Automatic texture-based analysis in ultrasound imaging of ovarian masses

PURPOSE

To assess the diagnostic accuracy of a new automatic texture-based algorithm (ATBA) in ultrasound imaging of ovarian masses and to compare its performance to subjective assessment by examiners with different levels of ultrasound experience.

MATERIALS AND METHODS

A total of 105 ultrasound images from three different groups of ovarian lesions (malignancies, functional cysts, and dermoid cysts) were evaluated using ATBA and by a total of 36 examiners with four different levels of experience (9 junior trainees, 8 senior trainees, 11 senior gynecologists, and 8 experts). Cohen's κ, Youden's indices, and the sensitivity and specificity of ATBA and of each observer were calculated for every subgroup of ovarian lesions.

RESULTS

ATBA classified 78 of the 105 masses correctly (κ = 0.62) - results that were significantly better than those of the junior and senior trainees (p = 0.02 and p < 0.01), while differences from the group of level II examiners did not reach statistical significance (p = 0.27). The best diagnostic performance (κ = 0.70) was obtained by the group of expert level III ultrasonographers. The best classification rates overall, including both ATBA and subjective assessments, were achieved in the detection of functional cysts (Youden's indices from 0.73 to 0.85), while the poorest diagnostic performance was obtained for the classification of dermoid cysts (Youden's indices from 0.28 to 0.55).

CONCLUSION

ATBA showed a significantly better diagnostic performance than observers with low or medium levels of experience, emphasizing its potential value for training purposes and in providing additional diagnostic assistance for inexperienced observers.

424 DOMINANT-NEGATIVE GLUCOSE-DEPENDENT INSULINOTROPIC POLYPEPTIDE (GIP) RECEPTOR (GIPRdn)TRANSGENIC PIGS — A LARGE ANIMAL MODEL FOR TYPE 2 DIABETES MELLITUS

The incretin hormones glucose-dependent insulinotropic polypeptide (GIP) and glucagon-like peptide-1 (GLP-1) are secreted in response to nutrients and enhance glucose-induced insulin secretion. The insulinotropic action of GIP is impaired in type 2 diabetes (T2D) whereas that of GLP-1 is preserved. To evaluate the role of an impaired GIP function in the pathogenesis of T2D in a large animal model, we generated transgenic pigs expressing a dominant-negative GIP receptor (GIPRdn) in the pancreatic islets. GIPRdn transgenic pigs were generated using lentiviral transgenesis. Metabolic tests and quantitative stereological analyses of the pancreas were performed in 3 different age groups to investigate the effects of an impaired insulinotropic action of GIP on glucose metabolism and pancreas morphology. The insulinotropic action of GIP was significantly reduced, whereas insulin secretion in response to the GLP-1 receptor agonist exendin-4 was enhanced in 11-week-old GIPRdn transgenic pigs compared with control pigs. Eleven-week-old GIPRdn transgenic pigs (n = 5) exhibited significantly reduced oral glucose tolerance (P < 0.05) with a delay in insulin secretion compared with controls (n = 5). The area under the insulin curve (AUC) during the first 45 min following glucose load was 31% smaller (P < 0.05) in transgenic pigs compared with controls. The total insulin secretion capacity was not different between the 2 groups indicating that GIPRdn expression initially only interferes with the incretin effect. This was supported by the fact that intravenous glucose tolerance and insulin secretion in transgenic pigs were not different from controls. Five-month-old GIPRdn transgenic pigs revealed markedly reduced insulin secretion in response to oral glucose challenge (P < 0.01), resulting in significantly elevated glucose levels (P < 0.05). Also, intravenous glucose tolerance and insulin secretion were diminished in 11-month-old transgenic pigs. To determine the reason for the alterations in glucose metabolism, quantitative-stereological analyses of the pancreas were performed. In 11-week-old pigs, transgenic and control groups showed similar β-cell mass (n = 5 in each group). However, pancreatic β-cell mass was reduced by almost 40% (P < 0.05) in 5-month-old pigs and by 60% (P < 0.01) in adult (1 to 1.4 years) GIPRdn transgenic pigs compared with controls. To investigate the reason for the progressive reduction of pancreatic β-cell mass in GIPRdn transgenic pigs, β-cell proliferation rate was determined performing a double-immunohistochemistry for insulin and the proliferation marker Ki67. Eleven-week-old GIPRdn transgenic pigs showed significantly fewer Ki67-positive cell nuclei compared with controls (P < 0.05). In conclusion, GIPRdn transgenic pigs exhibit a comparable situation to T2D, such as impaired insulinotropic action of GIP, disturbed oral and intravenous glucose tolerance, and progressive reduction of β-cell mass. These alterations are at least partly attributable to diminished proliferation of β-cells. Grant support: Deutsche Forschungsgemeinschaft (GRK 1029), Bayerische Forschungsstiftung (492/02).

433 NEW TRANSGENIC PIG DIABETES MODEL EXPRESSING THE MUTANT Insulinc93S FOR THE STUDY OF THE PANCREATIC BETA CELL DYSFUNCTION

We previously established a mutant mouse line showing diabetes which was caused by a point mutation in the Insulin 2 (Ins2) gene. The point mutation leads to the amino acid exchange C95S and the loss of the A6-A11 intrachain disulfide bond of insulin. Male heterozygous Ins2C95S mutant mice develop progressive diabetes mellitus with strong reduction of the total pancreatic islet volume and the total beta cell volume together with severe alterations of the beta cell structure. As pigs share many anatomical and physiological characteristics with humans, we aimed to establish a transgenic pig model expressing the mutant insulin by additive gene transfer for the subsequent study of beta cell dysfunction in diabetes mellitus. The transgene consisted of German Landrace insulin sequences including 1.3 kb of the insulin promoter and 1 kb insulin gene sequences with the 3 exons and the T to A (C93S) point mutation in exon 3 analogous to the mutant mouse insulin. A floxed neomycin resistance gene at the 3′ end of the transgene was used as a selection marker. After successful transfection of male fetal fibroblasts of both the German Landrace and the Schwaebisch- Haellisch pig breeds, pooled transgenic fibroblasts were used for somatic cell nuclear transfer (SCNT). Five hundred three reconstructed pig embryos were generated and endoscopically transferred to 5 synchronized recipients. One pregnancy with German Landrace genetic background and one pregnancy with Schwaebisch-Haellisch genetic background of the cloned embryos developed to term, which gave rise to 3 living offspring from each of the 2 pregnancies; 5 of the 6 piglets were transgenic. Southern blot analysis showed different transgene signal patterns in all animals examined. Transgenic pigs of both litters revealed unaltered fasting blood glucose levels up to an age of 8 months. However, disturbed intravenous glucose tolerance and reduced insulin secretion were detected in 1 transgenic pig of the first litter at 8 months of age. The area under the glucose curve of this transgenic pig was 75% larger (22 136 v. 12617) and the area under the insulin curve 53% smaller (1250 v. 2670) compared with the control. Ongoing analyses comprise glucose tolerance tests in the second litter as well as pathohistologic analysis of the pancreata of both litters. Cells from suitable transgenic founders will be used for recloning to establish a new transgenic pig model expressing the mutant InsulinC93S for an in-depth study of pancreatic beta cell dysfunction in diabetes mellitus. Supported by the Deutsche Forschungsgemeinschaft (GRK1029).

Assembly and melting of DNA nanotubes from single-sequence tiles

DNA melting and renaturation studies are an extremely valuable tool to study the kinetics and thermodynamics of duplex dissociation and reassociation reactions. These are important not only in a biological or biotechnological context, but also for DNA nanotechnology which aims at the construction of molecular materials by DNA self-assembly. We here study experimentally the formation and melting of a DNA nanotube structure, which is composed of many copies of an oligonucleotide containing several palindromic sequences. This is done using temperature-controlled UV absorption measurements correlated with atomic force microscopy, fluorescence microscopy and transmission electron microscopy techniques. In the melting studies, important factors such as DNA strand concentration, hierarchy of assembly and annealing protocol are investigated. Assembly and melting of the nanotubes are shown to proceed via different pathways. Whereas assembly occurs in several hierarchical steps related to the formation of tiles, lattices and tubes, melting of DNA nanotubes appears to occur in a single step. This is proposed to relate to fundamental differences between closed, three-dimensional tube-like structures and open, two-dimensional lattices. DNA melting studies can lead to a better understanding of the many factors that affect the assembly process which will be essential for the assembly of increasingly complex DNA nanostructures.

4 IMPAIRED INCRETIN EFFECT IN TRANSGENIC PIGS EXPRESSING A DOMINANT NEGATIVE RECEPTOR FOR GLUCOSE-DEPENDENT INSULINOTROPIC POLYPEPTIDE IN THE PANCREATIC ISLETS

Glucose-dependent insulinotropic polypeptide (GIP) and glucagon-like peptide-1 (GLP-1) are secreted by specific cell types in the intestine and are responsible for the so-called incretin effect, the phenomenon that an oral glucose load elicits a higher insulin response than does an intravenous glucose load. In patients with type 2 diabetes mellitus the overall incretin effect is reduced. This fact is mostly attributed to a lowered insulinotropic effect of GIP, while the effect of GLP-1 is preserved. In order to better understand the consequences of impaired function of GIP, knockout mice lacking a functional GIP receptor (GIPR–/–) as well as transgenic mice expressing a dominant negative GIPR (GIPRdn) were established. While GIPR–/– mice show only relatively mild changes in glucose homeostasis, GIPRdn mice display a distinct diabetic phenotype due to disturbed development of the endocrine pancreas (Herbach et al. 2005 Regul. Pept. 125, 103–117). To further clarify the underlying mechanisms, we used a novel, highly efficient gene transfer technology based on lentiviral vectors (Hofmann et al. 2003 EMBO Rep. 4, 1054–1060; Hofmann et al. 2006 Mol. Ther. 13, 59–66) to generate transgenic pigs expressing a GIPRdn under the control of the rat Ins2 promoter (RIP). RIP-GIPRdn transgenic pigs develop normally and do not display diabetes mellitus up to at least one year of age. Weekly measured fasting blood glucose levels in transgenic animals did not show a significant difference compared to control pigs. The same was true for monthly determined fructosamine levels. However, RIP-GIPRdn transgenic pigs exhibited reduced insulin release and higher glucose levels than non-transgenic littermate controls in an oral glucose tolerance test. The area under the curve (AUC) for insulin was 49% smaller (P < 0.01) and the AUC for glucose 26% larger (P < 0.05) in RIP-GIPRdn transgenic pigs (n = 5) than in their non-transgenic littermate controls (n = 5). These findings demonstrate that expression of a GIPRdn, which was shown by RT-PCR in isolated pancreatic islets, disturbs the function of GIP in transgenic pigs. Thus we have created a novel, clinically relevant animal model for studying the roles of the GIP/GIPR system. Quantitative morphological studies of the pancreas are being performed to clarify whether GIPR function is essential for pancreatic islet development and maintenance.