Patent Application
20070274871
For a better understanding of the invention and to show how the same may be carried into effect reference is now made by way of example to the accompanying drawings in which:
Before describing a well plate embodying the invention, some relevant general background regarding meniscus formation is summarized.
Surface tension forces arise due to the energy needed to form an interface. At the point where the air-water interface meets the solid, a contact angle θ is defined by the tangent 17 to the air-water interface. When the surface tension forces are balanced the contact angle is defined by Young's Equation which for the example here of an air, water, solid system reads:
γwater-solid−γair-solid=γair-water cos θ
where γair-water is the surface tension between the two fluids air and water, γair-solid is the surface tension between air and the solid, and γwater-solid is the surface tension between water and the solid. In the literature, the contact angle defined by Young's equation is often referred to as the static contact angle to differentiate it from the more complex situation of dynamic flow, as occurs in a capillary for example, where there is a dynamic contact angle.
The (static) contact angle is thus the standard parameter used to describe meniscus formation. In the figure, the illustrated meniscus is concave, which is generally associated with a hydrophilic surface. However, for some systems the meniscus is convex (i.e. bends down rather than up as viewed in the illustration), which is generally associated with a hydrophobic surface.
The terms hydrophilic and hydrophobic are used in the art generically for all fluids, not specifically to describe situations where water or aqueous solutions are involved. Some authors define the boundary between hydrophilic and hydrophobic surfaces rigorously as a contact angle of 90 degrees. More usually, a hydrophilic surface is regarded as one with good wettability, perhaps with a contact angle of less than 30-40 degrees, and a hydrophobic surface is regarded as one with poor wettability, perhaps with a contact angle in excess of 60-80 degrees. Surfaces with contact angles of approximately 90 degrees are often referred to as hydrophobic in the literature. In this document we do not follow an exact definition of the terms hydrophilic and hydrophobic, while nevertheless following predominant standard usage that: a hydrophilic surface has a contact angle in the range from a few degrees to 30-40 degrees; a hydrophobic surface has a contact angle of greater than 60-80 degrees, often in the range 80-100 degrees; and a strongly hydrophobic surface is a hydrophobic surface with a contact angle greater than 90 degrees, typically greater than 100 degrees.
Having described the underlying physical description of meniscus formation, an embodiment of the invention is now described.
The body part 26 has tubular structures that form the cylindrical side walls 30 of the wells 22, where the tubes are interconnected by webs 36 that laterally extend between adjacent side walls. The webs 36 may be formed midway up the side walls, or at another position, for example at the top to provide an upper surface flush with the top of the wells. Multiple web layers may be provided if desired.
The base part 28 is formed as a single featureless transparent sheet of constant thickness ‘t’. By forming preferably all the well bases, but at least multiple well bases, from a single sheet, the base thickness can be made low, for example less than about 400 microns, and can be as thin as 20 to 100 micrometers and can exhibit superior imaging optical properties compared to existing well plates. Zeonor™ which is a transparent cyclo-olefin polymer (COP) manufactured by Zeon Corporation is a preferred base material. Zeonor has >85% transmission from 450-800 nm. With Zeonor base material, the imaging under liquid optical properties of the well plates of the present invention can approach those of optical glass. The well plate is particularly useful for measuring fluorescent events that can take place within the wells. The combination of the materials for the well wall and the thin bottom material, provides for particularly low fluorescent background.
The well bases 32 are then constituted by circular shaped portions of the base part 28 defined solely by the side-wall defining tubes of the body part 26.
The liquid 25 is shown partially filling the well 22 to define a water-air interface 27. The water-air interface 27 is shown schematically as being completely horizontal over its full lateral extent so that where it meets the side wall 30 there is no meniscus. In other words, with reference to
The base part 28 is made of suitable transparent material, for example a COP such as Zeonor™, but possibly a polymethyl methacrylate (PMMA) material or some other polymeric material or a glass. The transparent base material is in any case selected having regard to its optical properties. Preparation of the base part 28 will in most cases include applying an appropriate surface treatment or surface coating to render it sufficiently hydrophilic to support cell adherence, e.g. to make its contact angle 10-30 degrees. However, if the base part is made of a material which is sufficiently naturally hydrophilic, or cell adherence is not important for the intended use of the well plate, then that will not be necessary.
The base part 28 is a substantially featureless generally rectangular sheet of constant thickness, although there are two small holes 38 at either end away from the well areas to assist assembly. The base part is fused to the body part with an ultrasonic bonding process. Other forms of bonding may be used depending on the materials. For example, if the base part is glass, then a contact adhesive may be used.
The body part 26 is made of a black material, such as Zeonor™ or another COP which naturally has a contact angle close to 90 degrees, so the conventional plasma or other treatment to render the surface hydrophilic is foregone to retain a hydrophobic surface. The untreated natural COP surface thus does not support a meniscus, or in reality only supports a small convex or small concave meniscus. The body part 26 has a regular 3×2 square grid array of six wells 22 interconnected and supported by its webs 36. A standard well plate outer form is provided by outer walls 40 and a lower skirt 42 which are also part of the body part 26 and connected to the webs 36.
It is convenient if the base part and the body part are made of the same substances, e.g. Zeonor, but are distinguished by their different surface energies. In particular, the combination of an untreated body part and a treated base part is highly convenient for manufacture.
The lid 24 has the general form of a standard well plate lid, but includes circular lips 44 on the underside of the lid 24 to locate with the tops of the well side walls. This may be done by giving the lips 44 an inner diameter slightly larger than the outer diameter of the tubular side walls 30.
The plasma treatment process for modifying the hydrophobicity of the base and/or wall parts of the wells is now described. Plasma treatment of a variety of plastics materials used for biological sample containers, such as well plates and Petri dishes, is well known. The plasma treatment is typically used to render the container surface hydrophilic (i.e. improved wettability) using oxygen or other oxidizing gases such as air, water (in vapor), or nitrous oxide.
As previously mentioned the principal materials used for commercially available well plates to date are PS and PP. Plasma treatment processes to render PS hydrophilic are well known [3, 4, 5, 6] as are plasma treatment processes to render PP hydrophilic [7, 12]. Moreover, plasma treatments for PMMA [3], PE [5, 12], polyamide (nylon) [12], polycarbonate (PC) [12] and PI [7, 12] are also known. Further, contact angles of various COPs is known [8, 9, 10, 11], in some cases also after UV treatment [10]. COPs include Zeonor™ and Zeonex™ from Zeon Corporation, Topas™ from Ticona, Arton™ from Japan Synthetic Rubber Corporation and APEL™ from Mitsui Chemical Corporation. The natural (pre-treatment) contact angles of a variety of materials to a water/air fluid combination is tabulated below together with the minimum contact angle achieved with plasma treatment or UV treatment, with literature references for the tabulated values indicated in superscript.
Any of these materials may be used for the well bases and/or side walls, either in surface treated or untreated (natural) form, to provide the desired combination of properties for well plates embodying the invention, as well as any of the materials and surface modification processes referred to in references [3-12] the relevant parts for which are incorporated herein by reference.
In all cases, the hydrophilicity is controllable in the process in that the contact angle gradually reduces from the natural level as a function of processing time. For example, assuming a typical gaseous ion density of say 1010-1012 cm−3, e.g. oxygen density, the contact angle will gradually reduce from the natural level to the minimum achievable processing times over a period of around 5-10 minutes. If the plasma reactor is RF assisted, the speed of the contact angle reduction will also scale with microwave power.
In the context of the invention, such surface treatments may be used to reduce the hydrophobicity of the body part from say 116 degrees (the table value for Topas™) to close to 90 degrees. Moreover, surface treatment may be used to reduce the hydrophobicity of the base part from any level to 10-40 degrees or less.
Plasma processes have also been developed to make surfaces hydrophobic (i.e. reduced wettability), for example many polymers can be rendered hydrophobic using fluorinated gases, such as yetrofluoromethane (CF4), sulfur hexafluorine (SF6), and perfluorohydrocarbons.
In the context of the invention, such surface treatments may be used to increase the hydrophobicity of the body part so that it attains close to 90 degrees.
It is noted that the wide variety of known combinations of materials and plasma treatment processes have varying degrees of long-term stability, so the choice of treatment for well plates will be influenced by this, with the more highly stable treatments with a shelf life of a year or more being preferred.
Any of the above-mentioned known processes and other known processes for modifying wettability may be applied to manufacture well plates according to the invention, either to the well base material, well wall material, or both.
Well plates of the invention are of particular interest for automated cell counting and confluence detection methods as described in our co-pending patent application U.S. Ser. No. 11/050,826, the full contents of which are incorporated herein by reference. For culturing in a well plate, confluence is the degree to which the cells have grown to fill the well or other biological sample container. One speaks of a well being 70% confluent, 80% confluent and so forth, where 100% confluence is when the cells all touch each other filling the plate. The term subconfluent is also used to refer to a plate in which the cell colony or other cell aggregate has not yet reached confluence. For many applications, full confluence must be avoided, since it is associated with cell death or other undesired activity. Typically one aims to replate when confluence reaches a threshold, such as 70%. We propose an automated cell counting and/or confluence detection method based on phase contrast microscopy in which the automated method is carried out using a well plate of the invention in combination with image capture with phase contrast microscopy followed by application of a suitable image processing algorithm.
The degree of confluence can be determined by an automated cell count which is translated into an area by multiplication of the cell count by an area representing an average area for the cell type being cultured. Alternatively, the degree of confluence is determined by processing the image to: establish cell boundaries, compute the area of each cell from the cell boundary, and sum the cell areas. Image processing software, or alternatively any mixture of software, firmware and hardware, can be used to perform the image processing.
The cell boundaries can be determined directly by contrast from the plasma membrane, or from the extent of the cytoplasm, or possibly in some cases from contrast provided by an extracellular fluid in which the cell is located. The method may be carried out with no fluorescence staining in many cell types of interest, but if desired or necessary to achieve sufficient contrast modifying the cells by inclusion of a fluorescent tag may be performed, e.g. to image other cell parts, such as the nucleus, or to assess the physiological state of the cell, such as cell cycle. For example, red lectin can be used to tag the cell membranes. Nuclear tags that do not kill the cells may also be suitable to provide contrast in the case that the aggregate area of the cells is determined by cell counting rather than by direct cell area calculation. Whole cell stains may also be considered, such as Phalloidin FM4-64. A variety of suitable dyes are known and can be selected, for example, from the Molecular Probes catalog.
To determine confluence for a given well, the process is as follows. The captured image of the cells adjacent the base of the well is divided into background and foreground by extracting the background. This is done by applying a large Gaussian blur filter to the image, then subtracting this from the original image before adding the mean of the original image to each pixel. After this operation, pixels with intensities close to the mean of the resulting image are considered background, the remainder are considered foreground. The closeness to the mean is adjustable to accommodate variations in lighting etc. A segmented binary image is then generated by assigning foreground pixels to white and background pixels to black.
We envisage measuring the degree of confluence in one of two ways.
The first way involves counting the cells, and then assuming a value for cell area. This is usually reliable, since the variance in average cell area of a given cell type is usually small. The degree of confluence is then calculated to by the number of cells multiplied by the cell area divided by the available area of the well or other substrate, plate or dish.
The second way is to directly measure the aggregate area of all the cells by image processing of each individual cell to determine its boundary and thus area. The area of the cells can then be scaled up by a packing factor, e.g. assuming hexagonal close packing, before being divided by the available area of the well to arrive at a degree of confluence.
It will be appreciated that an automated cell counting process is a special simplified case of the confluence determination process in which cell areas (and thus boundaries) are not generally relevant.
By using a well plate according to the invention confluence detection and/or cell counting processes are much less demanding to automate, since both the imaging and the image processing are made easier and quicker. Imaging is easier and quicker since cell growth on the well side walls can be inhibited thereby ensuring that the cells lie in a single common plane defined by the base. Image processing is easier and quicker, since the cells lie in a common plane, so the image processing task can be reliably assumed to be a two-dimensional problem. For example, multiple images taken at vertically offset focal planes are not required, since it can be reliably assumed that no cells have adhered to the (hydrophobic) side walls. In addition, the two-part well plate construction allows superior optical performance to be achieved, for the reasons described above, so the quality of the images acquired for the image processing will be high across the full area of the base. This will also improve the accuracy of co-ordinate determination of the cells, which will in turn improve the reliability of picking that may follow from replating decisions made on the basis of the confluence having reached a predetermined threshold. These processes may be carried out on a Genetix CloneSelect™ Imager, ClonePix™ or ClonePix FL™ products, for example.
Finally, although confluence detection and cell counting processes are described here primarily in terms of automated processes, the advantages gained by use of well plates according to the invention will also benefit manual or semi-automated confluence detection and cell counting processes, e.g. a user viewing through a microscope will see a clearer image with no black meniscus-induced ring and no optical distortion in the region of the base-wall junction.
In summary, applying the invention, it is possible to provide well plates with unique combinations of desirable properties. In particular, meniscus formation can be suppressed by suitable choice of the surface energy of the well side walls. Moreover, the well bases can be designed with superior optical properties made possible by forming the well bases from a different material than the side walls, and also by making the well bases from a separate part (or parts), which in particular allows the well bases to be made very thin while retaining sufficient structural integrity. Still further, regardless of whether optical properties are of interest, the separate formation of the well bases and side walls allows the degree of hydrophilicity and hydrophobicity or the well and base to be selected freely. In particular, the combination of a relatively hydrophilic bases and hydrophobic wells promotes cell growth on the bases and not the side walls. Using such well plates it is less demanding to obtain good data when applying existing protocols, such as the examples 1-8 of reference [1] which are incorporated herein by reference in their entirety. Moreover, use of plates with higher well density and smaller well volume becomes possible leading to savings in reagent volumes and increases in the degree of automation.
