MESENCHYMAL STEM CELLS FOR USE AS VEHICLES FOR THERAPEUTIC AGENTS

Abstract

Mesenchymal stem cells for use as vehicles for therapeutic agents. The present invention is related to a method for selecting or obtaining mesenchymal stem cells (MSCs) that can be used as transporters or vehicles for therapeutic agents, especially in cancer treatment.

Claims

1. Substantially pure population of mesenchymal stem cells, wherein the mesenchymal stem cells constitute at least 80% of the total cell population, characterized in that expression of the MAVS gene or other genes regulated by said gene is inhibited.

2. Substantially pure population of mesenchymal stem cells according to claim 1, wherein the mesenchymal stem cell is derived from bone marrow, placenta, umbilical cord, amniotic membrane, menstrual blood, peripheral blood, salivary gland, skin and prepuce, synovial fluid, amniotic fluid, endometrium, adipose tissue, umbilical cord blood and/or dental tissue.

3. Substantially pure population of mesenchymal stem cells according to either claim 1 or claim 2 for use as transporters or vehicles for therapeutic agents.

4. Composition which comprises: i) at least one cell according to either claim 1 or claim 2 and ii) at least one antigenic substance.

5. Composition according to claim 4, wherein the antigenic substance is an intracellular parasite.

6. Composition according to either claim 4 or claim 5, wherein the intracellular parasite is an oncolytic virus.

7. Composition according to any one of claims 4 to 6, wherein the intracellular parasite is ICOVIR-5.

8. Pharmaceutical composition which comprises a composition according to any one of claims 4 to 7 and pharmaceutically acceptable vehicles, adjuvants and/or excipients.

9. Composition according to any one of claims 4 to 7, or pharmaceutical composition according to claim 8, for use as a medicine.

10. Composition according to any one of claims 4 to 7, or pharmaceutical composition according to claim 8, for use in the treatment of solid tumors.

11. Composition for use in the treatment of solid tumors according to claim 10, wherein the solid tumor is brain cancer, lung cancer, renal cancer, ovarian cancer and/or liver cancer.

12. In vitro method for the production of a substantially pure population of modified mesenchymal stem cells which comprises: a. Obtaining mesenchymal stem cells from an isolated sample; b. Culturing the mesenchymal stem cells; c. Inhibiting expression of the MAVS gene, or other genes regulated by said gene, in the cultured cells.

13. Method according to claim 12 which also comprises infecting the cells with an intracellular parasite.

14. Kit which comprises: a. At least one mesenchymal stem cell; a. A system capable of inhibiting expression of the MAVS gene, or other genes regulated by said gene.

15. Kit according to claim 14, wherein the system capable of inhibiting expression of the gene is a small interfering RNA or gene therapy vector.

Description

BRIEF DESCRIPTION OF THE FIGURES

[0050] FIG. 1. Diagram illustrating the process of obtaining the MSCs according to the invention, characterized in that expression of the MAVS gene (or other genes regulated by said gene) is inhibited, or in that the MAVS protein is not expressed.

[0051] FIG. 2. Relative expression of the differentially expressed genes by ANN analysis for the immune response to the virus process (MAVS, p≤0.001; NDRG1, p≤0.0116).

[0052] FIG. 3. Quantification of the proteins differentially secreted in the supernatants of the Celyvir cultures. Of the five proteins analyzed, only IL-6 and MCP-1 were detected with the multiplex assays used (p≤0.05).

[0053] FIG. 4. Graphic illustration of the proposed model.

DETAILED DESCRIPTION OF THE INVENTION

Example 1. Materials and Methods

Example 1.1. Design of the Study

[0054] Celyvir is an advanced therapy medicinal product (ATMP) produced with autologous mesenchymal cells and the oncolytic adenovirus ICOVIR-5. Variability is associated with the individual and intrinsic characteristics of the cells of the ATMP which could have an impact on the quality of the medicine, which in turn may affect the clinical response. These biological characteristics in the different Celyvir products administered to the first cohort of patients were studied exhaustively, as shown in FIG. 1. Using an approximate analysis of biological systems, biological programs were identified that functioned differently between patients who responded to the treatment and those who did not. Moreover, molecules responsible for said differences were identified. These candidates were validated in a second cohort of patients treated with Celyvir.

Example 1.2. Obtaining Mesenchymal Cells from Patients

[0055] The MSCs were obtained by aspirating bone marrow from the iliac crest of the patients. Mononucleated cells were obtained by Ficoll gradient centrifugation (Ficoll-Paque™ PLUS, GE Healthcare) at 1500 rpm for 20 minutes at 20° C. The layer of mononuclear cells obtained after centrifugation were isolated from the rest of the components of the gradient using a Pasteur pipette. The cells were transferred to a clean Falcon tube and washed with phosphate buffered saline (PBS), Gibco. After centrifuging for one last time at 1500 rpm for 5 minutes, the mononucleated cells were resuspended in culture medium.

Example 1.3. Culture of the Mesenchymal Cells from Patients

[0056] The initial mononucleated fraction was cultured at a density of 500,000 cells/cm2 of surface with Dulbecco's Modified Eagle Medium (DMEM 1 g/L glucose, Gibco, Carlsbad, Calif.), supplemented with 10% fetal bovine serum (FBS, HyClone, Logan, Utah) which had previously been inactivated in a wet bath for 20 minutes at 56° C., and 1% penicillin-streptomycin (P/S, Gibco). The culture medium was replaced 48 hours later, which was sufficient time for the MSCs to adhere to the plastic while the rest of the mononucleated cells remained in suspension in the culture medium. After this first change, the medium was replaced every 3-5 days.

[0057] The MSCs were kept in an atmosphere at 37° C. and 5% CO.sub.2 until reaching an adequate confluency level (approximately 80%). At the moment of trypsinization, the cells were washed once with PBS and then trypsin was added thereto (TrypLE Express, Life Technologies, Carlsbad, Calif.). The cells that had detached from the plastic surface were collected using fresh culture medium and a pipette, and centrifuged at 1500 rpm for 5 minutes to remove the trypsin. The MSC pellet obtained was resuspended in fresh culture medium, and the cells were re-seeded at 3000-5000 cells/cm2.

[0058] In the third or fourth cell pass, the MSCs were characterized by flow cytometry.

Example 1.4. Icovir-5 Oncolytic Adenovirus

[0059] The characteristics of ICOVIR-5 have been described by its inventors (Alonso M M, Cascallo M, Gomez-Manzano C, Jiang H, Bekele B N, Perez-Gimenez A et al. ICOVIR-5 shows E2F1 addiction and potent antiglioma effect in vivo. Cancer Res 2007; 67:8255-8263). For this study, it was produced in the Vector Production Unit of the University of Barcelona.

Example 1.5. Celyvir Preparation

[0060] The preparation of the medicine for human use met good manufacturing practice (GMP) standards and the administration thereof to patients has already been described by the present team (Melen G J, Franco-Luzón L, Ruano D, Gonzalez-Murillo A, Alfranca A, Casco F et al. Influence of carrier cells on the clinical outcome of children with neuroblastoma treated with high dose of oncolytic adenovirus delivered in mesenchymal stem cells. Cancer Lett 2016; 371: 161-170).

Example 1.6. Celyvir Transcriptome Study (RNA Sequencing)

[0061] For the transcriptome study, the cell fraction of the Celyvir cultures was used from which the supernatants were also removed. The RNA was extracted using the Absolutely RNA Microprep Kit (Agilent, Santa Clara, Calif.) and each sample was treated with DNasa to avoid DNA contamination. 3 μg of RNA from each sample were sent to the Massive Sequencing Unit at the Madrid Science Park where massive sequencing was carried out. The raw data obtained were used for the subsequent statistical analysis.

Example 1.7. Validation of the Results Obtained in the Transcriptome Analyses

[0062] Validation of the Transcriptome Results Using qPCR

[0063] To validate the candidate genes identified by the bioinformatics analyses, a quantitative PCR was carried out using TaqMan technology (Applied Biosystems, Foster City, Calif.) on Celyvir samples from a cohort of patients that was independent of the initial cohort used during the identification process (responders, n=5; non-responders, n=10).

[0064] Table 1 shows the Applied Biosystems assays used.

TABLE-US-00001 TABLE 1 Gene Synonym Reference TRAF3 Factor 3 associated with the Hs00936781_m1 TNF receptor MAVS Mitochondrial antiviral-signaling Hs00920075_m1 protein PCBP2 Poly(rC)-binding protein 2 Hs01590472_mH FASN Fatty acid synthase Hs01005622_m1 NDRG1 N-myc regulated protein 1 Hs00608387_m1 GAPDH Glyceraldehyde 3-phosphate Hs02786624_g1 dehydrogenase

[0065] The qPCR reaction was carried out using TaqMan Gene Expression Master Mix (Applied Biosystems, Foster City, Calif.) and the proportion of the elements for a PCR reaction is that shown in Table 2 for a final volume of 20 μL:

TABLE-US-00002 TABLE 2 TaqMan Gene Expression Master Mix (2X) 10 μL TaqMan Gene Expression Assay (20X) 1 μL RNAsa-free water 4 μL cDNA 5 μL FINAL REACTION VOLUME 20 μL

[0066] Quantification of Secretome Candidates by Luminex

[0067] To functionally validate the results of the genes determined in the previous paragraph, the levels of some proteins in the Celyvir supernatants from a cohort of patients independent of the initial cohort used during the identification process (responders, n=5; non-responders, n=10) were quantified using Luminex xMAP technology. To do this, the following Luminex xMAP assays were used, in accordance with the instructions of the manufacturer: [0068] MILLIPLEX MAP Human Cytokine/Chemokine Magnetic Bead Panel (HCYTOMAG-60K, Merck, EMD Millipore Corporation, Billerica, Mass.) for measuring interleukin 4 (IL-4), interleukin 6 (IL-6), interleukin 10 (IL-10) and MCP-1 (CCL2). [0069] MILLIPLEX MAP Human Soluble Cytokine Receptor Panel (HSCRMAG-32K, Merck, EMD Millipore Corporation, Billerica, Mass.) for measuring the interleukin 6 receptor 6 (IL-6R).

[0070] The plates were analyzed using MAGPIX® equipment (EMD Millipore Corporation, Billerica, Mass.) and the results were interpreted using MAGPIX xPONENT software (EMD Millipore Corporation, Billerica, Mass.).

Example 2. Results

Example 2.1. Transcriptome Analysis

[0071] The data from the transcriptome were processed using conventional statistical analyses in order to identify genes with greater differential expression between the MSCs of responders and non-responders. In a first approximation, two groups of genes were considered: those that showed significant statistical enrichment with a level of error of 1% (p<0.01) and those to which a less restrictive criterion (p<0.05) had been applied in order to produce a more extensive list of potentially interesting results. Three statistical tests were carried out (ANOVA, Wilcoxon and T-student) to evaluate statistically significant differences in gene expression between responder and non-responder patients. The following positive results were obtained following the corrected multitest (FDR<0.05): only 15 results, and only those derived from the ANOVA test. Table 3 shows the candidate genes that in a first approximation showed significant differences between the responder and non-responder patient groups.

TABLE-US-00003 TABLE 3 ANOVA Uniprot Gene FDR ID name Secretome q-value Log.sub.2 FC FC P35573 AGL No 0.01280 6.49492 90.19145 P09238 MMP10 Yes 0.01280 8.65044 401.83075 Q12834 CDC20 No 0.01280 −8.88274 0.00212 P11597 CETP No 0.01280 −9.02274 0.00192 P19404 NDUFV2 No 0.01280 −10.89015 0.00053 P16284 PECAM1 No 0.01280 −7.35809 0.00610 A2AJT9 CXorf23 No 0.01280 −6.79375 0.00901 P17021 ZNF17 No 0.02209 −9.33322 0.00155 Q8IWZ8 SUGP1 No 0.02236 8.37531 332.06250 Q96Q89 KIF20B No 0.02236 6.68242 102.70893 Q8TE49 OTUD7A No 0.02236 −8.93858 0.00204 Q92990 GLMN No 0.02236 −8.50728 0.00263 Q9NZH4 PTTG3P No 0.02236 −10.71427 0.00060 O60930 RNASEH1 No 0.04146 8.51271 365.24350 O75153 CLUH No 0.04974 −1.65012 0.31861

[0072] Being less statistically restrictive, a total of 113 genes showed a significant uncorrected p-value (p-value<0.01) for at least one of the three tests carried out.

Example 2.2. Gene Set Enrichment Analysis (GSEA)

[0073] The above data were uploaded to the GSEA platform to identify the differential molecular pathways and processes between the MSCs of patients who responded to therapy using Celyvir and those who did not. This approximation provided an ordered list of genes where the more significant the difference, the higher the position in said list. In this case, the parameter used to classify said genes was the ratio between the signal emitted and the background signal for each sample. Those that had a background signal in all the samples were excluded from the GSEA. With these criteria, the expression levels of 13,715 genes in the signal-noise ratio parameter were transformed. A total of 5601 genes were left out of the analysis as no signal was detected in any of the samples.

[0074] 81 significantly enriched pathways were found for the cohort of patients who responded to Celyvir, and 59 pathways for the non-responder patient cohort.

[0075] The most relevant processes and pathways provided by GSEA were: [0076] 1. Viral infection and metabolism of nucleic acids. The GSEA results indicated that the expression of the genes involved in these processes was increased in the cohort of responder patients compared with non-responders. Some processes relating to nucleic acids appeared as representative of the cohort of responder patients, including groups of genes relating to mechanisms for regulating deoxyribonucleic acid (DNA) transcription, DNA replication, DNA-ribonucleic acid (RNA) interaction, ribosomal RNA, RNA splicing and translation. In addition, groups of genes directly relating to viral biology (positive regulation of viral release by the host cell, viral life cycle) appear overexpressed in the samples of responder patients. [0077] 2. Processes relating to the immune response. These processes appeared increased in the MSCs of both cohorts of patients. However, this type of process class differed depending on the cohort. Responder patients showed enrichment of the pathways relating to IFNγ, and also processes relating to T cell biology; both processes are known to constitute antiviral response mechanisms. In the case of the non-responder patients, there is a greater presence of processes relating to phagocytic capacity. Similarly, this group of patients show an increased amount/greater abundance of immune cell migration processes. The majority of these (diapedesis, monocyte adhesion, leukocyte extravasation, etc.) contain PECAM1, which appears expressed in all the samples of non-responder patients. The non-responder patients also showed an increased expression of proteins relating to increased permeability of the vessels. [0078] 3. MSCs regulatory signaling pathways. Differences were detected in the pathways associated with negative and positive regulation of MSC migration in both patient cohorts. These include the HGF signaling, tachykinin signaling, Hippo signaling, NOTCH signaling, Wnt signaling and MAPK signaling pathways. [0079] 4. Cell-cell adhesion. Both patient cohorts showed differences in gene groups relating to cell-cell adhesion: the responders in desmosomes and communicating junctions; whereas the non-responder patients mostly showed processes relating to cytoskeleton reorganization. [0080] 5. Mitosis. A large number of gene groups relating to mitosis appeared overexpressed in the MSCs of responder patients, compared with the MSCs of non-responder patients. Further, the samples from non-responder patients showed overexpression of the gene set that defines the regulation of stem cell proliferation. [0081] 6. Metabolism. Both patient cohorts showed differences in genes relating to metabolism, mainly glycogen, proteins and lipid metabolism. Both groups of patients showed increased glycogen metabolism, which occurred either by direct catabolism (mainly represented by AGL, which was expressed consistently in responder patients but not in non-responders) or by negative regulation of glycogenesis in non-responders. The samples of responder patients showed enrichment of the genes relating to protein metabolism, including peptidase activity and metabolism of peptides and amino acids. On the other hand, non-responder patients showed an increase in genes relating to lipid metabolism (isoprenoid production, and regulation and production of fatty acids and lipoproteins). [0082] 7. Stress response. Both groups of patients showed increased determined pathways relating to the stress response (organelle stress response; DNA damage induced by phosphorylation, regulation of the DNA damage response and endoplasmic reticulum stress). [0083] 8. Bone marrow fibrosis. The samples of non-responder patients showed an enrichment of genes relating to bone marrow fibrosis when compared with responder patients, including angiogenic factors (VEGFA, VEGFB, PDGFA), proinflammatory cytokines (IL 1A and IL 1B) and extracellular matrix remodeling factors (MMP14).

[0084] Next, an evaluation was made of whether the proteins that were differentially expressed in the secretome of the MSCs of responder and non-responder patients formed part of the processes that had appeared enriched in the GSEA (p-value<0.01). In this respect, it was seen that 27 of the 127 candidates participated in at least one of the enriched gene sets (12 out of 81 in the responder patients and 9 out of 59 in non-responder patients). A total of 17 proteins that were differentially expressed in the secretome of the MSCs between responder and non-responder patients formed part of the enriched sets in the samples of responder patients. However, 15 proteins that were differentially expressed and present in the secretome of both patient groups formed part of enriched protein sets in non-responder patients.

Example 2.3. Biological Systems Approximation Analysis

[0085] The object of the first analysis or approximation was simply to identify those genes or proteins that were expressed with greater or lesser intensity in each patient group. Next, a more in-depth analysis was carried out of the processes that could be most relevant in the efficacy or activity of Celyvir: those that were related to the immune response to the virus and to viral replication. To do this, the candidates were evaluated using mathematical analyses of biological systems and artificial neural networks (ANNs).

[0086] The ANN model evaluates the relationships that exist between sets of regions within the network or matrix generated (in the present case, genes and/or proteins), providing a predictive score which quantifies the probability of the existence of functional relationships or connections between the regions evaluated (in the present case, the sets of genes/proteins evaluated). Each score is associated with a p-value that indicates the probability of the result being statistically significant. Depending on the value of the p-value, groups of proteins are defined that have: [0087] A strong relationship with the processes studied at that time if the protein set has a strong or medium-strong predictive relationship with any of the sub-processes used for the characterization (p-value<0.05). [0088] A medium relationship with the processes studied at that time if said processes have at least a medium value relationship with any of the sub-processes used for the characterization (p-value<0.075-0.25). [0089] A weak or non-existent relationship with the processes studied at that time if said processes have a low or non-existent relationship with the rest of the sub-processes used in the characterization (p-value<0.25).

[0090] Given the considerable extent of the full analysis, only the most relevant results obtained for the two above-mentioned processes are detailed.

Example 2.4. Differential Genes Obtained in the Transcriptome Analysis

[0091] A total of 44 differentially expressed genes detected in the transcriptome of the MSCs infected with the oncolytic adenovirus ICOVIR-5 from both responder and non-responder patients showed a strong relationship with the following processes: [0092] Immune response to the virus (29 proteins) [0093] Viral replication (26 proteins)

[0094] For the immune response to the virus process, a total of five genes coding for the same number of proteins were strongly related to at least three different sub-processes involved in this overall process: TRAF3, MAVS, PCBP2, FASN and NDRG1 (Table 4). Table 4 shows genes differentially expressed by the patients for the immune response to the virus process. The genes that appear in the list showed a strong relationship to said process. Detailed from left to right are the name of the gene; its abundance in the responder patient group; if the gene has an effector role in the process described; the value of the ANN analysis score; and, finally, a description of the specific sub-processes in which said gene intervenes.

TABLE-US-00004 TABLE 4 In ANN ID responders Effector score Related processes TRAF3 Decreased Yes 84.19 Regulation of defense response to the virus Yes 74.32 Viral pattern-recognition receptor Yes 84.18 Toll-like receptor signaling No 75.06 Regulation of Toll-like receptor signaling Yes 74.46 Type I IFN response Yes 79.08 Regulation of type I IFN production No 74.15 Regulation of T cell differentiation No 77.06 Virus entry into host cell MAVS Decreased Yes 90.73 Regulation of defense response to the virus Yes 77.38 Defense response to the virus Yes 77.67 Regulation of signaling pathway mediated by type I IFN Yes 88.63 Regulation of type I IFN production Yes 92.29 Regulation of chemokine production PCBP2 Decreased Yes 82.50 Regulation of defense response to the virus Yes 81.04 Defense response to the virus Yes 82.13 Response to type I IFN Yes 82.46 Regulation of type I IFN production FASN Decreased No 78.50 Toll-like receptor signaling No 77.39 Regulation of type I IFN production No 74.14 Response to IFN-gamma NDRG1 Decreased No 77.28 Response to type I IFN No 78.37 Regulation of the adaptive immune response No 74.30 Leukocytic differentiation No 81.59 DNA replication

[0095] For the viral replication process, a total of eight genes coding for the same number of proteins were strongly related to at least two sub-processes involved in this overall process: CHMP2A, CLEC5A, XBP1, BMPR1B, MAVS, RAB29, TRAF3 and MEMO1 (Table 5). Table 5 shows genes differentially expressed by the patients for the viral replication process. The genes that appear in the list showed a strong relationship to said process. Detailed from left to right are the name of the gene; its abundance in the responder patient group; if the gene has an effector role in the process described; the value of the ANN analysis score; and, finally, a description of the specific sub-processes in which said gene intervenes.

TABLE-US-00005 TABLE 5 In ANN ID responders Effector score Related processes CHMP2A Increased No 75.01 Viral cycle Yes 77.50 Regulation of the viral cycle Yes 88.52 Viral assembly Yes 85.04 Regulation of viral release from the host cell CLEC5A Increased Yes 80.38 Viral cycle Yes 85.74 Virus entry into host cell XBP1 Increased Yes 81.54 Regulation of immunity mediated by immunoglobulins No 74.49 Viral cycle No 77.73 Regulation of translation BMPR1B Increased No 75.93 Viral cycle No 75.53 DNA replication MAVS Decreased Yes 82.30 Regulation of viral cycle No 77.23 Cell response to virus RAB29 Decreased Yes 80.00 Viral cycle No 78.42 Regulation of viral transcription TRAF3 Decreased No 77.06 Viral entry into host cell No 76.78 Cell response to virus MEMO1 Decreased No 74.24 RNA interference No 75.19 Regulation of viral transcription No 76.01 Regulation of translation

Example 2.5 Validation of the Candidates Obtained in the Transcriptome Studies

[0096] Once the molecules involved in the different biological behavior of the ATMP between both groups of patients regarding cell response to infection and permissivity for adenoviral replication were identified, said molecules were validated. To do this, an independent cohort of patients treated with the Celyvir ATMP was used.

Example 2.6. Validation of Candidate Genes Responsible for the Differences in the Innate Cellular Response to Infection

[0097] A quantitative PCR was carried out for the FASN, MAVS, NDRG1, PCBP2 and TRAF3 genes. The results revealed that only MAVS and NDRG1 met this condition (FIG. 2).

Example 2.7. Quantification of Secretome Candidates

[0098] The other two analytes (IL-6 and MCP-1) showed significantly lower values in the non-responder group (FIG. 3).

[0099] The validation results obtained in the study of the second cohort are consistent with what is known about MAVS signaling. The relationship between IL6 and CCL2 and MAVS has been described (Seth R B, Sun L, Ea C-K, Chen Z J. Identification and characterization of MAVS, a mitochondrial antiviral signaling protein that activates NF-κB and IRF3. Cell 2005; 122: 669-682) (Jiang C, Lin X. Regulation of NF-κB by the CARD proteins. Immunol Rev 2012; 246: 141-153), such that MAVS regulates the expression of IL6 and CCL2 via the NF-κB pathway. The inability to produce interferon in the context of an antiviral response in knockout (deficient) mice for MAVS (−/−) has been demonstrated (Sun Q, Sun L, Liu H-H, Chen X, Seth R B, Forman J et al. The specific and essential role of MAVS in antiviral innate immune responses. Immunity 2006: 24: 633-642). This role of MAVS has been detected in immune cells (mainly macrophages and dendritic cells) (Di Fiore I J M, Holloway G, Coulson B S. Innate immune responses to rotavirus infection in macrophages depend on MAVS but involve neither the NLRP3 inflammasome nor JNK and p38 signaling pathways. Virus Res 2015; 208: 89-97) (Roe K, Giordano D, Young L B, Draves K E, Holder U, Suthar M S et al. Dendritic cell-associated MAVS is required to control West Nile virus replication and ensuing humoral immune responses. PloS One 2019; 14: e0218928) (Vazquez C, Horner S M. MAVS Coordination of Antiviral Innate Immunity. J Virol 2015; 89: 6974-6977), but as far as is known, this is the first time the regulatory role of MAVS in mesenchymal cells has been described and confirmed.

[0100] In sum, the low expression of these two genes and these two proteins (resulting from lower signaling via the first) in responder patients indicates lower signaling of the intracellular presence of ICOVIR-5, or delayed signaling, with significantly less production of immune response mediators. In any event, the MSCs of responder patients show under-stimulation of the immune system so that said immune system develops an antiviral response and therefore an immune response to Celyvir. This allows the mesenchymal cells to migrate to the tumor sites without being neutralized by the patient's immune system at the time ICOVIR-5 is being replicated inside the MSCs. A graphic illustration of the hypothesis is that shown below (FIG. 4).