Results from the inaugural year of the Collaborative Islet Transplant Registry

The National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) established the Collaborative Islet Transplant Registry (CITR) in September 2001. In its inaugural year, CITR represents the efforts of 12 collaborating North American islet transplant centers reporting 86 islet transplant recipients' data (1999-2003) and 173 processed pancreata leading to 158 infusion procedures. Recipient median age was 42.2 years, duration of diabetes was 30 years, and over 66% of the recipients were female. Twenty-eight patients received a total of 1 islet infusion, 44 patients received 2 islet infusions, and 14 patients received 3 islet infusions. The median age of the deceased donor was 44 years and body mass index was 28.2. Median time from cross clamp to pancreas recovery was 27 minutes, while duration of cold ischemia was 7 hours. Over 77% of the processing facilities used a density gradient for islet purification. For recipients of only 1 infusion, approximately 8665 total islet equivalents/kg were infused; recipients of 2 infusions received 14,102 islet equivalents/kg, and recipients of 3 infusions received 22,922 islet equivalents/kg. At 6 months after the last infusion, 61.1% of the recipients were reported to be insulin independent; at 12 months, 57.9% were reported to be insulin independent. There have been no deaths reported to CITR; 45 serious adverse events were reported. Through its collaboration with the islet transplant community and its interaction with professional societies and federal agencies, CITR is positioned to provide current and comprehensive information on clinically significant outcome measures in islet transplantation.

Intravenous administration of replication-incompetent adenovirus to rhesus monkeys induces thrombocytopenia by increasing in vivo platelet clearance

A replication-incompetent adenovirus vector was administered to rhesus macaques at 1, 3 and 6 x 1012 particles/kg doses to investigate its toxicity. Platelet count decrements of 28%, 82% and 90%, respectively, were observed, with corresponding platelet half-lives of 69.0, 25.2 and 22.2 h (compared with 111 h in untreated animals). The platelet decline was equivalent for all three doses for 8 h, and platelet count recovery began as early as 8 h after infusion for low-dose recipients, or as late as 24 h for the medium and high dose recipients. These observations suggest that thrombocytopenia is a saturable, reversible consumptive process.

Collaborative iIslet Transplant Registry

The Collaborative Islet Transplant Registry (CITR) was established in 2001 with support from the National Institute of Diabetes and Digestive Diseases to collect and analyze information on islet transplants in North America. Thirteen of 27 invited islet transplant centers from the United States and Canada were already participants as of January 2004 with 5 additional centers awaiting IRB approval. In October 2003, CITR had preliminary data on 58 recipients of 107 islet transplants performed at the 13 member centers. These islet recipients averaged 41 years of age (SD 8.9), had diabetes for almost 28 years (SD 10.2) and were predominantly (67%) female. Islets were procured from 118 deceased donors whose average age was 42 (SD 12.2), weight was 90 kg (SD 23.5), and BMI was 30.5 (SD 7.9). The time from cross-clamp to pancreas recovery averaged 56 minutes (SD 134.6). An average of 7,127 islets (SD 2,973) were infused per kg of the recipient's weight. Outcome data are still pending as the Registry prepares its first annual report. Here we describe the organization, goals and design of the data collection instruments for CITR.

Phosphorylation is a regulatory mechanism in apolipoprotein B mRNA editing

The editing of apolipoprotein B (apoB) mRNA is under tissue-specific, developmental and metabolic regulation. We found that multiple protein kinase inhibitors or activators increased apoB mRNA editing up to 2.5-fold in Caco-2 cells and 3-8-fold in McA7777 and FAO rat cells respectively. The phosphorylation-agent-induced modulation is independent of the apolipoprotein B editing catalytic subunit 1 (APOBEC-1) and of apoB mRNA expression levels, indicating the involvement of a protein modification, such as phosphorylation, regulating the cellular editing of apoB mRNA. Transient expression of protein kinase C-θ more than doubled apoB mRNA editing in FAO cells. Chronic exposure to ethanol, a treatment known to increase the expression of protein kinases and to change protein phosphorylation status, increased apoB mRNA editing in FAO cells up to 2.5-fold without increasing the mRNA abundance of APOBEC-1. The elimination of potential phosphorylation sites 47 and 72 of human APOBEC-1 decreased its activity to approx. one-eighth of control levels by a Ser(47)-->Ala mutation, but more than doubled the activity by a Ser(72)-->Ala mutation. The activity modulation was reversed by a Ser-->Asp mutation at sites 47 and 72, which introduced a phosphorylation-like carbonic acid group. Both human APOBEC-1 dephosphorylated by alkaline phosphase and the Ser(47,72)-to-alanine double mutant protein demonstrated a shifted isoelectric focusing pattern compared with the wild type, indicating phosphorylation at these sites. Taken together, these results suggest that phosphorylation might be an important mechanism in the regulation of apoB mRNA editing.

Characterization of a 95 kDa high affinity human high density lipoprotein-binding protein

A new human 95 kDa high density lipoprotein (HDL)-binding protein (HBP) corresponding to a high affinity HDL-binding site with K(d) = 1.67 microg/mL and a capacity of 13.4 ng/mg was identified in human fetal hepatocytes. The HDL binding with the 95 kDa HBP plateaus at 2.5-5 microg/mL under reducing and nonreducing conditions. The association of HDL(3) with the 95 kDa HBP plateaued in 15-30 min while dissociation was complete in 30 min. HDL(3), apoA-I, and apoA-II were recognized by the 95 kDa HBP while low density lipoproteins (LDL) and tetranitromethane-modified HDL were not. The 95 kDa HBP predominantly resides on the surface of cells since trypsin treatment of HepG2 cells eliminated nearly 70% of HDL binding. All studied human cells and cell lines (HepG2, Caco-2, HeLa, fibroblasts, SKOV-3, PA-I) demonstrated the presence of the 95 kDa protein. Both RT-PCR and Western blotting for HB-2/ALCAM were negative in human fetal hepatocytes while Gp96/GRP94 was clearly differentiated from the 95 kDa HBP by two-dimensional electrophoretic mobility. Moreover, deglycosylation of HepG2 membrane preparations did not affect either HDL binding to the 95 kDa HBP or its size, while in contrast it affected the molecular weights of HB-2/ALCAM and SR-BI/CLA-1. We conclude that the 95 kDa HBP is a new HDL receptor candidate widely expressed in human cells and cell lines.

Apolipoprotein specificity for lipid efflux by the human ABCAI transporter

ABCAI, a member of the ATP binding cassette family, mediates the efflux of excess cellular lipid to HDL and is defective in Tangier disease. The apolipoprotein acceptor specificity for lipid efflux by ABCAI was examined in stably transfected Hela cells, expressing a human ABCAI-GFP fusion protein. ApoA-I and all of the other exchangeable apolipoproteins tested (apoA-II, apoA-IV, apoC-I, apoC-II, apoC-III, apoE) showed greater than a threefold increase in cholesterol and phospholipid efflux from ABCAI-GFP transfected cells compared to control cells. Expression of ABCAI in Hela cells also resulted in a marked increase in specific binding of both apoA-I (Kd = 0.60 microg/mL) and apoA-II (Kd = 0.58 microg/mL) to a common binding site. In summary, ABCAI-mediated cellular binding of apolipoproteins and lipid efflux is not specific for only apoA-I but can also occur with other apolipoproteins that contain multiple amphipathic helical domains.

Calcium increases apolipoprotein B mRNA editing

ApoB-100 and apoB-48 are major components of chylomicrons, very low-density lipoprotein (VLDL) and low-density lipoprotein (LDL). The two proteins are generated from a single apoB mRNA by apoB mRNA editing which induces an in-frame stop codon in apoB mRNA. Apolipoprotein B (apoB) mRNA editing is an important determinant of the proportion of full-length (apoB-100) and truncated (apoB-48) proteins in total apoB metabolism. Calcium is involved in the regulation of secretion and synthesis of VLDL and apoB. In this paper, we demonstrate for the first time that the amount of edited apoB mRNA in the cultured cells Caco-2 and McA7777 is markedly increased by calcium. Increasing extracellular calcium concentration, calcium ionophore (A23187 and ionomycin) treatment, and depleting calcium stores and raising cytoplasmic calcium concentration by thapsigargin increase apoB mRNA editing up to threefold in a dose dependent manner. Calcium has no direct stimulative effect on apoB mRNA editing in an in vitro editing system. The editing increase by extracellular calcium is not related to alterations of APOBEC-1 mRNA expression. These data suggest that calcium is not only involved in the regulation of apolipoprotein metabolism but also apoB mRNA editing.

Heat shock protein 60 is a high-affinity high-density lipoprotein binding protein

A new 55-kDa HDL/apolipoprotein binding protein was demonstrated in plasma membrane preparations of the human cell lines and primary cultured hepatocytes. Analysis of specific binding by ligand immunoblots of HDL, apoA-I, and apoA-II to a partially purified 55-kDa PA-I plasma membrane preparation demonstrated a K(d,HDL) = 50 nM (10 microg/ml), K(d,apoA-II) = 20 nM (0.4 microg/ml), and K(d, apoA-I) = 330 nM (10 microg/ml). Following preparative SDS-PAGE electrophoresis of a plasma membrane preparation isolated from human PA-I cells, fractions with apoA-II binding activity were collected, concentrated, and subjected to two-dimensional electrophoresis. Internal microprotein sequencing of the 55-kDa protein band revealed the binding protein as being heat shock protein 60 (hsp60). The hsp60 monoclonal antibody LK-1 blocked apoA-II binding to the 55-kDa HBP preparation. In summary, these results provide a potential mechanism to explain the known association between immunity developed against hsp60 and the development of atherosclerosis.

Detection of apolipoprotein B mRNA editing by peptide nucleic acid mediated PCR clamping

Apolipoprotein B (apoB) mRNA editing leads to a single base change in its mRNA and the production of apoB-48. Currently, the degree of apoB mRNA editing is analyzed by the RT-PCR primer extension method. While this method is quantitative, it is labor intensive, utilizes radioactivity for labeling and may not be sensitive enough to discriminate between low levels of editing and inherent assay background levels. Peptide nucleic acid (PNA) oligonucletides have been used in single point mutation detection through PCR clamping. In the present work, we developed a PCR based assay which can detect the single base change responsible for the apoB-48 production. We found that as low as 0.5% of the edited form can be clearly detected by PNA mediated PCR clamping. When combined with the primer extension assay, an approximately 180-fold enrichment of the edited percentage is observed, reflecting selected PCR amplification of templates containing the edited base.

Differential tissue-specific expression of human apoA-I and apoA-II

To evaluate the sources of high density lipoprotein (HDL) particles containing only apolipoprotein A-I (apoA-I), the synthesis of apoA-I and apolipoprotein A-II (apoA-II) was examined in human liver and small intestine as well as the human intestinally derived cell line, Caco-2. Human liver contained apoA-I, apoA-II as well as apolipoprotein B (apoB) mRNA. In contrast, human adult small intestine total and polyA+ RNA had little or no apoA-II despite the presence of apoA-I and apoB. Intestinal biopsies from normal individuals failed to show de novo apoA-II protein synthesis in the media of organ cultures during [35S]methionine pulse-chase labeling, whereas apoA-I could be readily detected. Caco-2 cells contained apoA-II mRNA and secreted apoA-II protein into the tissue culture media. These data indicate that the primary site of human apoA-II synthesis is in the liver and that the small intestine secretes apoA-I-containing high density lipoproteins.

Apolipoprotein E expression by human-monocyte-derived macrophages. Modulation by opsonised zymosan and cholesterol

The effects of opsonised zymosan and of acetylated low-density lipoprotein (AcLDL) on the synthesis and secretion of apolipoprotein E (apoE), and of apoE mRNA abundance, have been studied in human-monocyte-derived macrophages (MDM). Stimulation by opsonised zymosan led to a concentration-dependent increase in apoE secretion; non-opsonised zymosan was without effect. Incubation with AcLDL led to a concentration-dependent elevation in apoE synthesis which paralleled the increase in cellular cholesterol content. The opsonised-zymosan-induced stimulation of apoE production was additive to that resulting from cholesterol loading with AcLDL. Opsonised zymosan alone did not affect the cholesterol content of MDM. Cholesterol-loaded MDM remained responsive to opsonised zymosan stimulation, displaying a 3.5-fold elevation in apoE secretion as compared to their non-stimulated counterparts. Cell-associated apoE remained at trace levels under all conditions of cell treatment. Studies involving [35S]methionine incorporation showed de novo synthesis of apoE to be enhanced in both cholesterol-loaded and opsonised-zymosan-stimulated macrophages. Estimation of apoE mRNA in opsonised-zymosan-stimulated and control MDM by dot-blot analysis revealed similar message abundance; by contrast, elevation in cellular cholesterol content following incubation with modified LDL led to a significant increase in apoE mRNA levels. We conclude that the opsonised-zymosan-induced stimulation of apoE synthesis and secretion in human MDM may occur by a mechanism(s) independent of cellular cholesterol content.

Characterization of single base substitutions in edited apolipoprotein B transcripts

Mature RNA transcripts from a single eukaryotic gene may contain different nucleotide sequences, ranging from alternately spliced exons to transcripts from separate alleles differing by only one base. Our laboratory and others have recently reported another class of RNA sequence differences, occurring in transcripts from the single copy apolipoprotein B (apoB) gene. A unique RNA editing mechanism allows expression of the CAA glutamine codon encoded by the apoB gene at nucleotide 6666, or terminates translation by the introduction of a premature UAA translational stop codon. In this study, we used the polymerase chain reaction (PCR) to amplify and characterize edited apoB RNA transcripts differing by a single nucleotide. Amplification and sequence analysis from small quantities of total RNA will facilitate the study of RNA editing and transcription in general.

The single prostacyclin receptor of gel-filtered platelets provides a correlation with antiaggregatory potency of PGI2 mimics

Gel-filtered human platelets (GFP) display only a single binding site for [3H]-PGI2: KD = 61nM, 234 fmol/10(8) platelets (1410 sites/platelet). Platelet-rich plasma (PRP) displays the same receptor density but the KD value increases to 123 nM due to protein binding of PGI2 which lowers its effective concentration. The [3H]-PGI2/GFP binding assay has been used to evaluate the molecular basis of aggregation inhibition for prostacyclin analogs and mimics, three PGE type structures, and PGD2. Antiaggregatory IC50s and radioligand binding IC50s correlate for PGE2, E1, and six PGI2 analogs. PGD2, and to a lesser extent 6-oxo-PGE1, display greater antiaggregatory potency than expected based on PGI2-binding site affinity data.

Separate receptors for prostacyclin and prostaglandin E2 on human gel-filtered platelets

Human gel-filtered platelets (GFP) and radiolabeled prostacyclin (PGI2), prostaglandin (PG) E2 and PGE1 were used to ascertain whether PGI2 and PGs of the E series share a common receptor or have their own specific receptors on platelets. Attention was given to ensuring the proper experimental conditions to compensate for the rapid half-life of PGI2 at physiologic pH. Specific [3H] PGI2 binding to GFP was maximal at 5 min and pH 7.45. Scatchard analysis indicated a single class of binding sites with an apparent KD of 4.52 X 10(-8) M and 1130 sites per platelet. Approximately 90% of specifically bound [3H]PGI2 could be dissociated by excess unlabeled PGI2 by 5 min. The IC50 for PGI2 was 66 nM. By 5 min, PGE1 and PGE2 were only 7.17 and 0.03%, respectively, as potent inhibitors of binding. Maximal specific binding of either [3H]PGE2 or [3H]PGE1 to GFP occurred by 60 min. During 60-min incubations with [3H]PGE2, the IC50 values for PGE2 and PGE1 were 3 and 6 nM, respectively. When [3H]PGE1 was used, the IC50 values for PGE1 and PGE2 were 30 and 10 nM, respectively. To examine PGI2 competition for [3H] PGE2 and [3H]PGE1 binding sites, 5-min incubation periods were used. PGI2 was only 0.38% as potent an inhibitor of [3H]PGE2 compared to PGE2 and only 30% as potent an inhibitor of [3H] PGE1 compared to PGE1. Scatchard analysis of the 60-min competition experiments using [3H]PGE2 and [3H]PGE1 and the homologous unlabeled ligand yielded curvilinear plots in both instances.(ABSTRACT TRUNCATED AT 250 WORDS)

Interrelationships between PGE1 and PGI2 binding and stimulation of adenylate cyclase

The effects of prostaglandin E1 (PGE1) and prostacyclin (PGI2) on hepatic adenylate cyclase were studied in plasma membranes isolated from Sprague-Dawley rat livers. Both PGE1 and PGI2 stimulated this enzyme complex to the same maximal levels and with approximately the same EC50 (10(-7) M). Maximally stimulating concentrations of PGE1 and PGI2 were examined alone and together; their effects were not additive, indicating that the same enzyme complex was shared. Although a receptor for PGE1 could be demonstrated with a dissociation constant of 1 X 10(-8) M, PGI2 was only 1/100 as effective in competing for PGE1 binding sites (KD, 1 X 10(-6) M), indicating that these two prostaglandins may act via separate membrane receptors. PGI2 is known to be unstable at neutral pH; however, we have determined its half-life during these assays by a sensitive bioassay and concluded that the degradation of PGI2 is not sufficient to account for its inability to dissociate [3H]PGE1 binding. Further evidence that PGI2 might act through a distinct receptor was found in animals whose PGE1 receptors were 40% downregulated with a corresponding 28% decrease in PGE1-sensitive adenylate cyclase activity. These membranes had no such decrease in PGI2-sensitive adenylate cyclase activity. We conclude that 1) hepatic adenylate cyclase is equally sensitive to PGE1 and PGI2; 2) the same adenylate cyclase complex responds to both prostaglandins; and 3) PGE1 and PGI2 interact with separate membrane receptors in rat liver.

Molecular basis for prostaglandin potency. III. Tests of the significance of the "hairpin conformation" in biorecognition phenomena

Prostaglandin analogs of the PGF2 alpha, 15-epi-PGF2 alpha, and PGE2 type bearing the following methyl substitution patterns -- 15-Me, 16, 16-(Me)2, 17, 17-(Me)2, and 18, 18, 20-(Me)3 -- and analogs constrained to "hairpin" alignment [via 1, (omega-1)-olide formation] and to "non-hairpin" arrangements [via 1, 9- and 1, 15-olide formation] are compared in the following biological assays: contraction of uterine and gastro-intestinal smooth muscle strips, luteolytic antifertility potency in the hamster, binding affinity to two different PGF2 alpha-receptor preparations from bovine corpora lutea, binding to the PGE-specific receptors from rat kidney and liver, inhibition of ADP- induced aggregation of human platelet-rich-plasma, and the effect on rat blood blood pressure. The methylated prostaglandins were also concerted to the corresponding prostacyclins and examined as to action on the platelet and on rat blood pressure. All evidence points to topographically distinct receptors for F2 alpha-, E- and I2- type prostaglandins. Cross-reactivity is reduced in most of the analogs examined. Independent of the target organ or tissue, the receptors show common features based on the functional class of PG recognized. "Hairpin" alignment improves binding (and potency) only for the PGF2 alpha specific assays. PGE-specific binding and potency is disrupted to an increasing extent as the chain branching point is moved further from the 15-hydroxyl center. In contrast 16, 16-dimethylation is particularly disruptive for the PGI2/E1 platelet receptor interaction.

On the multiplicity of platelet prostaglandin receptors. I. Evaluation of competitive antagonism by aggregometry

Methods for the evaluation of competitive interactions at receptors associated with platelet activation and inhibition using aggregometry of human PRP have been developed. The evidence supports the suggestion that PGE1 and PGI2 share a common receptor for inhibition of platelet reactivity, but only a portion (if any) of the aggregation stimulation associated with PGE2 is the result of PGE2 binding (without efficacy) to this receptor. PGE2 (at .3-20 microM) is an effective antagonist of PGE1, PGI2, and PGD2 producing a shift of about one order of magnitude in the IC50-values obtained from complete aggregation inhibition dose response curves. The antagonism of PGD2 inhibition is particularly notable, 80 nM PGE2 levels are detectable. This and other actions of PGE2 indicate another platelet receptor for PGE2. PGE1 acts at both the PGE2 and PGI2 receptor. Other substances showing PGI2-like actions only at high doses (1-30 microM), display additive responses with PGI2 indicative of decreased affinity for the I2/E1 receptor and the absence of PGE2-like aggregation stimulation activity. PGI2 methyl ester has intrinsic inhibitory action not associated with in situ ester hydrolysis. The methyl ester is dissaggregatory showing particular specificity for inhibition of release and second wave aggregation.