Clinical Immunology
`Vol. 90, No. 3, March, pp. 425–430, 1999
`Article ID clim.1998.4654, available online at http://www.idealibrary.com on
`
`RAPID COMMUNICATION
`Detection of Intracellular Phosphorylated STAT-1 by Flow Cytometry
`
`Thomas A. Fleisher, Susan E. Dorman,* Jill A. Anderson,* Michael Vail,
`Margaret R. Brown, and Steven M. Holland*
`Immunology Service, Warren G. Magnuson Clinical Center, and *Laboratory of Host Defenses, NIAID, NIH, Bethesda, Maryland 20892
`
`We have applied flow cytometry to the investigation
`of interferon-g activation of human monocytes. This
`approach uses monoclonal antibodies that distinguish
`between the native and phosphorylated forms of
`STAT-1. It enables rapid and quantitative assessment
`of STAT-1 phosphorylation on a discrete cell basis and
`is both more sensitive and less time consuming than
`immunoblotting. Furthermore, it allows for discrimi-
`nation between a mixture of cells that differ in their
`response to interferon-g. This approach should allow
`for the evaluation of different intracellular signaling
`pathways using a combination of monoclonal reagents
`that are specific for native and activation modified
`proteins. Application of this form of testing should
`prove valuable in screening for signaling defects in
`selected patients with recurrent infections. In addi-
`tion, this technique should permit dissection of a full
`range of cellular signaling pathways at the protein
`level.
`
`We have employed a novel flow cytometric assay to
`investigate the interferon-g cellular activation path-
`way that is mediated by Jak–STAT signal transduc-
`tion. This method depends on specific monoclonal an-
`tibodies that distinguish between the native and
`phosphorylated forms of STAT-1. It enables rapid and
`quantitative assessment of STAT-1 phosphorylation on
`a discrete cell basis and is both more sensitive and less
`time consuming than immunoblotting. Furthermore, it
`allows for discrimination between a mixture of cells
`that differ in their response to an activation signal.
`General application of this technique in the evaluation
`of intracellular signaling proteins requires combina-
`tions of monoclonal antibodies that are specific for na-
`tive and activation modified proteins. It has the poten-
`tial of being applicable in any setting where
`immunoblotting has been useful to dissect intracellu-
`lar signaling pathways.
`
`MATERIALS AND METHODS
`
`INTRODUCTION
`
`Cell Preparation
`
`Intracellular protein phosphorylation is a critical
`step in cellular activation induced by the binding of
`different ligands to cell surface receptors. This process
`is initiated by ligand activation of a specific protein–
`tyrosine kinase(s) that is associated with intracellular
`domains of the respective ligand receptor. One impor-
`tant pathway in this cell activation process involves
`the Janus kinase (Jak) family linked to signal trans-
`ducer and activator of transcription (STAT) proteins
`(1). In this well-characterized signal transduction
`pathway, ligand-receptor binding activates a member
`of the Jak family associated with an intracellular do-
`main of the receptor. This is followed by phosphoryla-
`tion of one or more STAT proteins, homo- or het-
`erodimerization of the phosphorylated STAT protein,
`and movement of the dimer into the nucleus. This final
`step allows binding to DNA regulatory elements which
`affect gene transcription.
`
`Peripheral blood mononuclear cells (PBMC) were ob-
`tained by density gradient centrifugation of EDTA an-
`ti-coagulated whole blood as previously described (2).
`Mononuclear cells were prepared at 5 3 106 cells/ml in
`phosphate-buffered saline (PBS) with 2% fetal calf se-
`rum and 400 ml aliquots were cultured without or with
`interferon-g (100 or 1000 IU/ml in the standard assay)
`at 37°C for the times indicated. Following incubation,
`the cells were either treated with specific antibodies or
`subjected to fixation and permeabilization before anti-
`body addition. The latter involved adding 400 ml of
`fixation reagent (Reagent A, Fix & Perm, Caltag Lab-
`oratories, Burlingame, CA) to 400 ml of the interferon-g
`pretreated cells and holding the tubes at room temper-
`ature for 2–3 min. Following a wash step, 100 ml of
`permeabilization medium (Reagent B, Fix & Perm,
`Caltag) together with a specific antibody was added to
`each cell pellet followed by a 30-min incubation at room
`temperature. The tubes were then washed, incubated
`
`425
`425
`
`1521-6616/99
`
`Page 1 of 6
`
`Fluidigm
`Exhibit 1004
`
`

`
`426
`
`RAPID COMMUNICATION
`
`TABLE 1
`Conditions for Antibody Binding to Phosphorylated STAT-1
`in Interferon-g (1000 IU/ml) Stimulated Human Monocytes
`
`Fluorescence index
`
`Surface
`binding
`
`Fix 1 Perm
`
`Fix 1 Perm
`1 MeOH
`
`STAT-1
`Phosphorylated STAT-1
`
`0.8
`0.9
`
`8.0
`1.2
`
`29.8
`7.2
`
`with the appropriate second antibody for 30 min at
`room temperature, and again washed before being re-
`suspended in 200 ml of PBS for flow cytometry. An
`augmented fixation and permeabilization method in-
`volved the addition of 3 ml of cold methanol while
`vortexing in between the addition of Reagents A and B
`(3). The tubes were held for 10 min at 4°C, centrifuged,
`washed in PBS, and resuspended in permeabilization
`medium (Reagent B) with antibody as described above.
`The precultured unmodified PBMC, fixed and per-
`meabilized (F/P) PBMC, and fixed and permeabilized
`including methanol (F/P 1 MeOH) PBMC were incu-
`bated with 1 mg of murine monoclonal anti-human
`STAT-1 cytoplasmic terminus (Transduction Laborato-
`ries, Lexington, KY), 1 mg of murine IgG2b, 0.1 mg of
`rabbit anti-human phosphorylated STAT-1 (New En-
`gland Biolabs, Beverly, MA), or 0.1 mg of rabbit IgG for
`30 min as above. The second antibody consisted of
`either 1 mg of FITC-conjugated F(ab9)2 goat anti-mu-
`rine IgG (Caltag) or 1 mg of FITC conjugated F(ab9)2
`goat anti-rabbit IgG (Caltag) with a 30-min incubation
`at room temperature as above. Following a final wash
`step, the cells were resuspended in 200 ml of PBS and
`analyzed with a flow cytometer.
`
`FIG. 2. Flow cytometry histograms of anti-phosphorylated
`STAT-1 binding to control and interferon-g receptor-1 deficient pa-
`tient monocytes. The PBMC were preincubated with 1000 IU/ml
`interferon-gfor 15 min and then were subjected to F/P 1 MeOH and
`stained as described under Materials and Methods. Dashed lines,
`isotype control staining; solid lines, specific antibody staining.
`
`Flow Cytometry
`
`PBMC samples were evaluated following antibody
`exposure using a FACScan (Becton–Dickinson, San
`Jose, CA) equipped with Cell Quest Software (Becton–
`Dickinson). The nonfluorescent signals for forward an-
`gle (FSC) and side angle light scatter were collected on
`each sample to determine a monocyte gate (4). The
`fluorescent signal at 525 nM for each sample was col-
`lected and evaluated using log amplification. Antibody
`binding to monocytes was expressed as a fluorescence
`index (FI) defined as FI 5 geometric mean channel
`fluorescence stimulated cells/geometric mean channel
`fluorescence unstimulated cells. In studies using hu-
`man EBV transformed B cell lines the cell preparation
`
`FIG. 1. Flow cytometry histograms of anti-STAT-1 (left) and
`anti-phosphorylated STAT-1 (right) intracellular binding to control
`human monocytes. The PBMC were preincubated with 1000 IU/ml
`interferon-gfor 15 min and then were subjected to F/P 1 MeOH and
`stained as described under Materials and Methods. Dashed lines,
`isotype control staining; solid lines, specific antibody staining.
`
`FIG. 3. Flow cytometry histograms of anti-phosphorylated
`STAT-1 binding to control and interferon-g receptor-2 deficient pa-
`tient B cell lines. The cells were preincubated with 1000 IU/ml
`interferon-g for 15 min and then subjected to F/P 1 MeOH and
`stained as described under Materials and Methods. Dashed lines,
`isotype control staining; solid lines, specific antibody staining.
`
`Page 2 of 6
`
`Fluidigm
`Exhibit 1004
`
`

`
`RAPID COMMUNICATION
`
`427
`
`FIG. 4.
`(A) Interferon-g dose curve for STAT-1 phosphorylation detected by flow cytometry. The control monocytes were preincubated
`with doses of interferon-g(0.1 to 10,000 IU/ml) for 15 min and then subjected to F/P 1 MeOH and stained as described under Materials and
`Methods. Dashed lines, isotype control staining; solid lines, specific antibody staining. (B) Interferon-g dose curve for STAT-1 phosphory-
`lation detected by immunoblotting. Upper panel shows the binding of rabbit anti-phosphorylated STAT-1; lower panel shows the same blot
`stripped and reprobed with mouse anti-STAT-1. STAT-1 loading was equivalent in all lanes.
`
`was similar but no gating process was used; rather all
`cells above an FSC threshold were evaluated using the
`same criteria as above.
`
`Immunoblot
`
`PBMC from a control were prepared as previously
`described and resuspended at a concentration of 3 3
`106/mL in RPMI containing 10% fetal calf serum. One
`milliliter of cells was stimulated with various concen-
`trations of interferon-g or media for 15 min and then
`solubilized with lysis buffer as described (5). Following
`centrifugation to remove insoluble components, the su-
`
`pernatants were boiled for 8 min in an equal volume of
`SDS sample buffer. A 30-ml aliquot of each sample was
`resolved on a 4–12% SDS–polyacrylamide gel and trans-
`ferred to a nitrocellulose membrane. The membrane
`was blocked overnight at 4°C using nonfat dry milk in
`phosphate-buffered saline plus 0.1% Tween 20 (PBST)
`and probed with rabbit anti-phosphorylated STAT-1
`(New England Biolabs) for 6 h atroom temperature.
`The membrane was washed in PBST, incubated for 1 h
`with horseradish peroxidase (HRP)-conjugated donkey
`anti-rabbit immunoglobulin (Amersham Corporation,
`Buckinghamshire, England), and developed using the
`enhanced chemiluminescence method (ECL, Amer-
`
`Page 3 of 6
`
`Fluidigm
`Exhibit 1004
`
`

`
`428
`
`RAPID COMMUNICATION
`
`cell line derived from a patient with a defect in the
`interferon-g receptor-2 failed to show phosphorylation
`of STAT-1 following interferon-g activation (Fig. 3) (6).
`Absence of interferon-g induced STAT-1 phosphoryla-
`tion in this patient’s cells was confirmed by immuno-
`blotting (4).
`
`Interferon-g Induced STAT-1 Phosphorylation Is Dose
`Dependent
`
`Interferon-g dose titration (0.1 to 10,000 IU/ml )
`using control PBMC showed STAT-1 phosphorylation
`following exposure to 1 IU/ml for 15 min (Fig. 4A). The
`level of phosphorylation appeared maximal by 10
`IU/ml and remained relatively constant at doses up to
`10,000 IU/ml. Immunoblotting using mononuclear
`cells from the same donor showed a similar dose curve.
`However, the immunoblots failed to detect phosphory-
`lated STAT-1 after stimulation with 1 IU/ml interfer-
`on-g (Fig. 4B).
`
`Interferon-g Induced STAT-1 Phosphorylation Occurs
`Rapidly
`
`Within 2.5 min following exposure to interferon-g
`(100 IU/ml), phosphorylation was detected in control
`monocytes (Fig. 5). This appeared to be maximal by
`5–10 min and remained constant until 15 min. By
`30–60 min following activation the level of STAT-1
`phosphorylation began to decrease. Similar findings
`were observed in kinetic studies using a control B cell
`line (data not shown).
`
`Phosphorylated STAT-1 Detection By Flow Cytometry
`Is Sensitive
`
`In order to define the sensitivity of this assay in
`distinguishing between cells that differ in their capac-
`ities to phosphorylate STAT-1, we performed cell mix-
`ing studies using control and interferon-g receptor-1
`deficient PBMC. The proportion of normal to abnormal
`monocytes on the flow histograms correlated with the
`composition of the cell mixture (Fig. 6). Flow cytometry
`detected intracellular phosphorylated STAT-1 when
`10% control cells were mixed with 90% interferon-g
`receptor-1 deficient cells. However, neither normal nor
`abnormal cells could be detected at the 1% level (data
`not shown).
`
`DISCUSSION
`
`We describe a flow cytometric assay that detects
`interferon-ginduced phosphorylation of STAT-1 in hu-
`man monocytes and lymphocytes. This method de-
`pends on specific antibody reagents that discriminate
`
`FIG. 5. Time course for interferon-g induced STAT-1 phosphor-
`ylation detected by flow cytometry. PBMC were incubated with 100
`IU/ml of interferon-g for varying times and then were subjected to
`F/P 1 MeOH and stained as described under Materials and Methods.
`The data are expressed as a fluorescence index (FI) as defined under
`Materials and Methods.
`
`sham). The membrane was stripped and reprobed with
`monoclonal anti-STAT-1 (Transduction Laboratories)
`for 1 h, washed, incubated for 1 h with HRP-conjugated
`sheep anti-mouse immunoglobulin (Amersham), and
`developed using the ECL system (6).
`
`RESULTS
`Phosphorylated STAT-1 Can Be Detected Using Flow
`Cytometry
`
`Neither STAT-1 nor phosphorylated STAT-1 was de-
`tected on the surface of human monocytes (Table 1).
`Fixation and plasma membrane permeabilization were
`required for the detection of native STAT-1 in mono-
`cytes (Table 1) and B cell lines (data not shown). Phos-
`phorylated STAT-1 was not detected even after fixation
`and plasma membrane permeabilization of monocytes
`preincubated with interferon-g. However, the addition
`of methanol to enhance membrane permeabilization
`resulted in detection of phosphorylated STAT-1 follow-
`ing interferon-g activation (Fig. 1 and Table 1). Anti-
`body binding to native STAT-1 was not altered by prior
`cell activation with interferon-g (Fig. 1) but did in-
`crease with F/P 1 MeOH (Table 1).
`Interferon-g activation consistently generated de-
`tectable phosphorylated STAT-1 in adult control mono-
`cytes (n 5 10) as evidenced by increased fluorescence
`compared to the unstimulated cells (FI 5 2.6 –7.2,
`mean 5 4.5 6 1.4). In contrast, monocytes from a
`patient with a defect in interferon-greceptor-1 showed
`no increase in antibody binding to phosphorylated
`STAT-1 following interferon-g activation (Fig. 2) (7). A
`B cell line derived from an adult control subject dem-
`onstrated STAT-1 phosphorylation following interfer-
`on-g (1000 IU/ml) activation (Fig. 3). In contrast, a B
`
`Page 4 of 6
`
`Fluidigm
`Exhibit 1004
`
`

`
`RAPID COMMUNICATION
`
`429
`
`FIG. 6. PBMC mixing experiment using control cells and interferon-g receptor-1 deficient patient cells in varying proportions based on
`the monocyte count in the PBMC preparations. Two different populations of cells could be identified for STAT-1 phosphorylation following
`100 IU/ml interferon-g activation for 15 min. Dashed lines, isotype control staining; solid lines, specific antibody staining.
`
`between native and phosphorylated proteins, as well as
`techniques for cell fixation and permeabilization that
`are routinely used in intracellular antigen and cyto-
`kine assays (8, 9). Detection of phosphorylated STAT-1
`requires the addition of methanol, a step which also
`enhances the detection of native STAT-1 (3).
`The time course of STAT-1 phosphorylation using
`the flow cytometric assay is similar to that described
`using immunoblotting (10). The dose curve observed
`was similar in parallel
`immunoblotting studies.
`However, the flow cytometric assay was more sensi-
`tive at the lower doses of interferon-g. This method
`also can discriminate between populations of cells
`that differ in their capacity to respond to an activa-
`tion signal.
`The flow cytometric evaluation we describe repre-
`sents a new approach to the evaluation of intracel-
`lular changes associated with activation. Previously,
`flow cytometry has been used to evaluate changes in
`intracellular cations, pH, osmolality, and glutathi-
`one (11). Generally, these methods evaluate changes
`that are generic to cell activation and thus are not
`specific to an activation pathway. In contrast, the
`method we describe is sensitive to the pathway of
`activation. This technique should find applications in
`the study of multiple phosphorylation-dependent
`pathways such as those involving other Jak–STAT
`combinations, IkB, and MAP kinases. This approach
`has distinct advantages in terms of sensitivity,
`speed, and technical simplicity compared to immu-
`noblotting. In addition, multiple phenotypes can be
`assayed within the same sample based on the cellu-
`lar discrimination inherent to flow cytometry. This
`type of approach has been successfully applied in the
`intracellular assessment of cytokine production by
`
`adding surface staining with lineage-specific mono-
`clonal antibodies.
`Flow cytometry provides a sensitive and rapid method
`for the evaluation of activation-specific changes in
`intracellular proteins. It is technically less demand-
`ing than immunoblotting and can discriminate dis-
`tinct cell types in a heterogeneous sample. This ap-
`proach should be valuable in studying any activation
`pathway for which antibody reagents exist that dis-
`criminate between a native and an activation modi-
`fied protein. This approach should have broad utility
`whenever cell signaling generates novel intracellular
`protein epitopes. Clinical settings could include im-
`munologic (leukocyte activation pathways), rheuma-
`tologic (apoptotic pathways), and oncologic (growth
`and differentiation pathways) disorders. It could also
`prove to be useful in assessing the effectiveness of
`gene transduction in protocols directed at the genetic
`correction of cells from patients with disorders in cell
`signaling.
`
`REFERENCES
`
`1. Darnell, J. E. Jr., Kerr, I. A. M., Stark, G. R. Jak–STAT path-
`ways and transcriptional activation in response to IFNs and
`other extracellular signaling proteins. Science 264, 1415–1420,
`1994.
`2. Mardiney, M., III, Brown, M. R., and Fleisher, T. A. Measure-
`ment of T cell CD69 expression: A rapid and efficient means to
`assess mitogen or antigen induced proliferative capacity in nor-
`mals. Cytometry 26, 305–310, 1996.
`3. Kartner, N., Evernden-Porelle, D., Bradley, G., and Ling, V.
`Detection of P-glycoprotein in multidrug resistant cell lines by
`monoclonal antibodies. Nature 316, 820–823, 1985.
`4. Salzman, G. C., Crowell, J. M., Martin, J. C., et al., Cell classi-
`fication by light scattering: Identification and separation of un-
`stained leukocytes. Acta Cytol. 10, 374–377, 1975.
`
`Page 5 of 6
`
`Fluidigm
`Exhibit 1004
`
`

`
`430
`
`RAPID COMMUNICATION
`
`5. Kotenko, S. V., Izotova, L. S., Pollack, B. P., et al., Interaction
`between the components of the interferon gamma receptor com-
`plex. J. Biol. Chem. 270, 20915–10921, 1995.
`6. Dorman, S. E., Holland, S. M. Mutation in the signal trans-
`ducing chain of the interferon-gamma receptor and suscepti-
`bility to mycobacterial infection. J. Clin. Invest. 101, 2364 –
`2369, 1998.
`7. Holland, S. M., Dorman, S. E., Kwon, A., et al., Abnormal regu-
`lation of interferon-g, interleukin-12, and tumor necrosis fac-
`tor-a in human interferon-g receptor 1 deficiency. J. Infect. Dis.
`178, 1095–1104, 1998.
`8. Groeneveld, K., te Marvelde, J. G., van dem Beemd, M. W. M.,
`Hoojkaas, H., and van Dongen, J. J. M. Flow cytometric detection
`
`of intracellular antigens for immunophenotyping of normal and
`malignant leukocytes. Leukemia 10, 1383–1389, 1996.
`9. Prussin, C., Metcalfe, D. D. Detection of intracytoplasmic cyto-
`kine using flow cytometry and directly conjugated anti-cytokine
`antibodies. J. Immunol. Methods 118, 117–128, 1995.
`10. Lee, C.-K., Bluyssen, H. A. R., Levy, D. E. Regulation of inter-
`feron-a responsiveness by the duration of janus kinase activity.
`J. Biol. Chem. 272, 21872–21877, 1997.
`11. Rabinovitch, P. S., June, C. H., and Kavanagh, T. J. Measure-
`ments of cell physiology: Ionized calcium, pH, and glutathione.
`In “Clinical Flow Cytometry Principles and Application” (K. D.
`Bauer, R. E. Duque, and T. V. Shankey, Eds.), pp. 505–534,
`Williams & Wilkins. Baltimore, MD, 1993.
`
`Received October 22, 1998; accepted with revision, November 13, 1998
`
`Page 6 of 6
`
`Fluidigm
`Exhibit 1004