Flattop (FLTP) is a novel biomarker for beta cell maturation

Abstract

The present invention relates to the use of the biomarker Flattop (Fltp) for distinguishing mature cells from immature progenitor cells. The present invention further relates to a method for distinguishing a mature cell from an immature progenitor cell, the method comprising: determining the presence or absence of the biomarker Flattop (Fltp) in a cell; wherein the presence of Fltp in the cell indicates that the cell is a mature cell and wherein the absence of Fltp in the cell indicates that the cell is an immature progenitor cell. Furthermore, the present invention relates to a method of identifying a compound suitable for differentiating immature progenitor cells into mature cells as well as to a method of identifying a compound suitable for preventing the de-differentiating of mature cells.

Claims

1. A method of identifying a compound suitable for differentiating immature progenitor cells into mature cells, the method comprising: (a) contacting a cell population comprising immature progenitor cells with a test compound; and (b) subsequently determining the presence or expression level of the biomarker Flattop (Fltp) in the cells comprised in the cell population; wherein the presence of Fltp, or an increased expression level of Fltp, in the cells comprised in the cell population after the contacting with the test compound is indicative of a compound suitable for differentiating immature progenitor cells into mature cells.

2. A method of identifying a compound suitable for preventing the de-differentiation of mature cells, the method comprising: (a) culturing a cell population comprising mature cells in the presence of a test compound, wherein the cells are cultured under conditions that induce the de-differentiation of said mature cells; and (b) subsequently determining the expression level of the biomarker Flattop (Fltp) in the cells cultured in step (a), wherein an expression level of Fltp determined in step (b) that is substantially identical to the expression level of Fltp in the cell population comprising mature cells prior to the culture in step (a) is indicative of a compound suitable for preventing the de-differentiation of mature cells.

3. The method according to claim 1 or 2, wherein the test compound is a compound that activates planar cell polarity (PCP).

4. The method according to claim 3, wherein the compound that activates planar cell polarity (PCP) is an activator of the non-canonical Wnt/PCP pathway.

5. A method for distinguishing mature cells from immature progenitor cells, the method comprising: determining the presence or absence of the biomarker Flattop (Fltp) in a cell, comprising subjecting the cell to a process that detects the presence or absence of a Fltp protein or nucleic acid in the cell; wherein the presence of the Fltp protein or nucleic acid in the cell indicates that the cell is a mature cell and wherein the absence of the Fltp protein or nucleic acid in the cell indicates that the cell is an immature progenitor cell; and separating the immature progenitor cells from the mature cells.

6. The method according to claim 5, wherein the presence or absence of a Fltp protein or nucleic acid is determined (i) on the nucleic acid level, (ii) on the amino acid level, or (iii) a combination thereof.

7. The method of claim 6, further comprising inducing the expression of Fltp in the cell population comprising immature progenitor cells, wherein the expression of Fltp in the cells is induced in the presence of a compound that induces the non-canonical Wnt/PCP pathway.

8. The method of claim 7, wherein the compound that induces the non-canonical Wnt/PCP pathway is Wnt5a or Wnt4.

9. The method of claim 5, further comprising inducing the expression of Fltp in the cell population comprising immature progenitor cells, wherein the expression of Fltp in the cells is induced in the presence of a compound that induces the non-canonical Wnt/PCP pathway.

10. The method of claim 9, wherein the compound that induces the non-canonical Wnt/PCP pathway is Wnt5a or Wnt4.

Description

(1) The figures show:

(2) FIG. 1: Fltp reporter expression correlates with post-natal cell maturation and shows heterogeneity in all endocrine lineages.

(3) (A) Representative images of single immunofluorescence stainings of pancreatic tissue for Fltp reporter and Nkx6.1 on postnatal day 1 (P1) as well as of 12 weeks old Fltp.sup.ZV/+ animals. (B) Quantitative analysis of Fltp reporter expression in cells during cell maturation period in Fltp.sup.ZV/+ mice. Values are meanSEM (standard error of the mean); n=2 for P1, 3, 11, and n=9 for 12 weeks; ***P<0.001 in Fltp reporter and Nkx6.1 positive -cells of P1 vs. 12 weeks old animals. The percentage of Fltp reporter expressing cells increase from 45% (P1) over 70% (P11) to 80% (12 weeks). (C) Representative images of single immunofluorescence stainings of 4,6-diamidino-2-phenylindole (DAPI), Fltp reporter, and one endocrine hormone on adult pancreatic tissue. Hormones are arranged from upper to lower row as indicated (glucagon, insulin, somatostatin, and pancreatic polypeptide). (D) Quantitative analysis of Fltp reporter expression reveals heterogeneity in Fltp reporter expression in all endocrine cell types (=47.4%, =81.1%, =49.3%, PP=47.0%). Values are meanSEM; n>30 islets. Nuclear reporter of Fltp expression (Fltp) is marked by GFP antibody (A,C).

(4) FIG. 2: cell subpopulations exhibit differences in proliferative capacity in vivo.

(5) (A) Quantitative analysis of Ki67 expression in cells upon metabolic demand (pregnancy) in Fltp.sup.ZV/+ mice. Values are meanSEM; n=4 for control and pregnant mice embryonic day 15.5 (E15.5); **P<0.01 in Ki67 expression in cells which lack Fltp reporter expression vs. Fltp reporter expressing cells in pregnant (E15.5) mice. (B) Representative images of single immunofluorescence stainings of Fltp reporter, Nkx6.1 and Ki67 on adult pancreatic tissue of the control group and pregnant mice E15.5. (C) Quantitative analysis of cell proliferation upon metabolic demand (pregnancy) after 24 h in vivo pulse labeling using EdU in Fltp.sup.ZV/+ mice. Values are meanSEM; n=5 for control and pregnant mice; ***P<0.001 in EdU marked cells of pregnant mice (E15.5) which lack Fltp reporter expression vs. Fltp reporter expressing cells of pregnant mice and vs. Fltp reporter negative cells in control group. **P<0.01 in EdU marked cells of pregnant mice (E15.5) expressing Fltp reporter vs. Fltp reporter expressing cells of control mice. *P<0.05 in EdU marked cells of control mice which lack Fltp reporter expression vs. Fltp reporter expressing cells of control mice. (D) Representative images of single immunofluorescence stainings of Fltp reporter, Nkx6.1, and EdU on adult pancreatic tissue of the control group and pregnant group at E15.5. (E) Quantitative analysis of Ki67 expression in cell upon metabolic demand (pregnancy) in Fltp.sup.ZV/+ mice. Values are meanSEM; n=2 for P1, 3, 11 and n=4 for control and pregnant mice (E18.5); *P<0.05 in Ki67 expression in cells which lack Fltp reporter expression vs. Fltp reporter expressing cells in P1, P3, P11 and pregnant (E18.5) mice. (F) Qualitative analysis of Fltp reporter expression in cells upon metabolic demand (pregnancy E15.5). Values are meanSEM, n=5 for control and pregnant group; **P<0.01 between percentage of Fltp reporter expressing -cells compared to Fltp reporter negative cells. Nuclear reporter of Fltp expression (Fltp) is marked by GFP antibody (B,D).

(6) FIG. 3: Fltp.sup.ZV/ZV cells show 1.sup.st phase insulin secretion defects in vivo.

(7) (A) Fltp.sup.ZV/+ as well as Fltp.sup.ZV/ZV do not reveal altered glucose tolerance compared to Fltp.sup.+/+. The ability of Fltp.sup.ZV/+, Fltp.sup.ZV/ZV and Fltp.sup.+/+ to handle a glucose load was assessed by using a standard GTT. Fasted male mice were injected i.p. with glucose (2 g/kg of body weight), and blood glucose levels were measured at 0, 15, 30, 60, and 120 min after glucose injection. (B) Quantitative analysis of total pancreatic insulin content in Fltp.sup.+/+, Fltp.sup.ZV/+ and Fltp.sup.ZV/ZV mice shows no major difference (n=5 per genotype). Values are meanSEM. (C) Quantitative analysis of first phase insulin secretion reveals differences between Fltp.sup.ZV/ZV and Fltp.sup.+/+ (n=4 mice per genotype). Values are meanSEM. (D) Quantitative analysis of second phase insulin secretion shows no major difference. Values are meanSEM.

(8) FIG. 4: FLIP intronic SNP rs7515334 significantly associates with insulin secretion defects in human.

(9) (A) Depending on the metabolic status, the minor C1ORF192 allele rs7515334 associates with increase insulin secretion (BMI<25) and decreased insulin secretion (BMI>35) (B). Insulin secretion index adjusted for gender, age, and OGTT-derived insulin sensitivity. Genotype tested in the dominant inheritance model with XG=CG+GG. Bonferroni-corrected -level: p<0.0073.

(10) FIG. 5: Fltp.sup.ZV/ZV mice reveal altered susceptibility to STZ induced cell death.

(11) (A) Representative images of a single immunofluorescence stainings of DAPI, Fltp reporter and Nkx6.1 on adult pancreatic tissue from Fltp.sup.ZV/+ mice treated with streptozotocin (STZ), Fltp.sup.ZV/+ vehicle (citrate buffer) (control), Fltp.sup.ZV/ZV mice treated with STZ, and Fltp.sup.ZV/ZV vehicle (control). (B) After an overnight fast (16 h), Fltp.sup.ZV/+ and Fltp.sup.ZV/ZV both treated with STZ and their respective controls, mice were injected i.p. with glucose (2 g/kg of body weight) and blood glucose levels were measured at 0, 15, 30, 60, and 120 min after injection. Fltp.sup.ZV/ZV mice reveal an enhanced susceptibility to STZ induced cell damage and develop insulin resistance at an earlier time point compared to Fltp.sup.ZV/+. Values are meanSEM; n=4 for STZ mice and n=4 for control mice. *P<0.05 vs Fltp.sup.ZV/+. (C) Whole-blood glucose concentration in Fltp.sup.ZV/+ and Fltp.sup.ZV/ZV both treated with STZ-induced diabetic mice (40 mg/kg, 5 days) and their respective control mice. Nuclear reporter of Fltp expression (Fltp) is marked by GFP antibody (A).

(12) FIG. 6: Fltp reporter negative and positive islet subpopulations show distinct gene expression differences.

(13) (A) Schematic overview of experimental approach starting from islet isolation to Fluorescent activated cell sorting (FACS) to gene expression analysis using microarray. (B) FACS plot of endocrine subpopulations with side scatter (SSC) on y-axis and emission of 488 nm fluorescence on x-axis from isolated mouse islets of Fltp.sup.ZV/+ mouse (12 weeks old). (C) Representative images of single immunofluorescence stainings of Nkx6.1 and Fltp reporter on sorted Fltp reporter negative cells and Fltp reporter expressing cells. (D) Microarray analysis of Fltp mRNA expression in FACS sorted samples; two dimensional heat map of microarray transcriptional gene profile containing genes which are differential regulated and important in endocrine cell function, polarity, signaling and cell cycle; P<0.05 and 1.5 regulated, 3 samples left are Fltp reporter expressing samples and 2 samples right are Fltp reporter negative samples. (E) GO-term enrichments after analysis with Geps of Genomatix; bars represent significance by P-value. Nuclear reporter of Fltp expression (Fltp) is marked by GFP antibody (C).

(14) FIG. 7: Genetic lineage tracing reveals that Fltp negative progenitors give rise to Fltp expressing mature cells

(15) (A) Scheme of the Fltp.sup.T2A-iCre/+ mTmG.sup.+/ model. In this model all cells express membrane bond tomato (mT) except for the Fltp.sup.T2A-Cre expressing cells which express membrane bound GFP (mG). When a Fltp negative cell starts to activate the Fltp promoter the cells switch through a yellow state (mG and mT expression) into the green state (mG expression). (B) Fluorescent activated cell sorting plot of islets from Fltp.sup.T2A-iCre/+-mTmG.sup.+/ model. Nkx6.1 staining control (up) and Nkx6.1 stained cells (down). GFP and RFP expression are shown on the right (66.7% of Nkx6.1+ cell are GFP expressing cells, 12.3% are RFP positive and 20% are GPF and RFP positive). (C) Representative images from live imaging experiment of islet isolated from Fltp.sup.T2A-iCre-mTmG.sup.+/ which shows the conversion of a mT expressing cell into mG and mT expressing state (white arrow).

(16) FIG. 8: Fltp expression correlates with islet neogenesis and post-natal cell maturation.

(17) (A-C) Qualitative analysis of Fltp::H2B-Venus reporter activity in whole-mount stained and BABB cleared pancreata and analyzed by LSM analysis at E18.5 and P5. Fltp reporter activity is detected by anti-GFP antibodies. High magnification images of compacted islets and cord-like structures (C-C). Fltp reporter activity strongly increases while islets form and become compacted 3D structures, but is still undetectable in cord-like structures.

(18) FIG. 9: Fltp.sup.ZV/ZV animals are Fltp null mutations.

(19) (A) The Fltp.sup.ZV targeting strategy deletes the whole open reading frame ranging from exon 2 (E2) till E6 (Primers for genotyping: 5-AGCCATACCACATTTGTAGAGG-3, 5-CAGCATGGCATAGATCTGGAC-3, 5-GAGGCTGACTGGGAACAATC-3). The external 5- as well as the 3-Southern probe are indicated. Restriction enzyme sites for DraIII and EcoRV are shown. Homology regions for recombination of the targeting construct are indicated as 5- and 3-Retrieval (5- and 3-R). The figure is on scale. (B) Southern blot of WT embryonic stem (ES) cells versus Fltp.sup.ZV/+ ES cells digested with DraIII and hybridized with the external 5 Southern probe showing the BI6 (16443 bp) and 129 WT allele as well as the BI6 targeted allele (11469 bp). Notice the shift of the WT band due to restriction length polymorphism. (C) Genotyping PCR to discriminate between WT, Fltp.sup.ZV/+, and Fltp.sup.ZV/ZV (Primers used: 418, 565, and 566; WT band (317 bp); targeted neo band (387 bp)). (D) Western blot shows the absence of Fltp protein in testis lysate of Fltp.sup.ZV/ZV animals. Fltp protein band is detectable at around 25 kDa (calculated weight 20 kDa). Abbreviations: NLS-LacZ: nuclear localization signal-beta-galactosidase; 2A: viral T2A sequence; H2B: histon-2B; Venus: yellow fluorescent reporter gene; SpA: Simian Virus 40 polyadenylation signal; loxP: site of Cre mediate recombination; bpA: bovine Growth Hormone polyadenylation signal; neo: neomycin resistance cassette; PG: phospho-glycerate kinase; UTR: untranslated region; SP: Southern probe; VP: Venus probe.

(20) FIG. 10: Generation a validation of the Fltp::Venus Fusion mouse line.

(21) (A) The Fltp::Venus-Fusion targeting strategy fuses the open reading frame of Fltp to the fluorescent reporter gene Venus and the 3 FLAG tag (Primers for genotyping: 5-CAGCATGGCATAGATCTGGAC-3, 5-GAGGCTGACTGGGAACAATC-3, 5-CAAGATCCGCCACAACATCG-3). The external 5- as well as the 3 Southern-probe are indicated. Restriction enzyme sites for DraIII are shown. Homology regions for recombination of the targeting construct are indicated as 5- and 3-Retrieval (5- and 3-R). (B) Southern blot analysis of targeted mouse ES cells (129Sv/ C57Bl/6; IGD 3.2; Hitz et al., 2007). Genomic DNA was digested with DraIII and hybridized with the 5 Fltp external probe resulting in bands of 16443 bp (calculated size) for WT (Fltp.sup.+/+) and 19133 bp for Fltp::Venus Neo (Fltp.sup.V/+) allele. (C) Genotyping PCR to discriminate between WT, Fltp.sup.+/+, and Fltp.sup.V/V (Primers for genotyping: 5-CAGCATGGCATAGATCTGGAC-3, 5-GAGGCTGACTGGGAACAATC-3, 5-CAAGATCCGCCACAACATCG-3). WT band (317 bp); targeted neo band (430 bp)). (D) Western blot analysis of lung and testis lysates of 2 months old male WT or Fltp.sup.+/+ animals using anti-Fltp antibody to detect endogenous Fltp (24 kDa) as well as the Fltp::Venus fusion protein (50 kDa). Abbreviations: E1-6: exon 1-6; Venus: yellow fluorescent reporter gene; bpA: bovine Growth Hormone polyadenylation signal; EM7: bacterial promoter; loxP: site of Cre mediate recombination; neo: neomycin resistance cassette; PG: phospho-glycerate kinase; UTR: untranslated region; SP: Southern probe; VP: Venus probe; 3F: 3 FLAG tag.

(22) FIG. 11: Fltp with active Foxa2 binding sites in its promoter, antibody binding sites and its conservation among species.

(23) (A) Fltp shows Foxj1, Foxa1, and Foxa2 binding sites in its promoter (bar under the schematic gene). (B) Scheme of Fltp protein showing the two predicted proline rich repeats (PRRs) and the peptide sequences of the Fltp116-1 antibody as well as the Fltp1 antibody. (C, D) Western blot showing the specificity of the Fltp116-1 (C) as well as the Fltp1 (D) antibody in testis lysate of WT, Fltp.sup.ZVP, and Fltp.sup.ZV/ZV animals. Fltp protein band is detectable at around 25 kDa (calculated weight 20 kDa). (E) Immunohistochemistry on cryosections combined with LSM analysis to show that the Fltp116-1 antibody is specific. Fltp is localized at the apical plasma membrane and along cilia in multi-ciliated lung epithelial cells of WT (Ctrl) adult animals, but no Fltp immunoreactivity is detected in Fltp.sup.ZV/ZV lungs. Abbreviations: exonl-6 (E1-E6); TSS: transcriptional start site). Nuclei are marked by 4,6-Diamidin-2-phenylindol (DAPI), Fltp by Fltp116-1. Scale bars; 10 m.

(24) FIG. 12: monoclonal antibodies against human FLTP

(25) (A) Laser scanning microscopy (LSM) of EndoC- H1 human -cells stained with different monoclonal antibodies (clones #13, #28 and #43) against human FLIP. As negative control human embryonic stem cells (hESCs) were used (not shown). (B) Western Blot of Strep Flag-tagged human FLTP transiently transfected in HEK293T and detected by monoclonal antibodies (clones #13, #28 and #43). Non-transfected HEK293T cells lysate were used as negative control.

(26) FIG. 13: FLTP mRNA expression in EndoC- H1 human -cells Significant increase of human FLTP mRNA expression levels cultured under 3D vs 2D conditions in EndoC- H cells is shown.

(27) FIG. 14: 3D polarization and Wnt/PCP pathway activation in murine and human -cells

(28) (a,b) Laser scanning microscopy (LSM) of EndoC- H1 human -cells cultured in Matrigel (3D) show higher expression of -cell and maturation marker NKX6.1 compare to 2D conditions (b). For quantification see d (3D: n=2; 2D: n=2). (c,g) Diagram showing the fluorescent intensity of Nkx6.1 in one single cell in 3D compared to 2D (c) and in 3D treated with Wnt5a and 3D without Wnt5a treatment (g). White lines indicate the measured region of interest (ROI) (i.e. in a,b for c; e,f for g). (e,f) LSM of EndoC- H cells cultured Matrigel (3D) and treated with WNT5a show higher expression of NKX6.1 than under 3D conditions (f). For quantification see h (WNT: n=1; control: n=4). (i,j) LSM of isolated islets of P5 WT animals either treated with (i) or without (j) Wnt5a for 3 days showing significant increase in Ucn3 maturation marker staining. For quantification see I (Wnt5a treated: n=3; untreated: n=3). (k) Diagram showing the fluorescent intensity of Ucn3 of the re-aggregated islets+Wnt5a and without Wnt5a treatment. White lines in (i,j) indicate the measured ROI. (m) LSM picture of EndoC- H cells at day 7 showing compacted 3D pseudo-islets. (n) FLTP mRNA expression level in EndoC-H cells cultured under 3D and 2D conditions. Counting criteria for d,h are: The NKX6.1 fluorescent intensity of a certain number of cells of several independent experiments were counted in 2D and 3D and the median fluorescent intensity was calculated with IMARIS. The median fluorescent intensity was used as a threshold. All cells with higher intensity were counted as NKX6.1 high, all cells with lower intensity were counted as NKX6.1 low cells. For FIG. 14d (3D: 300 NKX6.1.sup.+ cells; 2D: 221 Nkx6.1.sup.+ cells) were counted. For FIG. 14h (WNT: 110 Nkx6.1.sup.+ cells; control: 550 Nkx6.1.sup.+ cells) were counted. Scale bars, 20 m (a,b,e,f), 10 m (m).

(29) FIG. 15: 3D polarization and Wnt/PCP pathway activation in murine and human -cells (additional data)

(30) (a,b) LSM of EndoC- H cells cultured in Matrigel (3D) show higher expression of -cell and maturation marker NKX6.1 than under 2D (b). For quantification see d (3D: n=1; 2D: n=1). (c) Diagram showing the fluorescent intensity of Nkx6.1 in one single cell in 3D and in 2D. White lines in a,b show the measured region of interest (ROI). (e,f) LSM of isolated islets of P5 WT animals revealed a higher amount of maturation marker Ucn3 protein in islets treated with Wnt5a for 12 h (e) compared to non-treated islets (f). For quantification see h (islets+Wnt5a: n=3; untreated islets: n=3). (g) Diagram showing the fluorescent intensity of Ucn3 of the re-aggregated islets+Wnt5a and without Wnt5a. White lines in (e,f) indicate the measured ROI. Counting criteria of c are: The NKX6.1 fluorescent intensity of a certain number of cells of several independent experiments were counted in 2D and 3D and the median fluorescent intensity was calculated with IMARIS. The median fluorescent intensity was used as a threshold. All cells with higher intensity were counted as NKX6.1 high, all cells with lower intensity were counted as NKX6.1 low cells. For FIG. 15d (3D: 571 NKX6.1.sup.+ cells; 2D: 265 Nkx6.1.sup.+ cells) were counted. For FIG. 15h (islets+Wnt5a: 680 cells; untreated islets: 877 cells) were counted. Scale bars, 20 m (a,b,e,f).

(31) The examples illustrate the invention:

EXAMPLE 1: MATERIALS AND METHODS

(32) Animal Studies

(33) Mice were kept in the animal facility under optimal conditions in a 12-h light cycle. Food and water were given ad libitum. Animal experiments were carried out according to the German animal care and ethics legislation and were approved by the local government authorities.

(34) Eight-week-old Fltp.sup.ZV/+ females were paired to C57BL/6J males and separated from the males after the appearance of the vaginal plug indicating day 0 of pregnancy. Pregnant females were euthanized on day 15.5 of pregnancy (E 15.5) and the pancreas removed.

(35) Glucose tolerance test (GTT) and insulin secretion test (IST) were carried out in 12 weeks-old mice after a 12 h fast. Briefly, a single dose of glucose was intraperitoneally administrated (2 g/kg body weight) to the mice and blood glucose level was measured using Freestyle Lite glucometer (Abbot Laboratory, USA), at 0, 15, 30, 60 and 120 minutes following glucose loading, by cutting off the tip of tail and squeezing it gently. For plasma insulin detection, blood sample were taken at 0, 2.5, 5, 10 and 20 minutes following glucose loading (Andrikopoulos, Blair et al. 2008, Ayala, Samuel et al. 2010). Plasma insulin was determined by using Ultra-sensitive mouse insulin ELISA kit (Chrystal Chem, USA) according to the manufacturer's instructions.

(36) For Streptozotocin (STZ)-mediated diabetes induction, freshly prepared STZ (Sigma Aldrich, Germany) in 50 nM sodium citrate (pH 4.5) was injected intraperitoneally (40 mg/kg) daily for 5 days. Blood glucose level was measured every 2 days using Freestyle Lite glucometer (Abbot Laboratory). On day 16 after the first STZ injection, GTT was carried out as described and mice were euthanized.

(37) Generation of Animal Models

(38) A Flattop-Venus fusion protein was generated by homologous recombination in mouse embryonic stem (ES) cells by removing the translational stop codon in exon 6 and directly fusing the Venus fluorescent protein to the open-reading frame of Flattop. From these ES cells a knock-in mouse was generated by germ line transmission of the targeted ES cells. These mice express the Flattop-Venus fusion protein in all tissues where Flattop is expressed in equal amounts to the wild-type Flattop protein. This Flattop-Venus fusion reporter has the advantage that it is expressed in physiological amount, shows normal protein turnover and shows normal subcellular localization.

(39) Antibody Generation

(40) To be able to analyse the Fltp protein in more detail, two different polyclonal rabbit antibodies were raised against a central and C-terminal epitope (FIG. 11). The specificity of these antibodies was confirmed in western blot analysis and immunocytochemistry on lysates and cells in which the Fltp gene was either over-expressed or knocked-out (FIG. 11). To be able to analyse the human FLIP protein in more detail, three different monoclonal rat antibodies were raised against a central and C-terminal epitope (FIG. 12). The specificity of these antibodies was confirmed in western blot analysis and immunocytochemistry on lysates and cells in which the Strep-Flag tagged human FLTP cDNA was over-expressed (FIG. 12). These antibodies can be used as primary antibodies to detect the protein in tissues or cell cultures and using secondary antibodies either conjugated to horseradish peroxidase, alkaline phosphatase or fluorescent dyes.

(41) Total Pancreatic Insulin Content

(42) After 12 h starving, the animals were euthanized and the pancreas rapidly removed. Pancreas was placed into 5 ml 0.2 M HCl in 70% Ethanol, homogenized and incubated over night at 20 C. Subsequently, the homogenized pancreas was again mixed with 0.2 M HCl in 70% Ethanol and incubated over night at 20 C. After centrifugation at 1000 g for 15 minutes the supernatant was diluted (1:2) with 1 M Tris pH 7.5 and then analyzed. Insulin detection was performed by using Ultra sensitive mouse insulin ELISA kit (Chrystal Chem, USA) according to the manufacturer's instructions. Total pancreatic protein content was estimated by Bradford assay (Harlow and Lane 2006). Total pancreatic insulin content is stated as insulin (ng)/total pancreatic protein (g).

(43) LacZ Staining and Immunohistochemistry

(44) Whole mount organ staining was performed as previously described (Huber, Kania et al. 2005). For whole mount imaging, embryos were cleared using BABB (1 part benzyl alcohol, 2 parts benzyl benzoate). For immunohistochemistry staining, pancreas samples were fixed in 4% formalin, cryoprotected by incubation in sucrose gradient for 1 h each (5%, 15%, 30%) and embedded in Optimum Cutting Temperature (OCT). Cells sorted on glass slide were fixed with 2% PFA.

(45) Nuclear staining was performed with DAPI (Life Technology, Germany). For histological assessment of islet -cell proliferation, mice were injected with EdU Solution (100 g/g of body weight) 24 h prior to being sacrificed. EdU staining was performed using the Click-iT Staining Kit (Life Technology) according to the manufacturer's instructions. Cryosection imaging was performed using Leica Confocal SP5 microscope. For quantification purposes, stained cells were counted manually on every tenth section (14-15 m thick frozen section). Quantification of whole mount organ staining was performed by using IMARIS software (Bitplane, Switzerland).

(46) Islet Isolation, FACS Analysis and Gene Profiling

(47) Islet isolation was carried out by collagenase P (Roche, Germany) digestion and centrifugation using Optiprep density gradient (Sigma). Isolated islets were handpicked two times under the microscope. After 1 to 3 h of culture, islets were washed with PBS and incubate with 0.25% Trypsin-EDTA (Invitrogen) to obtain single cell suspensions. Single cells were sorted using FACS-Aria III (BD Bioscience). The results were analyzed by using Flow Jo software. Total RNA was extracted using miRNeasy micro kit (Qiagen, Germany), amplified with the Ovation PicoSL WTA System V2 in combination with the Encore Biotin Module (Nugen, USA). Amplified cDNA was hybridized on Affymetrix Mouse Gene ST 1.0 arrays containing about 28,000 probe sets. Staining (Fluidics script FS450_0007) and scanning was done according to the Affymetrix expression protocol including minor modifications as suggested in the Encore Biotion protocol. Expression console (v.1.3.0.187, Affymetrix) was used for quality control and to obtain annotated normalized RNA gene-level data (standard settings including median polish and sketch-quantile normalization). Statistical analyses were performed by utilizing the statistical programming environment R (R Development Core Team) implemented in CARMAweb. Genewise testing for differential expression was done employing the (paired) limma t-test and Benjamini-Hochberg multiple testing correction (FDR<10%). Heatmaps were generated with CARMAweb and cluster dendrograms with the R script hclust. GO term and pathway enrichments were done for 1.5 regulated genes and a P-value<0.005 using the GePS module in the Genomatix Software Suite v3.1 (Genomatix, Munich, Germany). For lineage tracing studies, islet from eight-week-old Fltp.sup.T2A-Cre-mTmG were isolated as described above. For FACS analysis islet were incubate with 0.25% Trypsin-EDTA (Invitrogen) to obtain single cell suspensions and intracellular staining for Nkx6.1 was performed. For live imaging experiment isolated islets from Fltp.sup.T2A-Cre-mTmG were cultured in matrigel and imaging was performed using Leica Confocal SP5 microscope.

(48) Antibodies

(49) Primary antibodies used for immunofluorescence: Goat anti-Nkx6.1 (R&D system, Germany, AF5857 1:200); chicken anti-GFP (Ayes Labs, USA, GFP-1020 1:800); guinea pig anti-glucagon (Millipore, 4031-01F, 1:500); goat anti-somatostatin (Santa Cruz, USA, sc-7819, 1:300); rabbit anti-insulin (Thermo Scientific, USA, PA-18001, 1:300); guinea pig anti-insulin (Thermo Scientific, USA, PA-26938, 1:300); goat anti-Pancreatic Polypeptide (PP) (Abcam, USA, ab77192, 1:300); rabbit anti-Ki-67 (Abcam, ab15580, 1:200); rabbit anti-Urocortin 3 (Phoenix Pharmaceuticals, USA, H-019-29, 1:300); Alexa Fluor 546 phalloidin (Invitrogen, Germany, A22283, 1:200).

(50) Secondary antibodies used for indirect fluorescence staining (dilution 1:800 for all): Goat anti-chicken Alexa Fluor 488 (Dianova, Canada, 103-545-155); donkey anti-goat Alexa Fluor 488 (Invitrogen, Germany, A11055); donkey anti-mouse Alexa Fluor 555 (Invitrogen, A31570); donkey anti-goat Alexa Fluor 555 (Invitrogen, A21432); donkey anti-rabbit Alexa Fluor 555 (Invitrogen, A31572); donkey anti-goat Alexa Fluor 594 (Invitrogen, A11058); donkey anti-mouse Alexa Fluor 594 (Invitrogen, A21203); donkey anti-guinea-pig Alexa Fluor 649 (Dianova, 706-495-148). Nuclear staining was performed with DAPI (Life Technology, Germany). EdU staining was performed using the Click-IT Staining Kit (Life Technology) according to the manufacturer's instructions.

(51) Statistical Analysis

(52) Statistical analysis was performed using GraphPad Prism 6 Software (GraphPad Software, USA). Student's t-test was used for direct comparisons between two groups. A p value of <0.05 was considered as statistically significant. Data is expressed as meansSEM/SD.

EXAMPLE 2: THE EXPRESSION OF THE PLANAR CELL POLARITY (PCP) EFFECTOR GENE FLTP STRONGLY INCREASES DURING POST-NATAL CELL MATURATION AND SHOWS HETEROGENEOUS LEVELS IN ADULT ENDOCRINE CELLS

(53) Three-dimensional (3D) and self-organized tissue architecture is required for organ formation and function (Eiraku, Takata et al. 2011, Sasai 2013). To determine whether the acquisition of tissue polarity during islet neogenesis impacts on cell function and maturation, tissue polarity establishment was analyzed on the molecular level by using the Fltp::H2B-Venus reporter mice.

(54) The Fltp reporter activity accurately reflects planar cell polarity (PCP) activity in the inner ear and lung, tissues that depend on Fltp function as a modulator of the actin and MT cytoskeleton dynamics. Fltp reporter activity was first analyzed at embryonic day (E) 18.5 when cells are still organized in cord-like structures and vividly aggregate to form 3D sphere-like mini-organs. For this purpose, whole pancreata were isolated, the tissue was cleared and laser-scanning confocal microscopy (LSM) analysis was used to acquire the entire 3D tissue distribution of Nkx6.1.sup.+ cells (FIG. 8). Interestingly, Fltp reporter expression was confined to 3 cells that were found in compacted islets, but was not detectable in cells that were still organized in cord-like structures. Additionally, it seemed that high levels of Nkx6.1 also induced Fltp reporter activity, but only in already formed islet structures. Together these results confirmed that Fltp reporter activity is switched on during planar cell polarity acquisition comparable to the inner ear and the lung.

(55) To further investigate the expression of this PCP effector gene during post-natal cell maturation (Blum, Hrvatin et al. 2012), Fltp::H2B-Venus reporter expression was analyzed in pancreatic sections shortly after birth (FIG. 1A). At postnatal day 1 (P1), Fltp reporter activity was detected in less than 50% of Nkx6.1.sup.+ cells. Reporter activity rapidly increased in up to 70% of cells while they matured in newly formed islets during the first ten days of life. In adult mice, Fltp reporter activity was detected in approximately 80% of cells and around 50% of other endocrine cell populations (FIG. 1C). These results show that all endocrine cells express heterogeneous levels of PCP reporter activity. Additionally, the reporter activation during islet neogenesis and compaction likely reflects the changes in physical properties of the islet cell niche and might have a functional impact on all islet cells.

EXAMPLE 3: FLTP NEGATIVE CELLS SHOW INCREASED PROLIFERATIVE CAPACITY ESPECIALLY DURING CELL EXPANSION PERIODS

(56) For the further study, the focus was on the biological relevance of the PCP-related heterogeneity for cells. Therefore, first the proliferation rate of Fltp reporter negative and Fltp reporter expressing cells during homeostasis, as well as during pregnancy-induced and postnatal cell expansion periods was compared. Strikingly, the Fltp reporter negative Nkx6.1.sup.+ cells at P1, P3 and P11 showed an up to four-fold higher replication rate when compared to Fltp reporter expressing Nkx6.1.sup.+ cells, as measured by Ki67 immunoreactivity in pancreatic cryosections (FIG. 2E). A similar up to 4-fold difference in the proliferation rate of these two cell subpopulations was observed during pregnancy (FIGS. 2A and 2E). The concomitant and significant decrease of Fltp reporter positive cells from 80% to 70% (FIG. 2F) indicates that mainly the Fltp reporter negative cells proliferate and contributed to cell expansion during pregnancy. Even in adult mice fed ad libitum, a physiological state where 6 cells are in homeostasis and refractory to proliferation, Fltp reporter negative cells still showed an increased replication capacity (FIG. 2C). These proliferation studies were confirmed by EdU-pulse labeling during cell homeostasis and expansion (FIG. 2C). Thus collectively these data illustrate that the two subpopulations of cells show markedly different proliferative capacity depending on environmental conditions and cell polarity status.

EXAMPLE 4: LOSS OF MOUSE FLTP AND AN INTRONIC SNP IN HUMAN FLTP ASSOCIATES WITH INSULIN SECRETION DEFECTS

(57) It is well established that cells are functionally coupled and that insulin secretion depends on the actin and MT cytoskeleton (Kalwat and Thurmond 2013). To analyze if the PCP effector protein Fltp is necessary for adult cell function, a glucose tolerance test was performed using intra-peritoneal glucose stimulation (ipGTT) in adult males on a chow diet (FIG. 3A). Only a slight but not significant delay in glucose clearance was observed in Fltp.sup.ZV/ZV and Fltp.sup.ZV/+ mice, when compared to Fftp.sup.+/+ littermates. Also, no significant difference in total pancreatic insulin content was observed between Fltp.sup.+/+, Fltp.sup.ZV/+ and Fltp.sup.ZV/ZV mice (FIG. 3B). Interestingly, 1.sup.st (FIG. 3C), but not 2.sup.nd phase insulin secretion (FIG. 3D) seems to be delayed in homozygous mutants when compared to Fltp.sup.+/+ littermates, suggesting that the PCP activity and Fltp function is necessary for glucose-induced insulin secretion of j cells.

(58) To seek first evidence whether the human orthologue gene C1Orf192 is also associated with metabolic traits and 10+ SNPs were screened for genetic association in a cohort of 2100 human pre-diabetic and diabetic subjects. Interestingly, this revealed that the intronic SNP rs75715534 with a minor allele frequency of 0.2 significantly associated with increased insulin secretion on lean subjects (BMI<25), whereas the same SNP associated with decreased insulin secretion in obese subjects (BMI>35) (FIGS. 4A and B). These results suggest that human FLTP differentially reacts to metabolic demand and is generally important for glucose-induced insulin secretion. Together these data suggest that Fltp-dependent cytoskeletal rearrangements established during planar cell polarity are important for glucose-induced insulin secretion of cells.

(59) Association Analysis

(60) Subjects.

(61) The study population consisted of 2,228 Caucasians at risk for type 2 diabetes (family history of type 2 diabetes, body mass index (BMI) 27 kg/m.sup.2, impaired fasting glycaemia, and/or previous gestational diabetes) recruited from the ongoing Tubingen Family study for type 2 diabetes (1). All participants underwent assessment of medical history, smoking status, and alcohol consumption habits; the subjects furthermore agreed to undergo physical examination, routine blood tests, and oral glucose tolerance tests (OGTTs). Only individuals with complete phenotypic and genotypic data sets and documented absence of medication known to influence glucose tolerance, insulin sensitivity, or insulin secretion were included. All study participants gave informed written consent to the study which adhered to the Declaration of Helsinki.

(62) The study protocol was approved by the OGTT and laboratory measurements. A standardized 75-g OGTT was performed following a 10-hovernight fast. For the determination of plasma glucose, insulin, and C-peptide levels, venous blood samples were drawn at baseline and at time-points 30, 60, 90, and 120 min of the OGTT (Stefan, Machicao et al. 2005). Plasma glucose levels (in mmol/L) were measured with a bedside glucose analyser (glucose oxidase method, Yellow Springs Instruments, Yellow Springs, Ohio, USA). Plasma insulin and C-peptide levels (in pmol/L, both) were determined by commercial chemiluminescence assays for ADVIA Centaur (Siemens Medical Solutions, Fernwald, Germany). BMI was calculated as weight divided by squared height (in kg/m.sup.2). OGTT-derived insulin sensitivity was estimated as proposed earlier (Matsuda and DeFronzo 1999): 10,000/[c(Glc0)*c(Ins0)*c(Glcmean)*c(Insmean)]I (with c=concentration, Glc=glucose, and Ins=insulin). OGTT-derived insulin secretion was estimated as area under the curve (AUC) Cpep0-30/AUC Glc0-30 according to the formula: [c(Cpep0)+c(Cpep30)]/[c(Glc0)+c(Glc30)] (with Cpep=C-peptide). Ethics Committee of the Eberhard Karls University Tubingen.

(63) Selection of Tagging SNPs and Genotyping.

(64) Based on publicly available data from the 1000 Genomes Project (http://browser.1000genomes.org/index.html), we analysed in silico a genomic area on human chromosome 1q23.3 spanning the C1orf192 gene (3.143 kb, five exons, four introns, located on the reverse strand) and 2 kb of the gene's 5-flanking region. Within the analysed C1orf192 locus, 36 SNPs were found. Using the tagger analysis tool of Haploview (see the Wordl Wide Web at broadinstitute.org/scientific-community/science/programs/medical-and-population-genetics/haploview/haploview), seven tagging SNPs were identified that cover all the other common SNPs (minor allele frequency 0.01) with an r.sup.2 0.8. These SNPs were rs17399583 (C/T), rs11584714 (C/G), and rs57835711 (C/G) in the 5-flanking region, rs114482063 (A/T) and rs182840301 (C/A) in intron 1, and rs16832872 (G/A) and rs75715534 (C/G) in intron 3. For genotyping, DNA was isolated from whole blood using a commercial kit (NucleoSpin, Macherey & Nagel, Duren, Germany). The tagging SNPs were genotyped using the mass spectrometry system massARRAY from Sequenom and the manufacturer's iPLEX software (Sequenom, Hamburg, Germany). The call rates were 96.1%. The mass spectrometric results were validated in 50 randomly selected subjects by bidirectional sequencing, and both methods gave 100% identical results.

(65) Statistical Analyses.

(66) Continuous variables with non-normal distribution were log.sub.e-transformed prior to statistical analysis. Multiple linear regression analysis was performed using the least-squares method. In the regression models, insulin secretion was chosen as outcome variable, the SNP genotype (in the dominant inheritance model) as independent variable and gender, age, BMI, and insulin sensitivity as confounding variables. SNP-BMI interaction effects on insulin secretion were tested by analysis of covariance (ANCOVA) with gender, age, and insulin sensitivity as confounding variables. When testing all seven tagging SNPs in parallel, a Bonferroni-corrected p-value <0.0073 was considered statistically significant. For all analyses, the statistical software JMP 8.0 (SAS Institute, Cary, N.C., USA) was used.

EXAMPLE 5: LOSS OF FLTP MAKES CELLS MORE SUSCEPTIBLE TO STREPTOZOTOCIN TREATMENT

(67) To further test the requirement of PCP and Fltp in the context of cell survival and regeneration, a multiple low-dose streptozotocin (STZ) model was employed (Kolb 1987). Streptozotocin is a naturally occurring chemical that is particularly toxic to cells. For this purpose STZ was injected on five consecutive days and blood glucose regulation was measured every 2.sup.nd day thereafter in Fltp.sup.ZV/+ and Fltp.sup.ZV/ZV cohorts (FIG. 5C). This revealed that blood glucose control and cell function gradually decreased in a Fltp-dose dependent manner until day 16 after the first injection. An ipGTT performed at day 16 after the first STZ injection revealed that Fltp.sup.ZV/ZV mice were significantly more glucose intolerant when compared to Fltp.sup.ZV/+ littermates (FIG. 5B). The gradual loss of cells was confirmed by immunohistochemistry on pancreatic sections from vehicle and STZ-treated animals (FIG. 5A), which at the stage analyzed did not show any signs of cell regeneration. Thus, loss of Fltp and planar polarization in cells increases the vulnerability to STZ.

EXAMPLE 6: MOLECULAR PROFILING REVEALS INCREASED MATURATION STATE OF FLTP REPORTER EXPRESSING CELLS

(68) The data presented so far clearly indicated that Fltp-mediated planar polarization subdivided cells into proliferative and into more mature cells. To better understand these PCP-mediated differences on the molecular level, use was made of the Fltp::H2B-Venus reporter. Adult islets were purified from animals under physiological homeostatic conditions and the Fltp reporter negative and Fltp reporter expressing endocrine subpopulations were isolated using fluorescent activated cell sorting (FACS; FIGS. 6A&B). The purity of the Fltp reporter negative and Fltp reporter expressing endocrine cell populations were controlled by cytospins and reached almost 100% for the Venus fluorescent marker and approx. 80% of both populations were cells (marked by Nkx6.1) (FIG. 6C).

(69) The Fltp reporter negative and Fltp reporter expressing endocrine subpopulations showed markedly different levels of mRNA expression after cDNA amplification and expression profiling using Affymetrix gene arrays (FIG. 6D). 1887 genes are up- and 1800 genes are more than 1.5-fold significantly (p<0.005) down-regulated. Strikingly, unbiased gene ontology (GO) term analysis revealed that the Fltp reporter negative population shows a significant enrichment of genes that associate with cell proliferation, actin binding, Wnt/PCP-, TGF receptor-, G-protein coupled receptor-, ERK-signaling transduction, whereas the Fltp reporter expressing population shows significant enrichment of genes that are important for mature beta cell function, such as genes involved in metabolic processes, glucose metabolism, mitochondria and insulin secretion (FIG. 6E).

(70) Taken together, the molecular profiling of Fltp reporter negative and Fltp reporter expressing endocrine populations clearly revealed that Fltp reporter negative cells constitute a population of low polarized more nave progenitor cells, whereas the Fltp reporter negative cells are highly polarized, more mature and express higher levels of glycolysis enzymes, polarity markers and signaling receptors for major cell regulatory pathways.

EXAMPLE 7: FLTP REPORTER NEGATIVE CELL PROGENITORS DIRECTLY GIVE RISE TO FLTP REPORTER EXPRESSING MATURE BETA CELLS

(71) To directly test whether Fltp reporter negative cells are progenitors of Fltp reporter expressing cells, a Cre recombinase/loxP-mediated genetic lineage tracing approach was used. To this end, the previously established Fltp-T2A-iCre mouse line (Lange, Gegg et al. 2012) was crossed to the mTmG reporter mouse line (Muzumdar, Tasic et al. 2007). Upon Fltp promoter-driven Cre expression, the membrane Tomato (mT) fluorescent reporter gene switches to membrane GFP (mG), which is irreversible and therefore allows cell fate analysis. Islets of these crossings were isolated and cultured in vitro. To establish the culture conditions, it is first tested if cells depend on Wnt/PCP signaling to maintain Fltp reporter expression. Therefore, islets were cultured in absence or presence of ligands of the canonical Wnt/-catenin (Wnt3a) and non-canonical Wnt/PCP pathway (Wnt5a). These experiments indicated that Fltp-reporter gene expression indeed depend on Wnt/PCP, but not canonical Wnt signaling. Using these culture conditions in combination with live single cell imaging will allow us to follow new born Fltp-Cre expressing mT and mG expressing cells over a time-course of two to four days. Single-cell tracking over this time period and later fixation and staining for the cell marker Nkx6.1 clearly revealed that mT expressing cells (Fltp negative) cells give rise to mG expressing cells (Fltp expression), suggesting that adult islets contain a Fltp negative low polarized and highly proliferative progenitor population, which can give rise to Wnt/PCP-dependent highly polarized and less proliferative Fltp expressing mature cell population.

EXAMPLE 8: NKX6.1 EXPRESSING CELL PROGENITORS FORM CORD-LIKE STRUCTURES WHICH LATER ARRANGE IN ISLETS OF LANGERHANS

(72) Analysis of Nkx6.1 expressing cells in whole-mount stained and BABB cleared pancreata at E18.5 show cord-like structures. During post-natal maturation period they become compacted 3D structures. The Fltp reporter is absent in b cells located in cord-like structures compared to islets. Together with the increased Fltp reporter expression during maturation (FIG. 1B) and the change in morphology of cord-like structures to islets indicate that Fltp expression is connected with islets formation.

EXAMPLE 9: KNOCK-IN/KNOCK-OUT OF LACZ AND H2B-VENUS INTO THE FLTP LOCUS

(73) To explore Fltp expression and function in vivo, a Fltp.sup.ZV knock-in/knock-out allele was generated where the entire ORF was replaced by a multicistronic lacZ-Venus reporter cassette. This contained a nuclear localization signal (NLS)-tagged -Galactosidase (lacZ) reporter gene followed by an intervening viral Thosea Asigma 2A-peptide (T2A) for co-translational cleavage and a very bright Histone 2B (H2B)-Venus fluorescent reporter gene. Southern and western blot analysis confirmed the targeted homologous recombination and generation of a null allele.

(74) The knock-in/knock-out construct was designed as shown in FIG. 9. 5 and 3 HR for the Fltp gene were amplified by PCR (449 fwd 5 HR Ascl: 5-NNNGGCGCGCCAGTCAGGAAGTGGAAGAGAAGAACACAG-3; 450 rev 5 HR HindIII, SpeI: 5-NNNAAGCTTACTAGTGTGGTGGAGTGCCTGTCTACATGTG-3; 451 fwd 3 HR HindIII: 5-NNNAAGCTICACGACAGTCAAAGCTGCAATAGAAC-3; 452 rev 3 HR BamHI: 5-NNNGGATCCGGTAATTTGGCAATTATAGAACTCAGGC-3) using a C57BL/6J BAC clone (RP23-333P11) as template. These two PCR products were subcloned into the pL254 vector (Liao, Uetzmann et al. 2009) using Ascl and BamHI. The resulting vector was digested with HindIII, SpeI and electroporated into electrocompetent EL350 bacteria containing the Fltp BAC clone to retrieve the WT sequence between PCR homology arms resulting in the Fltp retrieval vector. For cloning of the knock-in/knock-out cassette in pBKS-5 and 3 HR for the knock-in into the ATG of exon two of Fltp were generated by PCR (453 fwd 5 HR SacII: 5-NNNCCGCGGAGCAGACTTAACTATGTTGGGGAAACAGC-3; 454 rev 5 HR SalI, NotI: 5-NNNGTCGACGCGGCCGCTGTTTACACTTGTTGCCTGGCAACTG-3; 455 fwd 3 HR SalI: 5-NNNGTCGACGGTCCTAGTCTAGCTGAGGTCCAGATC-3; 456 rev 3 HR KpnI: 5-NNNGGTACCATGCTGTGGGAGTCACTGACATTCTTG-3) using the previously mentioned BAC as a template and subcloned into pBKS-using the introduced restriction sites, resulting in pBKS-Fltp-HomArms. The first step to generate the targeting vector was to construct the pBKS-H2B-Venus-intron-SV40pA plasmid by subcloning an oligonucleotide for the H2B (histone 2B) that introduces a 5 NotI and a perfect Kozak sequence (025: 5-NNNGCGGCCGCGCCACCATGCCAGAGCCAGCG-3) and a 3 XbaI site (026: 5-NNNTCTAGACTTAGCGCTGGTGTACTTGGTGATGG-3). This PCR product was ligated into pBKS-(both cut with NotI, XbaI) resulting in pBKS-H2B. The next step was to introduce the Venus reporter gene (yellow fluorescent protein) also via PCR with the forward (fwd) primer containing an XbaI site (013: 5-NNNTCTAGAATGGTGAGCAAGGGCGAGGAGCTGTTC-3) and a reverse (rev) primer containing a SpeI site (014: 5-NNNACTAGTTTACTTGTACAGCTCGTCCATGCCGAGAG-3). This PCR and the vector pBKS-H2B were digested with XbaI and SpeI and ligated resulting in pBKS-H2B-Venus. To complete the construct an intron-SV40pA oligonucleotide was generated by using the fwd primer containing SpeI (011: 5-NNNACTAGTAGGTAAGTGTACCCAATTCGCCCTATAG-3) and the rev primer containing BamHI (012: 5-NNNGGATCCACGCGTTAAGATACATTGATGAGTTTGGAC-3). This oligonucleotide was subcloned into pBKS-H2B-Venus by cutting both with SpeI and BamHI resulting in the pBKS-H2B-Venus-intron-SV40pA plasmid. The next step was to introduce the loxP flanked neomycin (neo) resistance cassette by digesting the PL-452 vector (Liu, Jenkins et al. 2003) with SalI and BamHI. The digested vector was ligated into the pBKS-H2B-Venus-intron-SV40pA plasmid opened by cutting with SalI and BamHI resulting in pBKS-H2B-Venus-intron-SV40pA-loxP-bGHpA-neo-EM7-PGK-loxP (pBKS-H2B-Venus-neo). For following cloning steps it was necessary to destroy the MluI site located in the SV40pA by cutting with MluI, filling up the 5 overhang with Klenow polymerase and religating the vector. The T2A sequence from Thosea asigna virus was introduced into the NotI site of pBKS-H2B-Venus-neo by annealing the following oligos 2A_fwd (5-GGCCGCACGCGTTTGAAGGTAGAGGCTCTTTACTAACATGCGGCGACGTTGAGGAAAAC CCAGGACC-3) and 2A_rev (5-GGCCTGGTCCTGGGTTTTCCTCAACGTCGCCGCATGTTAGTAAAGAGCCTCTACCTTC AAACGCGTGC-3), which created a NotI compatible overhang resulting in pBKS-2A-H2B-Venus-neo. To clone the NLS-lacZ (nuclear localisation signal-b-galactosidase fusion protein) in front of the H2B-Venus construct we amplified the NLS-lacZ by PCR out of a NLS-lacZ containing vector. We used the fwd primer 340 (5-NNNGCGGCCGCGCCACCATGAACCTTGAAGCTCGAAAAACAAAG-3) with a NotI site at the 5 end and the rev primer 341 (5-NNNGGCGCGCCTTTTTGACACCAGACCAACTGGTAATGGTAGC-3), containing an Ascl site at the 3 end. The PCR product was digested with NotI and Ascl and ligated into the NotI and MluI digested pBKS-2A-H2B-Venus-neo vector resulting in pBKS-NLS-lacZ-2A-H2B-Venus-neo. For finishing the minitargeting construct we cloned pBKS-NLS-lacZ-2A-H2B-Venus-neo into pBKS-Fltp-HomArms (both cut with NotI and SalI). The minitargeting construct was cut out by SacII and KpnI, electroporated in EL350 bacteria and introduced into PL254 via bacterial homologues recombination resulting in the final targeting construct (PL254-Fltp-NLS-lacZ-2A-H2B-Venus-intron-SV40pA-loxP-bGHpA-neo-EM7-PGK-loxP) which was confirmed by sequencing and is ready for electroporating into embryonic stem (ES) cells (after linearization by Ascl).

EXAMPLE 10: FLTP-VENUS FUSION KNOCK-IN STRATEGY AND CONFIRMATION

(75) The knock-in construct was designed as shown in FIG. 10. The Fltp retrieval vector was generated as described in example 9.

(76) For cloning of the knock-in cassette in pBKS5 and 3 HR for the knock-in into the ATG of exon two of Fltp were generated by PCR with following primers: 1228 fwd 5 HR: 5-NNNGCGGCCGCGGTTGGATTCTGAGGCTGACTGGG-3, and 1229 rev 5 HR: 5-NNNTCTAGACTTGGTGCTCTTACAAGGGCTCGG-3, digested with NotI and XbaI; EP1230 fwd 3 HR: 5-NNNGAATTCGTCCTAGTCTAGCTGAGGTCCAGATCTATG-3, and EP1231 rev 3 HR: 5-NNNAAGCTTGTGGGAGTCACTGACATTCTTGTTAACC-3, digested with EcoRI and HindIII using a C57BL/6J BAC clone (RP23-333P11) as template. The STOP codon of the 5 HR was excluded resulting in a fusion construct with introduced downstream sequences. The 5 and 3 HR PCR products were subcloned into pBKS-using the introduced restriction sites, resulting in a plasmid named pBKS-Fltp-Hom Arms.

(77) To introduce the Venus fusion reporter gene and the 3 FLAG tag into the targeting construct the Venus sequence was amplified from pBKS-Venus vector (Nagai, Ibata et al. 2002) using primers 1126 fwd GCGGCCGCAGCCACCATGTCTAGAAT GGTGAGCAAGGGCGAGGAGCTGTTC-3) containing an XbaI site and 1201 rev (5-NNNACTAGTTCACTTGTCATCGTCATCCTTGTAATCGATGTCATGATCTTTATAATCACCG TCATGGTCTTTGTAGTCCTTGTACAGCTCGTCCATGCCGAGAGTGATCC-3) containing a SpeI site and the C-terminal 3 FLAG tag sequence. The resulting open reading frame of Venus-3 FLAG was cloned in frame with Fltp sequence into the pBKS-Fltp-HomArms vector after digestion with XbaI and SpeI.

(78) In the next step, the PGK promoter-driven neomycin resistance gene flanked by IoxP sites (loxP-bGHpA-neo-EM7-PGK-loxP) was cloned from the PL-452 vector (Liu, Jenkins et al. 2003) via BamHI, EcoRI and subcloned 3 of the Venus-3 FLAG sequence resulting in the pBKS-Fltp-HomArms-Venus-3 FLAG-fusion-loxP-Neo-loxP. Subsequently, the mini-targeting cassette was cut with NotI and HindIII and introduced into the Fltp retrieval vector pL254 via bacterial homologous recombination in EL350 bacteria resulting in final targeting construct pL254-Fltp-Venus-3 FLAG-fusion-loxP-bGHpA-neo-EM7-PGK-loxP. This was confirmed by sequencing. The targeting vector was linearized with Ascl and electroporated into ES cells.

EXAMPLE 11: FLTP ANTIBODY EPITOPES AND BINDING SPECIFICITY

(79) Fltp antibodies were generated as described previously (Lange, Gegg et al. 2012). To analyze the Fltp protein biochemically and cell biologically, two affinity purified polyclonal antibodies were raised in rabbit (Fltp1, Fltp116-1) against mouse Fltp using the peptide sequence: DNPDEPQSSHPSAGHT for Fltp1 and KPFDPDSQTKQKKSVTKTVQ for Fltp116-1 (Pineda, Berlin, Germany). The Fltp1 epitope lies in the PRR of the less well conserved C terminal part of the Fltp protein (FIG. 11 B). Nevertheless, the human and murine sequences are nearly completely similar. The Fltp116-1 epitope lies N terminal to the Fltp1 epitope and is less conserved in human.

(80) Additional rat monoclonal antibodies (clone #13, #28 and #43) against the human FLTP were prepared using the peptide sequence KPHDPDSQKKLRKKSITKTVQ (FIG. 12). EndoC- H1 were culture in adherence (2D culture) as described previously (Ravassard et al. (2011) and were stained as follows: cells were fixed for 10 min in 4% paraformaldehyde (PFA) at 37 C., washed 3 with PBS containing 0.2% Tween and 0.3% BSA (washing buffer) at room temperature (RT), and were permeabilized for 30 min on a shaker at RT. Subsequently the cells were washed again 3 with washing buffer at RT, and blocked for 1 h at RT in blocking buffer (PBS including 0.02% Tween, 10% FCS, 0.2% BSA, 3% serum). Then the primary antibody was added in blocking solution and incubated over night at 4 C. The cells were washed 3 in washing buffer, incubated with mouse IgG2b anti Rat IgG2b antibody (dilution 1:2) for 2 h at RT. Then the cells were washed 3 in washing buffer, incubated with donkey anti-mouse IgG 488 (Invitrogen, A21202) (dilution 1:800) for 2 h at RT, washed 3 in washing buffer and embedded in Elvanol for subsequent imaging. Imaging was performed with a Leica SP5 confocal microscope according to the manufacturer's guidelines. The 63 glycerol objective was used. FIG. 12A shows that endogenous FLTP protein is localized in the cytoplasm of EndoC H1 human -cells.

(81) For western blotting experiments (see FIG. 12B), HEK293T cells transiently transfected with Flag-tagged FLTP Strep were used, while non-transfected HEK293T cells were used as controls. The human FLTP coding sequence was obtain by PCR using cDNA from EndoC- H1 human -cells. The following primers were used:

(82) Forward (NotI) 5-GCGGCCGCGCCACCATGGCCACTAACTACAGTGCCAAC-3

(83) Reverse (Eco-RI) 5-NNNNGAATTCTAAGGATTTGGCTGGTCTTTGGGGACC-3.

(84) NotI and Eco-RI restriction enzyme were used to clone human FLTP coding sequence into the pCAG Strep Flag-Tag plasmid and to generate a FLTP Strep Flag-tagged plasmid. HEK 293T were transiently transfected with FLTP Strep Flag-tagged plasmid using polyethylenimine (PEI).

(85) For Western blotting, the samples were resuspended in RIPA buffer with protein inhibitor and incubated on ice for 20 min. The cell lysates were centrifuged at 14000 rpm for 30 min at 4 C. and the supernatant containing the protein was collected. Protein concentration was determined by using the Bradford Assay. For each sample, 20 g of protein were mixed 1:4 with 4SDS loading buffer with dithiothreitol (DTT) and denatured at 95 C. for 5 min. Proteins were separated on denaturing SDS polyacrylamide gel (10%) at 125V for approximately 1.5 h. The protein was transferred to nitrocellulose membrane by Semi-dry Blot for 30 min at 0.44 mA and 25V. Afterwards, the membrane was incubated in Ponceau-S solution to confirm the successful transfer and washed in PBS-T (PBS+0.2% Tween) to remove the color. The membrane was blocked with 5% milk powder in PBS-T for 2 h. Subsequently, the membrane was incubated with the primary antibody in PBS-T overnight at 4 C. while rolling. After washing the membrane 3 for 10 min with PBS-T, the secondary antibody in PBS-T was added and incubated for 1 h at RT. The membrane was washed again 3 with PBS-T while shaking. The ECL (enhanced chemiluminescent solution) was prepared by mixing solution 1 and 2 at a ratio of 1:1. The membrane was rinsed with the solution before wrapping it into plastic foil and placing it into an X-ray film cassette. The membrane was exposes to a film for 10s-5 min before the film was developed.

EXAMPLE 12: FLTP MRNA EXPRESSION IN ENDOC- H1 HUMAN 3-CELLS

(86) To investigate FLTP mRNA expression levels in human -cells cultured under 3D and 2D conditions, EndoC- H1 human -cells were cultured in Matrigel (3D) and compared to 2D conditions. EndoC- H1 human -cell line was culture in adherence (2D culture) as described previously (Ravassard et al. 2011). For 3D matrigel based cultures, EndoC- H1 were cultured in Matrigel Matrix Growth Factor Reduced (BD Bioscience, Germany) diluted 1:2 in their respective medium.

(87) For mRNA isolation, the miRNeasy Micro Kit (Qiagen) was used. First, 700 l QIAzol Lysis Reagent was added to the cell pellet of up to 1 million cells. The sample was disrupted by pipetting and vortexing and incubated for 5 min at RT. 140 l Chloroform was added to the sample and then vortexed for 15s and incubated for 3 min at RT. After 15 min of centrifuging at 12,000g at 4 C., the upper aqueous phase was collected in a new tube. 100% Ethanol was added to 1.5 times volume of the sample and mixed by pipetting. 700 l of the sample was transferred to the RNeasy Min Elute spin column in a 2 ml collection tube and centrifuged at 8000g for 15 s. The flow through was discarded. The DNase digest was performed by mixing 10 l DNaseI and 70 l RDD buffer and pipetting on the membrane of the column and incubated for 15 min at RT. 700 l Buffer RWT was added to the column and centrifuged for another 15s at 8000g. The flow-through was discarded, 500 l RPE Buffer was added and the column was centrifuged for 15 s at 8000g. After discarding the flow-through, 500 l of 80% ethanol was added and the column was centrifuged for 2 min at 8000g. The flow-through and the collection tube were discarded again and the column was placed in another tube. For drying the membrane, the column was centrifuged at full speed for 5 min with open lid. Flow-through and collection tube were discarded and the column was placed in a new collection tube. The RNA was eluted by placing 14 l RNase-free water in the center of the membrane and centrifuged at full speed for 1 min to collect the RNA. Purity and concentration of isolated RNA was determined by using nanodrop. For each sample, 1 g of RNA was synthesized into cDNA by using the Super Script Kit (Invitrogen, Germany). 4 l of 5 Vilo Reaction Mix and 2 l of Super Skript Enzyme Mix was added to the RNA and filled up to 20 l with nuclease-free water. The mixture in the Eppendorf tube was placed into a heat block with the following program: 25 C. 10 min, 42 C. 60 min 85 C. 5 min.

(88) TaqMan qPCR was assessed according with the manufacture instruction and the following probes were used GAPdh Hs-02758991_g1 and (Fltp) C1orf192 Hs01595277_g1.

(89) As is shown in FIG. 13, FLTP mRNA expression in EndoC- H1 human -cells is increased when these cells form mini-islets and cellular connections.

EXAMPLE 13: THE EXPRESSION OF FLTP CORRELATES WITH THE EXPRESSION OF OTHER MATURATION MARKERS AND IS INDUCED BY COMPACTION AND ADDITION OF THE NONCANONICAL WNT LIGAND WNT5A

(90) To confirm that FLTP is a novel maturation marker for cells, FLTP expression was correlated with the expression of another maturation marker, namely the maturation marker NKX6.1 in EndoC- H cells (described e.g. in Ravassard et al. 2011), as well as in re-aggregated postnatal islets (Ucn3).

(91) EndoC- H1 human -cell line were cultured in adherence (2D culture) as described previously (Ravassard et al. (2011)). For 3D matrigel based cultures EndoC- H1 were cultured in Matrigel Matrix Growth Factor Reduced (BD Bioscience, Germany) diluted 1:2 in their respective medium. For analysis of WNT5a induced -cell maturation, samples were stimulated with 400 ng/ml of WNT5a (R&D systems, Germany) for 12 h or 3 days.

(92) Staining of EndoC- H cells was performed as follows: cells were fixed for 10 min in 4% PFA at 37 C., washed 3 with PBS containing 0.2% Tween and 0.3% BSA (washing buffer) at RT, and permeabilized for 30 min on a shaker at RT. Subsequently the cells were washed again 3 with washing buffer at RT, blocked for 1 h at RT in blocking buffer (PBS including 0.02% Tween, 10% FCS, 0.2% BSA, 3% serum). Then the primary antibody was added in blocking solution and incubated over night at 4 C. The cells were washed 3 in washing buffer, incubated with the secondary antibody for 2 h at RT, washed 3 in washing buffer and embedded in Elvanol for subsequent imaging. Imaging was performed with a Leica SP5 confocal microscope according to the manufacturer's guidelines. The 63 glycerol objective was used.

(93) Murine pancreatic islets of Langerhans were isolated from 5 day old mice. After decapitation of the animals, the pancreas was dissected and supplemented with collagenase P (1 mg/ml) in HBSS (Hanks balanced salt solution) including 10% BSA (G-solution) and incubated for 15 min at 37 C. The pancreas was washed 2 in G-solution at RT and the islets were picked under the stereomicroscope (Leica). The islets were incubated in RPM 1640 plus 5% FCS (fetal calf serum) and 1% P/S (penicillin and streptomycin). Afterwards the islets were trypsinized with 0.05% Trypsin 15 at 37 C., washed twice with PBS plus 0.3% BSA and then seeded in iBidi 8 well chambers. For immunohistochemistry the protocol mentioned above was used. Fluorescent intensity was calculated by the Leica LAS-AF (Version 2.7.3.9723) software and the graphs were plotted with Excel.

(94) As is shown in FIGS. 14 and 15, maturation marker expression increases upon compaction of single cells to mini-islets in EndoC- H cells (FIG. 14a-d; FIG. 15a-d).

(95) Taken together, these findings of a correlation of up-regulated FLTP expression and another cell maturation marker confirm that FLTP is a novel maturation marker for cell maturation.

(96) To additionally confirm that Wnt/PCP signaling is important for the induction of -cell maturation, EndoC- H1 and neonatal isolated islets were stimulated with the noncanonical Wnt ligand Wnt5a. As shown in FIG. 14 e-l and FIG. 15 e-h, Wnt5a stimulation of also resulted in increased expression of maturation markers.

FURTHER REFERENCES

(97) Adams, M. D., Fields, C., Venter, J. C. (1994). Automated DNA Sequencing and Analysis. Academic Press. Alphey (1997). DNA Sequencing: from experimental methods to bioinformatics. Springer Verlag Publishing. Andrikopoulos, S., A. R. Blair, N. Deluca, B. C. Fam and J. Proietto (2008). Evaluating the glucose tolerance test in mice. Am J Physiol Endocrinol Metab 295(6): E1323-1332. Ayala, J. E., V. T. Samuel, G. J. Morton, S. Obici, C. M. Croniger, G. I. Shulman, D. H. Wasserman and O. P. McGuinness (2010). Standard operating procedures for describing and performing metabolic tests of glucose homeostasis in mice. Dis Model Mech 3(9-10): 525-534. Baarsma, H. A., M. Konigshoff and R. Gosens (2013). The WNT signaling pathway from ligand secretion to gene transcription: molecular mechanisms and pharmacological targets. Pharmacol Ther 138(1): 66-83. Bertschinger, J., D. Grabulovski and D. Neri (2007). Selection of single domain binding proteins by covalent DNA display. Protein Eng Des Sel 20(2): 57-68. Beste, G., F. S. Schmidt, T. Stibora and A. Skerra (1999). Small antibody-like proteins with prescribed ligand specificities derived from the lipocalin fold. Proc Natl Acad Sci USA 96(5): 1898-1903. Blum, B., S. S. Hrvatin, C. Schuetz, C. Bonal, A. Rezania and D. A. Melton (2012). Functional beta-cell maturation is marked by an increased glucose threshold and by expression of urocortin 3. Nat Biotechnol 30(3): 261-264. Bonner-Weir, S. (1988). Morphological evidence for pancreatic polarity of beta-cell within islets of Langerhans. Diabetes 37(5): 616-621. Brenk, R., A. Schipani, D. James, A. Krasowski, I. H. Gilbert, J. Frearson and P. G. Wyatt (2008). Lessons learnt from assembling screening libraries for drug discovery for neglected diseases. ChemMedChem 3(3): 435-444. Cole, S. P. O (1985). The EBV-Hybridoma Technique and its Application to human Lung Cancer. Monoclonal Antibodies and Cancer Therapy 77-96. Cooper G. M. (2000) Tools of cell biology. Sinauer Associates, Inc. Eberhard, D. and E. Lammert (2009). The pancreatic beta-cell in the islet and organ community. Curr Opin Genet Dev 19(5): 469-475. Eiraku, M., N. Takata, H. Ishibashi, M. Kawada, E. Sakakura, S. Okuda, K. Sekiguchi, T. Adachi and Y. Sasai (2011). Self-organizing optic-cup morphogenesis in three-dimensional culture. Nature 472(7341): 51-56. Eisenberg, E. and E. Y. Levanon (2003). Human housekeeping genes are compact. Trends Genet 19(7): 362-365. Ezan, J. and M. Montcouquiol (2013). Revisiting planar cell polarity in the inner ear. Semin Cell Dev Biol 24(5): 499-506. Feldwisch, J. and V. Tolmachev (2012). Engineering of affibody molecules for therapy and diagnostics. Methods Mol Biol 899: 103-126. Gao, T., B. McKenna, C. Li, M. Reichert, J. Nguyen, T. Singh, C. Yang, A. Pannikar, N. Doliba, T. Zhang, D. A. Stoffers, H. Edlund, F. Matschinsky, R. Stein and B. Z. Stanger (2014). Pdx1 Maintains beta Cell Identity and Function by Repressing an alpha Cell Program. Cell Metab 19(2): 259-271. Gebauer, M. and A. Skerra (2009). Engineered protein scaffolds as next-generation antibody therapeutics. Curr Opin Chem Biol 13(3): 245-255. Grabulovski, D., M. Kaspar and D. Neri (2007). A novel, non-immunogenic Fyn SH3-derived binding protein with tumor vascular targeting properties. J Biol Chem 282(5): 3196-3204. Granot, Z., A. Swisa, J. Magenheim, M. Stolovich-Rain, W. Fujimoto, E. Manduchi, T. Miki, J. K. Lennerz, C. J. Stoeckert, Jr., O. Meyuhas, S. Seino, M. A. Permutt, H. Piwnica-Worms, N. Bardeesy and Y. Dor (2009). LKB1 regulates pancreatic beta cell size, polarity, and function. Cell Metab 10(4): 296-308. Haque, K., J. Hehir, J. C. Fox, C. R. Newton and S. Little (1998). Amplification refractory mutation system linear extension: a novel, gel-free, enzyme-linked immunoassay method for DNA genotyping. Diagn Mol Pathol 7(5): 248-252. Harlow, E. and D. Lane (2006). Bradford assay. CSH Protoc 2006(6). Harlow, E. and D. Lane (1988). Antibodies, A Laboratory Manual. Cold spring Harbor Laboratory Press. Harlow, E. and D. Lane (1999). Using Antibodies, A Laboratory Manual. Cold spring Harbor Laboratory Press. Heimberg, H., A. De Vos, A. Vandercammen, E. Van Schaftingen, D. Pipeleers and F. Schuit (1993). Heterogeneity in glucose sensitivity among pancreatic beta-cells is correlated to differences in glucose phosphorylation rather than glucose transport. Embo 112(7): 2873-2879. Holzgraben and Bechtold (2000). Deutsche Apotheker Zeitung 140(8): 813. Huber, A. B., A. Kania, T. S. Tran, C. Gu, N. De Marco Garcia, I. Lieberam, D. Johnson, T. M. Jessell, D. D. Ginty and A. L. Kolodkin (2005). Distinct roles for secreted semaphorin signaling in spinal motor axon guidance. Neuron 48(6): 949-964. In't Veld, P. and M. Marichal (2010). Microscopic anatomy of the human islet of Langerhans. Adv Exp Med Biol 654: 1-19. Kakavas, K. V., A. Noulas, C. Chalkias, C. Hadjichristodoulou, I. Georgiou, E. Georgatsou and S. Bonanou (2006). Identification of the four most common beta-globin gene mutations in Greek beta-thalassemic patients and carriers by PCR-SSCP: advantages and limitations of the method. J Clin Lab Anal 20(1): 1-7. Kalwat, M. A. and D. C. Thurmond (2013). Signaling mechanisms of glucose-induced F-actin remodeling in pancreatic islet beta cells. Exp Mol Med 45: e37. Katsuta, H., C. Aguayo-Mazzucato, R. Katsuta, T. Akashi, J. Hollister-Lock, A. J. Sharma, S. Bonner-Weir and G. C. Weir (2012). Subpopulations of GFP-marked mouse pancreatic beta-cells differ in size, granularity, and insulin secretion. Endocrinology 153(11): 5180-5187. Kohler, G. and C. Milstein (1975). Continuous cultures of fused cells secreting antibody of predefined specificity. Nature 256(5517): 495-497. Kolb, H. (1987). Mouse models of insulin dependent diabetes: low-dose streptozocin-induced diabetes and nonobese diabetic (NOD) mice. Diabetes Metab Rev 3(3): 751-778. Kozbor, D. and Roder, J. C. (1983). The Production of monoclonal antibodies from human lymphocytes. Immunology today 4: 72-79. Kubinyi, H. (1992). Hansch Analysis and related Approaches. VCH Verlag, Weiheim. Landsman, L., A. Parent and M. Hebrok (2011). Elevated Hedgehog/Gli signaling causes beta-cell dedifferentiation in mice. Proc Natl Acad Sci USA 108(41): 17010-17015. Lange, A., M. Gegg, I. Burtscher, D. Bengel, E. Kremmer and H. Lickert (2012). Fltp(T2AiCre): a new knock-in mouse line for conditional gene targeting in distinct mono- and multiciliated tissues. Differentiation 83(2): S105-113. Lawrence, P. A. and J. Casal (2013). The mechanisms of planar cell polarity, growth and the Hippo pathway: some known unknowns. Dev Biol 377(1): 1-8. Liao, W. P., L. Uetzmann, I. Burtscher and H. Lickert (2009). Generation of a mouse line expressing Sox17-driven Cre recombinase with specific activity in arteries. Genesis 47(7): 476-483. Lickert, H. (2013). Betatrophin fuels beta cell proliferation: first step toward regenerative therapy? Cell Metab 18(1): 5-6. Liu, P., N. A. Jenkins and N. G. Copeland (2003). A highly efficient recombineering-based method for generating conditional knockout mutations. Genome Res 13(3): 476-484. Masters, J. R. W. (2000). Animal cell culture Oxford University Press. Matis, M. and J. D. Axelrod (2013). Regulation of PCP by the Fat signaling pathway. Genes Dev 27(20): 2207-2220. Melani, C., L. Rivoltini, G. Parmiani, B. Calabretta and M. P. Colombo (1991). Inhibition of proliferation by c-myb antisense oligodeoxynucleotides in colon adenocarcinoma cell lines that express c-myb. Cancer Res 51(11): 2897-2901. Meng, H., K. Hager, S. A. Rivkees and J. R. Gruen (2005). Detection of Turner syndrome using high-throughput quantitative genotyping. J Clin Endocrinol Metab 90(6): 3419-3422. Mouratou, B., G. Behar, L. Paillard-Laurance, S. Colinet and F. Pecorari (2012). Ribosome display for the selection of Sac7d scaffolds. Methods Mol Biol 805: 315-331. Murray, G. I. (2007). An overview of laser microdissection technologies. Acta Histochem 109(3): 171-176. Muzumdar, M. D., B. Tasic, K. Miyamichi, L. Li and L. Luo (2007). A global double-fluorescent Cre reporter mouse. Genesis 45(9): 593-605. Nagai, T., K. Ibata, E. S. Park, M. Kubota, K. Mikoshiba and A. Miyawaki (2002). A variant of yellow fluorescent protein with fast and efficient maturation for cell-biological applications. Nat Biotechnol 20(1): 87-90. Newton, C. R., A. Graham, L. E. Heptinstall, S. J. Powell, C. Summers, N. Kalsheker, J. C. Smith and A. F. Markham (1989). Analysis of any point mutation in DNA. The amplification refractory mutation system (ARMS). Nucleic Acids Res 17(7): 2503-2516. Niehrs, C. (2012). The complex world of WNT receptor signalling. Nat Rev Mol Cell Biol 13(12): 767-779. Pagliuca, F. W. and D. A. Melton (2013). How to make a functional beta-cell. Development 140(12): 2472-2483. Pan, F. C. and C. Wright (2011). Pancreas organogenesis: from bud to plexus to gland. Dev Dyn 240(3): 530-565. Pissard, S., L. T. Huynh, J. Martin and M. Goossens (2002). HFE genotyping by amplification refractory mutation system-denaturing HPLC. Clin Chem 48(5): 769-772. Puri, S., H. Akiyama and M. Hebrok (2013). VHL-mediated disruption of Sox9 activity compromises beta-cell identity and results in diabetes mellitus. Genes Dev 27(23): 2563-2575. Quinn, R. J., A. R. Carroll, N. B. Pham, P. Baron, M. E. Palframan, L. Suraweera, G. K. Pierens and S. Muresan (2008). Developing a drug-like natural product library. J Nat Prod 71(3): 464-468. Ramon, D., M. Braden, S. Adams, F. M. Marincola and L. Wang (2003). Pyrosequencing trade mark: A one-step method for high resolution HLA typing. J Transl Med 1(1): 9. Ravassard, P., Y. Hazhouz, S. Pechberty, E. Bricout-Neveu, M. Armanet, P. Czernichow and R. Scharfmann (2011). A genetically engineered human pancreatic beta cell line exhibiting glucose-inducible insulin secretion. J Clin Invest 121(9): 3589-3597. Salomon, D. and P. Meda (1986). Heterogeneity and contact-dependent regulation of hormone secretion by individual B cells. Exp Cell Res 162(2): 507-520. Sasai, Y. (2013). Cytosystems dynamics in self-organization of tissue architecture. Nature 493(7432): 318-326. Sawada, A., Y. Nishizaki, H. Sato, Y. Yada, R. Nakayama, S. Yamamoto, N. Nishioka, H. Kondoh and H. Sasaki (2005). Tead proteins activate the Foxa2 enhancer in the node in cooperation with a second factor. Development 132(21): 4719-4729. Schier, R. and J. D. Marks (1996). Efficient in vitro affinity maturation of phage antibodies using BIAcore guided selections. Hum Antibodies Hybridomas 7(3): 97-105. Schlatter, D., S. Brack, D. W. Banner, S. Batey, J. Benz, J. Bertschinger, W. Huber, C. Joseph, A. Rufer, A. van der Klooster, M. Weber, D. Grabulovski and M. Hennig (2012). Generation, characterization and structural data of chymase binding proteins based on the human Fyn kinase SH3 domain. MAbs 4(4): 497-508. Seifert, J. R. and M. Mlodzik (2007). Frizzled/PCP signalling: a conserved mechanism regulating cell polarity and directed motility. Nat Rev Genet 8(2): 126-138. Smukler, S. R., M. E. Arntfield, R. Razavi, G. Bikopoulos, P. Karpowicz, R. Seaberg, F. Dai, S. Lee, R. Ahrens, P. E. Fraser, M. B. Wheeler and D. van der Kooy (2011). The adult mouse and human pancreas contain rare multipotent stem cells that express insulin. Cell Stem Cell 8(3): 281-293. Steemers, F. J., W. Chang, G. Lee, D. L. Barker, R. Shen and K. L. Gunderson (2006). Whole-genome genotyping with the single-base extension assay. Nat Methods 3(1): 31-33. Stefan, Y., P. Meda, M. Neufeld and L. Orci (1987). Stimulation of insulin secretion reveals heterogeneity of pancreatic B cells in vivo. J Clin Invest 80(1): 175-183. Talchai, C., S. Xuan, H. V. Lin, L. Sussel and D. Accili (2012). Pancreatic beta cell dedifferentiation as a mechanism of diabetic beta cell failure. Cell 150(6): 1223-1234. Tung, J. W., K. Heydari, R. Tirouvanziam, B. Sahaf, D. R. Parks, L. A. Herzenberg and L. A. Herzenberg (2007). Modern flow cytometry: a practical approach. Clin Lab Med 27(3): 453-468. Turksen, K. (2004). Adult Stem cells. Humana Press Inc., Totowa. Velculescu, V. E., S. L. Madden, L. Zhang, A. E. Lash, J. Yu, C. Rago, A. Lal, C. J. Wang, G. A. Beaudry, K. M. Ciriello, B. P. Cook, M. R. Dufault, A. T. Ferguson, Y. Gao, T. C. He, H. Hermeking, S. K. Hiraldo, P. M. Hwang, M. A. Lopez, H. F. Luderer, B. Mathews, J. M. Petroziello, K. Polyak, L. Zawel, K. W. Kinzler and et al. (1999). Analysis of human transcriptomes. Nat Genet 23(4): 387-388. Wallingford, J. B. (2012). Planar cell polarity and the developmental control of cell behavior in vertebrate embryos. Annu Rev Cell Dev Biol 28: 627-653. Wallingford, J. B. and B. Mitchell (2011). Strange as it may seem: the many links between Wnt signaling, planar cell polarity, and cilia. Genes Dev 25(3): 201-213. Wang, Y. and J. Nathans (2007). Tissue/planar cell polarity in vertebrates: new insights and new questions. Development 134(4): 647-658. Weedon, M. N., I. Cebola, A. M. Patch, S. E. Flanagan, E. De Franco, R. Caswell, S. A. Rodriguez-Segui, C. Shaw-Smith, C. H. Cho, H. Lango Allen, J. A. Houghton, C. L. Roth, R. Chen, K. Hussain, P. Marsh, L. Vallier, A. Murray, S. Ellard, J. Ferrer and A. T. Hattersley (2014). Recessive mutations in a distal PTF1A enhancer cause isolated pancreatic agenesis. Nat Genet 46(1): 61-64. Weidle, U. H., J. Auer, U. Brinkmann, G. Georges and G. Tiefenthaler (2013). The emerging role of new protein scaffold-based agents for treatment of cancer. Cancer Genomics Proteomics 10(4): 155-168. Yang, Y. (2012). Wnt signaling in development and disease. Cell Biosci 2(1): 14.