Changes in expression of the lysosomal membrane glycoprotein, LAMP-1 in interdigital regions during embryonic mouse limb development, in vivo and in vitro.

Syndactyly, a failure of the digits to separate into individual units, affects about 8 to 9 per 1000 newborns and results from an aberration of the normal development of the interdigital tissues. Limb digit separation is the result of programmed cell death (apoptosis). Lysosomes play a role in the process of cell self-destruction. Our experiment was designed to test the hypothesis that the intensity of interdigital lysosomes increases during the separation of digits in vivo and in vitro. The primary mouse monoclonal antibody, 1D4B, detects the presence of lysosomes by identifying the LAMP-1 glycoprotein on the lysosome cell membrane. In our experiment this antibody immunodetected interdigital lysosome proteins in serial sections of limbs from Swiss-Webster mouse embryos, gestational ages E12.5 through E15, key developmental stages for digit separation. Digit separation was associated with an increase in intensity of lysosomal protein staining. In E12.5 limbs, the presence of lysosomes was enriched in the distal aspect of the interdigital tissue. However, the number of lysosomes markedly increased in the E13 and E14 limbs, including the entire length and width of the interdigital tissue in the E14 limbs. This lysosomal protein presence in E14 limbs was significant compared to E12.5, E13, and E15 limbs. By day 12.5, the mouse embryo limb is committed to digit separation. Addition of retinoic acid to the culture medium of limbs earlier in development, such as E12, results in induction of the process of digit separation. Cultured E12 limbs that did not receive an addition of retinoic acid, did not show digit separation. We conclude that in the limb development process, the enrichment in interdigit LAMP-1 proteins, may indicate a relationship between lysosomes, apoptosis, and digit separation. We also conclude that retinoic acid has an important role in digit separation in vivo, as shown in limb development, and demonstrated through the addition of retinoic acid to media of cultured tissues.

In vitro model of syndactyly replicates the morphologic features observed in vivo.

Syndactyly is a common congenital hand anomaly that may occur after exposure to teratogens. We have developed an in vitro model of syndactyly to investigate the molecular mechanisms underlying this malformation of digit development. Retinoic acid, which regulates pattern formation in vertebrate limb development and is associated with teratogenic malformations, was used in the development of this syndactyly model system. Pregnant Swiss-Webster mice were given retinoic acid by oral gavage on days 10 and 11 of embryonic development (E10 and E11, respectively). The mice were sacrificed on gestational days 13 and 17 (E13, E17) and immediately postnatally (PN). The fetuses were removed and the forelimbs dissected under the operating microscope. The E13 limbs were cultured for 4 days (E13+4) in an organ culture system using a serumless, chemically defined medium. The E17, PN, and E13+4 forelimbs were critically examined for malformations of digit separation and digit development. Retinoic acid-induced fetal mouse forelimb syndactyly was observed in all the groups; 81 percent of E17 limbs, 75 percent of PN limbs, and 77 percent of E13+4 limbs had syndactyly. The morphology of the digital malformations was similar in the E17, PN, and E13+4 limbs. This in vitro model permits further studies to characterize the molecular changes that occur during the development of a congenital hand anomaly.

Mesenchymal commitment to digital joint formation.

Temporal and spatial commitment of in vivo and in vitro mammalian digital joint development were characterized in a murine model. Alcian blue and alizarin red staining were used to label proteoglycans of cartilage matrix and mineralized matrix in both whole mounts and histological sections. Mesenchymal differentiation toward a joint fate was identified by a lack of matrix deposition in islands of joint precursor cells between phalangeal precursors, and localized lysosomal enzyme activity was later demonstrated in these regions during formation of the joint cavity. Organ-cultured forelimbs and in vivo specimens demonstrated analogous digital joint morphological trends. With a defined developmental window, reverse transcription, polymerase chain reaction, demonstrated differential gene expression of transforming growth factor-beta isotypes, aggrecan core protein, and type II collagen, suggesting a role for transforming growth factor-beta in directing digital joint development.

Human endothelial cells are defective in diabetic vascular disease.

Diabetic vascular disease is characterized pathologically by endothelial cell (EC) hyperplasia and basement membrane (BM) thickening. One key question regarding the pathogenesis of diabetic vascular disease is whether the EC or BM or both are primarily defective and responsible for these pathological changes. Previous studies, which took the approach of creating artificial diabetic conditions, have been inconclusive. It is known, however, that the extracellular matrix may be altered by glycosylation as a result of hyperglycemia, thereby altering EC function. To begin to address this question and more closely mimic the situation in vivo, we characterized human diabetic EC harvested from insulin-dependent diabetic mothers (IDDM) at the cellular and molecular levels. Human EC were isolated from both normal and IDDM umbilical cords and cellular functions evaluated using standard assays of attachment (% attached cells), proliferation (cpm/cell), resistance to detachment under shear stress (number of cells remaining attached), and glucose uptake (cpm/2 X 10(4) cells). Gene expression of major BM components (collagen type IV, laminin beta 1, and laminin beta 2) was quantified by Northern analysis. Diabetic EC demonstrated increased proliferation (two- to eightfold compared to normals), were 20-40% less resistant to shear stress and took up glucose 10-15% more slowly than normal EC. Furthermore, Northern analysis showed that the expression of major BM components was increased by an average of 10-18% in diabetic cells compared to normal cells. These results were consistent with in vivo observations and previously published data.(ABSTRACT TRUNCATED AT 250 WORDS)

Interdigital soft tissue separation induced by retinoic acid in mouse limbs cultured in vitro.

The temporal pattern of separation of the soft tissue between mouse digits was examined in an organ culture model system. Mouse limbs of different gestational age were cultured in vitro and the pattern of separation of the digits characterized. By gestational day 12.5 (E12.5) the limbs were committed to undergo separation of the soft tissue in the interdigital space when cultured in vitro. Prior to E12.5 digital separation did not occur and the limb tissues were not committed to this process. The addition of 10(-7) M retinoic acid (RA) to the media of E12 limbs was capable of inducing digit separation in the uncommitted limbs. Both the soft and hard tissue development of digits formed in vitro for either committed limbs or uncommitted limbs induced with RA was similar to the in vivo pattern.

Endogenous epidermal growth factor regulates limb development.

Mutations associated with genes of the EGF superfamily are implicated in limb malformations. To evaluate the potential role of EGF-mediated signal transduction in the control of early mammalian limb development, we developed a simple in vitro system which is permissive for morphogenesis and cytodifferentiation in serumless, chemically defined medium. Our experimental strategy was to ascertain if the EGF precursor gene was transcribed and translated into potentially bioactive growth factor. EGF mRNA transcripts are expressed in Swiss Webster mouse embryonic (42-44 somite pairs) forelimbs as determined by mRNA phenotyping. EGF transcripts are translated into precursor EGF polypeptides which were localized to limb covering epithelium and the chondrogenic mesenchymal cell lineages. EGF immunostaining patterns suggested a paracrine type of regulation for the cartilage blastema associated with forelimb development. To test whether EGF effects the timing and positional information required for limb-specific cartilage morphogenesis, we employed tyrphostin (RG 50864) which inhibits EGF receptor kinase activity in a concentration-dependent manner and severely retards limb development. These findings support our hypothesis that endogenous EGF or EGF-like proteins provide signaling for the size and shape of discrete forelimb cartilage formations during mouse embryonic morphogenesis.

Cellular responses to silicone and polyurethane prosthetic surfaces.

Prosthetic devices composed of silicone or polyurethane are commonly used in surgery. These devices elicit a soft tissue reaction which may frequently be complicated by capsule formation. Histologically the capsule comprises both cellular (fibroblasts and endothelial cells (EC)) and matrix components (predominantly collagen type I). We hypothesized that the function of the cellular elements is altered by exposure to prosthetic materials and that this alteration contributes to capsule formation. To test this hypothesis, we utilized specific in vitro assays of cell function (attachment, proliferation, matrix gel contraction), which closely mimic in vivo cellular events, in order to define the responses of EC and fibroblasts to prosthetic surfaces (foam polyurethane, flat silicone, and textured silicone). Morphologic changes were evaluated by scanning electron microscopy (SEM). Attachment of both cell types to all prosthetic surfaces was significantly decreased compared to control (HUVEC: control, 55 +/- 1; foam polyurethane, 19 +/- 4*; flat silicone, 25 +/- 3*; textured silicone, 36 +/- 1*; fibroblast: control, 93 +/- 6; foam polyurethane, 21 +/- 4*; flat silicone, 57 +/- 5*; textured silicone, 44 +/- 5* (*P < 0.05 = significant; units, percentage spread)). Fibroblast proliferation was significantly decreased on foam polyurethane (0.1 +/- 0.03*) and textured silicone (0.18 +/- 0.05*), but not on flat silicone (0.79 +/- 0.2; control = 0.96 +/- .2). In contrast, HUVEC proliferation was significantly decreased on both silicone surfaces but not on polyurethane (units, cpm/cell; control, 0.26 +/- 0.05; foam polyurethane, 0.15 +/- 0.05; flat silicone, 0.08 +/- 0.03*; textured silicone, 0.02 +/- 0.01*).(ABSTRACT TRUNCATED AT 250 WORDS)

Medial edge epithelium fate traced by cell lineage analysis during epithelial-mesenchymal transformation in vivo.

Vital cell labeling techniques were used to trace the fate of the medial edge epithelial (MEE) cells during palatal fusion in vivo. Mouse palatal tissues were labeled in utero with DiI. The fetuses continued to develop in utero and tissues of the secondary palate were examined at several later stages of palatal ontogeny. The presence and distribution of DiI was correlated with the presence of cell phenotype-specific markers. During the initial stages of palatal fusion the DiI-labeled MEE were present in the midline position. These cells were attached to an intact laminin-containing basement membrane and contained keratin intermediate filaments. At later stages of palatogenesis the DiI-labeled MEE were not separated from the mesenchyme by an intact basement membrane and did not contain keratin. In late fetal development, DiI-labeled cells without an epithelial morphology were present in the mesenchyme. The transition of the DiI-labeled cells from an epithelial phenotype to a mesenchymal phenotype is consistent with a fate of epithelial-mesenchymal transformation rather than programmed cell death.

New evidence and new hope concerning endothelial seeding of vascular grafts.

Endothelial cell (EC) seeding of prosthetic bypass grafts has been promoted as a method of improving graft patency. However, an efficient and reliable method of seeding vascular prostheses with ECs is lacking due to inefficient harvesting of ECs and poor attachment and proliferation of cells on the prosthetic surfaces. To investigate the effect of a commonly used prosthetic surface on EC attachment and proliferation, we measured the attachment and proliferation of ECs on polytetrafluoroethylene (PTFE) grafts uncoated or coated with gelatin, laminin, fibronectin, collagen type I and/or III, or RGD (arginine-glycine-aspartate)-containing peptide. EC attachment and proliferation were both significantly decreased on the untreated PTFE graft surface. Conversely, coating of PTFE with fibronectin, RGD, laminin, or gelatin significantly (p less than 0.05) improved the attachment of ECs, with the most striking increases occurring with laminin and gelatin. Similarly, all matrix components in this study improved EC proliferation compared with untreated PTFE, with RGD and gelatin producing the most significant improvement. PTFE adversely effects EC attachment and proliferation. These properties can be improved by treating PTFE graft surfaces with extracellular matrix components in relatively low concentrations. Future investigations are needed to determine whether there are combinations and concentrations of matrix components that will optimize these cellular functions on vascular prostheses.

Bilateral solitary glomus tumors of the hands.

Solitary glomus tumors of the digits are uncommon, comprising only about 2% of all hand tumors. In this report, we present a case report of a patient with bilateral glomus tumors that became symptomatic 4 years apart. No inheritance pattern was apparent for this patient. An extensive literature review did not uncover a similar patient. Because this condition does not appear to be widely recognized, we suggest that it be considered when patients present with bilateral digital pain.

The local effects of cachectin/tumor necrosis factor on wound healing.

Previous experimental studies have suggested that tumor necrosis factor (TNF) may have either a beneficial or a detrimental role in wound healing. Control and doxorubicin-treated (6 mg/kg, intravenously) rats underwent paired dorsal 5-cm linear wounds and had either vehicle or recombinant (r)TNF (0.5, 5, or 50 micrograms) applied locally to the wound. Paired wounds were harvested at 7 and 14 days after wounding and analyzed for wound-bursting strength (WBS) and activity of the gene for type 1 collagen and TNF. Doxorubicin treatment decreased WBS at 14 days but not at 7 days after wounding. Local application of 50 micrograms of rTNF decreased WBS in saline-treated rats and concentrations of 5 and 50 micrograms decreased WBS in doxorubicin-treated rats when measured 7 days after wounding. These effects dissipated when WBS was measured 14 days after wounding. Doxorubicin decreased wound collagen gene expression and local TNF treatment decreased wound collagen gene expression in saline-treated rats and further decreased it in doxorubicin-treated rats. The decrement in collagen gene expression induced by rTNF increased as the local dose of rTNF increased. The gene for TNF was not detectable in wounds from normal or doxorubicin-treated rats at 3, 7, 10, or 14 days after wounding. These data suggest that the gene for TNF is not expressed in wounds and that the local application of TNF is detrimental to wound healing as it decreases WBS and activity of the gene for collagen.

Increased calcium levels alter cellular and molecular events in wound healing.

Surgical morbidity is dictated directly by wound healing. We have studied the effects of elevated calcium levels using cultured keratinocytes in vitro on two of the rate-limiting steps of wound healing, chemotaxis (directed migration) and adhesion. We found that the increased calcium (10 mmol/L) significantly inhibited both keratinocyte chemotaxis and adhesion (p less than 0.05). The calcium effect on adhesion could be partially reversed by pretreatment with the calcium channel blocker verapamil. Based on these data, an animal model was formulated in which topical calcium (5 mmol/L/day) was added to linear incision wounds. This resulted in significantly (p less than 0.05) delayed wound contraction characteristic of a chronic or impaired wound. Wound contraction depends on the presence of fibroblasts that synthesize collagen. The chronic wound was characterized by increased collagenase activity (p less than 0.05) but little alteration in collagen I synthesis. The addition of verapamil to these chronic wounds resulted in improved wound closure. These studies define the molecular and cellular events occurring as a result of the addition of elevated levels of calcium both in vitro and in vivo. Calcium may play a key role in the pathogenesis of chronic wounds.

Cholinergic stimulation of pancreatic polypeptide release by a meal, bombesin, neurotensin, tetragastrin, cholecystokinin, and cerulein.

It has been demonstrated that pancreatic polypeptide (PP) release can be markedly impaired by vagotomy or anticholinergic drugs. The current studies examine the role of cholinomimetic stimulation on PP release in dogs. Eight conscious animals underwent a series of tests: (1) a test meal (10 g/kg Alpo); (2) tetragastrin infusion (4 micrograms/kg/hr); (3) bombesin infusion (1.0 microgram/kg/hr); (4) cerulein infusion (100 ng/kg/hr); (5) cholecystokinin octapeptide (CCK-OP) infusion (100 ng/kg/hr); (6) neurotensin infusion (3 ng/kg/hr). All the studies were repeated individually with intravenous bethanecol (100 micrograms/kg/hr) as the background stimulant. The mean increment of PP released by a meal (160 +/- 32 fmol/ml) was significantly increased by bethanecol infusion (316 +/- 49 fmol/ml) (P less than 0.05). Each individual peptide released a significant amount of PP; tetragastrin: 53 +/- 11; neurotensin: 58 +/- 14; CCK-OP: 42 +/- 9; cerulein: 42 +/- 12; bombesin: 118 +/- 24 (P less than 0.05). Bethanecol did not significantly augment PP release by any of the individual peptides (P greater than 0.05). This study indicates that PP release by a meal is sensitive to cholinomimetic stimulation and that the peptide involved is neither gastrin, neurotensin, CCK, bombesin, nor cerulein. These data support the possibility of the existence of a cholinergic stimulatable mechanism, possibly a peptide responsible for the release of PP.

Bombesin and insulin-stimulated pancreatic polypeptide release as a discriminator of vagal integrity.

Intact vagi after ulcer operations are often implicated in the cause of recurrent ulcer. The stimulation of gastric acid by insulin hypoglycemia is dangerous and the measurement of acid secretion after gastrectomy unreliable. This study was undertaken to assess and compare PP release by bombesin or insulin as an indicator of vagal integrity. Eight dogs with a chronic gastric fistula were tested with bombesin (100 nanograms per kilogram) and insulin (0.1 unit per kilogram) intravenous bolus after unilateral and, then, bilateral truncal vagotomy. Each study was 120 minutes, and blood was taken at one, three, five, seven and then ten minute intervals. Gastric acid was measured by autobiuret titration. Plasma was stored at minus 20 degrees C. until assayed for PP by radioimmunoassay. Bombesin-stimulated gastric acid secretion was not significantly altered by vagotomy (p greater than 0.05), whereas that stimulated by insulin was significantly inhibited by bilateral truncal vagotomy (p less than 0.05). Bilateral and right hemivagotomy significantly inhibited PP release by bombesin (p less than 0.05), however, only bilateral truncal vagotomy significantly inhibited PP release by insulin (p less than 0.05). These results suggest that the measurement of PP release by insulin or bombesin is a sensitive index of vagal integrity and that bombesin-released PP may specifically delineate the integrity of the right vagus. Since the measurement of gastric acid secretion after operation is both uncomfortable and often difficult to interpret, the value of a simple blood test to determine vagal integrity may be of considerable clinical relevance.

Opiate modulation of pancreatic polypeptide release by a meal in the dog.

It has been reported that morphine abolished the plasma pancreatic polypeptide (PP) response to a meal in man, but the mechanism of this action is unclear. This study was designed to investigate the effect of low doses of the endogenous opiate peptide. Met-enkephalin and naloxone on basal- and meal-stimulated PP release in order to examine the role of opioid modulation in the release of this hormone. Four gastric fistula dogs underwent a series of six studies, a test meal alone. Met-enkephalin infusion (40 microgram/kg/hr), naloxone infusion, meal plus naloxone infusion and meal plus Met-enkephalin plus naloxone. Gastrin and PP were measured by radioimmunoassay. Basal PP levels averaged 35.1 +/- 3.0 fmol/ml. Although Met-enkephalin had no effect on basal PP levels, it significantly (P less than 0.05) inhibited the mean peak increment of PP stimulated by a meal (control, 331 +/- 39 fmol/ml; Met-enkephalin, 145 +/- 49 fmol/ml; P less than 0.05). This inhibition was completely abolished by naloxone. Naloxone alone did not alter basal- or meal-stimulated plasma PP levels. Neither Met-enkephalin nor naloxone altered basal or stimulated plasma gastrin levels. This study demonstrated that opiate peptides play a role in the regulation of the release of PP by a meal; it thus suggests the possibility of an opioid modulatory mechanism for the release of this hormone.

The early diagnosis of gastrinoma.

Despite the increasing awareness of gastrinoma and its lethal peptic ulcer sequelae, the diagnosis is often initially missed or made as a terminal event. The authors screened all patients with peptic ulcer symptoms serious enough to warrant hospital admission or those associated with diarrhea, nephrolithiasis, hypercalcemia, or pituitary abnormality. In a one-year period (1979-1980) nine (of 14 suspected) new gastrinoma patients were identified using a sensitive and specific gastrin radioimmunoassay in combination with provocative tests including IV secretin, calcium, and food. Conventional upper GI series, CAT scan, arteriography, and endoscopy provided no additional information other than to confirm the presence of ulcer disease. Basal plasma gastrin levels were more than 200 pmol L-1 in only three of the nine (normal fasting plasma gastrin levels are less than 25 pmol L-1). Three patients presented with acute ulcer perforation, and the diagnosis of gastrinoma was suspected because of multiple ulcers and pancreatic masses. In three other patients, previous duodenal ulcer surgery had failed. One patient with dyspepsia, high basal plasma gastrin, negative secretin and calcium infusion studies, and a positive meal test was diagnosed as having G-cell hyperplasia; this was confirmed by biopsy and antral gastrin extraction. Antrectomy alone resulted in cure. In all patients tested, a positive calcium infusion or secretin bolus (greater than 100% rise over basal) strongly suggested the diagnosis of gastrinoma, which was confirmed at surgery. In the acute perforations, initial management with omental patch and cimetidine therapy allowed survival of two patients, while emergency total gastrectomy in the third resulted in death due to esophagojejunal leak. Elective patients were treated with cimetidine initially for at least two weeks before total gastrectomy. In this group there were no operative mortalities, and postoperative morbidity was minimal. This series illustrates three important points: (1) careful screening of an ulcer population using gastrin radioimmunoassay and provocative tests has enabled a high yield of gastrinomas while conventional investigations are of minimal values; (2) a high index of suspicion in appropriate cases is necessary; and (3) total gastrectomy performed under elective circumstances is safe and allows the patients to resume a normal and healthy life without the sequelae of aggressive peptic ulceration or daily drug administration.

Effect of methionine-enkephalin and naloxone on bombesin-stimulated gastric acid secretion, gastrin, and pancreatic polypeptide release in the dog.

In four dogs with chronic gastric fistulae, bombesin infusion was used to stimulate the release of gastrin and pancreatic polypeptide (PP) as well as rates of gastric acid secretion. Neither methionine-enkephalin (met-enkephalin) nor naloxone alone or the combination of these agents altered bombesin-stimulated gastrin release. met-enkephalin alone (but not naloxone) significantly inhibited the gastric secretory response to bombesin, but this inhibitory effect was not influenced by the simultaneous infusion of naloxone; the data suggested that the effect of met-enkephalin was indirect, and perhaps modulated by another inhibitory mechanism. Whereas PP release induced by bombesin was not affected by naloxone, it was significantly suppressed by met-enkaphalin; since this inhibition was virtually totally reversed by naloxone, the data suggested that the effect of opiate peptides on the release of pancreatic polypeptide was direct and mediated by a specific opiate receptor.

Evidence for an intestinal mechanism of pancreatic polypeptide release.

The purpose of this study was to define the existence of an intestinal phase of pancreatic polypeptide (PP) release and to assess whether it was mediated by a cholinergic-sensitive mechanism. Four conscious dogs with 20-cm upper intestinal Thiry-Vella loops and chronic gastric fistulas were used. The Thiry-Vella (T-V) loops were perfused with 10% liver extract or 0.154 M NaCl at a rate of 1 ml/min for 120 min. In a separate experiment, 240 ml of 10% liver extract was infused over a 5-min period into the stomach via the gastric fistula. Basal PP levels were 29 +/- 4 fmol/ml. The gastric infusion of liver extract caused a significant increase of plasma PP levels to a peak of 215 +/- 29 fmol/ml (P less than 0.05). The perfusion of the T-V loop with liver extract significantly increased plasma PP levels over basal to a peak of 73 +/- 14 fmol/ml (P less than 0.05). This value was significantly less than that released by gastric infusion of liver extract (P less than 0.05). Perfusion of the loop with NaCl did not significantly alter basal plasma PP levels (P greater than 0.05). PP release by perfusion of the T-V loop with liver extract was abolished by atropine intravenous bolus (0.2 mg/kg). Although the combination of bethanechol (100 microgram/kg/hr intravenous) and liver extract consistently increased the plasma levels of PP, the values did not attain statistical significance when compared to liver extract alone (P greater than 0.05). The data presented are thus consistent with the hypothesis that there is an enteric phase of pancreatic polypeptide release and that this enteropancreatic reflex is mediated by a cholinergic-sensitive mechanism which might be hormonal or neural.