Elsevier

Methods

Volume 30, Issue 1, May 2003, Pages 3-15
Methods

A two-photon and second-harmonic microscope

https://doi.org/10.1016/S1046-2023(03)00003-3Get rights and content

Abstract

Two-photon microscopy has revolutionized life sciences by enabling long-term imaging of living preparations in highly scattering tissue while minimizing photodamage. At the same time, commercial two-photon microscopes are expensive and this has prevented the widespread application of this technique to the biological community. As an alternative to commercial systems, we provide an update of our efforts designing custom-built two-photon instruments by modifying the Olympus FluoView laser scanning confocal microscope. With the newer version of our instrument we modulate the intensity of the laser beam in arbitrary spatiotemporal patterns using a Pockels cell and software control over the scanning. We can also perform simultaneous optical imaging and optical stimulation experiments and combine them with second harmonic generation measurements.

Introduction

The introduction of two-photon excitation [1] to life sciences has opened novel experimental territories [2]. Two-photon excitation occurs when two low-energy photons are simultaneously absorbed by a molecule in the ground state, resulting in an excitation similar to that produced by a single high-energy photon [3]. This process has important consequences for microscopy because it enables fluorescence with infrared excitation light, which can penetrate without major scatter through living tissue [4]. In addition, the nonlinear reaction confines the excitation essentially to the focal point [1], thus effectively solving a major problem in optical microscopy, that of out-of-focus excitation. These improvements of two-photon excitation over conventional fluorescence microscopy have proven to be of great practical advantage: two-photon microscopy has enabled, among other things, physiological analysis of dendritic spines [5], [6] as well as direct functional mapping of synaptic receptors [7] and channels [8] on living neurons in brain slices or in vivo.

The spread of two-photon microscopy, however, has been hampered by the high costs associated with commercially available two-photon systems. As a solution to this, over the last years our laboratory has designed and built two-photon microscopes based on modification of relatively low-cost confocal systems. In a previous publication we describe in detail our initial design of a custom-built two-photon microscope based on the Olympus FluoView scanning system [9]. We find that this strikes a good compromise between cost, flexibility, and ease of engineering the system.

The large interest generated by the first publication in the neuroscience community has stimulated us to provide a detailed update to the modifications and further improvements of our system, so that other investigators can also profit from them. In this report, we describe how we have implemented the control of the beam positioning and intensity, essential for fast imaging or photostimulation experiments, as well modifications of the microscope to enable the simultaneous measurement of second harmonic generation (SHG) [10], [11], a novel nonlinear microscopy technique that could have major implications for life sciences and that can be implemented with the same lasers as two-photon excitation.

Section snippets

Description of the system

Our current instrument consists of two commercial lasers, an external light path, a scanning head, and an optical microscope (Fig. 1). We describe in detail each part of it and the rationale behind each of our choices in design.

Modifications of optical pathway before the scanning unit

In our new system, it became necessary to modify the laser excitation pathway since we found that the laser beam, in reaching the back aperture of the objective, was not collimated (the Olympus BX50WI microscope is designed for an infinity-corrected objective lens). We diagnosed this problem by monitoring the beam profile at different points in the light path with a WM100 Omega Meter from Thorlabs (Newton, NJ). The lack of collimation is probably caused by incompatibility of our BX50WI upright

Summary

In this work, we provide a substantial update to our original modification of the Olympus FluoView system for two-photon microscopy [9]. The major improvements are spatial filtering and refocusing of the incident laser beam, direct control of the galvanometers via custom software enabling imaging or photostimulation of any arbitrary number of ROIs, and, finally, implementation of SHG imaging. The system we describe is flexible and can be easily used for a variety of experiments. Importantly,

Acknowledgements

We thank members of our laboratory and Hajime Hirase for their help; Kenji Matsuba and Yiwei Jia from Olympus USA for FluoView documentation; Huibert Mansvelder for his point measurements; Robert O’Hagan, Dattananda Chelur, and Martin Chalfie for supplying the C. elegans samples; and Les Loew and Aaron Lewis for encouragement to explore SHG. This work was supported by the Merck Fund, the NEI (EY 111787 and EY13237), the NINDS (NS40726), the John Merck Fund and the New York State STAR Center for

References (24)

  • W. Denk et al.

    J. Neurosci. Meth.

    (1994)
  • A. Lewis et al.

    Chem. Phys.

    (1999)
  • M. Ng et al.

    Neuron

    (2002)
  • P.J. Campagnola et al.

    Biophys. J.

    (1999)
  • L. Moreaux et al.

    Biophys. J.

    (2001)
  • P.J. Campagnola et al.

    Biophys. J.

    (2002)
  • O. Bouevitch et al.

    Biophys. J.

    (1993)
  • W. Denk et al.

    Science

    (1990)
  • M. Goeppert-Mayer

    Ann. Phys.

    (1931)
  • L.O. Svaaland et al.

    Photochem. Photobiol.

    (1983)
  • R. Yuste et al.

    Nature

    (1995)
  • R. Yuste et al.

    Nat. Neurosci.

    (2000)
  • Cited by (0)

    View full text