ARTIFICIAL NK CELLS AND USES THEREOF
1. CROSS REFERENCE TO RELATED APPLICATION
[0001] This application claims the benefit under 35 U.S.C. § 119(e) of provisional application no. 61/533,707, filed September 12, 2011, incorporated herein by reference in its entirety.
2. BACKGROUND
[0002] Precise identification and accurate quantification of rare cells present in blood specimens from patients with disease can pose significant challenges but is of significant value, especially where the rare cells are useful in monitoring the progression of disease and/or the efficacy of a therapeutic regimen. One example of a rare cell that is useful in disease monitoring is a lymphocyte Natural Killer (NK) cell called a CD56bnght NK cell.
[0003] Human NK cells, which account for about 15% of all peripheral blood lymphocytes, are characterized in part by the presence of CD56 and the absence of CD3. Of these, CD56bnght NK cells make up a very small percentage, with estimates ranging from 5-10%. See Cooper et al, 2001, Blood 97(10):3146-3151; Cooper et ai, 2001, Trends Immunol. 22(l l):633-640; Romagnani et ai, 2007, J. Immunol. 178:4947-4955. The rarity of these cells in circulation makes them difficult to detect and count accurately and reliably. Tools and methods that enable the rapid, reliable, and accurate identification CD56bright NK cells are needed.
3. SUMMARY
[0004] CD56bright NK cells, present in small numbers in blood, are characterized by an increased level of CD56 expression as compared to the levels observed in CD56dim NK cells. While CD56bri ht NK cells are classified and named by the level of CD56 expression, they differ from CD56dim cells in expression levels of a number of surface antigens, including CD16, CD177, CXCR3. See Romagnani et ai., 2007, J. Immun. 178:4947-4955. It has recently been reported that blockade of the human interleukin 2 receptor (ILR2) by the anti-CD25 monoclonal antibody daclizumab induces expansion of CD56bright regulatory NK cells. See, e.g., WO 2011/053777, disclosing that
administration of daclizumab to patients with multiple sclerosis caused expansion of CD56bright NK cells; Li et al., 2005, J. Immunol. 174:5187-5191, disclosing that
administration of daclizumab to patients with uveitis caused expansion of CD56brightNK cells. Accordingly, as noted in WO 201 1/053777, CD56bri ht NK cells may play an important role in monitoring the efficacy of therapies that block the ILR2 cascade.
[0005] As noted in the Background section, CD56bright NK cells are found in low numbers, comprising approximately 0.5-1.5% of the total peripheral blood lymphocyte population. Their low numbers makes them extremely difficult to detect and reliably count in cellular assays. For example, in fluorescence-based flow cytometry assays such as those utilizing fluorescence-activated cell sorting (FACS), their low numbers make it difficult to set the gating parameters necessary to detect the CD56bright NK cells such that they can be counted. In addition, once gating parameters have been set, their low signal- to-noise ratio can make it difficult to locate and count the CD56bnght NK cells in the resultant FACS scatter plots.
[0006] Further frustrating the ability to reliably detect and count CD56bright NK cells in flow cytometry assays is the fact that there is no consensus as to what markers, or the number of markers, are needed to identify the cells. The CD56bright NK cell
subpopulation is commonly defined in flow cytometry experiments based upon the intensity of CD56 staining in the CD56+/CD3~ population or by double staining as CD56bnght and CD 16 low Qr negative (Cooper et al, 2001, Trends Immunol. 22(11):633-
640). However, recent studies have reported that these criteria lead to ambiguous results when the CD56bright NK cell subpopulation is relatively small (see, e.g., Li et ai, 2005, J. Immunol. 174:5187-5191). In these circumstances, it has been suggested that additional markers be used, such as double staining of either CD56bnght/CX3CRr or
CD56bright/CXCR3+ in the lyn^hocytes or in the CD56+/CD3" NK cell population (Id.).
[0007] Applicants have discovered that "artificial cells" capable of generating two distinguishable detectable signals in an intensity ratio corresponding to the molar ratio of CD56 to CD16 (CD56:CD16) expressed on human CD56bri ht NK cells can be used to accurately and reliably identify CD56bnght NK cells in cellular assays. For example, as shown in the Examples below, when polymeric beads that mimic the profile of lymphocytes on a flow cytometer were coupled to anti-CD56 and anti-CD 16 antibodies, each labeled with a different fluorophore, in a molar ratio of anti-CD56 antibody to anti-
CD16 antibody of about 100:1, they reliably identified the position at which CD56bright NK cells would appear in a flow cytometry assay.
[0008] Applicants have also discovered that such "artificial cells" capable of generating two distinguishable detectable signals in an intensity ratio corresponding to the molar ratio of CD56:CD16 expressed on human CD56d,m NK cells can be used to accurately and reliably identify CD56dim NK cells in cellular assays. For example, applicants have shown that polymeric beads coupled to anti-CD56 and anti-CD 16 antibodies, each labeled with a different fluorophore, in a molar ratio of anti-CD56 antibody to anti-CD16 antibody of about 10:1 provides a marker for the detection of CD56dim NK cells in a flow cytometer.
[0009] Accordingly, the present disclosure provides, inter alia, artificial NK cells, for example artificial CD56bright NK cells and artificial CD56dim NK cells, methods of making these artificial NK cells, kits comprising the artificial NK cells and/or components and/or reagents for making the artificial NK cells, and methods of using the artificial NK cells to, among other things, calibrate flow cytometry instruments, optimize gating parameters of flow cytometry instruments, identify, detect, measure and/or count or quantify CD56bright NK cells. In some contexts, the artificial CD56bri ht NK cells and optionally the artificial CD56dim NK cells are useful in assays to detect, measure and/or count or quantify the CD56bnght NK cells in a sample in the context of a disease, condition, or therapeutic regimen affecting CD56bnght NK cell numbers.
[0010] The artificial NK cells generally comprise an artificial lymphocyte substrate, typically in the form of a polymeric bead or particle, that mimics a human lymphocyte in a flow cytometer and two distinct markers, a CD56 marker and a CD 16 marker. The CD56 and CD 16 markers are conjugated to the artificial lymphocyte substrate and are capable of generating detectable signals in defined intensity ratios. For artificial CD56bnght NK cells, the detectable CD56 marker:CD16 marker signal intensity ratio is in the range of about 95: 1 to about 105: 1, with a ratio of about 100: 1 being preferred.
Although, as will be discussed in more detail in the Detailed Description section, the signal intensity ratio may not in all embodiments correspond directly to the molar ratio of markers attached to the artificial lymphocyte substrate, in most embodiments this signal intensity ratio corresponds to a CD56 marker:CD16 marker molar ratio in the range of
about 95: 1 to about 105:1, with the molar ratio of about 100:1 being preferred. For artificial CD56dim NK cells, the detectable CD56 marker:CD16 marker signal ratio generated is in the range of about 9.5:1 to about 10.5: 1, with a ratio of about 10: 1 being preferred. Although, as will be discussed in more detail in the Detailed Description section, the signal intensity ratio may not in all embodiments correspond directly to the molar ratio of markers attached to the artificial lymphocyte substrate, in most embodiments this signal intensity ratio corresponds to a CD56 marker:CD16 marker molar ratio in the range of about 9.5: 1 to about 10.5:1, with the molar ratio of about 10:1 being preferred.
[0011] The artificial cells may optionally include additional markers capable of generating additional detectable signals, but such additional markers are not necessary.
[0012] The signals generated by the markers can be detected by a variety of means, including by way of example and not limitation, absorbance, fluorescence, luminescence, phosphorescence, colorimetry, isotopic difference, etc. In many embodiments, for example, artificial NK cells designed for use in the context of fluorescence-based flow cytometry assays, such as fluorescence-activated cell sorting (FACS) assays, fluorescence signals are preferred.
[0013] The markers may generate the detectable signals directly or indirectly. By directly it is meant that the marker itself generates the detectable signal, for example the marker includes a moiety capable of absorbing light, fluorescing, luminescing, phosphorescing, participating in a color-producing reaction, etc. As a specific, non- limiting example, the CD56 marker could comprise a first fluorophore that generates a first fluorescent signal, and the CD 16 marker could comprise a second fluorophore that generates a second fluorescent signal distinguishable from, or resolvable from, the first fluorescent signal. Markers that generate detectable signals directly are referred to herein as "direct markers."
[0014] By indirectly it is meant that the marker interacts with or binds another molecule that includes a moiety capable of absorbing light, fluorescing, luminescing,
phosphorescing, participating in a color-producing reaction, etc. or that interacts with or binds yet another molecule that includes a moiety capable of absorbing light, fluorescing, luminescing, phosphorescing, participating in a color-producing reaction, etc. As a
specific, non-limiting, example, the CD56 marker could comprise a first peptide or polypeptide that specifically binds a first binding partner, and the CD 16 marker could comprise a second peptide or polypeptide that specifically binds a second binding partner. Detectable signals are generated indirectly by incubating the artificial NK cell with labeled binding partners specific for the markers, for example, an anti-CD56 marker antibody labeled with a first fluorophore that generates a first fluorescent signal, and an anti-CD 16 marker antibody labeled with a second fluorophore that generates a second fluorescent signal distinguishable from, or resolvable from, the first fluorescent signal.
[0015] Detectable signals may also be generated indirectly via "sandwich" type binding assays by incubating the artificial NK cells with unlabeled binding partners specific for the CD56 and CD 16 markers, and labeled binding partners specific for the unlabeled binding partners. For example, the artificial NK cells could be incubated with unlabeled species-specific anti-CD56 marker and anti-CD 16 marker antibodies, for example, goat anti-CD56 marker monoclonal antibodies and rabbit anti-CD 16 marker monoclonal antibodies, and the resultant complex incubated with labeled anti-species IgG antibodies, for example an anti-goat IgG monoclonal antibody labeled with a first fluorophore that generates a first fluorescent signal and an anti-rabbit IgG monoclonal antibody labeled with a second fluorophore that generates a second fluorescent signal distinguishable from, or resolvable from, the first fluorescent signal. Markers that generate detectable signals indirectly are referred to herein as "indirect markers."
[0016] Skilled artisans will appreciate that in some instances, it may be desirable for the artificial CD56bri ht NK cells and/or human CD56dim NK cells to mimic as closely as possible human CD56bnght and/or human CD56dlm NK cells, respectively, under the assay conditions being used. As a specific non-limiting example, in assays where it would be beneficial to stain artificial NK cells utilizing indirect CD56 and CD 16 markers with the same reagents used to stain human NK cells from test samples, it may be desirable for the artificial NK cells to mimic as closely as possible human NK cells under the specific assay conditions. In such embodiments, the indirect CD56 and CD16 markers in the artificial NK cells could be polypeptides that correspond in amino acid sequence to the sequences of human CD56 and CD16, respectively, recognized by the reagent used to stain the test sample. In specific non-limiting embodiments, the CD56 and CD 16 markers
could be polypeptides corresponding in amino acid sequence to the amino acid sequences of all or a portion of the extracellular domains of CD56 and CD16, respectively.
[0017] For a single artificial NK cell, all the markers can be direct markers, all the markers can be indirect markers, or combinations of both direct and indirect markers can be used. In preferred embodiments, all the markers present on a single artificial NK cell are of the same type, i.e., either all direct or all indirect.
[0018] Whether direct or indirect markers are used, they should be conjugated to the artificial lymphocyte substrate in molar ratios such that under conditions of use, the artificial cell yields signal intensity ratios indicative of CD56bright NK cells or CD56dim NK cells, as described above.
[0019] The markers can be conjugated to the artificial lymphocyte substrate using a variety of different means and/or chemistries, as will be described in more detail in the Detailed Description section. The linkages may be direct or may be mediated via linkers, as will be described in more detail in the Detailed Description section. The linkages may be covalent or non-covalent, as will be described in more detail in the Detailed
Description section.
[0020] In another aspect, the present disclosure provides methods of making the artificial NK cells described herein, either CD56bright or CD56dim. Generally, the method comprises conjugating two distinct markers, a CD56 marker and a CD16 marker, each capable of generating distinguishable detectable signals, either directly or indirectly as discussed above, in defined ratios, to an artificial lymphocyte substrate as described herein. The particular methodologies and chemistries used to couple the markers to the artificial lymphocyte substrate, as well as the characteristics of the artificial lymphocyte substrate, will be discussed in more detail in the Detailed Description section.
[0021] Artificial NK cells as described herein can be provided in kits. Accordingly, in another aspect, the present disclosure provides kits including artificial NK cells and/or various reagents. The kits may include artificial CD56bnght NK cells that are ready for use, or reagents suitable for making artificial CD56bnght NK cells. The kits may optionally include artificial CD56dlm NK cells that are ready for use, or reagents suitable for making artificial CD56dlm cells, as well as reagents and buffer or buffer components
for using the artificial NK cells in assays. In embodiments where one or more markers on the artificial NK cells are indirect markers, the kit may optionally include reagents to stain the artificial NK cells, such as, for example, labeled primary antibodies that specifically bind the indirect markers, or unlabeled primary antibodies that bind the indirect markers and labeled secondary antibodies that bind the primary antibodies.
[0022] Fluorescence-based flow cytometry methods depend on proper calibration to ensure accurate identification, sorting, and measurement of cells of interest. The artificial NK cells described herein are useful to calibrate flow cytometry instruments, optimize gating parameters of flow cytometry instruments, identify, detect, collect, measure and/or count or quantify CD56bnght NK cells found in complex mixtures {e.g., blood).
Accordingly, in another aspect, the present disclosure provides methods of using the artificial NK cells described herein in flow cytometry assays. Generally, the methods comprise determining a location of artificial CD56bnght NK cells in a sample using a flow cytometry assay. Optionally, a location of artificial CD56dim NK cells is determined using the same assay. The location of artificial CD56bright, and optionally CD56dim, NK cells provides a positive, and optionally negative, reference point(s) for delimiting the location of, or for setting gating parameters for, CD56bnght NK cells in a test sample assayed under the same conditions. The methods can further include a step of collecting, counting, or measuring CD56bright NK cells from a test samplC) which may include multiple cell types, based on the location established for artificial CD56bnght, and optionally CD56dim, NK cells.
4. BRIEF DESCRIPTION OF THE FIGURES
[0023] FIG. 1 provides a scatter plot of the side scatter (Y axis) and forward scatter (X- axis) of human lymphocyte cells (outlined by hexagonal area in the plot) measured by flow cytometry.
[0024] FIG. 2 provides a scatter plot of the side scatter (Y axis) and forward scatter (X- axis) of exemplary artificial lymphocytes that can be used to make artificial NK cells, as measured by flow cytometry. The hexagonal box identifies the expected migration region of human lymphocyte cells.
[0025] FIG. 3 provides a representative flow cytometric scatter plot of a whole blood sample when double stained with an Alexa Fluor® 488 (AF488)-labeled anti-CD56
antibody and an Alexa Fluor® 647 (AF647)-labeled anti-CD 16 antibody. The fluorescence intensity of the AF647 signal is plotted on the Y-axis, the fluorescence intensity of the AF488 signal on the X-axis. Cells within the rectangular box are CD56bright NK cells.
[0026] FIG. 4 provides a scatter plot of exemplary artificial CD56bright NK cells. The exemplary artificial NK cells utilized comprise direct CD56 and CD 16 markers, in this instance an AF488-labeled anti-CD56 antibody (CD56 marker) and an AF647-labeled anti-CD16 antibody (CD 16 marker) covalently linked to an artificial lymphocyte cell in a CD56:CD 16 marker molar ratio of 100: 1. The rectangular box identifies the expected location of human CD56bnght NK cells in a whole blood specimen optimally stained with anti-CD56 and anti-CD 16 antibodies.
[0027] FIG. 5 provides a scatter plot of exemplary artificial CD56dim cells. The exemplary cells utilized comprise direct CD56 and CD16 markers, in this instance an AF488-labeled anti-CD56 antibody (CD56 marker) and an AF647-labeled anti-CD 16 antibody (CD 16 marker) covalently linked to an artificial lymphocyte cell in a
CD56:CD16 marker molar ratio of 10:1. The rectangular box identifies the expected location of human CD56bnght NK cells and the cluster of signals in the upper left-hand region defines the expected location of a CD56dim NK cells in a whole blood specimen optimally stained with anti-CD56 and anti-CD16 antibodies.
[0028] FIG. 6 provides cartoons illustrating certain exemplary embodiments of artificial NK cells as described herein.
[0029] FIG. 7 provides cartoons illustrating the use of certain exemplary embodiments of artificial NK cells as described herein.
[0030] FIG. 8 provides a cartoon illustrating a synthesis scheme for a specific exemplary embodiment of artificial NK cells as described herein.
[0031] FIG. 9 provides the amino acid sequence of human CD56140kD (N-CAM) (FIG. 9A; SEQ ID NO:l) and human CD16A (Low affinity receptor IIIA for Fc fragment of immunoglobulin G, FcylllA) (FIG. 9B; SEQ ID NO:2), with signal peptide residues indicated in italics, with residue +1 being the N-terminus of the mature polypeptide; the extracellular regions of each polypeptide sequence are underlined, and regions or residues
thought to be recognized by specific anti-CD56 antibody NCAM 13 (e.g., Product no. 556324, BD Pharmingen™, BD Biosciences) and anti-CD16 antibody 3G8 (e.g., Product no. 557710, BD Pharmingen™, BD Biosciences) are shown in bold, in FIG. 9A and FIG. 9B, respectively.
5. DETAILED DESCRIPTION
[0032] Applicants have discovered that "artificial cells" capable of generating two distinguishable detectable signals in intensity ratios that correspond to the molar ratios of
CD56 and CD16 expressed on human CD56bright NK cells or human CD56dim cells behave as human CD56bnght or human CD56dim NK cells, respectively, in flow cytometry assays.
The artificial CD56bnght NK cells, either alone or in conjunction with the artificial
CD56dim cells, can be used to reliably identify and/or count human CD56bright NK cells in flow cytometry cellular assays, such as FACS assays. As shown in the Examples below, artificial CD56bnght NK cells comprising polymeric beads that migrate as human lymphocytes in a flow cytometer covalently linked to CD56 markers (in this instance an anti-CD56 antibody labeled with Alexa Fluor® 488 (AF488)) and CD 16 markers (in this instance an anti-CD 16 antibody labeled with Alexa Fluor® 647 (AF647)) in a
CD56.CD16 marker molar ratio sufficient to generate a CD56:CD16 marker fluorescence signal intensity ratio that approximates that of human CD56bnght NK cells stained with anti-CD56 and anti-CD 16 antibodies reliably identified the position at which human CD56bright NK cells appear in a FACS flow cytometry assay (see Example 3). Similarly, artificial NK cells including a CD56:CD16 marker molar ratio sufficient to generate a CD56:CD16 marker signal intensity ratio that approximates that of human CD56dim NK cells stained with anti-CD56 and anti-CD 16 antibodies reliably identified the position at which human CD56dim NK cells appear under the same assay conditions (see Example 4).
5.1. Artificial NK Cells
[0033] The artificial NK cells will now be described in more detail with reference to a specific embodiment useful in cellular assays utilizing fluorescence-activated cell sorting (FACS). Skilled artisans will appreciate that while the specific exemplary embodiment focuses on the use of fluorescent signal moieties and fluorescence detection, the artificial NK cells described herein could include signal moieties detectable by other means, such as for example, absorption, luminescence, phosphorescence, colorimetry, etc. Thus, the
artificial NK cells, methods, assays and kits described herein are not intended to be limited to those involving fluorescent signal moieties and fluorescence detection.
[0034] The artificial NK cells useful in FACS-based assays generally comprise three components: (1) an artificial lymphocyte substrate that mimics the characteristics of a human lymphocyte in a flow cytometer, (2) a CD56 marker and (3) a CD 16 marker. The CD56 and CD 16 markers are linked to the artificial lymphocyte substrate, either covalently or by other means, in molar ratios such that under conditions of use, the artificial NK cells generate two distinguishable fluorescence signals having a defined intensity ratio, causing the artificial NK cells to migrate as human CD56b"8ht NK cells or human CD56dim NK cells in FACS-based assays, depending upon their fluorescence signal intensity ratio.
[0035] For certain applications, such as monitoring the response of patients to treatment or assessing disease state or progression, it may be desirable to detect additional markers on CD56bnght NK cells. Artificial NK cells useful for such applications may optionally include additional markers, such as, for example, one or more of CD122, CD25, CD127, CD44, NKG2D, NKp46, KIR2DL4, Perforin, Granzyme K, Granzyme B, CD2, CXCR3, CD94, TRAIL. For a description of various additional markers, as well as guidance on how to use them, see Bielekova et al, 2006, PNAS 103(15):5941-5946, Li et al, 2005, J. Immunol. 174:5187-5191.
[0036] As will be discussed in more detail below, the signal intensity ratios generated by the artificial NK cells depend upon a variety of different factors, such as, for example, the extinction coefficients of the fluorescent labels, potential quenching of fluorescent signal and, in the case of indirect markers, the affinities of the labeled molecules that bind the markers, and may not correlate identically with the molar ratios of the markers attached to the artificial lymphocyte substrate (as determined by the molar ratios of the marker synthons used to attach to the artificial lymphocyte substrates). However, as will be discussed in more detail below, and as will be recognized by skilled artisans, in many instances the markers and, in the case of indirect markers, labeled binding partners for the markers are selected such that the molar ratios of markers attached to the artificial lymphocyte substrate substantially correspond to the signal intensity ratios generated by the artificial NK cells.
[0037] As used herein, marker molar ratios substantially correspond to signal intensity ratios when the CD56:CD16 marker molar ratio is within about +/-10% of the
CD56.CD16 signal intensity ratio of the artificial NK cell, as measured under conditions of use.
[0038] It has been discovered that artificial NK cells bearing labeled anti-CD56 antibodies as direct CD56 markers and labeled anti-CD 16 antibodies as direct CD 16 markers in a CD56:CD16 molar ratio of about 100:1 migrate as human CD56bright NK cells in FACS-based cellular assays {see, Example 3). Similarly, artificial NK cells bearing these same labeled antibodies as direct CD56 and CD 16 markers, respectively, in a CD56:CD16 marker molar ratio of about 10: 1 migrate as human CD56dim NK cells in FACS-based cellular assays {see Example 4). Based on these results, it is expected that artificial NK cells bearing CD56 and CD 16 markers in molar ratios that generate CD56:CD16 marker signal intensity ratios of about 100:1 and about 10:1 will migrate as human CD56bright NK cells and CD56dim NK cells, respectively, in FACS-based assays.
[0039] It is also expected that artificial NK cells having CD56:CD16 marker molar ratios and/or signal intensity ratios within about +/- 10% or less, for example, within about +/- 9%, +/-8%, +/-7%, +/-6%, +/-5%, +/-4%, +/-3%, +1-2%, or +/-1%, of the stated marker and/or signal intensity ratios would likewise migrate as human CD56bnght or CD56d,m NK cells, respectively, in FACS-based assays.
[0040] As a specific embodiment, it is expected that artificial NK cells having a
CD56:CD16 marker molar ratio or generating a CD56:CD16 marker signal intensity ratio in the range of about 100:1 +/- 5%, including, for example, about 95:1, 96:1, 97: 1, 98:1, 99: 1, 100: 1, 101 : 1, 102: 1, 103:1, 104:1 or 105:1, will migrate as human CD56bright NK cells in FACS-based assays.
[0041] As another specific embodiment, it is expected that artificial NK cells having a CD56:CD16 marker molar ratio or generating a CD56:CD16 marker signal intensity ratio in the range of about 10:1 +1-5%, including, for example, about 9.5:1, 9.6:1, 9.7:1, 9.8:1, 9.9:1, 10:1, 10.1:1, 10.2:1, 10.3:1, 10.4:1 or 10.5:1, will migrate as human CD56dim cells in FACS-based assays.
[0042] The artificial lymphocyte substrates are typically polymeric beads or particles that mimic the characteristics of human lymphocytes in a flow cytometer, for example polymeric beads and particles that approximate the size, shape, volume, surface and/or internal complexity of a human lymphocyte. The beads or particles may be, but need not be, spherical in shape. Other possible shapes include discoidal, ovoid, elliptical. Suitable spherical beads or particles will typically have diameters in the range of about 6 μηι to about 9 um, and will have light scattering properties in a flow cytometer similar to human lymphocytes, as determined by measuring forward and side light scatter of the artificial lymphocyte substrates and comparing the light scatter to that of human lymphocytes, as described in Example 2 below.
[0043] Subject to the constraints above, the artificial lymphocyte substrate can be composed of any material suitable for use in FACS-based and other flow cytometry assays, including by way of example and not limitation, polymeric materials of natural origin, synthetic polymers, as well as silicates such as glass (coated or uncoated).
Exemplary polymers of natural origin include, but are not limited to, latex and cellulose. Exemplary synthetic polymers include, but are not limited to, acrylic resins (e.g., hexa- methylenediaminepolyacryl resins and related polymers), polyacrylamides (e.g., poly(acryloylsarcosine methyl ester), poly N-{2-(4-hydroxylphenyl)ethyl}acrylamide, dimethylacrylamide, optionally cross-linked with N,N'-65-acrylolyl ethylene diamine), polystyrenes (including poly(halomethylstyrene), poly(halostyrene),
poly(acetoxystyrene)), polystyrene cross-linked with divinylbenzene and grafted copolymers such as polyethylene glycol/polystyrene, polyethers, polyvinyls,
polymethacrylates (including polymethylmethacrylate), polyacylamides, polyurethanes, polycarbonates, polyesters, polyamides, and combinations thereof, poly (N- acryloylpyrrolidine) resins, or Wang resins. Exemplary silicates include, but are not limited to, silica, porous silicates (e.g., controlled pore-glass), glass coated with a hydrophobic polymer including cross-linked polystyrene or a fluorinated ethylene polymer which provides a material having a rigid or semi-rigid surface.
[0044] The material should either include reactive or other groups suitable for attaching the CD56 and CD 16 markers, or be capable of being treated to yield reactive or other groups suitable for effecting attachment. The identity of the reactive or other groups used
will depend upon the specific mode and/or chemistry desired for attachment, as will be described in more detail, below. A variety of beads and particles suitable for use as artificial lymphocyte substrates in the artificial NK cells described herein are available commercially, and include, by way of example and not limitation, beads or microspheres ranging in size from 5 to 10 μπι, made from any of the polymeric materials listed above, obtainable from a variety of sources including Bangs Laboratories, Inc., Fishers, IN, Polysciences, Inc., Warrington, PA, Invitrogen, Carlsbad, CA, and BD Biosciences, Inc., San Diego, CA.
[0045] The CD56 and CD 16 markers may comprise the actual human CD56 and CD 16 polypeptides, respectively, but need not. Rather, they may comprise virtually any type of molecular moiety capable of generating fluorescence signals, either directly or indirectly, under the desired conditions of use of the artificial NK cell. The expressions "CD56 marker" and "CD 16 marker" are used herein for convenience regardless of the actual composition of the markers, and in particular regardless of whether the markers include human CD56 and human CD16 polypeptide sequences, respectively. Thus, in embodiments of artificial NK cells where the CD56 and CD16 markers do not comprise human CD56 and human CD16 polypeptide sequences, respectively, the markers act as surrogates for the levels of CD56 and CD16, respectively, expressed on human NK cells.
[0046] The CD56 and CD 16 markers may generate detectable fluorescence signals directly or indirectly. Markers that generate fluorescence signals directly generally comprise a fluorescent moiety, such as a fluorescent dye. An exemplary embodiment 10a of an artificial NK cell comprising direct CD56 and CD16 markers is illustrated in FIG. 6A.
[0047] In FIG. 6A, direct CD56 marker 14a comprises a fluorescent dye moiety 18 linked to artificial lymphocyte substrate 12 by way of an optional linker Li. Direct CD 16 marker 16a comprises a fluorescent dye moiety 20 linked to artificial lymphocyte 12 by way of an optional linker L2, which may be the same or different as optional linker Lj. Fluorescent dye moieties 18 and 20 generate fluorescence signals that are distinguishable from one another. Artificial lymphocyte substrate 12 bears n number of CD56 markers 14a and m number of CD 16 markers 16a, providing a molar ratio of n:m as described herein for artificial NK cells. For example, the n:m ratio for an artificial CD56bnght NK
cell could be 100:1. In another example, the n:m ratio could be 10: 1, for an artificial CD56dim NK cell.
[0048] Markers that generate fluorescence signals indirectly generally comprise moieties that specifically bind to other molecules that comprise fluorescent dye moieties, or that specifically bind to molecules that in turn specifically bind yet other molecules that comprise fluorescent dye moieties. An exemplary embodiment 10b of an artificial NK cell comprising indirect CD56 and CD 16 moieties is illustrated in FIG. 6B.
[0049] In FIG. 6B, indirect CD56 marker 14b comprises a binding moiety 22 linked to artificial lymphocyte substrate 12 by way of an optional linker Li. Indirect CD16 marker 16b comprises a binding moiety 24 linked to artificial lymphocyte substrate 12 by way of an optional linker L2, which may be the same or different than optional linker Lj.
Binding moieties 22 and 24 are each capable of specifically binding other molecules (binding partners). The binding partner of binding moiety 22 is different than the binding partner of binding moiety 24. Artificial lymphocyte substrate 12 bears n number of CD56 markers 14b and m number of CD16 markers 16b, providing a molar ratio of n:m as described herein for artificial NK cells. For example, the n:m ratio for an artificial CD56bright NK cell could be 100: 1 In another exampie5 the n:m ratio could be 10:1, for an artificial CD56dim NK cell.
[0050] The binding moieties 22 and 24 can comprise virtually any type of molecule or moiety capable of specifically binding a binding partner. As specific, non-limiting examples, the binding moieties of the markers could comprise antigens for antibody binding partners, such as for example, specific peptide or polypeptide sequences for which antibody binding partners are known, enzymes or active binding pockets thereof for which substrates are known, or sequence specific oligo- or polynucleotides capable of hybridizing to complementary oligo or polynucleotide binding partners.
[0051] For certain types of assays, there may be advantages to utilizing artificial NK cells that approximate as closely as possible human NK cells. Accordingly, in a specific exemplary embodiment, indirect CD56 marker 14b comprises a polypeptide having an amino acid sequence that corresponds to all or a portion of a human CD56 polypeptide and indirect CD16 marker 16b comprises a polypeptide having an amino acid sequence that corresponds to all or a portion of a human CD 16 polypeptide.
[0052] CD56, also known as Neural Cell Adhesion Molecule ("NCAM"), is a founding member of a large family of cell surface glycoproteins that contain structural motifs related to immunoglobulin fibronectin type III domains (Brummendorf et al, 2001, Curr. Opin. Cell. Biol. 13:611-618; Walmod et al, 2004, Neurochem. Res. 29:2015-2035). CD56 generally has three main domains: an extracellular domain, a transmembrane domain, and an intracellular domain. Human CD56 is encoded by a single-copy gene on chromosome 11 that spans more than 314 kb and contains 19 major exons as well as 6 additional smaller exons (Walmod et al, 2004, Neurochem. Res. 29:2015-2035; Walsh et al, 1986, Brain Res. 387:197-200; Nguyen et al, 1986, J. Cell. Biol. 102:711-715).
Alternative splicing results in the expression of three major isoforms, referred to as CD56120kd, CD56I40kd, CD56180kd. The isoforms share the same extracellular domain but differ with respect to their intracellular domains and how they associate with the cell membrane {see, e.g., Gattenlohner et al, 2009, Am. J. Pathol. 174(4):1160-1171). For example, CD56120kd lacks transmembrane and intracellular domains, instead associating with the membrane via a glycosylphosphatidylinositol anchor. CD56140kd and CD56180kd both have transmembrane domains, but differ with respect to their intracellular domains (for a schematic representation of the three major isoforms of human CD56, see
Gattenlohner et al. , 2009, Am. J. Pathol. 174(4): 1160- 1171 , at FIG. 2 A and its associated description). Human NK cells predominantly express the 140 kD isoform (Lanier et al, 1991, J. Immunol. 146:4421-4426). The amino acid sequence of human CD56140kd is provided in FIG. 9A. The signal sequence is represented in italics. The sequence of the mature polypeptide spans 829 residues, with residue +1 being the N-terminus of the mature polypeptide. The extracellular domain spans residues 1 to 689 (underlined in FIG. 9A). Also shown in FIG. 9A, in bolded type is the epitope understood to be recognized by anti-CD56 monoclonal antibody NCAM-13 (Product no. 556324, BD Pharmingen™, BD Biosciences; and see Gattenlohner et al, 2009, Am. J. Pathol. 174(4): 1160-1171, Table 1 showing isoform specificity and epitope mapping for a number of commercially available anti-CD56 antibodies).
[0053] In embodiments of artificial NK cells 10b in which the binding moiety 22 of indirect CD56 marker 14b comprises a human CD56 (hCD56) polypeptide, the polypeptide may be the entire human CD56140kd sequence, as illustrated in FIG. 9A, or a portion or fragment of this sequence. In a specific embodiment, the binding moiety 22
comprises a polypeptide corresponding in sequence to the extracellular domain of human CD56i40kd (residues i to 689 of FIG. 9A), or a fragment or portion thereof. In another specific embodiment, binding moiety 22 comprises a polypeptide corresponding in sequence to an antigen used to raise an anti-CD56 antibody, or an epitope recognized by an anti-CD56 antibody. As a specific example, binding moiety 22 could comprise a polypeptide corresponding in sequence to residues 372 to 576 of FIG. 9A (epitope mapped for anti-CD56 monoclonal antibody NCAM-13 (Product no. 556324, BD
Pharmingen™, BD Biosciences)).
[0054] CD 16, also known as type III low affinity receptor for the Fc portion of immunoglobulin G (Fcylll), belongs to the immunoglobulin superfamily and, along with type I and II receptors for immunoglobulin G, plays an important role in inducing phagocytosis of pathogens (Fridman, 1991, FASEB J. 5 (12): 2684-90). Two major isoforms of CD 16, CD16A (or FcylllA) and CD16B (FcylllB) are encoded by different genes on Chromosome 1 (Ravetch et al, 1989, J. Exp. Med. 170:481). Human NK cells express CD16A, which contains an extracellular domain, a transmembrane domain, and an intracellular domain (Tamm et al, 1996, J. Immunol. 157:1576-1581; Peltz et al., 1989, Proc. Nat'l Acad. Sci. USA 86:1013). In contrast, CD16B is expressed exclusively on neutrophils and lacks transmembrane and intracellular domains, instead associating with the membrane via a glycosylphosphatidylinositol anchor (Tamm et al, 1996, J. Immunol. 157: 1576-1581; Simmons et al, 1988, Nature 333:568; Hundt et al, 1992, Eur. J. Immunol. 22:811). The amino acid sequence of CD16A is provided in FIG. 9B. The signal sequence is represented in italics. The sequence of the mature polypeptide spans 237 residues, with residue +1 being the N-terminus of the mature polypeptide. The extracellular domain spans residues 1 to 192 (underlined in FIG. 9B). Also shown in FIG. 9A, in bolded type, are the residues understood to be required for binding by anti- CD16 monoclonal antibody 3G8 (Product no. 557710, BD Pharmingen™, BD
Biosciences and see Tamm et al, 1996, J. Immunol. 157:1576-1581, Table 1).
[0055] In an embodiment of artificial NK cells 10b in which the binding moiety 24 of indirect marker CD16 marker 16b comprises a human CD16 (hCD16) polypeptide, the polypeptide may be the entire human CD16A sequence, as illustrated in FIG. 9B, or a portion or fragment of this sequence. In a specific embodiment, binding moiety 24
comprises a polypeptide corresponding in sequence to the extracellular domain of human CD16A (residues 1-192 of FIG. 9B), or a fragment or portion thereof. In another specific embodiment, binding moiety 24 comprises a polypeptide corresponding in sequence to an antigen used to raise an anti-CD 16 antibody, or an epitope recognized by an anti-CD 16 antibody. As a specific example, binding moiety 24 could comprise a polypeptide corresponding in sequence to residues 110 to 170 of FIG. 9B encompassing residues required for binding of anti-CD16 monoclonal antibody 3G8 (Product no. 557710, BD Pharmingen™, BD Biosciences).
[0056] In the exemplary artificial NK cells of FIG. 6B, fluorescence signals are generated indirectly, via binding partners for the binding moieties. Exemplary embodiments are illustrated in FIGS. 7A and 7B. With reference to FIG. 7A, artificial NK cell 10b is incubated with labeled binding partners 26 and 28. Labeled binding partner 26 comprises a binding moiety 30 specific for binding moiety 22 of indirect CD56 marker 14b and a fluorescent dye moiety 18, and labeled binding partner 28 comprises a binding moiety 32 specific for binding moiety 24 of indirect CD marker 16b and a fluorescent dye moiety 20. The nature and identities of binding moieties 30 and 32 of binding partners 26 and 28 will depend upon the nature and identities of binding moieties 22 and 24 on artificial NK cell 10b. As a specific, non-limiting example, binding partners 26 and 28 could be labeled anti-22 and anti-24 antibodies respectively, where the fluorescence signals generated by their respective fluorescent dye moieties 18 and 20 are distinguishable or resolvable from one another. In the specific exemplary embodiment of artificial NK cell 10b discussed above where binding moiety 22 is a human CD56 ("hCD56") polypeptide and binding moiety 24 is a human CD 16 ("hCD16") polypeptide, binding partners 26 and 28 could be labeled anti-hCD56 and labeled anti-hCD16 antibodies, respectively, such as, for example, FITC-labeled anti-hCD56 monoclonal antibodies and APC-labeled anti- hCD16 antibodies.
[0057] Suitable anti-CD56 and anti-CD 16 antibodies that are either labeled or that can be conjugated with custom labels are available commercially. Suitable anti-CD56 antibodies include but are not limited to, 1B6 (Novocastra), B159, NCAM 13 and 12F11 (BD Biosciences), H-300, 123C3 and C-20 (Santa Cruz Biotechnology, Inc.), NCAM-0B 11 (Sigma-Aldrich), MEM188 (BioDesign), MOC-1 (Millipore) available with or without
conjugated fluorophores. Suitable anti-CD 16 antibodies include, but are not limited to, J551 1, LNK16, MEM-154, MEM-168, and BL-LGL/1 (Thermo Scientific, Pierce antibodies), 3G8 and B73.1 (BD Biosciences), LNK16, 3G8 (Abeam).
[0058] An exemplary embodiment of artificial NK cells 10b in which signals are generated utilizing a sandwich-type assay is illustrated in FIG. 7B. With reference to FIG. 7B, artificial NK cell 10b is incubated with unlabeled binding partners 34 and 36 that are specific for, respectively, binding moieties 22 and 24 on artificial NK cell 10b. The resultant complex is then incubated with labeled binding partners 38 and 40 that are specific for, respectively, unlabeled binding partners 34 and 36. The nature and identities of the various unlabeled binding partners 34 and 36 and labeled binding partners 38 and 40 will depend upon the desired binding interactions. In the specific exemplary embodiment of artificial NK cell 10b discussed above where binding moiety 22 is a human CD56 polypeptide and binding moiety 24 is a human CD16 polypeptide, unlabeled binding partners 34 and 36 could be unlabeled anti-hCD56 and anti-hCD16 antibodies, respectively, that are derived from different species, for example, goat anti- hCD56 monoclonal antibodies and rabbit anti-hCD16 monoclonal antibodies. Labeled binding partners 38 and 40 could then be labeled species-specific antibodies, such as FITC -labeled anti-goat IgG monoclonal antibodies and APC-labeled anti-rabbit IgG monoclonal antibodies.
[0059] A variety of fluorescent dyes can be used to generate detectable signals from the artificial NK cells, provided that the CD56 and CD 16 markers generate distinguishable fluorescent signals. By distinguishable, it is meant that that the fluorescence emissions spectra generated by the CD56 and CD16 markers are sufficiently distinct, i.e., sufficiently non-overlapping under the conditions of use of the artificial NK cells, that they are distinguishable from one another using standard photodetection systems such as photodetectors employing a series of band pass filters and photomultiplier tubes, charged- coupled devices (CCD), spectrographs, etc., as exemplified by the systems described in U.S. Pat. Nos. 4,230,558 and 4,811,218 or in Wheeless et al, 1985, Flow Cytometry: Instrumentation and Data Analysis, pp. 21-76, Academic Press, New York. This is generally achieved by using fluorophores that have emissions peaks that differ by at least about 50 nm, although in some instances, fluorophores having emissions peaks that differ
by fewer than 50 nm may also be distinguishable or resolvable, depending upon the overall shape and width of their respective emissions spectra. Use of fluorophores having emission maxima that differ by at least about 75 nm, 100 nm, 125 nm, 150 nm, 175 nm, 200 nm, or even more, is preferred. As will be recognized by skilled artisans, fluorophores having emissions spectra with minimal or no overlap may be more readily distinguished or resolved than those having overlap. Generally, the greater the distance between the emissions maxima or peaks, the less the overlap. Accordingly, use of fluorophores having minimal or no overlap in their emissions spectra is preferred. When artificial CD56bright NK cells and artificial CD56dim NK cells are used together, the CD56- and CD16-generated fluorescent signals of artificial CD56bri8ht NK cells may be distinguishable, or resolvable, from the CD56- and CD16-generated fluorescent signals of artificial CD56dim NK cells.
[0060] Suitable dyes can selected from among fluorescein, rhodamine, Alexa Fluor® dyes, phycobiliproteins, cyanine (Cy) dyes, and other fluorescent dyes. Specific examples of fluorescein dyes include fluorescein isothiocyanate (FITC), 5- carboxyfluorescein (5-FAM), 6-carboxyfluorescein (6-FAM), hexachlorofluorescein (HEX), tetrachlorofluorescein (TET), 2', 4',5',7'-Tetrabromosulfonefluorescein and dichlorotriazinylamine fluorescein. Additional fluorescein dyes can be found, for example, in U.S. Pat. No. 5,840,999; U.S. Pat. No. 5,066,580; U.S. Pat. No. 4,439,356; U.S. Pat. No. 4,481,136; U.S. Pat. No. 5,188,934; and U.S. Pat. No. 5,654,442. Specific examples of rhodamine dyes include rhodamine B, 5-carboxyrhodamine, rhodamine X (ROX), 4,7-dichlororhodamine X (dROX), rhodamine 6G (R6G), rhodamine 110 (ROX), 4,7-dichlororhodamine 110 (dROX), tetramethyl rhodamine (TAMRA) and 4,7- dichlorotetramethylrhodamine (dTAMRA), REG, and Rhodamine Green. Additional rhodamine dyes can be found, for example, in U.S. Pat. No. 5,936,087; U.S. Pat. No. 5,750,409; U.S. Pat. No. 5,366,860; U.S. Pat. No. 5,231,191; U.S. Pat. No. 5,847,162; and U.S. Pat. No. 6,025,505. Specific examples of Alexa Fluor® dyes include Alexa Fluor® 350, Alexa Fluor® 405, Alexa Fluor® 430, Alexa Fluor® 488, Alexa Fluor® 500, Alexa Fluor® 514, Alexa Fluor® 532, Alexa Fluor® 546, Alexa Fluor® 555, Alexa Fluor® 568, Alexa Fluor® 594, Alexa Fluor® 610, Alexa Fluor® 633, Alexa Fluor® 647, Alexa Fluor® 660, Alexa Fluor® 680, Alexa Fluor® 700, and Alexa Fluor® 750. Specific examples of phycobiliproteins include phycocyanin, allophycocyanin,
phycoerythrin (PE), B-phycoerythrin, R-phycoerythrin, and C-phycoerythrin, perdinin chlorophyll protein (PerCP), phycoerythrin/cyanine 5 (PE/Cy5), and allophycocyanin (APC). Additional phycobiliprotein dyes can be found, for example, in U.S. Pat. No. 4,542,104; U.S. Pat. No. 5,171,846; U.S. Pat. No. 5,994,089; and U.S. Pat. No. 6,387,622. Specific examples of cyanine dyes include Cy2, Cy3, Cy5, and Cy7, as well as dye pairs such as phycoerythrin/cyanine 5 (PE/Cy5). Additional cyanine dyes can be found, for example, in U.S. Pat. No. 5,268,486; U.S. Pat. No. 5,616,502; U.S. Pat. No. 7,553,959; U.S. Pat. No. 7,709,653; U.S. Pat. No. 7,842,811 ; and U.S. Pat. No. 7,871,773. Other specific examples of fluorescent dyes are Texas Red® dyes, amine-reactive BODIPY dyes, such as BODIPY 493/503, BODIPY 530/550, BODIPY 558/568, BODIPY
564/570, BODIPY 576/589, BODIPY 581/591, BODIPY 630/650, BODIPY 650/655, BODIPY FL, BODIPY R6G, BODIPY TMR, and, BODIPY-TR; NED, LIZ, PET, 6- JOE, Oregon Green 488, Oregon Green 500, Oregon Green 514, and Pacific Blue.
[0061] As will be appreciated by one of skill in the art, any combination of fluorescent dyes that generate distinguishable signals can be used. Suitable combinations of dyes are known from the multiplex DNA detection and PCR arts and include combinations of FITC or Alexa Fluor® 488, with PE or PE Cy5, or with Alexa Fluor® 647 or APC;
combinations of HEX, NED, ROX, and/or 6-FAM; combinations of 6-FAM, VIC, NED, PET, and/or LIZ; or combinations of 5-FAM, JOE, NED and/or ROX, from which pairs, triplets, or quartets of dyes can be selected. A specific, non-limiting example of a suitable pair of fluorescent dyes is APC and FITC. Another specific, non-limiting example of a suitable pair is Alexa Fluor® 488 and Alexa Fluor® 647. Other possible combinations are known to the person of skill and are also possible.
[0062] Fluorophores bear, or can be modified to bear, a functional group that is used to couple them to a binding moiety, a binding partner, or to the artificial lymphocyte substrate. Techniques for functionalizing fluorophores are well known in the art.
Functionalized fluorophores are available commercially from a variety of sources. See e.g., Molecular Probes® Products, Invitrogen, Carlsbad, CA; Pierce Protein Research Products, Thermo Fisher Scientific, Rockford, IL. Techniques for conjugating fluorophores to a variety of molecules, including, but not limited to, peptides,
polypeptides, nucleotides, polynucleotides, and hormones, are also well known in the art.
See, e.g., U.S. Pat. Nos. 4,481136, 5,994,089, 6,191,278, 7,553,959, 7,927,830, the contents of each of which is incorporated by reference in their entireties herein.
[0063] The detectable signals generated by the CD56 and CD16 markers can be produced by fluorophore(s) having the same or different excitation wavelengths. For some applications, it may be desirable to use a single excitation wavelength for both fluorophores. For these applications, one or both markers can generate signals by way of fluorescence resonance energy transfer (FRET)-based fluorescent moieties, or energy- transfer dye pairs. In some embodiments, a FRET dye can be used in conjunction with a standard fluorescence dye. While a standard fluorescent dye and a FRET-based fluorescent dye are excited at the same wavelength, the FRET-based dye will emit fluorescent signals at a different, typically longer, wavelength than the standard fluorescent dye. A thorough discussion of the various structures, synthesis and use of certain energy-transfer dye pairs is provided in U.S. Pat. No. 5,188,934, U.S. Pat. No. 5,654,419, U.S. Pat. No. 5,775,409, U.S. Pat. No. 5,800,996, and U.S. Pat. No. 5,863,727, the disclosures of which are incorporated herein by reference in their entireties.
[0064] Fluorescent moieties that generate signals in very narrow emission spectra are particularly useful for the methods described herein, as they permit greater resolution between fluorescent signals and can be used for applications where detection of additional markers is desired. Inorganic crystalline nanoparticles made of semiconducting materials - referred to as quantum dots - have been developed and used in a variety of assays in which multiple fluorescent signals are simultaneously detected and distinguished from each other. Suitable quantum dots can be selected and used to label CD56 and CD16 markers. A thorough discussion of the various structures, synthesis and use of quantum dots is provided in WO 2001/77391, and references cited therein; see also Ibanez-Peral et al, 2008, Int'l J. Mol. Sci. 9:2622-2638, the disclosures of each of which are incorporated herein by reference in their entirety. For guidance on the use of quantum in flow cytometric assays, see Chattopadhyay, 2011, Methods Cell Biol. 102:463-77, Jennings et al, 2011, ACS Nano. 5(7):5579-93; Lim et al, 2009, Nanotechnology 20(47):475102,
[0065] As illustrated in FIGS. 6A and 6B, the markers are attached to artificial lymphocyte 12 via optional linkers Li and L2, which may be the same or different (generally referred to herein as linker "L"). The nature of linker L will depend upon the
particular application, point of attachment and type of conjugation desired. The linker can be hydrophilic or hydrophobic, long or short, rigid, semirigid or flexible, depending upon the particular application. The linker can be optionally substituted with one or more substituents or one or more additional linking groups, which may be the same or different, thereby providing a "polyvalent" linking moiety capable of conjugating with multiple molecules or substances. Preferably, however, linker L does not include such additional substituents or linking groups.
[0066] A wide variety of linkers L comprised of stable bonds are known in the art, and include by way of example and not limitation, alkyldiyls, substituted alkyldiyls, elkylenos, substituted alkylenos, heteroalkyldiyls, substituted heteroalkyldiyls, heteroalkylenos, substituted heteroalkylenos, acyclic heteroatomic bridges, aryldiyls, substituted aryldiyls, arylaryldiyls, substituted arylaryldiyls, arylalkyldiyls, substituted arylalkyldiyls, heteroaryldiyls, substituted heteroaryldiyls, heteroaryl-heteroaryl diyls, substituted heteroaryl-heteroaryl diyls, heteroarylalkyldiyls, substituted
heteroarylalkyldiyls, heteroaryl-heteroalkyldiyls, substituted heteroaryl-heteroalkyldiyls, and the like. Linkers may also be based upon biological or non-biological polymers, such as, for example, peptides, polypeptides, oligonucleotides, polynucleotides, polyalkylene glycols, etc. Thus, linker L may include single, double, triple or aromatic carbon-carbon bonds, nitrogen-nitrogen bonds, carbon-nitrogen, carbon-oxygen bonds and/or carbon- sulfur bonds, and may therefore include functionalities such as carbonyls, ethers, thioethers, carboxamides, sulfonamides, ureas, urethanes, hydrazines, etc. In one specific exemplary embodiment, linker L has from 1-20 non-hydrogen atoms selected from the group consisting of C, N, O, and S and is composed of any combination of ether, thioether, amine, ester, carboxamide, sulfonamides, hydrazide, aromatic and
heteroaromatic bonds.
[0067] Choosing a linker having properties suitable for a particular application is within the capabilities of those having skill in the art. For example, where a rigid linker is desired, L may be a rigid polypeptide such as polyproline, a rigid polyunsaturated alkyldiyl or an aryldiyl, biaryldiyl, arylarydiyl, arylalkyldiyl, heteroaryldiyl,
biheteroaryldiyl, heteroarylalkyldiyl, heteroaryl-heteroaryldiyl, etc. Where a flexible linker is desired, L may be a flexible polypeptide such as polyglycine or a flexible
saturated alkanyldiyl or heteroalkanyldiyl. Hydrophilic linkers may be, for example, polyalcohols or polyethers such as polyalkyleneglycols. Hydrophobic linkers may be, for example, alkyldiyls or aryldiyls.
[0068] Linkers suitable for use in most biological applications include (C1-C12)
alkyldiyls, particularly alkanylenos such as methano ( -CH2-), ethano (-CH2-CH2-), n- propano (-CH2-CH2-CH2-), H-butano (-CH2-CH2-CH2-CH2-), rc-pentano
-CH
2-CH
2-CH
2-CH
2-CH
2-) and rc-hexano (-CH
2-CH
2-CH
2-CH
2-CH
2-CH
2-); (C
5-C
20)
naphtha-2,6-diyl ( ^ ι ^Ύ^^-^τ
ν )
and naphtha-2,7-diyl (
arylalkyldiyls, for example those having the structural formula -(CH2), -Φ- or -(CH^-Ψ-, where each i is independently an integer from 1 to 6, Φ is phenyldiyl for example phena- 1,3-diyl or phena-l,4-diyl) and Ψ is naphthyldiyl (for example naphtha-2,6-diyl or naphtha-2,7-diyl). Rigid linkers include -(CH2)rNR"-C(0)-0-and -(CH2),-NR"-C(0)^-, where I, R", Φ and Ψ are as previously defined. Analogs of all of these linkers L containing one or more heteroatomic groups, particularly those selected from the group consisting of O, S, N and NR", where R" is hydrogen or (Ci-Ce) alkyl, can also be conveniently used. For further description of, and guidance on how to use, linkers, see Carmon et al, 2002, BioTechniques 32(2):410-420; Hermanson 1996, Bioconjugate Techniques, San Diego, Academic Press and Hermanson, 2008, Bioconjugate
Techniques, 2nd Edition, Academic Press; Jones, 2001, IVD Tech. Nov/Dec 39; Pierce Chemical Company, 2001, Double-agents™ cross-linking reagents selection guide;
Siiman et al., 2001, J. Colloid Interface Sci 234:44-58; Sjoroos et al., 2001, Clin. Chem 47(3):498-504, the contents of each of which are incorporated herein by reference in their entireties.
[0069] In a specific embodiment, linker L may be a polypeptide. Specific examples of artificial NK cells 10a (FIG. 6A) utilizing direct CD56 and CD 16 markers in which Li and L2 are polypeptides, in this instance anti-CD56 and anti-CD16 antibodies, respectively, are described in Examples 3 and 4.
[0070] The markers can be conjugated to artificial lymphocyte substrate 12, optionally via linkers, using a variety of different conjugation means. For example, the conjugation can be mediated via hydrophobic interactions, ionic attraction, through the use of pairs of specific binding molecules such as biotin and avidin/streptavidin or through covalent attachment. When conjugation via hydrophobic interactions is desired, optional linker L includes a hydrophobic moiety that is capable of forming hydrophobic interactions with a hydrophobic moiety on artificial lymphocyte substrate 12. Typical hydrophobic moieties include, but are not limited to, unsubstituted and substituted aryl, arylalkyl, arylaryl, heteroaryl, heteroarylaklyl and heteroaryl-heteroaryl groups. When the hydrophobic moiety is substituted, the substituents are preferably nonpolar, more preferably hydrophobic. Suitable hydrophobic moieties for forming non-covalent conjugates will be apparent to those of skill in the art.
[0071] When conjugation via ionic attraction is desired, optional linker L includes a charged moiety having a net charge of a polarity opposite to a net charge on the molecule or substance to be conjugated. Typical charged moieties include, by way of example and not limitation, quaternary ammoniums, carboxylates and sulfonates, including salts thereof. A variety of cyclic quaternary ammoniums that are suitable for use are described in U.S. Patent No. 5,863,753 {see, e.g., Cols. 8-9), the disclosure of which is incorporated herein by reference.
[0072] When conjugation via pairs of specific binding molecules such as biotin and avidin/streptavidin is desired, optional linker L will be coupled to one member of the binding pair. Artificial lymphocyte substrate cell substrate 12 will bear the other member of the binding pair. Where one of the members of the specific binding pair is a small molecule, such as biotin or a hormone, that member is preferably coupled to optional linker L, while the other member, e.g., avidin or streptavidin, is coupled to the artificial lymphocyte substrate. In such embodiments, biotin can be functionalized, i.e. it can bear a functional group that can be reacted with a complementary functional group on a fluorescent dye or a binding moiety, as will be described further below. A variety of functionalized biotins capable of being covalently linked to reactive groups such as amines or carboxyls are commercially available {e.g., Molecular Probes® Products, Invitrogen, Carlsbad, CA; Pierce Protein Research Products, Thermo Fisher Scientific,
Rockford, IL). Other suitable pairs of specific binding molecules are known, including, for example, Protein A from S. aureus which binds to the constant region of polyclonal and monoclonal antibodies with high affinity. For embodiments in which Protein A and an antibody are used as a binding pair, typically, artificial lymphocyte substrate 12 will bear Protein A, while optional linker L will be coupled to the antibody. Protein A is available from a variety of commercial sources, as are beads coated with Protein A and ready for conjugation to antibodies (see, e.g., ProActive® Protein A-Coated Microspheres by Bangs Laboratories, Inc., Fishers, IN).
[0073] A specific, non-limiting exemplary embodiment of an artificial NK cell 10c comprising direct CD56 and CD 16 markers where linkages are effected by specific binding pairs of molecules is illustrated in FIG. 6C. As shown, artificial lymphocyte substrate 12 is coupled to one member of the binding pair, streptavidin or SA, and the other member of the binding pair, biotin or B, is coupled, via optional linker Li or L2, to fluorescent dye moieties 18 and 20. Artificial lymphocyte substrate 12 bears n number of CD56 markers 14a and m number of CD 16 markers 16a, providing a molar ratio of n:m as described herein for artificial NK cells. For example, the n:m ratio for an artificial CD56bright NK cell could be 100:1. In another example, the n:m ratio could be 10:1, for an artificial CD56dim NK cell. A similar exemplary embodiment of an artificial NK cell lOd comprising indirect CD56 and CD16 markers is illustrated in FIG. 6D. The artificial lymphocyte substrate 12 is coupled to one member of the binding pair, streptavidin or SA, and the other member of the binding pair, biotin or B, is coupled, via optional linker Li or L2, to binding moieties 22 and 24. Artificial lymphocyte substrate 12 bears n number of CD56 markers 14b and m number of CD16 markers 16b, providing a molar ratio of n:m as described herein for artificial NK cells. For example, the n:m ratio for an artificial CD56bright NK cell could be 100:1. In another example, the n:m ratio could be 10: 1, for an artificial CD56dim NK cell. In FIGs. 6C and 6D, the various optional linkers Li and L2 may be the same or different.
5.2. Methods of Making Artificial NK Cells
[0074] The artificial NK cells of the present disclosure comprise a CD56 and a CD16 marker that are linked to an artificial lymphocyte substrate, either by covalent or other means, in molar ratios such that, under the conditions of use, the artificial NK cells
generate two distinguishable signals, e.g., fluorescent signals, having a defined ratio as discussed above.
[0075] Preferably, the markers are conjugated to artificial lymphocyte substrate 12 via covalent attachment. In this preferred embodiment, artificial lymphocyte substrate 12 and the various synthons used to synthesize the artificial NK cells will include complementary pairs of functional groups Rx and Fx capable of reacting with each other to yield covalent linkages (referred to herein as "complementary reactive functional groups"). For example, with reference to FIG. 8, exemplary artificial NK cell 10a of FIG. 6A in which CD56 and CD16 markers 14a and 16a are covalently attached to artificial lymphocyte substrate 12 can be obtained by reacting artificial lymphocyte substrate 12, which includes or is modified to include reactive functional groups Rx, with marker synthons 14a and 16a, which include complementary reactive functional groups Έχ, which may be the same or different, under conditions in which Rx and ¥x react to yield a covalent linkage ("LK").
[0076] The exact identities of R* and ¥x will depend upon the nature of the desired covalent linkage and the chemistry used to form the covalent linkage. Generally, reactive group Rx is a functional group that is capable of reacting with a complementary functional group F* under specified reaction conditions to form a covalent linkage. However, those of skill in the art will recognize that a variety of functional groups that are typically unreactive under certain reaction conditions can be activated to become reactive. Groups that can be activated to become reactive include, e.g., carboxylic acids and esters, including salts thereof. Such groups are referred to herein as "activatable precursors" and are specifically intended to be included within the expression "reactive group."
[0077] Pairs of complementary reactive groups Rx and Fx suitable for forming covalent linkages with one another under a variety of different reaction conditions are well-known. Any of these complementary pairs of groups can be used to covalently conjugate the markers to artificial lymphocyte substrate 12. As provided herein, assignment of a reactive group to the Rx or Fx member of a pair of complementary reactive functional groups is not intended to be limiting and members of Rx and Fx pairs are generally interchangeable. In one convenient embodiment, reactive group Rx and complementary functional group F* comprise complementary electrophiles and nucleophiles (or their
respective activatable precursors). In a preferred embodiment, reactive group Rx is a group that reacts with, or that can be readily activated to react with, an amine, a thiol or an alcohol.
[0078] In some embodiments, one of R* and Fx will include an activated ester. As understood in the art, "activated esters" generally have the formula -C(0)Q, where Ω is a good leaving group. Exemplary good leaving groups include, by way of example and not limitation: oxysuccinimidyl; N-succinimidyl; oxysulfosuccinimidyl; 1-oxybenzotriazolyl; and -ORa, where Ra is selected from the group consisting of (C4-C2o) cycloalkyl (e.g., cyclohexyl), heterocycloalkyl, (Cs-C2o) aryl, (Cs-C2o) aryl substituted with one or more of the same or different electron-withdrawing groups (e.g., -N02, -F, -CI, -CN, -CF3, etc.), heteroaryl, and heteroaryl substituted with one or more of the same or different electron- withdrawing groups, H-dialkylaminoalkyls (for example, 3-dimethylaminopropyl) and N- morpholinomethyl, or R° is used to form an anhydride of the formula -OCOR* or -OCNR* NHRC, where R* and Rc are each independently selected form the group consisting of (C\-Ce) alkyl (Ci-C6) perhaloalkyl, (Ci-C6) perfluoroalkyl and (Ci-C6) alkoxy. A preferred activated ester is NHS ester.
[0079] In a convenient embodiment, reactive group Rx is a photoactivatable group that becomes chemically reactive only after illumination with light of an appropriate wavelength and complementary functional group Fx is a group capable forming a covalent linkage with the chemically reactive species. Such photoactivatable groups can be conveniently used to photo cross-link the markers to artificial lymphocyte substrate 12. Exemplary photoactivatable groups suitable for conjugation via light-activated cross- linking include, but are not limited to, azido (-N3), 4-azido-phenyl and 2-nitro-4-azido- phenyl. Conjugation using photoactivatable groups typically involves illuminating a mixture comprising the photoactivatable dyes and the molecule or substance to be conjugated, followed by separation of unreacted dyes and byproducts.
[0080] As will be recognized by those of skill in the art, reactive group Rx can comprise any electrophilic, nucleophilic or photoactivatable groups. The selection of reactive group R* used to covalently conjugate artificial lymphocyte substrate 12 of the invention to the other molecule or substance typically depends upon the identity of the
complementary functional group F* on the molecule or substance to be conjugated. The
types of complementary functional groups typically present on molecules or substances to be conjugated include, but are not limited to, amines, thiols, alcohols, phenols, aldehydes, ketones, phosphates, imidazoles, hydrazines, hydroxylamines, mono- and disubstituted amines, halides, epoxides, sulfonate esters, carboxylic acids or carboxylates, or a combination of these groups. Specific exemplary pairs of complementary reactive functional groups R* and Έχ that can be reacted to yield covalent linkages are provided in Table 1, below:
Table 1
Exemplary complementary reactive groups
Rx (Activatable Linkage Exemplary precursor) activation
reagents
Activated ester Amine Amide bond l-Ethyl-3-(3- (Carboxyl) dimethylaminoprop yl)carbodiimide
Activated aldehyde Amine Amide bond Glutaraldehyde (Amino)
Activated cyanate ester Amine Isourea Cyanogen bromide (Hydroxyl) derivative
Hydrazide group Oxidized Hydrazone
carbohydrate bond
Chloromethyl group Amine Amide bond
[0081] A single type of complementary functional group may be available on the molecule or substance (which is typical for polysaccharides), or a variety of different complementary functional groups may be available (e.g. amines, thiols, alcohols, phenols), which is typical for proteins. Some selectivity can be obtained by carefully controlling the reaction conditions. Selectivity of conjugation is optimized by appropriate choice of reactive group Rx in light of the available complementary functional group(s) Fx. In instances where the marker synthon to be conjugated does not contain available complementary functional group(s) F*, it can be modified to contain such groups using any of a variety of standard techniques. The marker synthon may also be synthesized or modified to contain a unique complementary functional group(s) Fx, at a specific location for site-specific conjugation to an artificial lymphocyte substrate. See, e.g., U.S. Pat. No. 7,662,644.
[0082] In a particularly preferred embodiment, one of reactive group R* or
complementary functional group F* is a carboxylic acid (or a salt thereof) or an activated ester, for example, a N-hydroxysuccinimidyl (NHS) ester, and the other is an amine, preferably a primary amine. The NHS ester may be conveniently obtained by reacting marker synthon including a carboxylic acid reactive group Rx with N-hydroxysuccinimide
in the presence of an activating agent (e.g., dicyclohexylcarbodiimide (DCC) or 1-Ethyl- 3-(3-dimethylaminopropyl)carbodiimide (EDAC)) according to known methods. Such complementary groups are particularly useful for the covalent coupling of amine- terminated proteins, DNA, or other molecules to carboxyl-functionalized artificial lymphocyte substrates.
[0083] For a further discussion of the various reactive groups R* and respective complementary functional groups F* that can be conveniently used to covalently conjugate the markers to a variety of different types of artificial lymphocytes cell substrate 12, as well as reaction conditions under which the conjugation reactions can be carried out, see Haugland, 1996, Molecular Probes Handbook of Fluorescent Probes and Research Chemicals, Molecular Probes, Inc.; Brinkley, 1992, Bioconjugate Chem. 3:2 and Garman, 1997, Non-Radioactive Labelling: A Practical Approach, Academic Press, London, as well as the references cited in all of the above. Protocols for coupling the various reactive groups Rx and respective complementary functional groups ¥x are available and can readily be adapted to any protein of interest, such as a particular polypeptide (e.g., CD56 or CD16) or antibody, to nucleic acids, or to other biological molecules. See U.S. Pat. No. 5,194,300, Basinska et al, 1991, J. Biomater. Sci. Polym Ed 3(2): 1 15-125, Borque et al., 1994, J. Clin. Immunoassay 17 : 160- 165, Brinkley, 1992, Bioconjugate Chem 3 :2-13, Cantarero et al., 1980, Anal. Biochem 105 :375-382, Derango et al, 1996, J. Immunoassay 17(2) : 145- 153, Gregorious et al, 2001, Anal.
Biochem 299(1) :84-91, Nustad et al, 1982, Agents Actions Suppl. 9 -.207-212, Grabarek et al., 1990, Anal. Biochem 185(1) : 131-135, Bang's Laboratories, Inc., TechNote 204, Adsorption to Microspheres, revision 002, (April 2008), TechNote 205, Covalent coupling, revision 004 (April 2008) and Technote 101, revision 007 (April 2008) (for protocols and crosslinking agents suitable for peptides); Andreadis et al., 2000, Nucl. Acids Res. 28(2):e5, Armstrong et al., 2000, Cytometry 20 : 102-108, Peterson et al., 2001, Nucl. Acids Res. 29(24):5163-5168, Steel et al, 2000, Biophys J. 79:975-981, Jones, 2001, IVD Technology, Nov Dec:39 (for protocols and crosslinking agents suitable for nucleic acids), the disclosures of each of which are incorporated by reference herein in their entireties. For general protocols and further conjugation techniques, see Hermanson, 2008, Bioconjugate Techniques, 2nd Edition, Academic Press, the contents of which are incorporated herein by reference.
[0084] The artificial NK cells described herein comprise CD56 and CD 16 markers that are linked to an artificial lymphocyte substrate in molar ratios such that under conditions of use, the artificial NK cells generate two distinguishable detectable signals having a defined intensity ratio. The ratio of detectable signal generated by the CD56 marker and the CD 16 marker will depend on a number of factors, including the molar ratio of the detectable moieties used, the properties of the detectable moieties themselves, the properties of binding moieties and/or any binding partners to which the detectable moieties may be coupled, as well as the affinity of any binding partners used for their respective binding moieties. The identity and amount of each marker, including the amount of any component parts (e.g., linkers, binding moieties and/or detectable moiety), used to make the artificial NK cells is chosen to account for each of these factors as described further below and can be further adjusted to achieve the desired ratio of signal intensities.
[0085] The amount of detectable moiety used to achieve a desired signal intensity is determined by a variety of factors including: the number of available binding sites that are present in the binding moiety or partner to which the detectable moiety is being linked, the concentration of binding moiety or partner to be labeled and the extinction
coefficients of the detectable moieties used. When combining two detectable moieties to achieve a certain ratio of signal intensity, the amount of each detectable moiety is adjusted to take into account differences in molar extinction coefficients, sensitivity to quenching, and any differences in binding affinity to binding moieties or binding partners. Differences in signal generation between two different detectable moieties due to difference in extinction coefficient can be accounted for by adjusting the starting amounts of each fluorescent dye moiety used. Differences in signal generation between two different fluorescent dye moieties due to quenching can be assessed experimentally using known amounts of each fluorescent dye, and the amount of each fluorescent dye can be adjusted until the desired ratio of signals is achieved.
[0086] Typically, the number of fluorescent dye molecules is chosen to provide a molar ratio of fluorescent dye molecules per CD56 or per CD16 marker to achieve a desired signal intensity for each marker. The ratio of fluorescent dye molecules per CD56 or per CD 16 marker may be as low as 1:1, particularly when the fluorescent dye molecule has a
high extinction coefficient and the marker is small (e.g., molecular weight <10,000 daltons). In other situations, where the marker is large (e.g., molecular weight >69,000 daltons), a larger number of fluorescent dye molecules (e.g., from 1-15 or more) per marker are capable of being conjugated thereto. Markers of intermediate size would expectedly be capable of conjugating intermediate numbers of detectable moieties per marker. A working ratio of fluorescent dye molecules for each marker ranges from about 1 :1 to about 15: 1.
[0087] Second, detectable moieties can be differentially sensitive to their
microenvironment and the signal generated can be affected to different extents by factors such as pH, local concentration of the detectable moiety, whether the detectable moiety is in solution or bound to a particle. Differences in signal generation between two different detectable moieties due to differential affect of the microenvironment can be assessed experimentally using known amounts of each detectable moiety, and the amount of each detectable moiety can be adjusted until the desired ratio of signals is achieved.
[0088] In addition to factors affecting the detectable moieties themselves, the signals detected from an artificial NK cell will also depend on the efficiency of the coupling reaction(s) between the artificial lymphocyte substrate and the CD56 and CD 16 markers, including any optional linkers, and, in the case of indirect CD56 and CD16 markers, on the binding affinity of binding partners to their respective binding moieties. As provided herein, the efficiency of the coupling is measured by the amount of CD56 and CD 16 marker in the starting materials relative to the amount of CD56 and CD 16 marker in the finished artificial NK cells. For embodiments using direct markers, the measurement can be based on the amount of detectable signal in the starting materials versus the finished artificial NK cells.
[0089] For embodiments using indirect markers, the intensity of the detectable signals generated will depend in part on the amount of marker synthon bound to the artificial lymphocyte substrate, and the amount of detectable signal bound to each binding partner, as well as the affinity constants of each binding partner for its target binding moiety.
[0090] Each of the parameters - signal intensity, efficiency of coupling, binding affinity, etc. - can be measured and optimized, according to methods known to the skilled artisan, to yield an appropriate amount of detection signal for the CD56 and CD16 markers in a
desired ratio of CD56:CD16 signal intensity. For the artificial NK cells provided herein, it is not necessary to achieve a specific absolute amount of fluorescence signal intensity for each of the CD56 and CD 16 markers, only a specified ratio of CD56:CD16 signal intensities. A starting molar amount of each marker that yields an appropriate ratio of signal intensities once coupled to the artificial lymphocyte, is adequate. An exemplary process for preparing an artificial NK cell with a suitable amount of each marker is provided. First, the artificial lymphocyte substrate, and CD56 and CD 16 markers, are chosen. The amount of marker needed to form a monolayer on the particle of choice is calculated, based on the diameter and density of the artificial lymphocyte substrate, the molecular weight of the marker to be linked to the particle, and the binding capacity of the particle for the marker, according to equations and parameters known in the art and typically provided with any commercial reagents for preparing conjugated particles. See, e.g., Example 3; see also, Bang's Laboratories TechNote 205, supra, for an example of how to calculate the amount of protein needed to form a monolayer on a particle of a given size.
[0091] Having established the total amount of marker that can be linked to an artificial lymphocyte substrate, the skilled artisan can then determine the relative amounts of each of the two markers needed to provide the desired ratio of detectable signal, taking into account the factors listed above affecting the intensity of the detectable signals. For CD56 and CD 16 direct markers that include a detectable moiety, the markers can be combined in molar amounts in the ratio of signal intensity to be achieved, as determined by reference to standards of known signal intensity for each of the detectable moieties. For example, a CD56 marker bearing a first fluorescent dye capable of generating a first fluorescence signal and a CD 16 marker bearing a second fluorescent dye capable of generating a second fluorescence signal can be combined in amounts approximating the desired ratio of first fluorescent signal intensity to second fluorescent signal intensity, see, e.g., Examples 3 and 4, below. The markers are combined with artificial lymphocyte substrates and a coupling reaction is carried out. The artificial cells comprising the CD56 and CD16 markers are then examined to determine the resulting ratio of CD56:CD16 signal intensities on the artificial NK cells. Based on the observed ratio of signal intensities, the starting amounts of each marker can be increased or decreased to optimize the ratio of detectable signals.
[0092] In a specific example, an artificial CD56bright NK cell is made using a
commercially available kit, for example PolyLink Protein Coupling Kit for carboxylated microspheres from Bangs Laboratories (Product no. PL01N, Bangs Laboratories, Inc.). The artificial CD56bright NK cells are made using carboxyl-functionalized
poly(styrene/divinylbenzene) beads approximately 7 microns in diameter (e.g., Product no. PC06N-6319 or -5405, Bangs Laboratories) as artificial lymphocyte substrates, a AF488-labeled anti-CD56 antibody (e.g., Product no. 561905, BD Biosciences) as a direct CD56 marker, and an AF647-labeled anti-CD16 antibody as a direct CD16 marker (e.g., Product no. 557710, BD Biosciences). The anti-CD56 and anti-CD16 antibodies are mixed in a molar ratio of 100:1 of the anti-CD56 to anti-CD 16 antibodies. The amount of the mixture of anti-CD56 and anti-CD16 antibodies needed in the conjugation reaction can be based on information available from the manufacturer and as described in Cantarero et al, 1980, Anal. Biochem 105 :375-382, generally between 200 and 500 μg. The anti-CD56 and anti-CD 16 antibodies are conjugated to the carboxylated
microspheres by way of primary amines. Conjugation is then carried out according to the kit instructions. Specifically, 12.5 mg of polystyrene beads is centrifuged for 5 to 10 minutes at approximately 500-1000XG, resuspended in 0.4 mL coupling buffer (50 nM MES, pH 5.2, 0.05% Proclin® 300), and then re-centrifuged. The polystyrene beads are then resuspending in 0.17 mL of the coupling buffer. Immediately before the conjugation reaction, 200 mg/ml carboiimide (EDAC) solution is prepared by dissolving 10 mg EDAC in 50 μΐ coupling buffer. 20 μΐ of the EDAC solution is added to the polystyrene bead suspension, and mixed gently. The anti-CD56 and anti-CD16 antibody mixture is added and mixed gently. The suspension is then incubated for 30 to 60 minutes, followed by centrifugation for 10 minutes at 500-1000XG. Supernatant is saved and can be used to determine the amount of protein bound to the beads, by comparing the protein concentration of the starting mixture to that of the supernatant. The resulting conjugated polystyrene beads can be stored in appropriate buffer for use (e.g., 10 mM Tris pH 8.0 0.05% BSA, 0.05% Proclin® 300).
[0093] In another specific example, an artificial CD56bnght NK cell is made in which the CD56 marker is a polypeptide comprising an amino acid sequence corresponding to the extracellular domain of CD56 and the CD 16 marker is a polypeptide comprising an amino acid sequence corresponding to the extracellular domain of CD16 are conjugated
to carboxylated microspheres from Bangs Laboratories (Product no. PL01N, Bangs Laboratories, Inc.). A CD56 polypeptide having the amino acid sequence corresponding to amino acids 1 to 689 of SEQ ID NO: l and a CD16 polypeptide having the amino acid sequence corresponding to amino acids 1 to 192 of SEQ ID NO:2 are synthesized and purified, according to standard techniques, and modified to include a unique functional group (e.g., aminooxy-containing amino acids) at the amino terminus, for site-specific conjugation, as described in U.S. Pat. No. 7,662,644. A solution containing the CD56 polypeptide and CD 16 polypeptide in a molar ratio of 100: 1 is prepared and conjugated by way of aminooxy-containing amino acids to NHS-functionalized carboxylated microspheres as described above. Polystyrene beads bearing CD56 and CD 16 markers can then be incubated with an AF488-labeled anti-CD56 antibody (e.g., Product no. 561905, BD Biosciences) and an AF647-labeled anti-CD16 antibody (e.g., Product no. 557710, BD Biosciences) to generate detectable signals.
[0094] To verify that the artificial NK cells described above perform as artificial CD56 nght NK cells, the labeled, conjugated polystyrene beads can be compared to artificial CD56bright NK cells of Example 3 below or CD56bright NK cells by flow cytometry. Adjustment can be made in the starting ratios of each marker where optimization is desired in the ratio of signal intensities. To optimize the overall signal intensity from the beads, the amount of protein added to the beads can be adjusted, as can the incubation time.
5.3. Kits
[0095] The artificial NK cells described herein are useful as reference standards in a variety of fluorescence-based cell sorting assays, including FACS flow cytometry, for CD56bright NK cells_ These NK ceiis can be provided in kits, with suitable reagents and instructions for use.
[0096] In its simplest form, a kit includes a vial or other suitable container of artificial CD56 bnght cellSj optiona]]y a vial or other suita ie container of artificial CD56dim NK cells, one or more vials of reagents to prepare the artificial NK cells for use on a cell sorting apparatus (e.g., one or more diluents or buffers), and instructions for use.
[0097] For some applications, it may be desirable to have artificial NK cells labeled with the same reagents used to detect human CD56bnght or CD56dim NK cells in test samples,
e.g., blood samples. For these applications, kits can include, in addition to reagents and instructions for use on a cell sorting apparatus:
(a) an artificial lymphocyte substrate linked to a CD56 marker comprising a first binding moiety and a CD 16 marker comprising a second binding moiety, wherein the CD56 and CD16 markers are conjugated to the artificial lymphocyte substrate in a ratio appropriate for an artificial CD56 nght NK cell;
(b) optionally, in a separate vial, an artificial lymphocyte substrate linked to a CD56 marker comprising a third binding moiety and a CD 16 marker comprising a fourth binding moiety, wherein the CD5 and CD 16 markers are conjugated to the artificial lymphocyte substrate in a ratio appropriate for an artificial CD56dim NK cell;
(c) a first binding partner for the first binding moiety and a second binding partner for the second binding moiety, wherein the first binding partner is linked to a first detectable moiety and the second binding partner is linked to a second detectable moiety, the first and second detectable moieties generating first and second detectable signals that are distinguishable or resolvable from each other;
(d) optionally, a third binding partner for a third binding moiety and a fourth binding partner for the fourth binding moiety, wherein the third binding partner is linked to a third detectable moiety and the fourth binding partner is linked to a fourth detectable moiety, the third and fourth detectable moieties generating third and fourth detectable signals that are distinguishable or resolvable from each other; and
(e) instructions and reagents for binding the artificial CD56brlght NK cell to the first and second binding partners, and optionally, instructions and reagents for binding the artificial CD56dim NK cell to the third and fourth binding partners.
[0098] Exemplary embodiments for artificial lymphocyte substrates, binding moieties, and binding partners labeled with detectable moities, e.g., fluorescent dye moieties, are further illustrated in FIG. 7 A.
[0099] In some kits, reagents suitable for "sandwich-type" assays are provided, wherein the detectable moieties are linked to further binding partners which bind specifically to unlabeled first, second, and optionally third and fourth binding partners. Such kits can include:
(a) an artificial lymphocyte substrate linked to a CD56 marker comprising a first binding moiety and a CD 16 marker comprising a second binding moiety, wherein the CD56 and CD 16 markers are conjugated to the artificial lymphocyte substrate in a ratio appropriate for an artificial CD56bnght NK cell;
(b) optionally, in a separate vial, an artificial lymphocyte substrate linked to a CD56 marker comprising a third binding moiety and a CD 16 marker comprising a fourth binding moiety, wherein the CD56 and CD 16 markers are conjugated to the artificial lymphocyte substrate in a ratio appropriate for an artificial CD56dim NK cell;
(c) an unlabeled first binding partner for the first binding moiety and an unlabeled second binding partner for the second binding moiety;
(d) a further first binding partner for the first binding partner and a further second binding partner for the second binding partner, wherein the further first binding partner is linked to a first detectable moiety and the further second binding partner is linked to a second detectable moiety, the first and second detectable moieties generating first and second detectable signals that are distinguishable or resolvable from each other;
(e) optionally, an unlabeled third binding partner for a third binding moiety and an unlabeled fourth binding partner for the fourth binding moiety;
(f a further third binding partner for the third binding partner and a further fourth binding partner for the fourth binding partner, wherein the further third binding partner is linked to a third detectable moiety and the further fourth binding partner is linked to a fourth detectable moiety, the third and fourth detectable moieties generating third and fourth detectable signals that are distinguishable or resolvable from each other; and
(g) instructions and reagents for binding the artificial CD56bright NK cell to the first and second binding partners, and for binding the first and second binding partners to the further first and further second binding partners, and optionally, instructions and reagents for binding the artificial CD56dim NK cell third to the third and fourth binding partners, and for binding the third and fourth binding partners to the further third and further fourth binding partners.
[0100] Exemplary embodiments for artificial lymphocyte substrates, binding moieties, unlabeled binding partners, and further binding partners labeled with detectable moities, e.g., fluorescent dye moieties, are further illustrated in FIG. 7B.
[0101] In specific embodiments, the CD56 and CD 16 markers comprise CD56 and CD 16 polypeptides, respectively. The first and second binding partners, whether labeled or unlabeled, in such kits can comprise molecules that bind to CD56 and CD16 polypeptides and therefore can be used to label human NK cells in test samples as well as the artificial NK cells.
5.4. Methods of using artificial NK cells
[0102] The artificial NK cells described herein are useful to calibrate flow cytometry instruments, optimize gating parameters of flow cytometry instruments, identify, detect, collect, measure and/or count or quantify CD56bnght NK cells found in samples (e.g., blood). Samples can be from subjects who suffer from a disease or condition affecting the number or presence of CD56bright NK cells, or from subjects receiving a treatment regimen, the effect of which can be monitored based on the number or presence of CD56bnght NK cells. In some cases, it is desirable to measure or quantify the number or presence of CD56bnght NK cells in samples from patients receiving therapies that block ILR2 signaling.
[0103] In sorting cell populations to identify specific subsets, such as CD56bnght NK cells, it is desirable to establish the best gate or region that can be set that encompasses the target cells within the sample. The gate will exclude cells which are not within the selected region and will only retain information generated by cells which are within the gate. Typically, this is done by means of a "curser" whose movement is tied into the data retention device (i.e., computer) and controlled manually. Such an approach for identifying CD56bnght NK cells is time consuming and error-prone as it depends on visual inspection. Additionally, this approach can result in substantial variation across experiments conducted on different days by different operators. The artificial CD56bnght and CD56dim NK cells described herein provide a tool for reliably and reproducibly identifying the region within a test sample that contains the cells of interest, i.e.,
CD56bri ht NK cells.
[0104] Generally, the methods provided herein comprise running a first reference sample containing artificial CD56bnght NK cells, through a cell sorting apparatus, determining a location of the artificial CD56bnght NK cells in the reference sample, running a test sample containing CD56bnght NK cells through the cell sorting apparatus, and finding a region in the test sample corresponding to the location of the artificial CD56bnght NK cells in the reference sample, wherein a majority of the cells in the region of the test sample are CD56bright NK cells.
[0105] As provided herein, to further optimize the CD56bnght NK cell-containing region within the test sample, a second reference standard may be used: artificial CD56dim NK cells. When artificial CD56dim NK cells are used, a second reference sample may be run in advance of running a test sample, and a location of the artificial CD56dim NK cells in the second reference sample can be determined, which is then used to limit the CD56bright NK cell-containing region in the test sample. Optionally, artificial CD56bnght and CD56dim NK cells may be combined in a single reference sample, provided that artificial CD56bnght and CD56dim NK cell popuiations have different fluorophores, so as to be able to individually detect the fluorescence signals of each first and second fluorophore of the two types of artificial NK cells in the reference sample.
[0106] In a preferred embodiment, the CD56 marker and CD 16 marker of the artificial NK cells are CD56 and CD 16, respectively. Accordingly, the artificial NK cells and the test sample(s) can be labeled with the same antibodies, anti-CD56 and CD 16 antibodies, linked to two distinct fluorophores, before the reference and test samples are assayed on a flow cytometer.
[0107] Having established a region in a test sample wherein the majority of the cells are CD56bright NK cellj the methods
may further comprise a step of counting the CD56bnght NK cells and/or a step of collecting CD56bright NK cells for further study.
[0108] The artificial CD56bright and CD56dim NK cells described herein can be used as reference standards in any suitably calibrated and aligned fluorescence-based cell sorting assay, including flow cytometry assays, conducted on any cell sorting apparatus, including any flow cytometer apparatus (e.g., a fluorescence-activated cell sorting or FACS apparatus). The artificial CD56bright and CD56dim NK cells described herein can be used in many types of flow cytometry devices which measure light scatter, particle or cell
volume, fluorescence or any other optical parameters for the identification or
quantification of CD56bnght NK cell subpopulations in a sample. Flow cytometers are commercially available from a variety of manufacturers, including BD Biosciences, Inc., EMD Millipore, Miltenyi Biotec, Accuri Cytometers, Inc., and Beckman Coulter. For a particular example of the elements of a suitable flow cytometry apparatus, see
FACScan™ flow cytometer. The FACScan™ flow cytometer analyzes cell populations on the basis of light scatter and fluorescence in a wide variety of research and laboratory applications. Other details of a cell analysis and sorting apparatus useful in conjunction with the present invention are described in U.S. Patent No. 3,826,364, the disclosure of which is incorporated by reference herein. Other FACS flow cytometers include: BD FACSVerse™, FACSAria™ II, or FACSCanto™ by BD Biosciences, Inc., MAC SQUANT Analyzer by Miltenyi Biotec, and COULTER® EPICS XL and XL-MCL by Beckman Coulter.
6. EXAMPLES
Example 1 : Artificial lymphocyte cells migrate like human lymphocytes
as assayed by fluorescence-activated cell sorting (FACS)
[0109] This example shows an artificial lymphocyte substrate that mimics the characteristics of a human lymphocyte in a flow cytometer.
1.1 Methods & Materials
[0110] Blood samples. Whole blood was collected from volunteers, from whom informed consent was obtained. Whole blood was then anti-coagulated and stained with anti-CD56, anti-CD 16, and anti-CD3 antibodies as follows. Approximately 0.2 mL of anti-coagulated blood was placed in 10 mL plastic tubes. Two milliliters of IX red blood cell lysis buffer (from a stock solution of 10X red blood cell lysis buffer (Catalog No. 349202, BD Biosciences Inc.)) was added to each tube and incubated 10 minutes at room temperature (RT). Immune cells were then pelleted by centrifugation at 400X gravity for 5 minutes and supematants removed. Samples were washed and pelleted twice more with 3 mL of PBS. Approximately 0.1 mL of liquid was left in each tube after removing the supematants. The cell pellets were then resuspended. Cells were then incubated for 15 minutes at RT with appropriate amounts of (1) anti-CD56 antibody (N-CAM clone) conjugated to Alexa Fluor® AF488 (Catalog No. 561905, BD Biosciences Inc.), (2) anti-
CD16 antibody (3G8 clone) conjugated to Alexa Fluor® AF647 (Catalog No. 557710, BD Biosciences Inc.), and (3) anti-CD3 antibody conjugated to allophycocyanin APC-H7 (Catalog No. 641397, BD Biosciences Inc.) at appropriate concentrations. Samples were then suspended in PBS for analysis using a flow cytometer FACSCanto™ model and FACSDiva™ software from BD Biosciences Inc, according to manufacturer's specification.
[0111] Artificial lymphocytes. Spherical polymeric beads (Reagent no. 01910, Bang's Laboratories, Inc., Fishers, IN) with a mean diameter of 7.65 μπι were used to mimic lymphocytes. The polymeric beads were added to 0.2 mL PBS and prepared for use with a FACSCanto™ flow cytometer and FACSDiva™ software from BD Biosciences Inc, according to operating instructions for the equipment.
[0112] Flow cytometry. The flow cytometer was calibrated using Cytometer Setup and Tracking (CS&T) beads (Catalog No. 641319, BD Biosciences Inc.), followed by manual adjustment of instrument voltages. Optimum compensation settings were established to account for spectral overlap in signals from fluorochromes, using whole blood samples stained with fluorescently labeled antibodies as described above. For each sample, 10,000 to 100,000 gated events were acquired. Lymphocyte cells were identified based on characteristic forward scatter (FSC) and side scatter patterns (SSC). Artificial lymphocytes were found based on having the same pattern of forward and side scatter as lymphocytes from the whole blood samples.
1.2 Results
[0113] FIG. 1 provides a representative flow cytometry scatter or dot plot of forward light scatter (FSC) versus side light scatter (SSC), which is used to identify lymphocyte cells. Lymphocyte cells appear in within the hexagonal area based on their characteristic light scattering properties. FIG. 2 provides a representative scatter plot for artificial lymphocytes. Individual artificial lymphocytes appear within the hexagonal area based on their characteristic light scattering properties. As shown in FIG.s 1-2, the artificial lymphocytes closely mimic the profile of actual lymphocytes, which include NK cells, in flow cytometry experiments.
Example 2 : Detection of CD56bright NK cells by FAC S
[0114] This example shows the standard flow cytometry profile of human CD56bright NK cells labeled with anti-CD56 and anti-CD 16 monoclonal antibodies.
2.1 Materials & Methods
[0115] Whole blood was prepared, including immunostaining with anti-CD3, anti-CD56, and anti-CD 16 antibodies, and analyzed on a flow cytometer as described in Example 1 above. Light scatter was used to identify likely lymphocyte cells (see FIG. 1) and negative selection with APC-H7- labeled anti-CD3 antibodies (Catalog No. 641397, BD Biosciences Inc.) was used to enrich for NK and B cells. Analysis gates for measuring signal intensity of the fluorescent signals from AF488-labeled anti-CD56 and AF647- labeled anti-CD 16 were set around the region identified in Example 1 as marking the location of lymphocytes and artificial lymphocytes by FSC and SSC. Fluorescence signals from Alexa Fluor® AF488, detectable at the same wavelength as that used for FITC, and Alexa Fluor® AF647, detectable at the same wavelength used for APC, were then recorded and plotted on scatter plots.
2.2 Results
[0116] FIG. 3 provides a scatter plot of a blood sample containing rare CD56bright NK cells, stained with Alexa Fluor® AF488-labeled anti-CD56 and Alexa Fluor® AF647- labeled anti-CD 16 specific antibodies. Data points in the region identified with an inner, small rectangle represent those cells with very highest levels of CD56 protein and very low or negligible levels of CD 16 protein, the CD56bnght NK cells. The densely populated lower left region represents immune cells that are likely B cells, based on the lack of staining with antibodies to CD3, CD16 and CD56 proteins. Cells in the upper right and center region are likely NK cells that are not of the CD56bnght NK phenotype.
Example 3 : Detection of Artificial CD56bright NK cells by FACS
[0117] This example demonstrates the synthesis of an exemplary embodiment of artificial CD56 bright NK cells having direct CD56 and CD16 markers. Artificial lymphocyte substrates conjugated to fluorescently labeled anti-CD56 and anti-CD 16 antibodies in a molar ratio of about 100:1 generated a fluorescence signal profile approximating that seen
for typical CD56bright NK cells labeled with anti-CD56 and anti-CD16 antibodies in a flow cytometry assay.
3.1 Materials & Methods
[0118] Artificial CD56bright NK cells. Anti-CD56 antibody conjugated to Alexa Fluor® AF488 (N-CAM clone, Catalog No. 561905, BD Biosciences Inc.) and anti-CD 16 antibody conjugated to Alexa Fluor® AF647 (3G8 clone, Catalog No. 557710, BD Biosciences Inc.) were conjugated directly to artificial lymphocytes in an input ratio of 99 parts anti-CD56 antibody to 1 part anti-CD 16 antibody to generate a first type of multi- specific, multi-fiuorochrome (artificial CD56bnght NK cell). The amount of antibody conjugated to the artificial lymphocytes was determined based on the antibody binding capacity ("ABC") of the unconjugated artificial lymphocyte substrate used to make the artificial NK cells by reference to Quantum Simply Cellular® reference microspheres (see Product Data Sheet 814, Rev. 001, Bangs Laboratories, Inc.). Artificial CD56bright NK cells were prepared for flow cytometry analysis as described in Example 1 above. Flow cytometry was performed as described in Example 2 above.
3.2 Results
[0119] Light scatter properties were used to identify and select the artificial CD56bnght NK cells by reference to lymphocyte cell location, as shown in FIG. 2. Subsequently, fluorescence signals were used to locate cells with high levels of Alexa Fluor® AF488 signal and low or negligible levels of Alexa Fluor® AF647, providing the location of the artificial CD56bright NK cells. FIG. 4 provides a representative scatter plot for the artificial CD56bnght NK cells. The inner rectangle corresponds to the anticipated region where CD56bnght NK cells would be expected to appear if a sample, such as blood, was stained with a mixture of fluorescently conjugated antibodies as described in Example 2 above and shown in FIG. 3. As can be seen, the highlighted region is densely populated with data points, demonstrating the ability of these artificial CD56bright NK cells to identify the location of CD56bnght NK cells in a blood sample analyzed by flow cytometry.
[0120] By including in a cytometry procedure the use of an artificial NK cell, such as that described herein, bound with appropriately defined ratios of two different fluorescent
molecules, the operator of the instrument now knows exactly where the target immune population should appear on their instrument and the operator may also be able to confirm, after a sample run, that the instrument was performing satisfactorily at the time of the sample analysis. Using such an artificial NK cell to define an x and y coordinate relative to the anticipated location of the target cell population of interest can also facilitate the application of software increasingly used to automate the analysis portion of cytometry procedures when attempting to minimize the potential for human error.
Example 4: Detection of Artificial CD56dim NK Cells by FACS
[0121] This example demonstrates the synthesis of an exemplary embodiment of artificial CD56dim NK cells having direct CD56 and CD16 markers. Artificial lymphocyte substrates conjugated to fluorescently labeled anti-CD56 and anti-CD16 antibodies in a molar ratio of about 10:1 generated a fluorescence signal profile approximating that seen for typical CD56dim NK cells labeled with anti-CD56 and anti-CD16 antibodies that is distinguishable from CD56bnght NK cells in a flow cytometry assay.
4.1 Materials & Methods
[0122] Artificial CD56dim NK cells. Artificial CD56dim NK cells with a 10:1 ratio of anti-CD56 antibody to anti-CD 16 antibody were made as described above for artificial CD56bright NK cells, except for the ratio of anti-CD56 antibody to anti-CD16 antibody used. Specifically, starting amounts of the two antibodies was 9 parts fluorescently labeled anti-CD56 antibody to 1 part fluorescently labeled anti-CD16 antibody.
4.2 Results
[0123] FIG. 5 provides a representative flow cytometry scatter plot of artificial CD56dim
NK cells. The densely populated region that appears up left corresponds to the anticipated region where NK cells that did not have the characteristic phenotype of
CD56bright NK ceUs would be expected t0 appear if a tissue sample, such as blood, was stained with a mixture of fluorescently conjugated antibodies (for example, see FIG. 3). The region highlighted by the inner rectangle denotes the area where human CD56bright NK cells are expected to appear. As can be seen, this region is lacking data points, demonstrating the utility of these artificial CD56dm NK cells in identifying the location of CD56bnght NK cells in blood sample analyzed by flow cytometry.
[0124] By including in a cytometry procedure the use of an artificial NK cell bound with appropriately defined concentrations of two or more matching fluorescent molecules, as shown in FIG. 5, the operator of the instrument now has a second x and y coordinate to aid in identifying exactly where the target immune population should be anticipated to appear on their instrument and the operator may also be able to confirm that the instrument was performing satisfactorily at the time of the specimen analysis. Using such an artificial NK cell to define a second x and y coordinate relative to the anticipated location of the target cell population of interest can also facilitate the application of software increasingly used to automate the analysis portion of cytometry procedures when attempting to minimize the potential for human error.
[0125] All publications, patents, patent applications and other documents cited in this application are hereby incorporated by reference in their entireties for all purposes to the same extent as if each individual publication, patent, patent application or other document were individually indicated to be incorporated by reference for all purposes.
[0126] While various specific embodiments have been illustrated and described, it will be appreciated that various changes can be made without departing from the spirit and scope of the invention(s).