction to define particular subpopulations¹⁶. Additionally, mass cytometry or CyTOF can now be used to analyze up to 30 parameters simultaneously^17,18.

For neural stem cell applications as well as primary cultures^14,19,20 the heterogeneity of cells in vitro is a common phenomenon^21-23. The cells not representing the target population of interest embody a potentially confounding factor for experimental readout^24,25. Conveniently, the different cellular subsets present within a heterogeneous cell suspension bear distinct (known or yet to be deciphered) antigen expression profiles, which can be utilized to define these various populations. Flow cytometry can hence play a crucial role in resolving cellular heterogeneity and, thereby, facilitate biomedical applications (in vitro assays, cell therapy) and optimize quantitative readout by focusing on the most relevant subset^24,26. Various surface antigen combinations have been identified over the past few years to allow the quantitation and isolation of specific neural cell types. This includes CD133 for the enrichment of neural stem cells²⁷, a combination of the CD15/CD24/CD29 surface antigens for the isolation of NSC, differentiated neuron and neural crest cells²⁸ or CD15/CD24/CD44/CD184/CD271 to isolate neural and glial subsets²⁵, among other signatures^29,30. Beyond neurons, glial markers include A2B5³¹, CD44²⁵, NG2³² and GLAST³³. A recent publication has exploited the midbrain floorplate precursor marker CORIN^34,35 to enrich for dopaminergic precursors in Parkinson cell transplantation paradigms³⁶. CD molecules are not only markers, but functionally relevant mediators of cell-cell interactions and of a cell’s ability to respond to cues from extracellular matrix molecules and growth factors³⁷. One strategy of further enhancing the arsenal of combinatorial CD antigens to characterize neural lineage development is to use known intracellular markers to screen for and define CD antigen combinations for a particular cell type of interest. We have recently exploited such an approach and identified CD49f^–/CD200^high combinatorial expression patterns as a novel approach for enriching neuronal subsets from neurally differentiated induced pluripotent stem cell culture systems³⁸. Here, we include and discuss the latter protocol (and optional variations thereof) in which surface staining and intracellular staining can be used simultaneously for defining neural cell subpopulations by flow cytometry.

Figure 2. Flow diagram of experimental protocol options. The figure depicts a schematic representation of the major steps involved in the protocol. Optional steps (CFSE dye or intracellular antigen labeling) are indicated by light grey boxes. After harvesting, it is essential to assess the viability and cell number of neural cell suspensions prior to cell surface staining. Positive as well as negative controls need to be included in addition to the samples of interest. Samples can be analyzed by flow cytometric analysis and/or used in cell sorting paradigms. Please click here to view a larger version of this figure.

While we have previously used primary antibody in combination with secondary antibody for intracellular staining³⁸, we now introduce non-covalent labeling of the primary antibody via fluorescent Fab fragments (Zenon labeling) as a slight variation, thereby reducing the steps of cell manipulation³⁹. Moreover, as a further example of the protocol’s versatility, we employ an optional labeling of one experimental subset by carboxyfluorescein succinimidyl ester (CFSE) prior to surface antigen staining. Such CFSE pre-labeling enables the immediate direct comparison of two cell lines or experimental conditions (CFSE-labeled vs. unlabeled) within a single sample tube, reducing variance or subtle differences in incubation time and saving antibody. CFSE is an established fluorescent dye that is commonly used for cell tracking⁴⁰, in proliferation^41,42 and barcoding experiments^43,44. Finally, while actual sorting steps (FACS, immunomagnetic cell separation or immunopanning) are not part of this protocol, in principle, the harvesting and labeling procedures described here do yield samples that can be subjected to surface antigen- or intracellular labeling-based sorting applications^{15 25,28}.

With this article, we aim to: summarize a viable surface antigen staining protocol^25,28, summarize a protocol for detection of intracellular targets as well as combined surface and intracellular antigen analysis³⁸, present an intracellular CFSE dye labeling step^41,45 as an experimental option for comparative analyses of neural cell populations, and summarize approaches to flow cytometric analysis (appropriate controls^13,46, gating strategy and data presentation⁴⁷).

Protocol

1. Neural Cell Harvesting

1. Assessment by microscope:
   1. Prior to initiating an experiment, check the status of the culture with bright-field or phase contrast microscopy.
      NOTE: While primary neural tissue obtained from dissections is, in principle, equally amenable to flow cytometric analysis^14,28, please note that the protocol’s focus is on cells obtained from in vitro neural cell systems.
2. Harvesting cells⁷:
   1. Gently wash the dish/flask of adherent cells with Mg²⁺/ Ca²⁺ free phosphate-buffered saline (PBS) at RT (e.g., 5 ml for T75 flask, 3 ml per well for 6-well plate, or 10 ml for 10 cm dish).
      NOTE: The PBS used throughout the protocol is Mg²⁺/ Ca²⁺ free.
      1. For the video example, use a T75 flask of SH-SY5Y neuroblastoma cells at 80% confluency. Apply additional washing steps in cases where considerable debris is present in the dish.
      2. Consider washes with serum albumin-containing PBS⁴⁸, Percoll, or Ficoll⁴⁹ centrifugation gradients and/or commercially available beads^50,51. Removal of myelin and other lipid or other contaminants is critical, particularly when adult primary tissue sources are being used.
   2. Add pre-warmed (37 °C) trypsin replacement at an appropriate volume that covers the entire surface of the tissue culture vessel.
      NOTE: Alternatively, consider Accutase or other enzymatic digestion options. This critical step may negatively affect surface epitope expression (see ⁷).
   3. Incubate the dish/flask at 37 °C for 2 - 5 min (depending on the cell type) to allow cells to detach. Gently tap the tissue culture vessel or flush with a serological pipet to dislodge the cells. Avoid over digestion (as this can cause cell loss and clotting at later steps).
   4. Quench the trypsin replacement by adding twice the volume of flow buffer (2% FBS in PBS) and collect the cells in a 15 ml conical tube.
   5. Gently triturate the cell suspension using a microliter pipet (100 - 1,000 µl) or a 5 ml serological pipet to prepare a single cell suspension.
   6. Centrifuge the cells at 220 x g for 5 min at 25 °C. Carefully aspirate the supernatant leaving the pellet behind.
   7. Re-suspend the pellet in an appropriate volume of flow buffer, depending on the size of the pellet (e.g., for one confluent T75 flask of SH-SY5Y cells the typical yield is at least 10 x 10⁶ cells, in which case the cells are resuspended in 5 ml of flow buffer).
      NOTE: If larger chunks or clotting are observed, filter through a 30 - 100 µm mesh.
3. Cell counting⁵²:
   1. Transfer a small aliquot of the cell suspension to a microcentrifuge tube and dilute at a defined ratio in a volume of trypan blue or an alternative viability dye before transfer to a hemocytometer or automated cell counting system.
   2. Dilute the cell suspension to a concentration of 1 x 10⁶ viable cells/ml by adding the appropriate volume of flow buffer or PBS with 0.1% BSA (if proceeding with CFSE labeling).
      NOTE: Propidium iodide, 7-aminoactinomycin D, annexin V and commercially available fixable viability assay kits represent alternative options in order to assess the viability of cells. Also, apoptosis assays using caspase-3 fluorescence as described previously⁵³ may be used. Fluorescent channels will be “occupied” by these reagents which may limit the options for subsequent steps if included in all samples.

2. Intracellular Dye Labeling Using CFSE (Figure 3)

1. Dilute the CFSE to a desired stock concentration that can be easily used.
   NOTE: For these experiments a stock concentration of 0.01 mM was used. Determine the optimal working concentration of CFSE empirically.
2. Add 10 µl of 0.01 mM CFSE solution per ml of cells (Section 1.3.2, concentration of 1 x 10⁶ cells/ml in PBS + 0.1% BSA) for a final concentration of 0.1 µM. Briefly vortex to mix well.
   NOTE: CFSE concentrations used here are about ten-fold lower than the ones commonly applied in proliferation assays, hence cell toxicity of the dye is minimal. We do not observe negative effects on cell viability.
3. Incubate for 5 min at RT with constant shaking (200 rpm). Protect from light.
4. Quench the dye by adding 5 volumes of flow buffer to the tubes. Centrifuge at 94 x g for 5 min at RT.
5. Discard the supernatant leaving the pellet behind. Re-suspend the cells with 5 volumes of flow buffer.
6. Centrifuge at 94 x g for 5 min at RT. Discard the supernatant and re-suspend the cells in flow buffer at a concentration of 1 x 10⁶ cells/ml.
7. Add an equal number of unstained cells of interest to the stained cell suspension. Proceed to surface antigen staining protocol (Section 3).

Figure 3. Detection of differential CD surface antigen expression between two cell lines via CFSE dye labeling. SH-SY5Y neuroblastoma cells are pre-labeled with CFSE for subsequent identification in comparison to unlabeled BJ fibroblasts. Co-staining of the mixed sample (right panels) with surface markers CD24 or CD54 (both conjugated to APC) demonstrates that cell lines are easily distinguishable due to the CFSE staining (arrows = SH-SY5Y; arrowheads = BJ fibroblasts). The majority of SH-SY5Y cells express CD24 but not CD54 (ICAM-1). In contrast, BJ fibroblasts (CFSE-negative) are positive for CD54 but largely negative for CD24. Please click here to view a larger version of this figure.

3. Cell Surface Staining

1. Label tubes including samples and critical controls (see Table 1).

Tube no.	Sample name	Antigen-fluorophore	Dilution
1	Unstained cells	-
2	Single stained	CD24-APC	1:50
3	Single stained	TUJ1-Alexa fluor 488nm	1:2,000
4	Double stained	CD24-APC	1:50
		TUJ1-Alexa fluor 488nm	1:2,000
5	Single stained	Secondary only: Alexa fluor 488nm	1:2,000

Table 1. List of tubes to be included in a typical flow cytometry experiment. The table shows a minimal set of sample tubes required for a co-staining experiment described in this video article. An ideal experiment needs to include all necessary controls (negative, positive as well as compensation controls) for accurate interpretation of the results obtained.

2. Add 100 µl of cell suspension (from Section 2.7 or 1.3.2) to each 1.5 ml microcentrifuge tube.
   NOTE: Make sure a minimum of 0.1 x 10⁶ cells are present per 100 µl of cell suspension.
3. Add fluorophore conjugated antibody to the sample at an appropriate dilution.
   NOTE: Determine working dilution for each antibody prior to the experiment. See Table 2 for a list of neural surface antigens.

Antigen	Cell Type	Reference
CD15	Neural stem cells	[28, 67]
CD24	Neuronal cells	[28, 68]
CD29	Neural stem cells	[28, 69, 70]
CD44	Glial cells	[25]
CD49f	Neural stem cells	[38]
CD56 (NCAM)	Neuronal cells	[71]
CD133	Neural stem cells	[27]
CD184	Neural stem cells and glial cells	[25]
CD200	Neuronal cells	[38]
CD271	Neural crest stem cells	[25]
A2B5	Glial cells	[31]
CORIN	Dopaminergic precursors	[35, 36]
FORSE1	Neural stem cells (NSC)	[72 ]
GLAST	Glial cells	[33]
NG2	Glial cells	[32]

Table 2. Selection of neural surface antigens. This table provides a list of surface epitopes found to be expressed by various neural cell types to exemplify the increasing panel of surface antigens used to characterize the neural lineage. Note that this selection is far from being complete and that most of these markers are also expressed by a range of other neural and non-neural cells. Consequently, combinations of several markers will be required to better define and isolate the indicated neural subsets.

4. Incubate for 30 min on an orbital shaker (200 rpm) in the dark.
5. Flow buffer wash
   1. Add 1 ml of flow buffer to the tubes. Centrifuge at 380 x g for 4 min at 4 °C.
   2. Discard the supernatant leaving the pellet behind.
6. Repeat the wash step again.
7. After the second wash, decant the supernatant and re-suspend the cells in flow buffer to a final volume of 100 µl.
8. Use sample for flow cytometric analysis. Alternatively, proceed with Section 4 and 5.
   NOTE: If cells are to be sorted and put back into culture post-FACS (i.e., viable cell suspension without fixation or permeabilization), apply aseptic techniques during the harvest, staining and analytical steps.

4. Fixation and Permeabilization³⁸

1. Fixation using paraformaldehyde (PFA):
   1. Prepare fixation buffer containing 2% PFA in PBS.
      NOTE: PFA is harmful to humans and the environment. Use appropriate personal protective equipment and discard waste in accordance with local regulations.
   2. Add 500 µl of fixation buffer to 100 µl of cell suspension.
   3. Incubate the tubes at RT for 15 min on an orbital shaker (100 rpm) in the dark.
2. PBS wash:
   1. Add 1 ml of PBS to the tube. Centrifuge at 380 x g for 3 min at 4 °C.
   2. Discard/ decant the supernatant leaving approximately 100 µl in the tube.
3. Permeabilization with Tween-20:
   1. Prepare permeabilization buffer containing 0.7% Tween-20 in PBS.
   2. Add 500 µl of permeabilization buffer to 100 µl of cell suspension.
   3. Incubate the tubes at RT for 15 min on an orbital shaker (100 rpm) in the dark.
4. Wash the cells once with PBS (as described in Section 4.2) and remove the supernatant from the tubes completely, leaving only the pellet behind.

5. Intracellular Antigen Staining³⁸ (Figure 4)

1. Preparation of primary antibody solutions:
   1. Dilute the primary antibodies in dilution buffer containing 1% bovine serum albumin, 10% serum (e.g., normal donkey or goat serum) and 0.5% Tween-20 in PBS.
      NOTE: Choose the serum to be used depending on the species the secondary antibodies were raised in.
   2. Alternatively, use Zenon fluorescein labeling of the primary antibody according to the manufacturer’s instructions.
      1. Prepare 1 µg of primary antibody in PBS at an appropriate dilution (total volume ≤ 20 µl).
      2. Add 5 µl of Zenon fluorescein IgG labeling reagent (Component A) to the antibody solution.
      3. Incubate the mixture for 5 min at RT.
      4. Add 5 µl of Zenon blocking reagent (Component B) to the reaction mixture.
      5. Incubate the mixture for 5 min at RT. Apply the antibody to the sample within 30 min.
2. Primary antibody staining:
   1. Add 100 µl of the primary antibody solution to the cell pellet and gently triturate to mix.
   2. Alternatively, add Zenon fluorescein labeled antibody to the cell suspension at the appropriate dilution.
   3. Incubate the tubes at RT for 30 min on an orbital shaker (200 rpm), protected from light. Wash the cells once with PBS (as describes in Section 4.2) and remove the supernatant from the tubes completely, leaving only the pellet behind.
3. Secondary antibody staining (not required for the Zenon fluorescein labeled antibodies):
   1. Dilute the secondary antibodies in PBS at an appropriate concentration.
   2. Add 100 µl of the secondary antibody solution to the cell pellet and gently triturate to mix. Incubate the tubes at RT for 30 min on a shaker (200 rpm) in the dark.
4. Wash the samples two times with PBS (as described in Section 4.2).
5. Wash once with flow buffer (see Section 3.5).
6. Re-suspend the cells in approximately 150 µl of flow buffer and analyze on the flow cytometer.

Figure 4. Co-staining of surface and intracellular proteins. Flow cytometry data exemplifies a comparison between primary + secondary antibody-based versus Zenon fluorescein-based intracellular staining in combination with surface staining. Exclusive positivity on the y-axis (upper left quadrant) and the x-axis (lower right quadrant) shows cells stained for MAP2 and CD24, respectively. After co-staining, shared MAP2 and CD24 expression can be seen in the upper right quadrant (right panels). Comparison of using Alexa fluor 488 (top; AF488) versus Zenon fluorescein (bottom) for MAP2-labeling yields similar results. Please click here to view a larger version of this figure.

6. Flow Cytometric Analysis

1. Conduct flow cytometric analysis immediately after the completion of the staining protocol using a flow cytometer with appropriate filters for signal detection. Use 488 nm blue and 640 nm red laser with FL-1 (533/30), FL-2 (585/40), and FL-4 (675/25) bandpass filters.
2. Set up primary gates based on the forward and side scatter excluding debris and dead cells.
3. Set fluorescence gates for surface and intracellular antigen to ≤0.5% based on the unstained samples and compensation for spectral overlap using single stained controls.

Results

The protocol presented here allows for versatile experimental approaches (Figure 2). In its shortest version (steps 1, 3 and 6), it can be considered a guide for simple staining of surface antigens.  In its more complex form, a number of co-labeling paradigms with a range of intracellular antigens can be pursued (optional steps 2 and/or 4 to 5). Moreover, the CFSE-labeling step exemplifies its utility for cell label...

Discussion

The protocol presented here is well established for neural cell cultures derived from human stem cells, but can be equally applied to other neural cell sources including primary tissue or neural cell lines. In addition to embryonic sources, neural stem or progenitor cells can be extracted from the neurogenic regions of adult brain²⁷. Moreover, flow cytometry and FACS can be exploited to quantify, analyze and isolate various cell populations including mature neurons⁵⁴, astroglia³³, microgl...

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
DMEM/F12 (1:1) (Dulbecco's Modified Eagle Medium/Nutrient Mixture F-12)	Life Technologies	11330057
DPBS without Ca2+ Mg2+	Life Technologies	14190169
Fetal bovine serum, qualified, E.U.-approved, South America origin (FBS)	Life Technologies	10270-106
MEM Non-essential amino acids (100x)	Life Technologies	11140035
TrypLE Express	Life Technologies	12604013
Trypan blue solution, 0.4%	Life Technologies	15250061
Paraformaldehyde	Carl Roth	335.3
Bovine serum albumin (BSA) Fraction V	PAA Laboratories, Coelbe	K41-001
Tween-20 Detergent	Calbiochem	655205
Carboxyfluorescein succinimidyl ester (CFSE)	eBioscience	65-0850-84
DMSO	AppliChem	A1584
Bottle top filters express plus 0.22 µm, 250 ml	Millipore	SCGPU02RE
Cell culture treated flasks (T 25)	NUNC	156367
Cell culture treated flasks (T 75)	NUNC	156499
Conical tubes (15 ml)	Greiner Bio-One	188271
Conical tubes (50 ml)	Greiner Bio-One	227261
Pasteur pipet, glass (150 mm)	STEIN Labortechnik, Remchingen	S03710150
Pipet tips (0.1-10 µl)	Corning	4125
Pipet tips (1-200 µl)	Corning	4126
Pipet tips (100-1000 µl)	Corning	4129
Serological pipets, 5 ml	Corning	4051
Serological pipets, 10 ml	Corning	4101
Serological pipets, 25 ml	Corning	4251
Microcentrifuge tubes (0.5 ml)	Sarstedt	72,699
Microcentrifuge tubes (1.5 ml)	Greiner Bio-One	616201
Microcentrifuge tubes (2.0 ml)	Sarstedt	72,695,500
Anti-Human CD24 APC monoclonal antibody	eBioscience	17-0247-42	Working dilution 1:50
Anti-Human CD54 PE monoclonal antibody	eBioscience	12-0549-42	Working dilution 1:50
Neuronal Class III β-Tubulin (Tuj1) polyclonal antibody	Covance	PRB-435P	Working dilution 1:2,000
Alexa Fluor 488 Donkey anti Rabbit	Life Technologies	A21206	Working dilution 1:2,000
Zenon® Fluorescein Rabbit IgG Labeling Kit	Life Technologies	Z-25342
Neubauer-Improved counting chamber	Marienfeld	0640010
Vortex	Scientific Industries	G560E
Thermomixer comfort	Eppendorf	5355 000.001
Accuri C6 flow cytometer	Becton Dickinson (BD)	653118
Microcentrifuge refrigerated, PerfectSpin 24 R	Peqlab	91-PSPIN-24R
Orbital shaker, Unimax 1010	Heidolph	543-12310-00
Centrifuge refrigerated, Rotanta 96 RC	Hettich	4480-50
Class II Biological safety cabinet Safe 2020	Thermo Scientific	51026640
CO2 Incubator, Heracell 240i	Thermo Scientific	51026331
Vacuum system, Vacusafe comfort	Integra Biosciences	158320
Microscope, Axiovert 40 CFL	Zeiss	451212
Pipet controller, accu-jet pro	Brand	26303
Micropipet, Pipetman neo P20N (2-20 µl)	Gilson	F144563
Micropipet, Pipetman neo P200N (20-200 µl)	Gilson	F144565
Micropipet, Pipetman neo P1000N (100-1000 µl)	Gilson	F144566