A Far Red Laser for Confocal Excitation
A 690 nm Laser Expands the Range of Excitable Dyes
- Fig. 1: The LASX spectrophotometer GUI with a detection range of 350-800 nm. We have referred to the region from 700-800nm as “the lost bandwidth”. Note the 690nm laser AOTF in the upper left of the window. Although the label says “pulsed”, it is in fact a CW laser.
- Fig. 2: Image of multiply labeled tissue, collected sequentially with the internal detectors. There is minimal background fluorescence, and no crosstalk between the 633 and 700nm channels. GFP (green), PE (red), Alexa 633 (white), Alexa 700 (cyan).
- Fig. 3: A two color image of tissue demonstrating crosstalk free sequential imaging of Alexa 594 and Alexa 700 using the internal detectors. Alexa 594 (red), Alexa 700(cyan).
- Fig. 4: APD image collection. Images collected simultaneously for higher speed using both APD channels. mCherry (Red), and Indocyanine green (green), overlay. A 760nm dichroic was used to separate red from near IR emission.
A 690 nm CW laser coupled to a Leica SP8 confocal microscope allows for imaging of samples labelled with far-red probes. Detection with internal PMTs or HyDs gives an image with a high signal to noise ratio. Alexa 700 is bright even with low laser power, while crosstalk from Alexa 633 is minimal. Longer wavelength dyes such as Alexa 750 and Indocyanine Green are also excited well by the 690nm laser. Fluorescence emission beyond 750 nm requires the use of external APD detectors with selective barrier filters.
Since the earliest days of commercial confocal systems, the longest excitation wavelength has always been 633 nm. This was primarily determined by the output lines of the Kr-Ar mixed gas lasers first used on confocal systems, and later by the 633 nm HeNe which was widely available. More recently many fluorescent dyes have become available with excitation maxima ranging from 700 to 800 nm. Investigators who use these dyes for flow cytometry studies often ask if they can be used as probes on our confocal systems. Until this time, we could not excite these probes other than by using the 633 nm laser which was not well matched to the excitation curves of these dyes.
The Leica SP8 confocal system, due to its design, is ideal for extending the excitation and emission range of confocal microscopy. The spectrophotometer detection system is highly flexible system allowing detection from 350 to 800 nm. The bandwidth and range of each of the 5 detectors can be set independently without the use of interference filters. However, like all other confocal systems the longest available excitation wavelength is 633 nm. As a result, we often referred to the detection range of 700-800 nm as “the lost bandwidth” since it is available for detection, but there was no way to excite the available probes. The region of 700-800 nm represents nearly 20% of the total available detection range.
More recently, investigators, particularly those in immunology, have been eager to use as many probes as possible when imaging tissue. Some samples contain as many as 10 probes.
To accommodate this, use of the complete bandwidth of the instrument is essential to minimizing crosstalk between the channels. Use of the 700-800 nm portion of the spectrum is of major benefit to this effort.
A number of new far red excitable probes for microscopy have been released in recently years. These include Alexa 700, 750 and 800 from Invitrogen, Silicon Rhodamine (SiR) 700 probes, and others such as Indo-cyanine green. Many of these are already used routinely for flow cytometry, but rarely for microscopy.
To add a longer wavelength laser to our SP8 confocal system, we worked with the development group at Leica Microsystems, Mannheim, Germany. A 690 nm laser proved to be the best compromise given the required mechanical compatibility, and the required excitation range.
Hardware Required (Method)
The requirements for adding an additional laser include both physical compatibility with the existing optical components, but also control via the Leica LAS software. For investigators using the microscope, the laser must be fully integrated, that control is similar to the control of other lasers.
A Coherent Obis 690 laser was selected as it provided the required wavelength, and was also mechanically similar to other coherent lasers used in the confocal system. The laser power entering the confocal system is controlled by an AOTF, as are all other lasers in the system. The long wavelength of the laser is outside the range of the AOBS crystal, so the laser must be introduced downstream. The MFP port, often used for multiphoton laser coupling was the logical place to introduce the laser. A custom dichroic mirror which reflects the 690 nm laser, but allows the passage of both visible emission and longer wavelength emission was added to the port slider. A near IR fiber was used to couple the laser to the MFP port.
The internal detectors of the confocal, either PMT or HyD work well to detect near IR fluorescence emission from 700-750 nm. Internal detection is advantageous because it allows near IR data collection simultaneously with typical shorter wavelength probes. At 750 nm the detector sensitivity is around 7-8% for both the APD and HyDs making this wavelength near the upper limit for internal detection. For fluorescence detection in the range of 750-900 nm, avalanche photodiodes (APD) are used. APD detectors can collect with high efficiency up to nearly 1000 nm and are connected to the SP8 scanhead via the X1 port. Selecting the option “substrate” in the X1 port slider allows the signal to pass through and on to the APD detectors. Control of the ADP detectors is completely integrated into the LASX software. As these are not part of the internal spectrophotometer detector of the scanhead, barrier filters are used to select the wavelengths collected by each of the two ADPs. We have found that an 810/90 gives good results when using a single near IR probe. Other filters which have proven useful are 740/60 and 610/75 (for mCherry).
A large number of investigators have labelled their samples with near IR dyes, and imaged them on our confocal system. Emission from dyes excited by the 690 laser are bright, with little background fluorescence. There is little to no cross excitation with dyes such as Alexa 594. Some cross excitation of Alexa 633 and 647 has been observed when this is present with longer wavelength probes. When using the internal detectors of the scanhead, Alexa 700 fluorescence can be collected simultaneously with most other probes with the exception of Alexa 633 and 647. Because of the dichroic used to introduce the 690 laser line into the scanhead, emission wavelengths from these dyes are blocked. Sequential collection is necessary if these combinations are used. Detection with the APDs and internal detectors is necessarily sequential because the X1 port slider must be moved.
In the future it is anticipated that the 690 nm laser should become a standard option for many confocal systems. Additional wavelengths even further into the near IR might also be possible. With the decreasing cost of small near IR diode lasers, and the increasing diversity of probes available, many more labs should consider this option as an upgrade to an existing confocal system or as part of an order for a new instrument.
This research was supported by the Intramural Research Program of the NIH, and the National Institute for Allergy and Infectious Disease. Reference to any instrument manufacturers, or probes does not constitute endorsement by the US government, NIH, NIAID, or the Biological Imaging Facility.
Dr. Owen Schwartz
NIAID-National Institutes of Health
Bethesda, MD, USA