WO1999054439A1 - Isolation and purging of cells by means of osmotic pressure - Google Patents

Abstract

The present invention involves methods and apparatuses which enable a selected population of biological cells to be isolated or purged from a cell mixture on the basis of a ratio of nuclear volume to total cell volume. According to the invention, osmotic pressure can be utilized to selectively lyse and/or render non-viable selected undesired subpopulations of cells in a suspension, while not adversely affecting other desired subpopulations. In some embodiments of the invention, cells can be selectively lysed or rendered non-viable on the basis of a ratio of nuclear volume to total cell volume by exposing the cells to a solution having a predetermined osmolarity selected to inactivate a substantial fraction of cells having a ratio of nuclear volume to total cell volume below a threshold value. The invention enables effective cell separation utilizing a relatively rapid and easy to perform method involving changes in the osmolarity of cell suspensions. The inventive method has a variety of potential applications in clinical medicine, research, etc., with two of the more important foreseeable applications being stem cell enrichment/isolation, and cancer cell purging.

Description

ISOLATION AND PURGING OF CELLS BY MEANS OF OSMOTIC PRESSURE

Related Applications

This application claims priority from provisional specifications U.S. Ser. No. 60/082,195, filed April 17, 1998, and U.S. Ser. No. 60/103,984, filed October 13. 1998, the subject matter of which is incorporated herein by reference.

Field of the Invention

This invention relates to methods and apparatuses utilizing osmotic pressure for isolating selected cell types from cell suspensions.

Background The basis of many therapies for treating a variety of human diseases or for countering the effects of physiological insults involves the manipulation, expansion, and/or alteration of specific biological cells. Examples include autologous, syngenic, and allogenic stem cell transplants for immune system reconstitution following the myeloablative effects of severe high dose chemotherapy or radiation cancer treatments; severe exposure to chemical agents from either chemical weapons or accidents that disperse chemical agents into the environment; or severe exposure to radiation from nuclear weapons or accidents involving nuclear power generators. Based on effector cells born from genetically directed stem cells, pre-exposure prophylaxis or post-exposure therapies are under development for a variety of biological insults that may occur naturally (e.g., Ebola, etc.) or be inflicted by mankind (i.e., biological warfare agents). Gene therapies, also involving genetically manipulated stem cells, are under development for treating other blood diseases (e.g., AIDS, leukemia, etc.) and cancer, and may be useful in cloning animals. However, genetically manipulating stem cells is now a difficult procedure using viruses or carriers and does not have high yields. Current research findings indicate that the practical implementation of animal organ transplants into human recipients also requires procedures involving stem cells from both the donor and recipient. Cryopreservation of large numbers of specimens of human immune system cells, which can provide donors with a therapeutic basis should a health emergency occur later in life, demands that the specimens have minimal volume. Since very few stem cells are required to reconstitute and provide long term immune system function, the small specimen volumes required to ensure the feasibility of these banks rests on the availability of effective methods for isolating the trace numbers of stem cells found in either bone marrow aspirate, mobilized peripheral blood, umbilical cord blood, or fetal liver.

In order to achieve broad implementation of the therapies discussed in the previous examples, rapid and cost effective methods are needed to isolate, with high purity, the desired target cells from suspensions having a diverse mix of cell types and concentrations. The target cell highlighted in the examples given above is the hematopoietic stem cell. The present invention provides methods and apparatuses for performing rapid, cost-effective cell isolations and cell enrichments yielding, in some embodiments, highly purified cell suspensions. The invention has wide applicability for isolating a large variety of cell types from cell suspensions derived from a wide variety of sources. In one particular embodiment, the invention provides a method for isolating stem cells from any source of human or animal tissue containing stem cells (e.g., bone marrow aspirate, umbilical cord blood, mobilized peripheral blood, fetal liver, etc.). Furthermore, in other embodiments, the invention can provide a rapid and cost effective method for purging tumor cells from progenitor cell preparations that are required to rescue patients, via autologous stem cell transplants, after high dose chemotherapy or radiation.

Summary of Invention In particular, the invention involves, in one preferred form, stem cells being isolated from other mononuclear cells using osmotic pressure to selectively lyse unwanted cells. This technique can also be used for a variety of other cell separation applications, for example it can purge tumor cells from progenitor cell preparations for autologous bone marrow transplants. The selection parameter utilized is the ratio of the nuclear diameter or volume to the cell diameter/volume. After employing osmotic pressure to lyse cells having a ratio of nuclear diameter to overall cell diameter which is less than the corresponding ratio for selected cells, such as stem cells, gradient density centrifugation and/or filtration techniques (among other common procedures) may be used to isolate the selected cells from the resulting debris. More generally, the invention involves methods and apparatus for enriching at least one cell type from a mixture of cells using osmotic pressure to lyse cells, and in particular to - J lyse cells based on an average ratio of nuclear volume to cell volume.

In one aspect, the invention involves a self-selection method. The method comprises providing a mixture of cells in a suspension. The suspension includes at least a first cell type and a second cell type, the first cell type having an average ratio of nuclear volume to total cell volume of a first value, and the second cell type having an average ratio of nuclear volume to total cell volume of a second value that is different from the first value. The method further involves changing an osmolarity of the suspension so that a fraction of a population of cells of the first cell type that are viable is reduced to a greater extent than a fraction of a population of cells of the second cell type that are viable, thus yielding a mixture of viable cells that is relatively enriched in the second cell type.

In another aspect, the invention provides a method that involves providing a mixture of cell types in a suspension, where each cell type includes a plurality of viable cells. The mixture of cells includes at least one selected cell type that has an average of nuclear volume to total cell volume that is different from other cell types. The method further involves enriching at least one cell type relative to a second cell type on the basis of a difference in a ratio of nuclear volume to total cell volume.

In yet another aspect, the invention provides a method of enriching one cell type from a mixture of cells involving using osmotic pressure to selectively lyse cells on the basis of a ratio of nuclear volume to total cell volume. In one aspect, the invention provides a cell enrichment method. The method comprises providing a cell mixture containing a plurality of viable cells, and subjecting the cell mixture to conditions that create a change in osmotic pressure in the cells so that a substantial fraction of cells having an average ratio of nuclear volume to total cell volume within a first range of value are selectively made non-viable, while a second fraction, which is greater than the substantial fraction, of cells having an average ratio of nuclear volume to total cell volume within a second range of values are maintained in a viable state.

In another aspect, the invention provides a variety of cell suspensions. In one embodiment, the invention provides a cell suspension comprising a plurality of biological cells suspended in a liquid. The plurality of cells includes a first population of cells having a maximum characteristic of nuclear volume to total cell volume ratio of not less than a predetermined value, which are substantially viable. The plurality also includes a second population of cells having a maximum characteristic nuclear volume to total cell volume ratio of not more than the predetermined value, which are substantially non-viable or lysed. The cell suspension is obtained from a precursor cell suspension that comprises substantially viable cells, where the precursor cell suspension contains subpopulations, the first and second populations of cells. The cell suspension is obtained from the precursor cell suspension by subjecting the cells in the precursor cell suspension to an osmolarity sufficient to render non- viable a substantial fraction of the cells in the precursor cell suspension having a maximum characteristic nuclear volume to total cell volume ratio below the predetermined value.

In another embodiment, the invention provides a suspension comprising viable, human pluripotent lympho-hematopoietic stem cells, which are capable of differentiating into members of the lymphoid, erythroid, and myeloid lineages. The suspension is essentially free of mature and lineage committed cells and is derived from a precursor cell suspension that comprises substantially viable cells by subjecting the precursor cell suspension to a osmolarity sufficient to inactivate a substantial fraction of the mature and lineage committed cells in the precursor cell suspension. In yet another aspect, the invention provides a system for performing a cell selection.

The system comprises a cell suspension containment element that is adapted to contain a mixture of cells in a suspension, where the mixture includes at least a first cell type and a second cell type. The system also includes a dispenser adapted to dispense a predetermined quantity of a lysing solution having a predetermined osmolarity into the suspension. The system also includes a controller that is configured to determine the predetermined quantity of the lysing solution, which is required to change an osmolarity of the suspension so that a fraction of a population of cells of the first cell type in the suspension that are viable is reduced to a greater extent than a fraction of a population of cells of the second cell type that are viable. The controller is further configured to actuate the dispenser to deliver the predetermined quantity of lysing solution to the suspension.

In another embodiment, the invention provides a system for performing a cell selection. The system comprises a containment element that is adapted to contain a lysing solution having a predetermined osmolarity. The system also includes a dispenser adapted to dispense a predetermined quantity of a cell suspension into the lysing solution, where the suspension contains a mixture of cells including at least a first cell type and a second cell type. The system also includes a controller that is configured to determine the predetermined quantity of the cell suspension required to be dispensed into the lysing solution in order to change an osmolarity of the suspension so that a fraction of a population of cells of the first cell type in the suspension that are viable is reduced to a greater extent than a fraction of a population of cells of the second cell type that are viable. The controller is further configured to actuate the dispenser to deliver the predetermined quantity of cell suspension to the lysing solution.

Other advantages, novel features, and objects of the invention will become apparent from the following detailed description of the invention when considered in conjunction with the accompanying drawings, which are schematic and are not intended to be drawn to scale. In the figures, each identical or nearly identical component that is illustrated in various figures is represented by a single numeral. For purposes of clarity, not every component is labeled in every figure.

Brief Description of the Drawings Fig. la is a schematic illustration of a nucleated cell suspended in a suspending solution having a physiological osmolarity;

Fig. lb is a schematic illustration of the cell of Fig. la when suspended in a suspending solution having an osmolarity lower than the physiological osmolarity;

Fig. 2 is a graph showing predicted relative solute concentrations for lysis of mononuclear cells having particular ratios of nuclear diameter to overall cell diameter; Fig. 3 is a bar graph showing the volumetric proportions of the constituents comprising the lysing suspension as a function of lysing suspension relative osmolarity;

Fig. 4 is a bar graph showing the volumetric proportions of the constituents comprising an arresting buffer, as a function of the relative osmolarity of the lysing suspension; Fig. 5a is a schematic illustration of an automated batch system for performing the inventive methods;

Fig. 5b is a schematic illustration of an automated continuous flow system for performing the inventive methods;

Fig. 6 is a photocopy of a phase contrast micrograph image showing a hemacytometer field having lymphocytes and platelets therein;

Fig. 7 is a graph showing a fractional population of unstained lymphocytes as a function of relative solute concentration at two different times after dilution; Fig. 8 is a graph showing fractional populations of stained lymphocytes as a function of relative solute concentration at two times after dilution;

Fig. 9 is a graph showing the number of monocytes in various conditions as a function of relative solute concentration; and Fig. 10 is a photocopy of a photomicrograph showing a potential uncommitted primitive cell resting on a membrane filter having pores with a 2 μm nominal diameter.

Detailed Description The present invention provides methods for selecting, isolating, and enriching desired subpopulations of cells from a mixture of cells. The methods, in some embodiments, are based on differences in the ratio of the volume, or diameter, of the nucleus of cells, to the overall cell volume, or diameter. In a preferred embodiment, the invention involves using osmotic pressure to selectively lyse or render non-viable cells on the basis of a ratio of nuclear volume to total cell volume. In some embodiments of the invention using osmotic pressure, the invention involves providing a mixture of living cells and subjecting the cell mixture to conditions that create a change in an osmotic pressure in the cells of the mixture. A variety of known techniques to create a change in an osmotic pressure in the cells can potentially be employed in the scope of the invention. One technique involves providing the mixture of living cells in a suspension having a physiological osmolarity and subsequently changing the osmolarity of the suspension to create a resulting change in an osmotic pressure in the cells of the cell mixture. In preferred embodiments, the osmolarity of the suspension is reduced in order to create an increase in an osmotic pressure in the cells. As described in more detail below, changes in the osmolarity of the solution in which the cells are suspended, and the resulting changes in osmotic pressure in the cells suspended in the solution, can be utilized according to the invention to lyse, or render non-viable, a substantial fraction of the cells comprising one or more selected, undesirable subpopulations of cells in the cell mixture, while leaving substantially viable at least one other selected, desired subpopulation of cells in the mixture. A "substantial fraction" as used herein in describing a population of cells, some fraction of which is lysed or made non-viable, refers to at least 25% of the cells in the population being lysed or made non-viable, preferably 50%. more preferably 90%, more preferably 95%, and most preferably greater than 99%. A '^"significant fraction" as used herein in the context of selected cells in a population, which remain viable after treatment according to cell the selection method, refers to at least 10% of the cells in the population remaining viable, more preferably at least 25%, more preferably at least 50%. more preferably at least 90%, and even more preferably at least 95%. In all cases, the fraction of undesirable cells that are lysed. or made non-viable, will exceed the fraction of the desired, selected cells lysed. or made non- viable, by the treatment, and, conversely, the fraction of the selected cells in a population, which remain viable after treatment, will exceed the fraction of the undesired cells which remain viable, thereby effecting an enrichment of the desired cells.

The cell selection methods provided by the invention may be utilized for selecting, isolating, or enriching a wide variety of living cells, and may be utilized with a wide variety of cell mixtures. For example, the method may be used to select one or more subpopulations of cells from a mixture of cells, the mixture including at least two cell types, where an average ratio of nuclear volume to total cell volume differs between the cell types.

For embodiments involving isolation or enrichment of one or more selected cell types from a mixture including a plurality of cell types, the inventive method can involve subjecting the cells in a cell mixture to a solution having a osmolarity selected to render non-viable a substantial fraction of the cells in the cell mixture having an average ratio of nuclear volume to total cell volume that is less than a predetermined value, while maintaining a significant fraction of cells having an average of nuclear volume to total cell volume greater than the predetermined value in a viable state. The predetermined value of average ratio of nuclear volume to total cell volume is selected, according to theory and techniques described below, be equal to or less than the average ratio of nuclear volume to total cell volume of the desired cells to be selected, while being greater than the average ratio of nuclear volume to total cell volume of the undesired cells to be lysed or rendered non-viable.

The inventive methods may be used to separate subpopulations of nucleated cells from cell mixtures comprising a plurality of subpopulations of nucleated cells. In other embodiments, the methods may be employed to select one or more subpopulations of nucleated cells from a cell mixture comprising a plurality of cell types, including both nucleated and non-nucleated cells. In each case, the cells are selected based on a difference in the ratio of average nuclear diameter or volume to the overall cell diameter or volume (the ratio being zero for non-nucleated cells). The present invention may be advantageously employed for separating a wide variety of living cells from a wide variety of cell mixtures including, but not limited to, blood cells, cells of various body organs and tissues, cultured cells. cancer cells, stem cells, mixtures of cancer cells and stem cells, mixtures of cancer cells and blood cells, and many others as apparent to the skilled practitioner. While it is to be understood that the inventive cell separation methods have wide applicability to a variety of cell types and cell mixtures, the invention is illustrated below, in many instances, in the context of one illustrative embodiment involving the isolation of hemopoietic stem cells, or other stem cells, from other mononuclear cells, for example non-stem blood cells.

One preferred cell isolation method enables stem cells to be isolated from other mononuclear cells using osmotic pressure to selectively lyse the unwanted cells. The cell selection parameter is the ratio of the nuclear diameter or volume to the overall cell diameter/volume. In this embodiment, the method involves providing a mixture of living cells including stem cells, where the stem cells have an average ratio of nuclear volume to total cell volume that is greater than a predetermined value. The mixture is then subjected to conditions of osmolarity selected to render non-viable and/or lyse a substantial fraction of the non-stem cells in the mixture that have a ratio of nuclear volume to total cell volume that is less than the predetermined value. The method can be utilized to enrich or isolate a wide variety of stem cells including, but not limited to, hematopoietic stem cells, embryonic stem cells, mesenchymal stem cells, epithelial stem cells, gut stem cells, liver progenitor cells, endocrine progenitor cells, skin stem cells, neural stem cells, and other stem cells known to those of ordinary skill in the art. The inventive method is particularly advantageous in that it enables the isolation of very primitive stem cells including pluripotent stem cells that are capable of differentiating into members of various precursor, and mature cell lineages. For example, the method enables isolation of human pluripotent lympho-hematopoietic stem cells, which are capable of differentiating into members of lymphoid, erythroid, and myeloid lineages. As will be discussed in more detail below, such stem cells as mentioned above typically have an average ratio of nuclear volume to total cell volume that is greater than the corresponding ratio for more mature cell types. Accordingly, the inventive method enables isolation of such stem cells from a precursor suspension containing a variety of cell types, including mature cell types, to provide a suspension of stem cells that is essentially free of mature and lineage committed cells. A further advantage of the inventive method for isolating stem cells, when compared to traditional antibody-based methods for isolating stem cells, is that the present method does not depend on the stem cell displaying a particular protein marker or receptor on its membrane surface. This may provide a particular advantage for isolating hematopoietic stem cells, since it is believed that the most primitive pluripotent hematopoietic stem cells may not express the CD34 marker on their surface. Unlike traditional antibody-based methods for isolating hematopoietic stem cells which typically rely on antibody binding to the CD34 marker, the inventive method enables isolation of hematopoietic stem cells that express CD34, and also hematopoietic stem cells that are essentially free of cell surface CD34 markers.

In some embodiments, after employing osmotic pressure to lyse a substantial fraction of cells having a ratio of nuclear volume to total cell volume which is less than a predetermined value essentially equal to or greater than the corresponding ratio for a desired cell type, an arresting component may be added to the cell suspension to adjust the osmolarity of the solution containing the cells to a physiological osmolarity for the selected cell type or types. Typically, the arresting component is provided in the form of an arresting solution (also referred to as an '"arresting buffer") that can be added to the cell suspension after a predetermined period of time has elapsed, the predetermined period of time being selected to allow the undesirable cells to be effectively lysed or rendered non-viable by the osmotic pressure. In other embodiments, the arresting component can comprise a solute or mixture of solutes which can be added to the cell suspension. In some embodiments, after employing osmotic pressure to lyse a substantial fraction of undesired cells having a ratio of nuclear diameter to overall cell diameter which is less than a predetermined value essentially equal to or less than the corresponding ratio for stem cells, single or multiple gradient density centrifugation and filtration techniques (among other known procedures) may be used to isolate the selected cells, for example stem cells, from the resulting debris. More generally, the methods enable enriching at least one cell type from a mixture of cells using osmotic pressure to lyse, or render non-viable, cells, and in particular to lyse, or render non-viable, cells based on an average ratio of nuclear volume to cell volume.

Another preferred cell isolation method, according to the invention, involves purging mature cancer cells from a cell suspension, for example bone marrow aspirate or peripheral blood, for bone marrow transplantation. In this embodiment, the inventive method involves subjecting a mixture of living cells, which contains a concentration of viable cancer cells, to a change in osmolarity sufficient to lyse and/or render non-viable cells having an average ratio of nuclear volume to total cell volume less than a predetermined value, where the predetermined value is greater than or equal to the average ratio of nuclear volume to total cell volume of the cancer cells. The inventive method can be used to purge (render non-viable) a substantial fraction of the cancer cells initially present in the suspension. In preferred embodiments, greater than 90% of the cancer cells are rendered non-viable, in more preferred embodiments, greater than 99%, and in even more preferred embodiments, greater than 99.999% of the cancer cells initially present in the suspension are rendered non-viable.

The inventive method advantageously utilizes the fact that for typical mature nucleated cell types, the nuclear volume is typically relatively small compared to the total cell volume. For example even for the case of small lymphocytes, the nuclear diameter is only approximately 80% of the typical 8 μm cell diameter. However, in the case of typical stem cells, including, hematopoietic stem cells, the nuclear diameter is at least 90% of the typical cell diameter, which is typically on the order of 6 μm or less.

The present cell isolation method utilizes the physical principal that the lower the ratio of the nuclear diameter to the cell diameter, the more vulnerable to lysis the cell will be as the osmolarity of the external medium is changed from a physiological osmolarity for the cell, for example by dilution of the external medium with water or other solvent. The reason for this is that a comparatively small amount of water/solvent entering a cell with a small volume between the outer membrane and the nuclear membrane can lower the solute concentration by a much greater degree than is the case when the volume between the two membranes is larger. Consequently the osmotic pressure, the cell dilation, and the stress in the membrane are smaller if the volume between the outer membrane and the nuclear membrane is smaller. Following dilution of the external medium, or placement of the cells into a medium having an osmolarity lower than physiological (i.e., a hypotonic medium), the relative volume expansion of the nucleus can be substantially less than that of the cytosol of the cell. This is true for cells that satisfy either or both of the following conditions, A) and B): A) Most of the volume within the nuclear membrane is filled with material that is insoluble in water. In which case, only a relatively small volume of water needs to pass through the nuclear membrane in order to establish osmotic equilibrium, and the diameter of the nucleus remains essentially constant.

B) The permeability of the nuclear membrane to water is significantly less than that of the outer membrane of the cell. In particular, the permeability of the nuclear membrane must be small enough such that the nuclear volume does not substantially change during the time necessary to effect lysis or killing of undesired cells. Following lysis or killing of the undesired cells, further permeation of water into the remaining viable cells can be stopped and reversed by suspension of the remaining cells in a physiological osmolarity solution or adjusting the osmolarity of the suspending solution to a physiological level (e.g.. about 275-300 mOs/kg water for most mammalian cells), for example by addition of an arresting solution.

What follows is a discussion outlining the physical principles that are believed to be underlying the method. The basic process described below is shown schematically in Fig. la and Fig. lb. Fig. la shows a representative nucleated cell 2 suspended in a surrounding medium having a physiological osmolarity for the cell. The cell 2 has a total cell volume represented by V₀ and a characteristic diameter (the diameter of the cell when in a spherical state) represented as d₀. The cell 2 further includes a nucleus 4 having a volume represented by V₀'"' ^l and a characteristic diameter d„""^cl. The cell 2 is surrounded by a cell membrane 6, which is shown enlarged in the figure insert, having a thickness represented by t_mϋ. Fig. lb shows a swollen cell 8 which results after the cell 2 from Fig. la has been exposed to a suspending solution having a reduced osmolarity. The volume and diameter of swollen cell 8 are now represented as Fand d respectively. The volume of nucleus 10 is represented by V" ^l, and cell membrane 12 has a thickness represented by t_m.

The discussion that presents the theoretical description of selective cell lysis by osmotic pressure begins by considering the relation between the stress σ in the cell membrane and the pressure difference P across the membrane. It is assumed that the membrane is a spherical shell with a diameter d, and that it has a thickness t,„. Mentally dividing the sphere into two half shells, the force pushing the two shells apart is given by the left hand side of the Eq. 1 below and the force holding the two sides together is given by the right hand side of the equation. (Roark RJ and Young WC. Formulas for Stress and Strain; 5th ed., McGraw-Hill, New York, p. 451, (1982).

Solving for the pressure,

⁽ σ^

P = 4 t (2)

V a j Assuming lipid bilayer membranes are fluid-like, which is generally considered a more accurate approximation than assuming that they behave as completely elastic membranes (although the following analysis could be completed using either assumption), the stress in the membrane scales with surface area, whereas the thickness of the membrane scales inversely with surface area (Fung YC. Foundation of Solid Mechanics; Prentice-Hall, Englewood Cliffs, N.J. p. 171 (1965)). Thus, comparing a stretched membrane to an unstretched membrane (variables with subscripted zeros) yields:

— = (d/d₀)² and -^- = (d/d₀y² (3)

Applying the above membrane stress and thickness scaling relations to the pressure- stress relation of Eq. 2 shows that the pressure ratio corresponding to the stretched CP, d) and unstretched (P„, d states is inversely proportional to the diameter ratio:

P ^d _ϋ

— = — (4)

P d

Thus, Eq. 4 describes the coupling of cell dilation and transmembrane pressure differences. The pressure due to osmotic force can be expressed in terms of the difference between the volumetric molar concentration (moles per unit volume of solvent) of solute internal to the cell, X„ and the volumetric molar concentration, X_c. of the solute external to the cell. (Robinson RA and Stokes RH. Electrolyte Solutions, 2nd. Ed. Butterworths, England, p. 30 (1959)).

P = R T(X_t - X ) (5)

In this expression, R is the universal gas constant, 8.3 joule/mole/K, and T is the temperature in degrees Kelvin. When solvent or a solution having a solute concentration lower than that of the extracellular medium is added to the extracellular medium the external concentration X_e will decrease from its initial value X_m. The internal concentration X, will likewise decrease from its initial value X„, due to the diffusion of solvent through the membrane. The ratio of the osmotic pressures after versus before adding the solvent or solution to the extracellular medium is given by:

X X e X e(.)

P X - X e ^XeO ^X,()

(6) p₀ ^{' X},o eO

1 _ ^X*»

^X,

The ratio furthest on the right of Eq. 6 results from dividing the numerator and denominator in the middle ratio by the initial internal concentration X_m, and by multiplying the second term in the numerator of the middle ratio by X_e / X_eo. Assuming the amount of solute within the cell is essentially constant, the internal solute concentration will vary inversely proportional to the volume V of the cell that is available to the solvent.

^x, = x,o ⁽7⁾

Further it is assumed that the nucleus of the cell maintains an essentially constant volume as the solute concentration of the cytoplasm decreases in response to an imposed reduction in the extracellular solute concentration. This dictates that cell volume available to the solvent does not include the nuclear volume. This assumption is based on the fact that nuclear DNA in its native state is not highly water soluble; it is more colloidal in its native structure (Robyt JF and White BJ Biochemical Techniques, Theory and Practice, Waveland Press, Inc., Prospect Heights. IL, pp. 278-9 (1990)). This is believed to be due to its strong association with basic proteins such as histones and protamines which when stripped from the DNA structure, using for example strong surfactants such as SDS, enables the DNA to become somewhat water soluble (Redina G. Experimental Methods in Modern Biochemistry. W.B. Saunders, Philadelphia, pp. 110-1, (1971)). However, even when stripped from its associated basic proteins, it is unlikely that DNA's hydrophilic properties are sufficient to significantly drive or affect nuclear solvent migration. Thus, if the nuclear diameter is d ^{nu l} , the volume available for solvent influx is

V = - (d³ - d "^uc' ) (8)

and the ratio of the internal concentrations becomes

Dividing the numerator and denominator of the right hand side of Eq. 9 by d„ and including the assumption that the nucleus remains essentially unchanged in diameter, i.e. d -d_{ yields

This and subsequent expressions can be simplified by assigning symbols to represent the various ratios. The ratio of the cell diameter to its initial diameter is designated d/d„ - ξ and the ratio of the nuclear diameter to the initial cell diameter is d₀"^uc Id₀=a . With these definitions, the internal concentration ratio of Eq. 10 can be expressed as

X α^J

(1 1) ^x ₀ ξ³ - <*³

Further simplifications can be obtained by writing the ratio of the initial extracellular concentration to the initial intracellular concentration in the form r = X_el/X,„ and by writing the ratio of the external concentration to the initial external concentration as B = X/X^ The pressure ratio (Eq. 6) due to the addition of solvent or solution having a reduced solute concentration to the extracellular medium then becomes

^J

- B r

P_ ^J (12) P, 1 - r

This expression can now be equated to the pressure ratio that results from the membrane being dilated (Eq. 4). After minor manipulation, the following third order equation relating B, the reduction in the extracellular solute concentration relative to the isotonic concentration, to the cell diameter ratio ξ is obtained.

' - - ''' ξA³ -^ α³ - A r - ⁱ A ξ C³⁾

This equation can be simplified with no loss of predictive accuracy by recognizing that r is typically approximately unity. A value for r has been derived as follows. The osmotic pressure of a cell under typical physiological isotonic conditions for typical mammalian cells is ΔP-28 torr (Guyton AC. Text book of Medical Physiology. 8th Ed. W.O. Saunders, Philadelphia, pp. 211-9, (1991)). The difference in a molar concentration of solute across a cell membrane is given by AX = AP/RT, where R is the universal gas constant and T is the temperature. Using normal body temperature (7-310 °K), the excess concentration is about 1.5 millimoles/liter. Normal extracellular solute concentration for typical mammalian cells is -300 millimoles/liter (XJ. Thus, r = XJX_m = (X_m - AX)/X _m = 0.995. The following simplified working expression is obtained by setting r = 1 and substituting ξ = λ¹'³, where λ = V/V„, the stretched/unstretched volume ratio.

1 - α³ B - _«- (14) λ -

Cells typically lyse due to osmotic pressure if the total volume (i.e., including the nucleus) swells to approximately 150% of the original volume, although the percentage may vary somewhat with cell type and other factors. (Kinosita K. and Tsong TY. Hemolysis of Human Erythrocytes by a Transient Electric Field, Proc-Natl. Acad. Sci. USA, 74:1923-7 (1977)). The function given by Eq. 14 is plotted in Fig. 2 for three different values of typical of mononuclear cells. Monocytes have a nucleus that is small in comparison to the cell diameter. For these cells, a limiting value of = 0 is used. The estimates of a = 0.8 and = 0.9 are used respectively for lymphocytes and stem cells. In Fig. 2 a horizontal line is drawn at the volume expansion ratio (1.5) at which cell lysing is generally observed to occur. The intersection of this line with the curve of volume ratio vs relative solute concentration for a particular type of cell gives the solute concentration for which lysing could be expected for that particular type of cell. The figure predicts differences in the critical relative solute concentrations for the different cell types that can readily be achieved experimentally. In addition, for a given type of cell that lyses over a range of volume expansion ratios of order about plus or minus 0.1 about an average critical volume expansion ratio of 1.5, the critical relative solute concentrations for lysing shown on the figure do not overlap. Consequently an osmotic pressure-based cell lysing process can be effective for isolating and/or enriching stem cells and purging tumor cells from progenitor cell preparations as well as for performing cell isolation and/or enrichment involving a wide variety of other cell types.

While one of ordinary skill in the art would readily envision a variety of conventional techniques, materials, and analytical tools which could be useful for implementing the above described inventive methods for performing osmotic pressure-cell isolation, the following section presents one exemplary protocol, which may be used for performing the inventive cell isolation method. It is to be understood that the following protocol is purely exemplary and that the skilled practitioner may modify the following protocol in a variety of ways and for a variety of purposes without departing from the scope and spirit of the invention.

IMPLEMENTING OSMOTIC PRESSURE-BASED CELL SEPARA TION

The above theoretical model of cell selection utilizing osmotic pressure indicates that the parameter determining which cells will be lysed and which will remain viable when a heterogeneous population of cells is exposed to non-physiological osmolarity conditions (for example, a reduced osmolarity) is the ratio of nuclear diameter to cell diameter. This selection parameter can also be equivalently expressed in terms of the ratio of the volume of the nucleus to the total volume of the cell or in terms of the ratio of the volume of cytoplasm within a cell to the total volume of the cell. Many cells having a specific value for the selection parameter will undergo lysis when the osmolarity of the lysing suspension is reduced to a level where the total volume expansion of the cell exceeds about 150% of the volume of the cell when under standard physiological osmolarity conditions. Thus, for a specific type of cell, there is a direct correlation between cell lysis, hence cell death, and the reduced osmolarity of the lysing suspension. Cells with small nuclear to cell diameter ratios will lyse at osmolarities that are much higher than cells where the nuclear to cell diameter approaches unity. Due to this dependence, it is possible to select for and isolate a subset of cells within a heterogeneous cell population that have a nuclear to cell diameter ratio that is greater than other cell populations by exposing the heterogeneous mixture of cells to a lysing suspension that has a unique predetermined osmolarity. The lysing suspension comprises, in some embodiments, a lysing solution (also referred to herein as a "lysing buffer") which is added to solution-free cells, for example as obtained after centrifugation of the cells into a pellet and discarding of the supernatant or after collection of the cells on a filtration membrane. Alternatively, the solution-free cells (or a fraction thereof) may be added to the lysing solution. In both cases above, the lysing solution is referred to herein as being '"mixed with" the cells. In preferred embodiments, a predetermined quantity of a lysing solution is added to cells suspended in a standard physiological osmolarity (SPO) suspending solution to form a lysing suspension. In such embodiments, a predetermined quantity of the lysing solution, having a predetermined osmolarity, is dispensed into (mixed with) the cell suspension (or, equivalently, the cell suspension may be dispensed into the lysing solution) in order to change the osmolarity of the cell suspension to the unique predetermined osmolarity required to effect the desired cell isolation. The term "lysing solution" as used herein refers to any liquid that can be mixed with cells or a cell suspension to effect a change in an osmotic pressure in the cells. Lysing solutions, according to the invention, can include, but are not limited to, solutions of one or more solutes in a solvent or mixture of solvents, essentially pure solvents (e.g., distilled deionized water), or mixtures of essentially pure solvents, so long as the lysing solution has an osmolarity that is different from a physiological osmolarity for the cells with which it is mixed. In each case above, the suspension of cells that has a solution osmolarity of a unique predetermined value for performing the desired cell isolation is referred to herein as a "'lysing suspension."

For example, since tumor cells, such as epithelial breast cancer cells, have nuclear to cell diameter ratios that are smaller than the hematopoietic stem cell, reduced osmolarity conditions can be identified that will purge, by osmotic lysis, these tumor cells from contaminated autologous stem cell transplant specimens. Furthermore, since all other mononuclear cells present in bone marrow or peripheral blood progenitor cell preparations have nuclear to cell diameter ratios significantly less than the stem cell, the osmotic selection technique may be applied to these preparations to isolate the stem cells to high purity. As shown later in the examples, the osmotic pressure technique may also be used to select for lymphocytes by selective lysis of monocytes in human peripheral blood mononuclear cell preparations. The protocol for cell selection by the osmotic pressure technique is fundamentally similar for all of these applications, the differences being the target cell(s) to be selected, the origin or composition of heterogeneous population of input cells in the precursor suspension, and the predetermined reduced osmolarity required to achieve the desired cell selection. The following describes the main steps that comprise one embodiment of a general protocol for cell selection by osmotic pressure. Table 1 lists certain parameters that are important for certain cell selections by osmotic pressure. As noted, the selection parameter is the nuclear to cell diameter ratio =d„""^cl/d₀. Cell selection can be effected by exposing a heterogeneous mixture of cells to a reduced osmolarity in a lysing suspension (B < 1 ) causing cells with less than those to be selected to swell beyond the threshold for lysis (λ > λ_c) where rupture, and/or cell death occurs. Thus, cell selection is achieved by inactivation of the undesired cells. Since cell swelling is a diffusion based process driven by the solute concentration difference across a cell's membrane, it is expected to have an exponential dependence on time. For the exemplary cell selections presented below in the Examples section, the equilibration time is approximately one hour. Thus, the upper limit given in Table 1 (e.g., 3 hours) for lysing exposure time, which is greater than the equilibration time, is expected to yield substantially complete lysis of the cells swollen beyond the threshold for lysis.

Table 1. Osmotic Cell Selection Parameters.

The lysing suspension can be formulated using standard reagents. Standard media, such as phosphate buffered saline (PBS) or Iscove's Modified DulBecco^'s Medium (IMDM), both available from Fisher Scientific, Pittsburgh, PA, can be used as a constituent that provides a solute for the lysing suspension. Both PBS and IMDM have a standard physiological osmolarity ("SPO" ~ 300 mOsm kg-water) for most mammalian cells. Reduced osmolarity lysing solution can be obtained by dilution of a solute providing constituent, such as PBS or IMDM. with a solvent. Distilled/deionized water (DDI-water) is a suitable choice for the solvent. Alternatively, an essentially pure solvent, such as DDI-water, can be used as the "lysing solution." As the cells lyse in the lysing suspension, they will release cytoplasm. DNA, etc. If the cell density is very high, the release of these intracellular species can alter the effective osmolarity of the lysing suspension. This effect will be minimal if the volume fraction of cells in the lysing suspension is kept at or below about 1 %. Based on a typical 10 μm diameter cell, this corresponds to a preferred maximum cell density of about 2 x 10⁷ cells/ml. With this density, a 500 ml lysing suspension would contain, in one embodiment, the number of bone marrow mononuclear cells harvested for a typical autologous stem cell transplant. This indicates that the osmotic cell selection technique can easily scale to enable processing of clinical cell loads.

The temperature range over which the osmotic cell selection technique can be applied is based on the range over which cell viability will not be thermally compromised. Since membrane diffusion rates increase with temperature, a preferred temperature is about 37 °C for most cells, which will provide a relatively short equilibration time and processing time. Once the desired lysing exposure time has expired, an arresting solution (or arresting buffer) can be added, in some embodiments, to the lysing suspension to adjust the osmolarity of the suspension containing the cells to a physiological osmolarity for the selected cells, thereby terminating lysing. In preferred embodiments, the osmolarity of the arresting buffer will be greater than SPO, with the particular osmolarity, and thus composition, dependent on the predetermined osmolarity of the lysing suspension and the volume of arresting buffer added. Since lysis can result in the release of DNA from lysed nucleated cells, an enzyme is added to the arrested cell suspension in preferred embodiments to prevent any released DNA from coagulating cells and cell debris. DNAse (Sigma, DN-25), may be employed for such purpose as apparent to the skilled practitioner. In some embodiments, recovery of post-osmotic cell selection viable selected cells from the lysing suspension or arrested lysing suspension, which will typically contain an abundance of cellular debris, can be accomplished using standard single or multiple gradient density centrifugation techniques. For example, it has been found that Ficoll-Paque gradient density centrifugation can essentially completely remove the cell debris from the remaining viable cells, which will be resident at the density interface, for many applications.

For some preferred protocols, there are three suspensions/solutions utilized to perform the inventive osmotic cell selection method: 1) a precursor cell stock suspension, 2) a lysing solution, and 3) an arresting solution. The precursor cell stock suspension will contain the input cells to be treated by the osmotic cell selection method, typically suspended in PBS or equivalent SPO medium. The lysing solution/buffer can typically contain varying proportions of PBS and distilled-deionized water (DDI-water) so that when a fixed volume of cell stock suspension is added to a fixed volume of lysing buffer (or vice versa), the resulting lysing suspension will have a relative predetermined osmolarity (0 < B < 1.0) required to isolate desired target cells or purge undesired cells. The arresting solution/buffer preferably contains a predetermined concentration of solutes such that when the desired exposure time of the cells in the lysing suspension expires, the arresting buffer can be mixed with the lysing suspension to yield a suspension that has standard physiological osmolarity, thereby terminating osmotic lysing activity. Fig. 3 gives the proportions of cell stock suspension, DDI-water. and PBS contained in the lysing suspension as a function of lysing suspension relative osmolarity (B). The DDI-water and PBS proportions for the lysing buffers are obtained by considering just the relative proportions presented in Fig. 3 for these two constituents, i.e.. the cell stock constituent should be ignored. Fig. 4 gives the proportions of DDI-water and 2X PBS that can be used to make up the arresting buffers, as a function of lysing suspension relative osmolarity (B). The relative osmolarities on the abscissa in Fig. 4 correspond to the relative osmolarity of the lysing suspension (Fig. 3) to which the arresting buffer is to be added, not the relative osmolarity of the arresting buffer. Cell counting, utilizing trypan blue dye exclusion, under phase contrast microscopy can be used to adjust the density of viable cells in the cell stock aliquot. It should be emphasized that for any given heterogeneous population of cells, the osmotic cell selection parameter values for optimal performance must be selected based on routine experimentation and optimization with guidance from the theoretical development presented previously. The performance of the osmotic cell selection strategy can be evaluated using a number of well established assays, which include: histological staining characteristics examined under phase contrast microscopy, light scatter or antibody fluorescence characteristic examined using flow cytometry, and/or antibody or viability stain fluorescence examined under fluorescence microscopy. Various functional assays, which are numerous and specific to the cell-types contained in the input population, may also be applied to evaluate the cell-types remaining after selection, as apparent to the skilled practitioner.

Table 2 presents a flow chart summarizing the steps of one typical embodiment of an osmotic cell selection protocol according to the invention. Initially, before beginning the procedure, the predetermined reduced relative osmolarity (B) is selected based on the nuclear to cell diameter ratios of the cells in the sample to be isolated, as well as those desired to be eliminated by lysis, with guidance from the theory described above. Step 1 of the illustrated procedure is the preparation of the cell stock, or precursor, suspension. The precursor cell stock suspension is prepared by placing the heterogeneous mixture of cells, from which the target cells are to be isolated by the means of the osmotic cell selection technique, into a suspending medium that has standard physiological osmolarity, e.g.. PBS, IMDM, etc. The concentration of cells in the cell stock suspension is selected such that when this suspension is combined with the lysing solution, the resulting number density of cells in the lysing suspension does not exceed the value given in Table 1. By adhering to the cell density limit given in Table 1, cytoplasm, DNA, etc. released during cell lysis will typically not significantly increase the osmolarity of the lysing suspension, which otherwise can degrade the selectivity of the inventive cell selection strategy. The cell stock suspension is typically prepared by uniformly dispersing the cells in the suspending medium. For embodiments involving stem cell isolation and/or tumor purging, the tissue harvested from the donor, which may be for example bone marrow or mobilized peripheral blood, can be pre-processed to obtain the desired input cells. Typically, this can involve a gradient density centrifugation step to isolate the low density mononuclear cells from the abundant red blood cells. The mononuclear cell fraction so isolated can then be dispersed in a suspending medium to form the cell stock suspension for osmotic cell selection processing. In other embodiments, for example cells derived from solid tissue or organs, the tissue may need to be disaggregated and the resulting disaggregated cells harvested and washed using well established methods. The viability, concentration and identity of the input cells can be determined by a variety of methods know in the art, as mentioned previously. For example, the viability of many cells, such as mammalian cells, can be determined by trypan blue dye exclusion under phase contrast microscopy. Concentration may be determined by manual or automated cell counting techniques, for example manual counting on a hemacytometer, or automated counting by light scattering techniques. Individual cell types can be enumerated and marked for further tracking by dye-labeled antibodies (e.g., fluorescently labeled antibodies) that have a specificity for certain cell surface antigens specific to certain cell types. The labeled cells can then be quantified by standard techniques, for example, fluorescence microscopy or flow cytometry. -7i

Table 2. Typical steps in the osmotic cell selection protocol.

Steps 2 and 3 involve the preparation of the lysing and arresting buffers. Fig. 3 gives the relative proportions of components comprising the lysing suspension, which is the combination of the lysing buffer and the cell stock suspension, as a function of reduced relative osmolarity (B). The relative proportions of the constituents of the lysing buffer can be obtained from Fig. 3 by considering just the relative amounts of PBS and DDI-water shown on the figure. Fig. 4 provides the relative proportions of the constituents comprising the arresting buffer as a function of reduced relative osmolarity of the lysing suspension. After selecting the predetermined reduced relative osmolarity required to provide the desired osmotic cell isolation, the composition and proportions of constituents comprising the lysing and arresting buffers and lysing suspension can be obtained from Figs. 3 and 4. Although DDI-water and PBS are shown in Figs. 2 and 3 as suitable choices for the solvent and solute constituents, it is to be understood that other suitable solvents and solutes may be used as apparent to the skilled practitioner. The volume of lysing and arresting buffers required to effect a desired cell isolation scales directly with the volume of the cell stock suspension to be processed. In this exemplary protocol, we have arbitrarily imposed that the volume of arresting buffer be equal to the volume of the lysing suspension. However, other arresting buffer formulations and relative proportions could be used that do not have this constraint so long as the arresting buffer is formulated so that when it is combined with the lysing suspension it yields a standard physiological osmolarity for the viable cells in the resulting suspension.

Steps 4 through 6 comprise the osmotic cell selection part of the exemplary protocol. In step 4. the cell stock suspension is combined with a predetermined quantity of the lysing buffer, which initiates cell selection by the selective lysis of cell populations that have nuclear to cell diameter ratios that are at or below the critical predetermined value associated with the predetermined reduced relative osmolarity (B) of the resulting lysing suspension. The cell population(s) that have a nuclear to cell diameter ratio greater than the critical value will have a significant fraction of which that will remain viable. Once the cell stock suspension and lysing buffer have been combined they are preferably gently agitated to keep the cells uniformly dispersed. This can be achieved in one embodiment by splitting the lysing suspension into multiple tubes and installing the tubes in a rotator mixer (e.g., Fisher, PA, 14- 259-21), where the tubes can be slowly rotated during the lysing step, thereby keeping the cells uniformly distributed in each of the tubes. This mixing process comprises step 5 of the exemplary protocol in Table 2. To achieve a high cell lysis rate, the rotator mixer can be installed in an oven having a temperature of about 37°C. Operating at this temperature can provide a more rapid transport of solvent across the cell membrane without significantly perturbing cell viability. The time required to complete the cell lysis step of the exemplary protocol for many cell types is typically less than about three hours, more typically less then about two hours, and most typically less than about 1 hour. After the desired cell lysis period has expired, in preferred embodiments, an arresting buffer is combined with the lysing suspension in order to return the resulting suspension to a standard physiological osmolarity for the selected cells and to terminate cell lysis. This is shown as step 6 of the exemplary protocol.

Steps 7 through 9 show optional post-processing treatments that can be performed in order to recover viable cells remaining after osmotic lysis of undesired cells. In step 7. DNAse, an enzyme that inactivates DNA released from lysed cells, is added to prevent the released DNA from coagulating remaining viable cells and cell debris. Addition of the

DNAse can prevent remaining viable cells from becoming bound with coagulated cell debris and being lost with the cellular debris in any subsequent viable cell recovery step. In step 8, a conventional method for isolating cells from debris can be performed, such as gradient density centrifugation, using a suitable gradient or multiple gradients. A Ficoll-Paque gradient (Pharmacia Biotech, 17-0840-03) can be utilized for the purpose of spatially separating remaining viable cells from cell debris in a container. When Ficoll-Paque is used for such purpose, the cell debris tend to collect at the bottom of the centrifuge tube, whereas the remaining viable cells collect at the interface between the Ficoll-Paque and the cell suspending medium. In step 9, a pipette can be used to aspirate the fluid at the interface that contains viable cells. These cells can then be washed with PBS using standard centrifugation techniques. Finally, the washed cells can be spun down, the supernatant can be aspirated, and the cells can be resuspended in a supportive medium, e.g., IMDM, etc., which may, in some embodiments, contain penicillin, streptomycin, etc., to prevent infection if these cells are placed under cell culture conditions. In other embodiments, the gradient density centrifugation method for isolating viable cells from cell debris can be replaced or supplemented by a filtration method for harvesting viable cells. In one such embodiment, the cell suspension containing selected viable cells and cell debris can be passed through a filter including pores having diameters smaller than the average diameter of the viable selected cells. The cells collected on the filter may be washed by passing a washing solution through the filter, where the washing solution has a standard physiological osmolarity for the cells. The viable cells collected on the filter can then be resuspended by back-flushing the filter with a supportive medium. In yet other embodiments, the filtration method for washing and resuspending cells can be used as a supplement to the gradient density centrifugation method. In such a method, a cell suspension, after osmotic pressure cell isolation treatment, is subjected to a gradient density centrifugation step as described above. Remaining viable cells can then be collected at the interface, as previously described, and the collected aspirate can be passed through a filter to collect, wash, and resuspend viable cells as described above. The resulting suspensions can be placed, if desired, under cell culture conditions (e.g., 37^°C and 5% CO₂ for typical mammalian cells) to maintain their viability and function in the event the selected cells are to be used in the near term. The washed cells may alternatively be resuspended in a medium appropriate for subsequent cryopreservation (e.g., 10% DMSO (Fisher, D128-50), 90% fetal calf serum (Sigma, T- 2442)). They may then be stored in a -70°C freezer to preserve their viability for later use.

Step 10 of the exemplary protocol can involve a number of optional analytical and/or functional assays, which can be used to quantitate the yield and/or selectivity of the cell selection strategy. It is typically desirable to determine the number and viability of the selected cells and their cell type. Trypan blue dye exclusion under phase contrast microscopy can be used to determine cell numbers and viability. Histological stains can also be employed in conjunction with microscopic examinations to assess cell numbers and viability. As noted previously, fluorescence microscopy or flow cytometry can also be employed to determine cell numbers and viability using a spectrum of fluorescence viability stains, e.g., propidium iodide. Cell type may be determined using a variety of well established conjugated monoclonal antibody fluorescence markers, which can be examined/analyzed by means of fluorescence microscopy or flow cytometry as apparent to the skilled practitioner. For cells that do not have unique or well defined fluorescence antibody markers (e.g., some hematopoietic stem cells) cell functional characteristics can be examined by using a variety of Icnown culture-based assays. These can include colony forming cell culture assays and long term colony initiating cell culture assays. Animal model assays, e.g., NOD/SCID mouse model, are typically considered to be a reliable assay for the unambiguous assessment of the multi-lineage potential of selected cells believed to be hematopoietic stem cells. In general there are a broad spectrum of assays that can be applied to determine the number, viability, cell type, and function of the cells selected by means of the osmotic cell selection strategy, with the exact choice of assays dependent on the input cells and the subset of selected cells as apparent to one of ordinary skill in the art.

While the inventive cell isolation and/or purging method may be implemented manually using ordinary labware and equipment as apparent to one of ordinary skill in the art, in another aspect, the present invention also provides automated apparatuses for performing the inventive methods. The apparatuses can be designed and configured to implement the inventive cell isolation and/or cell purging methods in either a batch mode or a continuous flow mode. One embodiment of a contemplated automated apparatus for performing a cell isolation and/or cell purging in a batch mode is shown in Figure 5a. Figure 5a shows a system 30 including a cell suspension containment element or vessel 34 for containing a volume of a cell suspension 32 from which a selected cell population is to be isolated and/or purged. Cell suspension containment element 34 can be any volumetric container suitable for containing a volume of a cell suspension. In preferred embodiments, the container will be constructed of a material which is biocompatible and non-toxic to the cells contained in the cell suspension. In order to avoid contamination of the cell suspension with undesirable microorganisms, container 34 is preferably sterilizable, or more preferably sterile and disposable. A variety of disposable labware suitable for use in cell culture may be advantageously employed as the containment element 34 of the system 30. Examples include, but are not limited to, sterile disposable centrifuge tubes, culture tubes, tissue culture flasks, petri dishes, and multi-well plates.

System 30 also includes a dispenser 38 that is adapted to dispense a predetermined quantity of a lysing solution 36 into the cell suspension 32 upon actuation. In the embodiment illustrated in Figure 5a, the dispensing element 38 comprises a volumetric container suitable for containing a volume of a lysing solution 36. As discussed above, lysing solution 36 preferably comprises distilled water, or mixtures of distilled water and isotonic saline solutions, such as PBS. In some embodiments, other agents such as DNAse, proteolytic enzymes, or other agents apparent to one of ordinary skill in the art may also be added to the lysing solution 36 in dispenser 38. As discussed above, lysing solution 36 preferably has an osmolarity that differs from the osmolarity of the cell suspension 32. In preferred embodiments, lysing solution 36 has an osmolarity that is lower than the osmolarity of the cell suspension 32. The system operates to perform a cell isolation by dispensing a predetermined quantity of lysing solution 36 through line 40 and flow control valve 42 into container 34 containing cell suspension 32.

The system also includes a controller 46 which is configured to actuate and control the operation of the dispenser 38 via actuation of the flow control valve 42 included as a component of the dispenser so as to dispense lysing solution 36 into cell suspension 32 in sufficient quantity to change the osmolarity of cell suspension 32 to a level required to lyse or render non-viable non-desired cells while maintaining the viability of a desired cell type, as discussed above. Control valve 42 can be any of a variety of electrically-controllable or actuatable flow control valves or devices apparent to one of ordinary skill in the art. Flow control valve 42 should be controllable or actuatable via electrical input from the controller 46 through at least one electrical line 48.

Controller 46 is preferably a programmable or a pre-programmed device (including special purpose hardware or a hybrid special purpose and programmed general purpose device) having a user interface for inputting operating parameter values for the particular cell isolation to be performed. For example, in some embodiments, values to be input into the controller 46 by the user can include: the volume of cell suspension 32 to be treated; the initial osmolarity of cell suspension 32; the osmolarity of lysing solution 36; and the nuclear diameter to total cell diameter ratio of the selected cell type to be isolated. Controller 46, then utilizes the user input values to calculate the required volume of lysing solution 36 that the system 30 needs to add to cell suspension 32 to affect the desired cell isolation. Controller 46 can, in some embodiments, execute a computer program to determine the required predetermined volume of lysing solution 36 from the given user input values utilizing the above-developed theoretical model alone, or in combination with algorithms derived from regression of experimental data for critical osmolarities required to lyse cells having given nuclear diameter to total cell diameter ratios. After determining the required predetermined volume of lysing solution 36 to be added to cell suspension 32, the controller then sends a signal via line 48 to flow control valve 42 to dispense the required volume of lysing solution 36, in one or more discrete boluses 44, into container 34. In alternative embodiments of system 30, the system may further include osmolarity probes to monitor the osmolarity of cell suspension 32 and/or lysing solution 36. These osmolarity values may be fed to controller 46 as input data for determining the volume of lysing solution 36 to be added to cell suspension 32, thus eliminating the need for an operator to separately measure, and input these values to controller 46. In some embodiments including an osmolarity probe for measuring the osmolarity of cell suspension 32, the osmolarity probe may be used to continuously monitor the osmolarity of cell suspension 32 as the system slowly adds lysing solution 36 to the cell suspension 32. In such embodiments, controller 46 may utilize the measured osmolarity for determining the volume of lysing solution 36 to be added to cell suspension 32. In other words, in such embodiments, instead of controller 46 determining a predetermined volume of lysing solution 36 to be added to container 34 based on a set of initial input values, controller 46 operates flow control valve 42 so as to slowly add lysing solution 36 to container 34 until a desired predetermined osmolarity in cell suspension 32 is reached, as measured by the osmolarity probe in cell suspension 32.

While system 30 has been illustrated showing a volumetric container 38 containing lysing solution 36, which is added to a cell suspension 32 via a flow control valve 42, various other arrangements and systems for performing similar functions will be apparent to those of ordinary skill in the art and are included within the scope of and spirit of the invention. For example, instead of the system shown, where lysing solution 36 is fed through a flow control valve 42 under force of gravity or a positive pressure supplied to dispenser 38, in alternative embodiments, lysing solution 36 may instead be pumped from the dispenser, at a controlled rate determined by controller 46, into container 34 containing cell suspension 32. In such embodiments, controller 46 would control the volume of lysing solution 36 added to container 34 by controlling operation of a pump, included as part of the dispenser, rather than a flow control valve 42 as shown. In another alternative embodiment to that illustrated, vessel 34 can be utilized to contain the lysing solution, instead of the cell suspension, and dispenser 38 can be adapted to dispense a predetermined quantity of the cell suspension into vessel 34, which contains the lysing solution, upon actuation of the dispenser 38.

Figure 5b shows an alternative system for performing the inventive cell isolation and/or cell purging method which operates in a continuous flow mode instead of the batch mode of the system shown above in Figure 5a. The continuous flow system 60 of Figure 5b includes a vessel 64 for holding a volume of cell suspension 62, vessel 68 for holding a volume of lysing solution 66, and vessel 86 for receiving a volume of treated cell suspension 84. In system 60, cell suspension 62 is pumped through a containment element comprising a conduit 70 via pump 72 into vessel 86 for the treated cell suspension. Similarly, lysing solution 66 is pumped via a dispenser comprising conduit 74 and pump 76 to a connector 80, connecting conduits 70 and 74 in fluid communication. In alternative embodiments to that illustrated, the lysing solution can be contained in vessel 64 and the cell suspension can be contained in vessel 68, with the lysing solution being pumped through containment element 70, and the cell suspension being pumped via the dispenser comprising conduit 74 and pump 76. For the illustrated embodiment, in operation, lysing solution 66 and cell suspension 62 become intermixed at connector 80 before flowing via conduit 82 into vessel 86. Cell suspension 62 and lysing solution 66 are continuously, or alteratively intermittently, pumped to connector 80 at flow rates selected to yield a treated cell suspension 84 having a desired predetermined osmolarity required to affect a desired cell isolation and/or cell purging. The flow rates of cell suspension 62 and lysing solution 66 are controlled, via control of pump 72 and pump 76 by controller 88 to yield a desired final osmolarity of treated cell suspension 84. Controller 88 is preferably a programmable or a pre-programmed device (including special purpose hardware or a hybrid special purpose and programmed general purpose device) having a user interface for inputting operating parameter values for the particular cell isolation to be performed. Controller 88 can be programed in preferred embodiments to determined the appropriate pumping rates for pump 72 and pump 76 required to establish the appropriate mixing proportion of cell suspension 62 and cell lysing solution/buffer 66 for yielding the desired predetermined osmolarity of treated cell suspension 84. Similar to the embodiment described in Figure 5a, controller 88 determines these required flow rates of cell suspension 62 and cell lysing solution/buffer 66 based on input values for the desired flow rate of cell suspension 62 and/or treated cell suspension 84, the osmolarities of cell suspension 62 and cell lysing solution 66, and the desired final osmolarity of treated cell suspension 84, or alternatively, the nuclear diameter to total cell diameter ratio of the selected cells for isolation and/or purging in cell suspension 62. Based on these input values, and a computer program, circuit, or algorithm, controller 88 controls the pumping rates of pump 72 and pump 76 to enable the desired cell isolation. Similar to the embodiment described in Figure 5a, in some embodiments of system 60, input values for the osmolarity of cell suspension 62, cell lysing solution 66, and treated cell suspension 84 can be measured and continuously monitored by osmolarity probes (not shown) coupled to controller 88. Utilizing such osmolarity probes, system 60 could be configured to control the relative flow rates of pump 72 and pump 76 based on deviations of a measured osmolarity of treated cell suspension 84 from the desired predetermined osmolarity required to affect a particular cell isolation and/or purging.

In addition to the particular components shown above in connection with systems 30 and 60 for performing the inventive cell isolation method, either system may also include additional dispensers or pumps and conduits for adding other desirable components to the treated cell suspensions. For example, in some preferred embodiments, the systems may include components for adding a predetermined quantity of an arresting solution to the treated cell suspension in order to readjust the osmolarity of the treated cell suspension to a physiological level after a desired, predetermined time has elapsed after addition of the cell lysing solution. Other systems may include a detector for monitoring the relative size and/or concentration of cells contained in the treated and/or untreated cell suspensions. A detector for measuring the light scattering property of suspensions containing discrete particles may be used for such application, as apparent to one of ordinary skill in the art. Various other modifications and additions may be made to the exemplary systems described above as apparent to one of ordinary skill in the art without departing from the scope of the invention. The function and advantage of these and other embodiments of the present invention will be more fully understood from the examples below. The following examples are intended to illustrate the operation of the present invention, but not to exemplify the full scope of the invention.

Examples INTRODUCTION

The experimental examples that follow are presented as two separate cases. The two cases presented utilize the principles of the osmotic cell selection theory given above and apply that theory to the practical application of isolating lineage uncommitted primitive cells from human peripheral blood. Case 1 illustrates the selectivity of osmotic lysis based cell isolation by demonstrating differences in solute concentrations in cell suspending media required for lysis of monocytes versus lymphocytes. Case 2 involves isolation of potential uncommitted primitive cells from a mixed cell population by first lysing non-primitive cells, having larger ratios of cell diameter to nuclear diameter, followed by removing cell debris and then concentrating the remaining cells.

Sixty milliliters of whole human peripheral blood was used for each experiment. Each sample contained approximately 10⁸ lymphocytes, 10⁷ monocytes, and 10 uncommitted primitive cells. Because of the very small number of uncommitted primitive cells, it was impractical when performing the Case 1 experiments to determine a solute concentration useful for the isolation of uncommitted primitive cells. Instead, Case 2 was designed to investigate primitive cell isolation by the inventive osmotic cell selection strategy.

CASE I : CELL SELECTIVE OSMOTIC L YSIS

The first step performed for both cases was the isolation of the low density mononuclear cells from whole blood using a standard centrifugation technique. Fifteen milliliters of whole blood was carefully layered onto 15 ml of Ficoll-Paque in each of two 50 ml centrifuge tubes. The tubes were then centrifuged at 1500 rpm, with slow acceleration and no breaking, for 20 minutes. After centrifugation, the top of the tube contained a clear, yellow, liquid plasma layer. Below that layer was a hazy layer containing the mononuclear cells. Below the hazy layer was a clear layer of Ficoll-Paque. A red blood cell pellet was present at the bottom of the tube. The plasma layer was removed using a syringe. Another syringe was employed to extract the hazy layer from each of the two tubes, and the two extracts were merged into a single tube. A physiological phosphate buffered saline solution. (PBS), was added to the extract, and the combination was centrifuged for 20 min. to wash the mononuclear cells of the Ficoll-Paque. After centrifugation, the mononuclear cells formed a white pellet at the bottom of the centrifuge tube. The pellet was removed, placed in another tube containing 1 ml of PBS, and resuspended in the PBS using a vortex mixer, with 9μl of DNAase I (100 mg, 500 Kunitz-units per mg solid, Sigma, DN-25) then being added to prevent clumping. After addition of the DNAase I, the suspension was revortexed. The number of mononuclear cells in the suspension was determined by performing a manual cell count on a 20 μl sample using a phase contrast microscope and a hemacytometer. The number of mononuclear cells in suspension was determined to be about 2.5 x 10⁷.

For the present case, a protocol was utilized to allow diluted cell suspension (lysis suspension) samples to be prepared, having the same number of cells per unit area in the hemacytometer, despite having different degrees of dilution with deionized water. Briefly, twenty microliter volumes of the undiluted precursor cell suspension samples were distributed to an array of micro wells (Fisher, 96 well plate). To each well, different proportions of distilled deionized water and PBS were delivered. The proportions were chosen to give solute concentrations relative to pure PBS of 1.00, 0.90, 0.80, 0.75, 0.70, 0.65, 0.60, 0.55, 0.50, 0.45, 0.40, 0.35, and 0.20. The total volume in each well, including the additions of distilled deionized water and PBS, was held constant at 105 μl. This value was chosen to yield an average count of 10 cells in each intermediate (200 μm x 200 μm) hemacytometer square following the dilutions and after doubling the volume of fluid in the well by the addition of 105 μl of Trypan Blue dye. Addition of Trypan Blue dye enabled cells that had been lysed or were non-viable to be visualized and their number determined, since staining of a cell with the dye indicates that the cell membrane had been lysed or compromised.

Two replicate sets at each dilution were prepared. After a delay of 1 hour, the first set - JJ - had Trypan Blue dye added as discussed above. The addition of the Trypan Blue dye was expected to partially arrest the osmotic lysing process since it had approximately the same solute concentration (osmolarity) as the PBS. Consequently the dye was not added until the predetermined time of exposure of the cells to the solutions containing low relative concentrations of solute had passed. The dye addition was followed by cell counting as previously described, and the counted cells were categorized by type and by condition. Figure 6 shows the typical appearance of the lymphocytes on the hemacytometer. The granular background in this picture is due to the presence of platelets, which are not resolved in the photo reproduction. Cells in samples for each solute concentration were counted over a 1 mm x 1 mm square field. Cells were classified as monocytes or lymphocytes on the basis of their size. Each of these cell types was further classified as unstained, stained with a blue rim, or totally stained blue. The second replicate set was stained and counted in the same manner as the first replicate set but the Trypan Blue dye was not added to the samples until after 3 hours had passed since dilution. The fraction of unstained and stained lymphocytes versus relative solute concentration in the samples is presented in Figures 7 and 8, respectively. Figure 8 further shows the fraction of lymphocytes that are stained with a blue rim, or totally stained, as a function of relative solute concentration. The relative fraction plotted on the Y-axis in Figure 8 was calculated with respect to the total number of all lymphocytes present in each sample. Data points on the curves indicated by square symbols represent results for cells whose suspension media was diluted one hour before the addition of the Trypan Blue dye while data points indicated by triangular symbols represent results for cells whose media was diluted three hours before the addition of the Trypan Blue dye.

Figure 7 shows that the fraction of unstained, presumably viable, lymphocytes falls off markedly at relative solute concentrations of between about 0.5 and about 0.4. This result is in good agreement with the model predictions presented in Figure 2. The results are similar for cells having a 1 and 3 hour exposure to the lowered solute concentration before the addition of the dye, as indicated by Figure 7.

Further insights can be gained by observing the behavior of the fractions of stained lymphocytes (shown in Figure 8). The figure shows that after 1 hour exposure to a lowered solute concentration there is a sharp rise at relative solute concentrations of between about 0.5 and about 0.4 in the fraction of cells that are either totally or partially stained. Note that there are virtually no cells stained with a blue rimmed appearance at the lowest relative solute concentration of 0.2. It is believed that at this very low relative solute concentration, any blue rimmed cells initially present were converted into a totally blue stained cell before the cell count was performed. The curves showing data for cells subject to a 3 hour exposure show that almost no blue rimmed cells were present at any condition of relative solute concentration, and further show a sharp increase in totally blue stained cells at relative solute concentrations of between about 0.5 and about 0.4. For all the relative solute concentrations tested, the fraction of totally blue cells after three hour exposure was greater than the corresponding fraction for one hour exposure. These differences give a rough estimate of the characteristic time required for the cells to be permeated. The behavior of the fraction of cells which were stained with a blue rim as a function of relative solute concentration and time is consistent with an interpretation that permeation of the nuclear membrane is delayed with respect to permeation of the outer cytoplasmic membrane.

The data showing the number of stained and unstained monocytes as a function of relative solute concentration is presented in Figure 9. The individual curves show the number of monocytes observed with the hemacytometer that were unstained, stained with a blue rim, or totally stained as a function of relative solute concentration. The data in the figure are for samples which were stained after a 1 hr. exposure to a lowered solute concentration. Only one monocyte was observed in the samples having a 3 hour exposure, and that cell was totally stained. Possible explanations for the lack of monocytes observed after 3 hours of exposure are that either the cells had lysed in the intervening 2 hours, regardless of relative solute concentration, and/or that they had settled to the bottom or adhered to the walls of the sample wells. In the later case, they would not be present in the 20 μl sample that was transferred from the well to the hemacytometer. The number of unstained, presumably viable, monocytes was observed to drop sharply at relative solute concentrations of between about 0.8 and about 0.7. This result is consistent with the above-mentioned theoretical prediction of a threshold for lysis of about 0.67 for a cell whose nuclear diameter is negligible compared with the outer diameter of the cell. The number of blue rimmed cells was observed to increase sharply at a relative solute concentration of about 0.75, and to drop off sharply at a relative solute concentration just above about 0.6. As with the data for the lymphocytes, this drop-off at lower relative solute concentrations may be due to a conversion of blue rimmed cells into totally stained cells. No monocytes were observed at relative solute concentrations below about 0.5.

In addition to lymphocytes and monocytes, there were also a large number of platelets observed on the hemacytometer at relative solute concentrations above about 0.55; however, for relative solute concentrations at and below about 0.55, no platelets were observed.

CASE 2: ISOLATION OF UNCOMMITTED PRIMITIVE CELLS Isolation of uncommitted primitive cells required the use of a starting sample comprising 60 ml of normal peripheral blood, which was expected to contain on the order of 10 such cells. The isolation protocol employed involved first using the inventive osmotic pressure cell selection method to lyse essentially all cells having a ratio of nuclear diameter to cell diameter smaller than that of a stem cell. The theoretical model and the results of Case 1 above predicted that this could be done by exposing all of the mononuclear cells to a relative solute concentration of somewhat less than about 0.4. Second, any uncommitted primitive cells were separated from the debris resulting from the lysis of the other cells. The uncommitted primitive cells were separated from the red cells along with the other mononuclear cells using gradient density centrifugation. Uncommitted primitive cells were expected to float at the interface above the Ficoll-Paque layer. By contrast, mononuclear cell debris, expected to have a density significantly greater than that of the intact mononuclear cells, collected as a pellet at the bottom of the centrifugation tube, thereby allowing withdrawal of the uncommitted primitive cells essentially free of cellular debris. The final step entailed concentrating the uncommitted primitive cells onto the surface of a small filter having a controlled pore size for observation by microscopy.

The same procedure as outlined in Case 1 was employed for isolating the mononuclear cells from the red cells. Two diluted cell suspension (lysis suspension) samples were then prepared by the addition of appropriate amounts of PBS and distilled deionized water to yield relative solute concentrations of about 0.4 and about 0.3, respectively. The total volume of each sample was about 5 ml.

Cell counts were conducted as previously discussed at regular intervals (about every 15 min.) over a period of approximately 2 hours until essentially no unstained cells were observed with the hemacytometer. Each of the 2 samples was then carefully layered onto 5 ml of Ficoll-Paque in a separate centrifuge tube. The tubes were slowly accelerated, centrifuged at 2.000 rpm for 30 min, and then allowed to decelerate without braking. Each of the tubes contained a white pellet of mononuclear cell debris at the bottom. A vacuum pipette was carefully inserted into each tube, to a level just above the original 5 ml Ficoll-Paque fill line. All of the liquid above this line was extracted. The tip was then moved approximately a 1 mm below the 5 ml fill line, and the liquid above this point was also extracted.

The combined extracts from the 0.4 relative solute concentration sample were placed in a 5 ml syringe fitted with a stainless steel 13 mm syringe filter holder. The filter holder was loaded with a 13 mm diameter polycarbonate track-etched (PCTE), 2 micron nominal pore diameter, plain hydrophilic membrane filter (Osmonics, Livermore, CA). Only a small fraction of the 0.4 relative solute concentration sample could be passed through the membrane filter before clogging occurred. Similarly, the extract from the 0.3 relative solute concentration sample was loaded into a new, clean syringe, equipped with a fresh filter and the sample was passed through the filter. The filters were then removed from the filter holders and placed on the hemacytometer with the cover glass removed. After installing the cover glass on the hemacytometer, 20 μl of PBS was added to the hemacytometer, and the entire surface of the membrane filter was examined using the phase contrast microscope. Figure 10 shows the appearance of a potential uncommitted primitive cell in a sample treated with a 0.3 relative solute concentration resting on the membrane filter (arrow). The diameter of this cell is approximately six microns. A total of seven such cells were identified over the entire area of the 13 mm diameter filter for the 0.3 relative concentration sample. In addition, it appeared that there was very little, if any, cellular debris present on the filters.

DISCUSSION OF CASES 1 AND 2 The model predictions illustrated in Figure 2 indicated that a relative concentration of about 0.3 or below would be able to lyse uncommitted primitive cells having a ratio of nuclear diameter to cell diameter assumed to be about 0.9, and a critical volume expansion assumed to be about 150%. If either the diameter ratio or the critical volume expansion for some cells was actually greater than that assumed than such uncommitted primitive cells would lyse at a relative solute concentration lower than 0.3. Hence, the observed isolation of intact cells at a 0.3 relative solute concentration is consistent with theoretical predictions, given a reasonable variation in the assumed values of the model parameters. The fact that the membrane filter rapidly clogged at a relative solute concentration of 0.4 is consistent with the data shown in Figure 7, indicating that a non-negligible fraction of the lymphocyte population survived exposure to a relative solute concentration of 0.4.

Cases 1 and 2 indicate the inventive osmotic cell selection method can be useful for isolating and/or purging cells and that the behavior of the method is in general agreement with the predictions given by the above-developed theory. These examples also demonstrate that the invention can select cells based on a ratio of nuclear diameter to a total cell diameter, and show that the inventive method can be a useful strategy for relatively rapid and simple isolation of human hematopoietic stem cells.

While the invention has been shown and described above with reference to various embodiments and specific examples, it is to be understood that the invention is not limited to the embodiments or examples described and that the teachings of this invention may be practiced by one skilled in the art in various additional ways and for various additional purposes.

What is claimed is:

Claims

Claims

1. A cell selection method comprising: providing a mixture of cells in a suspension including at least a first cell type and a second cell type, the first cell type having an average ratio of nuclear volume to total cell volume of a first value, and the second cell type having an average ratio of nuclear volume to total cell volume of a second value different from the first value; and changing an osmolarity of the suspension so that a fraction of a population of cells of the first cell type in the suspension that are viable is reduced to a greater extent than a fraction of a population of cells of the second cell type in the suspension that are viable, yielding a mixture of viable cells relatively enriched in the second cell type.

2. The method of claim 1, wherein the first value is smaller than the second value.

3. The method of claim 2, wherein the change in osmolarity comprises reducing the osmolarity.

4. The method of claim 1, wherein the changing step comprises adding a lysing solution to the mixture of cells in the suspension, the lysing solution having an osmolarity different from that of the suspension before addition of the lysing solution.

5. The method of claim 1, wherein the changing step comprises isolating the mixture of cells from the suspension and subsequently adding a lysing solution to the mixture of cells, the lysing solution having an osmolarity different from a physiological osmolarity for the cells.

6. The method of claim 1 , wherein the changing step comprises isolating the mixture of cells from the suspension and subsequently adding at least a fraction of the cells to a lysing solution, the lysing solution having an osmolarity different from a physiological osmolarity for the cells.

7. The method of claim 1 wherein the mixture of living cells provided in the providing step contains a concentration of viable cancer cells which have an average ratio of nuclear volume to total cell volume less than a predetermined value, and wherein the changing step is performed in a manner that renders non-viable a substantial fraction of the cancer cells.

8. The method of claim 7 wherein said substantial fraction exceeds 90%.

9. The method of claim 7 wherein said substantial fraction exceeds 99.999%

10. The method of claim 1 wherein the mixture of living cells provided in the providing step includes stem cells having an average ratio of nuclear volume to total cell volume greater than a predetermined value, and other cells, wherein the changing step is performed in a manner that renders non-viable a substantial fraction of the other cells having a ratio of nuclear volume to total cell volume less than the predetermined value.

1 1. The method of claim 10, wherein said stem cells include at least one of the following: hematopoietic stem cells; embryonic stem cells; mesenchymal stem cells; epithelial stem cells; gut stem cells; liver progenitor cells; endocrine progenitor cells; skin stem cells: and neural stem cells.

12. The method of claim 1 1, wherein said stem cells include stem cells that are essentially free of cell surface CD34 markers.

13. The method of claim 1 1, wherein said stem cells include pluripotent stem cells.

14. The method of claim 1 further comprising after the changing step, the step of adjusting the osmolarity of the suspension to a physiological osmolarity for the second cell type after a predetermined period of time has elapsed.

15. The method of claim 14, wherein the step of adjusting the osmolarity of the suspension to a physiological osmolarity comprises adding a quantity of an arresting solution to the suspension, the arresting solution having an osmolarity greater than the physiological osmolarity.

16. The method of claim 1 further comprising: separating viable cells of the second cell type from cell debris resulting from osmotic cell lysis occurring as a result of said changing step by using at least one of single and multiple gradient density centrifugation, wherein the viable cells of the second cell type can be spatially separated in a container containing the suspension from the cell debris; and collecting the viable cells of the second cell type by aspiration from a spatial region containing the viable cells of the second cell type to form an aspirate containing the viable cells of the second cell type.

17. The method of claim 16 further comprising: passing said aspirate through a filter including pores having diameters smaller than the average diameter of the viable cells of the second cell type; washing cells collected on the filter by passing a washing solution through the filter, the washing solution having a physiological osmolarity for the cells; and resuspending the cells by back-flushing the filter with a suitable suspending solution.

18. The method of claim 16 further comprising: placing the aspirate in a centrifuge tube into which a washing solution is added, the washing solution having a physiological osmolarity for the cells of the second cell type; and centrifuging the mixture at least one time to form a cell pellet; and resuspending the cells comprising the cell pellet in a predetermined volume of suspending solution to yield a predetermined concentration of cells.

19. A suspension of cells obtained by practicing the method of claim 1.

20. A method comprising: providing a mixture of cell types in a suspension, each cell type including a plurality of viable cells, the average ratio of nuclear volume to total cell volume being different for at least one selected type of the cell types; and enriching at least one cell type relative to a second cell type on the basis of a difference in a ratio of nuclear volume to total cell volume.

21. The method of claim 20 wherein the enriching step includes changing an osmolarity of the suspension.

22. The method of claim 21 wherein the osmolarity of the suspension is reduced.

23. The method of claim 20 wherein the enriched cell type has a ratio of nuclear volume to total cell volume greater than the second cell type.

24. A suspension of cells obtained by practicing the method of claim 20.

25. A system comprising: a cell suspension containment element adapted to contain a mixture of cells in a suspension, the mixture including at least a first cell type and a second cell type; a dispenser adapted to dispense a predetermined quantity of a lysing solution having a predetermined osmolarity into the suspension; and a controller configured to determine the predetermined quantity of the lysing solution required to change an osmolarity of the suspension so that a fraction of a population of cells of the first cell type in the suspension that are viable is reduced to a greater extent than a fraction of a population of cells of the second cell type in the suspension that are viable, and further configured to actuate the dispenser to deliver the predetermined quantity of lysing solution to the suspension.

26. The system as in claim 25. wherein the containment element comprises a volumetric container.

27. The system as in claim 26. wherein the containment element comprises a multiwell plate.

28. The system as in claim 25. wherein the containment element comprises a conduit through which the suspension flows when the system is in operation.

29. The system as in claim 25, wherein the dispenser is adapted to dispense the predetermined quantity of lysing solution into the suspension in at least one discrete bolus.

30. The system as in claim 28. wherein the dispenser is adapted to continuously pump the lysing solution into the conduit through which the suspension flows.

31. The system as in claim 25, wherein the controller determines said predetermined quantity of lysing solution based on at least one of an average ratio of nuclear volume to total cell volume of the first cell type and the second cell type.

32. The system as in claim 31, further comprising a probe positionable to be in at least partially inserted in at least one of a cell suspension and the lysing solution, and adapted to measure an osmolarity of a solution.

33. The system as in claim 25, further comprising a second dispenser adapted to dispense a predetermined quantity of an arresting solution having a predetermined osmolarity into the suspension.

34. The system as in claim 33. wherein the controller is configured to determine the predetermined quantity of the arresting solution required to change an osmolarity of the suspension including an added quantity of the lysing solution to a physiological osmolarity for the cells of the second cell type, and further configured to actuate the second dispenser to deliver the predetermined quantity of the arresting solution to the suspension after a predetermined period of time has lapsed since a predetermined quantity of lysing solution has been dispensed to the suspension.

35. A system comprising: a containment element adapted to contain a lysing solution having a predetermined osmolarity; a dispenser adapted to dispense a predetermined quantity of a cell suspension into the lysing solution, the suspension containing a mixture of cells, the mixture including at least a first cell type and a second cell type; and a controller configured to determine the predetermined quantity of the cell suspension required to be dispensed into the lysing solution to change an osmolarity of the suspension so that a fraction of a population of cells of the first cell type in the suspension that are viable is reduced to a greater extent than a fraction of a population of cells of the second cell type in the suspension that are viable, and further configured to actuate the dispenser to deliver the predetermined quantity of cell suspension to the lysing solution.

36. The system as in claim 35, wherein the containment element comprises a volumetric container.

37. The system as in claim 35, wherein the containment element comprises a conduit through which the lysing solution flows when the system is in operation.

38. A method of enriching one cell type from a mixture of cells comprising: using osmotic pressure to selectively lyse cells on the basis of a ratio of nuclear volume to total cell volume.

39. A cell enrichment method comprising: providing a cell mixture containing a plurality of viable cells; subjecting the cell mixture to conditions creating a change in an osmotic pressure in the cells so that a substantial fraction of cells having an average ratio of nuclear volume to total cell volume within a first range of values are selectively made non-viable while a second fraction, greater than said substantial fraction, of cells having an average ratio of nuclear volume to total cell volume within a second range of values are maintained in a viable state.

40. The method of claim 39 wherein said change in osmotic pressure comprises an increase in osmotic pressure.

41. The method of claim 40 wherein said first range of values is less than said second range of values.

42. A suspension of cells obtained by practicing the method of claim 39.

43. A cell suspension comprising: a plurality of biological cells suspended in a liquid including a first population of cells having a maximum characteristic nuclear volume to total cell volume ratio of not less than a predetermined value that are substantially viable and a second population of cells having a maximum characteristic nuclear volume to total cell volume ratio of not more than the predetermined value that are substantially at least one of non-viable and lysed, the cell suspension being obtained from a precursor cell suspension comprising substantially viable cells, which precursor cell suspension contains as subpopulations the first and second populations of cells, the cell suspension being obtained by: subjecting the cells in the precursor cell suspension to an osmolarity sufficient to render non-viable a substantial fraction of the cells in the precursor cell suspension having a maximum characteristic nuclear volume to total cell volume ratio below the predetermined value.

44. The cell suspension as in claim 43, wherein said biological cells are derived from an animal.

45. The cell suspension as in claim 44, wherein said animal is a human.

46. The cell suspension as in claim 43, wherein said second population of cells include cancer cells.

47. The cell suspension as in claim 43, wherein said first population of cells include stem cells.

48. The cell suspension as in claim 47, wherein said stem cells include pluripotent stem cells.

49. The cell suspension as in claim 47, wherein said stem cells comprise at least one of the following: mesenchymal stem cells; embryonic stem cells; epithelial stem cells; gut stem cells; liver progenitor cells; endocrine progenitor cells; skin stem cells; or neural stem cells.

50. The cell suspension as in claim 47, wherein said stem cells include hematopoietic stem cells.

51. The cell suspension as in claim 47, wherein said stem cells include stem cells that are essentially free of cell surface CD34 markers.

52. A suspension comprising: viable, human pluripotent lympho-hematopoietic stem cells, which are capable of differentiating into members of the lymphoid, erythroid, and myeloid lineages, that is essentially free of mature and lineage committed cells, the suspension being derived from a precursor cell suspension comprising substantially viable cells by subjecting the precursor cell suspension to an osmolarity sufficient to inactivate a substantial fraction of the mature and lineage committed cells in the precursor suspension.

53. The suspension as in claim 52, wherein said stem cells include stem cells that are essentially free of cell surface CD34 markers.