Original Article

Fluorescence Lifetime Imaging of DAPI-

Stained Nuclei As a Novel Diagnostic Tool for
the Detection and Classification of B-Cell
Chronic Lymphocytic Leukemia

Gilad Yahav,1 Abraham Hirshberg,2 Ophira Salomon,3 Ninette Amariglio,4

Luba Trakhtenbrot,4 Dror Fixler1*

 Abstract
1
Faculty of Engineering and Institute of
Nanotechnology and Advanced
Materials, Bar Ilan University, Ramat
Gan, Israel
2
Department of Oral Pathology and Oral
Medicine, Maurice and Gabriela
Goldschleger School of Dental Medicine,
Tel Aviv University, Tel Aviv, Israel
3
Thrombosis Unit, Sheba Medical Center
and Sackler Faculty of Medicine, Tel
Aviv University, Tel Aviv, Israel
4
Laboratory of Hematology, Sheba
Medical Center, Ramat Gan, Israel
Received 16 November 2015; Revised 21
April 2016; Accepted 18 May 2016
Grant sponsor: Israel Cancer Association,
Grant number: 20150012
 Key terms
Additional Supporting Information may be
found in the online version of this article.
Correspondence to: Dror Fixler; Faculty of
Engineering and Institute of Nanotech-
nology and Advanced Materials, Bar-Ilan
University, Ramat Gan 5290002, Israel.
E-mail: Dror.Fixler@biu.ac.il
L. Trakhtenbrot and D. Fixler contributed
equally to this work.
Published online 17 June 2016 in Wiley
Online Library (wileyonlinelibrary.com)
DOI: 10.1002/cyto.a.22890
C 2016 International Society for
V
Advancement of Cytometry
B-cell chronic lymphocytic leukaemia (B-CLL) and B-cell precursor acute lymphoblastic
leukaemia (B-ALL) are the most common type of leukaemia in adults and children,
respectively. Today, fluorescence in situ hybridization (FISH) is the standard for detect-
ing chromosomal aberrations that reflect adverse and favorable outcome. This study
revealed a new, simple, and fast diagnostic tool to detect pathological cells by measuring
and imaging the fluorescence lifetime (FLT) using FLT imaging microscopy (FLIM) of
the peripheral blood (PB) cells of B-CLL samples that were labeled with the DNA
binder, DAPI. The FLT of DAPI in healthy individuals was found to be 2.66 6 0.12 ns. In
contrast, PB cells of B-CLL and BM cells of B-ALL patients were characterized by a spe-
cific group distribution of the FLT values. The FLT of DAPI was divided into four sub-
groups, relative to 2.66 ns: short1, normal, prolonged, and prolonged1. These
alterations could be related to different chromatin arrangements of B-CLL and B-ALL
interphase nuclei. Notably, extremely long FLT of nuclear DAPI correlate with the pres-
ence of extra chromosome 12, while moderate increases compared to normal character-
ize the deletion of p53. Such correlations potentially enable a FLT-based rapid automatic
diagnosis and classification of B-CLL even when the frequency of genetic and chromo-
somal abnormalities is low. VC 2016 International Society for Advancement of Cytometry
B-cell chronic lymphocytic leukaemia; B-cell precursor acute lymphoblastic leukaemia;
fluorescence in situ hybridization; fluorescence lifetime; fluorescence lifetime imaging
microscopy; bone marrow
INTRODUCTION
THE nucleus is a highly complex organelle containing most of the cell’s genetic
material, DNA. The precise organization of the nucleus includes the spatial arrange-
ment of chromatin, interchromatin compartment, and specialized structures (lam-
ina, nucleoli, and nucleoskeleton), which are completely connected and act as part of
the responsive cellular system (1). A series of genetic and epigenetic events that char-
acterize cancer initiation and progression-result in structural changes of the nuclear
architecture (2–5). These chromosomal alterations lead to gene deregulation and
instability of the genome and are important clinical parameters in various types of
cancer (6).
B-cell chronic lymphocytic leukaemia (B-CLL) is the most common type of leu-
kaemia that affects adults in developed countries (7–9). B-CLL results from the progres-
sive accumulation of mature-appearing clonal B cells with a characteristic
immunophenotype characterized by a common gene expression “signature.” The

Cytometry Part A  89A: 644652, 2016


clinical presentation can vary; some patients exhibit an indolent
course while others face an aggressive disease that can lead to
death within a short period after diagnosis (10,11).
B-cell precursor acute lymphoblastic leukaemia (B-ALL)
is the most common childhood cancer (12). It is an acute rap-
idly progressing form of leukaemia that is characterized by the
presence of large numbers of lymphoblasts—unusually imma-
ture white blood cells which typically differentiate to form
mature lymphocytes in the blood and bone marrow (13). The
rate of success in the treatment of B-ALL with the use of mod-
ern treatments is about 80% for children and 40% for adults
(14). Today, therapeutic decisions in both B-CLL and B-ALL
are based on several prognostic markers, such as cytogenetic
abnormalities and various chromosomal aberrations (15,16)
that are regularly detected by fluorescence in situ hybridiza-
tion (FISH) (16,17). In general, the FISH technique suggests
high-resolution detection of specific DNA sequences in indi-
vidual cells and, hence, is considered as a common tool for
clinical purposes (18). Today, despite many automated FISH
approaches, the detection of cytogenetic abnormalities is
mainly done manually, which is time consuming (19,20). In
addition, it is expensive and limited, and in about 30% of
patients FISH cannot predict adverse prognosis (21). Thus,
additional methods are required to improve prognosis of B-
CLL patients.
Fluorescence lifetime imaging microscopy (FLIM) is an
advanced spectroscopic technique used in biology and medi-
cine (22,23). The fluorescence lifetime (FLT) is considered a
preferable measure than the classical fluorescence intensity
(FI) on which FISH is based, since FLT avoids many of the
optical artifacts of FI (22,24). Furthermore, FLT provides a
quantitative means of probing changes in the local fluoro-
phore’s environment (22,24,25) and, therefore, FLIM can be
used for investigating the intracellular medium. In addition,
FLT measurement extracts a quantitative parameter that is not
subject to users interpretations like a FISH analysis. All of
these qualities lead many researchers to explore the numerous
potential applications of FLTs especially in biology (22,25).
Over the past decade, many studies explored mechanisms
that can change the FLT for in vivo and ex vivo diagnoses,
including cancer situations (25), mainly by using FLT from
endogenous fluorophores and cells stained with exogenous
probes. These studies include the detection of skin tumors
(26) and tracking apoptosis and stimulation in individual cells
(27). NAD(P)H auto fluorescence has been used to differenti-
ate cancer from noncancerous tissue (25). Very recent works
suggest using FLT to evaluate the efficiency of anticancer
drugs (used for chemotherapy) (28) and to differentiate
between metastatic cells from inflammatory cells in the cere-
brospinal fluid of patients with medulloblastoma (29). Based
on these works and additional previous studies (22,25,30,31),
we suggest that FLT, as a differential indicator of cellular activ-
ity, of DAPI bound to the DNA of lymphocytic cells nuclei
can be correlated with chromosomal alterations of B-CLL
cells.
The aim of the current study was to use FLIM to detect
and monitor the chromosomal abnormalities in B-CLL and
Cytometry Part A  89A: 644652, 2016
Original Article
B-ALL patients. This method may lead to a high throughput
diagnostic tool that can serve as a replacement or complement
to the traditional FISH diagnosis.
METHODS
Slide Preparation
Slide preparation and FISH detections were performed
according to the standard protocol developed in our labora-
tory (32). Direct preparations without any culturing of
peripheral blood (PB) cells from 19 adult B-CLL samples, two
children B-ALL samples and five samples of control subjects
who were not diagnosed with any cancer were used. All the
samples were harvested and analyzed as part of a routine clini-
cal evaluation of the patients.
The study protocol was approved by the Institutional
Review Board for Clinical Studies at the Sheba Medical Cen-
ter, in accordance with the Declaration of Helsinki.
All samples were diluted with red blood cell lysis buffer
(RBCLB) in a 15 mL tube, centrifuged at 1,400 rpm for 10
min, washed with phosphate buffer solution (PBS) (Biological
industries, Kibbutz Beit Haemek 25115, Israel) for 5 min and
centrifuged for 10 min at 1,400 rpm. Fixation with methanol
(Merck, Germany)/acetic acid (Frutarom, Israel) (3:1) was
repeated three times, cells were dropped on a slide, and air-
dried for several hours at room temperature or were stored at
2208C for later use. Slides were stained with 20 mm of DAPI
(4,6-diamidino 2-phenyl-indole) in antifade solution (Q-Bio-
gene, Cambridge, UK) and covered with a coverslip. Slides
were analyzed using an Olympus BH2 fluorescence light
microscope equipped with a Plan Apo objective 1003/1.4 oil,
an appropriate spectral filter (BH2-TFC1 Triple Band filter
DAPI/FITC/TRITC), and a 100 W mercury arc lamp. Two spe-
cifically defined FISH were performed using commercially
available probes to detect specific chromosomal abnormalities
that have predictive value and outcome for B-CLL namely:
Alpha Satellite chromosome 12 probe of Vysis (Vysis Down-
ers, Grove, IL) and P53 (TP53) (17p13.1) deletion probe of
Cytocell (Cytocell, Cambridge, UK). From this point, we
name these aberrations as FISH markers 112 and p53-,
respectively. In the present study, FLT measurements analyzed
one-karyotype aberration that has prognostic relevance only
in B-CLL: trisomy of chromosome 12 and one-karyotype
aberration that has prognostic relevance in both B-CLL and
B-ALL: 17p13/p53 deletion. The measurements were per-
formed on 19 B-CLL samples: four samples with one karyo-
type aberration of 112 alone; nine samples with simultaneous
two karyotype aberrations of 112 and p53; and six samples
with one karyotype aberration of p53-. These samples were
compared to five non-oncologic patients serving as controls
and to two B-ALL samples since they are characterized by the
same deletion of p53 gene.
It is noted that when we define a patient with markers
112 and p53- the meaning is that these chromosomal altera-
tions varied from 8 to 95% as analyzed via FISH and is pre-
sented Table S1 in Supporting Information.
645
Original Article
Figure 1. (A) An image of a cell from a non-oncologic patient by regular fluorescence microscopy of the FI (0–65355 A.D.U). (B) An image
of the same cell by the FLIM system with a FLT of 2.66 ns (a pixel with a 2.6 ns is marked with an arrow as for an example). [Color figure
can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
FLIM Measurements
The FLIM experiments were performed using an OLYM-
PUS IX-81 inverted microscope with a 103, NA 5 0.4 objec-
tive (Olympus, Tokyo, Japan) and the images were taken with
the Lambert Instruments FLIM system (LI-FLIM; Groningen,
Netherlands). Our system consists of three different modu-
lated LED excitation sources (403, 468, and 537 nm), a
LI2CAM MD intensified charge coupled device (CCD) camera
with a modulated image intensifier for detection of the fluo-
rescence images, a modulation signal generator/power supply
unit (Prior OptiScan I, Rockland, MA), a computer and the
LI-FLIM software package for complete system control, image
acquisition, and FLT calculation. All references were made
daily. Each measurement was repeated at least twice. The FLTs
of DAPI that were not from cells (from the background)
appeared to be 2 6 0.12 ns as known for free DAPI (33), and
the FLT of DAPI in normal cells is known to be 2.8 ns (34,35).
In our experiments, the excitation source was set between
three to eight repetition frequencies in the range of 1–120
MHz for all measurements and hence the modulation and
phase angle were measured across three to eight variable fre-
quencies from 1 to 120 MHz in every measurement. The
number of frequencies was chosen according to the minimum
chi-square (v2) and all FLT data was fit to a mono exponential
decay rate in a same manner. As the light modulation fre-
quency increases, the phase angle increases and the modula-
tion of the emission decreases. Figure 1 presents FI and FLT
imaging of a cell diagnosed from non-oncologic patient by a
regular microscopy system of FI (Fig. 1A) and by the FLIM
system (Fig. 1B). In the present study, the FLT value found in
normal cells was 2.66 6 0.12 ns.
Polar Plots
The polar plots were made by LI-FLIM software. The basic
theory regarding the polar plots has been previously discussed
(36). In our FD-FLIM system, the FLT is calculated from every
pixel of the sample. As a single pixel data is very noisy, it is vital
to take as large a region of interest as possible, which will ensure
646
that the average FLT of all the pixels combined will give the cor-
rect FLT. In our system, the polar plot is divided to 200 3 200
bins, and all pixels of the FLT image correspond to an appropri-
ate polar coordinate so that bins with many pixels are purple
and bins with few pixels are blue. Therefore, it is of our interest
to address the polar coordinates of the purple circle, based on
the previously defined normal average FLT. The blue points
indicate the polar coordinates where only a few pixels were
measured, and as mentioned this data is very noisy and there-
fore should not be addressed. It is valuable to mention that we
compared the results from the polar plot with those from the
multi-frequency analysis and received the same results. The
experimental polar plots of phase and modulation data were
measured at a single frequency of 60 MHz.
Group Classification
It is well known from the literature that the FLT of DAPI fol-
lows a heterogeneous distribution (34,35). In our study the FLT
measurements of DAPI bound to DNA in cells from B-CLL sam-
ples were divided to four subgroups: short1, normal, prolonged,
and prolonged1. The normal group correlated with the FLT of
DAPI from the controls: s 5 2.66 6 0.12 ns (range from 2.3 to
2.92 ns). The additional three abnormal groups: short1, pro-
longed, and prolonged1 were classified according to their relative
proximity to the normal group as s  norm (s < 1.8 ns),
s > norm (3 ns < s < 4.5 ns), and s  norm (s > 4.5 ns), respec-
tively. FLIM measurements revealed that the values between 1.8
ns < s < 2.2 ns corresponded to free DAPI (unbounded to DNA)
and therefore were not part of the analysis (2 6 0.12 ns).
RESULTS
Characterization of the FLTs Distribution by Different
Cytogenetic Alterations
In the first step, we investigated the FLT distribution in three
B-CLL samples classified according to their cytogenetic alteration
identified by FISH analyses: a sample with 112 aberration, a
sample with simultaneous 112 and p53- and a sample with p53-
FLIM for Identification of B-CLL
 Original Article

Figure 2. An example FISH analysis using Alpha Satellite chromosome 12 probe–hybridisation pattern. Two red signals denote normal
chromosome 12 cell (A, top) and three red signals denote trisomy of chromosome 12 (A, bottom). FISH analysis for four cells using the
p53 deletion probe; consist of TP53 probe located in 17p13 labeled in red and control probe for the centromere of chromosome 17 labeled
in green. A pattern of two red and two green signals denotes normal p53 cell (B, top) and a pattern of one red and two green signals
denotes deletion of p53 (B, bottom). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

. These three samples were compared to one healthy control sam- distribution in the range of 2.66 6 0.12 ns were classified as
ple. Figure 2A (top) presents an example of a control cell’s FISH normal (indicated as s 5 norm). An example polar plot of a
imaging using Alpha Satellite chromosome 12 probe. Two red region of one nucleus is presented in Fig. 4A. The polar coor-
signals signify normal chromosome 12. In a same manner, Fig. dinates (purple circle) reveal a single FLT component of 2.8 ns
2B (top) presents an example of two control cells’ FISH imaging that is present. Figure 5A presents a FLT analysis for all meas-
using p53 deletion probe that also detected a normal pattern. As ured cells (indicated by black points) along with the cells clas-
expected, all cells in the control sample revealed normal patterns sification, the average and the standard deviation (SD) of each
of chromosome 12 and P53 (the 11q- and 13q- were also checked group. Full data information for each control patient is pre-
and found normal but their analyses are beyond the scope of this sented in Table S1 in Supporting Information.
article). Figure 2A (bottom) presents a FISH image for a cell with
112 and simultaneous 112 and p53-, three red signals indicate Characterization of FLTs for B-CLL Samples With
the trisomy of chromosome 12 (in contrast to normal chromo- Chromosome 12 Trisomy (112)
somes with two red signals). Figure 2B presents a FISH image for The FLT of DAPI was measured in four B-CLL samples
p53- identification in cells and simultaneous 112 and p53-, one with variable concentrations of the chromosomal abnormal-
red and two green signals indicate the chromosomal abnormality ity 112. FISH analysis detected that three samples exhibited
of p53-. FLT results of the B-CLL sample groups are shown in
Fig. 3. While the median appeared identical in all of the samples,
each group displays a different FLT distribution. The 112 sample
spans to higher FLT values, the p53- spans to shorter FLT values,
and the sample with simultaneous 112 and p53- is located
approximately between the other two distributions. However, a
quantitative method of differentiating between the samples is not
evident from these results. Thus, in the next steps we checked the
FLT distribution for each cytogenetic mutation group.

Characterization of FLTs for the Control Samples

The control group consisted of 87 PB cells selected from
five non-oncologic samples. As expected, all 87 PB cells were
checked by FISH analyses and found normal. The FLT of
DAPI was consistently measured at 2.66 6 0.12 ns in all meas- Figure 3. Distribution of the FLT measurements from control
sample and three B-CLL samples categorized according to FISH
urements which agrees with the known decay time of DAPI analyses: 112 mutation alone, 112 together with p53-mutations
bound to DNA (s ffi 2.8 ns) (37). Accordingly, cells with FLT and p53- mutation alone.

Cytometry Part A  89A: 644652, 2016 647

Original Article

Figure 4. An experimental polar plot of phase and modulation data measured at 60 MHz. An example of a healthy cell nucleus is shown,
with a single component FLT that corresponds to the normal group (2.8ns) (A). In the B-CLL sample, one or more FLT components were
obtained. An example of two components of 2.5 and 4.8 ns that were taken from a region containing two nucleuses of cells from a 112
sample is shown (B). These results were additionally obtained by multi-frequency analysis (as presented in Methods section). Bins with
many pixels are purple and bins with few pixels are blue. [Color figure can be viewed in the online issue, which is available at wileyonline-
library.com.]

a high proportion of 112 (90–97% of the cells), while one
case was detected in only 10% of the cells. In a series of 217
cells measurements via FLIM in the four samples, the FLT of
DAPI presented a large heterogeneous distribution. Based on
the previous section’s findings, cells with a FLT distribution
within the range of 2.66 6 0.12 ns (between 2.3 and 2.92 ns)
were classified as normal, correlated with the FLT of DAPI
from the controls. Relative to this normal value, cells with a
FLT distribution a little higher than the normal group were
classified as the prolonged group and cells with a FLT distri-
bution a lot higher or a lot shorter than the normal group
were classified as the prolonged1 and short1 groups, respec-
tively. Each sample can be characterized by these four sub-
groups. Following the analyses of FLT values, we find that
relative to the control samples, which presents only one
group (Fig. S1A in Supporting Information), the four sam-
ples with 112 mutation correlated with the four subgroups.
This correlation was as follows: 27% were classified as
prolonged1, 27% normal, 26% prolonged, and 20% as
short1 (Fig. S1B in Supporting Information). Figure 4B
presents an example of polar plot of region contains cluster
of two cells nucleuses that are classified into both the normal
and the prolonged1. In this region, two FLT components
were found: 2.5 and 4.8 ns. Figure 5B presents the entire FLT
distribution along with the mean FLT and the SD for each
sample. All samples revealed the short1, the normal, and the
prolonged1 subgroups while the prolonged group appeared
in most of the samples. Full data description for the four
samples is presented in Table S1 in Supporting Information.
Characterization of FLTs for B-CLL Samples With
112 Together With p53-
Next, the FLT of DAPI was examined in nine B-CLL sam-
ples with simultaneous two karyotype aberrations of 112 and
p53-. FISH analysis detected that seven samples exhibited a
high proportion of both aberrations (80–95% of the cells)
while in two cases, it was detected in only 8–52% of cells. Fig-
ure 5C presents the full 258 cells’ FLT data along with the
average and SD for each group of each sample with 112 and
p53-. In a similar fashion as in 112 alone (Fig. 5B), samples
with the two chromosomal abnormalities (Fig. 5C) present
the same correlation as 112 samples with the same four sub-
group distribution, but with different percentages: 38% were
classified as short1, 31% normal, 20% prolonged1, and 11%
prolonged (Fig. S1C in Supporting Information). All samples
revealed the short1, the normal, and the prolonged1 groups
while the prolonged group appeared in most of the samples.
Full data description for the nine samples is presented in Table
S1 in Supporting Information.
Characterization of FLTs for B-CLL Samples With
p53- Alone
The next step in B-CLL samples was to examine the FLT
of DAPI in six samples with only the p53-. These six samples
exhibited variable concentrations of p53- between 21.5 and
90% by FISH analyses. This time DAPI measurements of 237
cells revealed smaller heterogeneity with high correlation to
only three subgroups: short1 with 33%, normal with 52%,
and prolonged with 15% (Fig. S1D in Supporting Informa-
tion). Figure 5D presents the mean FLT and SD for each sam-
ple in this group. All samples revealed the short1 and the
normal groups while the prolonged group appeared in most
of the samples. Full data description for the six samples is pre-
sented in Table S1 in Supporting Information.
Characterization of FLTs for B-ALL Samples With
p53- Alone
The final step was to measure two B-ALL samples with
only p53- in order to examine whether the p53- heterogeneity
is repeated in another disease. These two samples exhibited
high concentration of p53- (90 and 95%). DAPI measure-
ments of 74 cells revealed the same group heterogeneity as in
B-CLL patients with p53- alone only this time the difference

648 FLIM for Identification of B-CLL

 Original Article

Figure 5. FLT analysis vs. patient numbers. For each patient heterogenous distribution is presented. Each point represents a FLT value for
single cell along with the average and SD. Due to the fact that many FLT values were alike, we can only observe the FLT distribution and
not the total numbers of cells. For the normal group the total mean and SD for all the five patients is 2.66 6 0.12 ns (A). For B-CLL patients
with 112 abnormality, the total mean and SD for the short, normal, prolonged and prolonged1 groups are 1.23 6 0.2 ns, 2.63 6 0.18 ns,
3.31 6 0.24 ns, and 5.22 6 0.44 ns, respectively (B). For B-CLL patients with 112 together with p53- abnormalities, the total mean and SD
for the short, normal, prolonged, and prolonged1 groups are 1.2 6 0.19 ns, 2.64 6 0.15 ns, 3.44 6 0.3 ns, and 4.93 6 0.35 ns, respectively
(C). For B-CLL patients with p53- abnormality, the total mean and SD for the short, normal, prolonged, and prolonged1 groups are
1.21 6 0.27 ns, 2.6 6 0.13 ns, and 3.66 6 0.27 ns, respectively (D). For B-ALL patients with p53- abnormality, the total mean and SD for the
short, normal, and prolonged groups are 1.17 6 0.2 ns, 2.46 6 0.11 ns, and 3.49 6 0.22 ns, respectively (E). [Color figure can be viewed in
the online issue, which is available at wileyonlinelibrary.com.]

between groups was very observable: 69% of cells were classi- porting Information). Figure 5E presents the mean FLT and
fied as short1, while prolonged FLT was found only in 4% SD for the two patients. The short1 and the normal group
and the normal group appeared only in 27% (Fig. S1E in Sup- were very dominant in the two samples and the prolonged

Cytometry Part A  89A: 644652, 2016 649

Original Article
Table 1. A summary of the four types of patients (controls, B-CLL with 112 chromosomal aberration, B-CLL with 112 together with p53-
aberrations and B-CLL with p53- aberration. The total number of patients in each group is given in the headline brackets. The FLT distribu-
tion into the four subgroups is shown (short1, normal, prolonged, and prolonged1). In each group the mean FLT, the appropriate SD and
the total number of cells (inside brackets) are presented
CONTROLS
GROUP [5 PATIENTS] 112 [FOUR PATIENTS]
Short1 1.23 6 0.2 ns
[45 cells]
Normal 2.66 6 0.12 ns 2.63 6 0.18 ns
[87 cells] [56 cells]
Prolonged 3.31 6 0.24 ns
[56 cells]
Prolonged1 5.22 6 0.44 ns
[60 cells]
group barely appeared. Full data description for the six sam-
ples is presented in Table S1 in Supporting Information.
DISCUSSION
To the best of our knowledge, the current research is the
first attempt to apply FLT analysis in order to identify B-CLL
and B-ALL pathological cells. While FLT measurements of PB
cells from the 5 non-oncologic samples revealed homogeneity
of one FLT component with a mean of 2.66 6 0.12 ns (Table
S1 in Supporting Information) (34), high level of heterogene-
ity of FLT measurements were found in all of the B-CLL and
B-ALL samples. In addition to the normal FLT value, we dis-
tinguished three different FLT values: short1, prolonged and
prolonged1. This clearly indicates that DAPI is bounded in
the B-CLL and B-ALL samples to more than one environ-
ment. This finding agrees with previous studies (22). Table 1
presents the summary of each of the B-CLL and B-ALL
patients, the average distribution, the number of patients and
the total number of cells that were measured.
We suggest that trisomy 121 might be identified by close
examination of the heterogeneity of the FLT measurements.
Validation for this claim is given as all samples with the abnor-
mality of 112 (alone and together with p53-) show the
appearance of the prolonged1 group (s > 4.5 ns). In contrast,
most of the abnormal cells diagnosed with p53- whether in B-
CLL or B-ALL patients were classified in the short1 range
(Figs. S1D and S1E in Supporting Information).
The shape of the FLT distribution can help in identifying
problematic cases with imaging and cryptologic techniques
that are offered today, for instance, in cases with low percen-
tages of chromosomal abnormalities (below 30%). FLT meas-
urements showed that even a relatively low number of cells
were sufficient for detecting abnormal FLT values, suggesting
that even a small number of measurements are still sufficient
to detect pathological cells with one or two chromosomal
abnormalities (e.g., patient #6 in Table S1 in Supporting
Information).
Unlike the classical FI measurements which are more
commonly considered by cytometers for probing cell behavior
650
B-CLL
112 AND P53- P53- B-ALL P53-
[FOUR PATIENTS] [SIX PATIENTS] [TWO PATIENTS]
1.2 6 0.19 ns 1.21 6 0.27 ns 1.17 6 0.2 ns
[92 cells] [75 cells] [51 cells]
2.64 6 0.15 ns 2.6 6 0.13 ns 2.46 6 0.11 ns
[80 cells] [126 cells] [20 cells]
3.44 6 0.3 ns 3.66 6 0.27 ns 3.49 6 0.22 ns
[31 cells] [36 cells] [3 cells]
4.93 6 0.35 ns
[55 cells]
(25), the FLT is sensitive to the intrinsic biological environ-
ment and not affected by many of the adverse optical effects
of FI measurements (24). This makes the FLT more suitable
whenever exploring biological phenomena that associate with
changes in the intracellular microenvironment (27). Previous
research revealed the association between the FLT and progres-
sions of the PB mononuclear cell response to antigens, mito-
gens, and phorobol esters, as well as to the contraction cycle
of cardiac cells (38–40). This implies that the different DAPI
FLT groups that appeared in all of the B-CLL samples (relative
to the normal group) may be caused by changes in the mobil-
ity of the fluorescent molecules due to the viscosity alteration
of the cell medium (the fluorophores’ hosting area). We sug-
gest that the viscosity alteration might be related to different
chromatin organization of B-CLL interphase nuclei. In nor-
mal cells, the nuclear chromatin is organized in the form of
interphase chromosome territories (CTs) which are not ran-
domly positioned; they inhabit specific locations and link to
certain regions of the cell nucleus by the nuclear structures. In
addition, a non-random radial nuclear distribution was found
not only for CTs but also for many individual genes (reviewed
in Refs. 4 and (41–43). A dynamic model of the spatial rela-
tionships between chromatin and functional nuclear machi-
neries was recently suggested (44). In contrast, in cancer
nuclei, CTs and chromosomal loci differ in both radial nuclear
positioning and large-scale chromatin organization parame-
ters of nuclear architecture (2–5). It is also known that the
nuclear chromatin environment is complex and contains
many proteins, RNA and other components that have the
potential to influence chromatin structure and compaction.
The complex genomic and epigenomic changes in carcinogen-
esis result in nanoscale structural alterations arising from the
changes in the 3D spatial arrangement and the chromatin
density variation in the cell nucleus. The chromatin confor-
mation there can be modulated by divalent cations (45).
Moreover, it was supposed that the molecular mechanisms
that control nuclear organization might be considered as tar-
gets for future anti-cancer agents (46). This is validated by the
article of Ana Katrina Estandarte (2015) that demonstrates
FLIM for Identification of B-CLL

that DAPI FLT variations across interphase nuclei reflect dif-
ferentially compacted chromatin of CTs within the nucleus
(47). However, the situation of the chromatin itself does not
characterize any risk factor and therefore clinically it is not
measured routinely. The limitations of this study are the small
number of patients included such as only two samples of the
p53- B-ALL samples. In addition, our work did not overcome
one of the main concerns regarding cancer diagnostics, which
is the ability to detect cancer cells even in a low concentrations
(48). Our method is suitable for cases where the concentration
of the cancer cells is not negligible and the difference between
their FLTs reflects in the populations. However, even with
these limitations our results were statistically distinguished
between the different cytogenetic mutations (Fig. S1 in Sup-
porting Information) that were investigated and were found
in high correlation with the FISH analyses. Recent FLT based
cytometry work include the work presented by the Jenkins
group, which offers methods to present how the FLT can be
used as a fast sorting parameter for cytometry applications
such as to discriminate between ethidium bromide and propi-
dium iodide that spectrally overlap in cells stained with both
dyes (49,50). In our work, we can use the same platforms in a
similar manner and create a fast sorting technique to discrimi-
nate between the four FLT values and detecting the appropri-
ate cytogenetic mutation.
Today, it is concerning that there still is no automatic
imaging system that could replace the intensive work of the
special experts. Adding FLIM to the arsenal of detection as a
complement or replacement method could assist in early
detection, using a faster, easier, and more reliable detection
method. This article serves as an initial and basic principle
that shows the direction with which we wish to continue. The
next vital step will be to investigate our findings on a larger
cohort of patients. Therefore, additional investigations on
other B-CLL and B-ALL chromosomal abnormalities and
other types of leukemia are recommended.
ACKNOWLEDGMENTS
This publication is partly supported by Israel Cancer
Association (diffusion reflection: a novel nanophotonic
method for early detection of oral cancer) and with the gener-
ous assistance of the Irma and Jacques Ber-Lehmsdorf
Foundation. The authors thank Ms. Galina Yshoev for her
excellent technical assistance.
LITERATURE CITED
1. Starr DA. A nuclear-envelope bridge positions nuclei and moves chromosomes.
J Cell Sci 2009;122:577–586.
2. Grossmann C, Schwarz-Finsterle J, Schmitt E, Birk UJ, Hildenbrand G, Cremer C,
Trakhtenbrot L, Hausmann M. Variations of the spatial fluorescence distribution in
ABL gene chromatin domains measured in blood cell nuclei by SMI microscopy after
COMBO–FISH labelling. Microsc Sci Technol Appl Edu 2010; 688–695.
3. Zink D, Fischer AH, Nickerson JA. Nuclear structure in cancer cells. Nat Rev Cancer
2004;4:677–687.
4. K€olbl AC, Weigl D, Mulaw M, Thormeyer T, Bohlander SK, Cremer T, Dietzel S.
The radial nuclear positioning of genes correlates with features of megabase-sized
chromatin domains. Chromosome Res 2012;20:735–752.
5. Schwarz-Finsterle J, Stein S, Großmann C, Schmitt E, Trakhtenbrot L, Rechavi G,
Amariglio N, Cremer C, Hausmann M. Comparison of triple helical COMBO-FISH
and standard FISH by means of quantitative microscopic image analysis of abl/bcr
positions in cell nuclei. J Biochem Biophys Methods 2007;70:397–406.
Cytometry Part A  89A: 644652, 2016
Original Article
6. Stratton MR, Campbell PJ, Futreal PA. The cancer genome. Nature 2009;458:719–
724.
7. Hallek M, GCS Group. Chronic lymphocytic leukemia (CLL): First-line treatment.
New York, USA: American Society of Hematology. In ASH Education Program
Book, Vol. 1. 2005. pp 285–291.
8. Puente XS, Pinyol M, Quesada V, Conde L et al. Whole-genome sequencing identi-
fies recurrent mutations in chronic lymphocytic leukaemia. Nature 2011;475:101–
105.
9. Rodrıguez D, Bretones G , Arango J R., Valdespino V, Campo E. Molecular pathoge-
nesis of CLL and its evolution. Int J Hematol 2015;101:219–228.
10. Marti, G, Faguet GB, Stewart C, et al. Evolution of leukemic heterogeneity of human
B-CLL lymphocytes between and within patients. In Mechanisms in B-Cell Neopla-
sia, Berlin Heidelberg: Springer. 1992. pp 303–311.
11. Guarini A, Gaidano G, Mauro FR, Capello D, Chronic lymphocytic leukemia
patients with highly stable and indolent disease show distinctive phenotypic and
genotypic features. Blood 2003;102:1035–1041.
12. Shah, NN and Wayne AS. Leukemias and lymphomas. In Pediatric Psycho-Oncol-
ogy: A Quick Reference on the Psychosocial Dimensions of Cancer Symptom Man-
agement, Amsterdam. Netherlands: Elsevier B.V. 2015.
13. Pui CH, Robison LL, Look AT. Acute lymphoblastic leukaemia. Lancet 2008;371:
1030–1043.
14. Helms RA, Quan DJ. Textbook of Therapeutics: Drug and Disease Management, 8th
edition, Philadelphia: Lippincott Williams & Wilkins: 2006.
15. Moorman AV, Ensor HM, Richards SM, Chilton L, Schwab C, Kinsey SE, Vora A,
Mitchell CD, Harrison CJ. Prognostic effect of chromosomal abnormalities in child-
hood B-cell precursor acute lymphoblastic leukaemia: Results from the UK Medical
Research Council ALL97/99 randomised trial. Lancet Oncol 2010;11:429–438.
16. Dewald GW, Brockman SR, Paternoster SF, Bone ND, et al. Chromosome anomalies
detected by interphase fluorescence in situ hybridization: Correlation with significant
biological features of B-cell chronic lymphocytic leukaemia. Br J Haematol 2003;121:
287–295.
17. Chiorazzi N. Implications of new prognostic markers in chronic lymphocytic leuke-
mia. In ASH Education Program Book, 2012. pp 76–87.
18. Bartlett JM. Fluorescence in situ hybridization. In Molecular Diagnosis of Cancer,
Springer: 2004. pp 77–87.
19. Wang X, Zhang RR, Li S, Chen X, Mulvihill JJ, Lu X, Panf H, Liu H. Automated anal-
ysis of fluorescent in situ hybridization (FISH) labeled genetic biomarkers in assist-
ing cervical cancer diagnosis. Technol Cancer Res Treatment 2010;9:231–242.
20. Wolff DJ, Bagg A, Cooley LD, Dewald GW, Hirsch BA, Jacky PB, Rao KW, Rao PN,
and the Association for Molecular Pathology Clinical Practice Committee and the
American College of Medical Genetics Laboratory Quality Assurance Committee.
Guidance for fluorescence in situ hybridization testing in hematologic disorders.
J Mol Diagn 2007;9:134–143.
21. Rigolin GM, del Giudice I, Formigaro L, Saccenti E, et al. Chromosome aberrations
detected by conventional karyotyping using novel mitogens in chronic lymphocytic
leukemia: Clinical and biologic correlations. Genes Chromosomes Cancer 2015;54:
818–826.
22. Lakowicz JR. Principles of Fluorescence Spectroscopy, Springer: 2007.
23. Engelborghs Y, Visser AJ, Fluorescence Spectroscopy and Microscopy: Methods and
Protocols, Humana Press: 2014.
24. Chang C, Sud D, Mycek M. Fluorescence lifetime imaging microscopy. Methods Cell
Biol 2007;81:495
25. Berezin MY, Achilefu S. Fluorescence lifetime measurements and biological imaging.
Chem Rev 2010;110:2641–2684.
26. Cubeddu R, Pifferi A, Taroni P, Torricelli A, Valentini G, Sorbellini E. Fluorescence
lifetime imaging: An application to the detection of skin tumors. IEEE J Sel Top
Quantum Electron 1999;5:923–929.
27. Fixler D, Tirosh R, Zurgil N, Deutsch M. Tracing apoptosis and stimulation in indi-
vidual cells by fluorescence intensity and anisotropy decay. J Biomed Opt 2005;10:
034007–0340078.
28. Kawanabe S, Araki Y, Uchimura T, Imasaka T. Applying fluorescence lifetime imag-
ing microscopy to evaluate the efficacy of anticancer drugs. Methods Appl Fluores-
cence 2015;3:025006
29. Yahav G, et al. A novel method for sensing metastatic cells in the CSF of pediatric
population with medulloblastoma by frequency domain FLIM system. In SPIE BiOS,
International Society for Optics and Photonics: 2016.
30. Kobayashi H, Ogawa M, Alford R, Choyke PL, Urano Y. New strategies for fluores-
cent probe design in medical diagnostic imaging. Chem Rev 2009;110:2620–2640.
31. Zhang J, Campbell RE, Ting AY, Tsien RY. Creating new fluorescent probes for cell
biology. Nat Rev Mol Cell Biol 2002;3:906–918.
32. Cohen N, Novikov I, Hardan I, Esa A, Brok-Simoni F, Amariglio N, Rechavi G, Ben-
Bassat I, Trakhtenbrot L. Standardization criteria for the detection of BCR/ABL
fusion in interphase nuclei of chronic myelogenous leukemia patients by fluores-
cence in situ hybridization. Cancer Genet Cytogenet 2000;123:102–108.
33. http://www.iss.com/resources/reference/data_tables/LifetimeDataFluorophores.html.
34. Barcellona M, Cardiel G, Gratton E. Time-resolved fluorescence of DAPI in solution
and bound to polydeoxynucleotides. Biochem Biophys Res Commun 1990;170:270–
280.
35. Barcellona ML, Gratton E. Fluorescence lifetime distributions of DNA-40 , 6-diami-
dino-2-phenylindole complex. Biochim Biophys Acta Gen Subjects 1989;993:174–
178.
36. Redford GI, Clegg RM, Polar plot representation for frequency-domain analysis of
fluorescence lifetimes. J Fluorescence 2005;15:805–815.
651
Original Article
37. Barcellona M, Gratton E. The fluorescence properties of a DNA probe. Eur Biophys J
1990;17:315–323.
38. Fixler D, Tirosh R, Eisenthal A, Lalchuk S, Marder O, Irlin Y, Deutsch M. Prelytic
stimulation of target and effector cells following conjugation as measured by intra-
cellular fluorescein fluorescence polarization. J Biomed Opt 1998;3:312–325.
39. Zurgil N, Shafran Y, Fixler D, Deutsch M. Analysis of early apoptotic events in indi-
vidual cells by fluorescence intensity and polarization measurements. Biochem Bio-
phys Res Commun 2002;290:1573–1582.
40. Fixler D, Tirosh R, Shainberg A, Deutsch M. Cytoplasmic changes in cardiac cells
during a contraction cycle detected by fluorescence polarization. J Fluorescence
2001;11:89–100.
41. Cremer T, Cremer C. Chromosome territories, nuclear architecture and gene regula-
tion in mammalian cells. Nat Rev Genet 2001;2:292–301.
42. Misteli T. Higher-order genome organization in human disease. Cold Spring Harbor
Perspect Biol 2010;2:a000794
43. Meaburn KJ, Misteli T. Cell biology: Chromosome territories. Nature 2007;445:379–
381.
44. Cremer T, Cremer M, Hubner B, Strickfaden H, Smeets D, Popken J, Sterr M,
Markaki Y, Rippe K, Cremer C. The 4D nucleome: Evidence for a dynamic nuclear
652
landscape based on co-aligned active and inactive nuclear compartments. FEBS Let-
ters 589(2015) 2931–2943.
45. Visvanathan A, Ahmed K, Even-Faitelson L, Lleres D, Bazett-Jones DP, Lamond AI.
Modulation of higher order chromatin conformation in mammalian cell nuclei can
be mediated by polyamines and divalent cations. PLoS One 2013; 8(6): e67689.
46. Gorjanacz M. Nuclear assembly as a target for anti-cancer therapies. Nucleus 2014;5:
47–55.
47. Estandarte AK, et al. The use of DAPI fluorescence lifetime imaging for investigating
chromatin condensation in human chromosomes. Available at http://www.ucl.ac.uk/
~ucapikr/Ana_FLIM_Ver4.pdf at 11/5/2016.
48. Krivacic RT, Ladanyi A, Curry DN, Hsieh H, Kuhn P, Bergsrud DE, Kepros JF,
Barbera T, Ho MY, Chen LB. A rare-cell detector for cancer. Proc Natl Acad Sci USA
2004;101:10501–10504.
49. Jenkins PL, Freyer JP, Naivar MS, Arteaga A, Houston JP. Flow cytometric separation
of spectrally overlapping fluorophores using multifrequency fluorescence lifetime
analysis. In SPIE BiOS, Bellingham WA: International Society for Optics and Pho-
tonics: 2011.
50. Cao R, Pankayatselvan V, Houston JP. Cytometric sorting based on the fluorescence
lifetime of spectrally overlapping signals. Opt Express 2013;21:14816–14831.
FLIM for Identification of B-CLL