SEARCH

SEARCH BY CITATION

Keywords:

  • T and B lymphocyte;
  • electron microscope;
  • confocal laser scanning microscope;
  • hematology analyzer;
  • surface antigen

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS ANDMETHODS
  4. RESULTS
  5. DISCUSSION
  6. LITERATURE CITED
  7. Supporting Information

T and B lymphocytes are difficult to distinguish morphologically even with electron microscopy, and antibodies are generally used to make the distinction. A specific reagent, consisting of nonionic and cationic detergents, is used for leukocyte differentiation using the Sysmex automated blood analyzer. This reagent increases cell membrane porosity and enables the introduction of fluorescent dye into leukocytes. In this study, we investigated the effect of this specific detergent on the morphology of T and B lymphocytes. T and B lymphocytes were obtained by density gradient centrifugation and magnetic cell sorting, with a minimum of 90% isolation efficiency. T and B lymphocytes were then treated with the specific detergent and fluorescent dye, and their distribution was analyzed based on side scatter and fluorescence intensity using general-purpose flow cytometry (FCM). Fluorescent images were observed using a confocal laser scanning microscope (CLSM), cellular inner structures using a transmission electron microscopy (TEM), and cell surfaces using a scanning electron microscope (SEM). The ratio of cholesterol to total lipid in cell membranes of B and T lymphocytes was measured using a fluorescent assay kit. The distribution of fluorescence intensity was different between T and B lymphocyte clusters, according to the FCM analysis. CLSM observations revealed that the fluorescent dye mainly stained cytoplasmic organelles. FCM, TEM, and SEM observations revealed that B lymphocytes are more likely to lose surface antigens and intracellular organelles than T lymphocytes, which allows the visual distinction between T and B lymphocytes. The ratio of cholesterol to total lipid in T lymphocyte membranes had tendency higher than that in B lymphocyte membranes. In this study, we demonstrate that cells with differences in cell membrane cholesterol amounts, such as B and T lymphocytes could be identified using an inexpensive detergent, as an alternative to costly antibodies. © 2013 International Society for Advancement of Cytometry.

Lymphocytes are a type of white blood cells that derives from a common lymphoid progenitor. Most lymphocytes are small, featureless cells with few cytoplasmic organelles and mostly inactive nuclear chromatin, as indicated by its condensed state. When lymphocytes encounter foreign antigens, they proliferate, differentiate, and become activated, which allows them to play a central role in the immune response (1). Lymphocytes in peripheral blood are divided into three types: T lymphocytes or T cells, which express the T-cell receptor (CD3); B lymphocytes or B cells, which express membrane immunoglobulin's; and null cells, which express neither of these cell-surface markers. The percentage of T lymphocytes in peripheral blood of healthy individuals is 75% of total lymphocytes, which represents the majority of lymphocytes. B lymphocytes constitute 15–25%, and null cells constitute only 10% of the total lymphocytes. Therefore, lymphocytes can be roughly divided into T and B lymphocytes (2, 3).

T lymphocytes develop in the thymus and play an important role in controlling the immune system. T lymphocyte disorders are associated with several disease conditions such as immunodeficiency, autoimmunity, allergy, and proliferative disease. On the other hand, B lymphocytes develop in the bone marrow and differentiate into plasma cells that secrete antibodies when activated. Changes in the immune system can be detected using the ratio between T and B lymphocytes, since they are both strongly required for normal immune system function (3). An increase in T lymphocyte cell count is associated with diseases such as mononucleosis and T lymphocytic tumors, and an increase in B lymphocyte cell count is associated with diseases such as B lymphocytic tumors, tuberculosis (4), benign monoclonal gammopathy, and myasthenia gravis (5). Moreover, the ratio of B lymphocytes to T lymphocytes increases during the early stages of HIV, as the T lymphocyte count declines (6, 7).

While unactivated T and B lymphocytes are morphologically very similar even with electron microscopy (8, 9), they become morphologically distinct following activation by an antigen. They are usually distinguished from each other using flow cytometry, based on their differential expression of cell-surface proteins. T lymphocytes are identified by their expression of CD3, and B lymphocytes by their CD19 or CD20 expression (1).

The leukocyte differentiation scatter diagram obtained via Sysmex automated hematology analyzers classifies leukocytes into four types: lymphocytes, monocytes, neutrophils, and eosinophils. The differentiation mechanism involves exposing leukocytes to a specific reagent, STROMATOLYSER®-4DL (contains detergent) and STROMATOLYSER-4DS (contains fluorescent dye which is polymethine dye (10)), and displaying their position on a two-dimensional scatter diagram (DIFF scattergram) based on fluorescence intensity and side scatter. Leukocytes are then classified based on their position on the DIFF scattergram (11, 12). We recently observed that treatment of samples with this specific detergent differentially degraded the cell membranes of T and B lymphocytes, resulting in morphological differences between these two types of lymphocytes. Thus, we investigated the possibility of morphologically distinguishing between T and B lymphocytes following treatment with this detergent.

MATERIALS ANDMETHODS

  1. Top of page
  2. Abstract
  3. MATERIALS ANDMETHODS
  4. RESULTS
  5. DISCUSSION
  6. LITERATURE CITED
  7. Supporting Information

Blood

Peripheral blood was collected by venipuncture, into tubes containing EDTA (Terumo, Tokyo, Japan), from five healthy human subjects (three women and two men) following informed consent. The period of experiment was from 4 September, 2008 to 12 May, 2009. The mononuclear cell-rich fraction was prepared by density gradient centrifugation following the manufacturer's instructions, using the lymphocyte separation solution, d = 1.077 (Nacalai Tesque, Kyoto, Japan), and washed with PBS.

Magnetic Cell Sorting

T and B lymphocytes were isolated from the mononuclear cell-rich fraction by negative selection using a magnetic cell sorting (MACS) system (STEMCELL Technologies, British Columbia, Canada), following the manufacturer's instruction, and measured using Sysmex XE-2100™ (XE, Sysmex, Kobe, Japan).

Cell Staining with Antibodies

Isolated T and B lymphocytes were incubated with FITC-conjugated CD3 (T lymphocyte marker) and CD19 (B lymphocyte marker) monoclonal antibody (DAKO, Glostrup, Denmark) solutions at a concentration of 20 mg/L in PBS for 30 min at 4°C. FITC-conjugated mouse IgG1 antibody (DAKO) was used as negative control.

Cell Staining with Specific Reagent

Isolated T and B lymphocytes were exposed to 4DL (Sysmex, Kobe, Japan) and 4DS (Sysmex), in a manner similar to that used for XE (Sysmex), as follows: 18 μL of sample, 882 μL of 4DL, and 18 μL of 4DS.

Flow Cytometry

Stained cells were analyzed through flow cytometry (FCM), using a FACSCalibur™ (BD Biosciences, Franklin Lakes, NJ). Quality control measure performed before every use using BD Calibrite™ (BD Biosciences).

Confocal Laser Scanning Microscopy

Stained cells were immediately placed on poly-L-lysine-coated coverslips (Sigma, MI) and observed through a confocal laser scanning microscope (IX81, Olympus, Tokyo, Japan; CSU-X1, Yokogawa electric, Tokyo, Japan; ImagEM, Hamamatsu Photonics, Hamamatsu, Japan).

Transmission Electron Microscopy

Stained cells were fixed with a 1% glutaraldehyde (Electron Microscopy Sciences, Hatfield, PA) solution in PBS for 16 h at 4°C. Fixed cells were attached to silanized glass slides using Cytospin® (Thermo Fisher Scientific, MA), and postfixed in 1% osmium tetroxide for 45 min at 4°C. Following fixation with osmium, samples were dehydrated in a graded series of ethanol and invert-embedded in Quetol 812 (Nisshin EM, Tokyo, Japan). Samples were then cut into 80- to 100-nm thick sections with Ultracut UCT ultramicrotome (Leica Microsystems, Wetzlar, Germany), and observed under an H-7500 transmission electron microscope (Hitachi High-Technologies, Tokyo, Japan).

Scanning Electron Microscopy

For scanning electron microscopy (SEM), stained cells were also fixed with a 1% glutaraldehyde (Electron Microscopy Sciences, Hatfield, PA) solution in PBS for 16 h at 4°C. Fixed cells were attached to poly-L-lysine-coated glass slides. After fixation, samples were dehydrated following the same procedure used for transmission electron microscopy (TEM), but replacing ethanol with t-butyl alcohol (Wako Pure Chemicals, Osaka, Japan). Samples were then freeze-dried (Hitachi High-Technologies) and coated with osmium using Neoc (MEIWAFOSIS, Tokyo, Japan). Observation was carried out using a JSM-7500F electron microscope (Nippon-Denshi, Osaka, Japan).

Cholesterol Quantification

Cell membranes were isolated from T and B lymphocytes by density gradient ultracentrifugation. Total lipid was extracted from the cell membranes using the Bligh-Dyer method (13). Cholesterol was quantified using the Cholesterol/Cholesteryl Ester quantitation kit (BioVision, San Francisco, CA), according to the manufacturer's instructions.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS ANDMETHODS
  4. RESULTS
  5. DISCUSSION
  6. LITERATURE CITED
  7. Supporting Information

Isolated T or B lymphocytes consisted of at least 90% CD3-positive or CD19-positive cells, and the cells had lymphocyte specific morphology, which was uniformly observed by electron microscopy under low magnification (Fig. 1).

thumbnail image

Figure 1. Verification of T and B lymphocyte isolation. (A) Isolated T lymphocytes were stained with CD3-FITC and analyzed using FCM (left), or observed using TEM (right). CD3-positive cells comprised 97.1%. (B) Isolated B lymphocytes were stained with CD19-FITC and analyzed using FCM (left), or observed using TEM (right). CD19-positive cells comprised 93.1%. SSC: side scatter, Bar = 5 μm.

Download figure to PowerPoint

First, T lymphocyte or B lymphocyte counts were measured in whole blood samples using XE (Fig. 2). Both T and B lymphocytes distributions were positioned within the lymphocyte area in whole blood. T lymphocyte distribution appeared as a widespread area across 4DS fluorescence intensity (vertical axis) within the lymphocyte area (Fig. 2A, center), while B lymphocyte distribution appeared as a low narrow area across 4DS fluorescence intensity within the lymphocyte area (Fig. 2A, right). Similar distribution patterns were observed in FCM analysis for both T lymphocytes (Fig. 2B, left; mean fluorescence intensity [MFI] = 15.05) and B lymphocytes (Fig. 2B, right; MFI = 13.95). The histogram in Figure 2C plots 4DS fluorescence intensity on the horizontal axis, using the same data as Figure 2B. The differences in distribution between T and B lymphocytes are obvious in Figure 2C. Moreover, confocal laser scanning microscopy observations of both types of lymphocytes revealed that the regions that strongly stained with 4DS corresponded to the cytoplasmic organelles but not to the nucleus (Fig. 2D).

thumbnail image

Figure 2. Distribution of isolated T and B lymphocytes on the XE cytogram. (A) Sysmex automated hematology analyzer XE was used to obtain the leukocyte differentiation cytogram of whole blood (left), isolated T lymphocytes (center), and isolated B lymphocytes (right). (B) Isolated T or B lymphocytes were stained with 4DS and 4DL, following the same procedure used for XE, and analyzed using FCM. (C) The histogram plots 4DS fluorescence intensity on the horizontal axis, using the same data as (B). The difference in distribution between T and B lymphocytes is obvious. (D) 4DS stained images of lymphocytes using CLSM. 4DS stained cytoplasmic organelle but not the nucleus. SSC: side scatter, Bar = 2 μm.

Download figure to PowerPoint

Then, the extent of cytoplasmic organelle loss in T and B lymphocytes was observed using electron microscopy (Fig. 3). Both TEM (Figs. 3A and 3B, left) and SEM (Figs. 3C and 3D, right) observations, prior to treatment with the specific reagent, showed almost no morphological differences between the B and T lymphocytes, and the two types of lymphocytes could not be distinguished. However, TEM observations showed that, following treatment with the specific reagent, T lymphocytes maintained their cell membranes and cytoplasmic organelles (Fig. 3A, right), while these were almost completely lost in B lymphocytes (Fig. 3B, right). SEM observations revealed larger pores on B lymphocyte cell-surfaces (Fig. 3D, right) than that on T lymphocyte cell-surfaces (Fig. 3C, right), enough to discern between the two types of lymphocytes.

thumbnail image

Figure 3. Electron microscopy observations of isolated T and B lymphocytes. TEM images of T lymphocytes (A) and B lymphocytes (B). SEM images of T lymphocytes (C) and B lymphocytes (D). Left images in each panel represent cells prior to treatment with 4DL and 4DS. Right images in each panel represent cells following treatment with 4DL and 4DS. Bar = 2 μm.

Download figure to PowerPoint

Moreover, the effect of the specific reagent on T and B lymphocyte surface antigens was investigated. This was performed to verify the differences between T and B lymphocyte cell membrane loss, using FCM analysis and antibodies (CD3-FITC and CD19-FITC) to cell-surface antigens (Fig. 4). FITC fluorescence intensity was plotted on the vertical axis, side scatter (SSC) on the horizontal axis, and both types of lymphocytes were analyzed following treatment with the specific reagent for 0, 2, 5, 10, and 15 min at room temperature (Fig. 4A). A constant MFI of CD3-FITC in T lymphocytes was maintained until 15 min (Fig. 4A), while the MFI of CD19-FITC in B lymphocytes decreased from 28.25 to 8.85, the latter value being similar to the MFI of B lymphocytes stained with negative antibody (MFI = 8.03, data not shown) (Fig. 4B).

thumbnail image

Figure 4. Verification of cell membrane loss using antibodies to cell-surface antigens. Isolated T lymphocytes (A) and B lymphocytes (B) treated with 4DS and 4DL for 0, 2, 5, 10, and 15 min. The numbers in each cytogram represent the MFI corresponding to each cluster. SSC: side scatter.

Download figure to PowerPoint

Finally, the ratio of cholesterol to total lipids in cell membranes of T and B lymphocytes was measured to identify the reason behind the differences in resistance to detergent between the two types of lymphocytes (Fig. 5). Our results showed that T lymphocytes tended to contain more cholesterol in their cell membranes than B lymphocytes.

thumbnail image

Figure 5. Quantification of cholesterol in cell membranes of T and B lymphocytes. Total lipid was extracted from cell membranes of isolated T or B lymphocytes, and the ratio of cholesterol to total lipid was calculated, n = 4 (mean ± SE). P = 0.057.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS ANDMETHODS
  4. RESULTS
  5. DISCUSSION
  6. LITERATURE CITED
  7. Supporting Information

The ratio of T and B lymphocytes thus far could only be estimated through the use of costly antibodies (1–3), due to the lack of morphological differences between these two types of lymphocytes (8, 9). In this study, we revealed that a specific detergent differentially affects cell membranes of T and B lymphocytes (Figs. 3 and 4). Consequently, differences in the amount of residual cytoplasmic organelles translate into differences in the fluorescence intensity of the specific reagent that stains these cytoplasmic organelles. This results in a difference in fluorescence intensity distribution between T and B, as observed in the cytograms (Fig. 2). While T lymphocyte distribution appeared as a widespread area across 4DS fluorescence intensity (Fig. 2A, center), B lymphocyte distribution appeared as a low narrow area across 4DS fluorescence intensity (Fig. 2A, right).

Data points were randomly selected from XE cytograms for T (Fig. 2A, center) and B lymphocytes (Fig. 2A, right), shuffled, and presented in Figure 6. The corresponding T to B lymphocyte ratios were estimated as 10:0 (Fig. 6A), 8:2 (Fig. 6B), 6:4 (Fig. 6C), 4:6 (Fig. 6D), 2:8 (Fig. 6E), and 0:10 (Fig. 6F). Peripheral blood T lymphocytes in a healthy individual constitute 50–85% of lymphocytes, representing the majority of lymphocytes, while B lymphocytes constitute 5–20% of the total lymphocytes (1). Therefore, lymphocytes in healthy individuals have an elliptical distribution (Figs. 6B and 6C). However, the ratio of T to B lymphocytes is reversed as T lymphocyte counts decrease during the early stages of HIV (6, 7), or as B lymphocyte counts increase with B lymphocytic tumors (4, 5). The increased influence of B lymphocytes that distribute to a low narrow area of 4DS fluorescence intensity (Fig. 6E) shifts the lymphocyte distribution from elliptical to boots. Therefore, changes in the T to B lymphocytes ratio can be detected promptly and easily using mathematical analysis of the shapes of lymphocyte distributions on an automated hematology analyzer, with inexpensive detergent and staining dye as alternatives to expensive antibodies.

thumbnail image

Figure 6. Representation of various T to B lymphocyte ratios according to XE cytograms. Data points were randomly selected from the XE cytograms of T lymphocytes (Fig. 2A, center) and B lymphocytes (Fig. 2A, right), shuffled, and T to B lymphocyte ratios were estimated as 10:0 (A), 8:2 (B), 6:4 (C), 4:6 (D), 2:8 (E), and 0:10 (F). SSC: side scatter.

Download figure to PowerPoint

Moreover, in this report, we examined the amount of cholesterol compared to total lipid content in the cell membranes of B and T lymphocytes, to identify the reason behind the differences in resistance to detergent between the two types of lymphocytes. According to Anderson's report, the ratio of cholesterol to total lipid in the T lymphocytes was higher than that in the B lymphocytes. Our results were consistent with Anderson's report, too (14).

Cell membranes contain subdomains called lipid rafts that are rich in glycosphingolipids and cholesterol. Lipid rafts play a role in signal transduction and are referred to as detergent-resistant membranes (DRMs) because they are insoluble to nonionic detergents such as TritonX-100 (15–17). The role of DRMs in lymphocytes has been examined. A study has reported that DRMs exhibit protein compositions specific to the maturation stage of the B lymphocyte (18). Other studies have demonstrated that the recruitment of lipid rafts occurs at the site of cell–cell contact between T cells and antigen-presenting cells, following T receptor signaling (19–21). While CD2 was easily eluted from T cell membranes by treatment with nonionic detergent, CD48 and CD71 that are mainly located in DRMs were significantly resistant to the detergent (22). In addition, it has been suggested that flotillins, which are integral components of lipid rafts, are required for T lymphocyte uropod formation (23). The ratio of lipid rafts to cell membrane may vary according to the lymphocyte type, since lymphocytes exhibit specific traits depending on their in vivo function. This ratio may contribute to the difference in resistance to detergent that is observed between different types of lymphocytes.

In addition, it has been reported that cholesterol plays an important role in membrane resistance to nonionic detergent such as Triton X-100, according to artificial membrane experiments (24). This study demonstrates that the quantity of cholesterol in cell membranes is responsible for the differences in resistance to the Sysmex specific reagent that contains nonionic detergent, between the two types of lymphocytes.

Our study shows that T and B lymphocytes, which exhibit similar morphological structure even with electron microscopy, can be distinguished following treatment with detergent, due to the differential effect of the detergent on their cell membranes. This technique may possibly be applied to other cell types and could be further developed.

LITERATURE CITED

  1. Top of page
  2. Abstract
  3. MATERIALS ANDMETHODS
  4. RESULTS
  5. DISCUSSION
  6. LITERATURE CITED
  7. Supporting Information
  • 1
    Janeway CAJr,Travers P,Walport M, Shlomchik MJ. Immunobiology: The Immune System in Health and Disease: Flow cytometry and FACS analysis, 5th ed. New York: Garland Science; 2001. NBK10755.
  • 2
    Alberts B,Johnson A,Lewis J,Raff M,Roberts K,Walter P. Molecular Biology of the Cell: B lymphocytes Develop in the Bone Marrow; T Lymphocytes Develop in the Thymus, 4th ed. New York: Garland Science; 2002. NBK26921.
  • 3
    Blumenreich MS. The white blood cell and differential count. In: Walker HK,Hall WD,Hurst JW, editors. Clinical Methods,3rd ed. Boston: Butterworths; 1990. pp 724727.
  • 4
    Wada M. Flow cytometric analysis of peripheral T lymphocytes from patients with mycobacterial diseases. Kekkaku 1992; 67: 393407.
  • 5
    Ippoliti G,Piccolo G,Lombardi, M,Carnevale Maffe G,Cosi V,Pinelli P. Lymphocyte populations in patients with myasthenia gravis. Influence of thymectomy and immunosuppressive drugs. Ital J Neurol Sci 1982; 3: 273280.
  • 6
    Blime VB,Amiel C,Bene MC,May T,Canton P,Faure GC. Peripheral B cells with intracytoplasmic mu chains in HIV infection. Clin Immunol Immunopathol 1992; 63: 275279.
  • 7
    Bussmann BM,Reiche S,Bieniek B,Krznaric I,Ackermann F,Jassoy C. Loss of HIV-specific memory B-cells as a potential mechanism for the dysfunction of the humoral immune response against HIV. Virology 2010; 397: 713.
  • 8
    Alexander EL,Wetzel B. Human lymphocytes: Similarity of B and T cell surface morphology. Science 1975; 188: 732734.
  • 9
    Alexander E,Sanders S,Braylan R. Purported difference between human T-and B-cell surface morphology is an artefact. Nature 1976; 261: 239241.
  • 10
    Mizutani T,Ishiyama M,Sakata T. Research and method for classification and counting of leukocytes. U. S. Patent No. 6004816; 1999.
  • 11
    Fujimoto K. Principles of measurement in hematology analyzers manufactured by Sysmex Corporation. Sysmex J Int 1999; 9: 3144.
  • 12
    Ruzicka K,Veitl M,Thalhammer-Scherrer R,Schwarzinger I. The new hematology analyzer Sysmex XE-2100: Performance evaluation of a novel white blood cell differential technology. Arch Pathol Lab Med 2001; 125: 391396.
  • 13
    Bligh EG,Dyer WJ. A rapid method of total lipid extraction and purification. Can J Biochem Physiol 1959; 37: 911917.
  • 14
    Anderson RE,Standefer JC,Scaletti JV. The phospholipid and glycoprotein composition of T and B cells. Lab Invest 1977; 37: 329338.
  • 15
    Lingwood D,Simons K. Detergent resistance as a tool in membrane research. Nat Protoc 2007; 2: 21592165.
  • 16
    Holowka D. Biomembrane liquid-liquid phase separation and detergent resistance: A relationship strengthened. Biochem J 2009; 424: e5e6.
  • 17
    Schuck S,Honsho M,Ekroos K,Shevchenko A,Simons K. Resistance of cell membranes to different detergents. Proc Natl Acad Sci USA 2003; 100: 57955800.
  • 18
    Mielenz D,Vettermann C,Hampel M,Lang C,Avramidou A,Karas M,Jack HM. Lipid rafts associate with intracellular B cell receptors and exhibit a B cell stage-specific protein composition. J Immunol 2005; 174: 35083517.
  • 19
    Kabouridis PS. Lipid rafts in T cell receptor signalling. Mol Membr Biol 2006; 23: 4957.
  • 20
    de Wet B,Harder T. Are rafts involved in T-cell receptor signalling? Introduction to the talking point on the involvement of lipid rafts in T-cell activation. EMBO Rep 2008; 9: 523524.
  • 21
    Jin ZX,Huang CR,Dong L,Goda S,Kawanami T,Sawaki T,Sakai T,Tong XP,Masaki Y,Fukushima T, et al. Impaired TCR signaling through dysfunction of lipid rafts in sphingomyelin synthase 1 (SMS1)-knockdown T cells. Int Immunol 2008; 20: 14271437.
  • 22
    Gombos I,Bacso Z,Detre C,Nagy H,Goda K,Andrasfalvy M,Szabo G,Matko J. Cholesterol sensitivity of detergent resistance: A rapid flow cytometric test for detecting constitutive or induced raft association of membrane proteins. Cytometry A 2004; 61A: 117126.
  • 23
    Affentranger S,Martinelli S,Hahn J,Rossy J,Niggli V. Dynamic reorganization of flotillins in chemokine-stimulated human T-lymphocytes. BMC Cell Biol 2011; 12: 28.
  • 24
    El Kirat K,Morandat S. Cholesterol modulation of membrane resistance to Triton X-100 explored by atomic force microscopy. Biochim Biophys Acta 2007; 1768: 23002309.

Supporting Information

  1. Top of page
  2. Abstract
  3. MATERIALS ANDMETHODS
  4. RESULTS
  5. DISCUSSION
  6. LITERATURE CITED
  7. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
CYTO_22262_sm_SuppInfo.doc68KSupporting Information: MIFlowCyt
Gate_information_of_Tcell_Kono_R.tif47KSupporting Information: Tcell gate
Gate_information_of_Bcell_Kono_R.tiff48KSupporting Information: Bcell gate

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.