Side Population Analysis Using a Violet-Excited Cell-Permeable DNA Binding Dye - 《stemcells》干细胞学杂志干细胞之家



免疫细胞治疗专区	欢迎关注干细胞微信公众号

12 3 4 5 6 7 8 9 10 ... 23 下一页

返回列表

查看: 388188\|回复: 226	go Side Population Analysis Using a Violet-Excited Cell-Permeable DNA Binding Dye [复制链接]

江边孤钓

版主

Rank: 7 Rank: 7 Rank: 7

积分: 0
威望: 0
包包: 483

楼主

发表于 2009-3-5 00:55 |只看该作者 |倒序浏览 |打印

作者：William G. Telforda, Jolene Bradfordb, William Godfreyb, Robert W. Robeyc, Susan E. Batesc作者单位：aExperimental Transplantation and Immunology Branch andcCancer Therapeutics Branch, National Cancer Institute, National Institutes of Health, Bethesda, Maryland, USA; % O. w4 J$ a& k0 W( P& D
   ( a: F; Z$ Y9 M5 _

      1 {" s: m2 X, {; ^" X# Q


         % U$ d0 v+ C1 Q6 K1 K! ?
         2 [; Z, W# N3 q3 C5 Z

                  5 c3 b, F$ e) y' f  j/ H" _
      , V7 d# h# M- \' f5 t  k/ Y
      9 X) A0 {8 m4 [3 u
      ! W: z. k7 i5 e1 p, y; U8 ?# p
      【摘要】
   Hoechst 33342 side population (SP) analysis is a common method for identifying stem cells in mammalian hematopoietic and nonhematopoietic tissues. Although widely employed for stem cell analysis, this method requires an ultraviolet (UV) laser to excite Hoechst 33342. Flow cytometers equipped with UV sources are not common because of the cost of both the laser and optics that can transmit light UV light. Violet laser sources are inexpensive and are now common fixtures on flow cytometers, but have been previously shown to provide insufficient Hoechst dye excitation for consistent resolution of SP cells. One solution to this problem is to identify additional fluorescent substrates with the same pump specificity as Hoechst 33342, but with better violet excitation characteristics. DyeCycle Violet reagent has emission characteristics similar to those of Hoechst 33342, but with a longer wavelength excitation maxima (369 nm). When this dye is loaded into hematopoietic cells, a sharply resolved side population was also observed, similar in appearance to that seen with Hoechst 33342. Unlike Hoechst SP, DCV SP was similar in appearance with both violet and UV excitation. DCV SP could be inhibited fumitremorgin C, and showed the same membrane pump specificity as Hoechst 33342. Simultaneous immunophenotyping with stem cell markers in mouse bone marrow demonstrated that DCV SP was restricted to the stem cell lineage¨C Sca-1  c-kit  cells population, as is Hoechst SP. Pending confirmation by functional analysis of DCV SP cells, these results suggest that DCV efflux identified approximately the same stem cell population as did Hoechst 33342 efflux. Substituting DCV for Hoechst 33342 in the SP technique may, therefore, allow side population analysis on flow cytometers with violet lasers.
+ U6 o( c. p/ m, h2 b/ j
Disclosure of potential conflicts of interest is found at the end of this article. ) N+ a7 `. }% B3 S5 r. T/ H% H5 J
      【关键词】 Stem cell Hoechst Side population DyeCycle Bone marrow Cord blood
   INTRODUCTION# A3 R+ X8 v/ ~! f6 Y
+ ~* H  L8 Q2 u" F# ~8 H
The Hoechst side population (SP) technique is a critical method for identifying stem cells and early progenitors in rodent, nonhuman primate, and human hematopoietic and nonhematopoietic tissues .
% z4 i1 }" g0 i, U0 E, O
Although Hoechst SP is an extremely useful for identifying stem and progenitor populations, its widespread use is hampered by its requirement for an ultraviolet laser source. Flow cytometers with UV lasers are not common; traditionally, UV laser light has only been available from large ion lasers such as argon or krypton sources, which are large and expensive, and require considerable maintenance. Large-cell sorters were the only instruments that could accommodate such lasers, making the analysis of Hoechst SP inaccessible to many laboratories. More recently, smaller UV lasers such as diode and Nd:YVO4 solid-state sources have become available, and have been incorporated into smaller benchtop flow cytometers, increasing the number of instruments capable of Hoechst SP analysis . Nevertheless, most benchtop instruments are not equipped with laser-focusing optics that can transmit UV light, making it difficult to retrofit existing instrumentation for UV excitation. Flow cytometers that can accommodate UV excitation, therefore, remain uncommon, and are considerably more expensive than are conventional cytometers. In contrast, violet laser diodes (VLDs) have become common fixtures on flow cytometers. The emission of violet diodes ranges from approximately 395¨C415 nm; conventional glass laser focusing optics can, therefore, transmit violet laser diode light, in contrast to UV. VLDs are therefore simpler to incorporate into flow cytometers than UV sources. With their relatively low cost, VLDs are now included as standard equipment on many flow cytometers.

Since VLDs are now common fixtures on benchtop flow cytometers, several investigators have attempted to use them as an excitation source for Hoechst 33342 SP, as an economical alternative to UV laser-equipped flow cytometers . This latter observation is not surprising given the excitation/emission spectra for Hoechst 33342 (Fig. 1). The excitation maximum for Hoechst 33342 is approximately 350 nm; excitation at 405 nm (a common VLD wavelength), for example is predicted to be less than 5% of maximum, and even at 395 nm (the rarely attained lower wavelength limit for a VLD) is less than 7% of maximum. Laboratories that observe adequate Hoechst SP resolution using a VLD are likely using a high-power laser with a very short wavelength, and instruments with optimally aligned optics. Nevertheless, outstanding Hoechst SP resolution using a VLD is difficult to reproducibly achieve on most VLD-equipped systems. An ultraviolet laser is always preferable for consistent Hoechst SP analysis if it is available.# z1 V# k1 ?6 C6 M! l! W! D
4 x+ A+ o! H$ M. ?( @* ]  F
Figure 1. Excitation spectra of Hoechst 33342 (gray curve) and DCV (black curve). The estimated relative excitation at 405 nm for each dye is shown. Spectral data for Hoechst 33342 and DCV are courtesy of Molecular Probes Invitrogen. Abbreviation: DCV, DyeCycle Violet.; C( z+ f0 }/ X# R

For the many laboratories that do not have access to a UV-equipped flow cytometer, therefore, the problem of performing Hoechst SP analysis remains. In this technical report, we describe a novel DNA binding dye, DyeCycle Violet reagent (DCV; Molecular Probes Invitrogen Inc., Eugene, OR, http://probes.invitrogen.com), that has emission characteristics similar to Hoechst 33342, but with a longer wavelength emission maxima (approximately 369 nm; Fig. 1). This dye has cell permeability characteristics similar to those of Hoechst 33342, and was developed primarily as a violet-excited dye for vital cell cycle analysis (W. Godfrey, personal communication). When hematopoietic cells were loaded with DCV and analyzed on a VLD-equipped flow cytometer, a side population similar to the traditional Hoechst SP was observed. Unlike Hoechst 33342, DCV SP was similar in appearance using either violet or UV excitation. We have demonstrated that cells appearing in both the DCV SP and Hoechst SP population possess the lineage-negative, Sca-1  c-kit  (LSK) surface marker phenotype, and that DCV and Hoechst 33342 are effluxed by the same membrane pump mechanism. In some cases, DCV may, therefore, be able to replace Hoechst 33342 for identifying stem cells and progenitors on flow cytometers equipped with violet laser diodes. This probe should make SP analysis accessible to a far larger group of instruments, laboratories and investigators.
. E5 ?/ f! R! g0 L7 g
Methods  Z" Q# [% E- |+ H. F2 @6 u) e

Reagents2 @- N7 ?4 Z- F- m/ R( p
/ c0 r! g/ {) Z0 I+ c4 I& K' V8 u
DCV and Hoechst 33342 were obtained from Molecular Probes. Fumitremorgin C (FTC) was prepared by Thomas McCloud, Developmental Therapeutics Program, Natural Products Extraction Laboratory, National Institutes of Health (Bethesda, MD).2 G7 |9 p2 m: V/ h; E* Y
9 E- e  J  m1 }! r8 E
Cells
' V# R7 l5 ~6 P/ ]5 P% x
Male C57B/6 mice 8¨C20 weeks old were obtained from the National Cancer Institute (NCI) Laboratory Animal Facility. Mice were euthanized by gradual CO2 exposure according to NIH guidelines. Bone marrow was extracted from femurs and tibias and cryogenically preserved in 90% fetal bovine serum (FBS)/10% dimethyl sulfoxide in liquid nitrogen until use. Frozen human cord blood was obtained from AllCells LLC (Emeryville, CA, http://www.allcells.com) and thawed immediately before use. A549 lung carcinoma cells were grown in Dulbecco's modified Eagle's medium (DMEM) with 10% FBS. H460 cells were grown in Richter's medium with 10% FBS. HEK 293 cells were stably transfected with ABCG2 and were maintained in Eagle's medium with 10% FCS and G418 at 2 mg/ml .8 ^, I/ m' M+ Q) K2 F; _

Hoechst 33342 and DCV SP Labeling of Hematopoietic Cells and Cell Lines8 `' D2 @1 [: E5 `& q

Mouse and human hematopoietic cells and A549 cells were labeled with Hoechst 33342 according to the method of Goodell et al. with minor modifications . Cells were suspended in room temperature Hanks' balanced saline solution containing 2% FBS and 2 mM HEPES buffer (SP buffer) at 5 x 106 cells per milliliter. Cells were prewarmed to 37¡ãC, and Hoechst 33342 added to a final concentration of 5 µg/ml (8.1 µM), from a 1 mg/ml stock in ddH2O. Cells were incubated for 90 minutes, centrifuged, and resuspended in cold SP buffer at 5 x 106 per milliliter. Cells were then kept on ice and analyzed within 6 hours of labeling. Propidium iodide was added to the cells at 2 µg/ml immediately before analysis. For some experiments, cells were preincubated with the ABCG2 inhibitor fumitremorgin C at a concentration of 10 µg/ml at 37¡ãC for 30 minutes before Hoechst 33342 or DCV addition.- p/ [6 y& s  U3 J

Mouse and human hematopoietic cells and A549 cells were labeled with DCV using the same procedure as described herein. DCV was added to cells at a final concentration of 10 µM, from a 5 mM stock in ddH20. Titration experiments demonstrated that DCV SP was resolvable using lower concentrations of DCV (5 µM), but 10 µM gave more reproducible results over multiple experiments. SP data from multiple independent experiments with both Hoechst 33342 and DCV is shown in Table 1.

Table 1. Reproducibility of SP data from Hoechst 33342 and DCV efflux; J  ^& _7 q# C# {9 ^$ T: X
: x6 u+ I! M4 _  q9 U4 N& H. H
H460 and HEK293 cells were incubated in Richter's medium with 10% FCS with either Hoechst 33342 at 2.5 µg/ml or DCV at 10 µM in the presence or absence of 10 µM of the ABCG2 inhibitor FTC for 30 minutes at 37¡ãC. Cells were subsequently washed and allowed to incubate in fluorochrome-free medium for 60 minutes without or with FTC, and analyzed immediately.* M. I7 a! }4 D4 O9 f0 _  @/ {
: j/ n7 t3 R7 A# G( v# x4 R
Fluorescent Immunophenotyping) E/ E* }$ G' E8 o. d

Murine bone marrow was sometimes immunolabeled for stem cell and lineage markers after Hoechst 33342 or DCV labeling. After substrate labeling, cells were subsequently labeled with phycoerythrin (PE)-Cy7-conjugated anti-mouse c-kit (CD117) allophycocyanin (APC)-conjugated anti-mouse Sca-1, and biotin-conjugated antibodies against the lineage markers CD3, B220, Ly6C G, Gr-1, and Ter-119, followed by streptavidin conjugated APC-Cy7. All labelings were carried out in phosphate-buffered saline containing 2% FBS at 4¡ãC for 30 minutes each. All antibodies were obtained from BD Pharmingen (San Diego, http://www.bdbiosciences.com), eBioscience (San Diego, http://www.ebioscience.com) and BioLegend (San Diego, http://www.biolegend.com/). Fluorescein, PE, and PE-Cy5 were not used for surface marker detection due to a minor emission overlap by DCV (Fig. 2).- Z+ P7 L9 u" H" ~$ J3 l

Figure 2. Lineage ¨C Sca-1  c-kit  (LSK) phenotype of Hoechst and DCV SP cells. The same cells in Figure 4 are displayed as Hoechst or DCV SP with no phenotype gating (left column). The SP-gated populations for these cells are displayed as scatter plots for lineage versus c-kit expression (second column) and c-kit versus Sca-1 expression (right column). Percentages of SP-gated LSK cells are shown. Abbreviations: APC, allophycocyanin; DCV, DyeCycle Violet; LP, long pass; PE, phycoerythrin; SP, side population; UV, ultraviolet.8 |3 Z8 s. f. N6 X: c
9 U4 `1 \- L8 ]9 n1 t+ B% h: R; o
Flow Cytometry' Q2 V! Z2 W. }' h8 x3 d. M( L
. s1 ^. @% N& b) R/ ?6 B
Hoechst 33342- and DCV-labeled cells were subsequently analyzed on a BD LSR II cytometer (BD Biosciences) using one of the following lasers.
! X0 m: f$ Y1 Z' |
Nd:YVO4 Solid-State UV Laser. Emission was at 355 nm, power level at 22 mW (JDS Uniphase Lightwave, Milpitas, CA, http://www.jdsu.com). This high-frequency pulsed laser source (>50 MHz) functions as a quasi-continuous wave source, and was the UV source for all samples.% h! @' E* q1 d8 J# y) G
; x5 J: ]0 s, d2 l
Single-Module VLD. A single violet laser diode. Emission was at 407 nm, power level at 22 mW (Coherent, Mountain View, CA, http://www.coherentinc.com/).

Fiber-Coupled Single Module VLD. A single violet laser diode as above, but coupled to a single-mode fiber optic with collimation optics (Spectral Applied Research, Concord, ON, Canada, http://www.spectral.ca). Emission was also at 407 nm, with a power level at the fiber output of approximately 7.5 mW.: P+ Y4 m6 F2 L& X5 `

Some samples were also analyzed on a FACSVantage DiVa cytometer (BD Biosciences) equipped with a dual-module violet laser diode; this unit combines the polarized beams from two 55-mW VLDs lasers into a "single" 100-mW beam, using a polarized beamsplitter cube .

The Hoechst 33342 and DCV SP blue and red fluorescence signals from bone marrow, cord blood, and all cell lines cells were detected on the BD LSR II cytometer using a 450/50 nm bandpass and a 650 nm longpass filter, respectively, with a 580 nm longpass reflecting dichroic to separate the signals. For some experiments, SP was detected on the FACSVantage DiVa cytometer using 450/50 nm bandpass and 650 longpass filters, with a 610 nm shortpass reflecting dichroic. APC, PE-Cy7, and APC-Cy7 immunolabeling was detected using 488 and 633 nm laser excitation and 660/20 nm and 780/60 nm bandpass filters. Data from both instruments were analyzed using FlowJo for PC, version 6.0 (Tree Star Software, Ashland, CA, http://www.flowjo.com).2 E; E6 z" N& q4 I! j0 a6 q
7 h3 q8 _8 H  K" }! [! d' k( M
RESULTS- D+ N8 C; A. W: M, a+ U. m7 w
4 u% S- M1 C- J& M. G2 n
Hoechst SP identification of stem cells ideally requires a UV laser for optimal results. This can be predicted from the excitation spectra of Hoechst 33342 (Fig. 1), where its excitation maximum is approximately 350 nm, well within the range of most UV laser sources. Violet laser excitation, however, is predicted to be suboptimal for Hoechst 33342 excitation, on the basis of its emission spectra. Excitation within the range of most commercial VLDs (400¨C410 nm) is predicted to be less than 5% of the maximum. The excitation maximum for the cell-permeable DCV, however, is approximately 369 nm, with a predicted excitation of more than 20% at 405 nm. Although still not completely optimal, DCV would be expected to excite more efficiently with violet excitation than would Hoechst 33342.4 I& H% {/ i; E$ M. W* K) {9 f
7 Q4 r; ^' A- W4 ]
DCV could, therefore, theoretically substitute for Hoechst 33342 SP analysis on instruments equipped with violet laser, if (a) it were sufficiently cell-permeable, (b) it were a substrate for the same membrane efflux pump(s) as Hoechst 33342 and blocked by the same pump-specific inhibitors, and (c) it generated a recognizable SP that corresponded to the same stem cell/progenitor populations as generated with Hoechst 33342. To determine whether DCV could substitute for Hoechst 33342 in the traditional SP assay, mouse bone marrow and human cord blood were loaded with DCV at 10 µM using the same protocol (37¡ãC, 90 minutes) as previously described for Hoechst 33342. DCV-labeled cells were then analyzed on the flow cytometer using either UV or violet excitation, but using the same blue and red excitation filters traditionally used for Hoechst SP .
3 C4 M# X6 c! X5 ~# Z
The results are shown in Figure 3. Both DCV-labeled mouse bone marrow and human cord blood showed a recognizable side population when excited with either UV or violet excitation. Although the overall appearance of the DNA dye labeling profile differed somewhat between Hoechst 33342 and DCV, a clearly defined DCV SP tail population was readily apparent. More importantly, the DCV SP showed roughly equal resolution with either UV or violet excitation. This was in marked contrast to Hoechst 33342, for which resolution using the violet laser was poor compared to that using the UV. DCV, therefore, did produce a recognizable SP, and violet laser excitation was sufficient to resolve it. Titration experiments showed that 10 µM produced reasonably reproducible DCV SP appearance over multiple experiments. DCV concentrations of 5 µM still gave a resolvable SP region, but with less day-to-day reproducibility. SP reproducibility for multiple independent experiments is shown in Table 1.; j% ]5 d' k8 Y9 ]9 H& E
) b. K7 t( [3 L8 }" o
Figure 3. Hoechst 33342 and DCV side population (SP) in mouse bone marrow and human cord blood. Human cord blood (top row) or mouse bone marrow (bottom row) were incubated with either DCV (left four panels) or Hoechst 33342 (right four panels), analyzed with either UV or violet laser excitation (as indicated) and displayed as scatterplots for blue (450 nm) versus red (>650 nm) fluorescence. Percentage of estimated SP cells is indicated on each scatterplot. Abbreviations: DCV, DyeCycle Violet; LP, long pass; UV, ultraviolet.; Q- Z" |1 y5 A& F$ \8 t* D% P
6 e8 w. b: |9 n
The pump specificity of the DCV substrate was then determined using several cell lines expressing normal or transfected levels of the ABCG2 membrane transporter. Hoechst 33342 is an ABCG2 substrate, and ABCG2 has been previously demonstrated to be largely responsible for the Hoechst SP phenomenon in stem cells . DCV or Hoechst 33342 were loaded into A549 lung carcinoma and H460 cells (with high but endogenous ABCG2 expression), and HEK ABCG2 (drug-selected ABCG2 transfectants). These cell lines were then analyzed with UV excitation for efflux of both dyes (Fig. 4). In all cases, the DCV substrate was effluxed by all ABCG2 expressing cell lines, showing pump specificity identical to that shown with Hoechst 33342. The ABCG2 inhibitor fumitremorgin C was also able to inhibit both DCV and Hoechst 33342 efflux in all cell lines.
% O/ I/ F, v6 _) A0 w
Figure 4. Hoechst 33342 and DCV efflux in ABCG2 expressing cell lines. Mouse bone marrow (left two panels), A549 lung carcinoma cells (second from left), H460 cells (second from right), or HEK 293 cells (right two panels) were loaded with DCV in the presence or absence of fumitremorgin C (bottom and top rows, respectively). All cells were analyzed with ultraviolet excitation and displayed as scatterplots for blue (450 nm) versus red (>650 nm) fluorescence. Gray markers indicate the approximate boundary of efflux inhibition with fumitremorgin C. Abbreviations: DCV, DyeCycle Violet; LP, long pass.
7 H$ M9 J% Z; S$ U
Finally, Hoechst 33342 or DCV labeling were combined with fluorescent immunolabeling of mouse bone marrow stem cells to determine whether Hoechst SP and DCV SP were restricted to the same progenitor subsets. Mouse bone marrow was labeled with either Hoechst 33342 or DCV, and subsequently labeled with fluorochrome-conjugated antibodies against the lineage positive cells (B220, CD3, Gr-1, Mac-1, and Ter-119 erythroid), Sca-1, and c-kit. The cells were then analyzed with either violet or UV excitation, with simultaneous detection of the Hoechst 33342 or DCV and the stem cell markers. In Figure 5, the Hoechst 33342 or DCV fluorescence profile is displayed for all cells (left column). All of the cells were then analyzed for lineage versus c-kit expression (second column), and lineage-negative Sca-1 versus c-kit expression (third column). The resulting LSK cells are displayed for SP in the fourth column. As expected, expression of the Hoechst SP phenotype was strongly restricted to the lineage¨C Sca-1  c-kit , often termed the pluripotent progenitors (Fig. 5, top row) . In the DCV-labeled cells, the same lineage¨C Sca-1  c-kit  restriction was also precisely observed for the DCV SP (Fig. 5, middle and bottom rows). In fact, DCV SP seemed to be somewhat more strongly associated with the LSK phenotype; a somewhat larger percentage of Hoechst SP cells were non-LSK compared to DCV. This observation was difficult to confirm with certainty, however, because establishment of the SP-negative/-positive limit using inhibitors possesses a large amount of error. Nevertheless, these results strongly suggest that the Hoechst SP and DCV SP phenomena both recognized a similar subset of cells, namely the LSK subset.3 A$ @7 I2 n8 a% \$ c' I: q/ G

Figure 5. Hoechst 33342 and DCV efflux in mouse bone marrow stem cells. Mouse bone marrow was loaded with either Hoechst 33342 or DCV and subsequently labeled with lineage-specific markers, Sca-1, and c-kit, as indicated in Materials and Methods. Hoechst 33342 loaded cells were analyzed with UV excitation (top row), and DCV loaded cells with either UV (middle row) or violet excitation (bottom row). Cells are displayed as scatter plots of Hoechst or DCV side population (SP) with no phenotype gating (left column). The same cells are displayed as scatter plots for lineage versus c-kit expression (second column), and the gated lineage ¨C cells displayed for c-kit versus Sca-1 expression (third column). The Hoechst or DCV cells gated for lineage ¨C Sca-1  c-kit  (LSK) cells are shown in the right-most column. Percentages of LSK-gated Hoechst or DCV SP are shown. Abbreviations: APC, allophycocyanin; DCV, DyeCycle Violet; LP, long pass; PE, phycoerythrin; UV, ultraviolet.
$ _* d9 c, v) r3 `5 N
The inverse relationship is shown in Figure 2. Hoechst or DCV SP cells only (first column) were gated and displayed for lineage, Sca-1, and c-kit expression. Greater than 45% of both Hoechst and DCV SP cells displayed the lineage¨C Sca-1  c-kit  phenotype.# c% P6 j9 H  ~: c1 m8 O
# u$ w* R5 o& Y( Y7 ]
A minor practical drawback of DCV labeling was also observed in the context of simultaneous fluorescent immunolabeling. Hoechst 33342- or DCV-labeled mouse bone marrow analyzed at 488 nm and visualized for fluorescein (530 nm) and PE (585 nm) fluorescence is shown in Figure 6. Because of its longer-shifted excitation range, DCV exhibited a minor excitation at 488 nm that is not present with Hoechst 33342. As a result, DCV-labeled cells were somewhat fluorescent in the green to orange (520¨C600 nm) range, resulting in a high background signal that complicated simultaneous detection of fluorescein, PE, and PE-Cy5 tandem conjugate signals (right column). This phenomenon with DCV could be observed in the fluorescein versus PE scatterplots in the right column of Figure 2. This emission is not detectable with Hoechst 33342. Therefore, in Figure 5, fluorescein, PE, and PE-Cy5 were not used for surface marker labeling. Rather, the long red 488 nm-excited probe PE-Cy7 and the red laser-excited probes APC and APC-Cy7 were used to avoid this undesirable effect. DCV showed almost no detectable 488 nm-excited fluorescence beyond 650 nm and was not excited by a red laser source. This is apparent in Figure 2 (left and middle columns), where the bone marrow surface marker expression profiles using PE-Cy7, APC, and APC-Cy7 were identical in the presence of either Hoechst 33342 or DCV. These results, therefore, suggest that fluorescein, PE, and PE-Cy5 need to be avoided for surface marker labeling when using DCV, but that long red and red-excited probes including PE-Cy7, APC, APC-Cy5.5, and APC-Cy7 could be successfully substituted.

VLDs are found in several formats on modern flow cytometers. The results discussed herein were obtained using a 25-mW laser with an air-launched beam, allowing more than 80% of the total laser power to reach the cuvette-contained cell stream. Some flow cytometers now use fiber-coupled lasers for beam delivery; although this offers great flexibility in instrument design, significant light loss is inevitable with single-mode fiber optics. This is particularly true at shorter wavelengths such as violet, where more than 50% of the total beam power can be lost during fiber coupling and transmission. Instruments with fiber-coupled VLDs, therefore, provide less laser power than do traditional air-launched laser systems, raising concern about the effect of this power loss on dye excitation. In another scenario, more powerful VLDs can be incorporated into stream-in-air cell sorters, overcoming their lower signal collection efficiency with a higher power level. DCV SP was successfully resolved using both of these systems (Fig. 6, 7). Mouse bone marrow and human cord blood were again labeled with DCV and analyzed on (a) a BD LSR II cytometer with a fiber-coupled VLD, emitting at 7.5 mW postfiber (407 nm), and (b) a FACSVantage DiVa stream-in-air cell sorter with a dual-module VLD, producing a combined laser power of 100 mW (404 nm). In both cases, DCV SP could be easily resolved. DCV SP could, therefore, be detected both with lower-power fiber-coupled laser source, and with more powerful lasers using less efficient stream-in-air optics.' R  k# r$ f7 \# M0 _1 v1 E
' c% e$ i  d+ Y. c
Figure 6. DCV fluorescence with 488 nm excitation. Mouse bone marrow was loaded with either Hoechst 33342 or DCV and subsequently labeled with lineage-specific markers (allophycocyanin -Cy7), Sca-1 (PE-Cy7) and c-kit (APC) as indicated in Methods and in Figure 4. Hoechst 33342 loaded cells were analyzed with UV excitation (top row), and DCV loaded cells with either UV (middle row) or violet laser (bottom row). Cells are displayed as scatterplots for APC-Cy7-lineage versus APC-c-kit expression (left column) and PE-Cy7 Sca-1 versus APC-c-kit expression (middle column). Incidental fluorescein-range (530 nm) and PE range (580 nm) fluorescence resulting from Hoechst 33342 or DCV at 488 nm is shown in the right column, and indicated with arrows for DCV labeled samples. Abbreviations: DCV, DyeCycle Violet; FITC, fluorescein isothiocyanate; PE, phycoerythrin; UV, ultraviolet.2 a: F% w+ _/ q' q+ E1 w
2 m4 `; M2 s+ U: l
Figure 7. DCV side population (SP) with other violet laser sources. Human cord blood (top row) or mouse bone marrow (middle row) were incubated with DCV (left four panels) and analyzed with a 25-mW violet laser diode (VLD) on a BD LSR II (left column; BD Pharmingen), a fiber-coupled VLD on a BD LSR II with 7.5-mW power output post-fiber (middle column), or a 100-mW dual-module VLD on a BD FACSVantage DiVa (right column). InSpeck Blue sensitivity microphere arrays (Molecular Probes) were analyzed with each system as an internal comparison of instrument sensitivity (bottom row). Abbreviations: DCV, DyeCycle Violet; LP, long pass.

DISCUSSION

DCV produces a side population similar to that observed for Hoechst 33342 in both murine and human hematopoietic cells. DCV appears to be effluxed by the same molecular mechanisms as Hoechst 33342 and was also blocked with the ABCG2 inhibitor fumitremorgin C. Finally, DCV SP corresponded to roughly the same population of murine lineage¨C Sca-1  c-kit  stem cells as did Hoechst SP. In contrast to Hoechst 33342, DCV is more efficiently excited by violet laser diodes, allowing the use of this inexpensive and increasingly common laser and eliminating the need for UV excitation. Pending functional analysis of DCV SP cells, these results suggest that DCV may constitute a viable second HSC efflux dye, and that DCV can replace Hoechst 33342 when a UV laser source is not available.
# ?. X5 l; o( C1 S4 O; S( I
The low-level green to orange fluorescence emitted by DNA-bound DCV when it is excited at 488 nm is a disadvantage, because it limits the use of fluorescein, PE, and PE-Cy5, commonly used fluorochromes that are frequently combined with SP analysis to more thoroughly precisely discriminate stem cell and progenitor subpopulations. Although this is not a trivial problem, most violet diode-equipped flow cytometers are additionally equipped with red lasers for APC, APC tandem conjugate, and Cy5 detection. Similarly, polychromatic flow cytometers are now routinely equipped with four or more detectors aligned the 488 nm laser, allowing the detection of long red PE tandem conjugate fluorochromes such as PE-Cy5.5 and PE-Cy7. Antibodies directed against stem cell antigens conjugated with PE-Cy7, APC, APC-Cy5.5, and APC-Cy7 are now available from multiple sources. As demonstrated in Figure 5 (where APC, PE-Cy7, and APC-Cy7 were used to identify c-kit, Sca-1, and the lineage markers, respectively), this wide variety of detector options still permitted immunolabeling for multiple surface markers, despite the loss of the more usual 488 nm-excited probes. In cases in which no UV laser is available, DCV can be substituted with simultaneous immunophenotyping still available.

The validation of DCV as a reagent for detecting SP cells should also be taken in the context of progress in UV excitation for flow cytometry. UV excitation is now easier to achieve in flow cytometry than it was even a few years ago. Recently developed near-UV laser diodes are capable of exciting Hoechst 33342 with sufficient efficiency to resolve Hoechst SP . Hoechst 33342 will therefore remain a critical tool for SP analysis. Nevertheless, DCV has been demonstrated to be a useful analytical reagent for stem cell analysis on the ever-increasing array of violet laser-equipped cytometry instrumentation.
$ {) P  ?. `  b
DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST6 t+ H8 m% I! r3 l$ t
) g5 p. `0 t/ a7 o* L6 H1 g
W.G. and J.B. own stock in and are employed by Molecular Probes Invitrogen; the remaining authors have no financial interest in this corporation.
: l! ^( }% w, p2 p6 x" w# d
ACKNOWLEDGMENTS
5 t3 \; L4 ^9 l/ A7 i
This work was supported by intramural research funds provided by the Center for Cancer Research, the National Cancer Institute, and the National Institutes of Health. Preliminary data from this study were presented as an abstract and poster at the International Society of Analytical Cytometry Conference, May 21¨C25, 2006, Qu¨¦bec City, Qu¨¦bec, Canada.& V8 `) r) E6 y! s( @7 L. J
      【参考文献】
7 X. }' Z( l- e# s# F+ t6 g
9 B6 O( P: ]1 A) O3 e
Goodell MA, Brose K, Paradis G et al. Isolation and functional properties of murine hematopoietic stem cells that are replicating in vivo. J Exp Med 1996;183:1797¨C1806.

Goodell MA, Rosenzweig M, Kim H et al. Dye efflux studies suggest that hematopoietic stem cells expressing low or undetectable levels of CD34 antigen exist in multiple species. Nature Med 1997;3:1337¨C1345.

Goodell MA. Stem cell identification and sorting using the Hoeschst 33342 side population (SP). In: Robinson JP, Darzynkiewicz Z, Dean PN, eds. et al. Current Protocols in Cytometry.New York: John Wiley & Sons,2001;9.18.1¨C9.18.xx.

Challen GA, Little MH. A side order of stem cells: The SP phenotype. STEM CELLS 2006;24:3¨C12.  w! @/ S. ?1 O) f* j; y- W

Litman T, Brangi M, Hudson E et al. The multidrug-resistant phenotype associated with overexpression of the new ABC half-transporter, MXR (ABCG2). J Cell Sci 2000;113:2011¨C2021.
4 y& B4 U- I* w( U9 P% ?( u
Zhou S, Schuetz JD, Bunting KD et al. The ABC transporter Bcrp1/ABCG2 is expressed in a wide variety of stem cells and is a molecular determinant of the side-population phenotype. Nat Med 2001;7:1028¨C1034.' i, G: p2 L* V" N! ?
4 g* Q* u! P( @4 j$ [1 U6 r% U
Scharenberg CW, Harkey MA, Torok-Storb B. The ABCG2 transporter is an efficient Hoechst 33342 efflux pump and is preferentially expressed by immature human hematopoietic progenitors. Blood 2002;99:507¨C512.
+ P6 O8 C1 k+ m0 m
Uchida N, Leung FY, Eaves CJ. Liver and marrow of adult mdr-1a/1b(¨C/¨C) mice show normal generation, function, and multi-tissue trafficking of primitive hematopoietic cells. Exp Hematol 2002;30:862¨C869.

Kim M, Turnquist H, Jackson J et al. The multidrug resistance transporter ABCG2 (breast cancer resistance protein 1) effluxes Hoechst 33342 and is overexpressed in hematopoietic stem cells. Clin Cancer Res 2002;8:22¨C28.4 D. f, }( U& \6 Z6 w* ^4 F7 G+ D3 ^

Telford WG. Small lasers in flow cytometry. In: Hawley TS, Hawley RG, eds. Methods in Molecular Biology. 2nd edition London: Humana Press,2004;Volume 263:399¨C418 Flow Cytometry Protocols., i5 s$ N' S  w
4 K4 E) |# X& L$ y
Telford WG. Analysis of UV-excited fluorochromes by flow cytometry using a near-UV laser diode. Cytometry A 2004;61:9¨C17.

Telford WG, Frolova EG. Discrimination of Hoechst side population in mouse bone marrow with violet and near-UV laser diodes. Cytometry A 2004;57:45¨C52.

Simpson C, Pearce DJ, Bonnet D et al. Out of the blue: A comparison of Hoechst side population (SP) analysis of murine bone marrow using 325, 363 and 407 nm excitation sources. J Immunol Methods 2006;310:171¨C181.* Y/ v/ |+ A% s

Eaker SS, Hawley TS, Ramezani A et al. Detection and enrichment of hematopoietic stem cells by side population phenotype. Methods in Molecular Biology 2nd edition In: Hawley TS, Hawley RG, eds. London: Humana Press,2004;Volume 263:399¨C418 Flow Cytometry Protocols., _$ g% T3 }& G, C" M% l1 d
; K  D  j5 ]7 s9 f% a8 m( D' \9 r
Kawanabe N, Murakami K, Takana-Yamamoto T. The presence of ABCG2-dependent side positive cells in human periodontal ligaments. Biochem Biophys Res Comm 2006;344:1278¨C1283.5 r1 N; g! y" s

Robey RW, Honjo Y, Morisaki K et al. Mutations at amino-acid 482 in the ABCG2 gene affect substrate and antagonist specificity. Br J Cancer 2003;89:1971¨C1978.0 `7 d; C9 N0 X! p- U, o
, a$ w0 b2 e! T
Telford W, Kapoor V, Jackson J et al. Violet laser diodes in flow cytometry: An update. Cytometry 2006;69:1153¨C1160.( U% C0 [4 |! c: B- B

Schwarz BA, Bhandoola A. Circulating hematopoietic progenitors with T lineage potential. Nat Immunol 2004;5:953¨C960.8 e6 O  r4 W# D! F5 l: p' j% B# @
& Y/ p# d8 {4 w. I0 A; B$ d) h( C
Cabana R, Frolova EG, Thomas RA et al. The minimal instrumentation requirements for Hoechst side population analysis: Stem cell analysis on low-cost flow cytometry platforms. STEM CELLS 2006;24:2573¨C2581.

收藏0 分享0 顶0 踩0