Les résultats présentés dans ce chapitre sont présentés sous format article scientifique. Ce dernier est soumis dans le journal « The FEBS Journal » et il est actuellement en révision. L’article scientifique s’intitule « FUNCTIONAL BKCa CHANNEL IN HUMAN RESIDENT CARDIAC STEM CELLS EXPRESSING W8B2».
1 Cadre de l’étude
Le cœur humain adulte abrite plusieurs populations de cellules souches cardiaques qui expriment des marqueurs spécifiques. Récemment, une nouvelle population positive pour le marqueur W8B2 a été identifiée. Ces cellules souches cardiaques exprimant W8B2 (CSCs W8B2+) ont une origine mésenchymateuse et présentent un fort potentiel thérapeutique (Zhang et al., 2015a). Cependant, le profil électrophysiologique de ces cellules n'a pas encore été caractérisé.
Nous nous sommes donc focalisés sur la caractérisation des canaux fonctionnels exprimés par les CSCs W8B2+ et leurs éventuels rôles dans la régulation de leurs propriétés (auto-renouvellement, prolifération et migration cellulaire) qui sont importantes pour le maintien de leur nombre et de leur mobilité pour la réparation tissulaire tout au long de la vie. Grâce à la technique du patch-clamp, nous avons montré pour la première fois la signature électrophysiologique d’un courant BKCa dans ces cellules.
Les canaux potassiques à large conductance activés par le calcium, ou BKCa, sont des canaux qui s’ouvrent suite aux changements du potentiel électrique membranaire et/ou suite à l’augmentation du calcium intracellulaire. Ces canaux se trouvent dans de nombreux tissus et participent à divers processus cellulaires (Toro et al., 2014). Les canaux BKCa sont impliqués dans la modulation de l'excitabilité vasomotrice et nerveuse en régulant le potentiel de membrane et la signalisation calcique (Nelson and Quayle, 1995). En outre, cette large conductance potassique contribue au maintien du potentiel de membrane dans les petits vaisseaux myogéniques (Waldron and Cole, 1999).
Il est intéressant de noter que ce canal joue un rôle clé dans de nombreux processus biologiques tels que le métabolisme cellulaire, la prolifération, la migration et l'expression des gènes. En plus de ses rôles physiologiques, le canal BKCa est impliqué dans plusieurs
173
pathologies comme l'obésité ou le cancer (Toro et al., 2014). Cependant, l’expression fonctionnelle et le rôle des canaux BKCa dans les cellules souches cardiaques humaines, restent peu étudiés.
Après avoir identifié le canal fonctionnel BKCa dans notre modèle de CSCs W8B2+, nous avons donc voulu étudier son rôle physiologique dans la régulation des processus fondamentaux de ces cellules souches comme l'auto-renouvellement, la prolifération et la migration.
2 Résumé des résultats
Comme nous l’avons déjà décrit précédemment chapitre I résultat (p 114) Les CSCs W8B2+ isolées par tri magnétique suivi du tri par FACS, ont une morphologie étalée et fusiforme (comme celle des fibroblastes) et possèdent des capacités d’auto-renouvellement (démontrées par le test de formation de colonies « CFU-Fs »). De plus, ces cellules proliféraient rapidement avec un temps de dédoublement d’environ 27h. L'analyse par cytométrie de flux a révélé que les CSCs W8B2+ expriment fortement les marqueurs de surface associés aux cellules souches mésenchymateuses (W8B2, CD29, CD73 et CD105) et sont négatives pour les marqueurs hématopoïétiques (CD34, CD45 et CD133). Au niveau génique, les CSCs W8B2+ ont montré une forte expression des facteurs de transcription cardiaque spécifiques précoces comme GATA4 et MEF2C mais pas Nkx2.5. De plus, ces cellules expriment les transcrits codant pour la connexine 43 (isoforme cardiaque) mais pas les transcrits codant pour les structures contractiles cardiaques (ACTC1, TNNT2, b-MHC).
Les enregistrements de courants obtenus par la technique de patch clamp (en configuration cellule entière), ont révélé la présence d’un courant potassique de type BKCa
bloqué par la paxilline. L'identité moléculaire du canal BKCa a été confirmée au niveau transcriptionnel par RT-PCR et au niveau protéique par Western Blot dans les CSCs W8B2+. Pour évaluer le rôle du canal BKCa dans la prolifération cellulaire des CSCs W8B2+, la paxilline a été utilisée pour bloqueur spécifiquement ce canal. La paxilline (utilisée à 3 μM et 10 μM) a progressivement inhibé la prolifération à partir du 4ème jour en culture. D’autre part, l'analyse du cycle cellulaire par cytométrie de flux, a montré que la paxilline (3 μM et particulièrement à 10 µM) a significativement augmenté la fraction de la population cellulaire en phase G1 et considérablement diminué la fraction de la population cellulaire en phase S/G2. L’inhibition du canal BKCa a affecté l'auto-renouvellement des CSCs W8B2+. En effet, la paxilline (utilisée à
174
1, 3 et 10 μM) a considérablement réduit le nombre de colonies CFU-Fs après dix jours de culture.
Concernant la migration cellulaire évaluée par le test de blessure (wound healing), les résultats ont montré que le blocage du canal BKCa par la paxilline ne réprime pas les capacités migratoires des cellules.
L’ensemble de ces résultats démontre que le canal BKCa est probablement un régulateur clé de la prolifération et de l'auto-renouvellement des CSCs W8B2+ et fournit une base pour une meilleure compréhension de la physiologie de ces cellules. De nouvelles recherches, basées sur la relation entre le potentiel de membrane et l’activité calcique via les canaux BKCa, semblent primordiales pour ouvrir de nouvelles voies thérapeutiques.
3 Article scientifique: « FUNCTIONAL BKCa CHANNEL IN HUMAN RESIDENT CARDIAC STEM CELLS EXPRESSING W8B2».
175
FUNCTIONAL BKCa CHANNEL IN HUMAN RESIDENT CARDIAC STEM CELLS EXPRESSING
W8B2
Ayad O , Magaud C, Sebille S, Bescond J, Cognard C, Faivre JF, Bois P* and A Chatelier*
Equipe Transferts Ioniques et Rythmicité Cardiaque, Laboratoire Signalisation et Transports Ioniques Membranaires, ERL CNRS 7368, EA 7349, Université de Poitiers
*These authors have contributed equally to this work
ABSTRACT
Recently, a new population of resident cardiac stem cells positive for W8B2 marker has been identified. These cardiac stem cells (CSCs) are considered an ideal cellular source to repair myocardial damage after infarction. However, the electrophysiological profile of these cells has not been characterized yet. We first establish the conditions of isolation and expansion of W8B2+ CSCs from human heart biopsies using magnetic sorting system followed by flow cytometry cell sorting. These cells display a spindle-shaped morphology, are highly proliferative and possess self-renewal capacity demonstrated by their ability to form colonies. Besides, W8B2+ CSCs are positive for mesenchymal markers but negative for hematopoietic and endothelial ones. RT-qPCR and immunostaining experiments show that W8B2+ CSCs express some early specific transcription factors but lack the expression of cardiac-specific structural genes. Using patch-clamp in whole cell configuration, we show for the first time the electrophysiological signature of BKCa current in these cells. Accordingly, RT-PCR and western blotting analysis confirm the presence of BKCa at both mRNA and protein levels in W8B2+ CSCs. Interestingly, BKCa channel inhibition by paxilline decreases cell proliferation in a concentration-dependent manner and cell cycling progression by accumulating the cells at G0/G1 phase. The inhibition of BKCa also decreased self-renewal capacity but did not affect W8B2+ CSCs migration. Taken together, our results are consistent with an important role of BKCa channels on cell cycle progression and self-renewal in human cardiac stem cells.
Key words: W8B2+ CSCs, BKCa channel, cell proliferation, cell cycle; migration; patch-clamp; self-renewal.
INTRODUCTION
The adult human heart harbors several populations of cardiac stem/progenitor cells that express specific markers. Recently a new population was identified which is positive for W8B2 marker (also called mesenchymal stem cell antigen-1) and shows high expression of mesenchymal but not hematopoietic nor endothelial markers. W8B2 positive cardiac stem cells (W8B2+ CSCs) exhibit a strong therapeutic potential when transplanted into a chronic myocardial infarction rat model [1]. Self-renewal, proliferation and migration are important properties that allow the perpetuation of progenitor cells and their mobility for repairing processes during life. However, the mechanisms that control these properties still have to be characterized in W8B2+ CSC.
176
High-conductance calcium-activated potassium channels (BKCa) are found in many tissues and participate in variety of cellular processes [2]. These channels are gated open by both binding of intracellular calcium and membrane depolarization. BKCa channels are involved in the modulation of vasomotor and nerve excitability by regulating membrane potential and calcium signaling [3]. Moreover, this large potassium conductance contributes to the maintenance of the membrane potential in small myogenic vessels [3-5]. Interestingly, increasing evidence indicates that the channel plays a key role in numerous biological processes such as cell metabolism, proliferation, migration, and gene expression. It strongly impacts physiological cell functions, for example in human primary skeletal myoblast [6] or in pathologies like obesity or brain, prostate, and mammary cancers [2]. However, whether these high-conductance calcium-activated potassium channels are functionally expressed in human cardiac progenitor cells remains poorly understood. Although human cardiac W8B2+
progenitor cells have been well characterized, the physiological role of BKCa channel in these cells type has never been reported.
In the present study, we identify BKCa channel current expressed in human cardiac W8B2+
progenitor cells isolated from human atrial appendage. We also investigate the role of BKCa on proliferation, migration and self-renewal processes in these progenitor cells.
MATERIALS AND METHODS Isolation of W8B2+ CSCs
Human right atrial specimens were obtained from 10 adult patients (mean age 72.7 ± 3.2 years, 9 males and 1 female). The right human atrium samples are operational waste resulting from the implementation of extracorporeal cardiac surgery, such as coronary artery bypass surgery, and are obtained in cooperation with the University hospital of Poitiers. All procedures were carried out in accordance with the Declaration of Helsinki. Freshly harvested specimens were rinsed with PBS and manually minced into 1–2 mm3 fragments and subjected to enzymatic digestion with collagenase A (1 mg/mL, Sigma-Aldrich) for 20 minutes at 37°C. The tissue fragments were plated onto fibronectin-coated dishes (10 μg/mL, Roche Diagnostics) and cultured in explant medium containing IMDM medium (Lonza) supplemented with 20% fetal bovine serum (Biowest), 0.1 mM β-mercaptoethanol (Sigma-Aldrich), 2 mM L-glutamine, 100 μg/mL streptomycin (Sigma-Aldrich), 100 μg/mL penicillin (Sigma-Aldrich) and 0.25 μg/mL amphotericin B (Sigma-Aldrich), at 37 °C and 5% CO2.
Magnetic-activated cell sorting
After 2–3 weeks, monolayers of adherent cells outgrowth from the adherent tissue fragments were collected using enzymatic digestion with Accumax® (Sigma-Aldrich) and partially enriched for W8B2 marker by magnetic Cell Sorting System (Miltenyi Biotec, Bergisch Gladbach, Germany) using W8B2 antibody. The cells were incubated for 15 minutes at 4-8°C in PBS-BSA 0.5% -2 mM EDTA buffer containing FCR Blocking Reagent® (Miltenyi Biotec, Bergisch Gladbach, Germany), mouse anti-MSCA-1 monoclonal antibody coupled to magnetic micro-beads (Miltenyi Biotec, Bergisch Gladbach, Germany). After being washed, the cells were passed through a 40 μm porosity nylon cell sieve (Corning) placed into a MACS cell separation column (Miltenyi Biotec, Bergisch Gladbach, Germany) positioned on a magnet.
177
Inside the column, the retained cells were washed with PBS-BSA buffer 0.5% - 2 mM EDTA and then harvested in this same buffer by separating the column from the magnet. The positive fraction (W8B2+ cells) were seeded in growth medium containing 25% EGM-2 (Lonza) and 75% M199 (Lonza), supplemented with 10% fetal bovine serum (Biowest), 100 μg/mL streptomycin (Sigma-Aldrich) and 0.25 μg/mL amphotericin B (Sigma-Aldrich), 0.1 mM nonessential amino acids, 100 U/mL penicillin, cultured at 37°C 5% CO2.
Flow cytometry cell sorting
After magnetic cell sorting, confluent cells were labeled with W8B2 PE-conjugated antibody and sorted using a FACS Aria Flow cytometer and sorter. Cells were first rinsed with PBS and then harvested by Accumax® (Sigma-Aldrich) enzymatic digestion. The cells were taken up in PBS buffer - BSA 0.5% - 2 mM EDTA and labeled with the isotypic antibody coupled to phycoerythrin (1:11, Miltenyi Biotec, Bergisch Gladbach, Germany) or MSCA-1 (W8B2) antibody coupled to phycoerythrin (1:11, Miltenyi Biotec, Bergisch Gladbach, Germany) for 10 min, in the dark, at 2-8 ° C. After labelling, the cells were washed with PBS buffer - 0.5% BSA - 2 mM EDTA and sorted by a flow cytometer (FACS Aria III, BD). After cell sorting, the collected W8B2+ cells were seeded into flasks with growth medium.
Cell surface marker analysis
The cell surface markers of sorted human W8B2+ CSCs were analyzed with flow cytometry. The cells were rinsed twice with pre-cooled PBS and re-suspended in cold PBS containing specific human antibodies from Miltenyi Biotec, Bergisch Gladbach, Germany: Mouse IgG1 CD117-APC (1:11), Mouse IgG2a CD45-VioBlue (1:11), Mouse IgG1 CD133/1-VioBright FITC (1:11), Mouse IgG1 CD105-PE(1:11), Mouse IgG2a CD34-PerCP-Vio700 (1:11), Mouse IgG1 CD90-PE-Vio770(1:11), Rec Human IgG1 (S) CD106-FITC (1:11), Mouse IgG1κ CD29-PE-Vio770 (1:11), Mouse IgG1 CD73-APC (1:11). For each antibody, isotype control antibodies were used: Mouse IgG2a-VioBlue (1:11), REA Control (S)-FITC (1:11), Mouse IgG1-VioBright FITC (1:11), Mouse IgG1-PE (1:11), Mouse IgG2a-PerCP-Vio700 (1:11), Mouse IgG1-PE-Vio770 (1:11), Mouse IgG1-APC (1:11). For flow cytometry florescence compensation, MACS Comp Bead Kit, mouse Igk (Miltenyi Biotec, Bergisch Gladbach, Germany) and MACS Comp Bead Kit, anti-REA (Miltenyi Biotec, Bergisch Gladbach, Germany) were used.
The cells were incubated at 4 °C in the dark for 10 min, then washed with PBS, and re-suspended for flow cytometry analysis. The percentage of positive cells was analyzed with FACS Aria Flow cytometer based on over 20000 events.
Colony formation assay (colony-forming unit fibroblasts)
W8B2+ CSCs were seeded at a density of 5 cells/cm² in a 6-well plate, in growth medium. The growth medium was renewed every 3-4 days. After 10 days of culture, the cells were rinsed with PBS, fixed for 10 min with 4% PFA and finally stained with 0.5% Crystal violet for 30 min (Sigma-Aldrich). The colonies were photographed under a phase contrast microscope and the colonies were counted.
178
Cells were plated at a density of 133 cells/cm2 in 35mm dishes with growth medium for 9 days. Cells were then synchronized to G0/G1 phase with a culture medium containing 1% FBS for 12 h. Every day, cell number is counted for each condition (control vs paxilline treated) using Malassez cell chamber.
Cell cycle analysis with propidium iodide incorporation
Cells were plated and cultured in 35 mm cell culture dishes at a density of 5×102 cells/cm2. Cells were then synchronized to G0/G1 phase with a culture medium containing 1% FBS for 12 h. Briefly, the cells were cultured in normal culture medium with ion channel blocker paxilline for 9 days. The cells were lifted using 0.125% trypsin, washed with PBS, and fixed with 70% ethanol for 30 minutes, cells were incubated with the staining solution (0.1% triton, 50µg/mL of RNase A (Sigma-Aldrich) and 5 µg/mL of propidium iodide (Sigma-Aldrich) in PBS in the dark for 90 minutes. Data were acquired using FACS VERSE cytometer (BD) and the percentage of cell cycle phases was determined using and FLOWJO software (Flowjo LLC).
Immunostaining
W8B2+ cells were seeded on gelatin-fibronectin coated glass slide. Cells were then fixed, permeabilized and incubated overnight with primary antibodies directed against connexin-43 (1:500, mouse monoclonal antibody, BD Transduction Laboratories, San Jose, CA), GATA4 (1:250, mouse monoclonal antibody, Santa Cruz Biotech,Texas,U.S.A.) and Nkx2.5 (1:250, goat polyclonal antibody, R & D Systems Minneapolis,U.S.A). Cells were washed three times with PBS and incubated for 2 h with the secondary antibody against connexin-43 (1:250, goat anti-mouse alexa 555, Invitrogen,California , U.S.A), GATA-4 (1:250, goat anti-mouse alexa 555, Invitrogen,California , U.S.A) and NKX2.5 (1:250, donkey anti-goat, alexa 555, Invitrogen,California ,U.S.A). TOPRO (1:1000, Invitrogen,California ,U.S.A) was used for nuclei labelling. Finally, the slides were slide-mounted in Mowiol® (Sigma-Aldrich) and visualized using confocal microscopy.
Patch clamp
Cells were seeded at a density of 2.103 cells/cm2 in 35mm diameter culture dish 24 h prior to the patch clamp experiments. The ionic currents were recorded on human W8B2+ CSCs in the whole-cell configuration using an Axopatch 200A amplifier with a CV 202AU headstage (Molecular Devices, CA, USA). Voltage clamp were generated by a personal computer equipped with an analog-digital converter (Digidata 1322a, Molecular Devices) using pClamp software v10 (Molecular Devices). During recordings, cells were superfused with an extracellular solution containing (in mM) : 136 NaCl, 5 KCI, 1 MgCl2, 1.8 CaCl2, 0.3 NaH2PO4, 10 glucose , 10 HEPES, pH 7.3. Membrane ion currents were recorded via glass electrodes of borosilicate GC150-T manufactured on a vertical twin heating puller. Pipette resistance was between 2 and 3 MΩ when filled with the intrapipette solution which contained (in mM): 20 KCl, 110 K-asp, 1 MgCl2, 10 HEPES, 2 CaCl2, 2.5 EGTA, 0.1 GTP, 5 Na2-phosphocreatine, 5 Mg-ATP, pH 7.2 (PCa = 6.3). The currents were recorded in control (with 0.1% DMSO as vehicle) or in paxilline (10 μM) conditions. Analyses were performed using Clampfit 10 software. Cell
179
capacitance was measured by integrating the area under the capacitive transient elicited by 5 mV depolarizing steps from holding potential of 0 mV.
Reverse transcription and polymerase chain reaction
Total RNA from W8B2+ cells was isolated using Rnable reagent (Eurobio) followed by chloroform extraction and isopropanol precipitation. RNA integrity was evaluated by ethydium bromide staining on a 1% agarose gel. Total RNA was quantified by assessing optical density at 260 and 280 nm (NanoDrop ND-100 Labtech France). cDNA was synthesized using the Pd(N)6 random hexamer primer (Invitrogen). 10 μL of total RNA (1 µg) was added to 12 μL of reaction mixture (100 mM Tris–HCl pH 8.3, 150 mM KCl, 6.25 mM MgCl2, 20 mM DTT, 2 mM dNTPs (Invitrogen) and 1.5 μg Random Primer Pd(N)6 (Invitrogen). RNA were denatured at 65 °C during 2 min and then added to 40 U RNAse inhibitor (RNaseOUT-Invitrogen) and 400 U M-MLV Reverse Transcriptase (Invitrogen) to 25 μL final volume. cDNA was synthesized at 37 °C for 1h and then added with 25 μL sterile water. Remaining enzymes were heat-deactivated (100 °C, 2 min). After the RT procedure, 10 μL of cDNA (≈150ng) was added to 40 μL of PCR reaction mixture (22 mM Tris–HCl pH 8.4, 55 mM KCl, 2.2 mM MgCl2, 277.8 µM dNTPs, 12 pmol forward and reverse primers and 1.25 U of Taq Polymerase (Invitrogen). The PCR reaction was performed with a thermocycler (denaturation at 95 °C for 5 min, followed by 38 cycles at 57 °C for 30 s, 72 °C for 30 s, and 95 °C for 30 s. After the last cycle, samples were incubated at 57 °C for 2 min and at 72 °C for 10 min to ensure complete product extension. All primers are described in table 1. GAPDH cDNA was used as housekeeping. cDNA extracted from cardiac atrial tissue was used as a positive control and the negative control consisted of the PCR reaction without the addition of the template. Amplified products were separated by electrophoresis on 2% agarose gels (containing 0.01% ethidium bromide) in Tris-Acetate-EDTA buffer and visualised using a UV Transilluminator.
Quantitative RT-qPCR
After reverse transcription the quantitative PCR reaction was carried out on a 96 well- plate, with the sense and antisense primers allowing the amplification of GAPDH (used as reference gene), GATA4, MEF2C, ACTC1, TNNT2, and β-MHC (all primers are described in table 1). Each well contained a 25 mM primer pair (sense and anti-sense), cDNA at 100 ng/well final , as well as Power SYBR® Green PCR reagent Master Mix (Thermo Fisher Scientific) containing AmpliTaq Gold® DNA Polymerase. Each sample was polymerized in triplicate to best estimate the threshold cycle of fluorescence appearance and water was used as a negative control. The amplification reaction begins with initial denaturation of the cDNA for 10 min at 95 °C, allowing the activation of the DNA polymerase, and is followed by 40 successive cycles consisting of a denaturation step for 15 sec at 95 °C followed by a step of hybridization of the primers and elongation for 1 min at 60 ° C. The results were obtained and analyzed by the 7500 Fast Real-Time PCR System (Applied Biosystems USA).
Western blotting analysis
To determine KCa1.1 protein expression, cultured W8B2+ cells were washed with cold phosphate-buffered saline (PBS) and lysed by scraping the cells into a
180
radioimmunoprecipitation assay buffer (in mM: Tris-HCl 50 pH:8, NaCl 150, EDTA 5, 0.05% Igepal, 1% deoxycholic acid, 1% Triton X-100, 0.1% SDS) containing protease and phosphatase inhibitors (Protease Inhibitor Cocktail, Sigma-Aldrich - PhosSTOP Phosphatase Inhibitor Cocktail, Roche).
Soluble cell lysates were denatured 5 min at 37°C in 2× Laemmli sample buffer (in mM): Tris-HCl 125 pH 6.8, SDS 4%, glycerol 20%, bromophenol blue 0.004%, β-mercaptoethanol 10%). Protein samples (60 and 80 μg), obtained from cultured W8B2+ cells, were separated by SDS-PAGE using 8% polyacrylamide gels and transferred to nitrocellulose membranes. Membranes were blocked 90 min in TBS-Tween blocking solution (in mM: Tris 20 pH : 7.6, NaCl 150 and Tween-20 0.2%) with 5% non-fat milk at room temperature. Blots were then incubated overnight at 4°C with primary antibodies diluted in TBST-5% nonfat dry milk. We used polyclonal anti-rabbit KCa1.1 (1:500, Alomone Labs, Jerusalem, Israel). Vimentine was probed by rabbit polyclonal anti-vimentine (1:500, Santa Cruz Biotechnology, Dallas, U.S.A). Membranes were washed with TBS-Tween three times for 10 min and then incubated for 90 min at room temperature with specific anti-rabbit horseradish peroxidase-conjugated secondary antibodies (1:5000, Interchim, Montluçon, France). Membranes were revealed with enhanced chemiluminescence (ECL) chemiluminescent substrate (GE Healthcare, Velizy-Villacoublay, France). The results were analyzed by using the GeneGnome Imager (SynGene Ozyme, Montigny - le - Bretonneux, France).
Cell mobility determination
Cell migration was determined using a wound healing method to investigate the potential effect of BKCa channels on cell mobility in human W8B2+ CSCs. The wound healing assay was conducted when the cells grew to total confluence in 6-well plates in 0% FBS culture medium. A standard wound was created by scratching the cell monolayer with a sterile 200 μL plastic pipette tip. After removing cell fragments by washing cell monolayer gently with PBS, the cells were incubated at 37°C with the medium containing 0% FBS and 10 µM paxilline for 7 h. The cells were then fixed and nuclei were labelled with DAPI. Transmission images were taken just after the wound (time 0h) and after 7h of paxilline treatment. The images were generated with a spinning disk confocal station (Revolution, Andor) equipped with a high precision motorized XY stage (Marzhauser) and IQ3 software (Revolution, Andor) that allowed to memorize positions of each sample. The number of migrated cells on the images were counted (based on the nuclei labelling) to assess cell mobility under different conditions of treatments. Statistical analysis
Results were expressed as mean ± SEM. All statistical analysis were performed using GraphPad Prism (La Jolla, CA, USA). The tests used for the different assays are provided in the figure