ALBERT

Description: Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the Massachusetts Institute of Technology and the Woods Hole Oceanographic Institution February 1981

Description: The equatorial Pacific heat flow low, a major oceanic geothermal anomaly centered on the equatorial sediment bulge, was investigated using deeply penetrating heat flow probes (6-11 meters penetration) within three detailed surveys (400 km2) and along over 10,000 km of continuous seismic profiles (CSP). Previous heat flow measurements in this region defined a broad region characterized by a heat flux well below 1 HFU. We report 98 new measurements collected during cruises PLEIADES 3 and KNORR 73-4 that verify the anomalous nature of the heat flux and also define non-linear temperature gradients (concave down). Temperature field disturbances due to perturbations of a purely conductive heat transport regime are incapable of suitably explaining either of these observations . A simple model incorporating heat transport by both conduction and fluid convection through the sediments fits the observations. A volume flux of (hydrothermal) fluid in the range of 10-6 to 10-5 cm3/sec/cm2 (0.1 liter/yr/cm2) is required. The sense of the flow for all measurements exhibiting non-linear gradients is upward out of the sediment column; no evidence for the recharging of the system was observed. Investigation of a well-defined boundary of the low zone at 4°N and 114°W showed a transition from low and variable heat flow to values compatible with thermal models that correlated with a change in the nature of the basement from rough to smooth. A few outcrops occur in the area of rough basement, but otherwise the region is well-sedimented (greater than 200 meters). Measurements within a detailed survey centered at this transition showed a dramatic increase in heat flow from 1.21 HFU to values greater than 3 HFU over a horizontal distance of 10km. A similar transition from non-linear to linear temperature gradients was not observed as nearly every measurement was non-linear. Heat flow measurements located in well-sedimented, outcrop-free areas (A environments) were associated with linear gradients and a heat flux greater than 1 HFU, however, several of these values were well below the theoretical heat flow for the appropriate age crust. Values measured in environments other than A exhibited variable heat flow and non-linear gradients. The average value of measurements located in A environments within the equatorial Pacific heat flow low was 1.37±0.27 HFU. The previously reported average was 0.92±0.48 HFU based on several measurements from L-DGO cruise VEMA 24-3. The average heat flow measured at a survey located outside the low heat flow zone on crust of 55 ±5 m.a. was 1.76 ±0.30 HFU which is in good agreement with the theoretical value of 1.60. The measurements in this survey were not located in A environments suggesting that crustal convection has ceased or is greatly attenuated within crust of this age. Error analysis of the geothermal data reduction using the convective/conductive heat transport model suggests that the volume flux parameter is sensitive to temperature measurement errors greater than a few millidegrees. Volume fluxes less than 10-7 cm/sec are difficult to distinguish from the purely conductive case assuming instrumental accuracies of 0.001°C. Resolution of the volume flux deteriorates as heat flow decreases and is poor for values less than 0.5 HFU. A detailed survey located within the low zone confirmed previous measurements of low heat flow, however, due to the low value of heat flow (about 0.5 HFU) the small-scale variability could not be clearly defined.

Description: A method for freeze-drying red blood cells (RBCs) while maintaining a high degree of viability, has important implications in blood transfusion and clinical medicine. The disaccharide trehalose, found in animals capable of surviving dehydration can aid in this process. We are reporting a method for loading RBCs with trehalose followed by subsequent freeze-drying and rehydration. The loading of erythrocytes is based on the thermal properties of the RBC plasma membranes and provides efficient uptake of the sugar at 37°C in a time span of 7 hours. The data show that RBCs can be loaded with trehalose from the extracellular medium through a combination of osmotic imbalance and the phospholipid phase transition, producing an intracellular trehalose concentration of about 40 mM. Freeze-drying of trehalose loaded RBCs results in water contents in the range between 2 and 4 % and a level of survival of around 37 %, as measured by the extent of hemolysis. Surprisingly, freeze-dried and rehydrated RBCs showed high levels of ATP and 2,3-DPG and low methemoglobin. Biochemical analysis demonstrated that the activities of superoxide dismutase, catalase and acetylcholine esterase in freeze-dried RBCs are similar to those in fresh RBCs. These data provide an important step toward a stable erythrocyte product, which will be invaluable for transfusion and clinical applications. Supported by DARPA grant N66001-03-1-9827

Description: Sphingomyelin and cholesterol rich platelet membrane domains are organized into lipid rafts, which are present in the liquid ordered state, and which have been suggested to be key membrane components both during cold-induced human platelet activation (Gousset et al. 2002 Evidence for a physiological role for membrane rafts in human platelets. J. Cellular Physiol. 190:117–128) as well as during agonist induced activation. We have previous demonstrated that platelets have two membrane phase transitions, a raft transition at 34–40°C (Gousset et al., 2002 ) and a phospholipid phase transition at 10–20°C (Tablin et al. 1996 The membrane phase transition of intact human platelets: correlation with cold-induced activation. J. Cellular Physiol. 168:305–313). Using Fourier transform infrared spectroscopy (FTIR) we are able to sample all the key phases adopted by intact platelet phospholipids and sphingolipids (gel, liquid crystalline, liquid ordered (raft), hexagonal). This method offers unique advantages for the study of phospholipid acyl chain structures and interactions, as it is non-invasive and can detect molecular vibrations produced by dipole moment oscillations at infrared frequencies. Equally important, these vibrational frequencies are well characterized for individual phospholipids, permitting unambiguous assignment of the transitions to membranes in intact cells (Crowe et al. 1999. Are lipid phase transitions responsible for chilling damage in human platelets? Cryobiology 38:180-101). Most recently, using FTIR we have examined platelet lipid membrane dynamics associated with agonist induced human platelet activation. FTIR analysis of thrombin ( 1U/ml and 0.5U/ml) activated human platelets demonstrates that agonist stimulation results in the development of multiple (four or more) phase transitions, strongly suggestive of phase separation of lipids into like-lipid domains, also referred to as lateral phase separation. These like-lipid domains have phase transition temperatures very similar to that of isolated single lipid species, which, we have previously demonstrated combined to form the main phospholipids phase transition at between 10–20°C (Crowe et al. 1999). Furthermore, when human platelets are repeatedly scanned by FTIR - despite the temperature excursions which might normally result in the remixing of pure lipid populations, the thrombin treated platelets maintained their multiple phase transitions, suggesting that the lateral phase separation is an irreversible event during agonist induced activation. In addition to FTIR studies of human platelet populations, we have also examined the distribution of diIC18 in washed resting, and agonist stimulated human platelets. This dye, which preferentially partitions into liquid ordered or gel phases was uniformly distributed in resting platelets, but became aggregated when platelets were treated by either low concentrations of thrombin (0.01U/ml) or collagen (3.0ug/ml). In addition, platelets treated with higher concentrations of either agonist, demonstrated lipid rafts which further aggregated into larger liquid ordered domains. These results strongly suggest that platelet membrane lipid organization plays a key role in membrane function and further downstream signaling. Funded by NHLBI and DARPA.

Description: A wide variety of medical procedures require transfusion of red blood cells (RBCs). RBCs are currently preserved either at 4°C, at a higher hematocrit (70%) for up to 7–12 weeks or in a frozen state in the presence of glycerol at −80°C for several years. However, each procedure has its demerits. Storage in the dry state offers a possibility for storing the cells for long periods of time under conditions that are far easier to maintain (i.e. room temperature), making the transport to sites of immediate need feasible. We developed a method for freeze-drying RBCs using 15% hematocrit, resulting in a survival of 40% after rehydration, as assessed by the percent hemolysis. In this work, we report the effect of cell hematocrit, concentration of trehalose, salts and overall osmolality of the freeze-drying medium on the survival after freeze-drying and rehydration. Decreasing the percent hematocrit and trehalose in the freeze-drying buffer resulted in about 20% improvement in the post-rehydration survival. Freeze-dried and rehydrated RBCs showed high levels of ATP, 2,3-DPG and low percent methemoglobin. These data are discussed in terms of the glass transition properties of the freeze-drying buffer. This work provides an important step in formulating a freeze drying medium that will provide optimum RBC survival after freeze drying and rehydration.

Description: A 3 year old Thoroughbred horse presented to the Veterinary Teaching Hospital with a three week history of bleeding. Clinical laboratory findings showed normal complete blood count including a normal platelet count. In addition the horse had normal prothrombin and partial thromboplastin times, as well as normal coagulation factors. These data suggested a defect in platelet function. The common pathway of the coagulation cascade occurs at the platelet surface when phosphatidylserine (PS) is flipped to the outer leaflet of the platelet membrane, providing binding sites for the prothrombinase complex that catalyzes the production of thrombin, which carries out the proteolytic activation of fibrinogen. We characterized the capacity of horse’s platelets to carry out each of these steps. PS-flip in the patient’s platelets was induced using several physiological agonists and found to be significantly reduced after stimulation with thrombin (0.1U/ml). The PS was detected by labeling platelets with FITC-annexin V, and measuring FITC intensity by flow cytometry. In comparison to control platelets, approximately 50% of the patient’s platelets were positive for FITC-annexin V after stimulation with thrombin. The fluorescent intensity of the label on the patient’s positive platelets was 40% that of controls. The reduced PS on the patient’s platelets represents a potential loss of prothrombinase binding sites. Using a prothombinase assay based on the hydrolysis the chromogenic substrate S-2238, we determined the amount of thrombin produced by the patient’s platelets over time. In response to thrombin (0.1U/ml), the patient’s platelets produced thrombin at a rate of 0.023±0.002 NIH-U/min/1 X 108 platelets, which was significantly less than control platelets, 0.053±0.004 NIH-U/min/1 X 108 platelets. Neither resting platelets nor the thrombin used to stimulate them produced appreciable hydrolysis of S-2238. We correlated this reduced prothrombinase activity with the capacity of the patient’s platelets bind FITC-fibrinogen. After stimulation with thrombin (0.1U/ml), control platelets form platelet microaggregates that are intensely labeled with FITC-fibrinogen and can be observed by flow cytometry. The patient’s platelets do not form platelet microaggregates and the FITC-fibrinogen labeling is only 20% that of control platelets. This reduced FITC-fibrinogen labeling is not due to a defect in the function of the a2bb3a integrin, because the patient’s platelets bind normal amounts of FITC-fibrinogen in response to ADP. In this regard, the affected horse is unlike patients with Glannzmann’s thrombasthenia, who cannot bind fibrinogen. Instead, the dysfunction is more closely related to individuals with Scott syndrome, who cannot produce the PS surface that is required for assembly of the prothrombinase and ultimately activation of fibrinogen. Ongoing studies using the techniques described above demonstrate one affected and one normal progeny from the original patient.