MEASUREMENT OF CELL VIABILITY
Volume 40, June 2000 TRANSFUSION 693
A reliable method for rapidly determining the vi- ability of HPCs is essential for clinical cellular en- gineering. The trypan blue (TB) viability assay is a widely performed procedure to assess HPC vi-
ability before and after cryopreservation and before bonemarrow transfusion.1 The TB assay is based on the abilityof the membrane of viable cells to exclude the dye, whilenonviable cells are stained blue when viewed by light mi-croscopy. However, the TB assay can be difficult to interpretbecause of staining artifacts. In addition, the TB viabilityassay has never been validated for use with HPCs.
Alternative viability assays using the fluorescent dyesacridine orange (AO) and propidium iodide (PI) have beendeveloped for the simultaneous visualization of both viableand nonviable cells.2,3 AO is a membrane-permeable, cat-ionic dye that binds to nucleic acids of viable cells and thatat low concentrations causes a green fluorescence. PI isimpermeable to intact membranes but readily penetratesthe membranes of nonviable cells and binds to DNA orRNA, causing orange fluorescence. When AO and PI areused simultaneously, viable cells fluoresce green and non-viable cells fluoresce orange under fluorescence micros-copy. Results of the AO/PI assay in pancreatic islet cells havebeen shown to correlate with those of the TB viability as-say but have been easier to interpret.4
Here we describe a rapid fluorometric assay for HPCviability using AO/PI dyes and compare the results of this
HPC viability measurement:trypan blue versus acridine orange and propidium iodide
K. Mascotti, J. McCullough, and S.R. Burger
BACKGROUND: A reliable, validated method for rapidlydetermining HPC viability is essential for clinical cell en-gineering.STUDY DESIGN AND METHODS: A fluorometric cellviability assay using acridine orange and propidium io-dide (AO/PI) was compared to the current standard,trypan blue (TB) exclusion. Viable cells stained with AO/PI fluoresce green under darkfield fluorescence micros-copy, while nonviable cells fluoresce orange. Mixtures offresh and heat-killed bone marrow were prepared andused as viability standards for evaluation of both assays.The frequency of CFU–GM was determined for eachspecimen.RESULTS: Cell viability measured by AO/PI was ex-tremely linear, with measured and predicted viability inagreement from 0 to 100 percent of the viable cells anda coefficient of regression (r2) of 0.9921. The predicted-viability regression line fell within the 95% CI for AO/PI-measured viability. The coefficient of regression for TB-measured viability was 0.9584, with the predicted-viabilityregression line almost entirely outside the 95% CI. TBoverestimated the percentage of viable cells, particularlybelow the 50-percent level. CFU–GM frequency corre-lated better with cell viability measured by AO/PI (r2 =0.979) than with that measured by TB (r2 = 0.930).CONCLUSIONS: The AO/PI viability assay is a rapid,highly linear, functionally correlated assay that is supe-rior to conventional viability measurement by TB exclu-sion. ABBREVIATIONS: AO = acridine orange; α-MEM = minimum
essential medium alpha modification; PI = propidium iodide; r2
= coefficient of regression; TB = trypan blue.
From the Department of Laboratory Medicine and Pathology,
University of Minnesota Medical School, Minneapolis, Minne-
sota.
Address reprint requests to: Scott R. Burger, MD, Director,
Cell Therapy Clinical Laboratory, University of Minnesota, Box
609, Mayo Building, 420 Delaware Street SE, Minneapolis, MN
55455; e-mail: [email protected].
Received for publication October 21, 1998; revision re-
ceived August 23, 1999, and accepted September 2, 1999.
TRANSFUSION 2000;40:693-696.
B L O O D C O M P O N E N T S
MASCOTTI ET AL.
694 TRANSFUSION Volume 40, June 2000
assay with those of the TB viability assay. Both viability as-says were validated by the use of specimens of known vi-ability and clonogenic assays for HPCs. In addition, thestability over time of the TB and AO/PI staining was determined.HPCs derived from peripheral or umbilical cord blood werenot tested.
MATERIALS AND METHODS
Cells
Bone marrow was collected from the posterior iliac crest ofa normal donor under institutional review board approvaland by using standard techniques.5 Fresh, heparinized bonemarrow was washed into minimum essential medium al-pha modification (α-MEM, GIBCO, Grand Island, NY) with1-percent HSA (Baxter-Hyland, Glendale, CA). The cell sus-pension was divided into two equal aliquots of 10 mL. Cellsin one aliquot were heat-killed at 65°C × 15 minutes andthen washed and resuspended in a total volume of 10 mLof α-MEM with 1-percent HSA. These two aliquots weremeasured by both the AO/PI and TB assays; the cells in theheat-killed aliquot were 100-percent nonviable, and thosein the other aliquot were 100-percent viable. Cell concen-trations were measured in each aliquot and adjusted to anidentical value, 9.4 × 106 cells per mL.
Calibrated automated electronic pipettors (Rainin Elec-tronics, Woburn, MA) were used to prepare seven 1-mL cellsuspensions from these two stock aliquots. The suspen-sions represented predicted percentages (0, 5, 25, 50, 75, 95,and 100%) of viable cells. Cell viability for each of theseseven aliquots was evaluated by both TB exclusion and AO/PI staining. The percentage of viable cells was calculated as theaverage of five viability measurements, each of which re-quired the scoring of 100 cells. Measurements were repeatedfive times to assess the reproducibility of viability measure-ment. Measurements were performed by two individuals.
Assays
TB viability measurement was performed by standardmethods.1 TB solution (0.4% wt/vol, GIBCO) was mixedwith each of the seven predicted-viability aliquots, using190 µL of stain to 10 µL of cell suspension. The suspensionwas loaded into a Neubauer hemocytometer and scoredwith a Zeiss compound light microscope at 250×. Cells thatstained blue were scored as nonviable.
Cell viability was evaluated by AO/PI staining as previ-ously described by Banks.2,4 AO/PI stock solution was preparedas 1 mM AO and 1 mM PI in PBS (Sigma Chemical Co., St. Louis,MO). Samples of each of the predicted-viability aliquots (10 µL)were mixed with 190 µL of AO/PI stock solution. The suspensionwas loaded into a Neubauer hemocytometer and scoredwith a Zeiss compound fluorescence microscope at 250×.Cells fluorescing green were scored as viable. Cells fluorescingorange, either fully or partially, were scored as nonviable.
Progenitor assays were performed on each predicted-viability specimen by using serum-free methylcelluloseprogenitor assay medium (MethoCult GF HC4434, StemCell Technologies, Vancouver, BC, Canada) and a standard-ized progenitor assay method.6 One person performed allprogenitor assays. Cells were suspended in the MethoCultmedium and plated in duplicate 35-mm culture dishes ateach of two plating concentrations, 2.5 × 104 and 5.0 × 104
nucleated cells per plate, with 1.1 mL dispensed per plate.Plates were incubated at 37°C in a humidified 5-percent CO2
incubator for 14 days. CFU–GM were counted with a ster-eomicroscope (SZH-ILLD, Olympus Precision Instruments,Melville, NY) at 40× power. The plating concentration giv-ing optimal growth and readability was selected for eachpair of duplicate plates, and the mean concentration ofCFU–GM per 105 nucleated cells was calculated.
The stability of the TB and AO/PI viability assays overtime was determined by repeated measurement of the per-centage of viable cells in a single, fresh marrow specimenover 120 minutes. Two cell suspensions were stained withTB and AO/PI, respectively, by the methods describedabove. The percentage of viable cells was measured at 10-minute intervals over 60 minutes and at 15-minute inter-vals over an additional 60 minutes.
Statistical methodsSoftware programs (Excel for Windows 95, version 7.0, Microsoft,Redmond, WA; SigmaPlot for Windows, version 3.06, SPSS , Chi-cago, IL) were used for data analysis. Viability measure-ments performed by each method were analyzed by linearregression, regression coefficients were compared with theideal values of 0 and 1, and 95% CIs were determined.
RESULTSThe percentage of viable cells measured by TB exclusion forthe controlled-viability specimens is depicted in Fig. 1. Theequation for the TB measured-viability regression line was
y = 0.796× + 23.437,with a coefficient of regression (r2) of 0.9584. The slope(0.796) and y intercept (23.437) of this regression line weresignificantly different from the ideal values of 1 and 0. Theregression line for predicted cell viability was almost en-tirely outside the 95% CI for TB-measured viability. The CVfor the five viability measurements performed on eachspecimen ranged from 0 to 63.3 percent. TB consistentlyoverestimated the percentage of viable cells, particularly atcell viabilities below 50 percent.
The percentage of viable cells measured with AO/PI isshown in Fig. 2. Viability measured by AO/PI was extremelylinear, with measured and predicted viability consistentlyin agreement for the entire range of viability measurement,from 0 to 100 percent. The linear regression equation forAO/PI measured viability was
y = 1.009× + 2.148,
MEASUREMENT OF CELL VIABILITY
Volume 40, June 2000 TRANSFUSION 695
with r2 = 0.9921. Slope and intercept of this regression linewere not significantly different from 1 and 0, respectively,and the regression line for predicted viability fell within the95% CI. The CV for the five viability measurements per-formed on each specimen ranged from 1.3 to 68.8 percent.
The results of both TB and AO/PI viability assays werecompared with the frequency of CFU–GM (the actual num-ber of colonies grown from each of the two plating concen-trations), a measure of HPC function. Figure 3 depicts thepercentage of viable cells measured with TB, compared withthe frequency of CFU–GM in the same samples. The per-centage of viable cells measured by AO/PI, compared withCFU–GM frequency, is shown in Fig. 4. The measured CFU–GM frequency in each sample was proportionate to its pre-
dicted percentage of viability and was appropriately linear.The percentage of viable cells measured with TB correlatedmoderately well with CFU–GM concentration (r2 = 0.930),but it consistently overestimated HPC frequency. The per-centage of viable cells measured with AO/PI correlatedbetter with CFU–GM frequency (r2 = 0.979).
Figure 5 depicts the percentage of viable cells, mea-sured over 2 hours by TB and AO/PI. The AO/PI results weremore stable over time, consistently averaging 98.4 ± 0.67percent over 120 minutes, with a linear regression equationslope of –0.0063, which is not significantly different from anideal value of 0. TB-measured viability, by contrast, de-creased from an average of 96.2 ± 1.60 percent during thefirst 60 minutes, to an average of 89.8 ± 2.17 percent duringthe final 60 minutes with a linear regression equation slopeof –0.0862, which is significantly different from 0 (p<0.005).
Fig. 1. The percentage of viable cells measured by TB exclusion
(� ) compared with the predicted percentage of viable cells (—).
Mean ± 1 SD for n = 5 measurements, regression line (– –), and
95% CI (·····) are shown.
Percentage of viable cells predicted
Per
cent
age
of v
iabl
e ce
lls m
easu
red
Fig. 2. The percentage of viable cells measured by AO/PI fluo-
rescence (� ) compared with the predicted percentage of viable
cells (—). Mean ± 1 SD for n = 5 measurements, regression line
(– –), and 95% CI (·····) are shown.
Percentage of viable cells predicted
Fig. 3. The percentage of viable cells measured by TB exclusion
(� ) compared with the frequency of CFU–GM (� ). Mean ± 1
SD for n = 5 measurements and regression line are shown for
both the percentage of viable cells (- - -) and CFU–GM (—).
Fig. 4. The percentage of viable cells measured by AO/PI fluo-
rescence (� ) compared with the frequency of CFU–GM (� ).
Mean ± 1 SD for n = 5 measurements and regression line are
shown for both the percentage of viable cells (– –) and CFU–
GM (—).
Per
cent
age
of v
iabl
e ce
lls m
easu
red
Predicted percentage of viable cells
Mea
sure
d pe
rcen
tage
of
viab
le c
ells
CF
U–G
M per 100,000 cells
Predicted percentage of viable cells
CF
U–G
M per 100,000 cells
Per
cent
age
of v
iabl
e ce
lls m
easu
red
MASCOTTI ET AL.
696 TRANSFUSION Volume 40, June 2000
DISCUSSIONA simple, reliable, and reproducible assay for HPC viabilityis essential for clinical cellular engineering laboratories.Traditionally, the TB dye exclusion assay has been used todetermine cell viability before and after cryopreservationand before cell transfusion, but the TB assay can be diffi-cult to read and interpret. Moreover, despite widespreaduse, validation studies of the TB assay have not been pub-lished. The AO/PI assay, with its two fluorescent dyesviewed under darkfield fluorescence microscopy, permitsclear differentiation of viable and nonviable cells.
The AO/PI assay has a higher degree of linearity andits results more closely resemble predicted viability thanthose of the TB assay. The TB assay was particularly prob-lematic at lower percentages of cell viability. At 5-percentand 25-percent predicted viability, the TB viability measure-ments were 29.9 percent and 55.3 percent, respectively. Al-though most clinical specimens contain a higher percent-age of viable cells, some extensively manipulated specimenscan fall within this range. This overestimation of viabilitymeasurements by the TB assay can have clinical implica-tions.
The percentage of viable cells measured with the AO/PI assay showed better correlation with CFU–GM frequencythan did that measured by TB (r2 = 0.979 and r2 = 0.930, respec-tively). The TB assay overestimated progenitor cell function asmeasured by CFU–GM concentration. Clinical laboratoriestypically test specimens for cell viability more often than forCFU–GM frequency; it is helpful, therefore, to use a viabil-ity assay that correlates well with HPC assay results.
Fig. 5. The percentage of viable cells repeatedly measured by
AO/PI (–� –) and TB (··� ··) assays over time allows comparison
of assay stability. Regression lines (—) for each assay are
shown.
One reason that cell viability measured with AO/PI soclosely correlated with predicted viabilities and CFU–GMconcentration may be the assay’s greater readability andmore accurate interpretation. With the high background ofRBCs and debris as well as the variable staining or shadingof cells, readability and scoring of cells with the TB assay canbe ambiguous. In the AO/PI assay, RBCs and debris are notvisible during darkfield microscopy, the cellular detail andnuclear outline are distinct, and the simultaneous use oftwo dyes allows for clear differentiation of viable and nonvi-able cells. With AO/PI, therefore, there is no need to removeRBCs by lysis or ficoll density gradient centrifugation. Bothassays are rapid and easy to perform, although the AO/PIassay requires a fluorescence microscope. We believe the in-creased accuracy of results and ease of interpretation morethan compensate for the need for fluorescence microscopy.
TB is toxic to cells over time and affects the stability ofTB-stained cells.1 By contrast, the AO/PI assay has previ-ously been described as being stable over time, and AO/PIis nontoxic to isolated islet cells.2 A perfectly stable assaywould have a linear regression slope of 0 over time. The lin-ear regression equation slope for bone marrow cells stainedwith AO/PI was –0.0063 over 2 hours, which indicates anextremely stable assay. TB-measured viability was lessstable over time. The stability of the AO/PI assay permitsstaining and scoring of multiple specimens in batches andalso is helpful in training staff.
The AO/PI viability assay is a simple, rapid, highly lin-ear, functionally correlated assay that is superior to the con-ventional TB viability assay. The AO/PI assay has not, how-ever, been validated with other HPCs, such as those derivedfrom peripheral or umbilical cord blood.
REFERENCES1. Reeb BA. Dye exclusion test for bone marrow viability. In:
Areman EM, Deeg HJ, Sacher RA, eds. Bone marrow and
stem cell processing: a manual of current techniques. New
York: F.A. Davis Company, 1992:403-4.
2. Bank HL. Assessment of islet cell viability using fluorescent
dyes. Diabetologia 1987;30:812-6.
3. Jahanmehr SA, Hyde K, Geary CG, et al. Simple technique
for fluorescence staining of blood cells with acridine or-
ange. J Clin Pathol 1987;40:926-9.
4. Bank HL. Rapid assessment of islet viability with acridine orange
and propidium iodide. In Vitro Cell Dev Biol 1988;24:266-73.
5. Thomas ED, Storb R. Technique for human marrow graft-
ing. Blood 1970;36:507-15.
6. Burger SR, Kadidlo D, McCullough J. Improved progenitor
assay standardization using peripheral blood progenitor
cells from a donor treated with granulocyte–colony-stimu-
lating factor. Transfusion 1999;39:451-6.
Time (min)
Per
cent
age
of v
iabl
e ce
lls
��
