Microspectrofluorimeter

The foregoing and other objects, features, and advantages of the invention will be apparent from the following more particular description of preferred embodiments of the invention, as illustrated in the accompanying drawings in which like reference characters refer to the same parts throughout the different views. The drawings are not necessarily to scale, emphasis instead being placed upon illustrating the principles of the invention.

FIG. 1 is a perspective view of a microspectrofluorimeter embodying the present invention;

FIG. 2 is an enlarged view of the filter accessory in the system of FIG. 1;

FIG. 3 is a perspective view of the filter accessory removed from the system of FIG. 1 and shown from the rear of FIG. 2;

FIG. 4 is a perspective view of the filter accessory of FIG. 3 with the filter housing endcap removed;

FIG. 5 is an end view of an element of a bayonet optical connector typically mounted to a microscope;

FIG. 6 is a side view, partially in section, of the bayonet optical connector connecting the accessory to the microscope;

FIG. 7 is an exploded sectional view of the filter assembly;

FIG. 8 is a sectional view of the filter assembly;

FIG. 9 is a view of the synchronizing marks on the filter assembly;

FIG. 10 is an optical and electrical block diagram of the microspectrofluorimeter;

FIGS. 11 (A-D) are a set of timing charts for the microspectrofluorimeter;

FIG. 12 is a perspective view of the filter assembly mounted in the optical path between the sample and the detector.

FIG. 13 is a rear view of the filter assembly in the system of FIG. 12.

A conventional microscope 18 modified in accordance with the present invention is illustrated in FIG. 1. A sample slide is shown at 20. The sample may be illuminated from above by a light source 22. Alternatively, a sample may be illuminated from a conventional xenon arc lamp 24 which would typically be connected directly to the microscope at 25. In accordance with the present invention, however, a filter assembly 26 is positioned in the optical path between the light source 24 and the sample 20.

Using either light source, the specimen can be viewed through a binocular viewer 28, and a video output can be provided by a television camera 30. To provide a quantitative indication of the fluorescence from a sample, a photomultiplier tube 32 is coupled to receive most of the light from the sample. In a conventional microscope, a cross-hair assembly extends into the microscope housing to an image plane of the specimen from a closure 34. For the presen invention that cross-hair is replaced by a variable field stop aperture which defines the portion of the sample viewed by the photomultiplier tube.

An enlarged view of the filter accessory is shown in FIG. 2. It includes the female portion 36 of a connector for coupling the accessory to the xenon arc lamp 24. The accessory also includes a male portion 38 of a connector for connecting it to the microscope 18. A beam of light from the light source 24 is transmitted through an optical conduit 40 between the connectors. A filter housing 42 is eccentrically mounted relative to the optical conduit and houses a rotating filter disc to be described. The filter disc is rotated by a motor 44 which is mounted to the optical conduit by means of a block 46. A guide 48 is provided for a manual shutter for blocking the light source from the filter assembly.

The filter accessory removed from the microscope is illustrated in FIGS. 3 and 4. In these figures the accessory is viewed from the rear of FIG. 2. In FIG. 4 the end plate 50 of the filter housing is shown removed. A rigid filter disc 52 is mounted within the housing for rotation about a center axis. The filter disc includes two fanshaped filter regions 54 and 56 separated by opaque regions 58 and 60 in a unitary construction. In a system for determining the calcium ion concentration of the sample using Fura-2 dye, one filter has a pass band centered at 340 nanometers and the other has a pass band centered at 380 nanometers. The specific filters used have a 10 nanometer full bandwidth at half the maximum transmission. As the filter disc is rotated by the motor 44, the two filters are sequentially moved into the optical path through the conduit 40. As will be discussed further below, the opaque regions 58 and 60 are just wide enough to block the light beam in the optical conduit as each is centered at the conduit.

A pair of reflective marks 61 are positioned on the filter assembly and those marks are sensed with each revolution of the filter disc by a light emitting diode and photodetector assembly 63 mounted to the end plate 50 of the housing. The signals from the photo detector synchronize photon counting with rotation of the filter as will be descibed below.

In order that the filter accessory can be readily mounted between a conventional microscope and its conventional light source, bayonet optical connectors 36 and 38 as illustrated in FIGS. 5 and 6 are provided. FIGS. 5 and 6 actually show the female portion 62 of the connector typically found on a microscope and the male portion 38 on the accessory. The connector 64 on the xenon arc lamp would be substantially the same as the connector 38, and the conector 36 of the accessory would be functionally the same as connector 62 on the microscope housing.

As shown in FIG. 5, the female portion of the connector includes two fixed pins 66 and 68 which extend radially inward from the rim 70 of the connector socket. A third pin 72 is spring biased so that it can be pulled outward by a pinhead 74. With the pin 72 thus pulled away from the socket, the frustoconical male portion of the connector on the accessory 76 can be slipped behind the pins 66 and 68. When the pin 72 is then released, the pin presses against the conical surface to bring the male portion tightly into the socket. The pin can then be locked into place by manipulating the head 74.

The filter assembly is shown in greater detail in FIGS. 7 and 8. A drive shaft 78 extends into the housing of 42 from the drive motor 44. A filter disc carrier 80 is positioned over the drive shaft 70. A clearance of about one mil is provided between the outer diameter of the carrier 80 and the inner diameter of a hole 82 in the filter disc 52 so that the filter can be slid onto the hub of the carrier 80. The hub of the carrier 80 protrudes through the filter disc and a brass retainer 84 is slid over the hub. The retainer 84 is fixed to the carrier 80 by a set screw 86. The set screw also clamps the carrier 80, which is split along its length, onto the drive shaft 78.

The filter disc 52 is pressed snuggly against the flange 88 of the carrier 80 by means of an internally and externally threaded brass sleeve 90 which is threaded onto the retainer 86. The sleeve 90 is also fixed by a set screw. Each of the retainer 84 and sleeve 90 has a pair of holes in the front face thereof for tightening by a two pin wrench. Finally, an end plate 92 is positioned on the assembly and retained by a set screw. The end plate 92 is blackened as shown in FIG. 9 and carries the reflective synchronizing marks 61.

FIG. 10 is a block diagram of the overall system by which microspectrofluorimeter readings are obtained. Light from the light source 24 is directed through the rotating filter disc 52 to illuminate a sample on the microscope 18. Light emitted by the specimen in the microscope 18 is directed to the photomultiplier tube 32. The photomultiplier tube includes a bandpass filter 94 which has a center frequency at 500 nanometers when measuring calcium concentration with Fura-2. The filter passes the light of interest but blocks any light of the excitation wavelengths which may have been reflected from the sample. A shutter 96 is also positioned on the photomultiplier tube.

The output from the photomultiplier tube is amplified by a preamplifier 98 and filtered by a discriminator 100 to eliminate pulses due to noise. The filtered signals are then applied to a photon counter 102. The photomultiplier tube power supply 104, the amplifier 98, the discriminator 100 and the counter 102 are conventional PMT devices such as those provided by Ortec Corporation. The counter 102 has dual channels and it is switched to alternately count photons as the sample is alternately illuminated by the 340 nm and 380 nm light. The counter 102 is switched from one channel to the other by a microprocessor 106, and as photons are counted in one channel the count previously obtained in the other channel is read by the microprocessor and stored. The microprocessor then computes the calcium concentration based on the photon counts during exposure of the sample through each of the 340 nm and 380 nm filters normalized for disc speed. In addition, the microprocessor normalizes for variations in the intensity of the light source 24 by means of a detector 108 which senses the intensity of the light source through a fiber optic bundle 110.

The microprocessor controls switching of the counter 102 between channels based on timing derived from the reflective marks 61 detected by the photondetector 63. The timing of the switching of the counter 102 between the two channels is illustrated in FIG. 11. As already noted, the opaque regions 58 and 60 of the filter disc are just wide enough to completely block the light beam through the conduit 40 when the opaque regions are centered with respect to the conduit. When an opaque region is centered, the photomultiplier tube detects a minimum intensity of fluorescent light. By switching the counter 102 when the opaque regions are centered, each counter channel can count at or near a 50% duty cycle with a plateau of maximum intensity through a substantial portion of each cycle. Thus, near 100% of the photons received by the photomultiplier tube are counted for maximum efficiency. To this end, the pulses for switching the counter between the two channels are, as illustrated in FIG. 11B, centered at the minimum of the PMT output.

Synchronization of the switching with the rotation of the filter disc results from sensing the two reflective marks 61 to obtain the pulses illustrated in FIG. 11C. With the rise of the first pulse, a timing interval is initiated as illustrated in FIG. 11D. With the rising edge of the next pulse as the second mark is sensed, that timing interval is terminated. During the timing interval illustrated by the pulse FIG. 11D, the output of a megahertz clock is counted. The speed of the filter disc is identified by the number of pulses counted, that is the time T.sub.i, between sensing of the two marks. Because the location of the synchronizing marks relative to the opaque regions of the filter is known, the time T.sub.d at which the counter channel must next be switched and then the subsequent time T.sub.f at which the counter must again be switched can be readily identified. Because the speed of rotation is determined with each cycle, the system responds immediately to any fluctuations in the speed of rotation of the filter disc. On the other hand, by varying the speed of rotation the temporal resolution of the system can be controlled with the microprocessor 106 responding automatically in each cycle to maintain synchronization of the counter switching with the filter rotation.

Because the microprocessor responds to the filter rotation in controlling switching of the photon count relative to speed of rotation, the photon counting by the combined counter channels is always near one hundred percent. As a result, the system always operates at maximum light collecting efficiency; all light received by the PMT is counted.

A particular application of the present system is in the measurement of fluorescence of ion or molecular sensitive fluorescent probes at two excitation wavelengths to calculate ion or molecule concentrations. For example, calcium ions react with Fura-2 in a reversable reaction of the form ##EQU1## Where X represents the calcium ion and F the Fura-2 fluorescent dye. The constant of the reaction K.sub.d is defined by ##EQU2## That is, K.sub.d is equal to the product of the concentrations of the free calcium ion and free Fura-2 relative to the concentration of the bonded calcium and Fura-2. The free Fura-2 has its primary fluorescence in response to an excitation wave length of 380 nm. The bonded Fura-2, on the other hand, fluoresces principly in response to an excitation wave length of 340 nm. Both the bonded and unbonded Fura-2 fluoresce at wavelengths centered about 500 nm. Thus, by exciting a sample at separate times with the two excitation frequencies the relative concentrations of the free Fura-2 and the combined Fura-2 can be determined in order to calculate the concentration of calcium ion. The concentration of the calcium ion can be calculated from the following: ##EQU3## I is the sensed fluorescence normalized by the microprocessor to both speed of filter rotation and light source intensity. The subscripts represent the excitation wave lengths. The intensity differences are between that of the test specimen and that of a blank specimen such as a sample of saline solution. R.sub.min and R.sub.max are respectively the R measured from a sample with only the Fura-2 and no calcium ions and the R measured from totally saturated Fura-2 by providing a sufficiently large amount of calcium ions. .beta. is the ratio of fluorescence with only Fura-2 and with saturated Fura-2 each excited at 380 nm.

Thus, in an experimental procedure, a specimen of saline solution is positioned on the sample holder of the microscope and is illuminated through the rotating filter disc to obtain the blank intensity measurements at the respective excitation wavelengths. Then, samples of Fura-2 without calcium ions and Fura-2 saturated with calcium ions are separately positioned on the sample holder and separate measurements are made as each is excited through the rotating filter disc. Finally, a specimen of interest is positioned on the sample holder and is excited through the rotating filter wheel to complete the measurements. All of the data is stored by the microprocessor and is used to compute the calcium ion concentrations.

In many studies it is important to follow the dynamic changes in calcium ion concentration over short time periods. For example, while the calcium concentration in a muscle cell is being monitored, the cell may be electrically or chemically stimulated to contract, and the change in calcium ion concentration during the contraction can be measured. Because the process occurs in the order of milliseconds, it is important that the measurements at both excitation wave lengths be made in the order of milliseconds. To that end, the filter disc is rotated at a rate of up to 12,000 revolutions per minute. With the microscope, the rerevolutions action may be monitored with such rapid response in regions of a sample as small as 0.0625 square microns.

FIGS. 12 and 13 illustrate an alternative position of the filter accessory 26. The accessory is coupled between the microscope and the detector 32 using the optical connectors 36 and 38 as before. The light source 24 is coupled directly to the microscope. This arrangement may be used to detect different emission wavelengths from molecules with a given excitation wavelength. An example of the systems' use would be in the detection of Indo-1 which also binds to calcium.

The system of FIG. 10 would be unchanged but for the position of the filter wheel 52. Detection through the photon counting photomultiplier tube 32 would be synchronized to the continuously rotating filter as before. The photon counts could be normalized to the speed of the rotating filters and to the intensity of the radiation source.

While the invention has been particularly shown and described with reference to preferred embodiments thereof, it is understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.