Cytometry Part B (Clinical Cytometry) 78B:59–64 (2010)

Utility of Lyophilized PMA and Ionomycin to

Stimulate Lymphocytes in Whole Blood for
Immunological Assays
Shelley Sims Belouski,1 Julie Wilkinson,2 John Thomas,1 Keith Kelley,1
Shen-Wu Wang,1 Sid Suggs,1 and John Ferbas1*
1
Department of Medical Sciences, Amgen Inc., Thousand Oaks, California
2
Custom BioPharma Group, Beckman-Coulter Inc., Miami, Florida

Background: The need to implement robust biomarkers in clinical trials has never been greater, and
such efforts can be easily compromised by reagent instability or simple human error during assay set-
up. Many biotechnology and pharmaceutical companies are introducing efforts to conduct biomarker
studies under more rigorous settings, and the use of plates or tubes pre-loaded with stimulation or stain-
ing reagents could be of value for studies that involve flow cytometry.
Methods: Five reagents lyophilized from ethanol or CHAPS buffer stock solution of phorbol 12-myris-
tate 13-acetate (PMA) and ionomycin were benchmarked against standard DMSO liquid formulation for
their stimulation equivalency. The median fluorescence intensity of phosphorylated ribosomal protein S6
in lymphocytes was assessed on a BD FACSCaliburTM.
Results: We demonstrate here that tubes pre-loaded with lyophilized versions of the liquid reagents
can provide equivalent stimulation in healthy volunteer specimens.
Conclusions: The value of this approach is that it safeguards against omission or erroneous addition
of bulk liquid formulations of PMA and ionomycin to the reaction vessel (i.e., plate or tube) and also
lends itself to extended stability/shelf-life of these reagents. On the basis of this initial success, we plan
to expand our evaluation of lyophilized reagents so that they can be incorporated into our clinical bio-
marker campaigns as appropriate. VC 2009 Clinical Cytometry Society

Key terms: flow cytometry; phosphoprotein; whole blood; biomaker

How to cite this article: Belouski SS, Wilkinson J, Thomas J, Kelley K, Wang S-W, Suggs S, Ferbas J. Utility of ly-
ophilized PMA and ionomycin to stimulate lymphocytes in whole blood for immunological assays. Cytometry
Part B 2010; 78B: 59–64.

The positive impact that flow cytometers have made
in the clinical arena is unambiguous, and the value of
this analytical platform continues to rise. This is
reflected in part by the continued introduction of novel
fluorescent probes (e.g., Alexa FluorsV R , tandem dyes
and quantum dots) and robust cell permeabilization and
intracellular staining methods. Moreover, new commer-
cial instrument platforms that differentiate well-beyond
the 4–5 fluorochromes (e.g., BDTM LSR II, BD FACSCan-
toTM II) which have traditionally served our industry,
and integrated systems that automate set-up and analysis
(e.g., FC 500) are significant. Of equal importance, how-
ever, are the parallel activities in quality assurance and
experimental design that have accompanied instrument
and reagent development. Such efforts were noted even
in the earliest clinical applications of flow cytometry
(1,2), precluding skepticism in the field that could have
arisen had these efforts not been made.
The need for optimized approaches and harmonized
methods has increased because many of the recent
advances in fluorochrome and instrument configurations
represent such a large step forward that they can be
considered in many ways to be first-in-class. Moreover,
recent and increasing enthusiasm for biomarkers in early
*Correspondence to: John Ferbas, One Amgen Center Drive, Mailstop
30E-3-C, Thousand Oaks 91320-1799, CA.
E-mail: jferbas@amgen.com
Received 18 February 2009; Revision 7 July 2009; Accepted 9 July
2009
Published online 23 September 2009 in Wiley InterScience (www.
interscience.wiley.com).
DOI: 10.1002/cyto.b.20492

C 2009 Clinical Cytometry Society

V
60 BELOUSKI ET AL.
Table 1
Derivation of Lyophilized Preparations of PMA and
Ionomycin for this Study
Solvent !
Excipient ; EOH CHAPS
1 Ba Eb
2 Cb Fb
3 Ac Db
The reagents were solubilized in one of two solvents (etha-
naol or CHAPS) and combined with one of three proprietary
excipients for a total of six potential formulations. Visual ac-
ceptance criteria were used to assess the success of the ly-
ophilization across a titration. Lyophilates C, D, E, and F were
further characterized in this study (Fig. 2).
a
Partially successful.
b
Successful.
c
Failed.
stage clinical development programs have created a
need for highly specialized approaches which often
times must be conducted at the local site of a given clin-
ical trial, rather than by the pharmaceutical or biotech-
nology laboratory where the idea and method
development originated.
One often overlooked aspect of the process that is
emerging in clinical development programs for virtually
every biomarker campaign relates to the stability and
lot-to-lot variability of critical reagents. Traditionally, our
industry has not made a concerted effort to have over-
arching control of reagents, but have instead relied
upon reliable reagent manufacturers and trending ana-
lyzes when switching from one lot of reagent to another.
Unfortunately, there is little that can be done when a
sub-optimal reagent lot is identified mid-study if the man-
ufacturer does not have multiple reagent lots on hand,
and some reagents – including phorbol 12-myristate 13-
acetate (PMA) and ionomycin – are inherently unstable
in liquid form (3,4). A new idea to better control rea-
gent stability and lot-to-lot variability is implementation
of plates or tubes pre-loaded with lyophilized reagents
(5), where a supply of lyophilized reagent could be pre-
pared to span the entire clinical study even if it were of
a duration that would exceed the shelf-life of liquid
reagents. Moreover, pipetting errors in reagent delivery
(e.g., inadvertently fail to deliver reagent to a tube/well)
and/or volume are eliminated with the lyotube/lyoplate
approach, adding safeguards against generation of erro-
neous data. We view this approach as a step forward rel-
ative to occasional reports in the literature of frozen
plates pre-loaded with stimulation reagents (6), because
lyophilization does not require a commitment to freezer
space and dry-ice shipment to an off-site clinical center.
To evaluate the potential need for a lyotube/lyoplate
in a biomarker campaign, a study was designed that
involved ex vivo stimulation of whole blood, with subse-
quent detection of an intracellular phosphoprotein.
Detection of phosphorylated epitopes is relatively new
to the field but of high value in assessing pharmacody-
namic effects of therapeutic proteins or small molecule
therapeutics because of the proximity of pharmacody-
namic signal to the target, which is often a signal trans-
ducing cell surface receptor or an overactive/
overexpressed kinase. As such, we lyophilized PMA and
ionomycin and studied their ability to induce phospho-
S6 in lymphocytes in whole blood as a means to further
develop a biomarker assay/phamacodynamic marker of
rapamycin inhibition of the mTOR pathway (7). We
found that lyophilized PMA and ionomycin could be
manufactured to produce results equivalent to liquid
reagents. This proof-of-concept effort supports the
potential utility of implementing lyophilized reagents in
settings where extended reagent stability and safeguards
against pipetting error are important. Although we ly-
ophilized PMA and ionomycin to support a specific bio-
marker development effort by our group, the utility of
lyophilized PMA and ionomycin is of widespread impor-
tance as these are prototypical stimulants used in a vari-
ety of immunological assays. Provided that lyotubes/
lyoplates can be manufactured in an economical format,
we anticipate that they may eventually replace the use
of liquid reagents in many circumstances.
MATERIALS AND METHODS
Blood Donors
Specimens of heparinized whole blood were collected
from Amgen employees with prior written consent. The
blood was held at ambient temperature for less than 3 h
before stimulation on the day of collection.
Preparation of PMA and Ionomycin Solutions
PMA and ionomycin (Sigma-Aldrich, St. Louis MO, Cat.
No. P1585 and IO623) were diluted in DMSO to stock
concentrations of 27.8 lg/mL and 1.1 mM, respectively,
and stored frozen in single-use aliquots at
20 C. The
aliquots were thawed at ambient temperature and
diluted in PBS to a working concentration immediately
before stimulation of whole blood.
Preparation of PMA and Ionomycin lyotubes
Preparation of lyophilized reagents was performed by
the Custom BioPharma group at Beckman Counter.
(Miami FL). PMA and Ionomycin (Sigma-Aldrich) were
solubilized at 1 mg/mL in either ethanol or CHAPS
buffer to create stock reagents; working dilutions (PMA
50 ng/mL; ionomycin 5 ng/mL) were made on the day
of lyophilization and combined with three proprietary
excipients; this resulted in six possible varieties of stim-
ulant, A–F (Table 1). The ethanol solubilized formula-
tions (A and B) were less successful in lyophilization
than CHAPS solubilized formulations as determined by
visual inspection acceptance criteria. Lyophilates C, D,
E, and F were further characterized in this study.
The process involved lyophilization of 20 lL drops
of aqueous liquid (size range 0.376–0.907 cm diameter)
with no added colorant on metal trays. The trays were
hermetically sealed before removal from the lyophilizer
and held at 4 C until being equilibrated to room
Cytometry Part B: Clinical Cytometry
 LYMPHOCYTES IN WHOLE BLOOD FOR IMMUNOLOGICAL ASSAYS
temperature. The resulting lyophilized pellets were
packaged into 30 mL conical centrifuge tubes (Sarstedt
AG & Co., Newton NC), and sealed into foil pouches
with desiccant pads. Pellets were transferred to the
tubes in a special nitrogen-purged dry chamber (<2%
relative humidity) so as not to introduce de-stabilizing
moisture to the final product. The sealed foil pouches
were stored in the laboratory at ambient temperature
until use (<2 months). Once the pouch was opened,
the product was solubilized in 1 mL of whole blood
within 30 minutes.
Stimulation and Processing of Whole Blood
The blood was stimulated on the same day it was col-
lected. When liquid reagents were used, a 100 lL vol-
ume of PMA and ionomycin was delivered to 50 mL
polypropylene tubes at the concentrations indicated in
the text and then mixed with 1 mL of blood. The tubes
were placed in a water bath at 37 C for 60 min and
gently vortexed every 20 min. Next, 20 mL of pre-
warmed 1 PhosFlowTM Fix/Lyse buffer (Becton Dickin-
son Immunocytometry Systems; BDIS, Mountain View
CA) was added, the tubes were mixed by gentle inver-
sion and finally returned to the water bath for 10 addi-
tional minutes. When PMA and ionomycin lyotubes
were used for stimulation, the only difference to the
above procedure was addition of blood directly to the
30-mL lyotube. In this report, specimens were either im-
mediately processed and analyzed immediately or proc-
essed into the fix/lyse buffer, flash-forzen in dry ice and
stored at
80 C for later processing and analysis. The
ability to freeze specimens could be leveraged in the set-
ting of a clinical trial, where stimulation could occur at
the local site and analysis could be performed in a cen-
tral laboratory.
To prepare the cells for analysis, the specimens were
centrifuged for 5 min at 300g at 4 C, and then washed
twice in 10 mL cold staining buffer [Stain Buffer (FBS),
BDIS]. Cells were permeabilized by addition of 1 mL of
Permeabilization Buffer (BD Biosciences) while vortex-
ing, and the specimens were then incubated for 30 min
on ice. The cells were washed twice with 10 mL vol-
umes of cold staining buffer and then resuspended in 50
lL of staining buffer. The cells were labeled with 10 lL
of AlexaFluor 488-conjugated rabbit anti-human phos-
phorylated S6 ribosomal protein antibody (catalog no.
4854; Cell Signaling Technology, Danvers, MA), and
placed on an orbital shaker (Bellco Glass, Vineland, NJ)
at medium speed for 60 min. The cells were then
washed in 2 mL of cold staining buffer and resuspended
in 250–500 lL of staining buffer.
Flow Cytometry
The cells were analyzed on a FACSCaliburTM (BDIS)
within 4 h of labeling. Lymphocytes were light-scatter
gated according to their unique scatter properties (8).
The fluorescence intensity for the anti-human phospho-
rylated S6 ribosomal protein antibody was reported as
the assay readout. Basal fluorescence intensity in unsti-
Cytometry Part B: Clinical Cytometry
61
mulated cells was minimal (as shown in Figs. 1 and 3)
and served as the negative control for the assay. Fluores-
cence intensity on the FACSCalibur was standardized
with chicken red blood cells (Biosure, Grass Valley CA),
according to the manufacturer instructions.
RESULTS
Previous data from our group showed that exposure
of whole blood to liquid formulations of PMA/ionomycin
led to a dose-dependent induction of S6 phosphorylation
in lymphocytes and that rapamycin exerted a dose-de-
pendent antagonistic effect in this model (7). With this
historical backdrop, we decided to continue our focus
on this biologic axis to understand the utility of a lyophi-
lized reagent with the intent that this may serve as a
generic example for those interested in lyophilized mate-
rials for these and similar applications (e.g., cytokine
expression). To further refine the assay for potential
application as a clinical biomarker, two stimulant con-
centrations were initially tested on two serial specimens
of blood collected 1 week apart from seven volunteers
(Fig. 1). This experiment recapitulated our original
work, showing that a reliable signal could be generated
that was proportional to the concentration of stimulant
added to the whole blood. The assay readout was inde-
pendent of whether the stimulated specimens were ana-
lyzed immediately (Fig. 1A) or frozen for analysis at a
later time (Fig. 1B). Moreover, this experiment provided
data to show that the phospho-S6 signal may be suffi-
ciently stable (within person) to permit a legitimate
comparison of a baseline measurement to a post-treat-
ment measurement (e.g., after treatment with rapamy-
cin) in a person enrolled in a clinical trial. That is, the
ability to detect a change in phospho-S6 is a function of
the longitudinal stability of the signal in a given person.
A concordance in phospho-S6 induction was observed
when liquid reagents were benchmarked against a vari-
ety of lyophilized formulations of PMA/ionomycin (Fig.
2A). Although a relatively higher concentration (10)
was required to reach equivalence to the 25 ng/mL
PMA, 500 pM ionomycin used in the liquid format, this
higher dose was nontoxic as demonstrated by light scat-
ter properties (Fig. 2B) and did not impose manufactur-
ing constraints that would preclude production as an
analytical reagent. Overall, it was found that the stimula-
tion effect was successful in each of the excipient
formulations tested, allowing for flexibility in manufac-
turing these reagents in large-scale batches.
The specificity and overall integrity of this system was
evaluated by stimulating blood from two normal donors
in the presence or absence of rapamycin (Fig. 3). This
experiment further illustrated that a lyophilized prepara-
tion of PMA/ionomycin provides results equivalent to a
freshly prepared liquid formulation and that the intensity
of signal is independent of whether the specimens were
analyzed immediately or frozen for analysis at a later
time.
62 BELOUSKI ET AL.

FIG. 1. Measurement of intracellular phospho-S6 in whole blood

stimulated with liquid PMA and ionomycin. Blood from seven normal
volunteers was assayed in triplicate from two donations spaced 1
week apart. The mean and standard deviation of the triplicate fluores-
cence intensity values for each visit of the two visits are represented
for unstimulated (open circles), low stimulation (12.5 ng/mL PMA, 1
mM ionomycin; blue circles) and high stimulation (25 ng/mL PMA,
500 pM ionomycin; red circles). Black, blue and red reference lines
represent the overall mean of all data for no, low and high stimulation,
respectively. Panel B shows that these specimens can be frozen and
analyzed at a later time with no appreciable change in fluorescence in-
tensity. Note that in some cases, data from each of the two visits
resulted in overlap of the datapoints such that only one symbol can be
discerned.

DISCUSSION
Clinical trials have experienced an increasing reliance
on blood that is manipulated at a local site for the pur-
pose of biomarker or surrogate marker assessment. To
the fullest extent possible, these assays should be
designed to be simple, easy to use, self-contained, and
mistake-proof. Lyophilization gives unstable chemical
and protein solutions a long shelf life at room tempera-
ture and excellent solubility for rapid reconstitution.
Over the years, kit manufacturers have explored alter-
native types of containers for the processing and deliv-
ery of lyophilized (freeze-dried) reagents and have not
had success with vessels other than the traditional glass
vials with slotted rubber septa, which meets the rigid
vessel requirements essential to the lyophilization pro-
cess. Glass vials have excellent thermal conductivity
properties, can be tightly sealed at the end of the lyophi-
lization cycle, and provide an effective moisture barrier.
The goal here was to increase the usability of lyophi-
lized products and thus utilization of nontraditional con-
tainers was implemented. Three major points were
considered in this effort: (1) was the product high qual-
ity freeze-dried or simply vacuum-dried, (2) what was
FIG. 2. Calibration of lyophilized PMA and ionomycin to liquid
reagents with respect to the phospho-S6 readout. Four lyophilized for-
mulations (C, D, E, and F), at 1 (red), 2 (teal), 10 (green), and
20 (black) concentrations (relative to 25 ng/mL PMA, 500 pM iono-
mycin, high concentration) were tested for their ability to induce phos-
pho-S6 in lymphocytes in whole blood from four normal volunteers.
The mean and standard deviation of the triplicate fluorescence inten-
sity values for these four donors are represented (A). The dotted black
line serves to reference the level of stimulation with liquid PMA (25
ng/mL) and ionomycin (500 pM), where 100% is indicative of an
equivalent response (between liquid and lyophilate). Representative
scatter plots for each formulation demonstrate the absence of toxicity
to the cells even at the 20 concentration (B).
the shelf-life of the product, and (3) how does the ly-
ophilization process affect the activity of the reagents?
A refined and tightly controlled process of lyophiliza-
tion of individual 25–250 ul drops of aqueous liquid
(size range 0.376–0.907 cm diameter), allows delivery
and storage in vessels that are not limited to traditional
glass-stoppered containers. The reagents are formulated
exactly as if they were being lyophilized conventionally
in a glass vial, except the concentrations are usually

Cytometry Part B: Clinical Cytometry

 LYMPHOCYTES IN WHOLE BLOOD FOR IMMUNOLOGICAL ASSAYS 63

FIG. 3. Rapamycin-mediated antagonism of PMA/ionomycin-induced phospho-S6. As indicated in the figure, blood from normal donors (n ¼ 2) was
stimulated with liquid or lyophilized PMA/ionomycin and either analyzed immediately or after the specimens were frozen. Doses of PMA/ionomycin
were adjusted according to Figure 2 (25 ng/mL PMA, 500 pM ionomycin or the 10 lyophilized equivalent). Red tracings represent fluorescence in-
tensity signal from the stimulated blood; blue tracings represent the signal of blood stimulated in the presence of rapamycin, and black tracings rep-
resent signal from unstimulated, nonrapamycin-treated controls. The histograms of both donors are presented as overlays on a single plot to depict
the consistency of signal that can be observed in this model as well as the ability of rapamycin to reproducibly antagonize the stimulation effect.

higher due to the very small pellet volumes, thus the
pellet has a high chemical-to-excipient ratio. The self-
contained lyophilized pellets allow several components
that are incompatible as liquids to be either lyophilized
independently and then stored together in a single ves-
sel, or premixed in precise amounts before lyophiliza-
tion while maintaining their integrity and enhancing
stability. In addition to visual size inspection of the ly-
ophilized pellets, coloring agents could be added as a
convenient quality assurance step, adding value if differ-
ent concentrations of (e.g., stimulation) reagents would
be used in a clinical study, i.e., the concentrations would
be differentiated by the color of the pellet in addition to
the package. One can imagine several approaches to
incorporate internal quality controls to monitor perform-
ance of the assay in a clinical setting.
Although the ability to lyophilize antibodies and other
reagents for use in immunological assays is not new, con-
straints in the vessel type (e.g., glass vial) was viewed as
an inconvenience and the reagents were never commer-
cially developed and manufactured in a cocktail format or
in pre-loaded plates/tubes. Until recently, bulk reagents in
a liquid formulation dominated the marketplace, primarily
because there are not enough advantages offered from
rehydrating a bulk lyophilized reagent to essentially make
it the equivalent of the liquid formulation. Willingness of
reagent manufacturers to create custom antibody or rea-
gent cocktails combined with the flexibility of pre-loaded
formats has renewed interest in this method, particularly
in instances where clinical study endpoints could be
compromised by instable reagents and or human error
introduced during set-up for the analytical procedure.
Ultimately, commercialization approaches for lyophilized
reagents are not expected to deviate significantly from
those employed for traditional buffers and reagent sets.
Practical questions regarding qualitative appearance and
dissolution properties of the lyo-cake, ease of lyophiliza-
tion and excipient formulation are all important issues to
be addressed. Not unlike the variety of commercial lysis
buffers and fixatives relied upon in our industry, it is likely
that the exact details surrounding a commercial lyophilate
will no doubt remain proprietary to the provider, with
ease of use and the potential to better control sources of
technical variability of the data ultimately governing com-
mercial success. With these proprietary rights, vendors
will need to provide methods and characterization of util-
ity (eg, phosphoprotein stimulation and cytokine expres-
sion profiles) as well as long term and temperature
stability to fully realize the market opportunity.
PMA and ionomycin are cornerstone stimulants utilized
in a variety of immunologic assays and we envision that
this reagent could ultimately be employed in any setting
where this axis of stimulation is important. The experi-
mental approach of this study was feasible, as indicated in
Figure 2, and the possibility that whole blood stimulations
could be conducted at local laboratories without concern
that degraded reagents might be used or that pipetting
errors might introduce unwanted variability into the data-
set is very attractive. The ability to freeze the stimulated
specimens (Figs. 1 and 3) for transport to a central labora-
tory is a component of this model that could be exploited
to further minimize variability, especially if multiple clini-
cal sites are involved. Finally, the variety of manufacturing
options that could be considered for these reagents,
including a blister-packaged pellet, would provide a high
degree of flexibility for small scale experiments, while
retaining the stability profile of the lyophilized reagent(s).
As noted elsewhere in this manuscript, the lyophilized

Cytometry Part B: Clinical Cytometry

64 BELOUSKI ET AL.
pellets are manufactured according to the needs of the in-
vestigator and project and can be provided in any type of
plate, tube or other vessel.
The field of translational medicine will no doubt con-
tinue to evolve, mature and further leverage flow cytom-
etry for biomarker and surrogate marker endpoints. It is
anticipated that variability and error could be controlled
and minimized as parallel advances in reagent formula-
tion and experimental design are thoughtfully incorpo-
rated into biomarker campaigns where specimen
manipulation is an important component of the study.
ACKNOWLEDGMENTS
The authors thank Lisa Uskali and the Amgen Depart-
ment of Occupational Health for soliciting normal blood
donors, providing informed consent and collecting the
blood specimens for the study. They also thank Brian
Kotzin (Vice President, Amgen Medical Sciences), Scott
D. Patterson and Steven J. Swanson (Executive Directors,
Amgen Medical Sciences) for support of this project.
LITERATURE CITED
1. Giorgi JV, Cheng HL, Margolick JB, Bauer KD, Ferbas J, Waxdal M,
Schmid I, Hultin LE, Jackson AL, Park L, Taylor JMG, Multicenter
AIDS Cohort Study Group. Quality control in the flow cytometric
measurement of T-lymphocyte subsets: The multicenter AIDS cohort
study experience. Clin Immunol Immunopathol 1990;55:173–186.
2. Schenker EL, Hultin LE, Bauer KD, Ferbas J, Margolick JB, Giorgi JV.
Evaluation of a dual-color flow cytometry immunophenotyping panel
in a multicenter quality assurance program. Cytometry 1993;14:
307–317.
3. Schmidt R, Hecker E. Autoxidation of phorbol esters under normal
storage conditions. Cancer Res 1975;35:1375–1377.
4. Liu WC, Slusarchyk DS, Astle G, Trejo WH, Brown WE, Meyers E.
Ionomycin, a new polyether antibiotic. J Antibiot 1978;31:815–819.
5. Maecker H, Rinfret A, D’souza P, Darden J, Roig E, Landry C, Hayes P,
Birungi J, Anzala O, Garcia M, Harari A, Frank I, Baydo R, Baker M,
Holbrook J, Ottinger J, Lamoreaux L, Epling CL, Sinclair E, Suni M,
Punt K, Calarota S, El-Bahi S, Alter G, Maila H, Kuta E, Cox J, Gray C,
Altfeld M, Nougarede N, Boyer J, Tussey L, Tobery T, Bredt B, Roe-
derer M, Koup R, Maino V, Weinhold K, Pantaleo G, Gilmour J, Hor-
ton H, Sekaly R. Standardization of cytokine flow cytometry assays.
BMC Immunol 2005;6:13.
6. Reimann KA, Chernoff M, Wilkening CL, Nickerson CE, Landay AL.
The ACTG immunology advanced technology laboratories: Preserva-
tion of lymphocyte immunophenotype and proliferative responses in
cryopreserved peripheral blood mononuclear cells from human im-
munodeficiency virus type 1-infected donors: Implications for multi-
center clinical trials. Clin Diagn Lab Immunol 2000;7:352–359.
7. Mccaffery I, Fitzpatrick VD, Wang SW, Rossi J, Bao H, Suggs SV, Bao
H, Ferbas J, Patterson SD.Clinical Proteomics. From Diagnosis to
Therapy, Weinheim: Wiley VCH Verlag GmbH & Co KGaA; 2008.
633.
8. Hoffman RA, Hansen WP. Immunofluorescent analysis of blood cells
by flow cytometry. Int J Immunopharmacology 1981;3:249–254.
Cytometry Part B: Clinical Cytometry