GB2583978A

GB2583978A - Biological sample scanner

Info

Publication number: GB2583978A
Application number: GB1909278.2A
Authority: GB
Inventors: Michael Salmon Richard
Original assignee: FFEI Ltd
Current assignee: FFEI Ltd
Priority date: 2019-05-09
Filing date: 2019-06-27
Publication date: 2020-11-18
Anticipated expiration: 2039-06-27
Also published as: GB201909278D0; GB2583978B; GB201906559D0

Abstract

A method of detecting an x-ray fluorescence signal emitted from a biological sample, the method comprising: obtaining a biological sample 4a to be analysed, the biological sample 4a comprising at least one biomarker; obtaining at least one heavy-metal atom bound to an antibody to form a heavy metal-antibody conjugate, wherein the conjugate is arranged to emit a fluorescent x-ray signal when excited by an x-ray fluorescence radiation source 10; labelling the biological sample 4a with the heavy atom-antibody conjugate; illuminating the labelled biological sample with an x-ray fluorescence beam 2 emitted from an x-ray fluorescence radiation source 10; detecting an x-ray fluorescence signal emitted from the heavy metal antibody conjugate on the biological sample 4a using an x-ray fluorescence detector 7.

Description

BIOLOGICAL SAMPLE SCANNER

Field of the Invention

The present invention relates to apparatus and methods for detecting biomarkers using heavy atoms bound to antibodies.

Background

Pathological investigations of human tissue-section slides, such as those used for biopsies, have been widely used by light microscopes in medical history. This method is limited to diagnosis from either coloured digital microscope images or bright field scanner images. In order to render the very thin tissue sections visible, the tissue can be broadly-stained with inaccurate dyes such as haematoxylin and eosin (H&E). Alternatively, the tissue can be probed with more advanced enzyme-conjugated antibodies which non-linearly deposit a coloured chemical product localised to a disease-specific antigen, known as immunohistochemistry (I HC). The degree of colour deposition in IHC is reliant on various factors including time, substrate concentration, and biological activity before fixation. This high dependency on a large number of variables introduces great lab-to-lab variation. These two methods vary in the mid-low range of precision, but both suffer from some degree of low quantitafion and high background noise. The low signal-to-noise ratio combined with the limited control of staining accuracy also presents an obstacle to artificial intelligence (Al) diagnostic algorithms.

Fluorescent antibodies are an alternative way of making tissue visible and they are used for more precise binding with lower background noise. However antibody multiplexing, in which multiple measurements are taken in a single experiment, is required to obtain a significant volume of information. This is limited by the number of conjugated fluorophores available and the number of signals that can be reliably layered and deconvoluted. This often requires z-stacking and multi-pass image registration which increases difficulties for Al algorithm processing. Fluorescent scanning is also time-consuming and tissue auto-fluorescence can occasionally occur, introducing elements of randomised noise. Combined, these caveats prevent efficient and accurate automated processing of large caseloads.

It would be desirable to provide a method for detecting biomarkers which does not suffer from high background noise whilst allowing for multiple signals to be detected quickly, simultaneously, and accurately.

Summary of the Invention

According to a first aspect there is provided a method of detecting an x-ray fluorescence signal emitted from a biological sample. The method comprises obtaining a biological sample to be analysed, the biological sample comprising at least one biomarker, and obtaining at least one heavy-metal atom bound to an antibody to form a heavy metal-antibody conjugate, wherein the conjugate is arranged to emit a fluorescent x-ray signal when excited by an x-ray fluorescence radiation source. The method further comprises labelling the biological sample with the heavy atom-antibody conjugate, illuminating the labelled biological sample with an x-ray fluorescence beam emitted from an x-ray fluorescence radiation source, and detecting an x-ray fluorescence signal emitted from the heavy metal antibody conjugate on the biological sample using an x-ray fluorescence detector.

Advantageously, this method allows for the interrogation of biological samples with high precision antibody-labelling, where antibodies are conjugated to heavy metals, whilst allowing for multiple signals to be detected simultaneously.

Advantageously, there is very little background noise because the elemental composition of biological samples falls outside the XRF spectrum of interest A heavy-metal atom may be any atom with an atomic number greater than 19. Preferably, the heavy metal atom is selected from the transition metals, the poor metals, the lanthanides, the actinides, the alkali metals, or the alkaline metals.

A biomarker may be anything which can be used to indicate the presence of a particular disease state or some other physiological state of an organism. That is to say, a biomarker may be a measurable indicator of the severity or presence of a state. In some cases, the biomarker may indicate a change in expression or state of a protein that correlates with the risk or progression of a particular state, for example a disease state.

There may be more than one biomarker present in the biological sample. The biomarker may take a number of different forms, for example the biomarker may be an antibody, an antigen, or a protein. In some cases a first biomarker may be the same type, or form, of biomarker as a second biomarker. For example, there may be a first and a second antigen biomarker. In other cases, a first biomarker may be a different type, or form, of biomarker to a second biomarker. For example, there may be a first antigen biomarker and a second antigen biomarker, wherein the first and second antigen biomarkers are not the same. In cases where the biomarker is an antigen, the antibody may bind to specific antigens, meaning that, in this case, the antibody may bind to a specific biomarker present in the biological sample. Examples of biomarkers may include antigens such as proteins, nucleic acids, carbohydrates and lipids.

The biological sample may take on variety of forms, such as a destroyed cell in solution or in a bodily fluid containing secreted antigen (for example blood with blood cells removed to detect a hormone or drug or biomarker) or even pure antigen i.e. one that has been chemically purified and wants to be identified and/or its concentration determined.

In other examples, the biological sample may be a tissue sample. A tissue sample can take on a number of different forms including processed tissue, cell preparations with or without denaturation, cellular extracts or pure and semi-pure biomarkers. A biological sample may also be a pure, semi-pure, or crude solution containing antigens. In some cases, a biological sample may comprise antigens immobilised to a membrane or a plate or a surface.

The heavy atom within the heavy atom antibody conjugate fluoresces within the x-ray band of the electromagnetic spectrum, when illuminated by an illumination source forming an incident beam of x-ray radiation having a wavelength specific to the heavy atom excitation peak. The fluoresced x-ray radiation of specific energy is detected by the detector. The fluoresced x-rays are radiated from the heavy metal atoms and are picked up in one direction of emittance by the detector.

The detector comprises a planar detector surface and the emitted radiation from the heavy atom antibody conjugate impacts the detector surface substantially normal to the detector surface. The emitted beam of radiation can therefore be thought of as a beam incident to the planar detector surface. In general, the angle of incidence between the emitted x-ray beam and the planar detector surface is substantially 90 degrees, however in some cases this angle of incidence could be in the range 45-90 degrees.

In a similar manner, the biological sample can be approximated as a planar surface such that the x-ray radiation emitted from the x-ray source is incident on the biological surface approximately normal to the biological surface. The emitted beam of radiation from the source can therefore be thought of as a beam incident to the approximately planar surface of the biological sample. In general, the angle of incidence between the incident x-ray beam and the surface of the biological sample is in the range 45-90 degrees.

It should be noted that while the terms "angle of incidence" and "incident beam" have been used, these are merely labels used to help describe the function of the system. The beam emitted from the x-ray source illuminating the biological sample and the beam emitted from the biological sample are two separate beams. That is to say, the beam emitted from the x-ray source is not reflected off the biological sample. Rather, the biological sample emits its own separate beam after it has been illuminated by the x-ray source.

The detector is positioned off-axis with respect to the emitter such that a principal ray emitted from the x-ray source would not be detected by the x-ray detector. In other words, the principal emitted beam is not shone directly onto the x-ray detector.

An angle is therefore created between the beam emitted from the x-ray source and the position of the detector in relation to the passage of the non-absorbed primary beam. Since the emitted beam is not being shone directly onto a detecting surface of the x-ray detector, the smallest angle created between the beam emitted from the x-ray source and the beam incident on the x-ray detector is less than 180 degrees. Preferably, this smallest angle is substantially 65 degrees and more preferably this smallest angle is substantially 45 degrees.

Since the x-ray beam emitted from the heavy metal antibody conjugate on the biological sample tends to take the form of a radiated beam rather than a reflected or scattered beam, only the photons that travel in a straight line from the biological sample to the detector form the angle that is less than 180 degrees.

The labelled biological sample may be illuminated with a beam emitted from a secondary illumination (radiation) source, in addition to the main or primary illumination source. This may provide a means for cross-checking the x-ray signal detected by the detector using detection methods that are currently being used to detect biomarkers, such as optical microscopy.

The secondary radiation source may be distinct, by which we mean spatially and physically separate, from the first radiation source. The secondary radiation source may be the same type of radiation source as the x-ray radiation source.

By "type of radiation source", we are referring to the form of electromagnetic (EM) radiation emitted from the radiation source. For example an x-ray source emits radiation in the x-ray part of the electromagnetic spectrum as so would be referred to as an x-ray type radiation source. Thus, two radiation sources of the same type both emit the same form of electromagnetic radiation i.e. both emit radiation having wavelengths in the same pad of the electromagnetic spectrum.

It follows that two radiation sources of different types each emit electromagnetic radiation having wavelengths in different pads of the electromagnetic spectrum. Thus, when the secondary radiation source is the same type as the main light source (i.e. the x-ray source), the secondary light source is also an x-ray radiation source. Alternatively, the secondary radiation source may be a different type of radiation source to the x-ray radiation source, which means that in some cases the secondary radiation source is not an x-ray radiation source. The secondary radiation source can therefore be a radiation source capable of emitting radiation in any part of the electromagnetic spectrum. Examples of secondary radiation sources include a laser source, an infrared source, a UV source, a bright field or visible light source, or an x-ray source as previously mentioned.

A secondary radiation detector may be used to detect a radiation signal emitted from the heavy-metal antibody conjugate on the biological sample which has been illuminated with the secondary illumination source. The secondary radiation detector may comprise an objective lens and a sensor. For example, the secondary radiation detector may be part of a digital microscope.

Thus, the above described method of detecting an x-ray fluorescence signal emitted from a biological sample allows for the interrogation of biological samples with high precision antibody-labelling, where antibodies are conjugated to heavy metals, whilst allowing for multiple signals to be detected simultaneously with very little background noise. The output received by the X-ray fluorescence (XRF) detector provides a large amount of co-localisation data that may be quickly and efficiently interpreted by Al algorithms and/or humans.

Advantageously, there is very little background noise because the elemental composition of biological samples falls outside the XRF spectrum of interest. The elemental composition of the biological samples therefore has a different spectral response to the spectral response of the heavy-metal atoms. For example, the spectral response of the biological sample may be in a different spectral range to that of the heavy metal atoms. This means that there is little sample-to-sample variation other than of diagnostic significance. There is also less lab-to-lab variation, for example due to slight differences in equipment sensitivity or laboratory conditions (e.g. temperature, humidity) which improves reliability, validity and remote data analysis and telecommunication.

Additionally, XRF is non-destructive, which means that the structure of the biological sample is not affected by the incident x-ray beam illuminating the biological sample. The XRF signal from heavy atoms is quantitative and can be turned off/on like a light-switch with no damage to biological samples, no change in signal sensitivity over multiple readings, or risk of bleaching. The biological tissue sample can therefore be reused at a later date if needed, as the sample has not been damaged by the XRF beam.

Further, as signal detection is of a specific position or peak location on an XRF spectra, it is possible to extensively multiplex or layer antibodies to simultaneously capture all possible co-localised signals irrespective of the amount of overlap. In the above described method, it is the spectral position of the signal that matters. Unlike the case with optical imaging, a focussed image of the biological samples not required in order to detect the spectral peaks. Thus, spectral detection using XRF, rather than a variable 3D-position for imaging, allows for reduced focussing and mechanical tolerances and for increased reliability and speed. This has the advantageous effect that the target does not need to be 'in focus', just in a registered 2D position.

The present invention allows for the development of new antibody 'cocktails' for increased disease-specific diagnostic power and caseload efficiency, meaning that tissue or well-plates or blots can be simultaneously tested for a panel of disease antigens or biomarkers of interest. Data can be output in different forms depending upon the interpretation method that is used. For example, the data may be output as a simple 2D 'heat map', which can advantageously make to simplify interpretation by an analyst (pathologist, technician or researcher etc) more accurate, quicker, and more efficient. A numerical scale can be applied along with 2D coordinates to simplify Al diagnostics of intensity and distribution.

Brief Description of the Drawings

Embodiments of the present invention will now be described by way of example only with reference to the accompanying drawings in which: Figure 1 shows a schematic cross section view of an example biological sample scanning apparatus; Figure 2 is a flow diagram illustrating a method of using the scanning apparatus; Figure 3 shows a periodic table indicating example heavy atoms; Figure 4 shows an example of a tissue sample; Figure 5 shows an example of detected signals; Figure 6 shows a cross section view of an example biological scanning apparatus with a storage system; Figure 7 shows a cross section view of an example biological sample scanning apparatus and an example XRF western blot; and Figure 8 shows an example XRF western blot;

Specific Description

X-ray fluorescence (XRF) is an analytical technique used to determine the elemental composition of a material. It is a non-destructive technique meaning that the structure of the material is not affected by the X-ray beam being shone onto it. XRF determines the chemistry of a sample by measuring the fluorescent (or secondary) X-rays emitted from a sample when it is excited by a primary X-ray source.

When an electron-dense atom in the sample is hit with an X-ray of specific energy or wavelength, an electron from one of the atom's inner shells may be ejected from that shell, leaving behind an electron-hole. In order to maintain atomic structure, an electron of higher energy must occupy the lower energy electron-hole that has been left behind. This occurs by the higher-energy electron losing energy in the form of an X-ray. As the energy to be lost is proportional to the energy of the displaced low-energy electron, the emitted X-ray is of the same energy as the primary X-ray which hit the sample. The energy of this X-ray is equal to the specific difference in energy between the two orbital shells. It is this measurement of energy which forms the basis of XRF analysis.

Each element present in the sample produces characteristic X-ray spectra, which can be thought of as a fingerprint. The spectral fingerprint is unique to each specific element which allows the element(s) present in the sample to be identified based on the detected fingerprint(s).

Energy-dispersive XRF (EDXRF) analyses groups of elements simultaneously so that the elements present in a sample, and their relative concentrations, can be rapidly determined. EDXRF is therefore a means of determining the elemental composition of the sample. After exciting the sample with X-rays, a spectrum of fluorescent X-rays will be emitted having multiple XRF peaks with varying intensities. A graphical representation of X-ray intensity peaks as a function of energy peaks can be created. The peak energy indicates the element present in the sample while the peak height/intensity indicates the concentration of that particular element.

Figure 1 shows an example of a biological scanning system 100. The system 100 includes a sample holder 4 for receiving a biological sample 4a, a radiation source 10, for emitting a beam 2 of radiation, in the form of x-rays, to be shone onto the sample 4a, and a detector 7 for detecting a signal, in the form of x-rays, which are emitted from the sample 4a. In use, the radiation source 10 illuminates the biological sample 4a in the sample holder 4 which causes excitations in the biological sample 4a. These excitations are in the form of x-ray fluorescence and are detected as a signal by the detector 7.

The system 100 includes a main stage 30 in the form of a table having a flat upper surface 32, on which the experiment is carried out, supported by a plurality of legs 34.

The radiation source 10 is positioned above the main stage 30 and directed towards the flat upper surface 32 so that the radiation beam 2 is directed towards the surface 32 of the main stage, typically at a low angle of incidence such as along the normal to a plane defining the flat upper surface 32. The upper surface 32 includes an aperture 36 extending through the upper surface 32 to allow the beam 2 from the radiation source 10 to travel through the main stage 30 without impacting the upper surface 32 The radiation source 10 is an XRF emitter 1, for example a micro-XRF emitter or an energy dispersive XRF (ED-XRF) emitter. The beam of radiation 2 emitted by the XRF emitter 1 is an X-ray beam 2a. The radiation source 10 is fixed relative to the main stage 30 so that the radiation beam 2 is always directed towards the same part of the upper surface 32 of the main stage 30. Specifically, the radiation source 10 is fixed directly above the aperture 36 in the main stage 30. Fixing the radiation source 10 relative to the aperture 36 and the main stage 30 ensures that the main components of the scanning system 100 do not need to be arranged for different experiments. The main set-up can therefore remain the same and only the biological sample 4a to be scanned needs to be changed.

Underneath the main stage 30, below the aperture 36, is a beam stop 3 that is used to stop the primary X-ray beam 2a emitted from the radiation source 10 which has not been absorbed by the sample 4a. The radiation source 10 is positioned above the main stage 30 so that a sample 4a can be placed in the path of the X-ray beam 2a, having a known beam size, with no other obstacles between the radiation source 10 and the beam stop 3.

The system 100 further includes a transportation system 40 which is arranged to move the sample holder 4 relative to the radiation source 10 and main stage 30.

The transportation system 40 includes a support mechanism 42 for supporting the sample holder 4, a motion system 44 for moving the support mechanism 42 within the system 100 relative to the radiation source 10 and main stage 30, and a frame (not shown) to which the support mechanism 42 and the motion system 44 are attached.

The frame 46 extends substantially across the upper surface 32 of the main stage 30 underneath the radiation source 10 and above the main stage 30 so that the support mechanism 42 can be moved across the upper surface 32 of the main stage 30. The transportation system 40 is connected to a computerised control system 50 and motors which control and power the transportation system 40.

The computer control system 50 is connected to all the individual components of the scanning system 100 including the radiation source 10, the detector, 7, and the transportation system 40, and all the sub-components of these systems. All the individual components and sub-components of the system 100 are therefore computer controlled, providing a fully automated, computer controlled scanning system. A computer program runs on the computer control system 50, which can be programmed by a user. The user is able to input details of the experiment into the computer program so that when the computer program is run, the scanning system 100 carries out the required experiment without any further interaction from the user, until the experiment has been completed.

As shown in Figure 1 the support mechanism 42 is in the form of a receiving platform 48 and the motion system 44 is in the form of a shuttling mechanism 49. The sample holder 4 is therefore positioned on, and supported by, the receiving platform 48. The receiving platform 48 is attached to the shuttling mechanism 49, which is arranged to move the receiving platform 48 relative to the fixed components of the system 100 (namely the radiation source 10 and the main stage 30). The shuttling mechanism 49 moves the receiving platform 48 laterally within the system 100, the movement being confined to a single horizontal plane.

The receiving platform 48 can therefore move left and right, across substantially the entire length of the upper surface 32 of the main stage 30, as well as forwards and backwards, across substantially the entire width of the upper surface 32 of the main stage 30. The shuttling mechanism 49 is therefore a multi-directional shuttling mechanism, for example an X+Y shuttling mechanism, which spans the surface area of the upper surface 32 of the main stage 30.

The sample holder 4 including the biological sample 4a, which is supported by the receiving platform 48, is moved across the surface area of the main stage 30 relative to the radiation source 10 by the shuttling mechanism 49. The shuttling mechanism 49 ensures that the biological sample 4a is positioned underneath the radiation source 10, the shuttling mechanism allowing the position of the sample holder 4 and biological sample 4a to be generally adjusted in the x-and y-directions, if necessary. The shuttling mechanism 49 is generally capable of macroscopic movements, which are movements generally of the order of centimetres.

In order to finely tune the position of the biological sample 4a relative to the radiation source 10, the shuttling mechanism 49 comprises an adjustment mechanism, capable of "microscope" movements in the x-and y-directions. By "microscope", we mean fine movements as opposed to large movements. Whilst microscopic movements may include movements on the micrometer scale, they may also be up to an including the millimetre scale.

Biological samples 4a, either labelled with heavy-metal conjugated antibodies or containing naturally localised heavy atoms (such as iron naturally present in blood or calcium naturally present in bone), are loaded into a deposition system (not shown), which includes a deposition mechanism which deposits biological samples 4a onto the sample holder 4 so that they can be imaged. The heavy-metal metal antibody conjugate which is used to label the biological sample 4a is commercially available.

The sample holder 4 may be in the form of a pathology slide but any other suitable holder sample holder 4can be used to receive the biological sample 4a from the deposition mechanism 62. The sample holder 4 is then placed in the path of the X-ray beam 2a, using the transportation system 40, so that the path of the X-ray beam 2a impacts the biological sample 4a perpendicularly.

As the receiving platform 48 on which the sample holder 4 is located is confined to the horizontal plane, all biological samples 4a are positioned the same vertical distance 5 away from the XRF emitter 1. Keeping the vertical distance 5 constant between the biological samples 4a and the XRF emitter 1 maintains a constant sample illumination area between different experiments. That is to say the beam size, having any conal dispersion associated with it, is kept constant at the biological sample 4a In some embodiments, the receiving platform 48 is not confined to the horizontal plane but can additionally move in the vertical plane (the z-direction) as well as the horizontal plane. This means that the biological samples 4a can be positioned at different vertical distances from the XRF emitter 1. Adjusting the distance between the biological sample 4a and the XRF emitter 1 varies the sample contact area of the X-ray beam 2a, meaning that the beam size, and associated conal dispersion, vary according to the vertical distance. This allows the system 100 to either image substantially the whole of the sample holder 4 (by increasing the vertical separation between the biological sample 4a and the XRF emitter 1, which increases the beam size) or to image a localised region of the sample holder 4 (by decreasing the vertical separation between the biological sample 4a and the XRF emitter 1, which decreases the beam size). In an alternative embodiment, this decrease in beam size may be achieved by having two XRF sources at a similar vertical height (i.e. both at z = x). In this case one XRF source has limited beam focus, in order to hit the entire sample area, and the other XRF source has microfocus, in order to hit a precise sub-sample region. However, as will be appreciated, the beam focus could be anywhere between these two end points and so could fall anywhere in between, within this range.

The XRF emitter 1 emits a timed beam 2a of high energy x-rays, by which we mean that the XRF emitter 1 is configured to emit a beam of x-ray radiation for a defined period of time. When the XRF emitter 1 is emitting radiation, the beam is said to be "on" and when the XRF emitter 1 is not emitting radiation, the beam is said to be "off'. The duration over which the beam is on is controlled by the computer system 50 and may be in the micro-second range. However, in some cases, the duration is long enough that the beam can be considered to be in a constant on state for the duration of the experiment. The size of the area of the biological sample 4a illuminated by the x-ray beam is defined by the beam size, which can also be controlled by the computer system 50. During an experiment, the computer system 50 switches the XRF emitter 1 on so that it emits a first beam, for a specific duration, illuminating an area of the biological sample 4a which corresponds to the beam size. After the pulse has been emitted, the signal emitted from the biological sample 4a is detected by the detector 7. After this detection, the beam may be turned off and the data can be processed. The biological sample can then be moved relative to the XRF emitter 1 and the beam may be subsequently turned on again so that the sample is further illuminated by another beam. The subsequently emitted signal from the biological sample 4a is again detected by the detector 7 and the data is processed. This beam on, illumination, detection, processing, beam off, reposition detection cycle can be repeated for the duration of the experiment. As will be appreciated, in some cases, the biological sample is not repositioned between subsequent beam-on, beam-off cycles. An advantage of keeping the position of the biological sample constant between detections is that multiple detections at the same position may help validate detections. Within a particular experiment, the duration of each timed beam is typically kept constant. However, between experiments the duration of each timed beam, as well as the total duration of the experiment, can be adjusted, for example longer or shorter durations.

In some examples, rather than the computer system 50 controlling the XRF emitter to be either on or off, the computer system 50 controls the XRF emitter 1 in such a way that the beam is pulsed. In this case, a pulsed beam refers to a beam that very quickly alternates between an on and an off state within one overall "on" instruction given by the computing system. When a pulsed beam is used, multiple beams are emitted in quick succession to illuminate the biological sample and so multiple beams are emitted from the biological sample and detected by the detector. This all happens within one session, where a session is equivalent to one on-off cycle of the above described timed beam. Thus, in this case, all the pulses have the same duration and there is no repositioning between pulses. However, between subsequent pulse sessions, the duration of the pulses, as well as the position of the biological sample relative to the XRF emitter 1, may be adjusted as before.

As will be appreciated, the timing of the beam will depend on the particular experiment or application being carried out. For example, a low concentration of heavy-atom antibodies labelling a tissue sub-region may require a longer exposure time in order to generate an appropriate signal. Thus, in some cases the incident beam time may be in the second range or the multi-second range. Once the incident X-ray beam 2a has impacted the biological sample 4a in the sample holder 4 and excited the heavy atoms labelling the biological sample 4a, a secondary, fluoresced X-ray beam is emitted from the heavy-atoms in the sample 4a which is detected by the detector 7. In this case, the detector 7 is an energy dispersive (ED) detector 7a.

The ED detector 7a detects XRF energy peaks in the spectrum of the emitted secondary X-ray beam of any heavy metal which has a statistically significant concentration, down to ppm (parts per million) range (for example, 105 -100s ppm, or 1000s -10000s ppm for very high concentrations), below the emitter's maximum energy (subject to and selectable to experimental requirements) in one spectral reading. This detection reveals the presence of the target, i.e. the heavy metal atom, within the given beam size.

The general method of detecting an x-ray fluorescence signal emitted from a biological sample can be summarised as follows, in combination with Figure 2.

Firstly, a biological sample is obtained 3101, the biological sample comprising at least one biomarker. A heavy-metal atom antibody conjugate is then also obtained 3102, the heavy-metal atom antibody conjugate comprising at least one heavy-metal atom bound to an antibody and the heavy-metal atom antibody conjugate being arranged to emit a fluorescent x-ray signal when it is excited by an x-ray fluorescent radiation source. The biological sample is then labelled 3103 with the conjugate. After the biological sample has been labelled, the labelled sample is loaded 3104 into a deposition system which deposits 3105 the labelled biological sample onto a sample holder. The transportation system then moves S106 the sample holder with the labelled biological sample into an illuminating position, in which the biological sample is positioned in substantial alignment with an emitter axis of the XRF emitter so that the XRF emitter can illuminate the biological sample. Here, the emitter axis is an axis substantially perpendicular to a planar surface of the XRF emitter, along which the XRF beam is emitted. Once the sample has been moved to the illuminating position, the biological sample is illuminated 3107 with an x-ray fluorescent beam which is emitted from an x-ray fluorescence radiation source, which in this case is the XRF emitter. The x-ray fluorescent beam is a pulsed or timed (including constant) beam which is turned on, shone onto the biological sample to illuminate an area, then turned off. The x-ray fluorescent signal that is produced by the heavy metal atom conjugate on the biological sample is detected S108 by an x-ray fluorescence detector. In some cases, between subsequent beam illuminations the biological sample can be moved laterally relative to the radiation source so that the radiation source can illuminate a different area of the biological sample to the area that was illuminated by the previous beam.

Figure 3 is a periodic table of elements showing a first section 200 from which a heavy atom may be chosen and a second section 300 which includes elements that are not suitable choices for heavy metal atoms. As can be see, heavy metal atoms can be generally chosen from the alkali metals, the alkaline earth metals, the lanthanides, the actinides, the transition metals, the poor metals, and metalloids. The heavy metal atoms are generally not chosen from the noble gases, halogens, or the non-metals. This is because this group of elements, i.e. the elements in the second section 300, (for example C, H, 0, N, P, S) are "biological" atoms and so are not in the spectral response range of interest. That is, these elements generally appear "invisible" to the XRF detector 7a. Advantageously, therefore, labelling the sample 4a with a heavy metal atom means that the biological atoms present in the sample are not detected and only the heavy metal atom of interest is detected. This removes unnecessary and unwanted biological noise from the detected signal.

In many cases, the proportion of the total surface area of the sample holder 4 taken up by the biological sample 4a is larger than the size of the area illuminated by the X-ray beam, which is dependent on the beam size. That is to say, the biological sample 4a will typically extend in the x-and y-horizontal directions such that the total surface area taken up by the biological sample 4a on the sample holder 4 is larger than the cross section of the X-ray beam 2a. In order to perform a complete spectral analysis of the biological sample 4a, the entire biological sample 4a needs to be irradiated with the X-ray beam 2a. The biological sample 4a therefore needs to be repositioned relative to the XRF emitter 10 to ensure that the entire biological sample 4a has been irradiated.

Using the shuttling mechanism 49 of the transportation system 40, the biological sample 4a can be accurately repositioned relative to the XRF emitter 1. This is done in such a way that an area of the biological sample 4a which is currently being illuminated is immediately adjacent to the previously illuminated area. This ensures that the total area of the biological sample 4a is exposed to the X-ray beam 2a.

The tissues sample 4a can therefore be thought of as being made up of a series of target areas 11, each target area 11 being the same size as the beam size, or illumination area, of the X-ray beam, as shown in Figure 4. These individual target areas 11 can be thought of as forming a virtual 2-dimensional (2D) grid 12 over the area covered by the biological sample 4a.

The computer control system 50 moves the receiving platform 48 in the x-and y-directions so that all the individual target areas 11 which make up the virtual 2-dimensional grid 12 are successively placed underneath the XRF emitter 10 and irradiated by the X-ray beam 2a, until the entire biological sample area has been irradiated The XRF detector 7a is connected to an amplification and spectrum generation system 70 which forms part of the computer control system 50. The amplification and spectrum generation system 70 includes signal amplification and spectrum generating software which can present the secondary X-ray beams, returned from the biological sample 4a, as a visual relative heat-map or numerically with 2D-coordinates. The 2D-coordinates are the 2D-coordinates associated with each individual target area 11 of the virtual 2D grid 12. The heat-map, or 2D-coordinates, can be used to analyse the biological sample 4a as well as being used for Al. In particular, Al can be designed and used to interpret the xylocation with the intensity of each target heavy-metal atom detected at each xyposition. The Al receives quantitative and spatial data in a simple format that has significantly less of the "noise" of the other biological components of the biological sample 4a. This cleaner data, used by the Al, can be interpreted more accurately and quickly compared to intensity/spatial data acquired from bright field images and interpreted by a human pathologist.

Al could also be integrated into the workflow. For example, if a sample 4a were exposed to the XRF beam 2a for x seconds but, due to low sensitivity, analysis was difficult, the Al could automatically initiate exposure for 2x seconds or 3x seconds to create a more accurately measurable signal. The relative signal can be mathematically interpolated when compared between samples of differing sensitivity.

Al can therefore make data collection more efficient without needing human decision time because the XRF signal doesn't degrade with exposure time, leading to an improvement in workflow.

As each biological sample 4a may include more than one type (where "type" is used to refer to an element) of heavy metal atom either from heavy metal atom-antibody conjugates or from heavy metal atoms that are naturally occurring in the sample 4a, the 2D-grid 12 for a particular biological sample 4a is represented as layered heat maps, one heat map for each heavy metal atom that has been detected over the whole of the biological sample 4a. The different heavy metal atoms present in the biological sample 4a are detected substantially simultaneously by the detector 7 because the detector is able to receive the entire spectrum emitted by the heavy-metal atom conjugate and so is therefore able to receive each signal emitted by each heavy-metal atom present at substantially the same time. In this respect, the detector is an imaging detector in that the detector is able to pick up a spectrum from each target area 11 of biological sample 4a. If a heavy metal atom is not detected over the whole of the biological sample 4a but only over part of the biological sample 4a, this gives spatial information about that particular target heavy metal. For example on a liver sample an antibody against liver would be detected everywhere, however an antibody against cancer would only be detected in sub-areas of the liver where the cancer is present. The areas where the liver-antibody and the cancer-antibody co-locate confirm liver cancer in a specific pad of the liver sample, or in multiple parts if spreading (metastasising).

The different heat maps can then be superimposed on top of each other, using the grid coordinates to ensure that the heat maps have been correctly overlaid, to reveal regions within the biological sample 4a of target co-localisation, i.e. regions which have a particular heavy metal atom label in common with each other. There can be many heavy metal atom labels per biological sample 4a, the only restriction being that there is one particular element-antibody label combination per sample. For example, one sample may have both Au-antiliver and Pb-anfiliver because these are different label combinations but one sample cannot have Au-antiliverA and Au-antiliverB because these are two of the same label combinations within one sample. This latter example would be allowed across two separate samples, as two discrete experiments.

The intensity of the signal from the secondary X-ray beam emitted per target area 11 relates to the concentration of specific heavy metal atoms, as shown by the peaks taken from an example experiment in Figure 5. The intensity of the signal can therefore be quantified and numerically represented such as for Al diagnosis. This may facilitate auto-diagnosis of diseases if the quantity of biomarker indicates severity. For example, no cancer marker (i.e. Oppm) indicates all clear, cancer marker of an amount 100ppm could indicate stage 1, and cancer marker of an amount 1000ppm could indicate stage 2 etc. Furthermore, Al allows clear YES/NO decisions to be made in relation to a control heavy metal-antibody. For example, if the equipment is working correctly, in that there is a control heavy metal atom-antibody conjugate which gives a signal, but there is no disease marker then this indicates a null result. If there is a control present and lots of disease marker this indicates that emergency treatment is required. Finally, this technique could also be used to supplement current Al methods which are based upon bright field images, e.g. if Al is 75% sure a bright field diagnosis is correct a simple XRF 'augmented overlay' can be used to confirm or reject the 25% outstanding certainty.

In more detail, using EDXRF, for each target area 11 in the 2D grid 12 a single spectrum is produced containing peaks of all the specific XRF energies produced in the target area 11 which are below the maximum energy of the X-ray beam 2a. Every heavy metal atom fluoresces at a very specific energy, as a product of energy-specific dislocation of a lower-energy shell electron relative to the electron number of the heavy metal atom (as explained previously).

Therefore, each heavy metal present in the target area 11 will have a separate peak that can be detected by the XRF detector 7a, analysed, and then extracted.

As an example, Figure 4 shows an example heat map that might be generated and illustrates what might be detected if a theoretical "antibody cocktail" of Fe-anti-hepatocyte, Cu-anti-hepatic cancer 1, Pb-anti-hepatic cancer 2, Os-anticirrhosis, and Ag-anti-hepatitis is applied to a liver pathology slide.

From the heat map there can be seen an intense region 90 of XRF. The corresponding spectral reading, shown in Figure 5, shows that liver-generic Fe has bound, with a lower concentration of cancer-marker 1 Cu bound, and a high concentration of cancer marker 2 Pb bound. The two Pb peaks arise due to the difference in two electron shells (K or L) perturbed by the X-rays 2a at two different energies under the beam maxima. With pre-knowledge of the metals applied, this discrete XRF split-signal can be easily accounted for. In this case, as no signal is seen from cirrhotic or hepatitis markers, this confirms the presence of cancer, with a 'second-opinion' made automatically by using two discrete heavy-atom antibodies. The grade of cancer and the spread (metastasis) may possibly be diagnosed from the heat map distribution and intensity differences of the two positive cancer markers. Treatment and prognosis can therefore be inferred.

The deposition system 60 is arranged to deposit multiple biological samples 4a onto the sample holder 4, each biological sample 4a being confined to a specific, discrete area on the sample holder 4. After the current biological sample 4a has been processed and analysed, the overall process is repeated across all the remaining biological samples 4a on the sample holder 4.

In some embodiments, the scanning system 100 is connected to a storage system 80, shown in Figure 6, which stores the sample holders 4 which have had tissues samples 4a deposited on them, before the biological samples 4a are analysed. The storage system 80 is therefore also in communication with the deposition system 60 so that the prepared holders 4 can be passed from the deposition system 60, to the storage system 80, and then finally to the scanning system 100. It should be noted that the storage system 80 is optional in that the storage system 80 is not required for experiments to be carried out, however the presence of a storage system will increase the overall throughput. In some cases the storage system includes an incubation system, which is is not required for experiments using fixed tissue but may be required for experiments which use live cells.

The system 100 also includes an automated stacking system which allows multiple sample holders 4 to be stacked within a cavity inside. When the system 100 includes a storage system 80, the automated stacking system may be located within the storage system 80. The holders 4 are typically stacked and stored vertically within the storage system 80, however it will be appreciated that the sample holders 4can be organised in any other convenient arrangement. Storing multiple holders 4 in the storage system 80 allows the possibility of loading and scanning many different holders 4 one after another, allowing multiple experiments to be performed automatically one after another, without the need for user intervention between changing subsequent holders 4.

Thus, once all the biological samples 4a on the current sample holder 4 have been analysed, the sample holder 4 is returned to the cavity in the storage system 80 (or stacking system when the storage system 80 is not present), via a transfer system, and the next sample holder 4 in the stack of s is retrieved from the cavity, via the transfer system, and brought into the same exposure, or imaging, location relative to the XRF beam 2. The relative vertical distance the emitted XRF beam 2 has to travel from the emitter 1 to the sample holder 4 is therefore kept substantially constant between subsequent sample holders 4. The transfer system therefore provides a mechanism for high-capacity loading of biological samples 4a into the scanning system 100 for increased throughput.

In summary, the biological scanning system 100 incorporates a new use of a micro-XRF radiation source into the framework of a biological sample scanner, and in doing so allows for the exploitation of a new chemical signal for this type of analysis On particular, heavy-metals, such as in metal-antibody-conjugates) as a target and consumable for biological research detection.

The biological scanning system 100 for detecting heavy metal XRF excitations will reduce or substantially eliminate background noise because tissue, cells, antibodies, and borosilicate glass (an example material used for the sample slide) are 'invisible' to the XRF spectrum proposed, whilst presenting quantifiable data in a format that is simply interpreted by a human analyst and Al algorithms alike.

This scanning system increases the ability to detect multiple overlapping signals, compared to current systems and methods. Many different heavy metals can be simultaneously detected in one position at a single time, which allows for increased data from multiplexing and detection as well as quantitation of accumulated heavy metals in biological samples.

Advantageously, the scanning system can be contained within a compact, bench top scanner, providing a system that will be user friendly for all skill levels and simply integrate into current laboratory workflow.

In some arrangements, the scanning system 100 includes a camera and/or light source to provide a thumbnail image to register a sub-section of the sample 4a, for example a tissue boundary on a pathology slide or the inside of a microwell 4c. By not exposing an empty pad of the slide or a non-relevant sample holder to the XRF beam, the overall speed of the system 100 in increased.

In some arrangements, manual control of the XRF emitter can be provided so that a user is able to define a sub-region, or regions, of interest such as a specific area of tissue or particular section of a well-plate. In other arrangements, a user can control the area of interested using the computer system to define the sub-region or regions to be imaged and scanned.

A loading mechanism can also be incorporated for sample variation or larger slides (for example single-, double-and whole organ-width slide-sections, or 6-, 12-, 24-, 96-, 384-well plates from different manufacturers, or a frame for holding membranous material flat).

Many experiments require a compromise between beam-size versus sensitivity versus cost. For example cytology using a low-concentration of target-biomarker results in small beam size with high sensitivity but at a high cost. A larger biological sample with a low-concentration of target-biomarker results in a large beam size with high sensitivity at medium cost whereas a high concentration of target-biomarker results in low sensitivity at low cost. The proposed system allows for sample-format variation and disease specific variation, allowing the user to adjust the beam size, sensitivity, or cost as necessary. Beams sizes may, for example, range between the micrometre scale (pm x pm) at the small end of the range and the millimetre scale (mm x mm) at the large end.

Although the primary application of the above described technology is for digital pathology, the technology behind the XRF scanner can be used in any number of other biological applications.

In one alternative application, heavy-metal-conjugated antibodies can be used rather than IHC (immunohistochemistry) to localise and immobilise heavy metal atoms to known biomarkers for diagnostic and research purposes.

In another alternative application that uses heavy-atom antibodies on tissue-slides, a method that multiplexes heavy-metal-conjugated antibodies, potentially in the composition of an off-the-shelf, all-in-one, pre-validated antibody 'cocktail kit' could be used for the purpose of: a) providing at least one control antibody for a universal antigen to confirm scanner functionality and identify biomarkers that confirm specific cell-types; b) simultaneously detect multiple biomarkers co-occurring in specific diseases or cell-types using 2+ different antigen, disease/target-specific labels for data redundancy and/or a 'second opinion'; or c) providing disease specific antibody cocktails and scanning for combinations of signals to determine which disease is present for example applying a disease-specific antibody cocktail, i.e "Liver Cocktail A" (anti-cancer, -cirrhosis or -hepatitis) to liver tissue and scanning for combinations of "A" signals.

As the skilled person will appreciate, the above scanning system 100 can be adapted slightly for specific uses, some of which are described below.

Western Blotting with XRF One example of a specific use is western blot multiplexing. A western blot is an analytical technique for detecting specific proteins in a sample of tissue. Western blotting uses gel electrophoresis to separate the proteins, by length of polypeptide relative to mass, and then transfer the proteins to a membrane where they are probed with antibodies specific to the target protein. Western blot multiplexing is a form of lane-gridding analysis which allows simultaneous detection and analysis of multiple protein targets in a single sample on the same blot at the same time.

More specifically, using this technique, the first step is to separate the proteins in a sample. To do this, the sample proteins are denatured with a specific detergent (for example sodium dodecyl sulphate (SDS)) to linearise and net negatively charge them. Wien combined with loading buffer to aid in sample-load sinking and coloured tracking of the running front, the sample is then loaded into a well above a designated gel lane. The sample proteins are vertically pulled through a polyacrylamide gel matrix by an electrical current (Polyacrylamide Gel Electrophoresis (PAGE)). The speed at which the protein is pulled through the gel matrix is proportional to the size of the protein; the bigger the protein, the slower it transits to the positive electrode. This is known as SDS-PAGE.

The separated proteins are then transferred, or blotted, onto a second matrix (also referred to as a binding membrane), in order to facilitate detection of one protein species from a mix of thousands, such as but not limited to cell-extractions from cell biology experiments. The separated proteins are blotted onto a high-protein-binding 2D membrane (such as, but not limited to, nitrocellulose or polyvinylidene difluoride (PVDF)) by horizontal electrophoresis with the membrane acting as a barrier to the positive electrode. The remaining parts of the membrane not containing the proteins are then blocked or masked using a blocking agent in order to prevent non-specific binding of antibodies to the surface of the membrane. Any suitable blocking agent can be used, for example milk or bovine serum albumin. The transferred, or blotted, protein is then probed with a combination of antibodies, one primary antibody being specific to the protein of interest and another being specific to a host species of the primary antibody. This forms the biological sample that is used in the western blot.

In more detail, after blotting, the membrane is amenable to probing of individual proteins by antigen-recognition by protein-specific primary antibodies. The chosen antibody (made in a known source, for example a mouse) will specifically bind the target protein and in turn present a source-derived epitope (i.e. part of an antigen to which an antibody attaches itself) from the primary antibody for binding by a source-specific secondary antibody (for example, rabbit anti-mouse).

Using current methods, this secondary antibody is conjugated to a chemiluminescent enzyme such as horseradish peroxidase which, when supplied with substrate, will release a photon of light as a by-product. In a dark-room, X-ray film can be exposed to the blot, and when developed there will be a black band where the light has interacted with the film. This reveals a specific protein's presence out of the mix of proteins as well as the protein's mass when compared to the position of protein mass markers. The same probing process can then be applied to detect specific control proteins, such as those ubiquitously expressed in cells (alpha-tubulin/beta-actin etc.), which provides a normalisation signal for quantitation.

Different protein targets and different antibodies have different combinations of binding efficiency. The current signal detection of photons impacting photographic film means that the number of photons released relates to the intensity of the signal. This means that a poor-binding combination could result in exposure times of hours-to-days to get a low and noisy signal, or a very efficient combination may take seconds to give a massive signal that bleaches the film. This hinders efficient processing of multiple parallel blots due to exposure time variation as well as preventing simultaneous multiplexing of antibodies that have a range of efficiencies. Furthermore, the signals detected using this photographic method are identical and differ only by positional mass on the blot. This means that if two protein targets of extremely similar masses were to be simultaneously probed, the two bands will merge together and mask each other. This is a further barrier to multiplexing with the current methodology.

In addition, between each step (blotting, target-primary, target-secondary, target-exposure, control-primary, control-secondary, control-exposure) there are extensive washing and blocking steps required (30-60 min each), which create large time-delays. Furthermore, for efficient binding, antibodies need to be applied for a sufficient length of time (typically 1-2 hrs per antibody). This means that each blot can take up to 2 days to process, and typically only gives 1 target result per blot. Although blots for different targets can be performed in parallel, this requires greater planning capabilities, induces time-lag for multi-tasking, and is still subject to the practicalities of differences in secondary antibody signalling efficiency.

Finally, in order to be able to compare identical sample contents across multiple blots a reliable supply of sample material is required. However, typically samples from cell-biology are very low yield and are therefore unlikely to be able to provide enough material for more than 2-3 blots.

Instead of using chemiluminescence, as described above, the principle described earlier involving heavy-atom-tagged antibodies can be used which emit a fluorescent X-ray signal that can be detected, as explained previously.

As explained, protein antigens in a mix are separated into lanes according to mass by SDS_PAGE and subsequently blotted onto a membrane. Antibodies conjugated with enzymes that release photonic products are used to reveal mass (kDa) based on their position on an exposed X-ray film. Western blot multiplexing with heavy atom antibodies can simultaneously assess a very large number of proteins on one membrane. In particular, this technique allows simultaneous assessment of as many proteins as there are XRF heavy-atoms and suitable antibodies in multiple combinations. Advantageously, blotting with heavy-atom antibodies therefore allows the detection of multiple proteins on a single blot, due to the discrete and unique spectral fingerprint of each heavy-atom antibody combination. This is quicker, cheaper, and more accurate as it does not require the time or data alignment of carrying out an individual blot per target under investigation.

The apparatus used for XRF western blotting is generally the same as the system 100 described previously, as so will not be described in detail again. An example of a system used for XRF western blotting is shown in Figure 7 where like reference numerals are used for like features.

In general, the system provides a means for multiplexing antibodies to probe for multiple target proteins simultaneously on one blot, requiring only one blocking-primary-secondary protocol. This is achieved using X-ray fluorescence in which secondary antibodies are conjugated to specific heavy metal atoms rather than to a chemiluminescent enzyme. As each specific heavy metal will give a specific XRF peak when probed with X-rays, each secondary antibody gives a different signal that can be assessed simultaneously with the same detection equipment.

In some cases, it is possible to use just primary antibodies if the antibody-protein interaction has been proven to be specific and sufficiently reliable.

As mentioned previously, detailed experiments using chemiluminescent western blotting technology require multiple blots to avoid signal obscurity, consuming precious sample, with increased experimental time and experimental variability leading to inaccurate results and low reproducibility. XRF western blotting provides a new data source that can be easily multiplexed with no signal obscurity on one blot for increased speed, decreased workload, sample conservation and statistical relevance. In particular, when a western blot is probed with all the antibodies required to detect all the multiple proteins separated by mass under investigation from the mix in each sample, a simple gridding of XRF spectra across the western blot will allow identification of positive binding of probes, relative to a control antibody target for normalisation, and give positional distribution of the target-specific signals relative to each other, thus identifying relative masses in one exposure.

The western blot membrane is held flat by a clamping mechanism (not shown) and the sample lanes of the blot are gridded by the XRF beam 2a, in the same manner as described previously. The 2D membrane moves in the x-and y-directions, i.e. in the horizontal plane, within the path of the static XRF beam 2a. This ensures that each part of the 2D membrane is illuminated by the stationary XRF beam 2a, when it falls within the path of the XRF beam 2a, in order to grid the blot in one process.

As shown in Figure 7, the blot membranes are loaded into a hollow tray mechanism 15, comprising a plurality of hollow trays 16, that holds the membrane flat against the sample holder 4 so that the X-ray beam 2a impacts, or illuminates, the membrane substantially perpendicularly to the surface of the membrane. Here, the hollow trays 16 perform the function of the previously described receiving platform 48. The hollow trays 16 include an adjustable frame 17 which allows the clamping mechanism to move in the x-and y-directions (i.e. in the horizontal plane) in order that variations in the blot membrane dimensions can be taken into account. This enables the same sample holder 4 to be used with a variety of membranes having different dimensions.

In one particular example, referring to Figure 7, the proteolytic cleavage of two protein domains is under investigation. In this example, an Fe-bound secondary antibody is used to detect the first anti-domain 6 and an Ag-bound secondary antibody is used to detect the second anti-domain 13. As shown in Figure 7, a first region 6a within a particular lane 12 reveals that the first anti-domain 6 is present at a higher mass than that of the second anti-domain 13 in a second region 13a. This is because the first region 6a did not travel as far down the gel from which the domain was blotted so therefore has a greater mass than the second region 13a. A third region 14a within the same lane 12 reveals a larger mass that corresponds to a summation of the signals from the first region 6a and a second region 13a, therefore revealing partial proteolysis of the third region 14a into the first anti-domain 6 from the second anti-domain 13 under the experiment conditions. The relative XRF signal indicates the ratio of each domain within each band. The apparatus allows data to be collected quickly, simultaneously, and to be auto-processed for multiple band ID, quantification, and presentation of experimental results As can be seen, the ability to multiplex heavy-metal-bound antibodies and simultaneously detect them when the antibodies are co-localised means that more than one target protein can be assessed at a time, increasing experimental scope. For example, in biological systems such as cells, proteins can perform multiple biological functions or induce gene-expression by occupying different structural conformations.

This is illustrated in the theoretical example shown in Figure 8. In this example, the effect of the proteolysis of Protein 1 (P1) on the expression of Protein 2 (P2) is investigated. In this experiment, five antibodies have been applied simultaneously, each with different specificity and each conjugated to a different heavy metal, therefore giving individually quantifiable XRF peaks in each gridded-box.

The XRF-blot in Figure 8 shows that the control protein (such as alpha-tubulin or beta-actin), which is constantly expressed, is equal across cell treatments and so the results of the experiment can be normalised to this.

Column A shows the results for untreated cells. In the absence of any proteolysis stimulus, only full-length P1 is detected by anti-full-length P1.

Column B shows the results for cells that have been treated with a low concentration of P1 proteolysis stimulus. In the presence of a low stimulus, P1 is partially broken into lower masses la and lb, detectable by I a-and lb-specific antibodies, but this detection is below a threshold that induces gene-expression of P2.

Column C shows the results for cells that have been treated with a medium concentration of P1 proteolysis stimulus. In the presence of a medium stimulus, it can be seen that more P1 is converted to la and lb to a threshold which is just above the threshold that induces low expression of P2.

Column D shows the results for cells that have been treated with a high concentration of P1 proteolysis stimulus. In the presence of a high stimulus, all of Pus converted to la and lb, and P2 is highly expressed.

Figure 8 therefore shows that as P1 decreases and la increases! P2-expression increases proportionally. In column D when the strong la and P2 signals of similar mass overlap, their individual XRF signal-peaks can still be separately interpreted without masking each other.

With current methods, such as chemiluminescent western blotting, five antibodies would need four blots (using the same control protein on each blot), each blot requiring two stages of detection (control and target) of varying exposure times. This gives a total of eight reads using current methods. Alternatively, the same blot could be stripped and re-probed four times, however this is time-consuming and can leave artefacts if not properly stripped. In this case the detection signal of each probe is the same. On the other hand, using XRF western blotting requires only one detection stage from one blot, with each target simultaneously detected without any masking as each probe has a different XRF signal.

ELISA with XRF: "HALISA" A further example of a specific use of the scanning system 100 for detecting and measuring antibodies, antigens, proteins and glycoproteins in biological samples is based on the well-known enzyme-linked immunosorbent assay (ELISA).

ELISA (enzyme-linked immunosorbent assay) is a plate-based assay technique for detecting and quantifying biological elements, for example peptides, proteins, antibodies, and hormones. ELISAs use specific immobilised capture antibodies to bind the target antigen, and a detection antibody system to indicate the presence and quantity of antigen binding. In an ELISA, a capture antibody is immobilized to a solid surface, such as the base of a well-plate, and used to capture, immobilise, and present an antigen to a specific primary detection antibody, followed by a secondary antibody that is linked to an enzyme. Detection is accomplished by assessing the conjugated enzyme activity via incubation with a substrate to produce a spectrophotometrically measureable product. The most crucial element of the detection strategy is a highly specific antibody-antigen interaction.

ELISAs are typically performed on plates such as polystyrene plates, each plate having a plurality of individual wells, which will passively bind antibodies and proteins. Having the antigens immobilized to the microplate surface makes it easy to separate bound from non-bound material during the assay. Nonspecifically bound materials are then washed away so that detecting and measuring specific analytes can be done.

Using the commonly known 'sandwich' ELISA technique, the first step of the process is to coat the surface of a protein-binding well plate with a specific capture antibody. Unbound capture antibody is then washed off to prevent nonspecific binding of other molecules, which would create a high background-tonoise ratio which obscures the accuracy of the assay. Blocking buffers are used to coat the protein-binding binding regions of the plate that are not bound by the capture antibody. Common blocking buffers contain random non-relevant biomolecules such as bovine serum albumin (BSA). Thus, all unbound sites on the plate are blocked by coating these sites with the blocking agent. The plate is then washed to get rid of excess blocking agent so that a fluid biological sample can then be added to the plate. Samples may contain antigens specific to a particular health condition to which the antibodies can bind, or a purified antigen to be quantified or subject to kinetic analysis. This means that any antigen present in the fluid sample will bind to the capture antibody which is already coating the plate. Excess sample is then washed from the plate.

The next step is to add a primary antibody to the plate. This primary antibody is specific to the target antigen, and therefore will be immobilised by binding with the antigen immobilised to the plate by the capture antibody. The primary antibody thus labels the antigen with high specificity and presents a source-derived epitope that can be bound by a general secondary antibody. Once excess primary antibody has been washed off, a secondary detection antibody can be applied. Typically, this is an antibody labelled with an enzyme such as alkaline phosphatase to form an enzyme-antibody conjugate. After a final wash to remove any further unbound secondary antibody, a substrate is added which is able to produce a detectable signal, usually a colour signal as a result of a chromogenic reaction by the conjugated enzyme, which can be measured. Thus, if the antibodies 'sandwich' the target antigen, an enzymatic reaction can take place and a signal is produced which can be detected and measured, so as to determine the antigens present in the sample. The more chromogen that appears in the solution the more antigen is present. The change in chromogen concentration in the solution can be quantified through spectrophotometry.

There are other variations on ELISA. The simplest uses antigen directly bound to the surface of the plate, and only a specific primary antibody conjugated to a detection enzyme as a detection antibody. This direct assay is faster and cheaper, and removes chances of secondary antibody cross-reactivity, however immunoreactivity of the primary antibody may be compromised, with reduced antibody choice and attenuation of signal amplification. Another variation is an indirect assay, in which antigen is again directly immobilised to the plate surface, and the primary and secondary antibody detection system is used. This increases the range of antigens that a general secondary antibody can be used to detect whilst maintaining the immunoreactivity and sensitivity of the specific primary antibody. Though this may amplify the signal more, this indirect method requires an extra experimental step then a direct assay and increases the chance of secondary antibody cross-reactivity. A simplified variation of the sandwich ELISA may utilise a capture antibody and primary antibody conjugated to an enzyme as a detection antibody, offering the selectivity of a capture method with increased speed, subject to the immunoreactivity caveats of a direct primary detection antibody.

All ELISA methods are limited by only being able to measure one antibody-antigen interaction per well (i.e. equivalent to one XRF target area) due to the diffuse colorimetric signal. The biochemical development of the colourimetric signal also requires time for the enzyme-conjugate to provide coloured product, the efficiency of which is therefore also subject to chemical environment such as pH, salt, substrate concentration and temperature.

An alternative technique, which the applicant has coined the term "HALISA", is a form of full micro-well analysis, based on the ELISA technique but using heavy atoms to produce the detectable signal, rather than detection with enzyme-products. The heavy atoms are bound to antibodies forming a heavy atom-antibody conjugate as previously described. The sample with the heavy atom-antibody conjugate used as the detection antibody in any variation of ELISA is then illuminated by an XRF beam, as described previously and the subsequently emitted x-ray is detected by the XRF detector.

Advantageously, using heavy atoms to produce a detectable signal means that heavy atom antibodies could multiplex in one micro-well to analyse complex antigen mixes simultaneously with an XRF scanner. Starting from the apparatus shown for example in Figure 1, a user simply needs to switch the receiving platform 48, which is configured to support a sample holder 4, to one that is configured to support a well plate. The shuttling mechanism 49 is then able to move the switched receiving platform 48, holding the well plate, in the x-and y-directions, as before, so that all well plates, in a multi-well well plate, can be illuminated by a single XRF beam 2a, so that the whole of the well plate is processed in one go. HALISA is therefore able to analyse more data and faster than any comparable ELISA method with a reduced variance, as signal is stimulated only when excited by XRF for a known period of time and is not subject to biochemical conditions that require time to develop and may introduce well-to-well or day-to-day variation.

The general apparatus and system used for HALISA is the same as that described previously, for example that shown generally in Figures 1, 4, or 6, and so will not be described again.

As will be appreciated, any feature described with reference to one specific application can be used in conjunction with another application described herein.

Claims

Claims 1. A method of detecting an x-ray fluorescence signal emitted from a biological sample, the method comprising: obtaining a biological sample to be analysed, the biological sample comprising at least one biomarker; obtaining at least one heavy-metal atom bound to an antibody to form a heavy metal-antibody conjugate, wherein the conjugate is arranged to emit a fluorescent x-ray signal when excited by an x-ray fluorescence radiation source; labelling the biological sample with the heavy atom-antibody conjugate; illuminating the labelled biological sample with an x-ray fluorescence beam emitted from an x-ray fluorescence radiation source; detecting an x-ray fluorescence signal emitted from the heavy metal antibody conjugate on the biological sample using an x-ray fluorescence 15 detector.
2. The method of claim 1 wherein the antibody specifically targets the biomarker.
3. The method of claim 1 or claim 2 wherein the biomarker is an antigen.
4. The method of any preceding claim wherein the biomarker is a protein.
5. The method of any preceding claim wherein the biological sample is a tissue sample.
6. The method of any preceding claim wherein the x-ray fluorescence radiation source impacts the biological sample substantially perpendicularly to a planar surface area region of the biological sample.
7. The method of any preceding claim wherein, during the illuminating step, a spatial separation between the biological sample and the x-ray fluorescence radiation source is kept constant.
8. The method of any preceding claim wherein, before the illuminating step, the method comprises fixing a position of the biological sample relative to the x-ray fluorescence radiation source.
9. The method of any preceding claim further comprising moving the biological sample relative to the x-ray fluorescence radiation source, in order to grid the sample creating spatial information across the sample, during the illuminating step.
10. The method of claim 9 comprising illuminating a region of the biological sample, wherein the region is smaller than a total area of the biological sample, and the x-ray fluorescence source illuminates the entire biological sample by moving the biological sample relative to the x-ray fluorescence radiation source.
11. The method of any preceding claim wherein an angle created between the incident beam emitted from the x-ray fluorescence radiation source to illuminate the biological sample and the beam detected from the heavy metal antibody conjugate on the biological sample by the detector is less than 180 degrees.
12. The method of any preceding claim further comprising: illuminating the labelled biological sample with a radiation beam emitted from secondary illumination source; detecting a radiation signal emitted from the heavy-metal antibody conjugate on the biological sample using a secondary radiation detector; and overlaying the detected x-ray fluorescence signal and the detected radiation signal.
13. The method of any preceding claim further comprising illuminating the biological sample with a plurality of x-ray fluorescence radiation sources, at least one of x-ray fluorescence radiation sources having a beam size that is different to a least one other x-ray fluorescence radiation source.
14. The method of any preceding claim further comprising depositing the biological sample onto a sample holder.
15. The method of claim 14 further comprising depositing multiple biological samples onto the sample holder, each biological sample being deposited on its own distinct area, separate from the other sample areas.
16. The method of claim 15 further comprising illuminating the multiple biological samples substantially simultaneously.
17. The method of any of claims 14 to 16 further comprising moving the biological sample between a deposition position, in which the biological sample is located substantially opposite a depositing system, and an illumination position, in which the biological sample is located substantially opposite the x-ray fluorescence radiation source.
18. A biological sample x-ray fluorescence signal detection apparatus, comprising: an x-ray fluorescence radiation source arranged to illuminate a biological sample with an x-ray fluorescence beam; and an x-ray fluorescence detector arranged to detect an x-ray fluorescence signal emitted from the biological sample; wherein the biological sample comprises at least one biomarker and the biological sample is labelled with at least one heavy-metal atom bound to an antibody forming a heavy-metal atom antibody conjugate, and wherein the conjugate is arranged to emit a fluorescent x-ray signal when excited by an x-ray fluorescence radiation source.
19. The apparatus of claim 18 wherein the antibody is arranged to specifically target the biomarker.
20. The apparatus of any of claims 18 to 19 further comprising: a deposition system configured to deposit the biological sample onto a target area of a sample holder; and a transportation system arranged to move the target area between a deposition position, in which the target area is located substantially opposite the deposition system, and an illumination position, in which the target area is located substantially opposite the x-ray fluorescence radiation source.
21. The apparatus of any of claims 18 to 20 further comprising a computer system arranged to control at least one individual component of the apparatus, including the deposition mechanism, the transportation system, the x-ray fluorescence radiation source, and the x-ray fluorescence detector.
22. The apparatus of claim 21 wherein the computer system further comprises a user interface configured to allow a user to interact with at least one component of the apparatus, including the deposition mechanism, the transportation system, and the imaging system.
23. The apparatus of any of claims 18 to 22 further comprising a storage system configured to store at least one sample holder having a biological sample on a surface of the sample holder.
24. The apparatus of claim 23 wherein the transportation system is arranged to move the target area between the deposition position and/or the illumination position and a storage position, in which the target area is located substantially within the storage system.
25. The apparatus of any of claims 18 to 24 further comprising an incubating system configured to incubate at least one sample holder having a biological sample on a surface of the sample holder.
26. An apparatus configured to, in use, carry out the method steps of any of claims 1 to 17.
27. A computer program comprising instructions which, when executed by a computer, cause the computer to carry out the method of any of claims 1 to 17.
28. A computer readable storage medium comprising instructions which, when executed by a computer, cause the computer to carry out the method of any of claims 1 to 17.