CELLULAR
134,
IMMUNOLOGY
Long-Term
157-170
Tracking
(1991)
of Lymphocytes
of PKH-Labeled
in Vivo: The Migration
Lymphocytes’
GARY F.TEARE,* PA~JL K. HORAN,PSUSAN E. SLEZAK,~.
CHERYL-SMITH,* AND JOHN B. HAY*,+'
Studies of in rwo cell migration
using cell markers such as “Cr. “‘In, FITC. or XRITC
have
been limited to short time periods due to the elution. toxicity.
or raped loss of label detectability.
WC have labeled sheep lymphocytes
w virro with PKH-2. a new fluorescent
cell membrane
label,
and, after their intravenous
injection
back into donor sheep. have been able to detect them tn
efferent lymph.
using tlow cytometry.
for longer than 38 days. The PKH-2-lahclcd
lymphocytes
migrated
with similar kinetics.
efficiency,
and tissue specificity
as lymphocytes
labeled with cell
markers used previously.
PKH-24abeled
cells mediated graft versus host reactions indistinguishable
from those mediated
by unlabeled cells. and cell surface antigens were equally detectable
on the
surface of labeled and unlabeled lymphocytes.
According
to the slow. consistent
loss of fluorescence
intensity
of the labeled cells in r’,w, we predict that labeled lymphocytes
could remain detectable
by flow cytometry
for greater than 7 weeks with the labeling protocol
used in these experiments.
h 1991 Academic Press. Ini
INTRODUCTION
Understanding the role and timing of the arrival and departure of specific cell types
or subsets in the pathogenesis of disease states, or in the generation of beneficial immune
responses, may lead to rational strategies for immunoregulation.
While much has been
learned about lymphocyte localization and migration in animals using various cell
labeling technologies, these studies have been limited to short-term time frames due
to the instability or toxicity of the cell markers that have been employed. A new
fluorescent cell labeling molecule, PKH-2. recently became available. This cell marker
integrates into the lipid bilayer of cell membranes and has been used by various laboratories to label several mammalian cell types as well as bacteria and other cells ( 1).
Studies with this labeling technology have shown that PKH-2 incorporates stably into
the cell membrane of rabbit erythrocytes such that labeled cells become intensely
fluorescent, and remain so over long periods of time after reinfusion into the blood
’ Funding
for this work was provided
by the Medical Research Council of Canada. G. F. Team is a fellow
of the Medical
Research Council
of Canada.
’ To whom correspondence
should be addressed
at Department
of Immunology,
Faculty
of Medicine.
University
of Toronto.
I Kings College Circle. Toronto.
Ontario.
Canada M5S I AX.
157
0008-8749/9
copyright
I $3.00
(cl 1991 h) Academx Press. Inc
All rights of reproductmn I” any form resenrd.
158
TEARE
ET
AL.
circulation of the rabbits (2). Several other parameters of cell function after labeling
with PKH label were previously examined. The cytotoxic function of mouse spleen
natural killer cells and the target susceptibility of Yac- 1 cells in culture were unchanged
when either of these cell types were labeled with PKH label (3) and PKH-2-labeled
Yac- 1 cells grew in culture with identical kinetics to unlabeled controls. Labeled cells
did not transfer their label to unlabeled cells. Interestingly, with each doubling of cell
number the mean fluorescence intensity of the labeled population decreased by half.
Thus the PKH label has the potential to allow retrospective analysis of cell division.
(S. E. Slezak, P. IS. Horan, and B. D. Jensen, in preparation). Before this label could
be used to study cell migrations and interactions in vivo, the migratory capability of
labeled cells must be evaluated. Here we have compared the migratory function of
PKH-Zlabeled lymphocytes with that of lymphocytes labeled with radiolabels and
other fluorescent markers in previous studies, and have also determined that these
PKH-2-labeled lymphocytes are easily detectable over many weeks in vivo.
MATERIALS
AND METHODS
ANIMALS,SURGERY,ANDLYMPHCOLLECTION
Randomly bred ewe lambs, 9- 12 months of age, were used in all experiments.
During the course of the experiments they were kept in metabolism cages with free
access to food and water. The efferent lymphatics draining lymph nodes of the mesentery and subcutaneous (prefemoral, prescapular) lymph nodes were cannulated under
pentobarbital anesthesia as described previously (4-6).
Lymph was collected into sterile polyethylene bottles containing 0.25- 1.O ml of
normal saline containing 1000 IU/ml heparin sodium USP (Hepalean, Organon Canada Ltd., Toronto, Ontario) and 20,000 IU/ml penicillin-G potassium (Ayerst Laboratories, Montreal, Quebec) such that the final concentration of heparin and penicillin
in the lymph collection was approximately 5 IU/ml and 100 IU/ml, respectively.
Lymph volume was measured in sterile polypropylene graduated cylinders and lymphocyte concentration in each sample was determined by hemocytometer. All cells
for PKH-2 labeling were kept under strict sterile conditions throughout the labeling
procedure.
Lymphocytes for labeling were collected from the efferent lymph draining mesenteric
or subcutaneous lymph nodes. The cells were isolated from lymph by centrifugation,
washed once with Hank’s balanced salt solution (HBSS, pH 7.3) (GIBCO, Grand
Island, NY) and resuspended to a concentration of 4 X 107/ml in the labeling diluent
“A” which is provided with the PKH-2 dye (Zynaxis Cell Science, Malvern, PA). To
a volume of diluent “A” equal to that into which the cells were suspended, sufficient
PKH-2 dye was added to make a solution twice the desired labeling concentration.
This dye/diluent solution was added to the cell/diluent suspension and gently but
thoroughly mixed by agitation. The final labeling conditions were thus 2 X lo7 cells/
ml in diluent “A” with either 2.5 or 10 PM PKH-2 depending on the experiment.
After incubating the cells with label at room temperature for 2.5 min, the labeling
reaction was stopped by addition of a doubling volume of M- 199 medium containing
LONG-TERM LYMPHOCYTE TRACKING
159
Earle’s salts, 25 mM HEPES, L-glutamine (M- 199; Flow Laboratories Inc., McLean,
VA) with 10% lamb serum (GIBCO). All media for washes and incubations were used
at room temperature. The labeled lymphocytes were recovered by centrifugation, and
were resuspended in serum-free M- 199 for immunofluorescence,
or in M- 199/ 10%
lamb serum for in vivo experiments. An aliquot of labeled cells was always set aside
for verification of staining and viability testing.
The aliquot of labeled lymphocytes was analyzed by FCM3 for green (PKH-2) fluorescence intensity, percentage of cells labeled, and for viability by propidium iodide
(7) and/or trypan blue exclusion. Greater than 98% of the cells became labeled every
time we stained lymphocytes with PKH-2. Lymphocyte populations used for recirculation experiments were at least 85% viable when injected into the venous circulation
of the sheep.
INDIRECT IMMUN~FLU~RESCENCE
Lymphocytes were recovered from lymph by centrifugation and were washed twice
with HBSS. Some cells were labeled in 2.5 FM PKH-2 while controls were left unlabeled. Unlabeled cells were kept at room temperature in M-199 with 10% lamb serum
while the other cells were being labeled. Both PKH-2 labeled and unlabeled cells were
(separately) washed twice with HBSS at 4°C and resuspended to a concentration of 2
X 10’ cells/ml in serum-free M- 199. Two million cells ( 100 ~1) were placed into wells
of a round-bottom
96-well microtiter plate so that each monoclonal antibody and
each control was tested in duplicate. Two well-characterized monoclonal antibodies,
SBU-T4, and SBU-I, which specifically bind sheep CD-4, and a nonpolymorphic determinant of MHC class I (OLA I) cell membrane antigens, respectively, were used
in this study (8,9). The cells were suspended in 100 ~1 of a I:20 dilution ofthe primary
antibody for 30 min at 4°C washed three times with ice cold M-199, and incubated
in 100 ~1 of a 1:2 dilution (0.5 mg/ml) of goat anti-mouse Ig conjugated to R-PE
(Sigma Chemical Co., St. Louis, MO) for 30 min at 4°C. After the secondary incubation,
the cells were washed three times and were fixed in cold 1% paraformaldehyde for 30
min. Fixed cells were washed three more times and were resuspended in a storage
medium consisting of M-199 cell culture medium, 1% bovine serum albumin (Sigma
Chemical Co.), and 0.1% sodium azide. Cells were analyzed by FCM within 2 days.
ANALYSIS OF CELLS BY FLOW CYTOMETRY
Recirculution
Experiments
Cells collected for analysis by FCM were isolated from lymph by centrifugation
(600 g for 5 min at 20°C) washed once with phosphate-buffered saline (PBS, pH 7.3)
and resuspended in cold 1% paraformaldehyde at a concentration of 5 X 1O’/ml. After
30 min of fixation, the cells were again pelleted, the paraformaldehyde was decanted
off, and they were resuspended in storage medium. The samples were stored in the
dark at 4°C until analyzed.
’ ilhhrevia~iom
zr.wd~
FITC, fluoresceinisothiocyanate;XRITC, substitutedrhodamine isothiocyanate:
FCM. flow cytometry;MHC, major h&compatibility complex;lg. immunoglobulin;R-PE,R-phycoerythrin:
MTT, modal transit time.
TEARE ET AL
160
Five hundred thousand cells were analyzed for each sample on an EPICS V flow
cytometer (Coulter Electronics, Hialeah, FL) with an argon laser set at 500 mW power,
and 488 nm wavelength for excitation of the PKH-2 label which emits maximally at
504 nm. Green fluorescence of the PKH-2 was collected through a 525/40 bandpass
filter. Forward and right angle light scatter were used to exclude cell clumps and debris
from the analysis.
All cells were analyzed within 7 days of their collection from the animal. Paraformaldehyde-fixed, PKH-2-labeled lymphocytes could be stored at 4°C for up to 14
days without loss of fluorescence intensity. Even after 28 days of storage, the mean
fluorescence intensity did not vary by more than four channels when compared to
the same sample analyzed within 7 days of collection from the animal. Since green
fluorescence intensity was compared over a long period of time in these experiments,
it was crucial to ensure that the flow cytometer was carefully aligned and calibrated
for fluorescence intensity immediately prior to each analysis session. Immuno-Check
alignment fluorospheres (Coulter Electronics, Batch 505 1) were run such that exactly
the same fluorescence intensity (mean log channel number) was recorded prior to
starting the analysis of cells. At the end of each analysis session, the fluorescence
calibration was rechecked with the fluorospheres and was always found to be within
less than one channel of the starting value.
Immunojluorescence
Experiments
Ten thousand cells were analyzed from each sample on an EPICS V flow cytometer
with an argon laser run at 500 mW power at an excitation wavelength of 488 nm.
Green signal from PKH-2 fluorescence was collected through a 525/40 bandpass filter
while orange-red signal from R-PE was collected through a 575/30 bandpass filter
into separate photomultiplier tubes (PMT). Due to the high intensity of the PKH
label, it was necessary to adjust both the electronic pulse subtraction module and the
high voltage to the green PMT in order to achieve the necessary amount of compensation for spectral overlap.
FLUORESCENCEMICROSCOPY
PKH-2-labeled cells were examined and photographed using a Zeiss photomicroscope provided with an epifluorescence attachment and a phase contrast condenser.
STATISTICAL ANALYSIS
The paired-sample Student’s t statistic was used for comparison of the mean skin
thicknesses in the graft-versus-host reaction assay.
RESULTS
INTENSEFLUORESCENCELABELINGOFLYMPHOCYTES
PKH-2 was able to label sheep lymphocytes very intensely. In order to determine
the most suitable labeling conditions, a series of experiments were performed varying
the cell concentration, the PKH-2 concentration, or the time of incubation in the cell
labeling procedure. The cells labeled very rapidly, with greater than 98% of the cells
LONG-TERM
LYMPHOCYTE
161
TRACKING
becoming intensely fluorescent within as little as 60 set of incubation with PKH-2. It
was found that cellular fluorescence increased with increasing concentration of PKH2 but that at labeling concentrations greater than 10 pA4 significant cytotoxicity occurred. At cell concentrations greater than 2 X 107/ml cell loss also increased proportional to cell concentration even at PKH-2 concentrations less than 10 PM. As a
result of these experiments, we selected labeling conditions of 2 X IO7 cells/ml in 10
pA.4 PKH-2 for 2.5 min at 22°C for our in viva cell tracking experiments, in order to
label the lymphocytes as intensely as possible with a minimum of cytotoxic effect.
Cells labeled in this manner were approximately 3000 times as bright as unlabeled
cells and were easily detected by flow cytometry (FCM) (Fig. 1A) or fluorescence
microscopy (Fig. 2).
Fluorescence microscopy of freshly labeled lymphocytes demonstrates that this dye
labels the cell membrane and intracellular structures such as the nuclear membrane
(Fig. 2). PKH-2-labeled lymphocytes with the same staining pattern were also easily
visible in frozen tissue sections of lymph node. spleen. and Peyers patches (not shown).
The labeling was so intense that very small numbers of labeled cells could be unequivocally identified in lymph hours or weeks after injection into the circulation of
sheep. In one experiment, the efferent lymph draining a subcutaneous lymph node
1 Mean channel
I
122
:
I
Mean
channel
115
Mean
channel
15
i
lo3
Green
Fluorescence
FIG. I. Single parameter
histograms
showing
the fluorescent
intensity
of PKH-?-labeled
lymphocytes
immediately
prior to intravenous
injection
(A), and from efferent lymph 22-25.5
hr (B) and 501.5-5053.5
hr postinjection
(C). The channel numbers
from the O-256 logarithmic
fluorescence
intensity
scale of the
EPICS V have been converted
to linear fluorescence
units to simplify
the comparisons.
The “mean channel
number”
given is in these linear units. For histogram
(A). a total of 10’ cells were counted.
of which >9YX
were PKH+.
For each of histograms
(B) and (C), a total of 5 X IO5 ceils were counted.
Unlabeled
cells in
histograms
(B) and (C) constituting
98.94 and 99.7% respectively
of the total cells counted
are extremely
dim relative to the PKH-labeled
ceils. and have accumulated
in the first few channels
of the histograms.
162
TEARE
ET AL.
FIG.2. Appearance of freshly labeled PKH-2-labeled
40X oil-immersion objective.
lymphocyte under fluorescence microscopy using
collected during the fifth hour after intravenous injection of PIUI-2-labeled cells was
found to contain 2 1 labeled cells per million when analyzed by FCM (data not shown).
FUNCTIONAL
ABILITIES
OF PKH-2-LABELED
LYMPHOCYTES
Several parameters of lymphocyte migration were examined with respect to possible
effects of the PKH-2 cell label on the normal functioning of lymphocytes. When
lymphocytes collected from mesenteric lymph were labeled with PKH-2 and returned
to the venous circulation of sheep, the cells migrated into lymph in a nonrandom,
tissue-specific manner. In both experiments, PKH-labeled mesenteric efferent lymphocytes migrated preferentially into mesenteric efferent lymph (Fig. 3). The ratio of
the concentration of labeled mesenteric efferent lymphocytes returning in mesenteric
efferent lymph to their concentration in subcutaneous efferent lymph over the first 40
hr postinjection was 1.87 in one experiment and 1.3 1 in the other. As expected, lymphocytes collected from subcutaneous efferent lymph and labeled with PKH-2 migrated
randomly into subcutaneous efferent lymph after intravenous injection.
PKH-2-labeled lymphocytes migrated from blood to lymph with the same efficiency
as lymphocytes labeled with other cell markers as previously determined in this laboratory (Table 1). Lymphocytes enter efferent lymph at the rate of approximately 3
X 10’ cells per gram of lymph node each hour ( 10, 11). Since more lymphocytes will
traffic through a large lymph node in a given time period than through a small one,
we standardize and normalize our recovery values to the percentage of injected (labeled)
cells recovered per 10’ cells collected (labeled and unlabeled) over 40 hr starting at
LONG-TERM
LYMPHOCYTE
time
163
TRACKING
(hours
)
FIG. 3. Concentration
of labeled cells in the mesenteric
(e) and subcutaneous
(0) efferent lymph of a
sheep after intravenous
injection
of 3.0 X IO9 lymphocytes
collected
from mesenteric
lymph and labeled in
IO PM PKH-2.
The kinetics
and tissue specificity
demonstrated
by the recirculating
PKH-?-labeled
lymphocytes
are not significantly
different
from lymphocytes
labeled with radioisotopes
or other fluorescent
labels.
the time of injection. This normalization permits comparison of labeled cell recovery
regardless of the specific lymph node cannulated. The recovery of PKH-2-labeled
lymphocytes returning to the efferent lymph compartment from which they were
isolated was determined in four experiments and the mean recovery rate +- standard
error of the mean (SEM) was found to be 0.45 + 0.1 I %/ 10’ cells in the first 40 hr
after injection.
Another well-defined parameter of lymphocyte tralhc in sheep is the time at which
the peak output of labeled lymphocytes occurs in lymph relative to the time of injection
of the cells (“modal transit time”). In the four experiments reported here, the modal
transit time + SEM of PKH-2-labeled lymphocytes was 34.25 -t 3. I2 hr. Intravenously
TABLE
Recovery
of PKH-Labeled
Label
in Lymph
Comparison
Recovery”
(%/lo9 cells/40 hr)
(mean t SEM)
used
Indium-I
I1
Chromium-5
FITC
XRITC
PKH-2
Lymphocytes
I
1
0.30
0.27
0.61
0.56
0.45
i
f
f
k
f
0.04
0.03
0.10
0.06
0. I I
to Lymphocytes
with
Other
Labels
n”
I I
25
8
6
4
’ Percentage
of labeled cells recovered
in lymph per IO9 lymph cells collected
in the first 40 hr after
injection
of labeled cells into the venous circulation
of sheep. Normalization
of the data to recovery
per lo9
cells allows comparison
of recoveries
through
lymph nodes of different
sizes and thus differing
cell output
rates.
h See text for references.
164
TEARE
ET
AL.
injected PKH-2-labeled lymphocytes appeared in small numbers in the lymph within
a few hours. The concentration of these cells increased rapidly over the course of the
first 24 hr, peaked after 30 hr, and declined to 60-70% of the peak concentration over
the subsequent 60-120 hr. After this initial decline the concentration of labeled cells
remained steady, declining very slowly over time (Fig. 3).
As a further test of function of PIUS2-labeled lymphocytes we employed the graft
versus host (GVH) response as measured by change in skin thickness. This response
depends on the viability, immune responsiveness, and capacity to proliferate of the
injected allogeneic lymphocytes ( 12). As can be seen in Fig. 4, in all three experiments
there was no statistically discernible difference in the GVH response of either 10’
viable PKH-2-labeled or unlabeled cells. In two of these experiments, labeled and
unlabeled cells irradiated with 5000 rad of y radiation were tested for their GVH
response simultaneously with the unirradiated lymphocytes. Although the irradiated
cells were viable when injected, they failed to mediate a GVH response, confirming
the role of donor cell proliferation in the GVH reaction.
PKH
Sheep
DAY
ID#
0
5
5
406
0
I
SR
5R
0
682
5
5
0
I
SR
5R
0
5
5
0
48
I
0
2
4
skin thickness
6
8
(mm)
FIG. 4. IO7 viable PKH-2-labeled
(PKH+)
and unlabeled
(PKH-)
allogeneic lymphocytes
were injected
intradermally
into at least three sites each in the hairless skin of the groins of three sheep. Skin thickness
measurements
were taken before injection
(Day 0) and on the fifth day after injection
of the cells (Day 5).
In two of the sheep, labeled (PKH+
Day 5R) and unlabeled (PKFDay 5R) cells which had been irradiated
with 5000 rad of y radiation
were also injected into separate sites. There was no statistically
discernible
difference
(406 and 682: P % 0.3; 48: P z 0.09) between the skin thicknesses
of GVH responses within each
sheep whether
or not the allogeneic
cells were labeled with PKH-2.
Irradiated
cells, though viable by the
trypan blue exclusion
method when injected,
were unable to mediate a GVH reaction.
Lymphocytes
were
labeled in IO PM PKH-2.
Error bars show standard deviation.
LONG-TERM
LYMPHOCYTE
165
TRACKING
Since PKH-2 labels the cell membrane we wanted to determine whether it would
interfere with immunofluorescent
detection of cell surface antigens. Lymph cells from
the same collection were either labeled with PKH-2 or left unlabeled and were subsequently phenotyped using the SBU-T4 and SBU-1 antibodies to the sheep CD4
molecule and MHC class I products, respectively (Fig. 5). All tests were done in du-
96.72
0.95
0.06
D
c
2.33
1.0
0.
0.02
i 0.26
0.11
4.78
yj
1:
0.29
I
I!
0.08
0.
0.
1
G
H
8-1
99.86
1
0.17
!i
1
0.06
I
1’
I
0.3 r!
I
I
I
199.53
FIG. 5. Two-color
flow cytometric
analysis of lymphocytes
labeled with PKH-2
and by indirect
immunofluorescence.
Two million unlabeled (A. C. E) or PKH-2-labeled
(B. D. F) lymphocytes
were incubated
with either SBU-T4
(A, B) or SBU-I (C. D). monoclonal
antibodies
which identify the sheep CD4 molecule
and MHC class I products,
respectively,
or with a control murine monoclonal
antibody
produced
by ATCC
hybridoma
HB-65 (E, F), which recognizes
a determinant
on influenza nucleoprotein.
The secondary
reagent
for immunofluorescence
was goat anti-mouse-lg
conjugated
with R-PE. Two color histograms
of unlabeled
and PKH-2.labeled
cells which were not incubated
with antibodies
are also shown (G, H). Ten thousand
cells were analyzed for each histogram,
and the numbers
at the comers of the histogram
quadrants
indicate
the percentage
of total cells contained
within that quadrant,
166
TEARE
ET AL.
plicate. The mean percentage of CD4+ and OLA I+ cells in the PKH-labeled (PKH+)
and unlabeled (PKH-) groups were CD4+ PKH+, 50.7%; CD4+ PKH-, 51.9%;
OLA I+ PKH+, 94.1%; OLA I+ PKH-, 96.8%. The presence of the PKH fluorescent
label did not interfere with the detection of either of these cell surface molecules.
LONG-TERMLYMPHOCYTETRACKING
The mean fluorescence intensity of the labeled cells returning in the lymph in the
first 24 hr was not substantially different from the intensity of the injected cells (Figs.
1A and 1B). In each of three experiments in which the labeled cells returning to lymph
were monitored for greater than 48 hr the mean fluorescence intensity decreased at a
steady rate over the entire time course, with a half-life ranging from 7.7 to 9.8 days
in the three experiments (Fig. 6). As can be seen from Fig. 6a, PKH-2-labeled lymphocytes were readily detectable for greater than 38 days (-9 17 hr) in viva and indeed
many of the labeled cells could remain detectable for 7 weeks or more assuming that
the rate of fluorescence loss remained constant. As can be seen from Fig. 1, when the
fluorescence
half-life = 8.3 days
o! . . . . . . . . . . . . . . . . . ,......,..,._.,..,,.
0
b.
c.
0 !...
0
168
336
504
672
‘~~“~~“‘~~““~~~“““~‘~‘~~~~~“’
168
336
504
672
,.
__._,...
..,..
336
504
672
840
840
226
o! .__...
0
I.
168
,_,
..,,._...,
840
Time (hours)
FIG.6. Mean fluorescence intensity of recirculating lymphocytes labeled in 10 ~~ PKH-2. Lymphocytes
collected by chronic lymphatic drain were fixed in paraformaldehyde and analyzed by FCM. The figure
shows results from three experiments in two separate sheep; (a) sheep T04, first injection of labeled lymphocytes; (b) sheep TOClabeled lymphocytes injected 26 days after the first injection; (c) sheep T05. The
half-life of fluorescence intensity is remarkably consistent between experiments, with a mean of 8.6 days.
LONG-TERM
LYMPHOCYTE
TRACKING
167
lymphocytes are labeled with PKH-2 the distribution of logarithmic fluorescence intensities in the cell population is symmetric and Gaussian. However, as time passed,
the distribution of fluorescence intensity in the population of labeled lymphocytes
recovered from the lymph became increasingly skewed and broadened to the left.
DISCUSSION
In this paper we have described some of the functional properties of sheep lymphocytes labeled with a new fluorescent marker molecule, PKH-2, which incorporates
into the lipid bilayer of cell membranes and permits long-term cell tracking in rive.
The integrity of the cell membrane is of primary importance to the viability and
functionality of all cells. Previous studies had demonstrated that rabbit erythrocytes
labeled with a PKH fluorescent cell membrane label were indistinguishable from unlabeled erythrocytes in their susceptibility to volume changes and lysis due to hypoosmolar stress (2). Thus, incorporation of the PKH label into the cell membrane
did not physically weaken the membrane. Lymphocytes depend on the integrity and
plasticity of their cell membrane for all of their functions including immune recognition
and effector functions, and migration for the dissemination and surveillance of immunity. Slezak and Horan demonstrated that mouse spleen cells and a mouse lymphoma line, Yac- 1, could be labeled with a PKH cell marker molecule for use in a
flow cytometric cell-mediated cytotoxicity assay which was found to have excellent
correlation to the chromium release assay (3). Labeled Yac- 1 targets were neither more
nor less susceptible to lysis by NK cells in the effector spleen cell population than
were unlabeled targets. Similarly, PKH-labeled and unlabeled effector spleen cells
were equally capable of killing targets. This data indicates that in lymphocytes and a
lymphoid cell line, the presence of the label in the cell membrane neither interferes
with recognition, adhesion, and lytic events nor renders the target cell membrane
more susceptible to lysis. In their studies, the cells were labeled with comparable
amounts of PKH-label as were used in our studies of sheep lymphocytes.
Lymphocyte migration involves a complex series of events all of which are dependent
on the functional integrity of the cell membrane. Lymphocytes must first attach themselves to endothelium by means of strong, sometimes tissue-specific intercellular adhesive interactions, then must squeeze themselves between the endothelial cells and
cross through the basement membrane into the underlying tissue (13). Within tissues
they presumably interact with numerous cell types by means of molecules on their
cell surface until they enter the lymph which returns them to the blood circulation to
repeat the process. This lymphocyte migration process has been studied in sheep by
our laboratory and others for almost 20 years using cell markers such as Na2”Cr04,
‘“In-oxine
fluorescein, or substituted rhodamine isothiocyanate (FITC, XRITC) as
they became available ( 14- 18). The large body of data regarding the efficiency and
kinetics of labeled-lymphocyte migration from blood to lymph, and tissue-specific
migration patterns that has accumulated over the years, is remarkably consistent and
therefore probably physiologically sound. This body of data also provides a standard
against which we were able to compare the migration of PKH-2-labeled lymphocytes.
Sheep lymphocytes labeled with PKH-2 migrated from blood to lymph with similar
efficiency (see Table l), kinetics, and tissue specificity as lymphocytes labeled with
radiolabels or fluorescent cellular protein labels.
168
TEARE
ET
AL.
The length of time taken by the majority of intravenously injected, labeled lymphocytes to enter the efferent lymph has been variously reported at between 21 and
30 hr in sheep. This “modal transit time” (MTT) has been found to be relatively
consistent whether the label was 5’Cr ( 17, 19), ” ‘In ( 15) FITC ( 16, 18), or XRITC
( 18). However, some studies of lymphocyte recirculation in sheep have shown a later
time for the MTT, between 30 and 50 hr postinjection of “Cr- or FITC-labeled lymphocytes (20,2 1). The MTT we report here for PKH-2-labeled lymphocytes is slightly
higher than the upper end of the range typically reported for lymphocytes labeled with
the other markers. This may be a real phenomenon due to the label or labeling procedure, or may be due to our relative inexperience with this new labeling technique,
compared with the other more familiar labels.
As further evidence that the presence of the reporter molecules in the lipid bilayer
of the cell membrane did not interfere with major function of lymphocytes, PKH-2labeled allogeneic lymphocytes were found to mediate graft versus host responses
indistinguishable from those generated by unlabeled lymphocytes. The detectability
of cell surface antigens was unaffected by the label at the lower (2.5 PM) labeling
concentration, while at the higher concentration (10 p.M) the amount of spectral
overlap of the extremely intense PKH fluorescence (green) into the orange-red emission
region of the R-PE conjugated to the secondary antibody made adequate spectral
overflow subtraction impossible. Thus, detection of the surface antigens CD4 and
OLA I on the more intensely labeled cells was not feasible in our study.
Earlier experiments in our laboratory determined that after intravenous injection
lymphocytes taken from mesenteric efferent lymph and labeled with 51Cr or I1‘In
returned preferentially to mesenteric efferent lymph such that the ratio of the specific
activity in mesenteric efferent to that in subcutaneous efferent lymph (i.e., mesenteric
cpm/ 10’ cells/subcutaneous cpm/ 10’ cells) over the first 40 hr was (mean ? SD) 1.74
+ 0.2 1 (18). The analogous ratio in this study was the ratio of the concentration of
PKH-Zlabeled mesenteric efferent lymphocytes returning in mesenteric efferent lymph
to that in subcutaneous efferent lymph over the first 40 hr postinjection. In the two
experiments of this type which were performed, the ratios (% PKH+ in mesenteric
lymph/% PKH+ in subcutaneous lymph) were 1.87 and 1.3 1. There is no apparent
difference in the tissue-specific, nonrandom migratory behavior of lymphocytes labeled
with PKH-2 compared to those labeled with other labels.
PKH-2-labeled lymphocytes were easily detected by FCM in lymph samples from
sheep over a period of greater than 5 weeks from the time they were injected intravenously. The half-life of the mean fluorescence intensity of PKH-2-labeled lymphocytes in vivo was remarkably consistent from one experiment to the next, being on
average 9 days. Although we did not continue any of our experiments beyond 9 10 hr
after intravenous injection of the labeled cells (>5 weeks), we predict, on the basis of
the constant rate of fluorescence intensity decay and the fluorescence intensity distribution at 910 hr, that PKH-2-labeled lymphocytes would remain detectable in vivo
for at least 7 weeks. Published results from other laboratories, and our own experience
with lymphocytes labeled with FITC, RITC, or XRITC, indicate that detection of
lymphocytes labeled with these labels is generally limited to less than 2 weeks and
often to only a few days ( 16, 17,2 1,22). Intravenously injected FITC- or RITC-labeled
murine lymph node cells can be recovered from syngeneic recipient mouse lymph
nodes after up to 11 days in vivo and detected by FCM or fluorescent microscopy in
LONG-TERM
LYMPHOCYTE
TRACKlNG
169
cell suspension (22). However, by 1 1 days after injection only half of the injected cells
known to be in the lymph nodes could be detected, and these had a median fluorescence
intensity of only 0.9% of their initial intensity. The PKH-2-labeled sheep lymphocytes,
however, still retained 12% of their original intensity after 2 1 days in viva.
The reasons for the loss of fluorescence intensity from the PKH-2-labeled lymphocytes are not entirely clear. Part of the loss is likely due to leaching of dye from the
surface of the cells. Unlike what some of us (S.E.S., P.K.H.) found with Yac-1 cells,
the PKH-2 dye did transfer from labeled to unlabeled sheep lymphocytes in vitro.
When labeled lymphocytes were mixed with an equal number of unlabeled ones in
the wells of a round-bottom
microtiter plate, gently centrifuged to ensure close apposition of the cells, and incubated at 37°C for 3 hr in complete medium, the mean
intensity of the unlabeled cells approached 6% of that of the labeled cells. Since the
dye transferred in vitro, it is likely to do so in vivo as well. We saw evidence of increased
“background”
fluorescence on the unlabeled population of lymphocytes in many of
the lymph samples we collected after PKH-2-labeled cells were injected intravenously.
It is significant that the newer PKH-26 cell-labeling molecule. which can be exited at
488 nm and emits at 572 nm (red), does not transfer between labeled and unlabeled
sheep lymphocytes (unpublished observations). This new red-fluorescing cell label will
perhaps be even more useful than the green-fluorescing PKH-2, in that it can be used
with FITC-conjugated antibodies for two color phenotypic analysis of recirculating
lymphocytes. Since the chemistry of the two PKH molecules are very similar, we do
not anticipate any difference between PKH-2- and PKH-26-labeled lymphocytes in
terms of their capacity to recirculate.
Studies of lymphocyte migration depend on marker molecules to enable tracking
of the cells in the body. The various radiolabels and fluorochromes used in such
studies each have their own particular advantages and disadvantages; however. none
of the labels previously available enabled long-term tracking of lymphocytes. As we
have shown, PKH-2 enables detection of labeled lymphocytes for several weeks to
months. The observations of this study should be seen as early hndings in the application of this new cell labeling technology. We find that when labeled at the higher
doses of PKH-2 (i.e., greater than 5 PM) there is considerable variability in the viability
of sheep lymphocytes from one experiment to the next. The cell death is due to a
concentration-dependent
cytotoxic effect of the PKH-2 molecule (unpublished observations); however, we do not have an adequate explanation yet for the variable
susceptibility of lymphocytes from different sheep to the toxic effect. The PKH label
will likely be most useful for labeling lymphocytes at less than or equal to 5 PM
concentrations in suspensions of lymphocytes ranging from 2 to 5 X lO’/ml. Along
with recent information on the role of cytokines on the local recruitment or local
detention of lymphocytes (24-26) this new cell labeling technology makes possible a
new realm of experimentation and possibly the determination of the real lifespans
and long-term migration patterns of lymphocytes and their subsets in normal and
pathological conditions.
REFERENCES
1, Slezak,
2. Slezak,
3. Slezak,
S. E., and Horan.
S. E., and Horan,
S. E., and Horan.
P. K.. h’uruw (London)
340, 167. 1989.
P. K.. Blood 74, 2 172, 1989.
P. K., J. Immunol.
MethA
177, 205. 1989.
170
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
2 I.
22.
23.
24.
25.
26.
TEARE
ET AL.
Lascelles, A. K., and Morris, B., Q. J. Exp. Physiol. 46, 199, 1961.
Hall, J. G., Q. J. Exp. Physiol. 52, 200, 1967.
Heitmann, H. H., Zentrolbl.
Veterinarrned.
(A) 17, 517, 1970.
Krishan, A., J. Cell Biol. 66, 188, 1975.
Maddox, J. F., MacKay, C. R., and Brandon, M. R., Immunology
55, 739, 1985.
Gogolin-Ewens, K. J., Mackay, C. R., Mercer, W. R., and Brandon, M. R., Immunology
56,717,
1985.
Hall, J. G., and Morris, B., J. Exp. Med. 121, 901, 1965.
Hay, J. B., and Hobbs, B. B., J. Exp. Med. 145, 31, 1977.
Jones, M. A. S., and Lafferty, K. J., Aust. J. Exp. Biol. Med. Sci. 47, 159, 1969.
Schoefl, G. I., and Miles, R. E., J. Exp. Med. 136, 568, 1972.
Cahill, R. N. P., Poskitt, D. C., Frost, H., and Trnka, Z., J. Exp. Med. 145,420,
1977.
Issekutz, T., Chin, W., and Hay, J. B., Clin. Exp. Zmmunol. 39, 215, 1980.
Schnorr, K. L., Pearson, L. D., Knisley, K. A., and DeMartini, J. C., Znt. Archs. Allergy Appl. Zmmunol.
72, 239, 1983.
Chin, G. W., and Cahill, R. N. P., Immunology
52, 341, 1984.
Abernethy, N. J., Chin, W., Lyons, H., and Hay, J. B., Cytometry
6,407,
1985.
Chin, W., and Hay, J. B., Gastroenterology
79, 123 1, 1980.
Frost, H., Cahill, R. N. P., and Trnka, Z., Eur. J. Zmmunol. 5, 839, 1975.
Reynolds, J., Heron, I., Dudler, L., and Trnka, Z., Immunology
47, 4 15, 1982.
Butcher, E. C., Scollay, R. G., and Weissman, I. L., J. Zmmunol. Methods 37, 109, 1980.
Melnicoff, M. J., Morahan, P. S., Jensen, B. D., Breslin, E. W., and Horan, P. K., J. Leuk. Biol. 43,
387, 1988.
Kaalaji, A. N., Abemethy, N. J., McCullough, K., and Hay, J. B., Reg. Zmmunol. 1, 56, 1988.
Hein, W. R., and Supersaxo, A., Zmmunology64,
469, 1988.
Kaalaji, A. N., McCullough, K., and Hay, J. B., Immunology
Lett. 23, 143, 1989.