Upload
doanthien
View
220
Download
0
Embed Size (px)
Citation preview
Practical Analysis of Stem Cells by Flow Cytometry
Established Methods and Emerging Technologies
Bill Telford
Experimental Transplantation and Immunology BranchFlow Cytometry Laboratory
National Cancer InstituteNational Institutes of Health
Flow Cytometry in Stem Cell Biology
Flow cytometry and cell sorting are absolutely indispensable techniquesfor both the identification and isolation of embryonic and adult stem cells
in both bone marrow and other tissues.
• Fluorescent immunophenotyping for stem cell markers
• Aldehyde dehydrogenase activity detection usingfluorogenic substrates
• G0/G1 discrimination using Hoechst 33342/pyronin Y
• Hoechst side population analysis
Quiescent SCCD34-c-Kit+CD34-c-Kit-
Stem cell subpopulations and their identificationHematopoetic stem cells (HSCs)
Sel
f-ren
ewal
Turnover time: 30-60 days
Activated SC CD34+ c-Kit -
Sel
f-ren
ewal
More frequent turnover
Circulating CD34+ Stem Cell
CD34+ c-Kit+Progenitor Cell
Immature differentiating cells
From DS Donnely DS and Krause DS, Leukemia and Lymphoma, 40, 221-234 (2001)
Isolation of Stem Cells by Flow Cytometry
As with most things in biology, there is no single characteristicthat adequately identifies stem cells by itself.
We need to look at multiple phenotypic characteristics includingcell surface markers and biochemical / physiological characteristics.
And, if possible, we need to look at these multiple characteristicsin a single multiparametric assay.
And of course, flow cytometry is an ideal way to do this!
R1
R2
R3
R4
R5
FSC
SS
C
Bone Marrow Cell Differentiation by Scatter Measurement
RBCs(Ter119)
granulocytes/monocytes(Gr-1/Mac-1)
stem cells / progenitors
mature lymphocytes
(CD3/B220)
Phenotypic markers for mammalian hematopoetic stem cells
• Human stem cells
• Lineage depletion • CD34• HLA-DR• AC133 (CD133)• CD7
• Rodent stem cells
• Lineage depletion • CD34 • Sca-1• c-kit (CD117)• MAC-1 (+/-)
Phenotypic markers for mammalian stem cells
CD34
• A single chain transmembrane glycoprotein expressed on HSCs, vascular endothelium, embryonic fibroblasts and some cells in fetal and adult nervous tissue
• Promotes adhesive interactions between stem cells and stromalelements in the bone marrow
• Probably regulates association between stems cells and the “niche”microenvironment via indirect regulation of other adhesion factors
• The dominant marker for clinical identification and separation ofstem cells
Phenotypic markers for mammalian stem cells
Sca-1
• Stem cell antigen, a GPI-linked surface protein expressed on 30% of mouse bone marrow including pluripotent HSC and committed lymphoid and myeloid progenitors, also was found in osteoblastskidney apithelia(Ly-6A/E).
• HSCs from Sca-1 KO mice have impaired repopulation potential.
• Lower engraftment of secondary transplants in Sca-1 deficient animals also suggests a defect in HSC self-renewal.
• In addition to HSC deficiencies in Sca-1-deficient mice, specific cell lineages and progenitor subpopulations are also affected. (Ito CY, Li CY, Bernstein A, Dick JE, Stanford WL. Blood 101, 517-23, 2002).
Phenotypic markers for mammalian stem cells
c-kit (steel factor)
• Transmembrane glycoprotein, PTK-R in HSC, was found in melanocytes and primordial germ cells
• Mice lacking the receptor tyrosine kinase c-Kit (c-KitW/W) have hematopoietic defects causing perinatal death
• A viable c-KitW/W mouse shows an age-dependent, progressive decline of pro-T and pro-B cells accompanied by loss of common lymphoid progenitors in the bone marrow in adult mice lacking c-kit.
• Adult c-KitW/W hematopoietic stem cells can engraft in host bone marrow but fail to radioprotect, form spleen colonies, or establish sustained lymphopoiesis (Waskow C, Paul S, Haller C, GassmannM, Rodewald H., Immunity 3, 277-88, 2003)
Benchtop analyzers
• BD FACScan – one laser, three colors
• BD FACScalibur – two lasers, four colors
• Beckman Coulter EPICS XL – one laserfour colors
PE-Cy5-Sca-1
AP
C-C
-kit
Sca-1only
c-kit only
Sca-1 and c-kit expression in mouse bone marrowThree color analysis
APC-c-kit
FITC
-line
age
Lineage depletion with biotinylatedantibodies against B220, CD3, Ly6C+G,GR-1, Ter119, CD5 and NK1.1
FITC “dump” channel
APC-anti-c-kitPE-Cy5-anti-Sca-1FITC-Lineage
side scatter
PE-C
y7-L
inea
ge
FITC-c-kit
PE
-Cy5
.5-S
ca-1
PE-CD34
side
sca
tter
side
sca
tter
forward scatter forward scatter
PE-Cy7 labeling for lineage panel(CD3, B220, Ly6C + G, CD11b, Ter119, NK1.1, CD5)
FITC-c-kitPE-CD34PE-Cy5-Sca-1PE-Cy7-Lineage
Sca-1, c-kit and CD34 expression in mouse bone marrowFour color analysis
Discrimination of the Hoechst SP on the flow cytometer
Samples from Drs. Atsushi Terasumi and John Vogel, DB, NCI, NIH
Hoechst 33342 Red(675/20 nm)
Hoe
chst
333
42 B
lue
(440
/10)
Murine bone marrowunseparated(150,000 events)
0.04%
• When loaded with the fluorescent DNA dye Hoechst 33342, murinebone marrow stem cells preferentially pump out the dye via the ABCG2 membrane transporter.
• These “side population” (SP) cells can be distinguished onthe basis of their reduced Hoechst dye fluorescence(Goodell et al., JEM 183, 1797 (1996)).
• Both DNA-bound (Hoechst blue fluorescence) and unbound(Hoechst red fluorescence) used to distinguish side population cells.
• SP cells are enriched for Sca-1 expression and show nolineage marker expression. SP cells can reconstitute irradiated mice with both lymphoid and myeloid lineages.
Mouse L929Hoechst 33342 2 :g/ml
BD FACSVantage SE
Discrimination of the Hoechst SP on the flow cytometer
krypton-ion UV (tertiary)laser path
split mirror
argon-ion 488 nm(primary)laser path
675/20
440/
10HOblue
HOred
Mouse L929Hoechst 33342 2 :g/ml
Verapamil blocks the accumulation of Hoechst SP cells
Samples from Drs. Atsushi Terasumi and John Vogel, DB, NCI, NIH
Hoe
chst
333
42 B
lue
(440
nm
)
Hoechst 33342 Red (675 nm)
verapamil 10 :M30 m preincubationno priorenrichment
0.03%0.11%
no verapamilno priorenrichment
Negative selection of lineage-positive cells by magnetic bead depletion
Label bone marrow with haptenor PE-conjugated antibodiesagainst lineage markers
CD3B220Ly6C + GGr-1Ter119NK1.1CD5 PE
Discrimination of Hoechst SP on the flow cytometer
Murine bone marrowLineage+ antibodyretained fraction
Murine bone marrowLineage+ antibody columnflow through cell fraction
Samples from Drs. Paul Love and Ella Frolova, NICHD, NIH
Hoe
chst
333
42 B
lue
(440
nm
)
Hoechst 33342 Red (675 nm)
Lineage panel includes CD3, B220,CD11b, Ly6G, Ter-119
krypton-ion UV (tertiary)laser path
560
SP
575/
26
split mirror
Sca-1
argon-ion 488 nm(primary)laser path
Lin530/30
675/20
440/
10HOblue
HOred
Discrimination of Hoechst SP on the flow cytometer
Lineage “dump” channel(CD3, B220, Ly6C + G, CD11b, Ter119, NK1.1, CD5)
It is very possible (and highly desirable) to combine Hoechst SPanalysis and cell surface immunophenotyping.
Lineage marker expression in Hoechst SP cells
Lin expression
Lin+ fraction
Cells with increasing SP phenotypeshow decreased Lin expression
Hoe
chst
333
42 B
lue
(440
nm
)
Hoechst 33342 Red (675 nm)
Murine bone marrowLineage+ antibody columnflow through cell fraction
all cells
lineagenegative
Lineage exclusion and Hoechst SPanalysis in mouse bone marrow
side scatter
PE
-Cy7
-Lin
eage
PE-Cy7 labeling for lineage panel(CD3, B220, Ly6C + G, CD11b, Ter119, NK1.1, CD5)
Lineage exclusion enriches for stem cells, but is insufficient alone for good isolation.
4.2%
25.3%
63.9%
6.5%
Sca-1 expression in SP subpopulation cells
Sca-1 expression
Cells with increasing SP phenotypeare enriched for Sca-1 expressing cells
Hoe
chst
333
42 B
lue
(440
nm
)
Hoechst 33342 Red (675 nm)
Murine bone marrowLineage+ antibody columnflow through cell fraction
Lineage exclusion, Sca-1 and c-kit immunophenotyping in mouse bonemarrow Hoechst SP analysis
PE-Cy5-Sca-1
AP
C-C
-kit
Sca-1only
c-kit only
all cells
lineagenegative
lineagenegative
Sca-1+c-kit+
PE-Cy7 labeling for lineage panel(CD3, B220, Ly6C + G, CD11b, Ter119, NK1.1, CD5)
Sorting Hoechst SP cellsH
oech
st 3
3342
Blu
e (4
40 n
m)
Hoechst 33342 Red (675 nm)
Murine bone marrowLineage+ antibody columnflow through cell fractionsorted
Murine bone marrowLineage+ antibody columnflow through cell fractionunsorted
Lineage panel includes CD3, B220,CD11b, Ly6G, Ter-119
Tissue and species distribution of the Hoechst SP phenotype
Mouse skeletal-muscle cells
Mouse ES
Monkey BM
Mouse BM
A wide variety of stem cell types(hematopoetic and mesenchymal,
embryonic and adult) from bothnon-primate and primate tissues
exhibit some degree of ABCdependent SP activity.
HeNe
argon-ion
krypton-ion
Laser sources on the FACSVantage SE
Equipment required for analysis of Hoechst side population…
• Large scale cell sorter (i.e. FACSVantage DiVa,Beckman-Coulter Altra or Cytomation MoFlo
• High power argon-ion or krypton-ion laser (US$ 30,000)
• Total equipment cost = US$ 400,000
The equipment for analyzing Hoechst SP is prohibitivelyexpensive for most institutions.
Blue-green 488 nm laserdetectorconfiguration
• BD LSR II, Beckman-Coulter FC500,Cytomation CyAn
• Polychromatic cell sorting using a variety of laser sources
• Up to 10 colors simultaneously using up to four lasers
Polychromatic flow cytometers
Diode-pumped solid state488 nm (blue-green)
HeNe 633 nm (red)
VLD 408 nm (violet)
Near-UV laser diode 374 nm
BD Biosciences LSR IILaser sources
Novel laser sources for flow cytometry
• Laser diodes
• Near-infrared and red• Blue• Violet• Near-ultraviolet
• Diode-pumped solid state
• Diode pumping of a solid state laser medium (such as yttrium aluminum garnet (YAG), or neodymium-YAG
• Frequency doubling or tripling of can generate interestinglaser lines for flow cytometry
• DPSS green 532 nm• DPSS blue-green 460 – 490 nm
• Mode-locked Nd-YAG frequency-tripled 355 nm UV laser (quasi-CW)
Lightwave mode-locked Nd-YAG 355 nm laser
NdYag 355 nm 22 mWBD LSR II
Hoechst red fluorescence (650 LP)
Hoe
chst
blu
e flu
ores
cenc
e (4
50/5
0)
Cost is still high (about US$ 30,000)
Laser diodes
42062083012402500
InGaN/Al2O3
InGaP/GaAs
blue/violet/near-UV diodes
red/near-IR diodes
“orange gap”
AlGaAs/GaAs
InGaAs/GaAs
InGaAsP/InP
InGaAsSb/Ga/Sb
visible rangeLaser diodes currently cover the infraredto the near-ultraviolet, with a few critical gaps.
Wavelength (nm)
Single Transverse Mode Violet Laser Diode
• can emit from 397 to 408 nm• may provide a useful near UV laser
source for both flow and laserscanning cytometry
Violet lasers on benchtop flow cytometers
BD FACSVantage SEBD LSR II
Violet laser diode 405 nm 18 mW• Solid-state violet laser diodes (VLDs) are nowstandard equipment on a wide variety of flowcytometers
• BD LSR II and FACSAria• DakoCytomation CyAn• Compucyte LSC2 and iCys
• These small, reliable laser sources havebroadened the use of violet-excitedfluorochromes such as DAPI, Cascade Blueand Pacific Blue.
Hoechst SP Analysis using a Violet Laser DiodePower Technology 408 nm 15 mW
Hoe
chst
333
42 B
lue
(440
nm
)
Hoechst 33342 Red (675 nm) Murine bone marrowNo purification
Kr MLUV 100 mWFACSVantage DiVa
Violet laser diode 408 nmBD LSR II
Violet laser diodes allow detection of the SP population,but with very low resolution.
350 400 450 500 600 650300 550Wavelength (nm)
Rel
ativ
e flu
ores
cenc
e
450/50
Hoechst 33342
Spectral Properties of Hoechst 33342
N
NH+
NCH3+NH
N
NH+
O
Cl
H
Hoechst 33342
The spectra of Hoechst33342 suggests that it
would be poorly excited by violet laser light.
SP = 1.58% SP = 0.58%
SP = 0.28%SP = 5.72%
0.31%
Hoechst redH
oech
st b
lue
FITC-Sca-1A
PC
-c-k
it
2.14%
no inhibitor fumitremorgin CSca-1 (+) c-kit (+)
gated cells
Hoe
chst
blu
e
SP = 76.5%
SP = 65.8%
Hoechst red
Violet diode 408 nm 25 mW
mouse BMno separation
mouse BMlineage depletion(SC-enriched)
Hoechst SP Analysis using a Violet Laser Diode
Violet laser diodes allow detection of the SP population,but with very low resolution.
Is good Hoechst SP resolution necessary?
Yes, it is!
Hoechst 33342 labeledbone marrow oftenproduces more thanone hypodlploid population.
We aren’t sure what theselow-blue populations are,but they are NOT stem cells.
This is especially true ofprimate bone marrow andnon-hematopoetictissues.
Suboptimal excitation ofthe Hoechst-labeled cellsoften fails to distinguishtrue SP cells from othernon-stem SP populations.
Lineage (+)Sca-1 (-)c-kit (-)CD34 (-)
Lineage (+)c-kit (+)
Hoe
chst
333
42 B
lue
(440
nm
)
Hoechst 33342 Red (675 nm)
Hoe
chst
333
42 B
lue
(440
nm
)
Hoechst 33342 Red (675 nm) Murine bone marrowNo purification
Kr MLUV 100 mWFACSVantage DiVa
Violet laser diode 401 nmBD LSR II
Even short-wavelength violet diodes do not significantly improve SP resolution
Hoechst SP Analysis using a Violet Laser DiodePower Technology 401 nm 15 mW
Near-UV laser diodes (NUVLDs)on benchtop flow cytometers
NUVLD sources can be mounted on some benchtopinstruments like the LSR II and used for previously
troublesome UV- dependent applications
Near-UV laser diodes (NUVLDs) for Hoechst 33342 side population analysis
NUVLD 374 nm 8 mWBD LSR II
Krypton-ion MLUV 100 mWBD FACSVantage DiVa
Hoechst red fluorescence (650 LP)
Hoe
chst
blu
e flu
ores
cenc
e (4
50/5
0)
NUVLDs on cuvette instruments give better SP resolution than gas lasers on stream-in-air instruments.
all three
Sca-1c-kit
CD34c-kit
c-kit onlySca-1 only CD34 only
CD34Sca-1
lineage-negative
all
Six color bone marrow stem cell analysisRequirements for simultaneous Hoechst SP analysis
forward scatter
side
sca
tter
PE
Cy5
.5-S
ca-1
PE-C
D34
FITC-c-kit
PE
-Cy7
Lin
eage
FITC-c-kit PE-Cy5.5-Sca-1
forward scatter
side
sca
tter
PE-Cy7 labeling for lineage panel(CD3, B220, Ly6C + G, CD11b, Ter119, NK1.1, CD5)
FITC-c-kitPE-CD34PE-Cy5.5-Sca-1PE-Cy7-Lineage
Hoechst red
Hoe
chst
blu
e
Cost of producing a UV laser line for Hoechst SP analysis…
Mode-locked Nd-YAG 355 nm laser
= US$ 30,000
Near-UV laser diode 375 nm laser= US$ 7,000
Krypton-ion multiline UV 361 - 365 nm laser
= US$ 30,000
Cost of instruments capable of doing Hoechst SP analysis…
BD LSR II withNUVLD or Nd-YAG laser
or
Cytomation CyAnwith Enterprise II laser
= US$ 250,000
FACSVantage withAr or Kr ion UV laser
= US$ 400,000
Still pricey!
NPE Analyzer
Next-generation flow cytometry technologystill in the advanced development stage.
Utilizes a unique Coulter sizing transducerincorporated into a flow cytometry flowcell to allow both highly accurate electroniccell sizing and fluorescence analysis.
stepper motors
for X-Y lampalignment
mercury arc lamp
Coulter transducer/flow cell
cell inlet
photodiodedetector
stepper motors
for X-Y lampalignment
NPE AnalyzerTransducer flow cell assembly
chargereservoir
mercury arclamp
chargeplate
photodiode
flow cell
mirror
injectionnozzle
chargeplate
Can we use the NPE Analyzer to measure Hoechst SP?Hg arc lamp derived UV excitation
366
405
546
313
578
435
Wavelength
Hg arc lamp excitation
405 nm violet
435 nm blue
546 nm green
We can “extract” most of the high-output linesfrom the Hg arc lamp for excitation of as varietyof fluorochromes.
The Hg lamp emits a strong UV line at 365 nm.
mirror
450/
20650 LP
linearphotodiode
mirror
450/
20
High-sensitivityPMT
Hig
h-se
nsiti
vity
PM
T
• Photodiodes have severalunique advantages as opticaldetectors, are notas sensitive as traditionalphotomultiplier tubes
• NPE can be equipped with high-sensitivity photomultipliertubes sensitive to as few asa few hundred photons per total detector area
• NPE set up a custom opticalbench for us with two detectorpositions, both with PMTs.
Can we use the NPE Analyzer to measure Hoechst SP?Optical layout
(-)
(-)(-)
(-)
(+)
(+)
(-)(-)
(-)
(-)(-)
(-)
(+)
(+)
(+)(+)
(+)
Coulter sizing provides a far more accurate measurement of particle size thantraditional forward scatter measurement by flow cytometry
The NPE Analyzer works on the same principle as the Coulter Counter, via the electrical resistance generated across an orifice by the occlusion of a particle.
Can we use the NPE Analyzer to measure Hoechst SP?Electronic cell volume
Can we use the NPE Analyzer to measure Hoechst SP?Electronic volume measurement
2.0 µm3.2 µm
4.5 µm5.6 µm
5.6 µm
6.0 µm 6.4 µm
The NPE Analyzer can Measure electronic cell volume with a highdegree of precision.
Measurement of stemcell electronic volumemight provide a valuablenew phenotypic markerfor stem cell-ness (likescatter does).
forward scatter
side
sca
tter
electronic cell volumeH
O33
342
blue
mirror
450/
20HOred
HOred
675/20
610 SP
krypton-ion UV (tertiary)laser path
mirror
argon-ion 488 nm(primary)laser path
675/2044
0/10HO
blue
HOred
610 SP
Hoe
chst
333
42 B
lue
(440
/10)
SP population= 0.16%
SP population= 0.21%
Hoechst 33342 Red (675/20 nm)
SP population= 0.16%
NPE Analyzer
FACSVantage SE
NPE Analyzer
FACSVantage SE
With assistance from Drs. Paul Love and Ella Frolova, NICHD
Can we use the NPE Analyzer to measure Hoechst SP?
Can we use the NPE Analyzer to measure Hoechst SP?
Yes, we can.
But…the same problem as with the violet diode, namely suboptimalexcitation. Lower precision … and poorer “junk” separation … than morepowerful UV sources.
The Hg arc lamp is a thoeretical point source – the actual power levelof laser light reaching the flow cell is probably less than 1 mW, makingit marginal for reproducible SP analysis (we have the same problem withlow-power NUVLDs on cuvette flow cytometers)
This is plenty of UV light for DAPI cell cycle…
…but less than optimal for applications requiring stronger UV excitation.
mercury arclamp
injectionnozzle
flow cell
NUVLD
dichroicmirror
Mounting a NUVLD laser on the NPE Analyzer
• A NUVLD can be mounted inthe NPE Analyzer and thebeam steered to the flow cell
mirror
450/
10HOblue
HOred
675/20
610 SP
Mounting a NUVLD laser on the NPE Analyzer
dichroic mirrornear-UV diode laser
Mounting a NUVLD laser on the NPE AnalyzerMounting a NUVLD laser on the NPE Analyzer
dichroic mirror
near-UV diode laser
Hoechst SP on the NPE Analyzer
NPE AnalyzerHg arc lamp 365 nm80 WHO blue = 450/20HO red = 675/20
NPE AnalyzerNUVLD 374 nm7 mWHO blue = 450/20HO red = 675/20
NUVLDs not only give better Hoechst SP resolution, they give greater contrast between the SP population and other non-stem cell hypodiploid populations
Hoechst red
Hoe
chst
blu
e
Today’s Stem Cell Wet Workshop…
We will label whole mouse bone marrow cells and A549 cells (an ABCG2 overexpressing lung carcinoma cell line) with Hoechst 33342 for Hoechst SP analysis.
We will then analyze these cells on the NPE Analyzer, first with themercury arc lamp UV source, then with the NUVLD.
This is a very new and continually evolving technique … any resultswe obtain today may have real experimental value.
Acknowledgements
Atsushi TeranumiJohn VogelMark Udey
Dermatology Branch, NCI
Veena Kapoor
NCI ETIB Flow Lab
Michael BustinElla Frolova
Molecular Biology Branch, NCI
NPE Systems
Richard ThomasErnie ThomasRaquel CabanaMichael Brochu, Sr. Michael Brochu, Jr.
Larry DuckettJoe Trotter
BD Biosciences
James Jackson
Power Technology
Awtar Krishan
The University of MiamiMedical School
Susan BatesRobert Robey
Developmental TherapeuticsBranch, NCI
Visit our WWW site at…
http://home.ncifcrf.gov/ccr/flowcore/index.htm