1
Stimulated with anti-IgM, the activity based probe TB474 can be
located with immunofluorescence in Daudi B-cells in vivo and might
visualise the enzyme activity of alfa-galactosidase.
Pol van de, K., 10-04-2020
Abstract
Fabry’s disease is an X-linked lysosomal storage disease in which the activity of the enzyme alfa galactosidase A (α-GAL) is deficient due to a specific mutation in the GLA-gene. In Fabry’s disease, the glycosphingolipid gb3 is deacetylated by acid ceramidase, which can be visualised in cells using immunofluorescence. For the enzyme deficient in Fabry’s disease, alfa galactosidase A, an activity based probe is designed to locate the enzyme activity. However, it is unclear if it is possible to locate the enzyme activity of alfa galactosidase in B-cells in vivo, using the activity based probe TB474. Stimulation with IgM and migalastat could promote the endocytosis of TB474 in b-cells in vivo. It is also debated where acid ceramidase is located, since acid ceramidase consists of two subunits: the alfa and beta subunit. In this project,
research is focused on the location of acid ceramidase (ASAH1) in liver tissue, kidney tissue, fibroblasts and podocytes under influence of an alfa-galactosidase inhibitor and an acid ceramidase inhibitor. Research is also focused on the location of GLA and gb3 in B-cells and on the localisation of the enzyme activity of alfa galactosidase, using TB474. Results indicate that the beta subunit of ASAH1 is always colocalised with the alfa subunit of ASAH1 in what seems to be lysosomes, but the alfa subunit is not always colocalised with the beta subunit of ASAH1. Furthermore, with IgM-stimulation it is possible to locate the activity based probe TB474 with
immunofluorescence, although it is unclear if TB474 has truly bound with the enzymatic active part of alfa galactosidase. In conclusion, the activity based probe TB474 can be found with immunofluorescence in intact B-cells in vivo, using IgM stimulation.
Keywords
Acid ceramidase (ASAH1) – Activity based probe –– B-cells – Carmofur - Cytospin – Daudi – DNJ - Endocytosis - ERT - Fabry’s disease - Fibroblasts – Gb3 – GLA –
Glycosphingolipids - IgM stimulation – Immunofluorescence - Migalastat – Podocytes – TB474
2
Introduction
Fabry’s disease is an X-linked lysosomal storage disease in which the activity of the enzyme alfa galactosidase A (α-GAL ) is deficient due to a specific mutation in the GLA-gene (El-Abassi et al., 2014). Normally, α-GAL converts globotriaocylceramide (gb3) into lactosylceramide. Gb3, α-GAL and lactosylceramide are part of the
sphingolipid pathway (Turpin-Nolan & Brüning, 2020). The clinical manifestations of Fabry’s disease are cardiovascular, cerebral and renal defects (Branton et al., 2002). Fabry’s disease is treated with enzyme replacement therapy (ERT), such as
Fabrazyme® (Baigent & Barngrover, 2002). Together with Pompe disease, Fabry’s disease is known for its expensive medication. Even though the medication is
expensive, ERT does not correlate with a higher quality of life in patients with Fabry disease (Wyatt et al., 2012). Since it is debated whether medication for Fabry’s disease needs to be reimbursed or not, it is indispensable to investigate the mechanisms of this disease. Research about the molecular principles of Fabry’s disease can maybe contribute to understand these mechanisms.
Deacetylated forms of glycosphingolipids are responsible for a lot of lysosomal glycolipid storage diseases such as Gaucher disease and Krabbe disease (Aerts et al., 2008). This suggest that lyso-gb3, the deacylated form of gb3, could be responsible for the clinical symptoms of Fabry’s disease. Researchers indeed found elevated plasma-concentrations of lyso-gb3 in patients suffering from Fabry’s disease (Aerts et al., 2008). This could suggest that gb3 is converted into lyso-gb3 when alfa-galactosidase is deficient, see figure 1 (Ferraz et al., 2014). They also found that the accumulation of gb3 does not correlate with the clinical symptoms of the disease and that lyso-gb3 is a more accurate biomarker for the course of the disease (Ferraz et al., 2016). The
enzyme responsible for the conversion of gb3 to lyso-gb3 is ceramidase. The human acid ceramidase enzyme (ASAH1) is known for converting ceramides into
sphingosines and it is known for its ability to autocleave and for consisting out of two subunits, namely the alfa-subunit and the beta-subunit, see figure 2 (Gebai et al., 2018).
3 Figure 1: Normal deglycosylation of globotriaosylceramide to lactosylceramide by alfa galactosidase A (straight arrow down) and alternative deacylation in alfa-galactosidase A deficient cells to
globotriaosylsphingosine (Ferraz et al., 2014)
Figure 2: Structural overview of aCDase. A Domain organization of aCDase (SP, signal peptide). B Left, structure of the inactive form of recombinant aCDase from the naked mole rat, colored as shown in a. The location of the disulphide bridge linking the alfa and beta-subunits is indicated by a red arrow. The glycans produced from insect cell expression are displayed as white sticks. Right, structure of aCDase in a different view with secondary structure elements labelled (Gebai et al., 2018).
4 However, little is known about the function of ceramidase and the role of ceramidase in Fabry’s disease. Also, the activity based probe (ABP) for alfa galactosidase is not used in intact cells in vivo. Researchers often measure the enzyme activity of alfa-galactosidase in lysate of cells with an enzyme activity assay (EEA). They can also determine the location of alfa-galactosidase with anti-GLA. However, the location of enzyme activity of alfa-galactosidase in cells in vivo is not determined yet. The activity based probe TB474 is designed to show the location of the enzyme activity of alfa galactosidase in intact cells in vivo. (Kuo et al., 2018). However, it looks like TB474 does not enter cells at his own. To facilitate the endocytosis of TB474, migalastat could be added to the activity based probe. Migalastat is a
pharmacological chaperone for alfa-galactosidase, which can facilitate the transport of alfa galactosidase between inner cell compounds (McCafferty & Scott, 2019). It could be that TB474 can also be transported from the cell membrane towards the inner cell compounds. Another possible molecule that could facilitate the endocytosis of TB474 is anti-IgM, since B-cells form a specialized immune synapse architecture, in which small peripheral clusters are formed (Saez et al., 2019). Maybe these small peripheral clusters can facilitate the endocytosis of TB474.
Research towards acid ceramidase and using a probe that reveals the location of the enzyme activity of alfa galactosidase in cells can give insight in the glycosphingolipid metabolism.
In this project, research is focused on the location of both subunits of acid ceramidase (ASAH1) in tissue and cells under influence of an alfa-galactosidase inhibitor and an acid ceramidase inhibitor. Research is also focused on testing the activity based probe TB474 to locate the enzyme activity of alfa galactosidase (α-GAL) in B-cells in vivo. The localisation of alfa galactosidase with anti-GLA and the localisation of gb3 with anti-CD77 is also determined in B-cells in vivo.
First, the location of the alfa and beta subunit of acid ceramidase (ASAH1) is visualised in kidney and liver tissue of healthy subjects and in wildtype fibroblasts and podocytes using antibodies against the alfa and beta subunit of acid ceramidase. The location of acid ceramidase is compared with anti-EEA1, an antibody that stains early endosomes and alfa ceramidase is also compared with anti-LAMP1, an
antibody that stains lysosomes. In this way, it can made clear if acid ceramidase is located into the lysosomes or that it moves through the cell with early endosome vesicles. It is also tested whether an alfa-galactosidase inhibitor (DNJ) and an acid ceramidase inhibitor (carmofur) can alter the localisation of acid ceramidase in fibroblasts and podocytes. After that, the activity based probe TB474 is tested to locate the enzyme activity of alfa galactosidase in B-cells in vivo.
5 B-cells are used here for being easy to culture and for forming specialised immune synapses when stimulated with anti-IgM. DAUDI B-cells do express IgM on their surface (Loftin et al., 1985), which is probably necessary for anti-IgM stimulation. It is investigated if migalastat or anti-IgM can facilitate the endocytosis of TB474. To check whether the activity based probe TB474 has bound to alfa galactosidase, the location of the probe is compared with the location of gb3, the substrate of alfa galactosidase. The location of the probe is also compared with anti-GLA, to give an idea of which part of alfa galactosidase in a B-cell is enzymatic active.
It is expected that the alfa and beta subunit are located at the same places in the cell, namely in the lysosomes. It is therefore expected that acid ceramidase colocalises with anti-LAMP1, a marker for lysosomes. It is expected that is does not correlate with anti-EEA1, a marker for early endosomes, since it is expected that acid ceramidase does not move through the cell with early endosome vesicles. It could be that the inhibition of alfa galactosidase with DNJ causes less alfa
galactosidase, which could lead to more acid ceramidase. In this case, there is more acid ceramidase visible with anti-ASAH1. However, it could also be that the
inhibition of alfa galactosidase with DNJ causes an overexpression of alfa
galactosidase, which could lead to less expression of acid ceramidase. In this case, there is less acid ceramidase visible with anti-ASAH1. It could also be that the inhibition of alfa galactosidase with DNJ has no effect on the amount of alfa
galactosidase. In this case, no difference in amount of acid ceramidase can be found with anti-ASAH1. The same can be said about carmofur as an inhibitor of
ceramidase. It could be that carmofur causes less acid ceramidase, but it can also (compensatory) cause more of acid ceramidase. It could also be that no effect is visible. No hypothesis about the effect of DNJ and carmofur on the intensity of the staining of acid ceramidase with anti-ASAH1 can be formed.
For the activity based probe TB474 it is expected that TB474 cannot be found with fluorescence microscopy when it is added to cells in vivo on its own. It is expected that both migalastat and anti-IgM can facilitate the endocytosis of TB474 and that the probe TB474 is visible with fluorescence microscopy.
6
Material and Methods
Localisation of acid ceramidase in fibroblasts, podocytes, kidney tissue and liver tissue using fluorescence with antibodies.
Cell lines and tissue material
The following cell lines and tissue materials are used: human fibroblasts wildtype CD9/20, human podocytes wildtype AB8/13, kidney tissue of a healthy subject and liver tissue of a healthy subject from the archive of the department of pathology at the Universitair Medische Centra (UMC) location AMC in Amsterdam.
Cell culture
For this experiment, wildtype podocytes and fibroblasts were thawed and mixed with medium and added to a T75 culture bottle. Medium for fibroblasts consisted of 500 mL DMEM (ThermoFisher® 11965092), 45 mL Foetal Bovine Serum (FBS,
Sigma® F7524), 15 mL Penicillin/Streptomicin (PS, Sigma® P4333) and 15 mL
Glutamate (G, ThermoFisher® 25030081). Medium for podocytes consisted of 500 mL RPMI (ThermoFisher® 11875085), 5 mL Insulin-Transferrrin-Selenium (ITS,
Invitrogen ® 41400045), 45 mL Foetal Bovine Serum (FBS, Sigma® F7524) and 15 mL Penicillin/Streptomicin (PS, Sigma® P4333). All information about cell culturing can be found in the culture protocol in the lab journal.
Cell preparation and antibodies
After counting with a Bürker counting chamber, 500 cells were added to 4 coverslips. DNJ (10 microM), a alfa galactosidase inhibitor, is added to one coverslip. Carmofur (100 microM), an acid ceramidase inhibitor, is added to another coverslip. Both drugs are added to the last coverslip. One coverslip did not contain any drug (control). Paraformaldehyde in PBS (4%) was added to 4 coverslips with cells for 20 minutes. NaN3 (1/1000) was added to the cells for 20 minutes. After that, cells were incubated with saponin 0.05% in PBS/BSA 1% for 60 minutes. After washing, cells were
incubated with 5% Normal Goat Serum (NGS) in PBS/Tween20 (PBS/T20) for 20 minutes. The antibodies were added to the coverslips, following table 1.
Hoechst33342 is added to every coverslip. The coverslips were fixated on slides with Aquamount® Polyscience and analysed with fluorescence microscopy.
7
Coverslip Primary antibodies Secondary antibodies
1 Anti-ASAH1-MIgG1 (1/250) Anti-ASAH1-RbtIgG (1/500) Gt-anti-MIgG1-A594 (1/200) Gt-anti-RbtIgG-A488 (1/200) 2 Anti-ASAH1-MIgG1 (1/250) Anti-EEA1-RbtIgG (1/500) Gt-anti-MIgG1-A594 (1/200) Gt-anti-RbtIgG-A488 (1/200) 3 Anti-ASAH1-RbtIgG (1/500) Anti-LAMP1-RatIgG (1/500) Gt-anti-RbtIgG-A488 (1/200) Gt-anti-RatIgG-A538 (1/200) 4 PBS/T20 Gt-anti-MIgG1-A594 (1/200) Gt-anti-RbtIgG-A488 (1/200) Table 1: Primary and secondary antibodies added fibroblasts and podocytes. ASAH1 = acid ceramidase, CD77 = gb3, GLA = alfa galactosidase. Results can be viewed in appendix B and C.
Tissue slides preparation and antibodies
Four slides were made of formalin-fixed, paraffin-embedded healthy liver tissue and four slides were made of healthy kidney tissue. Slides (n=8) were deparaffinized in xylene (3x 100%) and graded ethanol series (100%, 90%, 70%) and washed with demi-water. 10 mM Natriumcitrate buffer (pH = 6) was added to 4 slides and 10mM
Tris/EDTA (pH = 9) was added to the other 4 slides. A heat induced antigen retrieval (HIER) was done for 20 minutes and after that, slides were blocked with
PBS/Tween20 0.01% and with 0.5 microliters of 1% NGS per slide for 30 minutes. 4 slides were incubated for 120 minutes with ASAH1-MIgG1 (1/250) and anti-ASAH1-RbtIgG (1/500) and washed with PBS/T(0.01%). After that, all 8 slides were incubated for 60 minutes with gt-anti-MIgG1-A594 (1/200), gt-anti-RbtIgG-A488 (1/200) and Hoechst33342 (1/2000). Coverslips were placed on the slides and the slides were fixated with Aquamount® Polyscience and analysed with fluorescence microscopy.
Localisation of the enzyme activity of alfa galactosidase with TB474 in B-cells. Pilot Cytospin®
B-cells (Daudi) were thawed and transferred to a culture bottle and plated on a 24-wells plate. Cells were counted with a Bürker chamber and around one million cells were added to 1 millilitres PBS. 2 drops of cell suspension per Cytospin® unit was added to 4 Cytospin® units with glass slide and tunnel. Cytospin® units were centrifuged in a Cytospin® at 1200 rpm for 5 minutes. Cytospin® units were removed and the position of the cells was marked. Slides were airdried o/n. 4% paraformaldehyde in PBS was added to the cells for 20 minutes. 0,3 M Glycine in PBS/Tween20 was added to the cells for 20 minutes. Cells were incubated 60 minutes in saponin 0.05% in PSB/BSA 1%. Then 5% NGS PBS/Tween20 was added to the cells
8 for 30 minutes. Antibodies are added, following table 2. Hoechst33342 (1/2000) is added to every well. Cells are fixated using Aquamount® Polyscience.
Slide Primary antibody (dilution) Secondary antibody (dilution) 1 Anti-ASAH1-Mouse (1/125) Anti-ASAH1-Rabbit (1/250) Goat-anti-Mouse (1/200) Goat-anti-Rabbit (1/200) 2 Anti-IgM-Goat (1/1000) Donkey-anti-Goat (1/200) 3 Anti-CD77-Rat (1/250) Anti-GLA-Rabbit (1/250) Goat-anti-Rat (1/200) Goat-anti-Rabbit (1/200) 4 - Goat-anti-Mouse (1/200) Goat-anti-Rabbit (1/200) Goat-anti-Rat (1/200)
Table 2: Primary and secondary antibodies added to Daudi B-cells. ASAH1 = acid ceramidase, CD77 = gb3, GLA = alfa galactosidase. Results can be viewed in figure 7. Incubating with activity based probe
Frozen B-cells (Daudi) were thawed and transferred into a culture bottle and into a 24-wells plate. 4 concentrations of the activity based probe TB474 were added to the wells: 0 microM, 10 microM, 1mM and 100mM. The duration of the incubation was 30 minutes, 1 hour, 2 hours, 4 hours, 24 hours and 48 hours. After incubation, a Cytospin® is done following the protocol described in ‘pilot Cytospin®’ and cells are fixated with Vectashield®. After that, cells are viewed with fluorescence microscopy. Incubation with activity based probe TB474 and migalastat
B-cells (Daudi) were transferred into a 24-wells plate. Four different combinations of substances are added to the wells, in this case migalastat and the activity based probe TB474 (1), migalastat (2) ABP (3) and no substance (4). The concentration migalastat added was 0.1 mM and the concentration of the activity based probe was 10nM. Cells are incubated for 6 hours. After incubation, a Cytospin® is done following the
protocol described in ‘pilot Cytospin®’ and cells are fixated. After that, cells are viewed with fluorescence microscopy.
Incubation with activity based probe TB474 and anti-IgM
B-cells are plated on a 24-wells plate. A biotin-SP-conjugated F(ab’)2 Goat anti-human IgM antibody (Jackson Immunoresearch, 1/1000), is added to every well. After that, cells are incubated for 30 minutes with the antigen binding probe TB474. Cells are diluted 1:1 with NaN3 (n=12) or with paraformaldehyde (n=12). A
Cytospin® is done, following the protocol described in ‘pilot Cytospin®’. Slides were airdried o/n.
First, slides incubated with NaN3 are fixated with paraformaldehyde in PBS (4%) for 20 minutes. Slides are fixed using Vectashield® with DAPI. Second, the slides
9 glycine in 50mL PBS/T20 for 20 minutes. Slides are incubated 60 minutes in saponin 0.05% in PSB/BSA 1% (Calbiochem #558255). This 5% NGS PBS/Tween20 solution is added to the slides for 30 minutes. Primary antibodies anti-CD77-Rat-IgM (1/200), anti-GLA-Rabbit-IgG (1/200) and Hoecst-33342 (1/2000) were added to every well. Secondary antibodies Goat-anti-Rat-IgM-BIO (1/200), Goat-anti-Rabbit-IgG-ATTO538 (1/200) and STRP-anti-BIO-FTC are added to all wells. Slides are fixated with
Vectashield®.
Incubation with activity based probe TB474 and anti-IgM at different time points
B-cells are plated onto a 24-wells plate with 0.6 mL cells per well. The plate is divided into 4 groups of 6 wells. The activity based probe with anti-IgM-FITC (1/500) is
added to the first group and the activity based probe with anti-IgM-FITC (1/1000) is added to the second group. Only anti-IgM-FITC (1/500) is added to the third group and only the activity based probe TB474 is added to the last group. Cells are fixed by diluting the cells with NaN3 (1:1) after different timepoints: 24h, 6h, 4h, 2h, 1h and 0h. From all the wells, two drops were added to a cytspin unit. Cytospin units were centrifuged in a Cytospin® at 1200 rpm for 5 minutes. Slides were airdried o/n and fixed with Vectashield®. Cells are viewed with fluorescence microscopy.
10
Results
Localisation of acid ceramidase in fibroblasts, podocytes, kidney tissue and liver tissue using fluorescence with antibodies.
Immunological staining of tissue
The location of the alfa and beta subunit of ASAH1 in human tissue and cells was investigated using immunostaining with antibodies. First, antigen retrieval was optimized by comparing the quality of the fluorescence staining of tissue incubated in an acid buffer with the quality of the fluorescence staining of tissue incubated in a base buffer. Following appendix A there is no difference in quality of the
fluorescence between an acid buffer and a base buffer.
Second, the staining of healthy liver tissue with rabbit and anti-ASAH1-mouse was compared with control healthy liver tissue without antibodies, see
appendix A. One can observe that anti-ASAH-1-rabbit (green) is located more diffuse
over the cell surface, which could reflect that it is located more into the cytoplasm than clustered in lysosomes for example. One can also observe that ASAH-1-mouse (red) is more clustered into lysosome-like dots inside the cytoplasm of cells, see figure 3 and appendix A. This can either be viewed when tissue slides were incubated in an acid buffer or be viewed when they are incubated in a base buffer. For kidney tissue, the same pattern can be seen, see figure 4 and appendix A.
Although both Anti-ASAH1-rabbit (green) and anti-ASAH1-mouse (red) are located in the cytoplasm of cells, it seems that the beta-subunit of ASAH1 (red) is always located at the same places were the alfa subunit of ASAH1 (green) is located, but the alfa-subunit is not always colocalised with the beta-subunit of ASAH1. For both liver and kidney tissue, it can be observed that the alfa subunit of ASAH1 can be found in a diffuse pattern in the cytoplasm of the cells and that the beta subunit of ASAH1 is located clustered at specific places in almost all the cells in the tissue slide. On top of that, it looks like the beta subunit of ASAH1 is always colocalised with the alfa subunit of ASAH1, but the alfa subunit is not always colocalised with the beta subunit of ASAH1.
Immunological staining of cells
The location of the alfa and beta subunit of acid ceramidase was also investigated in cells using immunostaining with antibodies. If we first look at the fibroblasts stained with anti-ASAH1-rabbit (green) for the beta subunit and anti-ASAH1-mouse (red) for the alfa subunit of acid ceramidase (ASAH), we can observe that anti-ASAH1-rabbit (green) is located in a diffuse dot-pattern for all the four conditions: control, carmofur, DNJ and DNJ + carmofur, see appendix B. In all conditions, we can
observe a striped pattern following the inner lining of the cell for anti-ASAH1-mouse (red), displayed in figure 5. If we compare the intensity of the four conditions, we
11 can observe a slightly more intense fluorescence for both anti-ASAH1-mouse (red) and anti-ASAH1-rabbit (green) when DNJ, a alfa-galactosidase inhibitor, is used. For all the conditions, the same pattern of anti-ASAH1-rabbit (green) and anti-ASAH-mouse (red) can be found in fibroblasts. If we then look at podocytes, the pattern of anti-ASAH1-rabbit (green) is the same as for fibroblasts: a clustered dot pattern, diffuse over the cytoplasm of the cell, see figure 6 and appendix C. However, for mouse (red), there is a difference in location. It seems than anti-ASAH1-mouse (red) is more located around the nucleus in a diffuse pattern more than a striped pattern at the cell-membrane. When the intensity of both antibodies is compared for the four different conditions, one can also observe a difference, see appendix C. For podocytes, there is no difference in intensity of anti-ASAH1-mouse (red) between the four conditions. However, the intensity of anti-ASAH1-rabbit (green) is increased when carmofur is used. A comparison of the intensity of the antibodies with different drugs compared to the control group is made in table 2 for fibroblasts and table 3 for podocytes.
If we look at the staining of fibroblasts with EEA1-rabbit (green) and anti-ASAH1-mouse (red), there is no difference between this staining and the control group. This is true for all the four conditions and the same can be seen in podocytes. We can conclude that the staining with EEA1 does not give us any results. If we look at the staining of fibroblasts with anti-ASAH1-rabbit (green) and anti-LAMP1-rat (red), we can observe the same dot-pattern in the cytoplasm of the cells for anti-ASAH1-rabbit (green), just like we already saw in the first slide. Anti-LAMP1-rat (red) is located around the nucleus of the cell. It cannot be found in the rest of the cytoplasm, where anti-ASAH1-rabbit (green) can be found. We also observe a lower intensity of anti-ASAH1-rabbit (green) when DNJ is added to the cells and when DNJ + carmofur is added to the cells compared with the control group and carmofur. For podocytes the same can be viewed. However, when DNJ + carmofur was added to podocytes, there was a higher intensity of anti-ASAH1-rabbit (green), whereas there was a lower intensity of anti-ASAH1-rabbit (green) when only DNJ was added. For both carmofur and DNJ + carmofur, the intensity of anti-LAMP1 (red) was lower than for the control and when DNJ was added.
If we look at the staining of fibroblasts without antibodies, no fluorescence can be found. There is no difference between the four conditions. This is also true for the podocytes.
All detailed results from the immunostaining of tissue can be found in appendix A and the results from the immunostaining of fibroblasts and podocytes can be found in respectively appendix B and appendix C.
12
Slide Antibody DNJ Carmofur DNJ + Carmofur
1 Anti-ASAH1-r Higher Lower Same
Anti-ASAH1-m Higher Lower Same
2 Anti-EEA1-r - - -
Anti-ASAH1-m - - -
3 Anti-ASAH1-r Higher Same Higher
Anti-LAMP1-r Same Same Same
4 - - - -
Table 3: Quality of the intensity of the antibodies with different drugs compared to the control group for fibroblasts. Quality of the intensity is scored with ‘higher’, ‘lower’ or ‘same’ compared to the control group (= slide 4). The figures compared in this table can be viewed in appendix B. EEA-1 = Early Endosome Antigen 1, LAMP1 = Lysosomal Associated Membrane Protein 1.
Slide Antibody DNJ Carmofur DNJ + Carmofur
1 Anti-ASAH1-r Same Higher Higher
Anti-ASAH1-m Same Lower Same
2 Anti-EEA1-r - - -
Anti-ASAH1-m - - -
3 Anti-ASAH1-r Same Same Higher
Anti-LAMP1-r Same Lower Lower
4 - - - -
Table 4: Quality of the intensity of the antibodies with different drugs compared to the control group for podocytes. Quality of the intensity is scored with ‘higher’, ‘lower’ or ‘same’ compared to the control group (= slide 4). The figures compared in this table can be viewed in appendix C. EEA-1 = Early Endosome Antigen 1, LAMP1 = Lysosomal Associated Membrane Protein 1.
13 Figure 3: Immunofluorescence of ASAH1 in healthy liver tissue prepared with an acid buffer. Antibodies used are ASAH1-rbt (green) for the beta-subunit of acid ceramidase, ASAH1-m (red) for the alfa-subunit of acid ceramidase and Hoechst33342 (H) for the nuclei. Scale bars = 50 µm.
Figure 4: Immunofluorescence of ASAH1 in healthy kidney tissue prepared with an acid buffer. Antibodies used are ASAH1-rbt (green) for the beta-subunit of acid ceramidase, ASAH1-m (red) for the alfa-subunit of acid ceramidase and Hoechst33342 (H) for the nuclei. Scale bars = 50 µm.
ASAH1-m
ASAH1-rbt ASAH1-rbt/ASAH1-m/H
ASAH1-m
14 Figure 5: Immunofluorescence of ASAH1 in human fibroblasts. Antibodies used are ASAH1-rbt (green) for the beta-subunit of acid ceramidase, ASAH1-m (red) for the alfa-subunit of acid ceramidase and Hoechst33342 (H) for the nuclei. Scale bars = 50 µm.
Figure 6: Immunofluorescence of ASAH1 in human podocytes. Antibodies used are ASAH1-rbt (green) for the beta-subunit of acid ceramidase, ASAH1-m (red) for the alfa-subunit of acid ceramidase and Hoechst33342 (H) for the nuclei. Scale bars = 50 µm.
DNJ + Carmofur
DNJ + Carmofur
ASAH1-rbt
/
ASAH1-m
/H
ASAH1-rbt ASAH1-mASAH1-rbt
/
ASAH1-m
/H
ASAH1-rbt ASAH1-m15 Localisation of the enzyme activity of alfa galactosidase with TB474 in B-cells. Pilot Cytospin®
The location of enzyme activity of alfa galactosidase in human B-cells was
investigated using immunostaining with antibodies and with TB474. The staining of B-cells with anti-ASAH1-alfa (red) and anti-ASAH1-beta (green) did not show any results. The staining of B-cells with anti-IgM shows that Daudi B-cells contain IgM receptors on their surface. The staining of B-cells with anti-CD77 (gb3) and anti-GLA shows that gb3 is located on the surface of the B-cells (green) and that α-GAL (red) is located inside the cytoplasm. The control slide is negative. The results of the pilot Cytospin® can be found in figure 7.
Incubation with activity based probe TB474
Although this experiment is done with different incubation times and with different concentrations of activity based probe TB474, all the slides showed the same result. Therefore, only two pictures are shown for all the 24 slides, see figure 8. One can see that the activity based probe cannot be found with fluorescence microscopy.
Incubation with activity based probe TB474 and migalastat
Results are shown in figure 9. Although this experiment is done with different incubation times and with different concentrations of activity based probe TB474, all the slides showed the same result. Therefore, only two pictures are shown for all the 24 slides. One can see that the activity based probe cannot be found with
fluorescence microscopy.
Incubation with activity based probe TB474 and anti-IgM
The location of enzyme activity of alfa galactosidase in human B-cells was investigated using immunostaining with TB474 and anti-IgM. It was also tested whether NaN3 or paraformaldehyde is de best substance for fixating cells before the Cytospin® is done. After using the Cytospin®, fixating with paraformaldehyde and staining with TB474 and anti-IgM, it seems that the quality of the staining was the same for both NaN3 as well as paraformaldehyde, according to figure 10. Both substances can thus be used for cell-fixation. We can see here that the activity based probe TB474 can be viewed with fluorescence microscopy. In figure 11, the result of the time-lapse of anti-IgM added to the probe TB474 can be viewed, fixated with NaN3. It can be stated that at least 2 hours of incubation with the activity based probe TB474 gives the best results. According to figure 12, IgM is located on the cell membrane. It also looks like there is clustering of IgM receptors on the surface of the cell membrane, when anti-IgM is added to the B-cells. This can best be viewed when the staining of IgM in Pilot Cytospin® is compared with the staining of IgM after adding TB474, see figure 12.
16 According to figure 13, It seems that gb3 and TB474 are colocalised. TB474 is also colocalised with anti-GLA, but anti-GLA is not always localised at places where TB474 can also be found.
Figure 7: Immunofluorescence of acid ceramidase, gb3, alfa galactosidase and IgM in DAUDI B-cells. Antibodies used are ASAH1-rbt (green) for the beta-subunit of acid ceramidase, ASAH1-m (red) for the alfa-subunit of acid ceramidase (A), anti-CD77 for gb3 and anti-GLA for alfa galactosidase (B) and anti-IgM for IgM (C). Hoechst33342 (H) is added for the nuclei (A, B, C & D). Scale bars = 50 µm. ASAH1-rbt ASAH1-rbt ASAH1-m ASAH1-m ASAH1-rbt/ASAH1-m/H H ASAH1-rbt ASAH1-m ASAH1-m ASAH1-rbt/ASAH1-m/H CD77/GLA/H IgM/H H
Human DAUDI B-cells
CD77 GLA IgM IgM A B C D
17 Figure 8: Immunofluorescence of TB474 in DAUDI B-cells (n=24). Displayed are the activity based probe TB474 (grey) and Hoechst33342 (H) for the nuclei. Scale bars = 50 µm.
Figure 9: Immunofluorescence of TB474 in DAUDI B-cells (n=24) stimulated with
migalastat. Displayed are the activity based probe TB474 (grey) and Hoechst33342 (H) for the nuclei. Scale bars = 50 µm.
Figure 10: Immunofluorescence of TB474 and IgM in DAUDI B-cells stimulated with anti-IgM, cells were fixed with NaN3 (left) or paraformaldehyde (right). Displayed are the activity based probe TB474 (grey), IgM (green) and Hoechst33342 (H) for the nuclei. Scale bars = 50 µm.
TB474/H TB474 TB474/H TB474 IgM IgM TB474 TB474 IgM/TB474/H IgM/TB474/H Paraformaldehyde NaN3
18 Figure 11: Immunofluorescence of IgM and TB474 in DAUDI B-cells at different time points (t=0, t=30m, t=1h and t=2h). Antibodies used are anti-IgM (green) and Hoechst33342 (H). The probe TB474 can be seen in grey. Scale bars = 50 µm.
Figure 12: Immunofluorescence of IgM in DAUDI B-cells with IgM stimulation (t=2h, t=1h and t=30m, upper 3 figures) compared to DAUDI B-cells without IgM stimulation (t=0, bottom 3 figures) Antibodies used are anti-IgM (green). Scale bars = 50 µm.
IgM/TB474/H IgM TB474 t = 2h t = 1h ASAH1-m ASAH1-rbt t = 30m t = 0m B-cells DAUDI IgM/TB474/H IgM TB474 IgM TB474 TB474 IgM IgM/TB474/H IgM/TB474/H IgM IgM comparison IgM IgM IgM IgM IgM t = 2h t = 1h t = 30m t = 0 t = 0 t = 0
19
Discussion
Figure 13: Immunofluorescence GLA, gb3 and TB474 in DAUDI B-cells. Antibodies used are anti-GLA (yellow), anti-CD77 (green) and Hoechst33342 (H). The probe TB474 can be seen in grey. Scale bars = 50 µm. GLA GLA GLA GLA GB3 GB3 GB3 GB3 TB474 TB474 TB474 TB474 GLA/GB3/TB474/H GLA/GB3/TB474/H GLA/GB3/TB474/H GLA/GB3/TB474/H
20
Discussion
Following the first part of the results in this project, it can be concluded that for acid ceramidase (ASAH1), there is a difference in location of the alfa subunit and the beta subunit of acid ceramidase (ASAH) in wildtype fibroblasts and podocytes. We can state that the beta subunit of ceramidase is located in the cytoplasm in a diffuse pattern, whereas the alfa subunit of ceramidase is located in a stripped pattern around the cell membrane. It can also be concluded that the beta subunit of acid ceramidase (ASAH1) always colocalised with the alfa subunit of acid ceramidase (ASAH1), but the alfa subunit is not always colocalised with the beta subunit. The association between acid ceramidase and early endosomes and lysosomes could not be determined. When DNJ, a alfa galactosidase inhibitor, is added to the cells, more acid ceramidase is visible. When acid ceramidase is blocked with carmofur, less acid ceramidase is visible. Following the second part of the results in this project, the activity based probe TB474 was not visible with immunofluorescence in B-cells when incubated alone or incubated with migalastat. Only with IgM stimulation and an incubation of at least 2 hours, a pattern of TB474 in B-cells is visible with
immunofluorescence. TB474 is colocalised with gb3 and α-GAL, which could confirm that TB474 binds to alfa galactosidase. Only the α-GAL around the cell membrane co-localised with TB474; the α-GAL in the cytoplasm of the cells did not.
The found location of acid ceramidase was not in line with the expectations. It was expected that there was no difference between the alfa and beta subunit of acid ceramidase. However, this is not the case. There may be an association between alfa-ceramidase and focal adhesion, since the alfa subunit of acid alfa-ceramidase is located in a striped-pattern, which resembles the pattern of focal adhesion sites (Ezratty et al., 2005). There could also be an association with early endosomes, since trafficking from the cell membrane towards the cytoplasm is mediated by actin fibres, which also resemble the striped pattern. It could also be that the alfa subunit of acid ceramidase moves through the cell in inactive form and that in the active form it is colocalised with the beta subunit of acid ceramidase. However, this has to be determined with an activity based probe for acid ceramidase.
It is in line with the expectations that there is more acid ceramidase when DNJ, an inhibitor of alfa galactosidase is added to the cells. It seems that DNJ binding to alfa galactosidase causes more acid ceramidase viewed with immunofluorescence. When carmofur, an acid ceramidase inhibitor, is added to the cells, there is less acid
ceramidase.
When added to B-cells without stimulation, the activity based probe TB474 alone cannot be found with immunofluorescence. It was however expected that migalastat could facilitate the endocytosis of the activity based probe, since migalastat promotes
21 endocytosis. An explanation for the deviation between the results and the
expectation can be found in the theory of chaperone-mediated autophagy: it could be that migalastat mediates autophagy of the endocytosed compounds, here TB474 (Yang et al., 2019). It could also be that the concentrations and incubation times of migalastat used are not optimal. It could namely be that in lower concentrations, migalastat promotes endocytosis, but in high concentrations, migalastat inhibits alfa galactosidase. Stimulation with IgM however did facilitate the endocytosis of the activity based probe TB474. This is due to the expected clustering and endocytosis of IgM receptors when stimulated with anti-IgM. This is also confirmed by the change in shape of the IgM-staining of B-cells when TB474 is added. It looks like IgM is more clustered on the surface, which indicates possible clustering of IgM-receptors. When viewed with fluorescence microscopy, the location of the activity based probe TB474 resembled the pattern of gb3, as expected, since active alfa galactosidase converts gb3 into lactosylceramide. It also colocalised with GLA, but only with the minority of the enzyme that was located close to the cell membrane. These statements could confirm that TB474 stains the enzyme activity of alfa galactosidase in vivo.
However, there are a lot of defects in the experiments. For the experiments with immunostaining of tissue and cells, it is unclear why there is a difference in location of the alfa and beta subunit of acid ceramidase. It may be due to defects in the
antibody or in the process of staining or in viewing the results under the fluorescence microscopy. However, the staining is repeated several times and gave the same results in all the experiments, so it is less likely that the experiment itself had a defect. For the experiments with B-cells, the quality of the cells from the Cytospin® was not the same in all the experiments. Besides that, a mistake in judging the results under the fluorescence microscope is easy made. However, the experiments with the activity based probes have been done more than once and the same results were collected, although sometimes results missed, because of failures in the process of the experiment. Finally, because of the fact that the experiments of the fibroblasts and podocytes differ from the experiments with B-cells in cell-type, experimental set up, antibodies, analyse technique and time, it is impossible to make comparisons
between both experiments. One cannot for example compare the pattern of the alfa and beta subunit of acid ceramidase with the location of the enzyme activity of alfa galactosidase and with the location of gb3 or GLA. The link between the location of enzyme activity of alfa galactosidase and the location of enzyme activity of acid ceramidase – for which there is not an activity based probe yet - is something for further research.
In this project, the results underline the statement that acid ceramidase differs in amount and localisation in cells when the amount of alfa-galactosidase is blocked (Sakuraba et al., 2018). This suggest that acid ceramidase plays an important role in
22 Fabry’s disease. This project also underlines the complex structure of acid ceramidase (Gebai et al., 2018).
Since now the activity based probe TB474 can be located with immunofluorescence, it is for further research to investigate if the probe truly binds to alfa galactosidase. In this way, the characteristics of alfa galactosidase in Fabry’s disease can be
determined in further research. One can think of experiments in determining the enzyme activity of alfa galactosidase in wildtype cells compared with cells with Fabry’s disease. Since activity based probes are known for being useful to screen drugs on their effect on different targets (Heal et al., 2011), one can think of
experiments around testing the effect of blocking alfa galactosidase in cells on the location of the enzyme activity of alfa galactosidase and thereby determining the effect of the drug. The same can be done for Fabrazyme®, by investigating the effect of ERT on the location of the enzyme activity of alfa galactosidase. On the other hand, further research can be focused on the explanation of the difference in location between the alfa and beta subunit of acid ceramidase in cells, since this is still not fully clear.
Now, combined with anti-IgM, the activity based probe TB474 can be found in Daudi B-cells with immunofluorescence. This result is an step forward in research around understanding the molecular and cellular basis of Fabry’s disease and can help in designing better treatment and in reducing unnecessary treatment costs.
23
References
Aerts, J.M., Groener, J.E., Kuiper, S., Donker-Koopman, W.E., Strijland, A., Ottenhoff, R., et al. (2008). Elevated globotriaosylsphingosine is a hallmark of Fabry disease. Proceedings of the National Academy of Sciences of the United States of America, 105(8), 2812-7.
Baigent, G., & Barngrover, D. (2002). Recombinant Interleukin-2 (aldesleukin) for oncology and HIV disease and recombinant protein treatment (Fabrazyme) for Farby’s disease. Journal of Biotechnology, 95:3, 277-283.
Branton, M.H., Schiffmann, R., Sabnis, S.G., Murray, G.J., Quirk, J.M., Altarescu, G., et al. (2002). Medicine, 81, 122-138.
El-Abassi, R., Singhal, D., & England, J.D. (2014). Fabry’s disease. Journal of neurological sciences, 344(1-2), 5-19.
El-Balat, A., Karn, T., Holtrich, U., Becker, S., Kommoss, S., et al. (2020). Histotype-specific analysis of acid ceramidase expression in ovarian cancer. Virchows Archiv, 2 january 2020 (doi: 10.1007/s00428-019-02728-0).
Ezratty, E.J., Partridge, M.A., & Gundersen, G.G. (2005). Microtubule-induced focal adhesion disassembly is mediated by dynamin and focal adhesion kinase. Nature Cell Biology, 7, 581-590.
Ferraz, M.J., Kallemeijn, W.W., Mirzaian, M., Herrera, M.D., Marques, A. (2014). Gaucher disease and Fabry disease: new markers and insights in pathophysiology for two distinct glycosphingolipidoses. Biochimica et Biophysica Acta, 1841, 811-825.
Ferraz, M.J., Marques, A.R.A., Appelman, M.D., Verhoek, M., Strijland, A., et al. (2016). Lysosomal glycosphingolipid catabolism by acid ceramidase: formation of glycosphingoid bases during deficiency of glycosidases. FEBS Letters, 560 (6), 716-725.
Gebai, A., Gorelik, A., Li, Z., Illes, K., Nagar, B. (2018). Structural basis for the activation of acid ceramidase. Nature Communication, 24:9, 1621.
Heal, W.P., Dang, T.H.T., & Tate, E.W. (2011). Activity-based probes: discovering new biology and new drug targets. Chemical Society Reviews, 40, 246-257.
Kuo, C.L., Meel van, E., Kytidou, K., Kallemeijn, W.W., Witte, M., Overkleeft, H.S., et al. (2018). Activity-based probes for glycosidases: profiling and other applications. Methods in Enzymology, 598, 217-235.
Loftin, K.C., Reuben, J.M., Hersh, E.M., & Sujansky, D. (1985). Cytoplasmic IgM in leukemic B cells by flow cytometry. Leukemia research, 9(11), 1379-1387.
24 McCafferty, E.H., & Scott, L.J. (2019). Migalastat: A review in Fabry disease. Drugs, 79, 543-554.
Ordonez, Y.F., Abad, J.L., Aseeri, M., Casas, J., Garcia, V., Casasampere, M., et al. (2019). Activity-based imaging of acid ceramidase in living cells. Journal of the American Chemical Society, 141(19), 7737-7742.
Saez, J.J., Diaz, J., Ibanez, J., Bozo, J.P., Cabrera, F.R., Alamo, M., et al. (2019). The exocyst controles lysosome secretion and antigen extraction at the immune synapse of B cells. Journal of Cell Biology, 218, 2247-2264.
Sakuraba, H., Togawa, T., Tsukimura, T., & Kato, H. (2018). Plasma lyso-gb3: a biomarker for monitoring fabry patients during enzyme replacement therapy. Clinical and Experimental Nephrology, 22(4), 843-849.
Turpin-Nolan, S.M., & Brüning, J.C. (2020). The role of ceramides in metabolic disorders: when size and localization matters. Nature reviews Endocrinology, 16(4), 224-233.
Wyatt, K., Henley, W., Anderson, L., Anderson, R., Nikolaou, V., Stein, K., et al. (2012). The effectiveness and cost-effectiveness of enzyme and substrate replacement therapies: a longitudinal cohort study of people with lysosomal storage disorders. Health Technology Assessment, 16(39), 1-543.
Yang, Q., Wang, R., & Zhu, L. (2019). Chaperone-mediated autophagy. Advanced Experimental Medical Biology, 1206, 435-452.
Yu, F.P.S., Amintas, S., Levade, T., & Medin, J.A. (2019) Acid ceramidase deficiency: Farber disease and SMA-PME. Orphanet journal of rare diseases, 13(1), 121.
Ziebold, T.O. (1967). Precision and sensitivity in electron microprobe analysis. Analytical Chemistry, 39-8, 858-861.
25
Appendix A: Results of staining healthy liver and kidney slides with
anti-ASAH1-rabbit (green) and anti-ASAH1-mouse (red).
Tissue is
incubated with base buffer or acid buffer. Scale bars = 50 µm.
ASAH1-rbt ASAH1-rbt ASAH1-m ASAH1-m ASAH1-rbt/ASAH1-m/H H ASAH1-rbt/ASAH1-m/H H ASAH1-rbt ASAH1-rbt ASAH1-m ASAH1-m ASAH1-rbt/ASAH1-m/H H ASAH1-rbt/ASAH1-m/H H
Human liver tissue, base buffer
Human liver tissue, acid buffer
Human kidney tissue, base buffer
26
Appendix B: Results of staining fibroblasts with anti-ASAH1-mouse,
anti-ASAH1-rabbit, EEA-1 and LAMP-1.
Scale bars = 50 µm. *intensified
= intensity of the staining is increased, using editing software.
ASAH1-rbt/ASAH1-m/H ASAH1-rbt/ASAH1-m/H
ASAH1-rbt ASAH1-m ASAH1-m ASAH1-rbt Control DNJ ASAH1-m ASAH1-m ASAH1-rbt ASAH1-rbt
ASAH1-rbt/ASAH1-m/H ASAH1-rbt/ASAH1-m/H
Carmofur DNJ + Carmofur
ASAH1-rbt & ASAH1-m
EEA-1/ASAH1-m/H EEA-1/ASAH1-m/H
EEA-1 ASAH1-m ASAH1-m EEA-1 Control DNJ ASAH1-m ASAH1-m EEA-1 EEA-1
EEA-1/ASAH1-m/H EEA-1/ASAH1-m/H
Carmofur DNJ + Carmofur
27 ASAH1-rbt/LAMP1/H ASAH1-rbt/LAMP1/H
ASAH1-rbt LAMP1 LAMP1 ASAH1-rbt Control DNJ LAMP1 LAMP1 ASAH1-rbt ASAH1-rbt
ASAH1-rbt/LAMP1/H ASAH1-rbt/LAMP1/H
Carmofur DNJ + Carmofur
ASAH1-rbt & LAMP1
H H
Control DNJ
H H
Carmofur DNJ + Carmofur
28 ASAH1-rbt/ASAH1-m/H ASAH1-rbt/ASAH1-m/H
ASAH1-rbt ASAH1-m ASAH1-m ASAH1-rbt Control DNJ ASAH1-m ASAH1-m ASAH1-rbt ASAH1-rbt
ASAH1-rbt/ASAH1-m/H ASAH1-rbt/ASAH1-m/H
Carmofur DNJ + Carmofur
ASAH1-rbt & ASAH1-m (intensified*)
EEA1/ASAH1-m/H EEA1/ASAH1-m/H
EEA1 ASAH1-m ASAH1-m EEA1 Control DNJ ASAH1-m ASAH1-m EEA1 EEA1
EEA1/ASAH1-m/H EEA1/ASAH1-m/H
Carmofur DNJ + Carmofur
29
ASAH1-rbt/LAMP1/H ASAH1-rbt/LAMP1/H
ASAH1-rbt LAMP1 LAMP1 ASAH1-rbt Control DNJ LAMP1 LAMP1 ASAH1-rbt ASAH1-rbt
ASAH1-rbt/LAMP1/H ASAH1-rbt/LAMP1/H
Carmofur DNJ + Carmofur
ASAH1-rbt & LAMP1 (intensified*)
H H
Control DNJ
H H
Carmofur DNJ + Carmofur
30
Appendix C: Results of staining podocytes with anti-ASAH1-mouse,
anti-ASAH1-rabbit, EEA-1 and LAMP-1.
Scale bars = 50 µm. *intensified
= intensity of the staining is increased, using editing software.
*Corrected = Intensity of the stainings is increased, using editing software. ASAH1-rbt/ASAH1-m/H ASAH1-rbt/ASAH1-m/H
ASAH1-rbt ASAH1-m ASAH1-m ASAH1-rbt Control DNJ ASAH1-m ASAH1-m ASAH1-rbt ASAH1-rbt
ASAH1-rbt/ASAH1-m/H ASAH1-rbt/ASAH1-m/H
Carmofur DNJ + Carmofur
ASAH1-rbt & ASAH1-m
EEA-1/ASAH1-m/H EEA-1/ASAH1-m/H
EEA-1 ASAH1-m ASAH1-m EEA-1 Control DNJ ASAH1-m ASAH1-m EEA-1 EEA-1
EEA-1/ASAH1-m/H EEA-1/ASAH1-m/H
Carmofur DNJ + Carmofur
31 ASAH1-rbt/LAMP1/H ASAH1-rbt/LAMP1/H
ASAH1-rbt LAMP1 LAMP1 ASAH1-rbt Control DNJ LAMP1 LAMP1 ASAH1-rbt ASAH1-rbt
ASAH1-rbt/LAMP1/H ASAH1-rbt/LAMP1/H
Carmofur DNJ + Carmofur
ASAH1-rbt & LAMP1
LAMP1 ASAH1-rbt H H Control DNJ H H Carmofur DNJ + Carmofur No antibodies
32 EEA1/ASAH1-m/H EEA1/ASAH1-m/H
EEA1 ASAH1-m ASAH1-m EEA1 Control DNJ ASAH1-m ASAH1-m EEA1 EEA1
EEA1/ASAH1-m/H EEA1/ASAH1-m/H
Carmofur DNJ + Carmofur
EEA1 & ASAH1-m (intensified*)
ASAH1-rbt/ASAH1-m/H ASAH1-rbt/ASAH1-m/H
ASAH1-rbt ASAH1-m ASAH1-m ASAH1-rbt Control DNJ ASAH1-m ASAH1-m ASAH1-rbt ASAH1-rbt
ASAH1-rbt/ASAH1-m/H ASAH1-rbt/ASAH1-m/H
Carmofur DNJ + Carmofur
33 ASAH1-rbt/LAMP1/H ASAH1-rbt/LAMP1/H
ASAH1-rbt LAMP1 LAMP1 ASAH1-rbt Control DNJ LAMP1 LAMP1 ASAH1-rbt ASAH1-rbt
ASAH1-rbt/LAMP1/H ASAH1-rbt/LAMP1/H
Carmofur DNJ + Carmofur
ASAH1-rbt & LAMP1 (intensified*)
H H
Control DNJ
H H
Carmofur DNJ + Carmofur
