Dynamic quantification of antigen molecules with flow cytometry.

Traditional methods for estimating the number of expressed molecules, based on the detection of target antigens bound with fluorescently labeled antibodies, assume that the antigen-antibody reaction reaches equilibrium. A calibration procedure is used to convert the intensity of the fluorescence signal to the number of target molecules. Along with the different limitations of every calibration system, this substantially limits the applicability of the traditional approaches especially in the case of low affinity antibodies. We address this problem here with studies in which we demonstrate a new approach to the antigen molecule quantification problem. Instead of using a static calibration system, we analyzed mean fluorescence values over time by flow cytometry during antibody-antigen binding. Experimental data obtained with an LSRII cytometer were fitted by a diffusion-reaction mathematical model using the Levenberg-Marquardt nonlinear least squares curve-fitting algorithm in order to obtain the number of target antigen molecules per cell. Results were compared with the Quanti-BRITE calibration system. We conclude that, instead of using experiment-specific calibration, the value of the binding rate constant for each particular antibody-antigen reaction can be used to quantify antigen molecules with flow cytometry. The radius of CD8 antibody molecule binding site was found, that allows recalculating the binding rate constant for other conditions (different sizes of reagent molecules, fluorescent label, medium viscosity and temperature). This approach is independent of specially prepared calibration beads, antibody reagents and the specific dye and can be applied to both low and high affinity antibodies, under both saturating and non-saturating binding conditions. The method was demonstrated on a human blood sample dataset investigating CD8α antigen on T cells in stable binding conditions.

Dynamic quantification of antigen molecules with flow cytometry.

Traditional methods for estimating the number of expressed molecules, based on the detection of target antigens bound with fluorescently labeled antibodies, assume that the antigen-antibody reaction reaches equilibrium. A calibration procedure is used to convert the intensity of the fluorescence signal to the number of target molecules. Along with the different limitations of every calibration system, this substantially limits the applicability of the traditional approaches especially in the case of low affinity antibodies. We address this problem here with studies in which we demonstrate a new approach to the antigen molecule quantification problem. Instead of using a static calibration system, we analyzed mean fluorescence values over time by flow cytometry during antibody-antigen binding. Experimental data obtained with an LSRII cytometer were fitted by a diffusion-reaction mathematical model using the Levenberg-Marquardt nonlinear least squares curve-fitting algorithm in order to obtain the number of target antigen molecules per cell. Results were compared with the Quanti-BRITE calibration system. We conclude that, instead of using experiment-specific calibration, the value of the binding rate constant for each particular antibody-antigen reaction can be used to quantify antigen molecules with flow cytometry. The radius of CD8 antibody molecule binding site was found, that allows recalculating the binding rate constant for other conditions (different sizes of reagent molecules, fluorescent label, medium viscosity and temperature). This approach is independent of specially prepared calibration beads, antibody reagents and the specific dye and can be applied to both low and high affinity antibodies, under both saturating and non-saturating binding conditions. The method was demonstrated on a human blood sample dataset investigating CD8α antigen on T cells in stable binding conditions.

Data management.

This unit presents a broad view of flow cytometry data management, emphasizing ways to create and manage data so as to maximize their immediate and long-term usefulness. The cost in time, money, and effort to create data files in the first place argues for the importance of making full use of files and keeping them from getting lost or becoming uninterpretable. As flow cytometry measurements become increasingly complex, researchers have more and more reasons to reexamine old data. Careful data management is clearly necessary to ensure that stored data contain enough information about an experiment to enable analysis and interpretation by others even years later.

Nine color eleven parameter immunophenotyping using three laser flow cytometry.

BACKGROUND: This study describes a three laser flow cytometer, reagents, and software used to simultaneously evaluate nine distinct fluorescent parameters on one cell sample. We compare the quality of data obtained with (1) full software compensation and (2) the use of partial spectral compensation of selected pairs of parameters in analog hardware, in combination with final software compensation. An application characterizing low frequency murine B cell subpopulations is given. METHODS: The fluorochromes used are: fluorescein (FITC), phycoerythrin (PE), Cy5PE and Cy7PE, excited at 488 nm by an argon laser; Texas Red (TR), allophycocyanin (APC), and Cy7APC excited at 595 nm by a pumped dye laser; and cascade blue (CB) and cascade yellow (CY) excited at 407 nm by a violet-enhanced krypton laser. Custom additions to commercial electronics and an extended optical bench allow the measurement of these nine parameters plus forward and side scatter light signals. RESULTS: We find the use of partial analog compensation reduces the variation in the background staining levels introduced by the compensation process. Novel B cell populations with frequencies below 1% are characterized. CONCLUSIONS: Nine color flow cytometry is capable of providing measurements with high information content. The choice of reagent-dye combinations and the ability to compensate in multi-parameter measurement space are crucial to obtaining satisfactory results.

Simultaneous fluorescence-activated cell sorter analysis of two distinct transcriptional elements within a single cell using engineered green fluorescent proteins.

Green fluorescent protein (GFP) is widely used as a reporter gene in both prokaryotes and eukaryotes. However, the fluorescence levels of wild-type GFP (wtGFP) are not bright enough for fluorescence-activated cell sorting or flow cytometry. Several GFP variants were generated that are brighter or have altered excitation spectra when expressed in prokaryotic cells. We engineered two GFP genes with different combinations of these mutations, GFP(S65T,V163A) termed GFP-Bex1, and GFP(S202F,T203I,V163A) termed GFP-Vex1. Both show enhanced brightness and improved signal-to-noise ratios when expressed in mammalian cells and appropriately excited, compared with wtGFP. Each mutant retains only one of the two excitation peaks of the wild-type protein. GFP-Bex1 excites at 488 nm (blue) and GFP-Vex1 excites at 406 nm (violet), both of which are available laser lines. Excitation at these wavelengths allows for the independent analyses of these mutants by fluorescence-activated cell sorting, permitting simultaneous, quantitative detection of expression from two different genes within single mammalian cells.

Cy7PE and Cy7APC: bright new probes for immunofluorescence.

We demonstrate the utility of indotricarbocyanine (Cy7) conjugates of the phycobiliproteins phycoerythrin (PE) and allophycocyanin (APC) in flow cytometry. This is the first demonstration of the use of an APC tandem dye for fluorescence measurements. These resonance energy transfer tandem dyes can be excited by the phycobiliprotein-specific excitation wavelengths and fluoresce at wavelengths above 780 nm. The tandem dyes, when conjugated to antibodies, are suitable for flow cytometry and other immunofluorescence applications. These conjugates are easily detectable above the very low autofluorescence in this part of the spectrum. Indeed, the Cy7-conjugated PE tandem (Cy7PE) has a "brightness" (fluorescence signal over cellular autofluorescence) comparable to that of fluorescein, and the Cy7APC tandem has a "brightness" comparable to that of APC. These tandems are also easily distinguished from other commonly used fluorophores, making them suitable for high-order multiparametric analysis. We show an example of six-color immunofluorescence analysis by flow cytometry, simultaneously measuring fluorescences from fluorescein, PE, Cy5PE, Texas red, APC, and Cy7APC.

Heterogeneous calcium flux in peripheral T cell subsets revealed by five-color flow cytometry using log-ratio circuitry.

Calcium flux measurements of different subpopulations of cells by flow cytometry are important in understanding complex interactions in the immune system. This paper discusses the use of the difference of Log signals as a preferred method for obtaining this information simultaneously with other immunofluorescence parameters. We describe simple modifications to a commercial instrument that enables the measurement of calcium flux in addition to three immunofluorescence parameters. Finally, we show an application of this technique to measuring calcium flux of T cell subsets in human blood. We show that different subsets of peripheral CD4 T cells have significantly different capabilities to flux calcium after CD3 stimulation. These differences are related to the functional capacities of the cells within these subsets.

Pattern sorting: a computer-controlled multidimensional sorting method using k-d trees.

Multidimensional binary trees provide a memory efficient and general method for computing sorting decisions in real time for a flow cytometer. Their fundamental advantage over conventional lookup table sorting techniques is that sort criteria in the full N-dimensional data space which cannot be described by projections onto two-dimensional parameter planes can be effectively implemented. This becomes particularly relevant when multidimensional analysis methods such as principal components or clustering are employed. We describe a prototype implementation of this method and point out other possible implementations.

Contribution of a single heavy chain residue to specificity of an anti-digoxin monoclonal antibody.

Two distinct spontaneous variants of the murine anti-digoxin hybridoma 26-10 were isolated by fluorescence-activated cell sorting for reduced affinity of surface antibody for antigen. Nucleotide and partial amino acid sequencing of the variant antibody variable regions revealed that 1 variant had a single amino acid substitution: Lys for Asn at heavy chain position 35. The second variant antibody had 2 heavy chain substitutions: Tyr for Asn at position 35, and Met for Arg at position 38. Mutagenesis experiments confirmed that the position 35 substitutions were solely responsible for the markedly reduced affinity of both variant antibodies. Several mutants with more conservative position 35 substitutions were engineered to ascertain the contribution of Asn 35 to the binding of digoxin to antibody 26-10. Replacement of Asn with Gln reduced affinity for digoxin 10-fold relative to the wild-type antibody, but maintained wild-type fine specificity for cardiac glycoside analogues. All other substitutions (Val, Thr, Leu, Ala, and Asp) reduced affinity by at least 90-fold and caused distinct shifts in fine specificity. The Ala mutant demonstrated greatly increased relative affinities for 16-acetylated haptens and haptens with a saturated lactone. The X-ray crystal structure of the 26-10 Fab in complex with digoxin (Jeffrey PD et al., 1993, Proc Natl Acad Sci USA 90:10310-10314) reveals that the position 35 Asn contacts hapten and forms hydrogen bonds with 2 other contact residues. The reductions in affinity of the position 35 mutants for digoxin are greater than expected based upon the small hapten contact area provided by the wild-type Asn. We therefore performed molecular modeling experiments which suggested that substitution of Gln or Asp can maintain these hydrogen bonds whereas the other substituted side chains cannot. The altered binding of the Asp mutant may be due to the introduction of a negative charge. The similarities in binding of the wild-type and Gln-mutant antibodies, however, suggest that these hydrogen bonds are important for maintaining the architecture of the binding site and therefore the affinity and specificity of this antibody. The Ala mutant eliminates the wild-type hydrogen bonding, and molecular modeling suggests that the reduced side-chain volume also provides space that can accommodate a congener with a 16-acetyl group or saturated lactone, accounting for the altered fine specificity of this antibody.

Altered hapten recognition by two anti-digoxin hybridoma variants due to variable region point mutations.

Two spontaneous variants of the murine anti-digoxin antibody-producing hybridoma cell line 26-10 were isolated by two-color fluorescence-activated cell sorting on the basis of altered hapten binding. The variable region sequences of the antibodies produced by the mutant lines revealed that each contains a single amino acid change in the heavy chain second complementarity determining region. A Tyr to His change at position 50 leads to a 40-fold reduction in affinity for digoxin. A Ser to Phe mutation at position 52 results in a 300-fold reduction in affinity for digoxin. A competition assay involving 33 digoxin analogues was used to examine the specificity of hapten binding of 26-10 and the two mutant antibodies. The position 50 mutant has a distinct specificity change; it exhibits a preference for digoxin congeners containing a hydroxyl group at the steroid 12 position, whereas the 26-10 parent does not. The affinities of all three antibodies for hapten are progressively lowered by substitutions of increasing size at the digoxin steroid D ring 16 position. Although 26-10 binds digoxin and its genin form equally, 12 and 16 steroid position substitutions which lower affinity also confer a preference for a sugar at the steroid 3 position. These results suggest that position 50 contributes to specificity of the antibody and that alterations of the hapten can lead to differences in recognition, possibly through a shift in hapten orientation within the binding site.

Improved FACS-Gal: flow cytometric analysis and sorting of viable eukaryotic cells expressing reporter gene constructs.

The previously reported FACS-Gal assay (Nolan et al., Proc Natl Acad Sci USA 85:2603-2607, 1988) measures E. coli lacZ-encoded beta-galactosidase activity in individual viable eukaryotic cells for a variety of molecular and cellular biological applications. Enzyme activity is measured by flow cytometry, using a fluorogenic substrate, which is hydrolyzed and retained intracellularly. In this system, lacZ serves both as a reporter gene to quantitate gene expression and as a selectable marker for the fluorescence-activated sorting of cells based on their lacZ expression level. This report details the following improvements of the original assay: 1) use of phenylethyl-beta-D-thiogalactoside, a competitive inhibitor, to inhibit beta-galactosidase activity; 2) reduction of false positives by two-color measurements; and 3) inhibition of interfering mammalian beta-galactosidases by the weak base chloroquine. We found an exponential relationship between fluorescence generated by beta-galactosidase in this assay and the intracellular concentration of beta-galactosidase molecules. Finally, we report conditions for optimal loading of the substrate (FDG) and retention of the product, fluorescein. Under these conditions, we found uniform loading of FDG in all cells of a clone in individual experiments. Together, these improvements make FACS-Gal an extremely powerful tool for investigation of gene expression in eukaryotic cells.

Variable region framework differences result in decreased or increased affinity of variant anti-digoxin antibodies.

Rare spontaneous variants of the anti-digoxin antibody-producing hybridoma 40-150 (Ko = 5.4 x 10(9) M-1) were selected for altered antigen binding by two-color fluorescence-activated cell sorting. The parent antibody binds digoxin 890-fold greater than digitoxin. The variant 40-150 A2.4 has reduced affinity for digoxin (Ko = 9.2 x 10(6) M-1) and binds digoxin 33-fold greater than digitoxin. A second-order variant, derived from 40-150 A2.4 (designated 40-150 A2.4 P.10), demonstrated partial regain of digoxin binding (Ko = 4.4 x 10(8) M-1). The altered binding of the variant 40-150 A2.4 was accounted for by a point mutation resulting in substitution of arginine for serine at position 94 in the heavy chain variable region. Antibody 40-150 A2.4 P.10 also contains this arginine but owes its enhanced antigen binding to deletion of two amino acids from the heavy chain amino terminus. This unusual sequence alteration in an immunoglobulin framework region confers increased affinity for antigen.

Isolation of human myoblasts with the fluorescence-activated cell sorter.

We have established procedures for the rapid and efficient purification of human myoblasts using the fluorescence-activated cell sorter. Our approach capitalizes on the specific reaction of monoclonal antibody 5.1H11 with a human muscle cell surface antigen. For each of the five samples analyzed, an enrichment of myoblasts to greater than 99% of the cell population was immediately achieved. Following 3 to 4 weeks of additional growth in vitro, sorted myoblast cultures remained 97% pure. Differentiation of the sorted myoblast cultures, assessed by creatine kinase activity and isozyme content, was comparable to that of pure myoblast cultures obtained by cloning, and was significantly greater than that of mixed fibroblast and myoblast cultures. An average of 10(4) viable myoblasts can be obtained per 0.1 g tissue, each with the potential to undergo approximately 40 cell divisions. Accordingly, if only two-thirds of this proliferative capacity is utilized, the potential yield approximates 10(12) myoblasts, equivalent to 1 kg of cells. Human myogenesis in vitro is no longer limited by cell number and is now amenable to molecular and biochemical analysis on a large scale.+

A single laser method for subtraction of cell autofluorescence in flow cytometry.

In flow cytometry cell autofluorescence often interferes with efforts to measure low levels of bound fluorescent antibody. We have developed a way to correct for autofluorescence on a cell-by-cell basis. This results in improved estimates of real staining and better separation of the fluorescence histograms of stained and non-stained cells. Using a single laser, two-color fluorescence measurement system and two-color compensation electronics, autofluorescence and one fluorescent reagent are measured (rather than two fluorescent reagents). With fluorescein-conjugated antibodies the signal in the 515 to 555 nm range (green fluorescence) includes both fluorescein emission and part of the cellular autofluorescence. In the cases we have investigated, autofluorescence collected at wavelengths above 580 nm ("red") is well correlated with the green autofluorescence of the cells. A fraction of this red fluorescence is subtracted from the green fluorescence to produce an adjusted fluorescein output on which unstained cells have zero average signal. Use of this method facilitates the selection of rare cells transfected with surface antigen genes. Culture conditions affect the level of autofluorescence and the balance between red and green autofluorescence. When applied with fluorescein-conjugated reagents, the technique is compatible with the use of propidium iodide for live/dead cell discrimination.

Murine B cell differentiation lineages.

Subpopulations of mouse B cells express different amounts of two antigens (BLA-1 and BLA-2) recognized by rat monoclonal antibodies (53-10.1 and 30-E2). Two-color immunofluorescence analysis on the fluorescence-activated cell sorter (FACS) shows that the 53-10.1 monoclonal antibody reacts with a similar proportion of splenic B cells from normal and CBA/N (xid) mice, whereas 30-E2 reacts with most CBA/N B cells but with only a fraction of normal B cells. Data from three- and four-color immunofluorescence analyses with xid, athymic (nude), and normal mice suggest that the order in which these antigens are lost during B cell differentiation distinguishes two B cell lineages: immature B cells express both antigens, intermediate-stage B cells of one or the other lineage express only BLA-1 or only BLA-2, respectively, and mature resting B cells express neither. CBA/N mice lack one of the putative intermediate populations (BLA-1+,2-); thus, this population apparently gives rise to the predominant mature B cell population, which is present in normal adult spleen and lymph node but is missing in CBA/N. The other putative intermediate population (BLA-1-,2+) is decreased by two- to threefold in spleens from nude mice compared with strain-matched controls. Both BLA-1 and BLA-2 antigens rapidly reappear after specific (antigen) or nonspecific (lipopolysaccharide) B cell activation. IgM plaque-forming cells (PFC) derived from such activated cells continue to express both antigens while IgG PFC express only BLA-1.

Three-color immunofluorescence analysis of mouse B-lymphocyte subpopulations.

We have modified a fluorescence-activated cell sorter (FACS) to make three independent immunofluorescence measurements on each cell and used this system to study mouse B-lymphocyte subpopulations. An argon-ion laser (emitting at 488 nm) excites fluorescein- and phycoerythrin-labeled reagents, and a tunable dye laser charged with rhodamine 6G (emitting at 615 nm) excites an allophycocyanin-labeled reagent. We report simultaneous measurements of IgM, IgD, and the recently-defined mouse B lymphocyte antigens BLA-1 and BLA-2 on splenic lymphocytes of CBA/J mice and mice of the congenic strain CBA/N (which have an X-linked immunodeficiency [xid]). These data provide information on relationships among the B-cell populations in CBA/J "normal" mice and the defective CBA/N that could not be derived from one- or two-color immunofluorescent measurements. We believe this is the first use of allophycocyanin as an immunofluorescence label.

Detection of fetal erythrocytes in maternal blood post partum with the fluorescence-activated cell sorter.

A study was made of the frequency and amount of fetal hemorrhage into maternal blood during labor and delivery as evidenced by the number of fetal cells present in the maternal circulation immediately after spontaneous vaginal delivery. A sensitive, indirect immunofluorescence was used with fluorescence-activated cell sorter analysis of erythrocytes. All of the 16 Rh-negative mothers studied after vaginal delivery of Rh-positive infants had circulating Rh-positive cells. The mean Rh-positive to Rh-negative erythrocyte ratio was 1:14, 100 in maternal blood, which corresponds to a mean fetal hemorrhage of 156 microliters. The test described is sufficiently sensitive to be used for the study of primary Rh isoimmunization and could be clinically applicable for antepartum screening to determine which patients require Rh immune globulin treatment before delivery.

Demonstration of B-cell maturation in X-linked immunodeficient mice by simultaneous three-colour immunofluorescence.

CBA/N mice carrying the X-linked immune deficiency gene (xid) have fewer splenic B cells than normal CBA mice and are unresponsive to a certain class of antigens. Studies of B-cell surface-marker expression and immune responsiveness have led to the commonly accepted idea that the B cells in adult xid mice are immature and resemble the B cells of young (1-3 week old) normal mice. That is, like young animals, xid mice lack cells in the most numerous of three IgM/IgD B-cell subpopulations (designated I in Fig. 1a, b) present in adult spleen. We now report, however, that this picture is an oversimplification and that in fact the B cells in adult xid mice differ from those present in either adult or young normal mice. Using quantitative three-colour fluorescence-activated cell sorter (FACS) analyses, we have compared the correlated expression of IgM, IgD and a newly discovered B-lymphocyte antigen (BLA-1) on splenic B cells in normal and xid mice. We show here (1) that most B cells in adult xid mice (as in normals) are BLA-1- whereas all B cells in young animals are BLA-1+; (2) that the major difference in the IgM/IgD B-cell subpopulations found between xid and normal mice is limited to the BLA-1- cells; and (3) that xid mice have increased numbers of BLA-1+ population III B cells.