Tissue oxygenation after exchange transfusion with ultrahigh-molecular-weight tense- and relaxed-state polymerized bovine hemoglobins

Abstract

Hemoglobin (Hb)-based O₂ carriers (HBOCs) constitute a class of therapeutic agents designed to correct the O₂ deficit under conditions of anemia and traumatic blood loss. The O₂ transport capacity of ultrahigh-molecular-weight bovine Hb polymers (PolybHb), polymerized in the tense (T) state and relaxed (R) state, were investigated in the hamster chamber window model using microvascular measurements to determine O₂ delivery during extreme anemia. The anemic state was induced by hemodilution with a plasma expander (70-kDa dextran). After an initial moderate hemodilution to 18% hematocrit, animals were randomly assigned to exchange transfusion groups based on the type of PolybHb solution used (namely, T-state PolybHb and R-state PolybHb groups). Measurements of systemic parameters, microvascular hemodynamics, capillary perfusion, and intravascular and tissue O₂ levels were performed at 11% hematocrit. Both PolybHbs were infused at 10 g/dl, and their viscosities were higher than nondiluted blood. Restitution of the O₂ carrying capacity with T-state PolybHb exhibited lower arterial pressure and higher functional capillary density compared with R-state PolybHb. Central arterial O₂ tensions increased significantly for R-state PolybHb compared with T-state PolybHb; conversely, microvascular O₂ tensions were higher for T-state PolybHb compared with R-state PolybHb. The increased tissue Po₂ attained with T-state PolybHb results from the larger amount of O₂ released from the PolybHb and maintenance of macrovascular and microvascular hemodynamics compared with R-state PolybHb. These results suggest that the extreme high O₂ affinity of R-state PolybHb prevented O₂ bound to PolybHb from been used by the tissues. The results presented here show that T-state PolybHb, a high-viscosity O₂ carrier, is a quintessential example of an appropriately engineered O₂ carrying solution, which preserves vascular mechanical stimuli (shear stress) lost during anemic conditions and reinstates oxygenation, without the hypertensive or vasoconstriction responses observed in previous generations of HBOCs.

the development of hemoglobin (Hb)-based O₂ carriers (HBOCs) represents a concentrated effort to engineer a replacement for red blood cells (RBCs). Although earlier attempts to develop HBOCs were imaginative and stimulated great research effort, few came to fruition, since satisfactory performance, as well as scientific knowledge about how O₂ is transported and extracted from the blood, had not been presented with sufficient evidence. The crisis of infectious disease transmission (in the 1980s) generated renewed interest in the HBOC research field, since the use of RBC replacement solutions allows the adverse consequences of transfusions to be avoided (34). The potential transmission of human immunodeficiency virus stimulated the development of a new generation of potentially useful solutions that could function as RBC substitutes. However, although many of the RBC substitutes showed promise in the initial preclinical testing phase, few survived the rigors of clinical regulatory oversight and safety considerations.

Once the challenges of HBOCs, i.e., the molecular stability of Hb and appropriate O₂ affinity, were partially addressed, some compounds entered clinical testing (16, 24). Through a variety of molecular modifications, more stable Hb polymers were created that effectively minimized renal clearance of the protein (33). Efforts to create modified Hb through recombinant technology were undertaken, with varying degrees of success (33). Unfortunately, these efforts proved to be prohibitively expensive. Scaling the process up to produce the quantities required to meet the anticipated demands were feared to exceed the costs that could be sustained in the eventual marketplace. Nevertheless, this recombinant technology remains an excellent approach for future research.

Current HBOCs result in vasoconstriction, which is evident as a rapid rise in peripheral vascular resistance and blood pressure, commonly associated with bradycardia, decreased cardiac output, and reduced perfusion to vital organs (15, 28, 36). Vasoconstriction is currently perceived to be the critical barrier hampering HBOC development (1, 41). Vasoconstriction appears to be directly linked to nitric oxide (NO) scavenging, to an oversupply of O₂ to the blood vessel wall, and extravasation (19). Regardless of the exact mechanism for vasoconstriction, the presence of acellular Hb in solution enables these various potential mechanisms of vasoconstriction (29).

Polymerization of Hb can be used to yield larger-sized Hb particles that can interact strongly with each other, especially when the polymerized Hb solution is highly concentrated. Hence, polymerization of bovine Hb (bHb) yields high-molecular-weight (MW) fractions that display an increase in viscosity with increasing cross-link density and low colloidal osmotic pressure (COP). At higher cross-link densities (40:1 and 50:1), the bHb polymers (PolybHbs) have the majority of intramolecular cross-linking sites saturated, which facilitates subsequent intermolecular cross-linking between adjacent Hb tetramers in solution. The molecular size of PolybHb has a direct impact on its proximity to the vascular endothelium and its molecular diffusivity, theoretically decreasing NO scavenging and vessel wall O₂ oversupply. These properties should prevent the induction of vasoconstriction.

A previous study (5) published by our group suggested a direct correlation between the hypertensive response and the constriction of resistance arterioles after the hypervolemic infusion of PolybHb solutions with MWs below 500 kDa. On the other hand, PolybHb solutions with high cross-link densities and MWs above 500 kDa increased the plasma O₂ carrying capacity and plasma viscosity without inducing vasoconstriction or microcirculatory disturbances (5). Under extreme anemic conditions, the O₂ delivered becomes insufficient to meet the O₂ demand, and high-MW PolybHb solutions could be a potential alternative to allogeneic blood transfusion. This study was designed to understand how the quaternary state, oxygenated [relaxed (R)] and deoxygenated [tense (T)], of high-MW PolybHbs during polymerization affects their O₂ transport and O₂ delivery characteristics during extreme anemic conditions. The extreme anemic state in the hamster window chamber model is attained at a hematocrit (Hct) of 11% and is a powerful tool to test the efficacy of HBOCs to maintain systemic and microvascular function as well as oxygenation. This Hct is below the threshold at which the animal organism becomes O₂ supply limited and magnifies the effects of transfused HBOCs.

METHODS

Materials.

Glutaraldehyde (70%), NaCl, KCl, NaOH, Na₂S₂O₄, NaCl (USP), KCl (USP), CaCl₂·2H₂O (USP), NaOH (NF), sodium lactate (USP), N-acetyl-l-cysteine (USP), NaCNBH₃, and NaBH₄ were purchased from Sigma-Aldrich (Atlanta, GA). Sephadex G-25 resin was purchased from GE Healthcare (Piscataway, NJ). KCN, KFe(CN)₆, and all other chemicals were purchased from Fisher Scientific (Pittsburgh, PA).

Hb purification.

Fresh bovine blood collected in 3.8% sodium citrate solution at a final concentration of 90:10 (vol/vol) (bovine blood-sodium citrate solution) was purchased from Quad Five (Ryegate, MO). bHb was purified from lysed bovine RBCs (bRBCs) via tangential flow filtration (12, 31). bRBCs were initially washed three times with 3 volumes of isotonic saline solution (0.9%) at 4°C. bRBCs were subsequently lysed on ice with 2 volumes of hypotonic 3.75 mM phosphate buffer at pH 7.4 for 1 h. The lysate was then filtered through a glass chromatography column packed with glass wool to remove the majority of cell debris. Clarified bRBC lysate was then passed through 50-nm and 500-kDa hollow fiber cartridges (Spectrum Labs, Rancho Dominguez, CA) to remove additional cell debris and impurity proteins. Purified bHb was collected and concentrated on a 100-kDa hollow fiber cartridge (Spectrum Labs) to yield the precursor material for PolybHb synthesis.

Polymerization of bHb.

T-state PolybHb was synthesized according to previously described methods (5, 12). To generate fully deoxygenated (T-state) bHb, 30 g of purified bHb were diluted with phosphate buffer (20 mM, pH 8.0) to yield 1,200 ml of bHb solution. The bHb solution was kept in a 2-liter glass bottle that was connected to a vacuum manifold and submerged in an ice water bath. The bHb solution was then subjected to several cycles of vacuum and argon purging to remove the majority of O₂ from the aqueous solution and the gas headspace of the bottle. After 4 h of vacuum and argon cycling, Na₂S₂O₄ solution (1.5 mg/ml) was titrated into the bHb solution while the Po₂ of the solution was simultaneously measured using a RapidLab 248 (Siemens, Malvern, PA) blood gas analyzer until the Po₂ of the bHb solution attained a value of 0 mmHg. At this point, an additional 30 ml of 1.5 mg/ml Na₂S₂O₄ solution were added to the T-state bHb solution to maintain the Po₂ at 0 mmHg during and after the polymerization reaction. A 30-ml syringe with a luer lock was used to titrate glutaraldehyde preequilibrated with argon in the sealed glass bottle under continuous stirring. For this polymerization reaction, the molar ratio of glutaraldehyde to bHb was 50:1.

Oxygenated (R-state) bHb was prepared in a similar manner to T-state bHb using the same vacuum manifold system. In this case, 1,500 ml of 0.3 mmol/l bHb solution were saturated with pure O₂ in the vacuum manifold system for 2 h in an ice water bath. The Po₂ measured was well over the 749-mmHg measurement range of the RapidLab 248 blood gas analyzer. Subsequently, a 30-ml syringe with a luer lock was used to titrate glutaraldehyde in the sealed glass bottle under continuous stirring. For this polymerization reaction, the molar ratio of glutaraldehyde to bHb was 40:1.

The resulting T-state or R-state bHb solutions were then allowed to react with glutaraldehyde at 37°C in a water bath for 2 h, which was stirred and equilibrated with either pure argon (50:1 T-state bHb) or O₂ (40:1 R-state bHb). At the end of this period, for the 50:1 T-state bHb solution, 20 ml of 2 M NaBH₄ in phosphate buffer (20 ml, pH 8.0) were injected into the glass bottle to quench the polymerization reaction. For the 40:1 R-state bHb solution, 5 ml of 8 M NaCNBH₃ in phosphate buffer (20 ml, pH 8.0) were injected into the glass bottle to reduce the Schiff base and reduce the methemoglobin (metHb) level. The PolybHb solution was continuously stirred for 30 min in an ice water bath. Subsequently, 20 ml of 2 M NaBH₄ in phosphate buffer (20 ml, pH 8.0) was injected into the glass bottle to quench the polymerization reaction. The Po₂ of the bHb solution before polymerization, after polymerization, and after quenching with NaBH₄ was measured using a RapidLab 248 blood gas analyzer (Siemens). All reactions were repeated in triplicate.

Separation of the PolybHb solution.

Initially, each PolybHb solution was clarified by passing it through a glass chromatography column packed with glass wool to remove large particles. The glass wool was previously autoclaved at 250°C for 30 min (39) before being used to clarify the PolybHb solution while all tubing, glassware, and plasticware were immersed in 1 M NaOH solution for >6 h to degrade any endotoxin present, followed by a thorough rinse with HPLC grade water (the mean conductivity value being 18.2 × 10⁶ Ω·cm). The clarified PolybHb solution was then separated into two distinct MW fractions with a 500-kDa hollow fiber cartridge (Spectrum Labs). The retentate mostly contained PolybHb molecules that were larger than 500 kDa. This fraction was used for the experiments described in this study.

Buffer exchange of PolybHb solution.

After polymerization, the PolybHb solution was suspended in phosphate buffer along with reduced glutaraldehyde and excess NaBH₄ (and NaCNBH₃ for R-state PolybHb). Glutaraldehyde, NaCNBH₃, and NaBH₄ are cytotoxic (17); therefore, the PolybHb solution was buffer exchanged with a modified lactated Ringer solution [115 mmol/l NaCl (USP), 4 mmol/l KCl (USP), 1.4 mmol/l CaCl₂·2H₂O (USP), 13 mmol/l NaOH (NF), 27 mmol/l sodium lactate (USP), and 2 g/l N-acetyl-l-cysteine (USP)]. The buffer exchange was conducted using an äKTA Explorer 100 system controlled by Unicorn 5.1 software (GE Healthcare). An XK 50/30 (length: 300 mm, inner diameter: 50 mm) column (GE Healthcare) was packed with 500 ml of Sephadex G-25 medium resin at room temperature. The column was balanced with modified lactated Ringer solution at a flow rate of 8 ml/min, and the PolybHb solution was injected into the XK 50/30 column via a superloop column (50 ml, GE Healthcare) at a flow rate of 5 ml/min. The sample (100 ml) was injected at a time and then eluted with modified lactated Ringer solution. The protein concentration was detected at a wavelength of 280 nm while the salt concentration was monitored with a conductivity detector. During the buffer exchange process, the ultraviolet signal increased as PolybHb eluted from the column, whereas the conductivity decreased when reduced glutaraldehyde and NaBH₄ (and NaCNBH₃ for R-state PolybHb) eluted from the column. The buffer-exchanged PolybHb solution was collected as the ultraviolet signal increased but before the conductivity signal decreased. The PolybHb fraction was concentrated with a 100-kDa hollow fiber cartridge (Spectrum Labs).

MetHb level and protein concentration of PolybHb.

The metHb level of PolybHb solutions was measured via the cyanomethemoglobin method (11, 32). Total protein concentration was measured using the Bradford method (3) using the Coomassie Plus protein assay kit (Pierce Biotechnology, Rockford, IL).

Size exclusion chromatography coupled with multiangle static light scattering.

The absolute MW distribution of bHb/PolybHb solutions was measured using a size exclusion chromatography column (Ultrahydrogel linear column, 10 μm, 7.8 × 300 mm, Waters, Milford, MA) driven by a 1200 HPLC pump (Agilent, Santa Clara, CA) and controlled by Eclipse 2 software (Wyatt Technology, Santa Barbara, CA) connected in series to a DAWN Heleos (Wyatt Technology) light scattering photometer and an OptiLab Rex (Wyatt Technology) differential refractive index detector. The mobile phase consisted of 20 mM phosphate buffer (pH 8.0), 100 ppm NaN₃, and 0.2 M NaCl (Fisher Scientific) in HPLC grade water (the mean conductivity value being 18.2 × 10⁶ Ω·cm), which was filtered through a 0.2-μm membrane filter. PolybHb solutions were diluted to 1 mg/ml with the mobile phase, and 60 μl of the sample were injected into the column via a 1200 Autosampler (Agilent). All data were collected and analyzed using Astra 5.3 (Wyatt Technology) software.

O₂-PolybHb equilibria.

The O₂ affinity and cooperativity coefficient of bHb/PolybHb solutions were regressed from O₂-PolybHb equilibrium curves measured on a Hemox Analyzer (TCS Instruments, Southampton, PA) at 37°C.

Samples were prepared by thoroughly mixing 100 μl of the sample with 5 ml Hemox buffer (pH 7.4, TCS Instruments), 20 μl Additive-A, 10 μl Additive-B, and 10 μl antifoaming agent. The PolybHb sample was allowed to equilibrate to a Po₂ of 145 ± 2 mmHg using compressed air. After the sample had been given sufficient time to equilibrate, the gas stream was switched to pure N₂ to deoxygenate the bHb/PolybHb sample. The absorbance of oxy- and deoxy-Hb (Abs) in solution was recorded as a function of Po₂ via dual-wavelength spectroscopy. O₂-PolybHb equilibrium curves (Y) were fit to a four-parameter (A₀, A_∞, P₅₀, n) Hill model (Eq. 1) as follows:

$$Y=\frac{{\text{Abs}-A}_0}{A_{\infty }-A_0}=\frac{{\text{Po}}_2^n}{{\text{Po}}_2^n+{\text{P}}_{50}^n}$$ (1)

where A₀ and A_∞ are the absorbance at 0 mmHg and full saturation, respectively; Abs is the absorbance of the sample; n is the cooperativity coefficient; and P₅₀ is the Po₂ at which the bHb/PolybHb is half-saturated with O₂ (O₂ affinity).

PolybHb viscosity and COP.

PolybHb viscosity was measured in a cone and plate viscometer DV-II plus with a cone spindle CPE-40 (Brookfield Engineering Laboratories, Middleboro, MA) at a shear rate of 160 s⁻¹. The COP of PolybHb was measured using a Wescor 4420 Colloid Osmometer (Wescor, Logan, UT) (40).

Animal preparation.

In vivo experiments were performed in 55- to 65-g male Golden Syrian hamsters (Charles River Laboratories, Boston, MA) fitted with a dorsal skinfold window chamber. The hamster window chamber model is widely used for microvascular studies in the unanesthetized state. The complete surgical technique has been described in detail elsewhere (9, 13). Arterial and venous catheters filled with a heparinized saline solution (30 IU/ml) were implanted into the carotid and jugular vessels. Catheters were tunneled under the skin, exteriorized at the dorsal side of the neck, and securely attached to the window frame. Animal handling and care followed the National Institutes of Health Guide for the Care and Use of Laboratory Animals. The experimental protocol was approved by the local animal care committee.

Inclusion criteria.

The microvasculature was examined 3–4 days after the window implantation surgery, and only animals that passed established systemic and microcirculatory inclusion criteria were used. Animals were considered suitable for experiments if 1) systemic parameters were within normal range, namely, heart rate (HR) > 340 beats/min, mean arterial blood pressure (MAP) > 80 mmHg, systemic Hct > 45%, and arterial Po₂ > 50 mmHg; and 2) microscopic examination of the tissue in the chamber observed under ×650 magnification did not reveal signs of low perfusion, inflammation, edema, or bleeding.

Experimental setup.

The unanesthetized animal was placed in a restraining tube with a longitudinal slit from which the window chamber protruded and then fixed to the microscopic stage for transillumination with the intravital microscope (BX51WI, Olympus, New Hyde Park, NY). Animals were given 20 min to adjust to the tube environment before any measurements were made. The tissue image was projected onto a charge-coupled device camera (4815, COHU, San Diego, CA) connected to a videocassette recorder and viewed on a monitor. Measurements were carried out using a ×40 (LUMPFL-WIR, numerical aperture: 0.8, Olympus) water-immersion objective.

Test solutions.

PolybHb solution concentrations were adjusted to 10 g Hb/dl. PolybHb solutions were subdivided by the quaternary state of the bHb during the polymerization reaction, namely, T-state PolybHb and R-state PolybHb. The biophysical properties of both PolybHb solutions are shown in Table 1.

Table 1.

Biophysical properties of the solutions

	Hb, g/dl	Viscosity, cP	COP, mmHg	P50, mmHg	n	MetHb, g/dl	Tetrameric Hb, %	Molecular Weight, MDa
Plasma expander
6% Dextran (70 kDa)		2.8	50					0.07
Hb-based O2 Carriers
T-PolybHb	10.0	11.4	1	40.0	0.9	3.3	0	16.59
R-PolybHb	10.0	7.8	7	1.0	0.7	3.4	0	26.33

Hb, hemoglobin; COP, colloid osmotic pressure; P₅₀, Po₂ at which 50% of the Hb is saturated with O₂; n, cooperativity coefficient; metHb, methemoglobin; T-PolybHb, tense-state bovine Hb polymers; R-PolybHb, relaxed-state bovine Hb polymers. Viscosity was measured at a shear rate of 160 ⁻¹ at 37°C.

Acute isovolemic exchange transfusion (hemodilution) protocol.

Acute anemia was induced by two isovolemic hemodilution steps. This protocol has been described in detail in our previous reports (4, 37, 38) (Fig. 1). Briefly, the volume of each exchange-transfusion step was calculated as a percentage of the blood volume, estimated as 7% of body weight. The acute anemic state was induced by lowering systemic Hct to 18% by two steps of progressive isovolemic hemodilution using 6% dextran (70 kDa, Pharmacia, Uppsala, Sweden). Blood was simultaneously withdrawn at the same rate from the carotid artery catheter according to a previously established protocol (4, 37). The first exchange was 40% of the blood volume, and the second exchange was 35% of the blood volume, respectively. Moderate hemodiluted animals were randomly divided into two experimental groups (2). The exchange-transfusion protocol was continued by exchanging 35% of the blood volume with the test solution (experimental groups are named accordingly). The duration of the experiments was 4 h. Each exchange and the respective observation time points after the exchange were completed in 1 h. Systemic and microcirculation data were taken after a stabilization period of 15 min.

Fig. 1.

Hemodilution was attained by means of a progressive, stepwise, isovolemic hemodilution protocol in which the red blood cell (RBC) volume was continuously decreased and the plasma volume was increased while the total blood volume constant was maintained (dotted line). The extreme anemic state was achieved by two hemodilution exchanges (first 40% followed by 35% of blood volume) and an exchange transfusion step (35% of the blood volume) using either tense-state or relaxed-state bovine hemoglobin (bHb) polymers (T-PolybHb and R-PolybHb, respectively; at 10 g/dl).

Experimental groups.

The two experimental groups were labeled as follows: R-state PolybHb and T-state PolybHb.

Systemic parameters.

MAP and HR were recorded continuously (MP 150, Biopac Systems, Santa Barbara, CA). Hct was measured from centrifuged arterial blood samples taken in heparinized capillary tubes. Hb content was determined spectrophotometrically (B-Hemoglobin, Hemocue, Stockholm, Sweden).

Blood chemistry and biophysical properties.

Arterial blood was collected in heparinized glass capillaries (0.05 ml) and immediately analyzed for arterial Po₂, arterial Pco₂, base excess, and pH (Blood Chemistry Analyzer 248, Bayer, Norwood, MA). The comparatively low arterial Po₂ and high arterial Pco₂ of these animals are a consequence of their adaptation to a fossorial environment. Blood samples for viscosity and COP measurements were quickly withdrawn from the animal with a heparinized 5-ml syringe at the end of the experiment for immediate analysis. Viscosity was measured in a cone/plate viscometer DV-II plus with a cone spindle CPE-40 (Brookfield Engineering Laboratories) at a shear rate of 160 s⁻¹. COP was measured using a Wescor 4420 Colloid Osmometer (Wescor) (40).

Functional capillary density.

Functional capillary density [FCD; number of capillaries per unit of area (in cm⁻¹)] is the total number of RBC-perfused capillaries divided by the area of the microscopic field of view (37). Capillary segments were considered functional if RBCs were observed to transit over a 60-s period. FCD was tabulated from the number of capillaries with RBC flow in an area composed of 10 successive microscopic fields (420 × 320 μm). Detailed mappings were made of the chamber vasculature to study the same microvessels throughout the experiment.

Microhemodynamic parameters.

Arteriolar and venular blood flow velocity were measured online using the photodiode cross-correlation technique (Fiber Optic Photo Diode and Velocity Tracker model 102B, Vista Electronics, San Diego, CA) (20). The centerline velocity was corrected according to vessel size to obtain the mean RBC velocity (V) (25). The video image shearing technique was used to measure vessel diameter (D) online. Blood flow (Q) was calculated from the measured parameters as follows: Q = πV(D/2)².

Microvascular Po₂ distribution.

High-resolution noninvasive microvascular Po₂ measurements were made using phosphorescence quenching microscopy (PQM) (21, 37). PQM is based on the O₂-dependent quenching of phosphorescence emitted by an albumin-bound metalloporphyrin complex after pulsed light excitation. PQM is independent of the dye concentration within the tissue and is well suited for detecting hypoxia because its decay time is inversely proportional to the Po₂ level, causing the method to be more precise at low Po₂ levels. This technique is used to measure both intravascular and extravascular Po₂ since the albumin-dye complex continuously extravasates from the circulation into the interstitial tissue (21, 37). Tissue Po₂ was measured in tissue regions in between functional capillaries. PQM allows for the precise localization of Po₂ measurements without subjecting the tissue to injury. These measurements provide a detailed understanding of microvascular O₂ distribution and indicate whether O₂ is delivered to the interstitial areas.

O₂ delivery and extraction.

The microvascular methodology used in our study allows the detailed analysis of O₂ delivery to the tissue. Calculations were made using the following equations for O₂ delivery (Do₂; Eq. 2) and O₂ extraction (Vo₂; Eq. 3) (4):

$${\text{Do}}_2=\left[\left({\text{RBC}}_{\text{Hb}}\times \gamma \times {\text{S}}_{\text{A}}\right)+\left({\text{Plasma}}_{\text{Hb}}\times \gamma \times {Š}_{\text{A}}\right)\right]\times \text{Q}$$ (2)

$${\text{Vo}}_2=\left[\left({\text{RBC}}_{\text{Hb}}\times \gamma \times {\text{S}}_{\text{A-V}}\right)+\left({\text{Plasma}}_{\text{Hb}}\times \gamma \times {Š}_{\text{A-V}}\right)\right]\times \text{Q}$$ (3)

where RBC_Hb is the Hb concentration derived from RBCs and is equal to the total Hb − plasma Hb (g Hb/dl blood), Plasma_Hb is the concentration of acellular Hb (g Hb/dl blood), γ is the O₂ carrying capacity of saturated Hb (1.34 ml O₂/g Hb), S_A (in %) is the arteriolar RBC O₂ saturation, and Š_A (in %) is the arteriolar PolybHb O₂ saturation. The subscript A-V indicates arteriolar/venular differences, and Q is the microvascular flow. Fresh hamster RBCs at pH 7.4 and 37.6°C had a P₅₀ (50% of the Hb is saturated with O₂) of 32 mmHg and Hill number of 2.9 measured using the Hemox Analyzer (TCS).

Data analysis.

Tabular results are presented as means ± SD. The box-whisker plot separates the data into quartiles, with the top of the box defining the 75th percentile, the line within the box giving the median, and the bottom of the box showing the 25th percentile; the upper whisker defines the 95th percentile, and the lower whisker defines the 5th percentile. Data within each group were analyzed using ANOVA for repeated measurements (Kruskal-Wallis test). When appropriate, post hoc analyses were performed with the Dunns multiple-comparison test. Microhemodynamic data are presented as absolute values and ratios relative to baseline values. A ratio of 1.0 signifies no change from baseline, whereas lower and higher ratios are indicative of changes proportionally lower and higher than baseline (i.e., 1.5 would mean a 50% increase from the baseline level). The same vessels and capillary fields were followed so that direct comparisons to their baseline levels could be performed, allowing for more robust statistics for small sample populations. All statistics were calculated using GraphPad Prism 4.01 (GraphPad Software, San Diego, CA). Changes were considered statistically significant if P < 0.05.

RESULTS

Po₂ of PolybHb solutions.

bHb was polymerized with glutaraldehyde in the T-state (50:1) and R-state (40:1). Subsequently, each PolybHb mixture was separated into two distinct MW fractions: <500 kDa and >500 kDa. The fraction above 500 kDa in MW was used in this study. For T-state PolybHb, the Po₂ was kept at 0 mmHg before and after polymerization and after quenching, which indicates that the bHb was maintained in the fully deoxygenated state (T-state) during the polymerization process. The Po₂ of the R-state PolybHb solution before and after polymerization was above the measurement range of the O₂ detector. This shows that the bHb was fully saturated with O₂ (R-state). After the R-state bHb polymerization reaction was quenched with NaCNBH₃ and NaBH₄, the Po₂ of the R-state PolybHb solution dropped to 0 mmHg, since the NaBH₄ generated H₂ gas, which displaced all O₂ from the solution.

MW distribution of PolybHb solutions.

The results shown in Table 1 demonstrate that the weight-averaged MW of T-state PolybHb was slightly smaller in magnitude compared with R-state PolybHb. Both PolybHb solutions possessed large MWs, ranging from 16.59 to 26.33 MDa, and did not show the presence of bHb in solution.

P₅₀ and n of PolybHb solutions.

The results shown in Table 1 demonstrate that R-state PolybHb had a lower P₅₀ compared with bHb, which itself had a lower value compared with T-state PolybHb. The n values of T-state and R-state PolybHb solutions were <1.

MetHb levels of PolybHb solutions.

The metHb levels of native bHb, T-state PolybHb, and R-state PolybHb are shown in Table 1. The metHb level of native bHb was very low (<1%), since it was purified from fresh bRBCs. T-state PolybHb and R-state PolybHb had similar metHb levels, which were below 5%.

Exchange transfusion with T-state and R-state PolybHb.

Twelve animals were entered into the study; all animals tolerated the entire protocol without visible signs of discomfort. Six animals were assigned to each experimental group. Groups were statistically similar (P > 0.30) in systemic and microcirculation parameters at baseline and moderate hemodilution. Systemic and microhemodynamic datasets for baseline and moderate hemodilution were obtained by combining data from all experimental groups.

Systemic parameters.

Blood chemistry parameters at baseline, moderate hemodilution, and after the exchange transfusion are shown in Table 2. Moderate hemodilution reduced the Hct and Hb concentrations. The Hct for all experimental groups was similar (11.3 ± 0.4%). Total Hb was increased from the moderate hemodilution level after exchange with T-state and R-state PolybHb. The Hct after the exchange transfusion remained stable over time until the end of the experiment for both experimental groups.

Table 2.

Systemic parameters

			Exchange Transfusion
	Baseline	MH	T-PolybHb	R-PolybHb
Number of animals	12	12	6	6
Hematocrit, %	48.6 ± 0.8	18.3 ± 0.6*	11.3 ± 0.5*†	11.4 ± 0.4*†
Total Hb, g/dl	14.8 ± 0.4	5.8 ± 0.3*	6.2 ± 0.3*†	6.1 ± 0.3*†
Plasma Hb, g/dl			2.9 ± 0.3	2.8 ± 0.4
Arterial Po2, mmHg	59.4 ± 5.2	77.2 ± 6.5*	94.6 ± 6.2*†	126.2 ± 8.4*†‡
Arterial Pco2, mmHg	53.8 ± 4.8	46.5 ± 5.4*	44.3 ± 4.5*†	37.6 ± 6.3*†‡
Arterial pH	7.338 ± 0.018	7.372 ± 0.022*	7.369 ± 0.024*	7.387 ± 0.026*
Arterial base excess, mmol	3.0 ± 1.5	1.2 ± 0.9*	0.6 ± 0.8*	−2.2 ± 1.7*†‡

Values are means ± SD. Baseline and moderate hemodilution (MH) measurements were made in all animals. No significant differences were detected between baseline or before exchange transfusion (MH) values between groups.

^*P < 0.05 compared with baseline;

^†P < 0.05 compared with MH;

^‡P < 0.05 between T-PolybHb and R-PolybHb.

Results from the systemic arterial blood gas analysis are shown in Table 2. The arterial Po₂ was statistically significantly increased from baseline after moderate hemodilution. The exchange transfusion with T-state and R-state PolybHb statistically significantly increased the arterial Po₂ compared with baseline and moderate hemodilution, respectively. However, the arterial Po₂ after the exchange transfusion with T-state PolybHb was statistically lower than after the exchange transfusion with R-state PolybHb. The arterial Pco₂ decreased significantly from baseline after moderate hemodilution. The exchange transfusion with T-state and R-state PolybHb further increased the arterial Pco₂ compared with baseline and moderate hemodilution. The exchange with R-state PolybHb had a more pronounced increase in arterial Pco₂ compared with T-state PolybHb. Arterial pH was higher than baseline after moderate hemodilution, and the exchange transfusion with T-state and R-state PolybHb maintained a higher arterial pH compared with baseline. The blood acid-base balance was statistically significantly decreased after moderate hemodilution and in both experimental groups compared with baseline. The exchange transfusion with R-state PolybHb produced severe blood acid-base unbalance compared with T-state PolybHb.

Changes in MAP and HR are shown in Fig. 2. MAP was statistically decreased from baseline after moderate hemodilution. The exchange transfusion with T-state PolybHb maintained the MAP achieved after moderate hemodilution and was statistically lower than baseline, whereas R-state PolybHb increased MAP compared with moderate hemodilution and remained statistically lower than baseline. HR was no different from baseline after moderate hemodilution, and it remained no different after the exchange transfusion with PolybHb solutions.

Fig. 2.

Relative changes from baseline in mean arterial pressure (MAP) and heart rate (HR) after moderate hemodilution and exchange transfusion with T-PolyHb and R-state PolybHb. The dotted line represents the baseline (BL) level. †P < 0.05 relative to BL; ‡P < 0.05 compared with moderate hemodilution; §P < 0.05 among groups. Baseline MAP values (means ± SD) for each group were as follows: 114 ± 6 mmHg for T-PolybHb and 111 ± 7 mmHg for R-PolybHb. Baseline HR values (means ± SD) for each group were as follows: 434 ± 29 beats/min for T-PolybHb and 426 ± 27 mmHg for R-PolybHb.

Blood biophysical properties after exchange.

Blood viscosity, plasma viscosity, and plasma COP after hemodilution for baseline and both experimental groups are shown in Table 3. Blood viscosities were lower than baseline for both exchange-transfused groups. Animals exchange transfused with T-state PolybHb had statistically higher blood and plasma viscosity than animals exchange transfused with R-state PolybHb. Plasma viscosity did not change from baseline. Plasma COP was lower for both exchange-transfused groups compared with baseline.

Table 3.

Blood rheological properties, shear rates, and shear stress

		Exchange Transfusion
	Baseline	T-PolybHb	R-PolybHb
Blood viscosity, cP	4.2 ± 0.2	2.9 ± 0.3*	2.5 ± 0.2*†
Plasma viscosity, cP	1.2 ± 0.1	2.1 ± 0.2*	1.7 ± 0.1*†
Plasma COP, mmHg	18 ± 2	17 ± 1	17 ± 2
Arteriolar WSR, s−1	628 ± 79	588 ± 102*	573 ± 95*†
Arteriolar WSS, dyn/cm2	27.5 ± 4.9	16.8 ± 5.6*	15.2 ± 6.2*†
Venular WSR, s−1	382 ± 68	454 ± 91*	404 ± 105*†
Venular WSS, dyn/cm2	15.4 ± 6.2	12.6 ± 4.6*	9.6 ± 4.2*†

Values are means ± SD. Shear rate was 160 s⁻¹ at 37°C for blood viscosity; COP was measured at 27°C. WSR, wall shear rate; WSS, wall shear stress.

^*P < 0.05 compared with baseline (nonhemodiluted diluted animals);

^†P < 0.05 between groups.

Microhemodynamics.

Microvascular parameters were characterized for large feeding and small arcading arterioles (range: 45–74 μm) and small collecting venules (range: 46–76 μm). Changes in diameter and blood flow from baseline, as well as absolute values, are shown Fig. 3. Arterioles after moderate hemodilution were dilated, and RBC velocity and blood flow were statistically increased from baseline. Arterioles were statistically constricted for both exchange-transfused groups compared with moderate hemodilution but no different from baseline. Arteriolar blood flows after moderate hemodilution were statistically higher than baseline. Arteriolar blood flows after exchange transfusion with T-state and R-state PolybHb were statistically lower compared with moderate hemodilution but no different from baseline. Animals exchange transfused with R-state PolybHb had statistically lower arteriolar blood flows than animals exchange transfused with T-state PolybHb.

Fig. 3.

Relative changes from baseline in arteriolar and venular hemodynamics after moderate hemodilution followed by an exchange transfusion with either T-PolybHb or R-PolybHb. The dotted line represents the BL level. †P < 0.05 relative to BL; ‡P < 0.05 compared with moderate hemodilution; §P < 0.05 among groups. Diameters (means ± SD) in each animal group were as follows: arterioles, 61.7 ± 7.9 μm, n = 34, and venules, 69.1 ± 8.3 μm, n = 36, for BL T-PolybHb; and arterioles, 63.5 ± 8.9, n = 32, and venules, 67.8 ± 9.2, n = 33, for BL R-PolybHb, where n is the number of vessels studied. RBC velocities (means ± SD) in each animal group were as follows: 4.2 ± 0.8 mm/s in arterioles and 1.9 ± 0.8 mm/s in venules for BL T-PolybHb; and 4.3 ± 0.9 mm/s in arterioles and 1.8 ± 0.9 mm/s in venules for BL R-PolybHb. Flow values (means ± SD) in each animal group were as follows: 13.9 ± 4.8 nl/s in arterioles and 7.2 ± 3.2 nl/s in venules for BL T-PolybHb and 14.7 ± 4.4 nl/s in arterioles and 7.3 ± 3.0 nl/s in venules for BL R-PolybHb.

Venular microvascular tone and blood flow changes are also shown in Fig. 3. Moderate hemodilution did not produce dilation; however, it statistically increased RBC velocity and blood flow compared with baseline. Venular blood flows after the exchange transfusion decreased compared with moderate hemodilution in both groups but were no different from baseline. Shear rate and shear stress values for all experimental groups are shown in Table 3.

FCD.

FCD values after moderated hemodilution and the exchange transfusion are shown in Fig. 4. Moderated hemodilution decreased FCD compared with baseline. The exchange transfusion with T-state and R-state PolybHb further decreased FCD compared with baseline and moderate hemodilution. The exchange transfusion with R-state PolybHb produced lower FCD than the exchange transfusion with T-state PolybHb.

Fig. 4.

Functional capillary density (FCD) relative to BL after moderate hemodilution followed by an exchange transfusion with either T-PolybHb or R-PolybHb. The dotted line represents the BL level. †P < 0.05 relative to BL; ‡P < 0.05 compared with moderate hemodilution; §P < 0.05 among groups. FCD at BL was 122 ± 16 capillaries/cm for T-PolybHb and 119 ± 17 capillaries/cm for R-PolybHb.

Microvascular O₂ tensions.

Microvascular and tissue Po₂ values are shown in Fig. 5. The exchange transfusion with R-state PolybHb produced lower arterial O₂ tensions than the exchange transfusion with T-state PolybHb. Similar results were found for interstitial tissue O₂ tensions and venular O₂ tensions. Interstitial tissue Po₂ in both experimental groups was lower compared with the normal tissue Po₂ without hemodilution (full Hct), which is 21.7 ± 3.5 mmHg (8).

Fig. 5.

Intravascular and extravascular Po₂ after hemodilution followed by an exchange transfusion with either T-PolybHb or R-PolybHb. §P < 0.05 among groups.

Microvascular O₂ delivery and extraction.

Figure 6 shows resulsts from the analysis of systemic and microvascular O₂ delivery and extraction. Systemic O₂ delivery was 8% higher for T-state PolybHb than for R-state PolybHb. Premicrocirculation O₂ extraction was statistically higher for T-state PolybHb compared with R-state PolybHb. Arteriolar O₂ delivery after extreme hemodilution with T-state or R-state PolybHb was no different. Microcirculation O₂ extraction was statistically higher for T-state PolybHb compared with R-state PolybHb. The venular O₂ reserve was statistically higher for R-state PolybHb compared with T-state PolybHb.

Fig. 6.

A: arterial O₂ supply (total O₂ transported by RBCs and PolybHb). B: arterial to arteriolar O₂ extraction (O₂ released from RBCs and PolybHb before the blood arrives to the microcirculation). C: arteriolar O₂ supply [O₂ transported by RBCs and PolybHb that is available (i.e., arrives) to microvascular tissues]. D: arterial to venular O₂ extraction (O₂ released at the microvascular level from RBCs and PolybHb). E: venular O₂ reserve (O₂ remaining in RBCs and PolybHb and returned to the central circulation). †P < 0.05, total O₂ RBCs plus PolybHb; ‡P < 0.05, O₂ from RBCs; §P < 0.05, O₂ from PolybHb.

Systemic O₂ extraction from the remaining RBCs after hemodilution was statistically higher after the exchange with R-state PolybHb compared with T-state PolybHb. Animals exchanged with R-state PolybHb extracted 98% of O₂ in the RBCs compared with 80% for the animals exchanged with T-state PolybHb. Systemic O₂ extraction from the polymerized Hb in the plasma was 70% for animals exchanged with T-state PolybHb compared with 16% for animals exchanged with R-state PolybHb. A similar result was observed at the microvascular level, where animals exchanged with T-state PolybHb extracted 75% of O₂ available in RBCs and 59% of O₂ available in T-state PolybHb compared with R-state PolybHb, in which 97% of O₂ available in RBCs and 12% of O₂ available in R-state PolybHb was extracted.

DISCUSSION

Biophysical properties of T-state and R-state PolybHb solutions.

The goal of this study was to examine the role of high-MW PolybHb O₂ affinity (P₅₀) in regulating tissue oxygenation. We hypothesized that the P₅₀ of a PolybHb solution is a major determinant of tissue oxygenation. To test this hypothesis, two types of high-MW PolybHb solutions were synthesized with different initial quaternary states (T and R states) and used for the in vivo evaluation of systemic and microcirculatory parameters.

bHb was polymerized under anaerobic and aerobic conditions. For T-state PolybHb, the polymerization process minimized the oxidation of the heme since bHb was polymerized in an anoxic environment (Po₂ = 0 mmHg) and quenched with the strong reducing agent NaBH₄. For R-state PolybHb, bHb was polymerized in a saturated O₂ environment (Po₂ > 749 mmHg), and some of the oxidized bHb was able to be reduced after quenching with both the mild reducing agent NaCNBH₃ and the strong reducing agent NaBH₄. As a further precaution against the autooxidation of the heme, PolybHb solutions were buffer exchanged against modified lactated Ringer solution containing N-acetyl-l-cysteine, an antioxidant (43), which limits heme oxidation. T-state and R-state PolybHb solutions possessed very similar weight-averaged MWs (Table 1), although R-state PolybHb had a slightly higher weight-averaged MW versus T-state PolybHb. This can be easily explained by the presence of Na₂S₂O₄ in the T-state bHb polymerization reaction. During the T-state bHb polymerization process, extra Na₂S₂O₄ was titrated into the bHb solution after the Po₂ attained a value of 0 mmHg to maintain the bHb solution in the T state during the glutaraldehyde polymerization process. The excess Na₂S₂O₄ quenched some of the glutaraldehyde in solution, thus reducing the weight-averaged MW of the T-state PolybHb solution compared with the R-state PolybHb solution.

For T-state PolybHb, the Po₂ of the initial bHb solution was maintained at 0 mmHg before polymerization. Therefore, all bHb molecules in solution were maintained in the low-O₂ affinity T state. After polymerization of T-state bHb, the reducing agent NaBH₄ was used to quench any remaining glutaraldehyde in solution as well as to reduce the resulting Schiff bases. Hence, T-state PolybHb was frozen in the low-O₂ affinity state via intra- and intermolecular glutaraldehyde cross-links within the bHb tetramer and between tetramers. The P₅₀ of high-MW T-state PolybHb (40 mmHg) was statistically different to that of unmodified bHb (26 mmHg) (30) and similar to the reported value of 38 mmHg for Hemopure (a glutaraldehyde cross-linked bHb manufactured by Biopure) (36). In the case of R-state PolybHb, bHb was maintained in the R state (high O₂ affinity) during the polymerization process and subsequently quenched with both NaCNBH₃ and NaBH₄. Therefore, R-state PolybHb was frozen in the high-O₂ affinity R state via intra- and intermolecular glutaraldehyde cross-links within the bHb tetramer and between tetramers. The P₅₀ of R-state PolybHb is 1 mmHg. The cooperativity of the two PolybHb solutions is much less compared with the reported value for unmodified bHb (2.5) (30). This behavior is explained by the glutaraldehyde cross-links present within and between the globin chains, which restrict the transmission of any quaternary changes in the Hb structure to other neighboring globin chains within the Hb tetramer. This results in a significant loss of cooperative binding of O₂ molecules to the Hb tetramer.

Exchange transfusion with T-state and R-state PolybHb.

The principal finding of this study was that under identical extreme anemic conditions, bHb polymerized in the T state (deoxygenated; T-state PolybHb: P₅₀ = 40 mmHg) significantly improved systemic and microvascular oxygenation compared with bHb polymerized in the R state (oxygenated; R-state PolybHb: P₅₀ = 1 mmHg). The appreciably increased tissue Po₂ attained with T-state PolybHb appears to be due to 1) the higher amount of O₂ released from the PolybHb; 2) the maintenance of macrovascular hemodynamics (e.g., blood pressure and HR); and 3) the microvascular hemodynamics (e.g., vascular tone and blood flow) compared with R-state PolybHb.

In the initial phase of the protocol to reach the moderate hemodilution state, blood was exchanged with dextran solution as a colloidal plasma expander to lower the intrinsic O₂ carrying capacity (Hct) of blood, which also changes the rheological properties of blood. The compensatory mechanisms responding to the acute decrease in Hct involve the increase of cardiac output due to a reduction in vascular resistance due to a lower blood viscosity (10, 14, 18, 23). During hemodilution, blood viscosity is an important factor in the responses of the cardiovascular system, as it affects endothelial shear stress, which activates the synthesis of vascular autocoids, such as prostacyclin and NO (22, 26, 27, 35, 42). Beyond the moderate hemodilution level, exchange transfusion using PolybHb at 10 g/dl increases blood O₂ content and plasma viscosity; however, blood viscosity remains lower than nonhemodiluted blood. Mechanistically, from a rheological standpoint, further decreases in Hct paired with an increase in plasma viscosity with high-viscosity O₂ carriers can partially preserve vascular shear stress stimuli plus sustain oxygenation. Thus, T-state PolybHb acted as a viscogenic O₂ carrier, maintaining both shear stress and oxygenation, without increasing vascular resistance consequent to an increase in plasma viscosity, leading to an improved transmission of central pressure to the microcirculation sustaining FCD.

Exchange transfusion with either T-state or R-state PolybHb resulted in a higher total Hb in the circulation (6.2 and 6.1 g/dl) than moderate hemodilution (5.8 g/dl). However, arterial blood gases (Po₂ and Pco₂) were drastically changed after the exchange transfusion, which suggests significant adjustments in the way that the respiratory system exchanged gases as a result of the presence of PolybHb in the blood. In an attempt to improve oxygenation, animals exchange transfused with R-state PolybHb increased arterial Po₂, which only increased the dissolved O₂ within the plasma without affecting O₂ transported or delivered by the remaining RBCs and R-state PolybHb. Moreover, the calculated O₂ extraction for animals exchanged transfused with R-state PolybHb showed that almost all the O₂ loaded into the RBCs was extracted, and only a small fraction of the O₂ in R-state PolybHb was ever released to hypoxic tissues. These results suggest that extremely high O₂ affinity of R-state PolybHb prevents the O₂ bound to R-state PolybHb from been used by the tissues. Since both groups had the same Hct after the exchange transfusion, changes in O₂ delivery and extraction are mostly due to the biophysical properties of the PolybHb. It appears that O₂ affinity appears to be a limiting factor for the effective transport and release of O₂.

The differences in oxygenation, tissue Po₂, MAP, and arteriolar blood flow between T-state and R-state PolybHb shows that in designing a HBOC, it is not sufficient to provide adequate O₂ carrying capacity. Once a suitable O₂ carrier is available, it must also be able to maintain or enhance other circulatory transport parameters, particularly blood flow. The large polymers of bHb used in this study are practically vasoinactive, and the difference in P₅₀ appears to be the determining factor involved in oxygenation. Introducing into the circulation PolybHb with P₅₀ slightly higher than the hamster blood (P₅₀ = 32 mmHg) will deliver O₂ earlier to the tissues and preserve O₂ in the RBCs to be released to the tissues that need it. On the other hand, the presence of even considerable amounts of strongly left-shifted Hbs would prevent O₂ transport in the organism.

In summary, the results presented here show that T-state PolybHb is an efficient high-viscosity O₂ carrier. When used during extreme anemic conditions, it produced a superior systemic and microvascular hemodynamic and oxygenation outcome compared with R-state PolybHb. The effects on microvascular hemodynamics were comparable, if not better, than those previously reported in a similar protocol (Table 4). These results show that maintaining blood rheological properties during extreme anemic conditions by means of the high-viscosity PolybHb increases oxygenation beyond the already observed benefit with a high-viscosity plasma expander (6% 500-kDa dextran). Previous results using high-viscosity plasma expanders include vasodilation and increased microvascular flow (37), sustenance of capillary pressure, maintenance of FCD (7), increased production of endothelium-mediated factors (35), and enhanced organ flow (6), effects that we speculate remain partially valid for T-state PolybHb. The potentially negative effect of increasing peripheral vascular resistance due to the increase in plasma and blood viscosity was compensated for by the already anemic state of the animals and improvement of microvascular function.

Table 4.

Exchange transfusion comparison

							Exchange Transfusion
		Solution Properties					Plasma properties			Hemodynamics and oxygenation
	References	Hb, g/dl	Viscosity, cP	COP, mmHg	P50, mmHg	n	Hb, g/dl	Viscosity, cP	COP, mmHg	MAP (relative to baseline)	FCD (relative to baseline)	Flow (relative to baseline)	Tissue Po2
Before hemodilution	4, 6									1.00	1.00	1.00	23.3
Plasma expander
6% Dextran (70 kDa)	6, 36		2.8	50				1.4	17	0.65	0.38	0.80	1.2
6% Dextran (500 kDa)	6, 36		6.3	38				2.1	16	0.81	0.72	1.26	2.4
10% Human albumin	4		1.1	42				1.3	20	0.78	0.71	0.64	1.7
Hb-based O2 carriers
Oxyglobin	38	13.0	1.8	38	48.0	1.2	3.7	1.4	20	0.93	0.37	0.75	2.2
PEGylated-Hb	38	4.0	2.2	49	5.4	1.2	1.1	1.4	23	0.87	0.68	0.62	1.8
Hb vesicles	4	8.6	2.9	20	29.0	1.4	2.1	1.0	18	0.82	0.74	1.52	8.6
T-PolybHb		10.0	11.4	1	40.0	0.9	2.9	2.1	17	0.86	0.76	1.09	7.4
R-PolybHb		10.0	7.8	7	1.0	0.7	2.8	1.7	17	0.91	0.54	0.91	4.5

Shear rate was 160 s⁻¹ at 37°C for blood viscosity; COP was measured at 27°C. MAP, mean arterial blood pressure; FCD, functional capillary density. Oxyglobin was from Biopure (Cambridge, MA).

GRANTS

This work was supported by Bioengineering Research Partnership (National Heart, Lung, and Blood Institute) Grant R24-HL-64395, Program Project P01-HL-071064, and Grants R01-HL-078840, R01-HL-62354, R01-HL-62318, R01-HL-76182, R01-HL-078840, and R01-DK070862.

DISCLOSURES

No conflicts of interest are declared by the author(s).