Advances and horizons for artificial intelligence of endoscopic screening and surveillance of gastric and esophageal disease

The development of artificial intelligence in endoscopic assessment of the gastrointestinal tract has shown progressive enhancement in diagnostic acuity. This review discusses the expanding applications for gastric and esophageal diseases. The gastric section covers the utility of AI in detecting and characterizing gastric polyps and further explores prevention, detection, and classification of gastric cancer. The esophageal discussion highlights applications for use in screening and surveillance in Barrett's esophagus and in high-risk conditions for esophageal squamous cell carcinoma. Additionally, these discussions highlight applications for use in assessing eosinophilic esophagitis and future potential in assessing esophageal microbiome changes.

Artificial intelligence and colonoscopy − enhancements and improvements

Artificial intelligence is a technology that processes and analyzes information with reproducibility and accuracy. Its application in medicine, especially in the field of gastroenterology, has great potential to facilitate in diagnosis of various disease states. Currently, the role of artificial intelligence as it pertains to colonoscopy revolves around enhanced polyp detection and characterization. The aim of this article is to review the current and potential future applications of artificial intelligence for enhanced quality of detection for colorectal neoplasia.

Role of microbial dysbiosis in the pathogenesis of esophageal mucosal disease: A paradigm shift from acid to bacteria?

Genomic sequencing, bioinformatics, and initial speciation (e.g., relative abundance) of the commensal microbiome have revolutionized the way we think about the “human” body in health and disease. The interactions between the gut bacteria and the immune system of the host play a key role in the pathogenesis of gastrointestinal diseases, including those impacting the esophagus. Although relatively stable, there are a number of factors that may disrupt the delicate balance between the luminal esophageal microbiome (EM) and the host. These changes are thought to be a product of age, diet, antibiotic and other medication use, oral hygiene, smoking, and/or expression of antibiotic products (bacteriocins) by other flora. These effects may lead to persistent dysbiosis which in turn increases the risk of local inflammation, systemic inflammation, and ultimately disease progression. Research has suggested that the etiology of gastroesophageal reflux disease-related esophagitis includes a cytokine-mediated inflammatory component and is, therefore, not merely the result of esophageal mucosal exposure to corrosives (i.e., acid). Emerging evidence also suggests that the EM plays a major role in the pathogenesis of disease by inciting an immunogenic response which ultimately propagates the inflammatory cascade. Here, we discuss the potential role for manipulating the EM as a therapeutic option for treating the root cause of various esophageal disease rather than just providing symptomatic relief (i.e., acid suppression).

Constitutive expression of heat shock proteins Hsp90, Hsc70, Hsp70 and Hsp60 in neural and non-neural tissues of the rat during postnatal development

Heat shock proteins (Hsps) are a group of highly conserved proteins, that are constitutively expressed in most cells under normal physiological conditions. Previous work from our laboratory has shown that neurons in the adult brain exhibit high levels of Hsp90 and Hsc70 mRNA and protein, as well as basal levels of Hsp70 mRNA. We have now investigated the expression of Hsp90, Hsc70, Hsp60 and Hsp70 in neural and non-neural tissues of the rat during postnatal development, a time of extensive cell differentiation. Western blot analysis revealed constitutive expression of these Hsps early in postnatal development. Developmental profiles of these Hsps suggest that they are differentially regulated during postnatal development of the rat. For example, while levels of Hsp90 decrease somewhat in certain developing brain regions, levels of Hsp60 show a developmental increase, and Hsc70 protein is abundant throughout postnatal neural development. Low basal levels of Hsp70 are also observed in the developing and adult brain. A pronounced decrease in Hsp90 and Hsc70 was observed during postnatal development of the kidney while levels of Hsp60 increased. In addition, tissue-specific differences in the relative levels of these Hsps between brain and non-brain regions were found. Immunocytochemical studies demonstrated a neuronal localization of Hsp90, Hsc70 and Hsp60 at all stages of postnatal development examined as well as in the adult, suggesting a role for Hsps in both the developing and fully differentiated neuron. The developmental expression of subunit IV of cytochrome oxidase was similar to that of Hsp60, a protein localized predominantly to mitochondria.

Time responses of cancellous and cortical bones to sciatic neurectomy in growing female rats

Effects of unilateral sciatic neurectomy on the responses of both cancellous and cortical bones were studied in growing female rats at 0, 1, 4, 8, and 12 weeks after operation. Using double-fluorescent labeling techniques, histomorphometric analyses were performed on longitudinal sections of proximal tibial metaphyseal secondary spongiosa (PTM) and on cross sections of tibial shaft (TX). In PTM, sciatic neurectomy not only inhibited the age-related bone gain, but also reduced the trabecular bone mass by 46%, which was accompanied by decreases in trabecular number, thickness, and node to node density, and an increase in trabecular separation and free end to free end density. The bone loss occurred mainly between 1 and 4 weeks after operation. A sharp increase in bone formation indices was observed during the first week after nerve section. However, these endpoints quickly dropped to levels lower than those of sham-operated controls at 4 weeks, and were not different from the control levels at 8 weeks after operation. Eroded surface increased progressively after sciatic neurectomy during the 12 weeks experimental period. In TX, sciatic neurectomy inhibited the age-related increase in total tissue area that maintained it at the basal control level. However, the cortical bone area in neurectomized legs was lower than that in sham-operated controls. Sciatic neurectomy also stimulated the bone formation indices on both periosteal and endocortical surfaces during the first week after operation. These endpoints declined sharply between 1 and 4 weeks and then maintained at control levels between 8 and 12 weeks post surgery. Endocortical eroded surface increased 1 week after neurectomy, reached the peak at 8 weeks, and then decreased thereafter. These findings suggest that (1) sciatic neurectomy not only inhibited age-related bone gain but also induced marked bone loss in cancellous bone site and inhibited age-related bone gain in cortical bone site, which mainly resulted from the decrease in bone formation and the increase in bone resorption; (2) the changes in both cancellous and cortical bones responded to sciatic neurectomy occurred mostly within the first 4 weeks and stabilized between 8 and 12 weeks after surgical intervention. In conclusion, the unilateral sciatic neurectomized rat is a complex model in which to study osteopenia. Despite sciatic neurectomy being a simple operation, the interactions of skeletal responses to postsurgical regional acceleratory phenomenon (RAP) and disuse and adaptation changes cannot be clearly differentiated. Furthermore, the complications from growth and aging should be avoided.

Contrasting effects of parathyroid hormone and insulin-like growth factor I in an aged ovariectomized rat model of postmenopausal osteoporosis

Agents that exert anabolic effects on bone have generally been tested in young or estrogen-replete animals. It is unclear whether these agents exert similar effects in older ovariectomized (Ovx) animals. In this single study we examined the effects of intermittent (daily) human PTH-(1-34) and continuous infusion of human recombinant IGF-I alone and in combination on bone resorption and formation over a 14 day period in an aged Ovx rat model of postmenopausal osteoporosis (2-year-old rats, Ovx at 1 year). Compared to Ovx controls, PTH treatment increased bone mineral content (BMC) and bone volume and stimulated bone formation but had no effect on bone resorption. In contrast, IGF-I treatment reduced BMC and stimulated resorptive activity as assessed by increases in marrow volume, cortical porosity, osteoclast-positive eroded surfaces, and urinary hydroxyproline excretion. IGF-I had no effect on bone formation, but when combined with PTH, IGF-I blunted the response to PTH on the periosteal and endocortical surfaces. In summary, PTH stimulated bone formation in a manner similar to that observed in younger animals and IGF-I stimulated bone resorption rather than formation and blunted the bone-forming response to PTH. The effects of IGF-I in older Ovx rats may differ from those observed in younger estrogen-replete animals.

Inhibition of bone resorption in vitro by human enkephalinase (EC 3.4.24.11), a neutral metalloendopeptidase

Bone metabolism is regulated by a wide variety of both circulating and locally produced peptides. The activity of such agents must be regulated, and one potential regulating mechanism is the inactivation of these peptides by locally produced proteolytic enzymes. One candidate for such a class of enzymes is enkephalinase (EC 2.3.24.11), a membrane-bound neutral metalloendopeptidase that inhibits the activity of a range of biologically active peptides, including interleukin-1 (IL-1), a potent bone-resorbing agent. In this study, we examined the effects of human enkephalinase on bone resorption in cultures of fetal rat long bones. We found that partially purified and highly purified enkephalinase inhibited bone resorption stimulated by parathyroid hormone (PTH) and IL-1 alpha. The effects on PTH-stimulated resorption were reversible, but enkephalinase did not inhibit prestimulated resorption. Enkephalinase also inhibited resorption induced by the nonpeptide stimulators 1,25-(OH)2D3, retinoic acid, and prostaglandin E2 (PGE2). In addition, preliminary studies confirmed a previous report of the presence of an enkephalinase-like activity in osteoblast-like osteosarcoma cells. These data are consistent with the hypothesis that proteolytic enzymes, such as enkephalinase, may play a role in the local regulation of bone resorption.

Effects of transforming growth factors beta 1 and beta 2 on a mouse clonal, osteoblastlike cell line MC3T3-E1

Transforming growth factors (TGF-beta 1 and TGF-beta 2) are polypeptide growth factors with a wide range of effects on the growth and differentiated function of a variety of cell types. Transforming growth factors of the beta class (TGF-beta) are found in large quantities in bone matrix and are synthesized by osteoblasts. For these reasons, it has been suggested that TGF-beta may play a major role in the regulation of bone cell metabolism. We have studied the effects of porcine TGF-beta 1 and the recently described porcine TGF-beta 2 in a mouse clonal, osteoblastlike cell line MC3T3-E1 that has previously been shown to have many characteristics of osteoblasts. In serum-containing medium, TGF-beta 1 inhibited alkaline phosphatase activity. The inhibition of alkaline phosphatase activity persisted for at least 72 h following a brief (24 h) exposure to TGF-beta 1. TGF-beta 1 also caused a marked change in cell morphology. High doses inhibited collagen synthesis; lower concentrations caused a small increase. Under serum-free conditions, TGF-beta 1 had biphasic effects on alkaline phosphatase activity inhibiting at high but stimulating at low concentrations and had only a slight stimulatory effect on collagen synthesis. Under the experimental conditions used, the effects of TGF-beta 1 on alkaline phosphatase activity and collagen synthesis were independent of effects on cell proliferation. In serum-containing medium, TGF-beta 2 inhibited alkaline phosphatase activity, an effect that was independent of changes in cell proliferation and caused shape changes in an identical fashion to that observed with TGF-beta 1.

Effects of transforming growth factor-beta on osteoblastic osteosarcoma cells

Transforming growth factor-beta (TGF beta), a polypeptide that controls growth and differentiation in many cell types and has recently been found in abundant amounts in bone, was examined for its effects on cells with the osteoblast phenotype using the clonal osteoblastic osteosarcoma cell line ROS 17/2.8. TGF beta increased alkaline phosphatase (AP) activity and the rate of collagen synthesis per cell. Cell proliferation was inhibited, and the morphological appearance of the cells was markedly changed. All effects were observed at concentrations as low as 0.1 ng/ml TGF beta. Increases in AP activity were detectable after 24 h and increased progressively with time. TGF beta increased AP activity under serum-free conditions and during thymidine-induced inhibition of DNA synthesis. The increase in AP activity mediated by TGF beta could be completely inhibited with actinomycin D and cycloheximide. 1,25-Dihydroxyvitamin D3 at 10(-7) M slightly increased AP activity in ROS 17/2.8 cells, but strongly inhibited AP activity when the cells were pretreated with TGF beta. The data suggest that TGF beta stimulates expression of the osteoblastic phenotype in ROS 17/2.8 cells and that TGF beta may be an important regulator of local bone remodeling.

Human synthetic calcitonin gene-related peptide inhibits bone resorption in vitro

The calcitonin gene-related peptide (CGRP) is a peptide which normally circulates. It is encoded by the calcitonin gene, whose precise function is unknown. Since it has recently been shown that human CGRP (hCGRP) lowers plasma calcium levels in both the rat and the rabbit, we examined the in vitro effects of human synthetic CGRP on bone resorption (as measured by 45Ca release) stimulated by PTH, prostaglandin E2, and 1,25-dihydroxyvitamin D3. CGRP caused a dose-dependent inhibition of PTH-stimulated resorption, with 50% inhibition at approximately 5 X 10(-8) M CGRP. The inhibitory effects of CGRP on PTH-mediated bone resorption were not due to toxic effects, since bones preincubated with CGRP for 48 h were subsequently able to respond to PTH. The inhibitory activity of CGRP in the rat was approximately 3 orders of magnitude less potent than that of human calcitonin. In contrast to the effects of calcitonin, a marked inhibition of PTH-stimulated bone resorption was still observed after 96 h in the continued presence of CGRP. CGRP (10(-6)-10(-8) M) also inhibited resorption stimulated by prostaglandin and 1,25-dihydroxyvitamin D3 in a dose-dependent manner, but had no significant effect on basal bone resorption. In conclusion, these data show that hCGRP inhibits hormone-stimulated bone resorption in vitro. Although it is less potent than calcitonin in the rat, CGRP has been shown to have potency comparable to that of calcitonin in other species, and therefore, a role for CGRP as a therapeutic agent in states of increased bone resorption cannot be ruled out.

Stimulation of bone resorption in vitro by synthetic transforming growth factor-alpha

Experiments were conducted to test the hypothesis that tumor-derived transforming growth factor-alpha (TGF-alpha) is responsible for the increased bone resorption and hypercalcemia seen in some malignant diseases. Homogeneous synthetic TGF-alpha prepared by the solid-phase synthesis method stimulated bone resorption directly in vitro in a concentration-dependent manner. Incubation times of 72 hours or more were required to stimulate resorption, which is similar to the time course of bone resorption by epidermal growth factor.

EGF receptor antiserum inhibits bone resorbing activity produced by a rat Leydig cell tumor associated with the humoral hypercalcemia of malignancy

The humoral hypercalcemia of malignancy (HHM) is a syndrome caused by tumor cells releasing unknown circulating factors which stimulate osteoclastic bone resorption. In the D6 variant of the rat Leydig cell tumor model of HHM, we found that tumor extracts and tumor cell conditioned medium contained a macromolecular bone resorbing factor which coeluted on column chromatography with transforming growth factor activity (TGF). This observation led to the hypothesis that the tumor-derived bone resorbing factor was a TGF which interacts with the epidermal growth factor (EGF) receptor. To test this hypothesis, we examined the effects of two classes of antisera to the EGF receptor on bone resorption stimulated by conditioned medium from Leydig D6 tumor cells using organ cultures of fetal rat long bones. The antiserum which blocks the binding of EGF to its receptor inhibited bone resorption stimulated by tumor conditioned medium and by EGF. The second antiserum to the EGF receptor which does not block EGF binding or biological activity had no effect on bone resorption stimulated by either tumor conditioned medium or EGF. Neither antiserum had any effect on bone resorption stimulated by parathyroid hormone (PTH). These results indicate that the tumor-derived bone resorbing factor is dependent upon the availability of EGF receptors for its activity and are consistent with it being a TGF.

Production of a macromolecular bone-resorbing factor by the hypercalcemic variant of the walker rat carcinosarcoma

The Walker rat 256 carcinosarcoma is a spontaneous rat breast tumor which has two unique variants, one of which is always associated with severe hypercalcemia and one which is not associated with hypercalcemia when the tumor is carried im. To determine the factor responsible for hypercalcemia, we established tumor cells derived from both variants in culture and found that the tumor cells from the hypercalcemic variant release an acid-soluble and acid-stable bone-resorbing factor of approximately 25,000 mol wt. Cells from the normocalcemic variant produce no bone-resorbing activity. These data are consistent with the notion that hypercalcemia in this animal model of the humoral hypercalcemia of malignancy is due to a macromolecular bone-resorbing factor which is acid and heat stable and requires disulfide bonds for biological activity.

Failure of parathyroid hormone antagonists to inhibit in vitro bone resorbing activity produced by two animal models of the humoral hypercalcemia of malignancy

The humoral hypercalcemia of malignancy (HHM) is caused by tumor cells that release a circulating factor which stimulates osteoclastic bone resorption. Recently, it has been reported that tumors associated with HHM contain factors that stimulate renal and bone cell adenylate cyclase. The activity was inhibited by parathyroid hormone (PTH) antagonists, and this led to the hypothesis that hypercalcemia is due to bone resorbing factors that engage PTH receptors in bone. Since it is not known whether the bone resorbing factors act via PTH receptors in bone, we examined the effects of PTH antagonists on PTH-stimulated bone resorption and bone resorbing activity that was produced by two tumor models of HHM which also release these adenylate cyclase stimulating factors. The PTH antagonists [8,18norleucine, 34tyrosine]bovine PTH (3-34) amide and [34tyrosine]bovine PTH (7-34) completely inhibited PTH-stimulated bone resorption. Neither antagonist inhibited bone resorption that was stimulated by the conditioned medium from cells that were derived from the Walker rat 256 tumor model of HHM. Both antagonists also failed to inhibit bone resorption that was stimulated by culture media from cells that were derived from the rat Leydig cell tumor. These data suggest that in these two models of HHM, the bone resorbing factors do not exert their effects by interacting with PTH receptors on bone.

Factors associated with humoral hypercalcemia of malignancy stimulate adenylate cyclase in osteoblastic cells

The culture media of three cell lines, a human prostate carcinoma (PC3), a rat Leydig cell tumor (Rice-500), and a rat carcinosarcoma (WRC-256), that were derived from tumors associated with humoral hypercalcemia of malignancy (HHM), were examined for stimulation of adenylate cyclase in ROS 17/2.8 osteoblastic cells and for bone resorptive activity in culture. Cells from a nonhypercalcemic variant of the WRC256 tumor served as control. Extracts from three solid human tumors, a lung adenocarcinoma from a patient with HHM and two adenocarcinoma from normocalcemic patients (lung and colon), were also examined for adenylate cyclase stimulation. We found excellent correlation between stimulation of cyclic AMP accumulation in ROS 17/2.8 cells and bone resorbing activity in culture, or production of HHM in vivo. Stimulation of adenylate cyclase by HHM factors was inhibited by the parathyroid hormone competitive inhibitor, [8norleucyl, 18norleucyl, 34tyrosinyl] bovine parathyroid hormone (3-34) amide.

Tumor-derived growth factor increases bone resorption in a tumor associated with humoral hypercalcemia of malignancy

Evidence is presented that a tumor-derived transforming growth factor is responsible for stimulating bone resorption and causing hypercalcemia in an animal tumor model of the hypercalcemia of malignancy. Both conditioned medium harvested from cultured tumor cells and tumor extracts of the transplantable rat Leydig cell tumor associated with hypercalcemia contained a macromolecular bone resorbing factor with the chemical characteristics of a tumor-derived transforming growth factor.

Absence of parathyroid hormone messenger RNA in nonparathyroid tumors associated with hypercalcemia

We used a sensitive and specific hybridization assay that detects evidence of parathyroid hormone synthesis in tumors to investigate whether this hormone mediates the hypercalcemia of malignant disease. The assay uses radiolabeled, cloned parathyroid hormone DNA to hybridize selectively with parathyroid hormone messenger RNA. We assayed 13 human and 3 animal tumors of diverse cell origins that are frequently associated with the hypercalcemia of cancer. Five of the human tumors were obtained from patients known to be hypercalcemic at the time of tumor excision, two were from normocalcemic patients, and six were from patients with breast cancer whose serum calcium levels were unknown. Messenger RNA was prepared from cultured cell lines or tumors; active RNA fractions were hybridized with either human or bovine cloned parathyroid hormone DNA that had been labeled to a high specific activity with [32P]nucleotide. We were unable to detect parathyroid hormone RNA transcripts in any of the tumors. Our results indicate that parathyroid hormone rarely, if ever, causes hypercalcemia in malignant disease.

Stimulation of production of prostaglandin E in gingival cells exposed to products of human blood mononuclear cells

1. Supernatant media from cultures of unstimulated human peripheral blood mononuclear cells contained one or more factors that increased by several hundred-fold the production of prostaglandin E by fibroblast-like cells derived from both inflamed and normal human gingival tissue. 2. This stimulation occurred in a dose-dependent manner and was completely inhibited by 14 microM-indomethacin. 3. Responsiveness to the factor declined as the age of the cell culture increased. 4. An increase in prostaglandin E production was first observed after a 2h exposure to the mononuclear cell factor(s) and could be prevented by cycloheximide. 5. Brief exposure (0.5 and 1.0 h) to mononuclear cell factor did not increase prostaglandin E production by the cells in a subsequent 72 h incubation in the absence of mononuclear cell factor. 6. Addition of arachidonate (10 microM and 15 microM) further enhanced stimulation of prostaglandin E production in response to mononuclear cell factor. 7. The stimulatory activity was resistant to digestion by trypsin, but was heat-labile, so that only 17% remained after treatment at 56 degrees C for 30 min.