Published July 1, 1985
Normal and Perturbed Chinese Hamster
Ovary Cells: Correlation of DNA, RNA, and
Protein Content By Flow Cytometry
H. A. CRISSMAN, Z. DARZYNKIEWlCZ,* R. A. TOBEY, and J. A. STEINKAMP
Los Alamos National Laboratory, Los Alarnos, New Mexico 87545, and "Memorial Sloan-Kettering Cancer
Center, New York City, New York 10021
DNA, RNA, and protein comprise the bulk of macromolecules in cells. Interrelationships in synthesis and accumulation
of these moieties appear to play an important role in regulating cell cycle traverse capacity, cell division, growth, and size.
Consistency in the size range of the population volume distribution and cycle generation time is controlled by transcriptional and translational processes that rigidly couple temporal
metabolism of DNA, RNA, and protein. With few rare exceptions (e.g., histone proteins), rates of RNA and protein synthesis in cells are constant, and increases in cellular content
are linear and proportional across the cell cycle of exponentially growing mammalian populations (1, 2). However, a
variety of cycle-perturbing agents, including drugs, induce a
differential uncoupling of normal synthetic patterns, causing
a disproportionate accumulation of these cellular constituents. These conditions may lead to states of unbalanced
growth, loss of long-term viability, and eventual cell death. A
technique which could directly measure and correlate cellular
DNA, RNA, and protein content in single cells would be
THE JOURNAL OF CELL BIOLOGY . VOLUME 101 JULY 1985 141-147
© The Rockefeller University Press • 0021-9525185/0710141]07 $1.00
useful in detecting alterations in normal metabolic patterns
and assessing the functional state of cells under a variety of
experimental conditions.
Flow cytometry (FCM) ~ provides a rapid and precise
method of performing multiple biochemical measurements
on single cells, thereby allowing for subsequent correlation of
the various metabolic parameters. Simultaneous FCM analysis of cellular DNA and RNA (3, 4) and DNA and protein
(5, 6) contents, for example, permits distinction between
quiescent and cycling cells (7, 8), cells in various stages of
differentiation (4), and determination of the relationship between cell growth and the cell division cycle (9). To date, no
method has been described for rapid, simultaneous measurement of DNA, RNA, and protein in cells. Such measureAbbreviations used in this paper. AdR, adriamycin; CHO, Chinese
hamster ovary; FCM, flow cytometry; FITC, HO, and PY fluorescence, fluorescein-isothiocyanate(green), Hoechst 33342 (blue), and
pyronin Y (red) fluorescence,respectively.
141
Downloaded from on June 15, 2017
ABSTRACT Quantitative, correlated determinations of DNA, RNA, and protein, as well as RNA
to DNA and RNA to protein ratios, were performed on three-color stained cells using a
multiwavelength-excitation flow cytometer. DNA-bound Hoechst 33342 (blue), proteinfluorescein isothiocyanate {green), and RNA-bound pyronin Y (red) fluorescence measurements were correlated as each stained cell intersected three spatially separated laser beams.
The analytical scheme provided sensitive and accurate fluorescence determinations by minimizing the effects of overlap in the spectral characteristics of the three dyes. Computer analysis
was used to generate two-parameter contour density profiles as well as to obtain numerical
data for subpopulations delineated on the basis of cellular DNA content. Such determinations
allowed for analysis of RNA to DNA and RNA to protein ratios for cells within particular
regions of the cell cycle. The technique was used to study the interrelationship of DNA, RNA,
and protein contents in exponentially growing Chinese hamster ovary cells as well as in cell
populations progressing the cell cycle after release from arrest in G1 phase. The sensitivity of
the method for early detection of conditions of unbalanced growth is demonstrated in the
comparison of the differential effects of the cycle-perturbing agent, adriamycin, on cells
treated either during exponential growth or while reversibly arrested in G~ phase.
Published July 1, 1985
ments, when combined with assessment of ratios of RNA to
DNA and RNA to protein per cell, can provide a sensitive
gauge on metabolism of cells located at particular stages of
the cell cycle.
This report describes the development of an FCM method
for direct determination and correlation of DNA, RNA, and
protein in individual cells. The fluorescent labeling protocol
incorporates modifications of procedures for fluorochroming
DNA with Hoechst 33342 (HO) (10), RNA with pyronin Y
(PY) (11-14), and protein with fluorescein isothiocyanate
(FITC) (5, 15). Analysis is performed in a three-laser excitation system that has been described recently (16).
Using this technique, patterns of DNA, RNA, and protein
contents, and ratios of RNA to DNA and RNA to protein
were analyzed in exponentially growing Chinese hamster
ovary (CHO) cells and in populations reinitiating cycle traverse following reversible arrest in the G; phase (17). The
sensitivity of the method for early detection of potential
metabolic impairment is demonstrated in analysis of the
differential effect of the antineoplastic drug, adriamycin
(AdR), on these two cell populations.
MATERIALS AND METHODS
Cell Harvest and Fixation: At various time intervals ceils from
untreated (control) and drug-treated cultures were harvested by centrifugation.
After removal of culture medium, the cell pellet was resuspended in one part
cold saline GM (g/L; glucose 1.1, NaCI 8.0, KCI 0.4, Na~HPO~. 12 H~O 0.39,
KH2PO4 0.15) containing 0.5 mM EDTA, and then three parts of ice cold 95%
ethanol was added, with mixing of the cell suspension. The final ethanol
concentration was ~70%. Sample tubes were chilled on ice for at least 1 h and
then stored at 4"C for at least 12 h prior to staining. The volume of fixative
solution was adjusted to maintain cell density at ~10 ~ cells/ml.
Spectrophotofluorometric Analysis: Excitation and emission
curves for dyes in phosphate-buffered saline (PBS) solution were obtained with
an automatic recording Aminco-Bowman Spectrophotofluorometer (American
Instrument Company, Silver Spring, MD) using a xenon lamp for excitation
and an R446 photomultiplier tube (Hamamatsu Corp., Middlesex, NJ). Data
collected represent the uncorrected fluorescence spectra. Dye solutions analyzed
contained HO (Calbiochem-Behring Corp., San Diego, CA) and calf thymus
DNA; FITC (isomer 1) (BBL Microbiology Systems, Cockeysville, MD) and
bovine serum albumin (BSA); and PY (purified) (Polysciences, Inc., Warrington, PA) and RNA.
Three-colorStainingfor DNA, RNA,and Protein:
Fixed cells
were removed from ethanol by centrifugation. The cell pellet was resuspended
in PBS containing the DNA stain, HO (0.5 ug/ml), and the general protein
stain, FITC (0.1 ug/ml), for 20 min to 2 h at room temperature. 5 min prior
to flow analysis, an equal volume of cold PBS solution containing HO and
FITC at the same concentrations as above and PY at 2.0 gg/ml was added to
the cell suspension. Tubes were thereafter maintained on ice during staining (5
rain) and cells were analyzed in equilibrium with the stain in solution. The
concentration of stained cells was maintained at ~7.5 × 105 cells per ml which
ensured an excess of dye per binding site, at least for the dyes whose binding
142
THE JOURNAL OF CELL BIOLOGY • VOLUME 101, 1985
FCM Analysis: The three-laser excitation flow system previously described (16) was used with the laser beams tuned to the UV (333.6-363.8 nm),
457.9 nm, and 530.9 nm. The beams were focused onto the cell stream at three
different points, with spacing of-250 gin. The three-color fluorescence detector
consisted of colored glass and dichroic color separating filters for measuring
blue (400-500 nm), green (515-575 nm), and red (>580 nm) fluorescence
emission from each stained cell. These emission measurement regions were
preselected for each dye so that HO-bound DNA (blue), FITC-protein (green),
and PY-bound RNA (red) fluorescence, respectively, were monitored at the
UV 457.9 nm and 530.9 nm excitation points on the cell stream. Fluorescence
measurements were correlated on a cell-to-cell basis. Fluorescence signals and
fluorescence ratios signals (5) were input to signal processing dectronics for
subsequent storage (i.e., list mode fashion) in a DEC PDP-11/23 computer,
and subsequently displayed as single-parameter frequency distribution histograms and two-parameter contour diagrams (18).
Histograms for forward angle light scatter detected from 0.5 to 2.0", three
colors of fluorescence, and two fluorescence ratios were calculated from the
data which were collected and stored in list mode for a given stained population
of cells.
Gated analysis, as described previously (5, 18), was used to derive numerical
data (i.e., relative values) for comparison of the ratios of RNA to DNA and
RNA to protein for cells within specific regions of the cell cycle of the control
and AdR-treated populations. Using such analysis, a gate or window may be
set between preselected regions (i.e., channel number x to channel number y)
of the DNA histogram, encompassing, for example, cells in G~ phase. Reprocessing of the data can then provide the numerical mean ratio value for cells
within the G; region of the cell cycle. Since data are collected and stored in list
mode fashion, data for any of the individual measurement can be retrieved in
a similar manner for cells in any phase of the cell cycle. Potentially, gated
analysis can be performed on any of the distributions, (i.e., protein or RNA
content) as well, and data can be correlated as described above.
RESULTS
Fluorescence Analysis and Spectral Resolution
Quantitative analysis of DNA, RNA, and protein were
derived from measurements of riO (blue), PY (red), and FITC
(green) fluorescence, respectively. The spectral characteristics
of the three dyes (Fig. 1, A and B) required sequential excitation of each dye at the wavelengths indicated (Fig. 1A) followed by emission analysis over preselected fluorescence measurement (wavelength) regions (Fig. I C). This sequential
analysis scheme minimized potential problems in spectral resolution due to the overlap in fluorescence emission of dyes
(Fig. 1B).
Analysis of populations of cells stained with all three dyes
provided distribution profiles for DNA, protein, and RNA
that were nearly identical to distributions obtained for cells
stained with only one of each respective dye (Fig. 2). Cells
stained with all three dyes (Fig. 2D) showed a slight decrease
in (FITC) green fluorescence and a slight increase in (PY) red
fluorescence as compared with respective single dye stained
cells (Fig. 2, B and C, respectively). This, in all probability, is
due in part to energy transfer from FITC to PY. The DNA
profiles in Figs. 2A and 2D are similar, indicating no change
in HO-DNA fluorescence intensity in the three-color stained
cells. Data in Fig. 2 were all obtained using the same electronic
gain and laser power settings for all three lasers. Cells pretreated with RNase prior to PY staining and analysis showed
an 85-90% decrease in red fluorescence compared with distributions for non-RNase-treated cultures (Fig. 2 C). Collectively the data above demonstrate the sensitivity and reliability
of multicolor staining and fluorescence analysis for accurate
assessment of DNA, RNA, and protein in single cells under
the staining and analysis conditions described.
Downloaded from on June 15, 2017
Cell Culture: Suspension (spinner) cultures of CHO cells were maintained in exponential growth phase in Ham's F-10 medium containing 15%
newborn calf serum, streptomycin, and penicillin. Populations of CHO cells
were reversibly arrested in the G; phase by growth in isoleucine-deficient
medium for 36 h as previously described by Tobey and Ley (17). The synchronized G; populations were subsequently cultured in complete (isoleucinecontaining) medium to initiate cell cycle traverse.
Drug Treatment: AdR (Adria Laboratories, Columbus, OH) was
added for a single 2-h interval at a final concentration of 6 /zg/ml to both
exponentially growing and Grarrested populations. Cells were then centrifuged,
washed briefly, and cultured in drug-free complete medium.
Cell Survival Studies: At 15 h after drug treatment and subsequent
culture in drug-free, complete medium, cells were collected from the AdRtreated and control populations for determination of colony-forming ability.
Cells were placed in dishes, allowed to attach, and cultured for 7 d prior to
staining and scoring of colonies. Each colony contained at least 50 cells. Cell
membrane integrity was also assayed by trypan blue staining and microscopic
examination; under such conditions, only damaged cells are stained.
sites could be estimated (e.g., HO, PY). In some studies, cell populations were
stained with only one of each dye, at concentrations described above, prior to
analysis. Ethanol-fixed cells were, in some instances, treated with RNase (50
~g/ml) (Worthington Biochemical Corp., Freehold, NJ) (Code R) for 1 h at
37"C prior to PY staining and analysis.
Published July 1, 1985
FLUORESCENCE E X C I T A T I O N
(A)
100
HOECHST 33342 -FITC
PYRONIN Y
~
i
|
i
355nm
4 g O n7
545 nm
Jr\ 4'°
IZ
W
It
i
I
t
W
I100
uJ
F L U O R E S C E N C E EMISSION
o,
A
I\l
_
I
\'
N
tt
~I~
FLUORESCENCE MEASUREMENT
(C)
REGIONS
•
iiiiiiiiiiiiii!iiiiiiiiiiiiiiiiiiii!iiiiiii!~
i!ii!ii!!iii!iii~iiiiiiiiiiiiii[iii[iiiiiiiiiiiiiiii!i!i!~.
m
o.
0
- @@iiiiiii!iii!ii
300
I
400
500
WAVELENGTH
600
700
800
(nm)
FIGURE 1 Fluorescence excitation and emission spectra, as well as
fluorescence measurement regions for D N A - b o u n d H O fluorescence, FITC-protein, and pyronin Y-RNA fluorescence. The arrows
in A indicate the laser beam wavelengths selected to preferentially
excite the respective dyes in stained cells. Approximate peak wavelengths values are designated. Fluorescence measurement regions
(C) are predetermined by the selective filter arrangement described
in Materials and Methods.
Analysis of Asynchronous and G1-arrested
C H O Cells
The DNA, RNA, and protein staining and analysis techniques were used to characterize and compare populations of
exponentially growing CHO cells and the noncycling CHO
cells in cultures deprived of isoleucine. Isoleucine deprivation
is a method routinely used for synchronization of rodent cells
in G~ phase (17).
The cell distribution patterns with respect to RNA and
protein contents are surprisingly similar both in exponentially
growing (Expon.) as well as noncycling (Ileu-) cell populations
(Fig. 3, A and B). However, the populations show greater
heterogeneity in protein rather than RNA content. This is
evident from comparisons of the width of the RNA and
protein distributions in the G~ or G2 + M clusters.
CRtSSMAN ET At. Normal and Perturbed Chinese Hamster Ovary Cells
143
Downloaded from on June 15, 2017
100
Z
0
z
/~
One characteristic feature of the G~ population is the apparent presence of a threshold RNA or protein content (Fig.
3A, arrows). The data indicate that cells with a subthreshold
content of either RNA or protein do not immediately enter
the S phase. As described before, this threshold discriminates
two different compartments of G~ phase, G~A and G~B, which
are believed to have different functions (7). The exponentially
growing population shows about half of the G~ cells in the
G~A compartment (i.e., cells characterized by the subthreshold RNA or protein content) (Fig. 3A). A remarkably constant relationship was observed between RNA and protein
content of individual cells in the cycling population (i.e.,
RNA vs. protein) (data not shown).
Analysis of the RNA to DNA ratio (Fig. 3A) in relation to
cell cycle position, as based on DNA content, reveals a
characteristic pattern of changing rates of DNA replication
vs. transcription. The data show that during Gt, when DNA
content is constant, cells accumulate increasing quantities of
RNA which is reflected in high heterogeneity of the RNA to
DNA ratio. During progression through S phase, the rate of
DNA replication exceeds the rate of accumulation of RNA
giving rise to a nonvertical, negative slope of the RNA to
DNA ratio for the S phase cell cluster. It can be noted that
cells in G2 + M have, on average, an RNA to DNA ratio that
is similar to that of the majority of G~ cells.
The ratio of RNA to protein is a novel parameter. Simultaneous measurements of DNA, RNA, and protein made it
possible not only to estimate this ratio, but also to assess the
RNA to protein relationship for cells at various positions in
the cell cycle. The RNA/protein ratio thus provides a useful
parameter which would detect unbalanced growth when the
rates of RNA and protein accumulation (reflecting DNA
transcription and RNA turnover as well as protein synthesis
and degradation) vary with respect to each other. These
studies detect no such variability during the cell cycle; in fact,
the RNA to protein ratio remains strikingly constant and
uniform for all cells regardless of their DNA content (Fig.
3A). Such a pattern is expected because of the very good
correlation between RNA and protein content in individual
cells mentioned previously.
Noncycling cells such as those shown in Fig. 3 B represent
CHO cells arrested in G~ by isoleucine deprivation. Cells from
these cultures show a much higher heterogeneity in RNA and
protein content than G~ cells in the cycling population. The
threshold RNA or protein content of the G~ population
(similar to that observed in the cycling population) is also
apparent in the respective profiles of cells deprived of isoleucine (Fig. 3B, arrows). Based on gating analysis, >90% of
isoleucine-deprived cells have the subthreshold RNA content
and may be characterized as G~A ceils (7). A good correlation
was also observed between RNA and protein content of
individual cells (data not shown) although the distribution is
more asymmetrical in comparison with cycling cells.
Both the RNA to DNA and RNA to protein ratio distributions of noncycling cells show higher heterogeneity of the
G~ population in comparison to respective distributions for
exponentially growing G~ phase cells. The distribution with
respect to the RNA to protein ratio is less symmetrical than
in cycling populations and the skew of the distribution indicates the presence of G~ cells with lower protein content but
still relatively high RNA content. The RNA and protein
contents, however, are somewhat decreased in the cells deprived ofisoleucine. Thus, on the basis of the RNA to protein
Published July 1, 1985
I
CHO CELLS --I
STAINED WITH-
--LASER
No.1 - - 1
E X C I T A T I O N (uv)
BLUE
FLUORESCENCE CHANNEL
--LASER
No.2 --~
EXCITATION (457nm)
GREEN
FLUORESCENCE
CHANNEL
I~536
,
LASER No.3 - EXCITATION (530nm)
RED
FLUORESCENCECHANNEL
384
I
(A)
HOECHST
/
33342
ONA CONTENT
TOTAL
PROTEIN
RNA
CONTENT
ONLY
0
I
I
I
I
I
I
1
I
I
L
I
84
I
128
t
192
258
I
1
I
I
I
1
1
l
1
I
I
J
ISOTHIOCYANATE
(FITC) ONLY
g.
o
la.
o
o
~
1
I
64
128
192
....
0
t
I
i
256
I
0
64
128
192
256
CHANNEL NUMBER
FIGURE 2 Single parameter frequency distributions for populations of ethanol-fixed CHO cells stained only with HO (0.5 p.g/ml)
for DNA (A); with FITC (0.1 #g/ml) for protein (B); with PY (I .0/~g/ml) for RNA (C); or with a combination of all three dyes (D) as
described. All populations (A-D) were analyzed in the three-laser system at all three wavelengths indicated, and fluorescence
was monitored in each channel.
ratio data and the diminished RNA to DNA ratio levels, the
noncycling cells are in a moderate state of unbalanced growth
after isoleucine deprivation.
Analysis of Cycle Progress in
Synchronized Cultures
When transferred to complete (isoleucine-containing) medium, the G~-arrested cells (Fig. 3B) will initiate DNA synthesis and progress through the cell cycle (Fig. 3, C-F). By 6
h (Fig. 3 C), cells with elevated RNA and protein contents are
the first to enter S phase. The RNA to DNA ratio is increased
in this subpopulation but the RNA to protein ratio profile is
similar to the subpopulation of cells still residing in G, phase
at that time. At 9 h (Fig. 3 D), cells have reached (32 + M
phase, and by 12 h (Fig. 3E), the single parameter DNA
profile for the population (Fig. 3 E, left panels) closely resembles the DNA profile for exponentially growing CHO cells in
Fig. 3A. However, assessment of the RNA and protein contents profiles as well as the RNA to DNA and RNA to protein
ratios in Fig. 3, B - E indicate greater heterogeneity in cellular
levels of these constituents in cultures initially arrested in GI
phase. By 24 h, the respective distributions patterns (Fig. 3 F)
begin to more closely resemble those shown in Fig. 3A,
especially the RNA to DNA ratio profde. At 36 h, patterns of
the two cultures are indistinguishable (data not shown).
144
THe JOURNALOF CELt B,OLOCY• VO'UME 101, ~985
Analysis of Drug-treated Populations
AdR treatment induced a differential response in metabolism of DNA, RNA, and protein in exponentially growing
cells and cells initially in GI phase at the time of treatment.
Analysis of the single parameter DNA profiles at 15 h after
drug treatment (Fig. 4B, left panel) showed a large accumulation of cells from the exponential culture arrested in the G2
+ M phase (i.e., 85%). Based on numerical data (Table I),
this subpopulation had mean ratio values for RNA to DNA
and RNA to protein that were 44% and 31% elevated, respectively, above control ratio values. Few cells are observed
in S phase, but the cells remaining in GI phase (i.e., 15%) had
RNA to DNA and RNA to protein ratio values almost identical to control values (Table I). Survival studies showed that,
compared to the exponential control CHO cell population,
the AdR-treated exponential population had only a 12%
surviving fraction. Analysis with trypan blue revealed that, at
the time the survival studies were performed (l 5 h after AdR
treatment), >97% of the drug-treated cells excluded the dye.
By comparison, the AdR-treated GI cells, when placed in
drug-free, complete medium (Fig. 4 D), show only a moderate
accumulation of cells in (32 + M phase with RNA to DNA
and RNA to protein mean ratio values that are only 5% and
11% below respective values for the untreated control (Table
I). Comparison of these values and those in Table I for cells
Downloaded from on June 15, 2017
o
l
Published July 1, 1985
Protein
RNA
::~,,
tI : tI; II= "t
t t '1
'
I
I
I~
/
I
I
I
I
I
I
.... i
/
/
I
I
, ........
I
/i
I~
i
I
,I
i
I
,,i -I
-
24
l
........
J
I
Downloaded from on June 15, 2017
,
I
I
tl ' t
J
@
z
..
I
e~
~
RNA/Protein
2+ M
IG 1
<
RNA/DNA
i
: .
/
I
I
I
I
/
I
I
I
..
FIGURE 3 Single parameter DNA distributions (left panels) and two-parameter contour density profiles for DNA {y axis) vs.
protein, and RNA as well as the RNA to DNA and RNA to protein ratios (left to right) for exponentially growing CHO cells (A),
and for cells initially arrested in G~ phase (B), and at the designated time periods (hours) following release from cell synchrony
(C-F). Single parameter DNA profiles are displayed to correspond with the DNA content (i.e., y axis) of the two parameter profiles
as, for example, shown as G1, S, and G2 + M regions in the protein distribution (see Fig. 3A). At least 30,000 cells are represented
in each two-parameter distribution. Dotted areas represent at least five cells and sequential contours represent increasing
isometric equivalents of 10, 50, 250, 500, and 1,500 cells. Regions designated by the arrows are discussed in the text.
in all phases of the cell cycle show that, based on these criteria,
the G, drug-treated population is in a more nearly balanced
state of metabolic equilibrium 15 h after drug treatment than
the drug-treated exponential culture at 15 h after AdR treat-
ment. Cells in the former population continue to progress
through S phase (Fig. 4 D), and based on cell count data, the
cell number of the population had doubled by 30 h compared
to an increase of only 8% in the drug-treated asynchronous
C~ISSM^N ET ^L. Normal and Perturbed Chinese Hamster Ovary Cells
145
Published July 1, 1985
Protein
<
Z
tExp°n
t
t
I
J
I
i
I
RNA/Protein
I
II
I
I
I
L''* t
/
A
..
I I
B1
*:Qii~i["
I I
t
':i~":::/i~iii
~i
I
L
/
I
• • I'L.
I.
ii~•~:.'
I
[
I
/
<
Z
~
I
I
I
I
I
/
~
:
I
'
"~
]
I
I
J
I
[
[
zl
I
ti c t
:. :7..;:•.
I
I
I
!
I
l
l
FIGURE 4 Single parameter DNA distribution (left panels) and two-parameter contour density profiles for DNA (y axis) vs. protein
and RNA, as well as the RNA to DNA and RNA to protein ratios (left to right) for control (untreated), exponentially growing CHO
cells (A) and control CHO cells at 15 h after release from synchrony in G1 phase (C). Corresponding populations of AdR-treated
cells at 15 h after drug treatment (i.e., 6 #g/ml AdR for 2 h) are shown in B and D, respectively.
TABLE I.
Ratios*
Mean Values of the RNA to DNA and RNA to Protein
RNA to DNA
G1
Exponential control 64.0
AdR treated
65.4
Synchronized
(G~) phase control
112.4
AdR-treated
101.9
RNA to Protein
S
G2+M
62.6
--
58.0
83.5
46.1 48.2
45.9 - -
G1
S
G2+M
47.3
62.2
101.5
94.6
91.0
86.7
71.6 72.3
57.2 58.3
65.0
57.4
* Obtained for cells at various stages of the cell cycle after a 2-h treatment
with AdR (6 #g/ml) and resuspension in drug-free medium for 15 h prior to
analysis•
Mean values of ratios are in arbitrary units (i.e., mean channel number) and
derived from the same ratio data used in Fig. 4. Numerical values were
calculated using gated analysis (5, 18) which allows for analysis of correlated
measurements made within preselected regions of the DNA histograms•
culture. The G)-arrested, untreated (control) population doubled in cell number by 24 h after release from isoleucine
deficiency blockade.
The isoleucine synchronization technique had no effect on
cell survival; cells released from G, arrest for periods at least
as long as 24 h exhibited identical plating efficiencies to those
obtained with exponentially growing control cultures. As reported previously (19), AdR was much less toxic for G~146
THE JOURNAL OF CELL BIOLOGY . VOLUME 101, 1985
arrested than for actively cycling ceils. In contrast to the
nearly two log reduction in survival of exponentially growing
cells treated with AdR and then released into drug-free medium for 15 h prior to plating, the isoleucine deficient, drugtreated cells released into drug-free medium for 15 h prior to
plating exhibited a survival value of 84%. Trypan blue measurements indicated that virtually all of the cells from the
cohort yielding an 84% survival fraction maintained an intact
plasma membrane.
DISCUSSION
Simultaneous analysis of DNA, RNA, and protein by FCM
provided for direct correlation of cellular levels of these macromolecules in exponential and Grarrested populations of
CHO cells. Ratios of RNA to DNA and RNA to protein on
a cell-to-cell basis were useful in assessing both graphically
and numerically the metabolic condition of cells at distinct
phases of the cell cycle. The technique was effective in demonstrating the accumulation of the respective cell constituents
during the progression of synchronized cells from Gt to S and
G2 + M phases of the cell cycle. Although the DNA to RNA
and the DNA to protein profiles in this initially synchronized
population appeared quite similar, the dispersion in the RNA
to protein ratio indicated a considerable heterogeneity in cells
Downloaded from on June 15, 2017
t
t
:..,...t
t
L
,%.:
<
7_
<
Z
RNA/DNA
RNA
Published July 1, 1985
The work was performed under the auspices of the Los Alamos
National Flow Cytometry and Sorting Research Resource, funded by
the Division of Research Resources of the National Institutes of
Health (grant P41-RR01315-02), the Department of Energy, and the
United States Public Health Service (grants IROCA23296 and CA
28704).
Received for publication 28 November 1985 and in revisedform 25
March 1985.
REFERENCES
I. Mitchinson, J. M. 1971. The Biology of the Cell Cycle. University Press, Cambridge.
29-33.
2. Pardee, A. B., R. Dubrow, J. L. Hamlin, and R. A. K]etzien. 1978. Animal cell cycle.
Annu. Rev. Biochem. 47:715-750.
3. Darzynkiewicz, Z., F. Traganos, T. Sharpless, and M. R. Melamed. 1975. Conformation
of RNA in situ as studied by acridine orange staining and automated cytofluorometry.
Exp. CelI Res. 95:143-153.
4. Darzynkiewicz, Z., F. Traganos, T. Sharpless, and M. R. Melamed. 1976. Lymphocyte
stimulation: a rapid, multiparameter analysis. Proc. Natl. Acad. Sci. USA. 76:358-362.
5. Crissman, H. A., and J. A. Steinkamp. 1973. Rapid simultaneous measurement of
DNA, protein, and cell volume in single cells from large mammalian cell populations.
.L Cell. BioL 59:766-771.
6. Crissman, H. A., J. V. Egmond, R. G. Holdrinet, A. Pennings, and C. Haanen. 1981.
Simplified method for DNA and protein staining of human hematopoietic cell samples.
Cytometry. 2:59-62.
7. Darzynkiewicz, Z., T. Sharpless, L Staiano-Coico, and M. R. Melamed. 1980. Subcompartments of the the Gt phase ofceU cycle detected by flow cytometry. Proc. NaIL Acad.
Sci. USA. 77:6696-6699.
8. Darzynkiewicz, Z., F. Traganos, and M. R. Melamed. 1980. New cell cycle compartments identified by multiparameter flow cytometry. Cytometry. 1:98-108.
9. Darzynkiewicz, Z., H. Crissman, F. Traganos, and J. Steinkamp. 1982. Cell heterogeneity
during the cell cycle. J. Cell. Physiol. I 13:465-474.
10. Arndt-Jovin, D., and T. M. Jovin. 1977. Analyses and sorting of living cells according
to deoxyribonucleic acid content..L Histochem. Cytochem 25:585-589.
I I. Brachet, J. 1940. La detection histochimique des acides pintose nucleiqoes. Comptes
Rendus Des Seances De La Societe De Biologie. 133:88-90.
12. Tankc, H. J., A. B. Nieuwenhuis, G. J. M. Kopcr, J. C. M. Slats, and J. S. Ploem. 1981.
Flow cytometry of human reticulocytes based on RNA fluorescence. Cytometry. 1:313320.
13. Shapiro, H. M. 1981. Flow cytometric estimation of DNA and RNA content in intact
cells stained with Hoechst 33342 and pyronin Y. Cytometry. 2:143-150.
14. Pollack, A., D. L. Prudhomme, D. B. Grecnstein, G. L. Irvin III, A. J. Claflin, and N.
L. Block. 1982. Flow cytometric analysis of RNA content in different cell populations
using pyronin Y and methyl green. Cytometry. 3:28-35.
15. G6hde, W., t. Spies, J. Schumann, T. Buchner, and G. Klein-Dopke. 1976. Two
parameter analysis of DNA and protein content of tumor cells. In Pulse-Cytophotometry. T. Buchner, W. Gtihde, and J. Schumann, editors. European Press, Ghent, Belgium.
27-32.
16. Steinkamp, J. A., C. C. Stewart, and H. A. Crissman. 1982. Three-color fluorescence
measurement on single cells excited at three laser wavelengti~s. Cytometry. 2:226-231.
17. Tobey, R. A., and K. D. Ley. 1971. Isoleucine-mediated regulation ofgenome replication
in various mammalian cell lines. Cancer Res. 31:46-51.
18. Salzman, G. C., S. F. Wilkins, and J. A. WhitfiU. 1981. Modular computer programs
for flow cytometry and sorting: the LACEL system. Cytometry. 1:325-336.
19. Tobcy, R. A., H. A. Crissman, and M. S. Oka. 1976. Arrested and cycling CHO cells as
a kinetic model: studies with adriamycin. Cancer Treat. Rep. 60:1829-1837.
CRISSMAN ET AL.
Normaland Perturbed Chinese Hamster Ovary Cells
147
Downloaded from on June 15, 2017
entering S phase. The RNA to protein ratio is a new parameter
not previously demonstrated on a per cell basis. The analysis
correlates transcriptional and translational activity and can
also be useful for detecting conditions of unbalanced cell
growth. For example, Gl-arrested cells were initially in a
moderate state of unbalanced growth after 36 h of isoleucine
deprivation; however, 24 h after growth in complete medium,
the population regained equilibrium and metabolic patterns
resembled those of exponentially growing cells. At 36 h, the
profiles for the two populations were virtually identical (data
not shown).
These FCM analyses further revealed conditions of gross
imbalance of cellular levels of DNA, RNA, and protein induced by the cycle-perturbing agent, AdR. As early as 15 h
after drug treatment, metabolic patterns showed uncoupling
of DNA synthesis and cell division associated with continued
but disproportionate accumulation of RNA and protein, particulady in the exponential drug-treated population. Based
upon the relatively greater degree of imbalance induced in
exponentially growing populations in Fig. 4, we would predict
that the toxicity of AdR would be greater for cells of this type
than cells that were initially arrested in Gl at the time of drug
addition. Survival measurements indicated that this is indeed
the case. In the present study, cells assayed for colony-forming
ability at 15 h after adriamycin treatment showed an 84%
surviving fraction for cells exposed to the drug while in G,arrest compared to 12% for the drug-treated exponentially
growing cells. At the time of plating, both drug-treated populations exhibited >97% exclusion of trypan blue, indicating
that samples processed for and analyzed by FCM were composed predominately of intact cells. Thus, in this instance at
least, there is excellent correlation between cell survival and
early (i.e., 15 h after drug treatment) indications of cellular
metabolic impairment and unbalanced growth (reflected in
the abnormal ratios of RNA to DNA and RNA to protein,
and drug toxicity).
The multiparameter analysis procedure outlined in this
report should be useful in yielding information on the metabolic state of cell populations treated with a variety of experimental agents. In addition, the technique should provide new
insight into normal metabolic processes associated with cell
cycle progression-related events.