Summary Document

This document is a summary of message from the Purdue University Cytometry Laboratories Cytometry Discussion list.

http://www.cyto.purdue.edu/hmarchiv/index.htm

Dear Flow Community.
Here are the responses I received to my question, posted below.
Thanks very much to all of you who took the time to respond. I was able to clarify what I wanted and prepare my proposal accordingly.

Best wishes,
Joanna

Flow Cytometry Core Facility
EPFL, Swiss Federal Institute of Technology
SV-SG, Station 15
CH 1015, Lausanne
Switzerland

+41 21 69 39 547

Original Question:

From: Roberts Joanna
Sent: jeudi, 14. juin 2007 17:11
To: 'cytometry@flowcyt.cyto.purdue.edu'
Subject: Choosing filter configurations for Qdots in Flow

Dear Flow Cytometry Community,

I have a question about filter configurations for Qdots in flow cytometry.

I would like to order a new instrument with a violet laser to measure the full range of qdot fluorochromes. >From quick talks with BD, I have a list of the filters they suggest. They represent slightly narrower signal ranges in just about all cases than the filters used by the fellows in the paper by Chattopadhyay et al in Nature Immunology last year.

For those of you working with these reagents in flow, could you offer your opinion on this matter? Would you recommend collecting more signal (and probably doing a bit more ‘inter-qdot’ compensation) or would you suggest taking the narrower band pass, collecting less signal and needing less compensation?

Here is a list of the bandpass filters suggested from BD as well as those used in the paper mentioned above.

Chattopadhyay et al Suggestion from BD

QDOT800 780/60 (750LP) 800/30

QDOT705 705/70 (670LP) 710/50

QDOT655 660/40 (640LP) 660/20

QDOT605 605/40 (595LP) 610/20

QDOT585 585/42 (570LP) 585/42

QDOT565 560/40 (557LP) 560/40

QDOT525 515/20 525/50

Any and all recommendations, ideas and advice are appreciated.

Thanks a lot to the organisers and moderators who run this list and all the people who take the time to dish out their 2 cents- it is great!

Best wishes,

Joanna

Responses:

Qdots are very bright, so if you're getting a new instrument with a good laser, I guess a narrower filter would do the job and reduce compensation issues. Of course, as always, it depends on the applications intended. If there are low-signal studies, then get the wider bands. Bear in mind that a Qdot is excited by all lines shorter than it's emission. Therefore, exactly due to the fact that Qdots are very bright, if doing multi-laser works you might encounter some trouble there, with a red Qdot being excited by UV, V, 488, and maybe the red laser as well. Check out Mol Probes spectra viewer. It could be small emission, but then you have compensation all over the place, which might or might not be an issue for you - again, depends on your application and instrument. Maybe people with more experience with Qdots can share their experience on this respect? This characteristic has been keeping me from getting more into Qdots because I commonly use all 4 lasers on the LSR II.

Good luck,

Actually, the best strategy for each Qdot detector is the use of a sharp dichroic mirror with a wide band pass filter. After much trial and error the attached filter configuration illustrates this point. You can also visit our web site at; http://www3.niaid.nih.gov/labs/aboutlabs/VRC/flowCytometryCoreLaboratory/ for additional information. If your current system has a green laser, the QD525 will be very dim because of the narrow 515/20 filter. I would recommend using QD545 with the filters as listed in the attachment.

SP

LSR Detector Configuration

v.070508

Laser	Detector Name	Fluorochrome Detected	Dichroic LP Filter	BP Filter
Blue 488nm	FSC	Forward Scatter
Blue 488nm	B710	Cy55PerCP, PerCP*	685LP	710/50
Blue 488nm	B515	FITC, CFSE, GFP, Alexa 488, GrViD	505LP	515/20
Blue 488nm	SSC	Side scatter	488LP
Green 532nm	G780	Cy7PE, Alexa 750PE	740LP	780/40
Green 532nm	G710	Cy55PE, Alexa 700PE	690LP	710/50
Green 532nm	G660	Cy5PE, Alexa 647PE	640LP	660/40
Green 532nm	G610	TRPE, TR, OrViD,RFP, Alexa 594	600LP	610/20
Green 532nm	G560	PE, Cy3	empty	575/25
Green 532nm	Empty	na
Green 532nm	Empty	na
Green 532nm	Empty	na
Red 633nm	R780	Cy7APC, Alexa 750	740LP	780/60
Red 633nm	R710	Cy55APC, Alexa 680, 700	685LP	710/50
Red 633nm	R660	APC, Alexa 647	empty	660/20
Violet 407nm	V800	QD800	740LP	780/60
Violet 407nm	V705	QD705	670LP	705/70
Violet 407nm	V655	QD655	630LP	660/40
Violet 407nm	V605	QD605	595LP	605/40
Violet 407nm	V585	QD585	570LP	585/42
Violet 407nm	V565	QD565	557LP	560/40
Violet 407nm	V545	QD545, Pacific Orange, Aqua Blue**	535 LP	550/40
Violet 407nm	V450	CBlue, ViViD, Pacific Blue	empty	450/50

* Requires filter change: 640LP + 660/50

**Requires filter change: 505LP + 515/20

Some rules for Q dots, generally speaking the compensation required is less than for conventional fluorochroomes , remember the Q dots are excited by any laser so you need not only consider what filters you have on a specific laser , but possibly more importantly , what similar filters you have on other laser lines.

I would always take the option of collecting more signal, although the extinction coefficient of the Q dots is relatively high the rate defining step very often is the antigen concentration on your cell , so you need to optimise signal in cases where antigen expression is just above autofluorescence.

To give definitive answers regarding filters means a lot more information is required as to laser power, type of cells etc, however I have found that Q dots will fit in to most filter configurations rather than re configuring the instrument specifically for the Q dots, then perhaps having to re configure things for conventional fluorochromes

We have not had a lot of experience with Qdots but what we have done shows that there are significant differences in average quantum yield (QY). In our samples qdot655 was much brighter than qdot585 (compared like-for-like). If you look at the company's literature I recall that they seem to suggest different QY is a general property of specific wavelength qdots. It may be this which is colouring BDs thinking - obviously brighter qdots would need lower bandwidth.

However, a paper by Yao et al (PNAS doi 10.1073) suggests that there is significant batch-to-batch variation of QY (and that this is related to the dark, ie inactive, fraction of particles).

Whatever, they are extremely bright by normal flow standards so assuming your samples will have a reasonable labelling density, narrow band filters will work perfectly well and make sense by reducing compensation issues.

We're using the Qdots on a UV laser, but emission is the same. We have a FACSAria, and we have a Trigon on that laser, so we're limited to 3 colors. We've chosen to use Q800, Q585 and Q525, as we felt we could get the best separation with available dichroics. We use as follows...

Q800 - 780/60 (750lp)

Q585 - 585/40 (555lp)

Q525 - 525/25

We got all the filters from Chroma (bp's & dcm's), and we get good strong signals and easy compensation with this combination. (FWIW, we get a bit more excitation using the UV laser than you will with the violet laser.)

Generally, I like to use narrower band pass filters and do less compensation. I just feel that it's more "natural". By that I mean that compensation is an electronic manipulation of the data, and taking a narrower window allows me to manipulate the data less. So, if you're looking to use all of the Qdots (and their spectra overlap quite a bit at times), I'd go with narrower band pass filters. If you're only going to be using a couple of them at a time (as we do), wider is probably fine. That's my two cents!

LSR Detector Configuration

Green 532nm