Control algorithms for artificial beta cell.

The evolution of control algorithms for closed-loop regulation of blood glucose levels is described. Because of the rapid response time of the BIOSTATOR Glucose Analyzer, a derivative algorithm can be applied to replace the previous generation of "predictor" algorithms for the calculation of dynamic insulin infusion rates.     

Insulin responses to mixed meals: comparison of an artificial beta cell and normal beta cells.

The peripheral glucose and free insulin levels seen following a mixed meal in six insulin-dependent diabetic patients whose insulin was administered by a glucose-controlled insulin infusion system (GCIIS) were compared to those of normal subjects who received the same mixed meal or who were given separately carbohydrate, protein, or fat in amounts equivalent to those contained in the mixed meal. Patients treated with the GCIIS achieved nearly normal glucose levels immediately after the mixed meal, but this was accompanied by marked hyperinsulinemia. In the period from 120 to 240 minutes after the start of the mixed meal, the GCIIS duplicates the insulin levels produced by the normal pancreas after a glucose meal and, with appropriate algorithm constants, closely approximates those seen after a mixed meal.     

Influence of insulin infusion kinetics of an artificial beta cell on blood glucose control in insulin-dependent diabetics.

The influence of different control modes for insulin infusion with an artificial beta cell was examined in 41 insulin-dependent diabetics. In 21 Patients, oral glucose tolerance tests were performed with control modes characterized either by low dynamic and high static gain (type I, 10 patients) or high dynamic and low static gain (type III, 11 patients). The change from type I to type III control mode effected an increase of initial insulin infusion rates (91 +/- 59 to 313 +/- 81 mU/min 10-20 min after glucose ingestion) and a decrease of infusion rates during the following phase of the 3-hour observation period (28.2 +/- 4.2 to 18.1 +/- 2.8 U) in patients whose blood glucose curves were completely normalized. Suppression of plasma glucagon levels, observed in 5 healthy control subjects, was not fully restored to normal in these patients. In another 20 insulin-dependent diabetics, daily insulin requirements form the artificial beta cell were determined by employing two control modes (types II and III) comparable in static control but different in dynamic control. Gain of dynamic control, especially in the range of falling glucose levels, was higher in type III control mode (15 patients) than in type II mode (5 patients). These insulin requirements were compared to the insulin doses necessary for subcutaneous treatment. While intravenous insulin requirements were much higher when type II control mode was employed (78.2 +/- 10.2%), during application of type III mode, intravenous insulin requirements were only 10.8 +/- 5.5% higher than subcutaneous doses. We conclude from these data that early increases in insulin infusion rates followed by a rapid decrease seem to reduce insulin requirements after glucose ingestion. A high-gain dynamic control is the basis for this insulin infusion profile.     

Use of an artificial endocrine pancreas in diabetic pregnancy.

Insulin demand of 12 pregnant diabetics has been investigated with an artificial endocrine pancreas. A rise in insulin requirement during pregnancy which can be attributed to the effort of reaching normoglycemia and to the effect of contrainsular hormones has also been observed by this objective method. Assessment of basal insulin demand during the night might be helpful in optimizing conventional therapy by using long-acting insulins for supplementing basal insulin need. According to our results, pregnancy complicated by diabetes can be foreseen as one of the main applications of a glucose-controlled insulin infusion system.     

The use of artificial beta cell in diagnosis and treatment of insulinoma.

The glucose-controlled insulin infusion system (GCIIS), the socalled artificial beta-cell, is an important and useful device for detecting and treating hypoglycemic reactions. The dangers of several diagnostic tests such as the tolbutamide or the insulin response test may successfully be avoided. Patients suffering from severe hypoglycemia are kept in normoglycemia by the feedback-controlled dextrose infusion before and during operation.     

Towards an artificial cell

We are on the verge of producing "synthetic cells," or protocells, in which some, many or all of the tasks of a real biological cell are harnessed into a synthetic platform. Such advances are made possible through genetic engineering, microfabrication technologies, and the development of cellular membranes from new surfactants that extend beyond phospholipids in stability and chemical control, and can be used to introduce designer functionality into membranes and cells. We review some of the recent advances in the development of synthetic cells and suggest future exciting directions.     

The artificial beta cell (Biostator) in the adjustment of instable diabetics--results after 20 months.

In 55 poorly controlled insulin-dependent diabetics, we tried to discover criteria for an improvement of metabolism by means of the "artificial beta-cell" (Biostator). To this end, during the first 24 h of hospitalization, blood glucose was monitored continuously under conventional insulin therapy (monitoring period). Insulin requirement was determined during the next 24 h by the artificial beta-cell (feedback period). Corrections of diabetes regimen were made with reference to the insulin consumption during the feedback period and to the extent of the postprandial blood sugar increases and decreases during the monitoring period. The resulting new diabetes regimen led to a significant improvement of the daily blood sugar profiles.     

Use of blood in elective general surgery: an area of wasted resources.

A prospective audit of the blood bank was carried out in Portsmouth to examine how efficiently blood was used in elective general surgery. The routine crossmatching of blood was found to be unnecessary for certain procedures such as cholecystectomy, thyroidectomy, mastectomy, and vagotomy. The most efficient use of blood was seen in vascular surgery despite the greater risk of severe haemorrhage. Completed detailed questionnaires returned by members of surgical teams in Wessex supported the Portsmouth data as being fairly typical and confirmed that substantial and unintentional overordering occurred, apparently due to poor communication, lack of discussion, and "force of habit." The place of a "half hour crossmatch" as a substitute for many routine orders has been explored and seems to be acceptable to most anaesthetists and surgeons. This study, by showing how inefficiently blood is used, has underlined the value of local audit. It also supports the general experience of centres throughout the world who have invoked similar methods and achieved considerable savings without harm to patients.     

Energetics of assembling an artificial heterodimer with an alpha/beta motif: cleaved versus uncleaved Escherichia coli thioredoxin.

We have studied the folding/binding process between the N- and C-fragments (1-73, 74-108) of oxidized Escherichia coli thioredoxin (Trx) to compare the energetics between the cleaved and uncleaved Trx. Sedimentation equilibrium analysis in 0.1 M potassium phosphate, pH 5.7, shows (i) the strong and weak self-association of the N- and C-fragments, respectively, (ii) a heterodimer with a small dissociation constant (K(d)) ca. 100 nM, and (iii) monomeric Trx. To avoid self-association, measurements were carried out in 10 mM potassium phosphate, pH 5.7. Far-UV CD spectra of the fragments at variable temperature show an isodichroic point at 208 nm and a non-cooperative cold induced disordering transition without concentration dependence. Deconvolution of these spectra indicates the presence of residual structure. Titration of the N-fragment with an excess of C-fragment indicates a 1:1 stoichiometric complex with an apparent K(d) ca. 49 nM. Analysis of this complex by CD and hydrogen exchange/2D-NMR (Tasayco and Chao (1995) Proteins: Struct., Funct., Genet. 22, 41-44) spectroscopy indicates the reassembly of the alpha/beta motif of Trx. GnHCl induced unfolding measurements give DeltaG(0) values of 9.5 +/- 0.2 and 10.0 +/- 0.4 kcal/mol at 20 degrees C for the uncleaved and cleaved Trx, respectively. The far-UV CD melting curve of uncleaved Trx indicates an intriguing non-cooperative upward baseline trend. CCA analysis of these spectra indicates the presence of a native-like folded intermediate. A three-state thermodynamic analysis of the thermal transition curves gives a total DeltaH(0) of unfolding of 121 +/- 4 kcal/mol at the T(m) (88 degrees C), while the two-state analysis for cleaved Trx gives 122 +/- 6 kcal/mol at 88 degrees C. Analysis of the chemical and thermal unfolding of both proteins indicates a value of ca. 1 M for the apparent effective concentration (C(eff)) of cleaved Trx.     

Therapeutical application of artificial beta-cell in surgery and obstetrics.

An artificial beta-cell (Biostator) was used to control blood glucose concentration in six insulin-dependent diabetics who underwent surgical procedures and in five diabetic women during both vaginal childbirth and Caesarian section. In all cases, an optimal glycometabolic control was achieved before, during, and after surgery and delivery.     

Transport and metabolism of glucose in an insulin-secreting cell line, beta TC-1.

Kinetic characteristics of glucose transport and glucose phosphorylation were studied in the islet cell line beta TC-1 to explore the roles of these processes in determining the dependence of glucose metabolism and insulin secretion on external glucose. The predominant glucose transporter present was the rat brain/erythrocyte type (Glut1), as determined by RNA and immunoblot analysis. The liver/islet glucose transporter (Glut2) RNA was not detected. The functional parameters of zero-trans glucose entry were Km = 9.5 +/- 2 mM and Vmax = 15.2 +/- 2 nmol min-1 (microL of cell water)-1. Phosphorylation kinetics of two hexokinase activities were characterized in situ. A low-Km (0.036 mM) hexokinase with a Vmax of 0.40 nmol min-1 (microL of cell water)-1 was present along with a high-Km (10 mM) hexokinase, which appeared to conform to a cooperative model with a Hill coefficient of about 1.4 and a Vmax of 0.3 nmol min-1 (microL of cell water)-1. Intracellular glucose at steady state was about 80% of the extracellular glucose from 3 to 15 mM, and transport did not limit metabolism in this range. In this static (nonperifusion) system, 2-3 times more immunoreactive insulin was secreted into the medium at 15 mM glucose than at 3 mM. The dependence of insulin secretion on external glucose roughly paralleled the dependence of glucose metabolism on external glucose. Simulations with a model demonstrated the degree to which changes in transport activity would affect intracellular glucose levels and the rate of the high-Km hexokinase (with the potential to affect insulin release).     

Mapping of an inducible element in the T cell receptor V beta 2 promoter.

The murine V beta 2 promoter was analyzed for an element regulating phorbol ester inducibility of the TCR beta chain gene. In transient expression analysis of 5' nested deleted fragments of the V beta 2 promoter, the TPA-inducible element mapped between -85 and -42. The -85 to -62 oligo conferred 12-0-tetradecanoylphorbol-13-acetate (TPA) inducibility to the heterologous TPA-uninducible thymidine kinase promoter. The -85 to -62 region contained an AP-1 site (-85 to -72) and inverted repeat motif (-72 to -62). The AP-1 site required the 3' flanking inverted repeat region for conferring optimal inducibility. In vitro transcribed and translated jun/fos heterodimers bind to the V beta 2 AP-1 motif with a 16-fold lower affinity as compared to the collagenase AP-1 motif. This explains the inability of the V beta 2 AP-1 motif to confer optimal TPA inducibility by itself. The affinity of jun/fos heterodimers for the V beta 2 AP-1 motif was not increased by the presence in cis of the inverted repeat motif. The 3' flanking inverted repeat binds the ets transactivator but not jun/fos heterodimers. The demonstrated cooperativity between the AP-1 and the 3' flanking sequence to confer TPA inducibility can thus be explained by the individual contributions of jun/fos and ets transactivators.     

Toward an artificial cell based on gene expression in vesicles

We present a new experimental approach to build an artificial cell using the translation machinery of a cell-free expression system as the hardware and a DNA synthetic genome as the software. This approach, inspired by the self-replicating automata of von Neumann, uses cytoplasmic extracts, encapsulated in phospholipid vesicles, to assemble custom-made genetic circuits to develop the functions of a minimal cell. Although this approach can find applications, especially in biotechnology, the primary goal is to understand how a DNA algorithm can be designed to build an operating system that has some of the properties of life. We provide insights on this cell-free approach as well as new results to transform step by step a long-lived vesicle bioreactor into an artificial cell. We show how the green fluorescent protein can be anchored to the membrane and we give indications of a possible insertion mechanism of integral membrane proteins. With vesicles composed of different phospholipids, the fusion protein alpha-hemolysin-eGFP can be expressed to reveal patterns on the membrane. The specific degradation complex ClpXP from E. coli is introduced to create a sink for the synthesized proteins. Perspectives and subsequent limitations of this approach are discussed.     

Desensitization of the insulin-secreting beta cell.

In human diabetes, inherent impaired insulin secretion can be exacerbated by desensitization of the beta cell by chronic hyperglycemia. Interest in this phenomenon has generated extensive studies in genetic or experimentally induced diabetes in animals and in fully in vitro systems, with often conflicting results. In general, although chronic glucose causes decreased beta-cell response to this carbohydrate, basal response and response to alternate stimulating agents are enhanced. Glucose-stimulated insulin synthesis can be increased or decreased depending on the system studied. Using a two-compartment beta-cell model of phasic insulin secretion, a unifying hypothesis is described which can explain some of the apparent conflicting data. This hypothesis suggests that glucose-desensitization is caused by an impairment in stimulation of a hypothetical potentiator singularly responsible for: 1) some of the characteristic phases of insulin secretion; 2) basal release; 3) potentiation of non-glucose stimulators; and 4) apparent "recovery" from desensitization. Review of some of the pathways that regulate insulin secretion suggest that phosphoinositol metabolism and protein kinase-C production are regulated similarly to the theoretical potentiator and their impairment is a major contributor to glucose desensitization in the beta cell.     

Use of evolved artificial regulatory networks to simulate 3D cell differentiation

Cell differentiation has a crucial role in both artificial and natural developments. This paper presents results from simulations in which a genetic algorithm (GA) was used to evolve artificial regulatory networks (ARNs) to produce predefined 3D cellular structures through the selective activation and inhibition of genes. The ARNs used in this work are extensions of a model previously used to create 2D geometrical patterns. The GA worked by evolving the gene regulatory networks that were used to control cell reproduction, which took place in a testbed based on cellular automata (CA). After the final chromosomes were produced, a single cell in the middle of the CA lattice was allowed to replicate controlled by the ARN found by the GA, until the desired cellular structures were formed. Two simple cubic layered structures were first developed to test multiple gene synchronization. The model was then applied to the problem of generating a 3D French flag pattern using morphogenetic gradients to provide cells with positional information that constrained cellular replication.