title2.JPG (17589 bytes)

Violet laser diodes in flow cytometry.

A variety of useful fluorescent probes require violet excitation, including the phenotyping fluorochromes Cascade Blue, Pacific Blue, Alexa Fluor 430 and Cascade Yellow, the DNA binding binding dye Hoechst 34580 and the expressible fluorescent protein Cyan Fluorescent Protein (CFP).   Some fluorescent probes traditionally excited by UV sources can also be used with violet excitation, including the DNA dyes Hoechst 33342 and DAPI and the alkaline phosphatase substrate ELF-97.  Violet excitation on the flow cytometer has traditionally been provided by water-cooled krypton-ion gas lasers, which emit strongly at 407 and 413 nm and more weakly at 415 nm.  The high purchase and operating cost, large size and demanding physical requirements for these lasers has led to a search for smaller, more cost-effective substitutes.

In 1995, Nakamura and colleagues at Nichia Corporation (Japan) developed a violet laser diode with an emission range of approximately 385 to 410 nm.  These lasers are promising candidates as flow cytometric sources for violet probe excitation.  We currently have a Power Technology 30 mW 408 nm violet laser diode in the laboratory and have evaluated it on our FACStar Plus and FACSVantage flow cytometers, in comparison with our krypton-ion emitting at 407 nm.  We have tested this laser with Cascade Blue, Pacific Blue, Cyan Fluorescent Protein and the fluorogenic alkaline phosphatase substrate ELF-97.  The data shown below were collected with the VLD on the same FACSVantage DiVa with the different lasers in the same position, with detection at the same PMT voltage settings (unless otherwise indicated); in this format we found comparable sensitivity and signal-to-noise ratios for the above probes compared to both high- and low-power krypton-ion laser excitation.  These results indicate that VLDs can be substituted for more expensive and maintenance-intensive gas lasers for some applications.

(Below).  Power Technology 30 mW violet laser diode emitting at 408 nm.  This laser requires two cylindrical optical elements to circularize the normally multimodal diode laser beam.  Maximum emission for this laser post-optics is approximately 18 mW.  For the evaluation below, power level was generally set to 15 mW.

wpe4.jpg (23186 bytes)

(Below).  Power Technology 30 mW violet laser diode beam profile.  Profile was collected with a WinCamD CCD beam profiling system (Dataray, Inc.) with a ND4.0 filter between the laser and the CCD element.  The VLD was mounted on the FACSVantage in the beam orientation shown.

wpe5.jpg (11591 bytes)

(Below).  VLD mounting system.  We used a laser mounting post system constructed from Newport Instruments components (vertical rod, rack-and-pinion rod clamp and adjustable laser mounting bracket with micrometer rotators) mounted on a single-axis dovetail linear stage and a adjustable X-Y base, with the entire setup mounted on a 15 cm x 15 cm aluminum breadboard.  All parts are metric standard and use M6 mounting bolts.  Newport part numbers are indicated in parentheses.  This assembly was extremely stable and could be easily mounted and removed from the FACSVantage optical bench.

wpeA.jpg (30939 bytes)

(Below).  Laser positioning.  The laser position adjustments are shown below.  This system allows both coarse and fine adjustment of the X-position of the laser (perpendicular to the beam), course Z-adjustment (vertical) and both horizontal and vertical rotational adjustment.

wpe5.jpg (25737 bytes)

(Below).  VLD mounted on the FACSVantage DiVa.  The VLD was mounted in the third (middle) position on the FACSVantage optical bench, the position normally occupied by the water-cooled krypton-ion laser.

wpe7.jpg (30780 bytes) 

(Below). FACSVantage DiVa optical bench configuration for violet-excited fluorochromes. Fluorochromes were detected in the FL11 detector position with no intervening dichroic.

wpe1.jpg (42885 bytes)

(Below). Quality control beads analysis with the VLD.    Polyscience 2 um yellow-green beads (left histograms) and Molecular Probes InSpeck Blue linearity beads (right histograms) were analyzed with krypton-ion excitation at 75 or 15 mW (top and middle rows of histograms) and with VLD excitation at 15 mW (lower two rows) on the same FACSVantage DiVa at the same detector voltage settings.  MFI values for the Polyscience beads are shown.  Relative fluorescence of the InSpeck beads compared to an arbitrary maximum control bead (100%) are shown in the right histograms.  

wpe8.jpg (29728 bytes)

VLD excitation of Cascade Blue and Pacific Blue.  EL4 cells were labeled with biotin-anti-CD44 or biotin-anti-CD90 followed by either Cascade Blue or Pacific Blue conjugated streptavidin.  Samples were then analyzed on the FACSVantage DiVa equipped with either a VLD 15 mW emitting at 408 nm or or a krypton-ion 407 nm 100 mW (or attenuated 15 mW) emitting at 407 nm.  Cascade Blue fluorescence was detected through a 440/10 nm filter, Pacific Blue through a 463/50 nm filter sandwiched with a 488 nm notch filter.  VLD excitation of both probes gave similar sensitivity to the krypton-ion laser.

(Below).  Cascade Blue detection with krypton-ion or VLD excitation.  Top panel, excitation/emission spectrum for Cascade Blue.   Krypton-ion excitation at 75 or 15 mW (next two rows of histograms) was compared to VLD excitation at 15 mW (lower two rows) on the same FACSVantage DiVa at the same detector voltage settings except where indicated.   MFIs for background (open) and specific labeling (filled) peaks are shown, and the ratios in boldface.  The VLD-excited sample voltage was increased by 100 V in the bottom row to bring the autofluorescence peak off scale and increase accuracy of the ratio measurement.

wpe9.jpg (44281 bytes)

(Below).  Pacific Blue detection with krypton-ion or VLD excitation.  Top panel, excitation/emission spectrum for Pacific Blue.   Krypton-ion excitation at 75 or 15 mW (next two rows of histograms) was compared to VLD excitation at 15 mW (lower two rows) on the same FACSVantage DiVa at the same detector voltage settings except where indicated.   MFIs for background (open) and specific labeling (filled) peaks are shown, and the ratios in boldface.  The krypton-ion and VLD-excited sample voltages were increased by 100 V in the two bottom rows to bring the autofluorescence peak off scale and increase accuracy of the ratio measurement.

wpeA.jpg (38685 bytes)

VLD excitation of Cyan Fluorescent Protein.  NIH 3T3 cells expressing CFP were similarly evaluated for VLD excitation.  pECFP, pEGFP-1 and pDsRed1-1 plasmids encoding the ECFP, EGFP and DsRed FPs, respectively were obtained from Clontech Laboratories, Inc. (Palo Alto, CA). The MCIN, MGIN retroviral vector containing the EGFP gene expressed from the murine stem cell virus (MSCV) retroviral long terminal repeat on a bicistronic transcript, which also contains a downstream neomycin phosphotransferase gene linked via an internal ribosome entry site (IRES) from the encephalomyocarditis virus. The MCIN, MGIN and MRIN retroviral vectors containing the ECFP, EGFP and DsRed genes were similarly constructed by inserting the respective FP genes into the same MSCV-based IRES-neo retroviral vector backbone. Stable helper-free retroviral producer lines were generated by transduction of GP+E-86 ecotropic packaging cells followed by G418 selection as described previously.  CFP-expressing cells were analyzed using a 485/22 nm filter and a 488 RB restriction bandpass blocker. "Cocktails" of GFP, CFP and DsRed-expressing cells were also analyzed on the FACSVantage SE to determine whether this laser was compatible with multicolor FP analysis.  As for Cascade Blue and Pacific Blue, the VLD excited CFP at a similar level of efficiency as the the krypton-ion laser.

(Below).  Cyan Fluorescent Protein analysis with krypton-ion or VLD excitation.      Top panel, excitation/emission spectrum for CFP.  Krypton-ion excitation at 75 or 15 mW (next row of histograms) was compared to VLD excitation at 15 mW (lower two rows) on the same FACSVantage DiVa at the same detector voltage settings except where indicated.   MFIs for background (open) and specific labeling (filled) peaks are shown, and the ratios in boldface.  The VLD-excited sample voltage was increased by 100 V in the bottom row to bring the autofluorescence peak off scale and increase accuracy of the ratio measurement.  Multicolor FP analysis on the FACSVantage SE with VLD in the secondary position is shown in the right-most cytograms.  Software compensation was carried out post-analysis with WinList ver. 4.0 (Verity Software House).

wpeB.jpg (66640 bytes)

VLD excitation of ELF-97.  The fluorogenic alkaline phosphatase substrate ELF-97 is useful for detecting endogenous alkaline phosphatase in connective tissue and may be a useful UV-excited phenotyping fluorochrome for flow cytometry.  In this experiment, EL4 cells were labeled with biotin-anti-CD44 or biotin-anti-CD90 followed by alkaline phosphatase conjugated streptavidin.  UMR-106 rat osteosarcoma cells expressing high levels of endogenous AP were also labeled with the ELF-97 substrate alone.  Samples were then analyzed on the FACSVantage DiVa equipped with either a VLD 15 mW emitting at 408 nm or or a krypton-ion 351 nm UV 100 mW or 407 nm 100 mW (or attenuated 15 mW) emitting at 407 nm.  ELF-97 was detected through a 535/45 nm filter.  All violet excitation sources gave similar detection sensitivity for ELF-97.

(Below).  ELF-97 detection with krypton-ion or VLD excitation.  Top panel, excitation/emission spectrum for ELF-97.   Krypton-ion excitation at 75 or 15 mW (next two rows of histograms) was compared to VLD excitation at 15 mW (lower two rows) on the same FACSVantage DiVa at the same detector voltage settings.   Both phenotyped EL4 cells (first two columns) and UMR-106 cells (right-most column) are shown.   MFIs for background (open) and specific labeling (filled) peaks are shown, and the ratios in boldface.

wpeC.jpg (52063 bytes)

So, violet laser diode excitation gave comparable detection sensitivity for Cascade Blue, Pacific Blue and Cyan Fluorescent Protein.  Excitation of ELF-97 was also comparable between both laser sources.

References.

Hawley, T.S., Telford, W.G., Ramezani, A. and Hawley, R.G. 2001.  Four-color flow cytometric detection of retrovirally expressed cyan, green, yellow and red fluorescent proteins. BioTechniques 30, 1028-1034.

Nakamura, S. and Fasol, G.  1997.  The blue laser diode.  GaN based light emitters and lasers.  Berlin: Springer.

Shapiro, H.M. and Perlmutter, N.G.  2001.  Violet laser diodes as light sources for flow cytometry. Cytometry 44, 133-136.


Go here for the violet laser diode evaluation on the BD LSR II.

Go here to see a comparison of Coherent and Power Technology violet laser diodes on the FACSVantage DiVa.

We have also evaluated a violet laser diode installed in our Compucyte laser scanning cytometer; the results can be seen here.

AdobePDF.gif (1172 bytes)  VLD mounting information and part list.

AdobePD2.gif (2536 bytes) Download Adobe Acrobat Reader here .

Return to the NCI ETI Flow Lab home page.