- Email: [email protected]
cyclin expression during the cell cycle in normal and leukemic cells
1.eukemta Re.~earch Vol. 15, No, 11. pp. 965 q74. 1991. Printed in Great Britain.
0145-2126/91 $3.00 + .00 Pergamon Press plc
P R O L I F E R A T I N G CELL N U C L E A R A N T I G E N ( P C N A ) / C Y C L I N E X P R E S S I O N D U R I N G T H E CELL CYCLE IN N O R M A L A N D L E U K E M I C CELLS MONICA GIORDANO,* MARCO DANOVA,* CARLO PELI,ICCIARI,t GEORGE D. WILSON,~: GIULIANO MAZZINI,§ ANNA M. FUHRMAN CONTI,II GIOVANNI FRANCHINI,* ALBERTO RICCARDI* and MARIA G. MANFREDI ROMANINIt *Department of Internal Medicine, Section 2nd Medical Clinic; tDepartment of Animal Biology; University and I.R.C.C.S. San Matteo, Pavia; :~Gray Laboratory, Mount Vernon Hospital, Northwood, U.K. ; §Centro Studio Istochemica C.N.R., University and I.R.C.C.S. San Matteo, Pavia and IIBioiogy Medical Sciences, University of Milan, Italy.
(Received 11 April 1991. Accepted 18 May 1991) Abstract--Bivariate flow cytometric analysis of the cell proliferation-associated nuclear protein, identified as the "proliferating cell nuclear antigen" (PCNA)/cyclin and of nuclear DNA content, was performed in quiescent and mitogen-stimulated human peripheral blood lymphocytes, in EUE (human embryonic epithelium) cells, before and after a long-term exposure to a hypertonic (HT) medium, in 4 human leukemic cell lines and in fresh bone marrow (BM) cells from 10 patients with untreated acute non-lymphoblastic leukemia (ANLL). The PCNA/cyclin was detected using both an autoantibody extracted from sera of systemic lupus erythematosus patients and the recently produced mouse monoclonal antibody (MoAb) IgG, named 19F4. The distribution of cells in the different phases of the cycle and the percentage of S-phase cells were obtained in duplicate samples, by DNA flow cytometry (FCM) and by dual parameter FCM of DNA content and bromodeoxyuridine (BUDR) incorporation. In all cell types, the non-specific cytoplasmic background fluorescence was significantly lower with the MoAb compared to that obtained with the polyclonal Ab. The percentage of PCNApositive cells (both with the autoantibody and the 19F4 MoAb) was always higher than that of Sphase cells by DNA FCM and of BUDR-labeled cells. The pattern of PCNA-expression in both normal proliferating cells and acute leukemia cells, showed that most G0/G1 cells did not express significant amounts of PCNA; an increase in PCNA immunofluorescence was found in late GI cells, and further increases were observed in S- and Gz--M phase cells. PCNA/cyclin, as revealed both with autoantibodies and with the 19F4 MoAb, is associated with all actively or potentially dividing (i.e. GI, S and Gz-M) cells thus identifying the proliferative cellular compartment. Combined with the use of muitiparameter FCM techniques, the PCNA immunolocalization offers a useful tool to study cell kinetics in normal and leukemic human cell populations.
Key words: PCNA/cyclin, acute leukemia, cell proliferation. DNA content, flow cytometry, Sphase, growth fraction.
[1,2]. Dual parameter flow cytometry (FCM) of
INTRODUCTION
DNA content and of the immunolabeling of these proteins allows protein expression to be precisely related to each cell cycle phase in populations with different proliferative activity [3-7]. Sera from certain patients with systemic lupus erythematosus (SLE) contain autoantibodies to a proliferating cell nuclear antigen (PCNA) [8-12] also called cyclin [13], which is an evolutionary highly conserved 36 kD acidic nuclear polypeptide (identified as a polymerase delta accessory protein) [1417]. The expression and synthesis of PCNA are enhanced in proliferation compared to quiescent cells [18-19]. cDNA clones corresponding to the human PCNA have been isolated and described [20-22].
THE study of nuclear and surface proteins expressed in variable amounts in the different phases of the cell cycle may permit a better understanding of the mechanisms regulating cell proliferation providing additional parameters for describing the cell cycle
Abbreviations: BM, bone marrow; ANLL, acute noniymphoblastic leukemia; MoAb, monoclonal antibody; PI, propidium iodide; PBL, peripheral blood lymphocytes; PHA, phytohemagglutinin; FCM, flow cytometry; PCNA, proliferating cell nuclear antigen; BUDR, bromodeoxyuridine. Correspondenceto: Marco Danova, M.D., Ph. D., Clinica Medica 2, Universit~ di Pavia, Policlinico S. Matteo 27100 Pavia, Italy. 965
966
M. GIORDANO et al.
Travali et al. [23] obtained a clone containing the entire human PCNA gene from a lambda phage library, that was subsequently functionally characterized using a c D N A probe [24]. PCNA expression was found to increase during the G 1 phase of the cell cycle, reaching its maximum in the S-phase, and then declining during the G2-M phase. Exposure of exponentially growing cells to antisense oligodeoxynucleotides to PCNA resulted in complete suppression of D N A synthesis and mitosis, indicating an important role for this protein in cell proliferation [25, 26]. The SLE autoantibodies have been noted to be a useful marker for detecting proliferating cells in the peripheral blood of patients with leukemia [27], in solid tumors [28] and in normal tissues [29], both through immunocytochemistry and bivariate FCM [3]. However, anti-PCNA-positive SLE sera without other antibodies directed against nuclear components are extremely rare. The development of monoelonal antibodies (MoAbs) made it possible to improve the specificity of PCNA detection [30, 31]. In this study, we aimed to compare the patterns of PCNA expression resulting from the application of either polyclonal or MoAbs on different cell systems, i.e. quiescent and PHA-stimulated peripheral blood lymphocytes, E U E (human embryonic epithelium) cells grown in either isotonic (IT) or hypertonic ( H T ) medium, human hematopoietic cell lines of different origin and with different phenotypes and fresh bone marrow (BM) cells from 10 patients with untreated acute non-lymphoblastic leukemia (ANLL). In order to perform dual parameter FCM of PCNA expression and DNA content, a fixation/permeabilization procedure was developed to obtain a satisfactory binding of the anti-PCNA Abs, and the simultaneous staining of DNA with propidium iodide (PI). Results (in terms of percentages of positivity to PCNA and patterns of expression) were compared with those obtained from experiments conducted in duplicate samples with single parameter DNA-FCM and twoparameter FCM (DNA vs bromodeoxyuridine, B U D R , incorporation). MATERIALS AND METHODS Phytohemagglutinin
(PHA)-stimulated
lymphocytes.
Human peripheral blood lymphocytes (PBS) of healthy donors, were separated by Ficoll-Hypaque density gradient centrifugation. The preparation contained more than 90% lymphocytes. Cells were suspended in RPMI 1640 (Flow Laboratories, Inglewood, CA, U.S.A.), supplemented with 2raM glutamine, vitamins, non-essential amino acids, sodium pyruvate, 100 U/ml penicillin, 100 lxg/ml streptomycin, and 15% fetal bovine serum to a concentration of 1 x 106/ml. The cell suspension was divided into two groups. One was cultured in presence of
30 Ixg/ml of PHA (Wellcome Reagents Ltd, Greenville, NC, U.S.A.) for 72 h in 5% CO2/95% air at 37°C and the other without PHA as control. Cell lines. EUE cells [32] were selected as a suitable normal cell model in which the long-term exposure (4 days) to a culture medium with increased toxicity was found to affect cell kinetic parameters, inducing a Go block in most cells, and a slowing down of the S-phase in the still cycling cells [32]. Cells were grown for 4 days in either IT or HT in minimum essential Eagle's medium, supplemented with 12% newborn bovine serum and 1% gtutamine. To obtain the HT medium, the concentration of NaCI was raised, from 0.136M to 0.274 M, in the Hank's saline used as a component of the culture medium (the final osmolarities were 305 mOsm and 495 mOsm, in IT and HT respectively). The culture media were changed every two days, to avoid the influence of nutrient shortage. Four human hemopoietic cell lines were studied during their exponential growth phase in liquid culture: acute lymphoblastic leukemia cell line MOLT 4; Burkitt's lymphoma cell line (Daudi); myeloid cell lines: HL-60 (promyelocytic leukemia), and KGI (myeloblastic leukemia). The cells were continuously cultured (in 25 c m 3 plastic flasks, Corning, MD, U.S.A.) in RPMI medium (GIBCO, Grand Island, New York, U.S.A.) supplemented with 10% fetal calf serum (FCS, Flow laboratories, Rockville, MD, U.S.A.), 1% L-glutamine and non-essential amino acids in 5% CO2/95% air. MCF-7 cells were maintained in Dulbecco's modified Eagle's medium (DMEM, GIBCO) supplemented with 20 mM Hepes buffer, 10% FCS and 200 U/ml penicillin (Eurobio, Paris, France). Bone marrow (BM) cells. The source of BM leukemic cells was a group of 10 patients with ANLL studied at the time of initial diagnosis. All patients were advised of procedures and attendant risks in accordance with the institutional guidelines, and gave informed consent. Lightdensity mononuclear cells (LDMNCs) were collected after centrifugation on a Ficoll-Hypaque gradient, density 1077g/cm 3 (Pharmacia Fine Chemicals, Piscataway, NJ, U.S.A.) in Iscove's modified Dulbecco's medium (IMDM, GIBCO). The cells were further purified by removing adherent cells and T lymphocytes. A 5 ml suspension of light density cells at a concentration of 5 × 106 cells/ml in IMDM plus 15% FCS was incubated in 25 cm 3 tissue culture ltasks for 60min at 37°C, and non-adherent cells carefully collected. This procedure was repeated twice. Tlymphocyte-depleted LDMNCs (T-LDMNCs) were obtained by rosetting MNC suspensions (5 × 106 cells) with 2-aminoethylisothiouronium bromide-treated (AET, Sigma Chemical Co, St Louis, U.S.A.) sheep red blood cells in a 5% suspension with IMDM. Non-rosetting cells were separated by a second Ficoll-Hypaque density centrifugation. Viability after cell separation was evaluated by the trypan blue dye exclusion test. Anti-PCNA antibodies. The anti-PCNA-positive Ab, obtained from patients with SLE was purchased in fluid form from Delta Biological Inc. (Pomezia-Roma, 00040, Italy). The anti-PCNA mouse MoAb Ab 19F4-1gG raised against purified cyclin from rabbit thymus isolated and characterized by Ogata et al. [24], was purchased from American Biotech Inc, (Plantation, FI, 33313. U.S.A.) in an ascites fluid form. Indirect immunofluorescence procedures. The following fixation/permeabilization procedures were tested for both the polyclonal Ab and MoAb: 20 min in 70% methanol at
PCNA and cell cycle in acute leukemia 4°C; 20 rain in 70% ethanol at 4°C; 10 min in 90% acetone at 4°C plus 0.01% Triton X 100 (5 rain); 2 min in 1% paraformaldehyde at room temperature followed by 5 min in 70% methanol at -20°C; 10 min ethanol 70% at -20°C followed by 0.01% Triton X 100 (5rain); 10min 70% ethanol at 4°C, followed by 0.1% Triton X 100 (15 min) and the subsequent incubation (30 min) with the antiPCNA diluted in PBS plus lysolecytin (500 lag/ml). Cells were incubated with a 1:100 dilution of the anti-PCNA polyclonal Ab in PBS (Hoechst AG, F.R.G.) containing 0.5% normal rabbit serum (NRS, Flow Laboratories) and a 1:50 dilution of the MoAb in PBS and 0.5% normal goat serum (NGS, Flow Laboratories), for 30 rain at room temperature. Cells were then centrifuged and incubated for 15 min in either PBS/NRS or PBS/NGS, respectively. After two washes in PBS, the bound antibodies were labeled with a 1:10 dilution of either or fluorescein isothiocyanate (FITC)-conjugated rabbit anti-human IgG (Sigma Chem.) or FITC-conjugated goat anti-mouse IgG (Sigma Chem.) MoAbs, respectively. The levels of non-specific background staining (zero threshold) were established for each measurement using control cells processed at the same time but without exposure to primary antibodies. In all samples, counterstaining of double-stranded DNA was done with PI (5 lag/ml, containing 50 Kunitz units/ml RNAse for at least 30 min). DNA staining. For single-parameter DNA measurements, cells were suspended in 2 ml of phosphate-buffered saline (PBS, Hoechst AG, F.R.G.) and drawn through needles of decreasing diameter. Cell counts were made so as to have more than 1 x 1(16 ceils/ml per sample. The suspension was centrifuged and the pellet stained with propidium iodide (PI, Calbiochem, Behring Corp., San Diego, CA, U.S.A.) at a concentration of 50 lag/ml in PBS; 0.1% Nonidet P40 (Calbiochem.) and 50 Kunitz units/ml RNAse (type 1A, from bovine pancreas, Sigma) were included in the staining solution. A 30 min staining time at room temperature provided the best histogram resolution. Finally, cells were filtered through a 35 p,m nylon mesh to remove aggregates prior to flow analysis. BUDR incorporation stud),'. In all samples pulse-labeling with BUDR (Sigma Chem.) was performed by incubating the cells for 20rain at a final BUDR concentration of 10 p,M; the medium was then washed out, and pre-warmed fresh medium was added. After incubation with BUDR, the cells were harvested and immediately placed in an icecold bath, washed once with PBS, fixed in cold 70% ethanol and stored at -20°C. The BUDR incorporated into Sphase cells was detected with the MoAb against BUDR (Becton Dickinson) by the indirect immunofluorescencc technique which involves pretreatment with HCI (2 N for 10 min at room temperature), incubation with the antiBUDR MoAb (1:10) and then with a FITC-conjugated goat anti-mouse IgG MoAb (1:50). Cells were finally resuspended in 5 p,g/ml PI in PBS containing 50 Kunitz units/ml RNAse, and stained for at least 30 min. Bivariateflow cytometry. Two parameter FCM analyses (FITC-green vs PI-red) were performed with a FACStar Cell Sorter (Becton Dickinson, Sunnyvale, CA, U.S.A.). Excitation was accomplished with an argon ion laser tuned to 488 nm and operated at 300 mW. Emitted fluorescence was split into two bands by means of a dichromatic mirror DM 560. Green fluorescence was measured through a 525 nm interference filter (30nm h.b.w.) while the red fluorescence was measured using a 620 nm long pass filter.
967
Instrumentation setting and calibration were done daily using fluorescent microbeads. Data were collected with a Consort 30 software program running on a dedicated Hewlett Packard computer and displayed as dual parameter contour density plots. At least 20000 cells were analyzed for each sample. Evaluation of FCM data on two-parameter plots was done with box analysis. Early S-(ES) phase cells were operationally defined as FITC-positive cells falling in the first half of the S-phase scale of DNA content, i.e. between the G0/Gl peak and half the distance between the G0/G l and G2-M peaks. Late S-(LS) phase cells were those in the second half of this interval. The levels of expression of the antigens studied in the G0/G l, ES, LS and G2-M phases of the cell cycle were approximately evaluated from the mean green fluorescence intensity of the corresponding FITC-labeled cells. The green fluorescence values for the corresponding boxes in matched control samples were also calculated in each FCM run. This fluorescence (which was set at the same position on the cytogram each time) was subtracted as an aspecific background to obtain the "'net" fluorescence intensity associated with the specific labeling of PCNA. DNA flow cytometry. Single FCM-DNA analyses were performed with a Partec PAS II (Basel, Switzerland) arc lamp flow cytometer, with data displayed as frequency histograms. For each sample, 10 0(~-40 000 cells were analyzed. The measuring conditions included the following: an HBO I(X)W/2 (Osram) excitation source with KGI (2 mm) and BG38 (4 mm) filters; an interference filter of 546 ± 12 nm; a TK 590 dichromatic mirror and a K 610 barrier filter to select the emitted red fluorescence. The percentage distribution of cells in the Go/G l, S- and G2-M phases of the cell cycle was obtained using Fried's mathematical model (which fits the data with a Gaussian distribution) adapted for a dedicated Hewlett Packard computer.
RESULTS We performed a total of 30 experiments: 6 with normal PBL, 6 with the E U E cell line, 8 with the hemopoietic cell lines and 10 with leukemic blasts from the patients with A N L L and the results obtained are shown in Figs 1-5 and in Table 1. We tested several methods of fixation/ permeabilization in order to optimize the reaction with the anti-PCNA polyclonal and/or MoAb. The best results were obtained by treating cells with 70% cold ethanol for 10 min, followed by 0.5% Triton X 100 (15 min) and the subsequent incubation (30 min) with the anti-PCNA diluted 1:50 in PBS plus lysolecytin (500 ~tg/ml). This procedure was then used in all the experiments. As expected [8], in resting lymphocytes, the P C N A was not detectable. After PHA-stimulation, the results obtained with two-parameter FCM of cells labeled with the FITC-conjugated polyclonal A b or 19F4 M o A b anti-PCNA antibody and simultaneously stained for D N A with PI, were consistent with the observed changes in the D N A distribution and in the
968
M. GtORDANOet al. 13 (a--
IT
HT .
.
¢
i
.
•
,~•
"
:
:
i( I
1
1
'
1
'
1
f
3~
~
&O
J
'
t
'
I
I
0
'
1
'
I
'
30
I [it~
DNA CONTENT FIG. 1. Representative flow cytometric cytograms of twoparameter contour plot of data collected after dual staining for bromodeoxyuridine incorporation (BUDR, green fluorescence), in ordinate and DNA content (red fluorescence in abscissa) of EUE cells in isotonic (IT) and hypertonic (HT) conditions of culture.
IT
~ :~i,:i!! ~
I t~
'
I
HT
J
i
'
I 3t~
1
'
1
,
ot
'
I
'
I
'
1
'
I
'
I
DNA CONTENT FIG. 2. Bivariate distribution of anti-PCNA polyclonal antibody (stained with FITC) vs DNA content (stained with propidium iodide) of EUE cells in the two different conditions of growth. percentage of BUDR-labeled cells. The percentage of PCNA-positive cells was always higher than that of BUDR-labeled cells. As already shown by Kurki et al. [33], FCM analysis shows that both the percentage of PCNA-positive cells and the pattern of PCNA expression (as evaluated from the intensity of green fluorescence in the two-parameter contour plot) progressively increased with time after stimulation and was correlated with cell progression through the cell cycle. In EUE cells, FCM analysis of DNA distribution shows that
the content of cells in the different cell cycle phases was changed after growth in HT medium: the mean frequency _+ S.D. of cells in S-phase was reduced (from 30% - 3.0 in IT, compared with 11.5% - 2.2), while G2-M phase cells were more frequent (about 20%-+ 4.0, instead of 10%-+ 2.5). Simultaneous FCM measurement of incorporated B U D R (after 20 min pulse labeling) and DNA content confirmed this feature: the mean percentage -+ S.D. of green fluorescent cells in S-phase was reduced (from about 20.5%-+ 1.5 to 12% +-2.2) (Fig. 1). Also in this
p(TNA and cell cycle in acute leukemia
969
HT
IT I"I
!, I
Il'
'
I
'
|
I
61:1
30
'
I
I
'
I
I
3(a
~t
60
DNA CONTENT FIG. 3. Same experiment as depicted in Fig. 2. with PCNA recognized by means of the 19F4 monoclonal antibody.
C.G., 5 7 y r s AnLL
! i '~
' 'I ' ' ' ' I' I O0
'~
' i 'f''y ' = 2~30
I o
I
'
I 30
'
I
'
I
~
1
DNA CONTENT FIG. 4. Representative DNA histogram (left) and 2-parameter contour plot (anti-PCNA monoclonal antibody 19F4 vs propidium iodide DNA distribution) of bone marrow cells in one of the studied patients with acute non-lymphoblastic leukemia (ANLL). model, the percentage of PCNA-positivc cells was always higher than that of BUDR-labeled cells. With 19F4 MoAb, the mean percentage _+ S.D. of PCNApositive cells( 85% --+ 5.3), decreased significantly in HT medium (60% _ 6.5) where, the expression of PCNA was predominantly present in GI cells. The cytometric pattern of PCNA expression in IT EUE showed that the amount of PCNA increased from GI through S- and G2-M phases. With the polyclonal Ab, the background fluorescence was lowest in G 1and gradually increased during the cell cycle to a level in Gz--M that was approximately twice the Gl value (Figs 2 and 3).
Among the tested hemopoietic cell lines a differcnce was observed in the percentage of PCNApositive cells (both with the polycional Ab and the 19F4 MoAb) between Daudi and MOLT4 vs HL-60 and KG1, with different proliferative activity. However, the percentage of PCNA-positive cells (mean values -+ S.D. of 80% +-- 4.4 for Daudi and MOLT4 and of 95 % +__3.8 for HL-60 and KG 1respectively) was always higher than that of cells in S-phase. The mean values of frequency _ S.D. of cells in Sphase as evaluated from the DNA histograms were in fact 10% +-- 2.2 for Daudi and MOLT4 and 30% --- 4.2 for HL-60 and KG 1 while the mean
970
M. GIORDANO et al.
[7'] DNA FCM S-PHASE CELLS 100 ¸ 90 t~
80 ¸
,d "<
70
Z '< r.'.
6O 50
,.d ,.3
4O
k)
30 20 l0
EUE IT
DAUD! MOLT 4
HL-60
BM BLASTS
K(; 1 FIG. 5. Relationship between the mean percentages of: (i) S-phase cells (by DNA flow cytometry), (ii) BUDR-positive cells and (iii) PCNA-positive cells (as evaluated with thc 19F4 MoAb) in EUE cells, in hemopoietic cell lines and in bone marrow cells from acute leukemia paticnts (IT, isotonic medium; BM, bone marrow). T A B L E 1. N E T FI.UORESCENCE INTENSITY (ARBrFRARY UNITS) OF CEt.I.S STAINED FOR P C N A CONTENT (BY MEANS OF 1 9 F 4 A N T 1 - P C N A MONOCLONAI. ANTIBODY) DURING DIFFERENT PtlASES O F T I I E (TELl. CYCLE (TItE NET FLUORESCENCE INTENSITY WAS OBTAINED BY SUBTRAf_*rlNG THE BACKGROUND FLUORESCENCE GIVEN BY STAINING WlTII NORMAL I g G )
Net fluorescence intensity in cell cycle phases (arbitrary units, mean z S.D.) Cell type PBL EUE Daudi MOLT4 HL-60 KG~ BM cells
No. of experiments
Go/G~
3 3 2 2 2 2 5
14.2 (+-5) 15.9 (-+8) 24.0 (-+-3) 22.0 (+-2) 18.0 (+-2) 21.2 (-+2) 23.8 (--+5)
Earlv S 18.3 21).7 27.2 28.3 23.1 25.4 28.1
-+4) +-3) -+3) •+5) -+2) ±3) -+3)
Late S 20.4 25.9 32.4 31.2 27.7 27.5 30.3
(-+4) (-+2) (-+1) (±4) (+-2) (-+2) (-+1)
G 2+
26.8 27.8 35.2 34.3 29.2 29.8 32.7
M
(---4) (-+3) (-*-2) (-+4) (-+-3) (-+3) (-+1)
PBL, PHA-stimulated cells; EUE, cells grown in isotonic medium; BM, bone marrow; ES, early S; LS. late S. percentages--+ S.D. of BUDR-positive cells were 10% +- 2.5 and 25% +- 5.5. In BM cells from the 10 A N L L patients, the mean value --+ S.D. of PCNA-positive cells was 94% +- 3.2, the mean DNA-FCM S-phase value +-S.D. was 15% -+ 2.1 while the mean value -+ S.D. of B U D R positive cells was 10% - 3.5. Figure 4 shows the relationship between the mean values of percentages of D N A - F C M S-phase cells of BUDR-positive cells and of PCNA-positive cells (as
evaluated with the 19F4 MoAb) in E U E cells, in hemopoietic cell lines and in acute leukemia cells. The mean PCNA-fluorescence levels (expressed as arbitrary intensity units) observed for all hemopoietic cell lines and in leukemic blasts from all patients were significantly higher than for PHA-stimulated PBL and E U E cells but the pattern of PCNA expression (both with the polyclonal Ab and the MoAb) confirmed that, while reactivity with anti-PCNA MoAb was observed in all phases of the cell cycle, immunofiuorescence
PCNA and cell cycle in acute leukemia levels increased as cells progressed from the G ~phase into the S- and G2-M compartments (Fig. 5). Table 1 summarizes the mean green fluorescence values (obtained with the anti-PCNA MoAb 19F4) in each cell cycle phase for the different cell types studied. DISCUSSION The molecular mechanisms, in addition to DNA synthesis, which are linked with normal and malignant cell proliferation have been the subject of investigation since 1980. This has led cell kinetics to change somewhat from a pure description of classic kinetic parameters into a more subtle dynamic picture, where cell progression through the cell cycle is also described by the successive appearance of different molecules [2,34]. These molecules (which include growth factors, receptors for growth factors and other specific proteins, some of which are oncogene products) are termed "cell cycle-dependent", in that they are expressed in a cell cycle-dependent manner [1]. Their transcription and/or synthesis and/or accumulation change, in fact, according to the proliferation state and may also differ in the different phases of the cell cycle. Relatively simple immunocytochemical techniques are now available for detecting "cell cycle-related" proteins. Bivariate FCM permits a direct correlation of protein expression and DNA distribution, i.e. the evaluation of differences in protein expression in the different cell cycle phases [3]. PCNA is a 36 kD nuclear protein that is upregulated in activated proliferating cells from a variety of tissucs and species [13]. The use of anti-sense oligonucleotides suggests that PCNA is an essential requirement for DNA synthesis [24] and has also been shown to be required for leading strand synthesis in SV40 virus replication [35]. PCNA functions as a co-factor for DNA polymerase delta in DNA synthesis [ 14-17, 36], but may also be involved in unscheduled DNA synthesis [37-38]. The protein is highly conserved throughout phylogeny, being present in plants, yeast and higher eukaryotes [39]. PCNA is regulated in a complex manner with the gene being transcribed efficiently in both quiescent and proliferating cells, but PCNA mRNA normally only accumulates in proliferating cells [40]. The absence of stable PCNA mRNA in quiescent cells is associated with the prescence of intron 4 in the genc; removal of this intron leads to high levels of accumulation of PCNA mRNA in such cells [41]. Accumulation of the PCNA mRNA and the synthesis of high levels of the protein is stimulated by growth factors, but is not necessarily associated with DNA synthesis: PCNA accumulate in the presence of hydroxyurea, which inhibits DNA synthesis [21, 42-45].
971
PCNA/cyclin is the target antigen for autoantibodies in the sera of about 3% of patients with SLE [8-12]. Although these autoantibodies occur in low frequency in SLE, they have become valuable reagents for the study of a key cellular component of the cell nucleus in the late G I and early S-phase of the cell cycle [28, 40]. The nature of the autoimmune response against PCNA has been extensively investigated, and the PCNA-encoding-DNA cloned, so that extensive information is now available concerning its molecular nature and function. In this field, two murine MoAbs (IgG 19F4 and IgM 19A2) raised against purified rabbit PCNA were obtained and tested for specificity with eight SLE sera with antiPCNA activity. From this study, antigenic domains reacting with SLE sera were shown to be located in both the NH2- and COOH-terminal halves of PCNA, whereas the two murine MoAbs reacted with a different domain located between the two lupus domains [30]. These MoAbs exhibited other interesting differences when compared with human PCNA-specific autoantibodies. While PCNA autoantibodies react with cells from late G 1to G 2-M phases of the cell cycle, Kurki et al. [30] reported that PCNA/cyclin is expressed primarily in the S-phase, relative to both G 1 and G2/M phases. Moreover, a recent study reported that by combining the properties of PCNA/cyclin (which was considered as S-phase related) and Ki-67 MoAb (which has increasing expression during S- and maximal expression in G t-M phases) a pattern could emerge that might make it possible to discriminate all cell cycle phases based on the differential expression of the antigens in a two-parameter flow cytometric analysis [46]. One of the technical difficulties that can affect the flow cytometric analysis of intracellular antigens is sample fixation and permeabilization. Hence, the reported difference in the cell cycle-phase distribution between human Abs and mouse MoAbs could partially depend on the fixation/permeabilization procedures employed. This has been shown both with enzyme [47, 48] and with immunofluorescence [31] immunocytochemistry. An additional consideration is that quantitation by indirect immunofluoresccnce could depend on modification of the PCNA/cyclin protein. Since no cell cycle-dependent modifications of PCNA have been detected [49] the most likely possibility is a modified immunogenicity resulting from association with other cellular components. In particular, immunofluorescence studies revealed two populations of cyclin, one probably associated with the sites of DNA replication and another that is homogeneously spread in the nucleoplasm and which is easily extracted by using non-ionic detergents [50].
972
M. GIORDArqOet al.
We tried several fixation/permeabilization techniques to evaluate PCNA/cyclin as a marker for the detection of a proliferative state of human cells of normal and leukemic origin, in suspension and utilizing bivariate FCM. Extensive experiments were conducted to identify the fixation/permeabilization procedure which fournishes: (1) specific and quantitative immunofluorescence staining patterns; (2) high resolution D N A profiles and (3) minimal cell clumping. A higher background was observed using polycional Ab both in normal and in leukemic culture and fresh cells. This could be, at least in part, due to an aspecific reaction against proteins other than PCNA. The combination of 70% ethanol, 0.5% Triton X 100 and lysolecytin permitted a satisfactory immunofluorescence analysis of nuclear PCNA content, with both polyclonal and MoAbs and the simultaneous quantitation of DNA. In all cell types we studied, the PCNA "profile" during the cell cycle was similar (with the highest PCNA signal obtained from S-phase cells but also with a continuous increase during G 1and a subsequent decrease in G2-M). Therefore, compared with the polyclonal anti-PCNA, we did not find a selective S-phase distribution for the 19F4 MoAb, but only a higher specificity resulting in a better FCM signal. On the basis of our experiments the total frequency of PCNA-positive cells (both with the polyclonal and with the MoAb) was always found to be larger than that of BUDRpositive cells and we were always able to demonstrate that the 19F4 M o A b reacts with cells from late G1 to G2-M phase of cell cycle. We conclude that the MoAb 19F4, represents a suitable reagent for the FCM evaluation not only of Sphase cells but of all the proliferating compartment of human cell populations. In this view, the detection of PCNA expression and of B U D R incorporation should be regarded as complementary approaches for studying cell proliferation in normal and leukemic cells [5153]. The application of the 19F4 MoAb and perhaps of other recently-produced specific MoAbs [54, 55] will certainly stimulate further investigation of PCNA as a possible proliferation marker with prognostic value in human leukemia. Acknowledgements--We thank the technician staff of the Department of Animal Biology at the University of Pavia, for growing the cell lines. Part of this work was done while Dr M. Giordano was a Visiting Fellow at the Gray Laboratory, Northwood, U.K., supported by an International Cancer Research TechnologyTransfer award from U. I.C.C. (International Union Against Cancer, Gen~ve). Research supported by C.N.R. (Consiglio Nazionale delle Ricerche, Roma, Progetto Finalizzato Oncologia, Grant No. 88.01.01422.44), by A.I.R.C. (Associazione Nazionale per la Ricerca sul Cancro, Milano), by I.R.C.C.S. Policlinico
San Matteo (Pavia) and by M.U.R.S.T. (Ministero Universit/~ e Ricerca Scientifica, Roma) "40%" and "60%" funding.
REFERENCES 1. Kaczmarek L. (1986) Protooncogene expression during the cell cycle. Lab. Invest. $4, 365. 2. Pardee A. B. (1986) Molecules involved in proliferation of normal and cancer cells. Ann. Ass. Cancer Res. 27, 433. 3. Danova M., Riccardi A., Giordano M., Girino M., Mazzini G., Dezza L. & Ascari E. (1988) Cell cycle-related proteins: a flow cytofluorimetricstudy in human tumors. Biol. Cell 64, 23. 4. Giordano M., Danova M., Riccardi A. & Mazzini G. (1989) Simultaneous detection of cellular ras p21 oncogene product and DNA content by two parameter flow cytometry. Anticancer Res. 9, 799. 5. Danova M., Giordano M., Mazzini G. & Riccardi A. (1990) Expression of p53 oncoprotein during the cell cycle in human leukemia: a flow cytofluorimetric study. Leukemia Res. 14, 417. 6. Stewart C. C. (1988) Flow cytometric analysis of oncogene expression in human neoplasias. Arch. Pathol. Lab. Med. 113, 634. 7. Danova M., Riccardi A. & Mazzini G. (1990) Cell cyclerelated proteins and flow cytometry. Haematologica 75, 252. 8. Miyachi K., Fritzler M. J. & Tan E. M. (1978) Antibody to a nuclear antigen in proliferating cells. J. lmmun. 121, 2228. 9. Takasaki Y., Fishwild D. & Tan E. M. (1984) Characterization of proliferating cell nuclear antigen recognized by autoantibody in lupus sera J. expl Med. 159, 981. 10. Ogata K., Ogata Y., Nakamura R. N. & Tan E. M. (1985) Purification and N-terminal amino acid sequence of proliferating cell nuclear antigen (PCNA)/cyclin and development of ELISA for anti-PCNA antibodies. J. Immun. 135, 2623. 11. Tan E. M., Chan K. L., Sullivan K. F. & Rubin R. L. (1988) Antinuclear antibodies (ANAs): diagnostically specific immune markers and clues toward the understanding of systemic autoimmunity. Clin. Immun. Immunopathol. 47, 121. 12. Tan C. K. (1982) Autoantibodies to nuclear antigens (ANA) their immunobiology and medicine. Adv. Immun. 33, 167. 13. Mathews M. B., Bernstein R. M., Franza B. R. & Garrels J. I. (1984) The identity of the proliferating cell nuclear antigen and cyclin. Nature 303, 374. 14. Tan C. K., Castillo A. G., So A. G. & Downey K. M. (1986) An auxiliary protein for DNA polymerase delta from fetal calf thymus. J. biol. Chem. 261, 12310. 15. Bravo R., Frank R., Blundell B. A. & MacDonaldBravo H. (1987) Cyclin/PCNA is the auxiliary protein of DNA polymerase. Nature 326, 515. 16. Prelich G., Tan C. K., Kostura M., Mathews M. B., So G. A., Downey K. M. & Stillman B. (1987) Functional identity of proliferating cell nuclear antigen and a DNA polymerase auxiliary protein. Nature 326, 517. 17. Bauer G. & Burgers P. M. J. (1985) The yeast analog of mammalian cyclin/proliferating cell nuclear antigen
PCNA and cell cycle in acute leukemia interacts with mammalian DNA polymerase delta. Proc. natn. Acad. Sci. U.S.A. 85, 7506. 18. Celis J. & Bravo R. (1984) Synthesis of the nuclear protein cyclin in growing, senescent and morphologically transformed human skin fibroblasts. FEBS Letts 165, 21. 19. Celis J. E. & Celis A. (1985) Cell cycle-dependent variation in the distribution of the nuclear protein cyclin proliferating cell nuclear antigen in cultured cells. Subdivision of S phase. Proc. natn. Acad. Sci. U.S.A. 82, 3262. 20. Almendral J. M., Huebsch D., Blundell P. A., MacDonald-Bravo H. & Bravo R. (1987) Cloning and sequence of the human nuclear protein cyclin: homology with DNA-binding proteins. Proc. Am. Ass. Sci. 84, 1575. 21. Jaskulski D., Gatti G., Travali S., Calabretta B. & Baserga R. (1988) Regulation of the proliferating cell nuclear antigen cyclin and thymidine kinase in RNA levels by growth factors. J. biol. Chem. 263, 10175. 22. Huff J. P., Roos G., Peebles C. L., Houghten R., Sullivan K. F. & Tan E. M. (1989) Insight into native epitopes of proliferating cell nuclear antigen using recombinant DNA protein products. J. expl Med. 172, 419. 23. Travali S., Ku D. H., Rizzo M. G., Ottavio L., Baserga R. & Calabretta B. (1989) Structure of the human gene for the proliferating cell nuclear antigen. J. biol. Chem. 264, 7466. 24. Rizzo M. G., Ottavio L., Travali S., Chang C., Kaminska B. & Baserga R. (1990) The promoter of the human proliferating cell nuclear antigen (PCN A) is bidirectional. Expl Cell Res. 188, 286. 25. Jaskulski D., DeRiel J. K., Mercer W. E., Calabretta B. & Baserga R. (1988) Inhibition of cellular proliferation by antisense oligodeoxynucleotides to PCN A. Science 240, 1544. 26. Zuber M., Tan E. M. & Ryoji M. (1989) Involvement of proliferating cell nuclear antigen (cyclin) in DNA replication in living cells. Mol. Cell Biol. 9, 57. 27. Takasaki Y., Robinson W. A. &Tan E. M. (1984) Proliferating cell nuclear antigen in blast crisis cells of patients with chronic myeloid leukemia. J.N.C.I. 73, 655. 28. Smetana K., Gyorkey F., Chan P. K., Tan M. & Busch H. (1983) Proliferating cell nuclear antigen (PCNA) and human malignant tumor nucleolar antigens (HTNA) in nucleoli of human hematological malignancies. Blur46, 133. 29. Miygawa S., Okada N., Takasaki Y., Kitano Y., Yoshikawa K., Sakamoto K. & Steinberg M. L. (1989) Expression of proliferating cell nuclear antigen/cyclin in human keratinocytes. J. Invest. Dermatol. 93, 678. 30. Ogata K., Kurki P., Celis J. E., Nakamura R. M. & Tan E. M. (1987) Monoclonal antibodies to a nuclear protein (PCNA/cyclin) associated with DNA replication. Expl Cell Res. 168, 475. 31. Kurki K., Ogata K. & Tan E. M. (1988) Monoclonal antibodies to proliferating cell nuclear antigen (PCNA/ cyclin) as probes for proliferating cells by immunofluorescence microscopy and flow cytometry. J. Immun. Meth. 109, 49. 32. Pellicciari C., Mazzini G., Fuhrman Conti A. M., De Grada L. & Manfredi Romanini M. G. (1989) Effects of hypertonic medium on human cell growth: III. Changes in cell kinetics of EUE cells. Cell Biol. Int. Rep. 13,345355.
973
33. Kurki P., Lotz M., Ogata K. & Tan E. M. (1987) Proliferating cell nuclear antigen (PCNA/cyclin) in activated human T lymphocytes. J. Immun. 138, 4114. 34. Baserga R. & Szurmak E. (1987) Oncogenes, cell cycle genes and the control of cell proliferation. Biotechnology 5, 355. 35. Prelich G. & Stilman B. C. (1988) Coordinate leading and lagging strand synthesis during SV40 DNA replication in vitro requires PCNA. Cell 53, 117. 36. Fairman M. P. (1988) DNA polymerase delta/PCNA: actions and interactions. J. Cell Sci 94, 1. 37. Celis J. & Madson P. (1986) Increased nuclear cyclin/ PCNA antigen staining of non-S phase transformed human amnion cells engaged in nucleotide excision DNA repair. FEBS Letts 209, 277. 38. Toschi L. & Bravo R. (1988) Changes in cyclin/PCNA distribution during DNA repair synthesis. J. Cell Biol. 107, 1623. 39. Suzuka I., Daidoji H., Matsuoka M., Kadowaki K., Takasaki Y., Nakane P. K. & Moriuchi T. (1989) Gene for proliferating cell nuclear antigen (DNA polymerase delta auxilliary protein) is present in both mammalian and higher plant genomas. Proc. natn. Acad. Sci. U.S.A. 86, 3189. 40. Celis J. E., Bravo R., Larsen P. M. & Fey F. T. (1984) Cyclin: a nuclear protein whose level correlates directly with the proliferative state of normal as well transformed cells. Leukemia Res. 8, 143. 41. Ottavio L., Chang C. D., Rizzo M. G., Travali S., Casadevall C. & Baserga R. (1990) Importance ofintrons in the growth regulation of mRNA levels of the proliferating cell nuclear antigen gene. Mol. Cell Biol. 10, 3(13. 42. Bravo R. & MacDonald-Bravo H. (1984) Existence of two populations of cyclin/proliferating cell nuclear antigen during the cell cycle: association with DNA replication sites. E M B O J. 3, 3177. 43. Bravo R. & MacDonald-Bravo H. (1985) Induction of the nuclear protein cyclin in quiescent mouse 3T3 cells stimulated by serum and growth factors. Correlation with DNA synthesis. E M B O J. 4, 655. 44. Chang C. D.. Ottavio g., Travali S., kipson K. E. & Baserga R. (1990) Transcriptional and post-transcriptional regulation of the proliferating cell nuclear antigen gene. Mol. Cell Biol. 10, 3289. 45. Liu Y. C., Marracino R. g., Keng P. C., Bambara R. A.. Lord E. M., Chart W. G. & Zain S. B. (1989) Requirement for proliferating cell nuclear antigen expression during stages of the Chinese hamster ovary cell cycle. Biochemistry 28, 2967. 46. Lanberg G., Tan E. M. & Roos G. (1990) Flow cytometric multiparameter analysis of proliferating cell nuclear antigen cyclin and Ki-67 antigen: a new view of the cell cycle. Expl Cell Res. 187, 111. 47. Galand P. & DegraefC. (1989) Cyclin/PCNA immunostaining as an alternative tritiated thymidine pulse labeling for marking S phase cells in paraffin sections from animal and human tissues. Cell Tissue Kinet. 22, 383. 48. Garcia R. L., Coltrera M. D. & Gown A. M. (1989) Analysis of proliferative grade using anti-PCNA/cyclin monoclonal antibodies in fixed, embedded tissues. Comparison with flow cytometric analysis. Am. J. Pathol. 134, 733. 49. Bravo R. & Celis J. E. (1985) Changes in the nuclear distribution of cyclin (PCNA) during S phase are not triggered by post translational modifications that are
974
M. GIORDANOet al.
expected to moderately affect its charge. FEBS Letts 185, 435. 50. Bravo R. & MacDonald-Bravo H. (1987) Existence of two populations of cyclin/proliferating cell nuclear antigen during the cell cycle: association with DNA replication sites. J. Cell Biol. 105, 1549. 51. Keim D. K., Hailat N., Hodge D. & Hanash M. (1990) Proliferating cell nuclear antigen expression in childhood acute leukemia. Blood 76, 985. 52. Kurki P., Vanderlaan M., Dolbeare F., Gray J. & Tan E. M. (1986) Expression of proliferating cell nuclear
antigen (PCNA/cyclin) during the cell cycle. Expl Cell Res. 166, 209.
53. Andreef M. (1986) Cell kinetics of leukemia. Semin. Hemat. 23, 300. 54. Murashima A., Takasaki Y., Ohgaki M., Hashimoto H., Shirai T. & Hirose S. (1990) Activated peripheral blood mononuclear cells detected by murine monoclonal antibodies to proliferating cell nuclear antigen in active lupus patients. J. clin. lmmun. 10, 28. 55. Waseem N. H. & Lane D. P. (1990) Monoclonal antibody analysis of the proliferating cell nuclear antigen (PCNA) J. Cell Sci. 96, 121.
0145-2126/91 $3.00 + .00 Pergamon Press plc
P R O L I F E R A T I N G CELL N U C L E A R A N T I G E N ( P C N A ) / C Y C L I N E X P R E S S I O N D U R I N G T H E CELL CYCLE IN N O R M A L A N D L E U K E M I C CELLS MONICA GIORDANO,* MARCO DANOVA,* CARLO PELI,ICCIARI,t GEORGE D. WILSON,~: GIULIANO MAZZINI,§ ANNA M. FUHRMAN CONTI,II GIOVANNI FRANCHINI,* ALBERTO RICCARDI* and MARIA G. MANFREDI ROMANINIt *Department of Internal Medicine, Section 2nd Medical Clinic; tDepartment of Animal Biology; University and I.R.C.C.S. San Matteo, Pavia; :~Gray Laboratory, Mount Vernon Hospital, Northwood, U.K. ; §Centro Studio Istochemica C.N.R., University and I.R.C.C.S. San Matteo, Pavia and IIBioiogy Medical Sciences, University of Milan, Italy.
(Received 11 April 1991. Accepted 18 May 1991) Abstract--Bivariate flow cytometric analysis of the cell proliferation-associated nuclear protein, identified as the "proliferating cell nuclear antigen" (PCNA)/cyclin and of nuclear DNA content, was performed in quiescent and mitogen-stimulated human peripheral blood lymphocytes, in EUE (human embryonic epithelium) cells, before and after a long-term exposure to a hypertonic (HT) medium, in 4 human leukemic cell lines and in fresh bone marrow (BM) cells from 10 patients with untreated acute non-lymphoblastic leukemia (ANLL). The PCNA/cyclin was detected using both an autoantibody extracted from sera of systemic lupus erythematosus patients and the recently produced mouse monoclonal antibody (MoAb) IgG, named 19F4. The distribution of cells in the different phases of the cycle and the percentage of S-phase cells were obtained in duplicate samples, by DNA flow cytometry (FCM) and by dual parameter FCM of DNA content and bromodeoxyuridine (BUDR) incorporation. In all cell types, the non-specific cytoplasmic background fluorescence was significantly lower with the MoAb compared to that obtained with the polyclonal Ab. The percentage of PCNApositive cells (both with the autoantibody and the 19F4 MoAb) was always higher than that of Sphase cells by DNA FCM and of BUDR-labeled cells. The pattern of PCNA-expression in both normal proliferating cells and acute leukemia cells, showed that most G0/G1 cells did not express significant amounts of PCNA; an increase in PCNA immunofluorescence was found in late GI cells, and further increases were observed in S- and Gz--M phase cells. PCNA/cyclin, as revealed both with autoantibodies and with the 19F4 MoAb, is associated with all actively or potentially dividing (i.e. GI, S and Gz-M) cells thus identifying the proliferative cellular compartment. Combined with the use of muitiparameter FCM techniques, the PCNA immunolocalization offers a useful tool to study cell kinetics in normal and leukemic human cell populations.
Key words: PCNA/cyclin, acute leukemia, cell proliferation. DNA content, flow cytometry, Sphase, growth fraction.
[1,2]. Dual parameter flow cytometry (FCM) of
INTRODUCTION
DNA content and of the immunolabeling of these proteins allows protein expression to be precisely related to each cell cycle phase in populations with different proliferative activity [3-7]. Sera from certain patients with systemic lupus erythematosus (SLE) contain autoantibodies to a proliferating cell nuclear antigen (PCNA) [8-12] also called cyclin [13], which is an evolutionary highly conserved 36 kD acidic nuclear polypeptide (identified as a polymerase delta accessory protein) [1417]. The expression and synthesis of PCNA are enhanced in proliferation compared to quiescent cells [18-19]. cDNA clones corresponding to the human PCNA have been isolated and described [20-22].
THE study of nuclear and surface proteins expressed in variable amounts in the different phases of the cell cycle may permit a better understanding of the mechanisms regulating cell proliferation providing additional parameters for describing the cell cycle
Abbreviations: BM, bone marrow; ANLL, acute noniymphoblastic leukemia; MoAb, monoclonal antibody; PI, propidium iodide; PBL, peripheral blood lymphocytes; PHA, phytohemagglutinin; FCM, flow cytometry; PCNA, proliferating cell nuclear antigen; BUDR, bromodeoxyuridine. Correspondenceto: Marco Danova, M.D., Ph. D., Clinica Medica 2, Universit~ di Pavia, Policlinico S. Matteo 27100 Pavia, Italy. 965
966
M. GIORDANO et al.
Travali et al. [23] obtained a clone containing the entire human PCNA gene from a lambda phage library, that was subsequently functionally characterized using a c D N A probe [24]. PCNA expression was found to increase during the G 1 phase of the cell cycle, reaching its maximum in the S-phase, and then declining during the G2-M phase. Exposure of exponentially growing cells to antisense oligodeoxynucleotides to PCNA resulted in complete suppression of D N A synthesis and mitosis, indicating an important role for this protein in cell proliferation [25, 26]. The SLE autoantibodies have been noted to be a useful marker for detecting proliferating cells in the peripheral blood of patients with leukemia [27], in solid tumors [28] and in normal tissues [29], both through immunocytochemistry and bivariate FCM [3]. However, anti-PCNA-positive SLE sera without other antibodies directed against nuclear components are extremely rare. The development of monoelonal antibodies (MoAbs) made it possible to improve the specificity of PCNA detection [30, 31]. In this study, we aimed to compare the patterns of PCNA expression resulting from the application of either polyclonal or MoAbs on different cell systems, i.e. quiescent and PHA-stimulated peripheral blood lymphocytes, E U E (human embryonic epithelium) cells grown in either isotonic (IT) or hypertonic ( H T ) medium, human hematopoietic cell lines of different origin and with different phenotypes and fresh bone marrow (BM) cells from 10 patients with untreated acute non-lymphoblastic leukemia (ANLL). In order to perform dual parameter FCM of PCNA expression and DNA content, a fixation/permeabilization procedure was developed to obtain a satisfactory binding of the anti-PCNA Abs, and the simultaneous staining of DNA with propidium iodide (PI). Results (in terms of percentages of positivity to PCNA and patterns of expression) were compared with those obtained from experiments conducted in duplicate samples with single parameter DNA-FCM and twoparameter FCM (DNA vs bromodeoxyuridine, B U D R , incorporation). MATERIALS AND METHODS Phytohemagglutinin
(PHA)-stimulated
lymphocytes.
Human peripheral blood lymphocytes (PBS) of healthy donors, were separated by Ficoll-Hypaque density gradient centrifugation. The preparation contained more than 90% lymphocytes. Cells were suspended in RPMI 1640 (Flow Laboratories, Inglewood, CA, U.S.A.), supplemented with 2raM glutamine, vitamins, non-essential amino acids, sodium pyruvate, 100 U/ml penicillin, 100 lxg/ml streptomycin, and 15% fetal bovine serum to a concentration of 1 x 106/ml. The cell suspension was divided into two groups. One was cultured in presence of
30 Ixg/ml of PHA (Wellcome Reagents Ltd, Greenville, NC, U.S.A.) for 72 h in 5% CO2/95% air at 37°C and the other without PHA as control. Cell lines. EUE cells [32] were selected as a suitable normal cell model in which the long-term exposure (4 days) to a culture medium with increased toxicity was found to affect cell kinetic parameters, inducing a Go block in most cells, and a slowing down of the S-phase in the still cycling cells [32]. Cells were grown for 4 days in either IT or HT in minimum essential Eagle's medium, supplemented with 12% newborn bovine serum and 1% gtutamine. To obtain the HT medium, the concentration of NaCI was raised, from 0.136M to 0.274 M, in the Hank's saline used as a component of the culture medium (the final osmolarities were 305 mOsm and 495 mOsm, in IT and HT respectively). The culture media were changed every two days, to avoid the influence of nutrient shortage. Four human hemopoietic cell lines were studied during their exponential growth phase in liquid culture: acute lymphoblastic leukemia cell line MOLT 4; Burkitt's lymphoma cell line (Daudi); myeloid cell lines: HL-60 (promyelocytic leukemia), and KGI (myeloblastic leukemia). The cells were continuously cultured (in 25 c m 3 plastic flasks, Corning, MD, U.S.A.) in RPMI medium (GIBCO, Grand Island, New York, U.S.A.) supplemented with 10% fetal calf serum (FCS, Flow laboratories, Rockville, MD, U.S.A.), 1% L-glutamine and non-essential amino acids in 5% CO2/95% air. MCF-7 cells were maintained in Dulbecco's modified Eagle's medium (DMEM, GIBCO) supplemented with 20 mM Hepes buffer, 10% FCS and 200 U/ml penicillin (Eurobio, Paris, France). Bone marrow (BM) cells. The source of BM leukemic cells was a group of 10 patients with ANLL studied at the time of initial diagnosis. All patients were advised of procedures and attendant risks in accordance with the institutional guidelines, and gave informed consent. Lightdensity mononuclear cells (LDMNCs) were collected after centrifugation on a Ficoll-Hypaque gradient, density 1077g/cm 3 (Pharmacia Fine Chemicals, Piscataway, NJ, U.S.A.) in Iscove's modified Dulbecco's medium (IMDM, GIBCO). The cells were further purified by removing adherent cells and T lymphocytes. A 5 ml suspension of light density cells at a concentration of 5 × 106 cells/ml in IMDM plus 15% FCS was incubated in 25 cm 3 tissue culture ltasks for 60min at 37°C, and non-adherent cells carefully collected. This procedure was repeated twice. Tlymphocyte-depleted LDMNCs (T-LDMNCs) were obtained by rosetting MNC suspensions (5 × 106 cells) with 2-aminoethylisothiouronium bromide-treated (AET, Sigma Chemical Co, St Louis, U.S.A.) sheep red blood cells in a 5% suspension with IMDM. Non-rosetting cells were separated by a second Ficoll-Hypaque density centrifugation. Viability after cell separation was evaluated by the trypan blue dye exclusion test. Anti-PCNA antibodies. The anti-PCNA-positive Ab, obtained from patients with SLE was purchased in fluid form from Delta Biological Inc. (Pomezia-Roma, 00040, Italy). The anti-PCNA mouse MoAb Ab 19F4-1gG raised against purified cyclin from rabbit thymus isolated and characterized by Ogata et al. [24], was purchased from American Biotech Inc, (Plantation, FI, 33313. U.S.A.) in an ascites fluid form. Indirect immunofluorescence procedures. The following fixation/permeabilization procedures were tested for both the polyclonal Ab and MoAb: 20 min in 70% methanol at
PCNA and cell cycle in acute leukemia 4°C; 20 rain in 70% ethanol at 4°C; 10 min in 90% acetone at 4°C plus 0.01% Triton X 100 (5 rain); 2 min in 1% paraformaldehyde at room temperature followed by 5 min in 70% methanol at -20°C; 10 min ethanol 70% at -20°C followed by 0.01% Triton X 100 (5rain); 10min 70% ethanol at 4°C, followed by 0.1% Triton X 100 (15 min) and the subsequent incubation (30 min) with the antiPCNA diluted in PBS plus lysolecytin (500 lag/ml). Cells were incubated with a 1:100 dilution of the anti-PCNA polyclonal Ab in PBS (Hoechst AG, F.R.G.) containing 0.5% normal rabbit serum (NRS, Flow Laboratories) and a 1:50 dilution of the MoAb in PBS and 0.5% normal goat serum (NGS, Flow Laboratories), for 30 rain at room temperature. Cells were then centrifuged and incubated for 15 min in either PBS/NRS or PBS/NGS, respectively. After two washes in PBS, the bound antibodies were labeled with a 1:10 dilution of either or fluorescein isothiocyanate (FITC)-conjugated rabbit anti-human IgG (Sigma Chem.) or FITC-conjugated goat anti-mouse IgG (Sigma Chem.) MoAbs, respectively. The levels of non-specific background staining (zero threshold) were established for each measurement using control cells processed at the same time but without exposure to primary antibodies. In all samples, counterstaining of double-stranded DNA was done with PI (5 lag/ml, containing 50 Kunitz units/ml RNAse for at least 30 min). DNA staining. For single-parameter DNA measurements, cells were suspended in 2 ml of phosphate-buffered saline (PBS, Hoechst AG, F.R.G.) and drawn through needles of decreasing diameter. Cell counts were made so as to have more than 1 x 1(16 ceils/ml per sample. The suspension was centrifuged and the pellet stained with propidium iodide (PI, Calbiochem, Behring Corp., San Diego, CA, U.S.A.) at a concentration of 50 lag/ml in PBS; 0.1% Nonidet P40 (Calbiochem.) and 50 Kunitz units/ml RNAse (type 1A, from bovine pancreas, Sigma) were included in the staining solution. A 30 min staining time at room temperature provided the best histogram resolution. Finally, cells were filtered through a 35 p,m nylon mesh to remove aggregates prior to flow analysis. BUDR incorporation stud),'. In all samples pulse-labeling with BUDR (Sigma Chem.) was performed by incubating the cells for 20rain at a final BUDR concentration of 10 p,M; the medium was then washed out, and pre-warmed fresh medium was added. After incubation with BUDR, the cells were harvested and immediately placed in an icecold bath, washed once with PBS, fixed in cold 70% ethanol and stored at -20°C. The BUDR incorporated into Sphase cells was detected with the MoAb against BUDR (Becton Dickinson) by the indirect immunofluorescencc technique which involves pretreatment with HCI (2 N for 10 min at room temperature), incubation with the antiBUDR MoAb (1:10) and then with a FITC-conjugated goat anti-mouse IgG MoAb (1:50). Cells were finally resuspended in 5 p,g/ml PI in PBS containing 50 Kunitz units/ml RNAse, and stained for at least 30 min. Bivariateflow cytometry. Two parameter FCM analyses (FITC-green vs PI-red) were performed with a FACStar Cell Sorter (Becton Dickinson, Sunnyvale, CA, U.S.A.). Excitation was accomplished with an argon ion laser tuned to 488 nm and operated at 300 mW. Emitted fluorescence was split into two bands by means of a dichromatic mirror DM 560. Green fluorescence was measured through a 525 nm interference filter (30nm h.b.w.) while the red fluorescence was measured using a 620 nm long pass filter.
967
Instrumentation setting and calibration were done daily using fluorescent microbeads. Data were collected with a Consort 30 software program running on a dedicated Hewlett Packard computer and displayed as dual parameter contour density plots. At least 20000 cells were analyzed for each sample. Evaluation of FCM data on two-parameter plots was done with box analysis. Early S-(ES) phase cells were operationally defined as FITC-positive cells falling in the first half of the S-phase scale of DNA content, i.e. between the G0/Gl peak and half the distance between the G0/G l and G2-M peaks. Late S-(LS) phase cells were those in the second half of this interval. The levels of expression of the antigens studied in the G0/G l, ES, LS and G2-M phases of the cell cycle were approximately evaluated from the mean green fluorescence intensity of the corresponding FITC-labeled cells. The green fluorescence values for the corresponding boxes in matched control samples were also calculated in each FCM run. This fluorescence (which was set at the same position on the cytogram each time) was subtracted as an aspecific background to obtain the "'net" fluorescence intensity associated with the specific labeling of PCNA. DNA flow cytometry. Single FCM-DNA analyses were performed with a Partec PAS II (Basel, Switzerland) arc lamp flow cytometer, with data displayed as frequency histograms. For each sample, 10 0(~-40 000 cells were analyzed. The measuring conditions included the following: an HBO I(X)W/2 (Osram) excitation source with KGI (2 mm) and BG38 (4 mm) filters; an interference filter of 546 ± 12 nm; a TK 590 dichromatic mirror and a K 610 barrier filter to select the emitted red fluorescence. The percentage distribution of cells in the Go/G l, S- and G2-M phases of the cell cycle was obtained using Fried's mathematical model (which fits the data with a Gaussian distribution) adapted for a dedicated Hewlett Packard computer.
RESULTS We performed a total of 30 experiments: 6 with normal PBL, 6 with the E U E cell line, 8 with the hemopoietic cell lines and 10 with leukemic blasts from the patients with A N L L and the results obtained are shown in Figs 1-5 and in Table 1. We tested several methods of fixation/ permeabilization in order to optimize the reaction with the anti-PCNA polyclonal and/or MoAb. The best results were obtained by treating cells with 70% cold ethanol for 10 min, followed by 0.5% Triton X 100 (15 min) and the subsequent incubation (30 min) with the anti-PCNA diluted 1:50 in PBS plus lysolecytin (500 ~tg/ml). This procedure was then used in all the experiments. As expected [8], in resting lymphocytes, the P C N A was not detectable. After PHA-stimulation, the results obtained with two-parameter FCM of cells labeled with the FITC-conjugated polyclonal A b or 19F4 M o A b anti-PCNA antibody and simultaneously stained for D N A with PI, were consistent with the observed changes in the D N A distribution and in the
968
M. GtORDANOet al. 13 (a--
IT
HT .
.
¢
i
.
•
,~•
"
:
:
i( I
1
1
'
1
'
1
f
3~
~
&O
J
'
t
'
I
I
0
'
1
'
I
'
30
I [it~
DNA CONTENT FIG. 1. Representative flow cytometric cytograms of twoparameter contour plot of data collected after dual staining for bromodeoxyuridine incorporation (BUDR, green fluorescence), in ordinate and DNA content (red fluorescence in abscissa) of EUE cells in isotonic (IT) and hypertonic (HT) conditions of culture.
IT
~ :~i,:i!! ~
I t~
'
I
HT
J
i
'
I 3t~
1
'
1
,
ot
'
I
'
I
'
1
'
I
'
I
DNA CONTENT FIG. 2. Bivariate distribution of anti-PCNA polyclonal antibody (stained with FITC) vs DNA content (stained with propidium iodide) of EUE cells in the two different conditions of growth. percentage of BUDR-labeled cells. The percentage of PCNA-positive cells was always higher than that of BUDR-labeled cells. As already shown by Kurki et al. [33], FCM analysis shows that both the percentage of PCNA-positive cells and the pattern of PCNA expression (as evaluated from the intensity of green fluorescence in the two-parameter contour plot) progressively increased with time after stimulation and was correlated with cell progression through the cell cycle. In EUE cells, FCM analysis of DNA distribution shows that
the content of cells in the different cell cycle phases was changed after growth in HT medium: the mean frequency _+ S.D. of cells in S-phase was reduced (from 30% - 3.0 in IT, compared with 11.5% - 2.2), while G2-M phase cells were more frequent (about 20%-+ 4.0, instead of 10%-+ 2.5). Simultaneous FCM measurement of incorporated B U D R (after 20 min pulse labeling) and DNA content confirmed this feature: the mean percentage -+ S.D. of green fluorescent cells in S-phase was reduced (from about 20.5%-+ 1.5 to 12% +-2.2) (Fig. 1). Also in this
p(TNA and cell cycle in acute leukemia
969
HT
IT I"I
!, I
Il'
'
I
'
|
I
61:1
30
'
I
I
'
I
I
3(a
~t
60
DNA CONTENT FIG. 3. Same experiment as depicted in Fig. 2. with PCNA recognized by means of the 19F4 monoclonal antibody.
C.G., 5 7 y r s AnLL
! i '~
' 'I ' ' ' ' I' I O0
'~
' i 'f''y ' = 2~30
I o
I
'
I 30
'
I
'
I
~
1
DNA CONTENT FIG. 4. Representative DNA histogram (left) and 2-parameter contour plot (anti-PCNA monoclonal antibody 19F4 vs propidium iodide DNA distribution) of bone marrow cells in one of the studied patients with acute non-lymphoblastic leukemia (ANLL). model, the percentage of PCNA-positivc cells was always higher than that of BUDR-labeled cells. With 19F4 MoAb, the mean percentage _+ S.D. of PCNApositive cells( 85% --+ 5.3), decreased significantly in HT medium (60% _ 6.5) where, the expression of PCNA was predominantly present in GI cells. The cytometric pattern of PCNA expression in IT EUE showed that the amount of PCNA increased from GI through S- and G2-M phases. With the polyclonal Ab, the background fluorescence was lowest in G 1and gradually increased during the cell cycle to a level in Gz--M that was approximately twice the Gl value (Figs 2 and 3).
Among the tested hemopoietic cell lines a differcnce was observed in the percentage of PCNApositive cells (both with the polycional Ab and the 19F4 MoAb) between Daudi and MOLT4 vs HL-60 and KG1, with different proliferative activity. However, the percentage of PCNA-positive cells (mean values -+ S.D. of 80% +-- 4.4 for Daudi and MOLT4 and of 95 % +__3.8 for HL-60 and KG 1respectively) was always higher than that of cells in S-phase. The mean values of frequency _ S.D. of cells in Sphase as evaluated from the DNA histograms were in fact 10% +-- 2.2 for Daudi and MOLT4 and 30% --- 4.2 for HL-60 and KG 1 while the mean
970
M. GIORDANO et al.
[7'] DNA FCM S-PHASE CELLS 100 ¸ 90 t~
80 ¸
,d "<
70
Z '< r.'.
6O 50
,.d ,.3
4O
k)
30 20 l0
EUE IT
DAUD! MOLT 4
HL-60
BM BLASTS
K(; 1 FIG. 5. Relationship between the mean percentages of: (i) S-phase cells (by DNA flow cytometry), (ii) BUDR-positive cells and (iii) PCNA-positive cells (as evaluated with thc 19F4 MoAb) in EUE cells, in hemopoietic cell lines and in bone marrow cells from acute leukemia paticnts (IT, isotonic medium; BM, bone marrow). T A B L E 1. N E T FI.UORESCENCE INTENSITY (ARBrFRARY UNITS) OF CEt.I.S STAINED FOR P C N A CONTENT (BY MEANS OF 1 9 F 4 A N T 1 - P C N A MONOCLONAI. ANTIBODY) DURING DIFFERENT PtlASES O F T I I E (TELl. CYCLE (TItE NET FLUORESCENCE INTENSITY WAS OBTAINED BY SUBTRAf_*rlNG THE BACKGROUND FLUORESCENCE GIVEN BY STAINING WlTII NORMAL I g G )
Net fluorescence intensity in cell cycle phases (arbitrary units, mean z S.D.) Cell type PBL EUE Daudi MOLT4 HL-60 KG~ BM cells
No. of experiments
Go/G~
3 3 2 2 2 2 5
14.2 (+-5) 15.9 (-+8) 24.0 (-+-3) 22.0 (+-2) 18.0 (+-2) 21.2 (-+2) 23.8 (--+5)
Earlv S 18.3 21).7 27.2 28.3 23.1 25.4 28.1
-+4) +-3) -+3) •+5) -+2) ±3) -+3)
Late S 20.4 25.9 32.4 31.2 27.7 27.5 30.3
(-+4) (-+2) (-+1) (±4) (+-2) (-+2) (-+1)
G 2+
26.8 27.8 35.2 34.3 29.2 29.8 32.7
M
(---4) (-+3) (-*-2) (-+4) (-+-3) (-+3) (-+1)
PBL, PHA-stimulated cells; EUE, cells grown in isotonic medium; BM, bone marrow; ES, early S; LS. late S. percentages--+ S.D. of BUDR-positive cells were 10% +- 2.5 and 25% +- 5.5. In BM cells from the 10 A N L L patients, the mean value --+ S.D. of PCNA-positive cells was 94% +- 3.2, the mean DNA-FCM S-phase value +-S.D. was 15% -+ 2.1 while the mean value -+ S.D. of B U D R positive cells was 10% - 3.5. Figure 4 shows the relationship between the mean values of percentages of D N A - F C M S-phase cells of BUDR-positive cells and of PCNA-positive cells (as
evaluated with the 19F4 MoAb) in E U E cells, in hemopoietic cell lines and in acute leukemia cells. The mean PCNA-fluorescence levels (expressed as arbitrary intensity units) observed for all hemopoietic cell lines and in leukemic blasts from all patients were significantly higher than for PHA-stimulated PBL and E U E cells but the pattern of PCNA expression (both with the polyclonal Ab and the MoAb) confirmed that, while reactivity with anti-PCNA MoAb was observed in all phases of the cell cycle, immunofiuorescence
PCNA and cell cycle in acute leukemia levels increased as cells progressed from the G ~phase into the S- and G2-M compartments (Fig. 5). Table 1 summarizes the mean green fluorescence values (obtained with the anti-PCNA MoAb 19F4) in each cell cycle phase for the different cell types studied. DISCUSSION The molecular mechanisms, in addition to DNA synthesis, which are linked with normal and malignant cell proliferation have been the subject of investigation since 1980. This has led cell kinetics to change somewhat from a pure description of classic kinetic parameters into a more subtle dynamic picture, where cell progression through the cell cycle is also described by the successive appearance of different molecules [2,34]. These molecules (which include growth factors, receptors for growth factors and other specific proteins, some of which are oncogene products) are termed "cell cycle-dependent", in that they are expressed in a cell cycle-dependent manner [1]. Their transcription and/or synthesis and/or accumulation change, in fact, according to the proliferation state and may also differ in the different phases of the cell cycle. Relatively simple immunocytochemical techniques are now available for detecting "cell cycle-related" proteins. Bivariate FCM permits a direct correlation of protein expression and DNA distribution, i.e. the evaluation of differences in protein expression in the different cell cycle phases [3]. PCNA is a 36 kD nuclear protein that is upregulated in activated proliferating cells from a variety of tissucs and species [13]. The use of anti-sense oligonucleotides suggests that PCNA is an essential requirement for DNA synthesis [24] and has also been shown to be required for leading strand synthesis in SV40 virus replication [35]. PCNA functions as a co-factor for DNA polymerase delta in DNA synthesis [ 14-17, 36], but may also be involved in unscheduled DNA synthesis [37-38]. The protein is highly conserved throughout phylogeny, being present in plants, yeast and higher eukaryotes [39]. PCNA is regulated in a complex manner with the gene being transcribed efficiently in both quiescent and proliferating cells, but PCNA mRNA normally only accumulates in proliferating cells [40]. The absence of stable PCNA mRNA in quiescent cells is associated with the prescence of intron 4 in the genc; removal of this intron leads to high levels of accumulation of PCNA mRNA in such cells [41]. Accumulation of the PCNA mRNA and the synthesis of high levels of the protein is stimulated by growth factors, but is not necessarily associated with DNA synthesis: PCNA accumulate in the presence of hydroxyurea, which inhibits DNA synthesis [21, 42-45].
971
PCNA/cyclin is the target antigen for autoantibodies in the sera of about 3% of patients with SLE [8-12]. Although these autoantibodies occur in low frequency in SLE, they have become valuable reagents for the study of a key cellular component of the cell nucleus in the late G I and early S-phase of the cell cycle [28, 40]. The nature of the autoimmune response against PCNA has been extensively investigated, and the PCNA-encoding-DNA cloned, so that extensive information is now available concerning its molecular nature and function. In this field, two murine MoAbs (IgG 19F4 and IgM 19A2) raised against purified rabbit PCNA were obtained and tested for specificity with eight SLE sera with antiPCNA activity. From this study, antigenic domains reacting with SLE sera were shown to be located in both the NH2- and COOH-terminal halves of PCNA, whereas the two murine MoAbs reacted with a different domain located between the two lupus domains [30]. These MoAbs exhibited other interesting differences when compared with human PCNA-specific autoantibodies. While PCNA autoantibodies react with cells from late G 1to G 2-M phases of the cell cycle, Kurki et al. [30] reported that PCNA/cyclin is expressed primarily in the S-phase, relative to both G 1 and G2/M phases. Moreover, a recent study reported that by combining the properties of PCNA/cyclin (which was considered as S-phase related) and Ki-67 MoAb (which has increasing expression during S- and maximal expression in G t-M phases) a pattern could emerge that might make it possible to discriminate all cell cycle phases based on the differential expression of the antigens in a two-parameter flow cytometric analysis [46]. One of the technical difficulties that can affect the flow cytometric analysis of intracellular antigens is sample fixation and permeabilization. Hence, the reported difference in the cell cycle-phase distribution between human Abs and mouse MoAbs could partially depend on the fixation/permeabilization procedures employed. This has been shown both with enzyme [47, 48] and with immunofluorescence [31] immunocytochemistry. An additional consideration is that quantitation by indirect immunofluoresccnce could depend on modification of the PCNA/cyclin protein. Since no cell cycle-dependent modifications of PCNA have been detected [49] the most likely possibility is a modified immunogenicity resulting from association with other cellular components. In particular, immunofluorescence studies revealed two populations of cyclin, one probably associated with the sites of DNA replication and another that is homogeneously spread in the nucleoplasm and which is easily extracted by using non-ionic detergents [50].
972
M. GIORDArqOet al.
We tried several fixation/permeabilization techniques to evaluate PCNA/cyclin as a marker for the detection of a proliferative state of human cells of normal and leukemic origin, in suspension and utilizing bivariate FCM. Extensive experiments were conducted to identify the fixation/permeabilization procedure which fournishes: (1) specific and quantitative immunofluorescence staining patterns; (2) high resolution D N A profiles and (3) minimal cell clumping. A higher background was observed using polycional Ab both in normal and in leukemic culture and fresh cells. This could be, at least in part, due to an aspecific reaction against proteins other than PCNA. The combination of 70% ethanol, 0.5% Triton X 100 and lysolecytin permitted a satisfactory immunofluorescence analysis of nuclear PCNA content, with both polyclonal and MoAbs and the simultaneous quantitation of DNA. In all cell types we studied, the PCNA "profile" during the cell cycle was similar (with the highest PCNA signal obtained from S-phase cells but also with a continuous increase during G 1and a subsequent decrease in G2-M). Therefore, compared with the polyclonal anti-PCNA, we did not find a selective S-phase distribution for the 19F4 MoAb, but only a higher specificity resulting in a better FCM signal. On the basis of our experiments the total frequency of PCNA-positive cells (both with the polyclonal and with the MoAb) was always found to be larger than that of BUDRpositive cells and we were always able to demonstrate that the 19F4 M o A b reacts with cells from late G1 to G2-M phase of cell cycle. We conclude that the MoAb 19F4, represents a suitable reagent for the FCM evaluation not only of Sphase cells but of all the proliferating compartment of human cell populations. In this view, the detection of PCNA expression and of B U D R incorporation should be regarded as complementary approaches for studying cell proliferation in normal and leukemic cells [5153]. The application of the 19F4 MoAb and perhaps of other recently-produced specific MoAbs [54, 55] will certainly stimulate further investigation of PCNA as a possible proliferation marker with prognostic value in human leukemia. Acknowledgements--We thank the technician staff of the Department of Animal Biology at the University of Pavia, for growing the cell lines. Part of this work was done while Dr M. Giordano was a Visiting Fellow at the Gray Laboratory, Northwood, U.K., supported by an International Cancer Research TechnologyTransfer award from U. I.C.C. (International Union Against Cancer, Gen~ve). Research supported by C.N.R. (Consiglio Nazionale delle Ricerche, Roma, Progetto Finalizzato Oncologia, Grant No. 88.01.01422.44), by A.I.R.C. (Associazione Nazionale per la Ricerca sul Cancro, Milano), by I.R.C.C.S. Policlinico
San Matteo (Pavia) and by M.U.R.S.T. (Ministero Universit/~ e Ricerca Scientifica, Roma) "40%" and "60%" funding.
REFERENCES 1. Kaczmarek L. (1986) Protooncogene expression during the cell cycle. Lab. Invest. $4, 365. 2. Pardee A. B. (1986) Molecules involved in proliferation of normal and cancer cells. Ann. Ass. Cancer Res. 27, 433. 3. Danova M., Riccardi A., Giordano M., Girino M., Mazzini G., Dezza L. & Ascari E. (1988) Cell cycle-related proteins: a flow cytofluorimetricstudy in human tumors. Biol. Cell 64, 23. 4. Giordano M., Danova M., Riccardi A. & Mazzini G. (1989) Simultaneous detection of cellular ras p21 oncogene product and DNA content by two parameter flow cytometry. Anticancer Res. 9, 799. 5. Danova M., Giordano M., Mazzini G. & Riccardi A. (1990) Expression of p53 oncoprotein during the cell cycle in human leukemia: a flow cytofluorimetric study. Leukemia Res. 14, 417. 6. Stewart C. C. (1988) Flow cytometric analysis of oncogene expression in human neoplasias. Arch. Pathol. Lab. Med. 113, 634. 7. Danova M., Riccardi A. & Mazzini G. (1990) Cell cyclerelated proteins and flow cytometry. Haematologica 75, 252. 8. Miyachi K., Fritzler M. J. & Tan E. M. (1978) Antibody to a nuclear antigen in proliferating cells. J. lmmun. 121, 2228. 9. Takasaki Y., Fishwild D. & Tan E. M. (1984) Characterization of proliferating cell nuclear antigen recognized by autoantibody in lupus sera J. expl Med. 159, 981. 10. Ogata K., Ogata Y., Nakamura R. N. & Tan E. M. (1985) Purification and N-terminal amino acid sequence of proliferating cell nuclear antigen (PCNA)/cyclin and development of ELISA for anti-PCNA antibodies. J. Immun. 135, 2623. 11. Tan E. M., Chan K. L., Sullivan K. F. & Rubin R. L. (1988) Antinuclear antibodies (ANAs): diagnostically specific immune markers and clues toward the understanding of systemic autoimmunity. Clin. Immun. Immunopathol. 47, 121. 12. Tan C. K. (1982) Autoantibodies to nuclear antigens (ANA) their immunobiology and medicine. Adv. Immun. 33, 167. 13. Mathews M. B., Bernstein R. M., Franza B. R. & Garrels J. I. (1984) The identity of the proliferating cell nuclear antigen and cyclin. Nature 303, 374. 14. Tan C. K., Castillo A. G., So A. G. & Downey K. M. (1986) An auxiliary protein for DNA polymerase delta from fetal calf thymus. J. biol. Chem. 261, 12310. 15. Bravo R., Frank R., Blundell B. A. & MacDonaldBravo H. (1987) Cyclin/PCNA is the auxiliary protein of DNA polymerase. Nature 326, 515. 16. Prelich G., Tan C. K., Kostura M., Mathews M. B., So G. A., Downey K. M. & Stillman B. (1987) Functional identity of proliferating cell nuclear antigen and a DNA polymerase auxiliary protein. Nature 326, 517. 17. Bauer G. & Burgers P. M. J. (1985) The yeast analog of mammalian cyclin/proliferating cell nuclear antigen
PCNA and cell cycle in acute leukemia interacts with mammalian DNA polymerase delta. Proc. natn. Acad. Sci. U.S.A. 85, 7506. 18. Celis J. & Bravo R. (1984) Synthesis of the nuclear protein cyclin in growing, senescent and morphologically transformed human skin fibroblasts. FEBS Letts 165, 21. 19. Celis J. E. & Celis A. (1985) Cell cycle-dependent variation in the distribution of the nuclear protein cyclin proliferating cell nuclear antigen in cultured cells. Subdivision of S phase. Proc. natn. Acad. Sci. U.S.A. 82, 3262. 20. Almendral J. M., Huebsch D., Blundell P. A., MacDonald-Bravo H. & Bravo R. (1987) Cloning and sequence of the human nuclear protein cyclin: homology with DNA-binding proteins. Proc. Am. Ass. Sci. 84, 1575. 21. Jaskulski D., Gatti G., Travali S., Calabretta B. & Baserga R. (1988) Regulation of the proliferating cell nuclear antigen cyclin and thymidine kinase in RNA levels by growth factors. J. biol. Chem. 263, 10175. 22. Huff J. P., Roos G., Peebles C. L., Houghten R., Sullivan K. F. & Tan E. M. (1989) Insight into native epitopes of proliferating cell nuclear antigen using recombinant DNA protein products. J. expl Med. 172, 419. 23. Travali S., Ku D. H., Rizzo M. G., Ottavio L., Baserga R. & Calabretta B. (1989) Structure of the human gene for the proliferating cell nuclear antigen. J. biol. Chem. 264, 7466. 24. Rizzo M. G., Ottavio L., Travali S., Chang C., Kaminska B. & Baserga R. (1990) The promoter of the human proliferating cell nuclear antigen (PCN A) is bidirectional. Expl Cell Res. 188, 286. 25. Jaskulski D., DeRiel J. K., Mercer W. E., Calabretta B. & Baserga R. (1988) Inhibition of cellular proliferation by antisense oligodeoxynucleotides to PCN A. Science 240, 1544. 26. Zuber M., Tan E. M. & Ryoji M. (1989) Involvement of proliferating cell nuclear antigen (cyclin) in DNA replication in living cells. Mol. Cell Biol. 9, 57. 27. Takasaki Y., Robinson W. A. &Tan E. M. (1984) Proliferating cell nuclear antigen in blast crisis cells of patients with chronic myeloid leukemia. J.N.C.I. 73, 655. 28. Smetana K., Gyorkey F., Chan P. K., Tan M. & Busch H. (1983) Proliferating cell nuclear antigen (PCNA) and human malignant tumor nucleolar antigens (HTNA) in nucleoli of human hematological malignancies. Blur46, 133. 29. Miygawa S., Okada N., Takasaki Y., Kitano Y., Yoshikawa K., Sakamoto K. & Steinberg M. L. (1989) Expression of proliferating cell nuclear antigen/cyclin in human keratinocytes. J. Invest. Dermatol. 93, 678. 30. Ogata K., Kurki P., Celis J. E., Nakamura R. M. & Tan E. M. (1987) Monoclonal antibodies to a nuclear protein (PCNA/cyclin) associated with DNA replication. Expl Cell Res. 168, 475. 31. Kurki K., Ogata K. & Tan E. M. (1988) Monoclonal antibodies to proliferating cell nuclear antigen (PCNA/ cyclin) as probes for proliferating cells by immunofluorescence microscopy and flow cytometry. J. Immun. Meth. 109, 49. 32. Pellicciari C., Mazzini G., Fuhrman Conti A. M., De Grada L. & Manfredi Romanini M. G. (1989) Effects of hypertonic medium on human cell growth: III. Changes in cell kinetics of EUE cells. Cell Biol. Int. Rep. 13,345355.
973
33. Kurki P., Lotz M., Ogata K. & Tan E. M. (1987) Proliferating cell nuclear antigen (PCNA/cyclin) in activated human T lymphocytes. J. Immun. 138, 4114. 34. Baserga R. & Szurmak E. (1987) Oncogenes, cell cycle genes and the control of cell proliferation. Biotechnology 5, 355. 35. Prelich G. & Stilman B. C. (1988) Coordinate leading and lagging strand synthesis during SV40 DNA replication in vitro requires PCNA. Cell 53, 117. 36. Fairman M. P. (1988) DNA polymerase delta/PCNA: actions and interactions. J. Cell Sci 94, 1. 37. Celis J. & Madson P. (1986) Increased nuclear cyclin/ PCNA antigen staining of non-S phase transformed human amnion cells engaged in nucleotide excision DNA repair. FEBS Letts 209, 277. 38. Toschi L. & Bravo R. (1988) Changes in cyclin/PCNA distribution during DNA repair synthesis. J. Cell Biol. 107, 1623. 39. Suzuka I., Daidoji H., Matsuoka M., Kadowaki K., Takasaki Y., Nakane P. K. & Moriuchi T. (1989) Gene for proliferating cell nuclear antigen (DNA polymerase delta auxilliary protein) is present in both mammalian and higher plant genomas. Proc. natn. Acad. Sci. U.S.A. 86, 3189. 40. Celis J. E., Bravo R., Larsen P. M. & Fey F. T. (1984) Cyclin: a nuclear protein whose level correlates directly with the proliferative state of normal as well transformed cells. Leukemia Res. 8, 143. 41. Ottavio L., Chang C. D., Rizzo M. G., Travali S., Casadevall C. & Baserga R. (1990) Importance ofintrons in the growth regulation of mRNA levels of the proliferating cell nuclear antigen gene. Mol. Cell Biol. 10, 3(13. 42. Bravo R. & MacDonald-Bravo H. (1984) Existence of two populations of cyclin/proliferating cell nuclear antigen during the cell cycle: association with DNA replication sites. E M B O J. 3, 3177. 43. Bravo R. & MacDonald-Bravo H. (1985) Induction of the nuclear protein cyclin in quiescent mouse 3T3 cells stimulated by serum and growth factors. Correlation with DNA synthesis. E M B O J. 4, 655. 44. Chang C. D.. Ottavio g., Travali S., kipson K. E. & Baserga R. (1990) Transcriptional and post-transcriptional regulation of the proliferating cell nuclear antigen gene. Mol. Cell Biol. 10, 3289. 45. Liu Y. C., Marracino R. g., Keng P. C., Bambara R. A.. Lord E. M., Chart W. G. & Zain S. B. (1989) Requirement for proliferating cell nuclear antigen expression during stages of the Chinese hamster ovary cell cycle. Biochemistry 28, 2967. 46. Lanberg G., Tan E. M. & Roos G. (1990) Flow cytometric multiparameter analysis of proliferating cell nuclear antigen cyclin and Ki-67 antigen: a new view of the cell cycle. Expl Cell Res. 187, 111. 47. Galand P. & DegraefC. (1989) Cyclin/PCNA immunostaining as an alternative tritiated thymidine pulse labeling for marking S phase cells in paraffin sections from animal and human tissues. Cell Tissue Kinet. 22, 383. 48. Garcia R. L., Coltrera M. D. & Gown A. M. (1989) Analysis of proliferative grade using anti-PCNA/cyclin monoclonal antibodies in fixed, embedded tissues. Comparison with flow cytometric analysis. Am. J. Pathol. 134, 733. 49. Bravo R. & Celis J. E. (1985) Changes in the nuclear distribution of cyclin (PCNA) during S phase are not triggered by post translational modifications that are
974
M. GIORDANOet al.
expected to moderately affect its charge. FEBS Letts 185, 435. 50. Bravo R. & MacDonald-Bravo H. (1987) Existence of two populations of cyclin/proliferating cell nuclear antigen during the cell cycle: association with DNA replication sites. J. Cell Biol. 105, 1549. 51. Keim D. K., Hailat N., Hodge D. & Hanash M. (1990) Proliferating cell nuclear antigen expression in childhood acute leukemia. Blood 76, 985. 52. Kurki P., Vanderlaan M., Dolbeare F., Gray J. & Tan E. M. (1986) Expression of proliferating cell nuclear
antigen (PCNA/cyclin) during the cell cycle. Expl Cell Res. 166, 209.
53. Andreef M. (1986) Cell kinetics of leukemia. Semin. Hemat. 23, 300. 54. Murashima A., Takasaki Y., Ohgaki M., Hashimoto H., Shirai T. & Hirose S. (1990) Activated peripheral blood mononuclear cells detected by murine monoclonal antibodies to proliferating cell nuclear antigen in active lupus patients. J. clin. lmmun. 10, 28. 55. Waseem N. H. & Lane D. P. (1990) Monoclonal antibody analysis of the proliferating cell nuclear antigen (PCNA) J. Cell Sci. 96, 121.