Skip to main content

Cellometer Auto 2000 Cell Viability Counter

The Cellometer™ Auto 2000 integrated touchscreen, dual-fluorescence cell counter is optimized for analysis of primary cells from peripheral blood, cord blood, bone marrow, and other complex samples. The counter features preconfigured one-touch assays for a variety of samples.

For research use only. Not for use in diagnostic procedures.
Part number: CMT-A2K
Unit Size: 1 unit

Overview

The Cellometer Auto 2000 utilizes brightfield imaging and dual-fluorescence imaging to quickly and accurately identify and count individual cells. Cell count, concentration, diameter, and % viability are automatically calculated and reported.

The Cellometer Auto 2000 enables users to:

  • Increase throughput
  • Increase accuracy
  • Improve consistency
  • Count difficult cells (clumpy, irregular-shaped)
  • Minimize judgment errors, miscounts, interference from red blood cells and user-to-user variability

Additional product information

Primary cell analysis: PBMCs, stem cells, and more

The Cellometer Auto 2000 performs analysis of primary cells from peripheral blood, cord blood, bone marrow, and other complex samples for use in a wide range of research areas, including: 

  • Nucleated cells for transplantation
  • PBMCs for immunology
  • Splenocytes for vaccine development
  • Stem cells for cellular therapy
  • Tumor cell suspensions for oncology

Dual-color fluorescence allows for staining of live and dead nucleated cells, generating accurate viability results even in the presence of debris, platelets, and red blood cells. Accurate analysis of both ‘messy’ and 'clean' samples enables the Cellometer Auto 2000 to evaluate samples at a variety of points throughout sample processing - from initial collection to separation, to cryopreservation.

One-touch assays for analysis of a wide range of primary samples, including:

Nucleated cells from whole blood

2000 icon immune cell high rbc

Stem cells

2000 icon stem cells

PBMCs / Splenocytes following separation

2000 icon immune cells low rbc

Tumor and tissue cell suspension

2000 icon primary cells cell lines
Reduced interference from red blood cells, platelets, or debris

The dual-fluorescence AO/PI method utilizes nuclear staining dyes that bind to nucleic acids in the cell nucleus. Because most mature mammalian red blood cells do not contain nuclei, only live and dead mononuclear cells produce a fluorescent signal. There is no need to lyse red blood cells, saving time and removing an extra sample preparation step. Red blood cells, platelets, and debris are not counted in the fluorescent channels.

The advantage of fluorescent counting for primary cells

These images (right) demonstrate the advantage of fluorescent counting for primary cells. The brightfield image shows the combination of nucleated cells, red blood cells, and platelets present in the sample. Only the live and dead nucleated cells are visualized and counted in the green and red fluorescent channels.

Sample Measurement Total nucleated All RBC % RBC n
PBMC+RBC

Mean

CV

1.26E+07

6.2%

1.39E+07

8.8%

1.23E+06 8.9% 10
PBMC+1/2RBC

Mean

CV

1.21E+07

4.8%

1.27E+07

5.4%

5.82E+05 4.6% 10
PBMC+1/4RBC

Mean

CV

1.22E+07

7.9%

1.24E+07

7.3%

2.27E+05 1.8% 10

 

Fresh human PBMCs (peripheral blood mononuclear cells) were spiked with varying amounts of RBCs (red blood cells.) All cells (nucleated + RBC) were counted in the brightfield channel. Nucleated cells were then counted in the green fluorescent channel. Varying amounts of RBCs (1.8%, 4.6%, and 8.9%) did not affect the nucleated cell count.

2000 exclude red blood cells

Several red blood cells are indicated in the brightfield image (above, left). The red blood cells are not visible in the fluorescent image (above, right) detecting cells stained with nuclear staining dye.

Live / dead nucleated cell counts using dual-fluorescence

Why dual-fluorescence?

Because brightfield cell counting does not differentiate nucleated from non-nucleated cells and trypan blue staining is not as easy to detect as fluorescent staining, dual-color fluorescence is strongly recommended for accurate viability analysis for primary cells. The Auto 2000 is equipped with standard assays for dual-fluorescence analysis of primary cells stained with AO/PI.

The AO/PI method

Acridine Orange, AO, is a nuclear staining (nucleic acid binding) dye permeable to both live and dead cells. It stains all nucleated cells to generate green fluorescence. Propidium iodide, PI, can only enter dead cells with compromised membranes. It stains all dead nucleated cells to generate red fluorescence. Cells stained with both AO and PI fluoresce red due to quenching, so all live nucleated cells fluoresce green and all dead nucleated cells fluoresce red.

Green fluorescent live cell image

2000-aopi-method-live-cells-200x190.png

Red fluorescent dead cell image

2000 aopi method dead cells

 

 
Cellometer precision

The Cellometer Auto 2000 offers excellent reproducibility, with a %CV (Coefficient of Variation) of <10% for fluorescent concentration and viability measurements. The data (below) is based on four preparations of Jurkat cells stained with propidium iodide, a fluorescent nuclear-staining dye.

Sample N value Average live cell concentration % viability CV of concentration CV of viability
A 4 4.20E+06 91.1 10% 2%
B 4 1.06E+06 22.7 7% 1%
C 4 3.27E+06 57.5 7% 7%

 

Cell line total cell concentration

The results indicate the accuracy of the Cellometer Auto 2000 instrument cell concentration. Serial dilutions (2-, 4-, and 8-fold) were prepared using PBS and Jurkat cells. Four measurements were performed for each sample and the averages were calculated and plotted. A linear correlation between the concentration measurements and 1/dilution factor was observed. The results show the reliability and accuracy of the Cellometer Auto 2000 in measuring cell concentration.

2000 stem primary cell line

Jurkat cell concentrations at various sample dilution factors. N=4. Error bars correspond to the Standard Deviations for cell concentration.

Sample Dilution factor Average cell concentration CV of concentration Standard deviation N value
A 1 5.80E+06 1% 4.8E+04 4
B 2 3.39E+06 2% 8.12E+04 4
C 4 1.63E+06 6% 9.54E+04 4
D 8 8.27E+05 6% 5.33E+04 4

Table of results for total cell concentration for the Cellometer Auto 2000 instrument.
 

Specifications for Cellometer Auto 2000 cell viability counter
Includes

Cellometer Auto 2000 instrument with integrated computer and touch-screen and dual-fluorescent optics

Cellometer software

Two USB 2.0 ports

Power supply

Ethernet port

Phone/online applications support during set-up

Imaging performance

Cell Size: 5 - 300* microns

Conc. Range: 105 – 107 cells/ml brightfield imaging, fluorescent imaging and pattern-recognition software to quickly and accurately decluster, identify and count individual cells.

*Cellometer CHT4-PD300 Slides are required for cells > 80 microns in diameter

Instrument specifications

Weight: 26 lbs. (11.8 kg)

Dimensions: Width: 11.1″ (28.3 cm)

Depth: 12.8″ (32.4 cm)

Height: 13.4″ (34.0 cm)Input to Power Adapter: 100-240 VAC, 50/60 Hz, 1.5A

Output to instrument: 12 VDC, 7.5A

Display 10.4″ touch screen
Fluorescent optics

Excitation / Emission: 470 nm/535 nm

Example fluorophores:

  • Acridine orange (+DNA)
  • Calcein AM
  • CFDA
  • SYTO® 9
  • SYTO® 13

Excitation / Emission: 540 nm/605 nm

Example fluorophores:

  • Propidium iodide
  • Ethidium bromide

Specifications

Brand
Cellometer
Unit Size
1 unit

Resources

7-12 of 25 Resources
Application Note Icon Image
Application Note
Comparison of Trypan blue and fluorescence-based cell viability detection methods via morphological observation.
Technical Note
Concentration and viability measurement of canine stromal vascular fractions using the Cellometer image cytometer.
Application Note Icon Image
Application Note
Detection of multiplexed GFP reporters in primary articular cultures using the Cellometer automated cell counter.
Application Note Icon Image
Application Note
Enumeration and viability of nucleated cells from bone marrow, cord blood and mobilized peripheral blood using the Cellometer Auto 2000 cell viability counter
Whitepaper
Exploring factors impacting and improving cell counting precision

Cell counting precision is a critical parameter for determining the overall quality of cell counting results. Many factors...

Application Note Icon Image
Application Note
Isolation, quantitation and viability analysis of neonatal cardiomyocytes using the Cellometer Auto 2000 cell viability counter.
SDS, COAs, Manuals and more Illuminator

SDS, COAs, Manuals and more

Are you looking for technical documents for this product. We have housed them in a dedicated section., click on the links below to explore.

Scroll Icon