QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) All Versions of this Article: dmd.107.019174v1 36/5/937    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Chow, E. C. Y. Articles by Pang, K. S. Search for Related Content PubMed PubMed Citation Articles by Chow, E. C. Y. Articles by Pang, K. S.

Role of Haptoglobin on the Uptake of Native and β-Chain [Trimesoyl-(Lys82)β-(Lys82)β] Cross-Linked Human Hemoglobins in Isolated Perfused Rat Livers

Departments of Pharmaceutical Sciences (E.C.Y.C., L.L., K.S.P.) and Chemistry (N.S., R.K.), University of Toronto, Toronto, Ontario, Canada

(Received October 9, 2007; Accepted February 13, 2008)

	Abstract

Top Abstract Materials and Methods Results Discussion Appendix References

 
The role of haptoglobin in liver cell entry of acellular native hemoglobin, and cross-linked human hemoglobin, a potentially useful oxygen-carrier alternative in transfusion medicine, was examined in the recirculating, perfused rat liver preparation. Doses of tritiated native human or β-chain [trimesoyl-(Lys82)β-(Lys82)β] cross-linked human hemoglobin were preincubated with haptoglobin-containing rat plasma or Krebs Henseleit bicarbonate buffer for 30 min and used for perfusion. Concentrations (dpm/ml) in reservoir, before and after separation of the hemoglobins and metabolites by gel filtration fast protein liquid chromatography column chromatography, were similar, showing mostly the presence of intact hemoglobin. Each hemoglobin species underwent a rapid distribution phase, followed by a protracted elimination phase. The radioactivity in bile at 3 h consisted of low molecular weight metabolites, and cumulative excretion was slightly higher when rat plasma was present: for native hemoglobin, 7.1 ± 1.6% versus 9.2 ± 2.1% dose; for cross-linked hemoglobin, 5.0 ± 1.7% versus 7.2 ± 0.8% dose. Data fit to a two-compartment model and physiologically based model revealed a significantly faster influx clearance (CL_influx) over the metabolic intrinsic clearance (CL_{int, met}). The ratios of CL_influx/CL_{int, met} were 125 and 535 for native hemoglobin in the absence and presence of rat haptoglobin, respectively, according to compartmental analyses; the ratios were 25 and 53, respectively, according to physiological modeling. The corresponding ratios for cross-linked hemoglobin in the absence and presence of rat haptoglobin were 55 and 81, respectively, and 24 and 70 for compartmental and physiological modeling. Although haptoglobin enhanced the hepatic internalization of the hemoglobins, the impact on the net clearance was lessened since degradation was the rate-limiting step.

In cases of transfusion of acellular hemoglobin or severe hemolysis, high levels of hemoglobin are present and will easily saturate the haptoglobin present in plasma (Keene and Jandl, 1965; Hershko et al., 1972). Excess hemoglobin in plasma exists in the form of its constituent β-dimers, which are removed primarily by renal filtration, a mode of elimination that results in iron deposits in the renal tubules, leading to renal damage (Chang, 1999; Gburek et al., 2002; Roach et al., 2004). Due to the large molecular size, the bound hemoglobin-haptoglobin complex is protected from renal excretion (Keene and Jandl, 1965). The complex is mainly eliminated through the liver (Goldfischer et al., 1970; Bissell et al., 1972; Hershko et al., 1972; Ship et al., 2005) by binding to receptors on the surface of hepatocytes for endocytosis (Bissell et al., 1972; Kino et al., 1980, 1987; Zuwala-Jagiello and Osada, 1998). Hepatic elimination is the primary process for the clearance of acellular hemoglobin, and hepatocyte uptake may occur by a haptoglobin-independent pathway (Hershko et al., 1972; Weinstein and Segal, 1984). The internalized hemoglobin is then metabolized in lysosomes (Graversen et al., 2002), generating hemoglobin metabolites such as bilirubin, globin chains, and iron bound to transferrin (Goldfischer et al., 1970; Hershko et al., 1972; Higa et al., 1981). Bilirubin is then further glucuronidated by UDP-glucuronosyl-transferase 1A1 (UGT1A1) and excreted into bile, whereas the iron transferrin and globin chains are returned back to the plasma (Clarke et al., 1997; Huang et al., 2004).

Aspects of the processes of blood transfusion have stimulated research toward the development of red cell substitutes (Creteur and Vincent, 2003). Acellular hemoglobin, a tetrameric protein, rapidly dissociates into its constituent β-dimers, which are subsequently cleared by filtration in the kidney (Bunn et al., 1969). It is therefore associated with short circulation times (1–4 h) and kidney toxicity, limiting its usefulness (Roach et al., 2004). Chemically cross-linked hemoglobins that prevent the dissociation to dimers tend to exhibit prolonged circulation times (Palaparthy et al., 2001). Formation of the hemoglobin-haptoglobin complex results in liver clearance (Bissell et al., 1972; Kino et al., 1980; Zuwala-Jagiello and Osada, 1998), fueling the postulate that haptoglobin binding of synthetically prepared hemoglobins may result in increased hepatic internalization and faster clearance. Indeed, Ship et al. (2005) found a correlation between binding and clearance. The lower in vitro binding of β-chain [trimesoyl-(Lys82)β-(Lys82)β] cross-linked human hemoglobin to rat haptoglobin (30% bound) resulted in a lower distribution and reduced plasma clearance and biliary excretion when compared with native human hemoglobin, which is almost completely bound (100%) in rats in vivo. Both native human hemoglobin and β-chain [trimesoyl-(Lys82)β-(Lys82)β] cross-linked human hemoglobin exhibited short half-lives (23–37 min) (Ship et al., 2005).

The contribution of the liver, in the absence of other eliminating organs such as the kidneys, to the removal of hemoglobin is unknown. The hepatic handling of hemoglobin versus the [trimesoyl-(Lys82)β-(Lys82)β] cross-linked human hemoglobin has not been directly compared. In the present recirculating liver perfusion study, we tested the hypothesis that haptoglobin binding to hemoglobin plays a role in enhancing hepatic internalization and clearance of native human hemoglobin and β-chain [trimesoyl-(Lys82)β-(Lys82)β] cross-linked human hemoglobin. Trace amounts of the radiolabeled hemoglobins were administrated in buffer in the presence or absence of rat plasma haptoglobin for hemoglobin binding in liver perfusion studies. This condition ensured that haptoglobin, when present, was in excess in relation to hemoglobin to ensure maximum binding (Ship et al., 2005).

	Materials and Methods

Top Abstract Materials and Methods Results Discussion Appendix References

 
Materials. Radiolabeled [1-³H]acetic anhydride (50 mCi/mmol) was purchased from American Radiolabeled Chemicals (St. Louis, MO). Bovine serum albumin in Tyrode's solution was supplied by Sigma-Aldrich (Mississauga, ON, Canada), and dextrose (50%) was obtained from Abbott Laboratories (Montreal, QC, Canada). Dextran T-40 was purchased from Amersham Biosciences (Québec, QC, Canada). Human hemoglobin was purified from pooled red cells provided by Hemosol BioPharma Inc. (Mississauga, ON, Canada) and stored liganded to carbon monoxide. The oxygen binding curves and metHb content were determined for each batch, and the hemoglobin was found to be active and cooperative. All other reagents were of the highest available grade.

Synthesis of Radiolabel Human Hemoglobin and Human Hemoglobin Cross-Linked with Trimesoyl Tris(3,5-Dibromosalicylate). β-Chain [trimesoyl-(Lys82)β-(Lys82)β] cross-linked human hemoglobin was prepared according to Kluger et al. (1992) and described previously (Ship et al., 2005). Carbon-monoxide-liganded hemoglobin (2 ml, 4 mM) was exposed to bright white light and bubbled with oxygen (BOC Gases, Mississauga, ON, Canada) for 2 h at 0°C to exchange the carbon monoxide with oxygen. After gel chromatography, hemoglobin was deoxygenated with nitrogen at 37°C for 2 h before the cross-linking at the two β-Lys82 residues with a 2-fold excess of trimesoyl tris(3',5'-dibromosalicylate) (16 µmol or 16 mg) for 16 h at 37°C (Kluger et al., 1992). The cross-linking reaction was stopped by bubbling oxygen to the reaction mixture. The tetrameric hemoglobin product was separated from unreacted trimesoyl tris(3',5'-dibromosalicylate) by size-exclusion chromatography and stored at 4°C under carbon monoxide (Ship et al., 2005). The purity of the cross-linked hemoglobin was >99% according to high-pressure liquid chromatography and previously described methodology (Kluger et al., 1992). Absence of monomeric β-chains was demonstrated by matrix-assisted laser desorption/ionization time-of-flight mass spectroscopy using an Applied Biosystems (Foster City, CA) Voyager-DE STR equipped with a 337-nm nitrogen laser. The concentration of hemoglobin was determined by the absorbance of cyanmethemoglobin in the sample prepared by reaction with potassium ferricyanide. Procedures similar to those of Ship et al. (2005) were followed for the radiolabeling of hemoglobin and cross-linked human hemoglobin by acetylation with [1-³H]acetic anhydride.

Recirculating Liver Perfusion Studies. Male Sprague-Dawley rats (320 ± 18 g) were obtained from Charles River Laboratories (St. Constant, QC, Canada). The protocols were approved by the University of Toronto Animal Care Committee. Rats were given water and food ad libitum and kept under a 12:12-h dark-light cycle. [³H]Radiolabeled native or cross-linked human hemoglobin (20–100 nmol) was preincubated with either Krebs Henseleit bicarbonate buffer (KHB) or rat plasma in a ratio of 1:25 (mol/mol) for hemoglobin to haptoglobin for 30 min at 37°C. The haptoglobin concentration in rat plasma was determined in vitro by titration with rat hemoglobin (Bunn, 1967), and this was found to be 1 mg/ml.

Liver perfusion. The surgical procedure and TWO/TEN Perfuser (perfusion apparatus; MX International, Aurora, CO) has been described in detail previously; perfusate consisted of KHB that was oxygenated with 95% O₂ and 5% CO₂ (BOC Gases) at 1 l/min (Tirona et al., 1999). Erythrocyte-free perfusate was prepared by mixing the hemoglobin doses with 150 ml of 1% bovine serum albumin and 3% dextran T-40 in KHB. After induction of anesthesia by intraperitoneal injection of a mixture of ketamine (90 mg/kg) and xylazine (10 mg/kg), the abdomen was opened with a V-section. The hepatic artery was ligated, and the portal vein and the bile duct were cannulated by a 14-gauge needle double catheter and PE50 tubing, respectively. The rat liver was removed from the carcass and placed onto a home-built glass perfusion tray. The hepatic vein was not cannulated, and the venous perfusate outflow was allowed to drain directly into the reservoir. KHB perfusate recirculated the liver at a flow rate of 35–40 ml/min at 37°C for 20 min for equilibration. Then, KHB perfusate (usually 150 ml), containing the designated dose from a second reservoir, was used for continued recirculation for 3 h. The reservoir perfusate (1 ml) was sampled at 0, 3.5, 10, 20, 30, 45, 60, 75, 90, 120, and 180 min, whereas bile was collected at 30-min intervals. At the end of the experiment (180 min), the liver was flushed with ice-cold, blank KHB to remove any residual blood. The liver was weighed, added KHB (1:2.5 v/v), homogenized by the Ultra-Turrax T25 homogenizer (Janke and Kunkel, Staufen, Germany), and stored at -20°C until further analysis. Reservoir perfusate data were expressed in terms of concentrations and amounts (concentration x volume of reservoir), normalized to dose; for bile, the cumulative amounts in bile, normalized to dose, were presented.

Sample analyses. The disappearance of the radiolabeled hemoglobin species in buffer perfusate and the appearance of hemoglobin metabolites in bile were studied by column chromatography of the samples. Perfusate and bile samples were applied to a Superdex G-75 (Amersham Biosciences) size-exclusion column, eluting with 0.1 M phosphate buffer, pH 7.4, at 0.4 ml/min according to Ship et al. (2005). Eluted fractions, collected at 1.5-min intervals, were added to 5 ml of Ready Protein (Beckman Coulter, Fullerton, CA) and subjected to liquid scintillation spectrometry (model 5801; Beckman Coulter). The total radioactivity of each sample was also ascertained by liquid scintillation counting. The liver was homogenized (1:3 v/v with KHB) and centrifuged at 9,000g for 5 min, and an aliquot of the supernatant was subjected to liquid scintillation spectrometry, as previously described (Ship et al., 2005). A calibration curve was constructed by addition of known dpms of radiolabeled hemoglobin to blank liver homogenate tissue (1:3 v/v with KHB); the standards were processed in an identical manner as that for the liver samples. The calibration curve was used to relate to the total dpm recovered in liver samples.

Modeling. Compartmental model. Perfusate concentrations (dpm/ml) of native and cross-linked human hemoglobins were normalized to the dose.

(1)

The above equation adequately describes the disposition profile of intact hemoglobin. Since elimination is assumed to occur from the peripheral compartment only, the micro rate constants, k₁₂, k₂₁, and k₂₀ in eqs. 2, 3, 4, may be estimated from the coefficients and hybrid constants, and β (Gibaldi and Perrier, 1982).

(2)

(3)

(4)

The "central" volume, V₁, is estimated as

(5)

whereas the liver volume, V₂, is estimated as

(6)

with the assumption that the transfer clearances, V₁k₁₂ and V₂k₂₁, are equal. The estimated volumes of distribution are then used to estimate influx (CL_influx) and metabolic intrinsic (CL_{int, met}) clearances.

(7)

(8)

Physiologically based pharmacokinetic model. A simple, physiologically based pharmacokinetic (PBPK) model for a recirculating liver perfusion system is shown in Fig. 1B. The model comprises three compartments: reservoir (R), extracellular plasma (EC), and liver tissue (L). In this model, the flow rate (Q) and volumes (V_R, V_EC, and V_L) are used. The transport clearances of the hemoglobin species across the sinusoidal membrane, from extracellular plasma to tissue and from tissue to extracellular plasma, are characterized by the influx (CL_influx) and efflux (CL_efflux) clearances, which represent the summed transport of both haptoglobin-dependent and haptoglobin-independent pathways. Hemoglobin is metabolized by enzymes with a metabolic intrinsic clearance, CL_{int, met}, within the liver tissue.

View larger version (8K): [in this window] [in a new window]   FIG. 1. Modeling approaches. A, a two-compartment model with elimination from the peripheral compartment (elimination rate constant k20) was used. The transfer rate constants between the compartments are denoted by k12 and k21, respectively. B, a physiologically based pharmacokinetic model consisting of the reservoir compartment (of volume VR and concentration CR) and extracellular compartment (of volume VEC and concentration CEC), which are interconnected by flow (Q), was used. Entry and exit of hemoglobin into and out of the liver compartment (of volume VL and concentration CL) and the EC space occur with influx (CLinflux) and efflux (CLefflux) clearances, respectively. Hemoglobin is metabolized by the liver with metabolic intrinsic clearance (CLint, met).

Compartment model. Equation 1 was used to yield A, B, , and β for the two-compartment model (Fig. 1B). Fitting yielded the coefficients A and B and the hybrid constants and β according to eq. 1 (Gibaldi and Perrier, 1982). V₁ and V₂, the volumes of distribution in reservoir and liver compartments, respectively, and k₁₂, k₂₁, k₂₀, CL_influx, and CL_{int, met} were estimated from eqs. 2, 3, 4, 5, 6, 7, 8.

Physiological model. Mass-balanced rate equations in the Appendix were used to yield CL_influx, CL_efflux, and CL_{int, met} with the physiological model (Fig. 1B). The CL_influx and CL_{int, met} derived from the compartment model were used as initial estimates. The extracellular plasma volume, sum of sinusoidal blood volume and sucrose Disse space, and value of the cellular water space (approximately 60% of liver weight) were obtained from Pang et al. (1988). First, data from each study was fitted individually, but this yielded very poor estimates due to the limited data and the large number of parameters that needed to be estimated. Hence, the aggregated data (all data within the same set of experiments) were used for each fit.

Data Analyses. Data were presented as mean ± standard deviation. The two-tail Student's t test was used to determine the significance among each hemoglobin species, with or without haptoglobin. The Wilcoxon-Mann-Whitney test and analysis of variance were used to test differences of the means among data sets, native versus cross-linked hemoglobin when haptoglobin was present and absent. A P value of 0.05 was set as the level of statistical significance.

	Results

Top Abstract Materials and Methods Results Discussion Appendix References

 
Separation of Hemoglobins from Metabolites. Size-exclusion chromatography adequately separated the HBs (bound or unbound to Hp) that eluted between 20 and 40 min from the metabolites (from 40 min onward) (Figs. 2 and 3). The nature of the radioactivity differed between perfusate and bile samples. Most of the dpms existed as intact HBs in reservoir perfusate, and there were only minor amounts of HB metabolites present. The concentration of native human HB in reservoir perfusate (retention time, 32 min) decreased over time (Fig. 2A). The same was observed for native human HB in the presence of rat plasma Hp (data not shown). By contrast, bile samples obtained from the hemoglobin studies, with and without plasma Hp, showed absence of intact, human HB or its bound complex (Fig. 2B). The majority of radioactivity in bile consisted of metabolite species of molecular weights that were much lower than those of HB and its Hp-bound complex. Likewise, cross-linked human HB was only found in reservoir perfusate, and metabolite species were absent (Fig. 3A); the same was observed in cross-linked HB in the presence of rat plasma (data not shown). Again, the radioactivities in bile were mostly metabolites of cross-linked human HB (Fig. 3B); the same was observed in cross-linked HB in the presence of rat Hp (data not shown). Moreover, acellular native human and cross-linked human HB and their bound complexes exhibited virtually identical retention time between 20 and 40 min in the chromatographic procedure.

View larger version (14K): [in this window] [in a new window]   FIG. 2. Size-exclusion chromatography of acellular, 3H-labeled native human hemoglobin in reservoir perfusate (A) and 3H-labeled metabolites in bile (B) when KHB was preincubated with the dose. For bile, only the data for 0, 30, 60, 120, and 180 min were shown. Similar results were obtained when rat plasma was preincubated with the dose of 3H-labeled native human hemoglobin.

View larger version (12K): [in this window] [in a new window]   FIG. 3. Size-exclusion chromatography of acellular, 3H-labeled cross-linked human hemoglobin in reservoir perfusate (A) and 3H-labeled metabolites in bile (B) when KHB was preincubated with the dose. For bile, only the data for 0, 30, 60, 120, and 180 min were shown. Similar results were obtained when rat plasma was preincubated with the dose of 3H-labeled cross-linked human hemoglobin.

Accumulation of Hemoglobin Metabolites into Bile and Liver. Radioactivity was observed in bile, usually after a short delay of approximately 10 min. Radioactivity, representing mostly metabolites of both native and cross-linked hemoglobins, was found excreted into bile (Fig. 4). Data resulting from dosing of the native human HB with added rat plasma Hp were associated with higher amounts of radioactivity in bile than when the dose was incubated with KHB (9.2 ± 2.1% versus 7.1 ± 1.6% dose in 3 h), although the difference was not significant (P > 0.05). The radioactivity accumulated in bile (7.2 ± 0.8% dose in 3 h) for cross-linked human HB in Hp-KHB was significantly higher than that incubated with KHB (5.0 ± 1.7%; P < 0.05) (Table 1). The amount of radioactivity in bile at 3 h found for the free-native human HB dose was higher than that for cross-linked human HB (7.1 ± 1.6% versus 5.0 ± 1.7%; P = 0.04, Wilcoxon-Mann-Whitney test). Moreover, a slightly longer delay in the biliary excretion/appearance of radioactivity in the first 30 min was observed for the studies of cross-linked human HB in KHB. The amounts of native human HB remaining in livers at 3 h, with or without rat Hp, were similar (5.8 to 6.1% dose; Table 1), whereas slightly higher values were observed for cross-linked human HB (7.6 to 8.3% dose; Table 1). The recovery of total radioactivity (summed amounts in perfusate, liver, and bile) was generally good and accounted for 87 to 99% dose (Table 1).

View larger version (12K): [in this window] [in a new window]   FIG. 4. Biliary excretion. Accumulation of total radioactivities of hemoglobin metabolites in bile for doses of [3H]native human hemoglobin (A) and [3H]cross-linked human hemoglobin (B) preincubated with KHB or rat haptoglobin. The data shown were for 3 h of rat liver perfusion. *, P < 0.05; data in KHB dose versus data in rat plasma dose.

View this table: [in this window] [in a new window]   TABLE 1 Model-independent data associated with (3H-acetylated) native human and cross-linked human hemoglobin doses in the presence or absence of rat plasma (haptoglobin) in rat liver perfusion studies Data are given as mean ± S.D. AUC(0) was calculated by the trapezoidal rule method: AUC(0180 min) + AUC, extrapolated to infinity, or (C(180 min)/β), where β is the terminal decay constant. CL was calculated as dose/trapezoidal AUC(0).

Disappearance of the Hemoglobins from Perfusate. On average, unchanged native HB represented >95% of the total radioactivity at each time point in studies in which KHB buffer or rat plasma was added to the hemoglobin doses. Thus, the total radioactivity in perfusate was used for estimation of the pharmacokinetic parameters. The decay patterns of unchanged HB in perfusate, normalized to the dose, were similar for the unbound form of native, human HB and that which had complexed with plasma Hp in 3-h perfusion studies (Fig. 5A). In both of the studies on free HB and Hp-bound HB, the distribution phase of the native HB was rapid and completed within 10 min; this was followed by a protracted decay. The patterns of disappearance of radioactivity of free and the Hp-bound cross-linked human HB in liver perfusion studies were similar; the unchanged cross-linked HB (free and Hp-bound) represented >95% of the total radioactivity at each time point in studies in which buffer or rat plasma was added to the dose (Fig. 5B). The areas under the curves (AUCs), estimated by trapezoidal rule plus extrapolated area, were not significantly different, although the trend of a lower AUC being associated with rat plasma was observed for both human HB and cross-linked human HB studies (Table 1). Hepatic clearance (CL), derived from dose/AUC, showed that the clearance of cross-linked human HB in the presence of Hp was less than that of human HB in the presence of Hp (P < 0.05, Wilcoxon-Mann-Whitney test; Table 1).

View larger version (11K): [in this window] [in a new window]   FIG. 5. Compartmental analyses. Fitting of the perfusate data of intact hemoglobin obtained upon recirculation of (A) [3H]native human hemoglobin and (B) [3H]cross-linked human hemoglobin doses in KHB and rat plasma to the two-compartment model. The lines represent the best fit to the data.

Compartmental Modeling. The biexponential equation (eq. 1) with a weighting scheme of unity adequately described the data of acellular native HB (Fig. 5A) and cross-linked HB (Fig. 5B) in the presence and absence of Hp. The fit described a rapid initial decay during the first 10 min and a protracted elimination phase (Fig. 5). There was no significant difference in the fitted coefficients, A and B, for native or cross-linked HB, with and without added rat Hp (Table 2). Although , the hybrid constant for the distribution phase, was smaller for the native hemoglobin compared with that of cross-linked hemoglobin, the difference was not significant because of the high variability. There was also a difference for β, the hybrid constant for the elimination phase. The terminal half-life for native HB tended to be shorter in comparison with that for cross-linked HB, and there was a tendency toward a shorter t_1/2 when Hp was present (Table 2). Again, these differences were not significant because of the high variability. Expectedly, the shorter half-lives for native human HB were associated with lower AUCs and higher total body clearances (CL) compared with those for cross-linked HB. The AUCs, when estimated as (A/ + B/β), were similar to those from the trapezoidal rule (Table 1). The AUCs were smaller and the CLs were higher for each hemoglobin species when rat haptoglobin was present, but, again, these differences were not significant (Table 2). The livers cleared native human HB better than the cross-linked human HB, and the values of clearance tended to be higher with Hp present. However, these CL values were not significantly different.

View this table: [in this window] [in a new window]   TABLE 2 Pharmacokinetic parameters of total native human (3H-acetylated) and cross-linked (3H-acetylated) human hemoglobin, based on compartmental analyses, in the presence or absence of rat plasma (haptoglobin) AUC(0) was calculated as A/ + B/β. CL was calculated as dose/(A/ + B/β). V1 was estimated from eq. 5, and V2 was estimated from eq. 6.

The intercompartmental rate constants (k₁₂ and k₂₁) and elimination rate constant (k₂₀) were estimated from A, B, , and β with eqs. 2, 3, 4 and V₁ and V₂ from eqs. 5 and 6 for elimination from the peripheral compartment only. These rate constants were converted to the influx clearance (CL_influx or k₁₂V₁) and the metabolic intrinsic clearance (CL_{int, met} or k₂₀V₂). The influx clearances were slightly higher in the presence of Hp; values of CL_influx were 60 and 56% higher for native hemoglobin and cross-linked HB, although the difference was not significant (Table 2). By contrast, the CL_influx was significantly higher than the CL_{int, met} (P < 0.05, Wilcoxon-Mann-Whitney), and it was 125- and 535-fold of CL_{int, met} for native HB in the absence and presence of Hp, respectively. The CL_influx was 55-fold and 81-fold for cross-linked HB in absence and presence of Hp, respectively (Table 2). From this comparison, the much higher value for CL_influx over CL_{int, met} suggests that uptake is much faster than elimination. The CL_efflux was not estimated from compartmental modeling since the inherent assumption of the bidirectional clearances being equal was not validated (Table 2).

Parameters for PBPK Model. As shown in Fig. 6, good fits were observed with the PBPK model (Fig. 1B). The fitted parameters are summarized in Table 3. A weighting factor of unity was shown to be optimal. Upon comparison, the value assigned to V₂ was noted to be similar to that estimated with the compartmental model (Table 2). When Hp was present, CL_influx was 4.7-fold higher for native HB and 2.4-fold higher for cross-linked HB. The influx clearance (CL_influx) was 2- to 3-fold greater than the efflux clearance (CL_efflux) (Table 3). No trend, however, was found among the data for CL_{int, met}. The ratios of CL_influx/CL_{int, met} were 25 and 53 for native hemoglobin, in the absence and presence of Hp, respectively, and 24 and 70 for cross-linked human HB, in the absence and presence of Hp, respectively (Table 3). These trends were similar to those projected from compartmental analyses (Table 2).

View larger version (14K): [in this window] [in a new window]   FIG. 6. Physiological modeling. Fitting of the perfusate data of intact hemoglobin obtained upon recirculation of (A) [3H]native human hemoglobin and (B) [3H]cross-linked human hemoglobin doses in KHB and rat plasma to the physiologically based pharmacokinetic model. The lines represent the best fit to the data.

	Discussion

Top Abstract Materials and Methods Results Discussion Appendix References

 
The presence of specific HB-Hp receptors on liver parenchymal cells has led to the postulate that binding of HB to Hp promotes the internalization of HB into the liver (Bissell et al., 1972; Kino et al., 1980; Zuwala-Jagiello and Osada, 1998). Ship et al. (2005) have shown that, because acellular HB was almost completely bound to Hp in excess, whereas binding of trimesoyl-(Lys82)β-(Lys82)β cross-linked human HB was considerably lower (30% bound), total body clearance of native human HB was indeed greater than that of cross-linked human HB in the rat in vivo (Table 4). The observation was explained by the higher binding affinity of native human HB toward rat Hp (Lockhart and Smith, 1975; Ship et al., 2005). Similarly, absence of the vitamin D binding protein for the binding of 25-hydroxyvitamin D₃ (25-OHD₃), an inactive metabolite of vitamin D, was found to alter the pharmacokinetics of 25-OHD₃ in mice (Safadi et al., 1999). The 25-OHD₃ complex, bound with vitamin D binding protein, interacted with the endocytic receptor, megalin, to facilitate reabsorption by endocytosis in the kidney (Nykjaer et al., 1999). These examples of protein-bound species promoting clearance contrasted with those ordinarily observed for drugs. Normally, drug clearance is decreased with increased protein binding (Smallwood et al., 1988), unless saturation of protein binding occurred prior to saturation of the elimination process (Chiba and Pang, 1993). Because of the inhibitory nature of protein binding on clearance, bilirubin (Oie and Levy, 1975), tolbutamide (Schary and Rowland, 1983), and diclofenac (Evans et al., 1993) exhibited decreased hepatic clearances when protein binding was increased.

View this table: [in this window] [in a new window]   TABLE 4 Comparison of total hepatic clearance, volume, and t1/2 from the liver perfusion studies and those in vivo (from Ship et al., 2005)

The scenario of enhanced entry and therefore increased clearance for Hp-bound HB was expected to be evident in the present rat liver perfusion study, since the amount of native or cross-linked human HB administered was below the amount of Hp present to ensure maximal binding of hemoglobin (Hershko et al., 1972). Results for this perfusion study were consistent with the hypothesis that haptoglobin binding increased hemoglobin entry. A rank order was found for CL_influx, the influx clearance that denotes entry of the hemoglobins into the liver: hemoglobin (+ haptoglobin) > free hemoglobin > cross-linked hemoglobin (+ haptoglobin) > free cross-linked hemoglobin with compartmental modeling (Table 2). A similar trend was found with the PBPK model, in which sinusoidal entry of the Hp-bound forms of native human HB and cross-linked human HB were >5- and >2-fold faster, respectively, than those when Hp was absent (Table 3). The trend was more apparent with physiological modeling for which fewer assumptions were taken.

We anticipated that binding to Hp would trigger increased hepatic entry of HB (CL_influx) in the rat liver preparation and enhance clearance. The CL_influx for Hp-bound HB is 4.6 times that of free HB, whereas that for Hp-bound cross-linked HB is 2.4 times that of free cross-linked HB (Table 3). Despite the higher CL_influx for Hp-bound hemoglobin, changes in clearance were modest (20–30%) (Table 1). A significant difference in clearance was found only between the Hp-bound forms of HB and cross-linked HB (Table 1), confirming previous results in vivo that a higher clearance existed for hemoglobin due to the higher binding to haptoglobin (Ship et al., 2005). A significant change in amounts excreted in bile at 3 h was found between cross-linked HB in the absence and presence of Hp (Table 1), but no change was found for human hemoglobin. The total radioactivity recovered for native HB was higher than that for cross-linked HB, especially when Hp was present (Fig. 4; Table 1). The slightly improved biliary excretion of radioactivity, mostly as metabolites, failed to parallel the 2- to 5-fold change in CL_influx in the presence of Hp (Tables 1 and 3). Upon closer examination, we noted that CL_influx was much greater than the metabolic intrinsic clearance (CL_{int, met}) in both physiological and compartmental modeling (Tables 2 and 3), suggesting that CL_{int, met} is the rate-limiting step in clearance. The trends for CL_influx and CL_{int, met} were similar for both compartmental and physiological modeling (Tables 2 and 3), suggesting that, although the Hp played an important role in the entry of HB into liver, metabolism of hemoglobin in the liver remains the rate-limiting step. Even without Hp present, the CL_influx of native and cross-linked hemoglobin is still at least 23-fold greater than that of the CL_{int, met}. Thus, Hp, expected to enhance the entry of native and cross-linked human hemoglobin by increasing CL_influx, exerts only a blunted impact, since CL of HB is limited by metabolism and not influx.

Similarities were found between the present data and observations in vivo (Table 4). In both the rat in vivo and perfused liver preparation, only unchanged HB was found in the blood/perfusate, whereas only small molecular weight metabolites were found in liver/bile (Figs. 2 and 3), as confirmed by others (Takami, 1993). Differences were also found. Values of hepatic CL (0.0109 and 0.0061 ml · min^-1 · g^-1 for the Hp-bound HB or cross-linked HB) from the perfusion study (Table 1) were only 30–50% of the total CL found in vivo (Ship et al., 2005). Clearing organs other than the liver are probably involved in the removal of the hemoglobins in vivo. The hepatic clearance is likely to be lower in the KHB-perfused liver, since other tissues that contain the macrophages are absent. The chosen volume for perfusion (150 ml) was significantly larger than those expected for the liver in vivo (12 and 6 ml for native human HB and cross-linked human HB, respectively); the larger volume in the perfusion system constituted artificially higher half-lives (Table 4). In contrast to the monoexponential decay (instantaneous distribution) observed following the intravenous injection of native human and cross-linked human hemoglobins in the rat in vivo (Ship et al., 2005), a sharp distribution phase followed by a prolonged, elimination phase was observed in the rat liver preparations (Fig. 5). The biexponential characteristics suggest that there are at least two compartments, a central and a peripheral (liver) compartment (Fig. 1). Inasmuch as observations for the isolated preparation reflected events occurring strictly in the liver, there appeared to be increased sensitivity in the system to reveal the distribution phase in liver, a process which was likely masked or obscured by the presence of multiple organs in vivo.

In summary, native and cross-linked human hemoglobins both showed similar biexponential decay profiles in the perfused rat liver preparation. Perfusate contained mostly unchanged hemoglobin, and bile, low molecular weight metabolites; these were also observed in vivo. However, a distribution phase was further revealed in liver perfusion studies. A substantial influx was found for both native and cross-linked human hemoglobins in the absence of Hp (Tables 2 and 3). Upon fitting to the open, two-compartment model and to the PBPK model, the influx clearance, CL_influx, was found to be much faster in the presence of Hp and greatly exceeded the metabolic intrinsic clearance, CL_{int, met}. For this reason, changes in CL_influx with Hp failed to directly impact the clearance of the hemoglobins as had been previously envisioned, despite the observed trend of enhanced hemoglobin entry caused by binding to rat haptoglobin. The hypothesis that haptoglobin binding to hemoglobin enhanced entry was correct. However, the improved entry failed to strongly influence the clearance, since degradation of the hemoglobin was rate-determining. Hence, the kinetic analyses resulting from this liver perfusion study lend direct insight into entry of the free versus the bound forms of native and cross-linked hemoglobins and the rate-limiting step in hepatic removal of the hemoglobins. This process for the native human and cross-linked human hemoglobins is metabolism within the rat liver and not entry.

	Appendix

Top Abstract Materials and Methods Results Discussion Appendix References

 
Mass Balance Rate Equations for Liver as the Only Eliminating Organ. The following are first-order rate equations that describe the rate of change of hemoglobin species in the reservoir, the extracellular plasma space, and the liver.

For rate of change in reservoir:

(A1)

For rate of change in extracellular plasma:

(A2)

For rate of change in liver:

(A3)

	Footnotes

 
This work was supported by grants from the National Sciences and Engineering Research Council of Canada and a Strategic Project (234768-00) and a Collaborative Health Research Program (237960-00) Grant to K.S.P. and R.H.K.

Article, publication date, and citation information can be found at http://dmd.aspetjournals.org.

doi:10.1124/dmd.107.019174.

ABBREVIATIONS: , hybrid constant associated with the distribution phase in compartmental modeling; β, hybrid constant associated with the elimination phase in compartmental modeling; C_R, C_EC, and C_L, concentrations of hemoglobin in reservoir, extracellular plasma, and liver compartments, respectively; CL, hepatic clearance; CL_influx and CL_efflux, influx and efflux clearances, respectively, at the sinusoidal membrane; CL_{int, met}, hepatic metabolic intrinsic clearance; k₁₂, micro rate constant denoting entry from central compartment to peripheral compartment; k₂₁, micro rate constant denoting entry from peripheral compartment to central compartment; k₂₀, micro rate constant denoting elimination from peripheral compartment; Hp, haptoglobin; HB, hemoglobin; KHB, Krebs Henseleit bicarbonate buffer; Q, total hepatic flow rate; V_R, V_EC, and V_L, volumes of reservoir, extracellular plasma, and liver, respectively; PBPK, physiologically based pharmacokinetic; AUC, area under the curve.

¹ Current affiliation: Genentech Inc., South San Francisco, CA.

Address correspondence to: Dr. K. Sandy Pang, Leslie Dan Faculty of Pharmacy, University of Toronto, 144 College Street, Toronto, ON, Canada M5S 3M2. E-mail: ks.pang{at}utoronto.ca

	References

Top Abstract Materials and Methods Results Discussion Appendix References

 


Bissell DM, Hammaker L, and Schmid R (1972) Hemoglobin and erythrocyte catabolism in rat liver: the separate roles of parenchymal and sinusoidal cells. Blood 40: 812-822.[Abstract/Free Full Text]

Bunn HF (1967) Effect of sulfhydryl reagents on the binding of human hemoglobin to haptoglobin. J Lab Clin Med 70: 606-618.[Medline]

Bunn HF, Esham WT, and Bull RW (1969) Renal handling of hemoglobin. I. Glomerular filtration. J Exp Med 129: 909-924.[Abstract]

Chang TM (1999) Future prospects for artificial blood. Trends Biotechnol 17: 61-67.[CrossRef][Medline]

Chiba M and Pang KS (1993) Effect of protein binding on 4-methylumbelliferyl sulfate desulfation kinetics in perfused rat liver. J Pharmacol Exp Ther 266: 492-499.[Abstract/Free Full Text]

Clarke DJ, Moghrabi N, Monaghan G, Cassidy A, Boxer M, Hume R, and Burchell B (1997) Genetic defects of the UDP-glucuronosyltransferase-1 (UGT1) gene that cause familial non-haemolytic unconjugated hyperbilirubinaemias. Clin Chim Acta 266: 63-74.[CrossRef][Medline]

Creteur J and Vincent JL (2003) Hemoglobin solutions. Crit Care Med 31: S698-S707.[CrossRef][Medline]

Evans AM, Hussein Z, and Rowland M (1993) Influence of albumin on the distribution and elimination kinetics of diclofenac in the isolated perfused rat liver: analysis by the impulse-response technique and the dispersion model. J Pharm Sci 82: 421-428.[CrossRef][Medline]

Gburek J, Verroust PJ, Willnow TE, Fyfe JC, Nowacki W, Jacobsen C, Moestrup SK, and Christensen EI (2002) Megalin and cubilin are endocytic receptors involved in renal clearance of hemoglobin. J Am Soc Nephrol 13: 423-430.[Abstract/Free Full Text]

Gibaldi M and Perrier D (1982) Pharmacokinetics, 2nd ed, Marcel Dekker, New York.

Goldfischer S, Novikoff AB, Albala A, and Biempica L (1970) Hemoglobin uptake by rat hepatocytes and its breakdown within lysosomes. J Cell Biol 44: 513-529.[Abstract/Free Full Text]

Graversen JH, Madsen M, and Moestrup SK (2002) CD163: a signal receptor scavenging haptoglobin-hemoglobin complexes from plasma. Int J Biochem Cell Biol 34: 309-314.[CrossRef][Medline]

Greer J, Liao WD, and Brown WE (1981) Haptoglobin-hemoglobin complex. Subunit interaction probed by cross-linking. J Biol Chem 256: 8771-8774.[Abstract/Free Full Text]

Hershko C, Cook JD, and Finch CA (1972) Storage iron kinetics. II. The uptake of hemoglobin iron by hepatic parenchymal cells. J Lab Clin Med 80: 624-634.[Medline]

Higa Y, Oshiro S, Kino K, Tsunoo H, and Nakajima H (1981) Catabolism of globin-haptoglobin in liver cells after intravenous administration of hemoglobin-haptoglobin to rats. J Biol Chem 256: 12322-12328.[Abstract/Free Full Text]

Huang MJ, Kua KE, Teng HC, Tang KS, Weng HW, and Huang CS (2004) Risk factors for severe hyperbilirubinemia in neonates. Pediatr Res 56: 682-689.[CrossRef][Medline]

Keene WR and Jandl JH (1965) The sites of hemoglobin catabolism. Blood 26: 705-719.[Abstract/Free Full Text]

Kino K, Mizumoto K, Watanabe J, and Tsunoo H (1987) Immunohistochemical studies on hemoglobin-haptoglobin and hemoglobin catabolism sites. J Histochem Cytochem 35: 381-386.[Abstract]

Kino K, Tsunoo H, Higa Y, Takami M, Hamaguchi H, and Nakajima H (1980) Hemoglobin-haptoglobin receptor in rat liver plasma membrane. J Biol Chem 255: 9616-9620.[Abstract/Free Full Text]

Kluger R, Song Y, Wodzinska J, Head C, Fujita TS, and Jones RT (1992) Trimesoyl tris(3,5-dibromosalicylate): specificity of reactions of a trifunctional acylating agent with hemoglobin. J Am Chem Soc 114: 9275-9279.[CrossRef]

Lim SK, Kim H, Lim SK, bin Ali A, Lim YK, Wang Y, Chong SM, Costantini F, and Baumman H (1998) Increased susceptibility in Hp knockout mice during acute hemolysis. Blood 92: 1870-1877.[Abstract/Free Full Text]

Lockhart WL and Smith DB (1975) Cross-linking of hemoglobin, haptoglobin, and hemoglobin-haptoglobin complex with bifunctional imidoesters. Can J Biochem 53: 861-867.[Medline]

McCormick DJ and Atassi MZ (1990) Hemoglobin binding with haptoglobin: delineation of the haptoglobin binding site on the alpha-chain of human hemoglobin. J Protein Chem 9: 735-742.[CrossRef][Medline]

Nykjaer A, Dragun D, Walther D, Vorum H, Jacobsen C, Herz J, Melsen F, Christensen EI, and Willnow TE (1999) An endocytic pathway essential for renal uptake and activation of the steroid 25-(OH) vitamin D₃. Cell 96: 507-515.[CrossRef][Medline]

Øie S and Levy G (1975) Effect of plasma protein binding on elimination of bilirubin. J Pharm Sci 64: 1433.[CrossRef][Medline]

Palaparthy R, Kastrissios H, and Gulati A (2001) Pharmacokinetics of diaspirin cross-linked haemoglobin in a rat model of hepatic cirrhosis. J Pharm Pharmacol 53: 179-185.[CrossRef][Medline]

Pang KS, Lee WF, Cherry WF, Yuen V, Accaputo J, Fayz S, Schwab AJ, Goresky CA (1988) Effects of perfusate flow rate on measured blood volume, disse space, intracellular water space, and drug extraction in the perfused rat liver preparation: characterization by the multiple indicator dilution technique. J Pharmacokinet Biopharm 16: 595-632.[CrossRef][Medline]

Polfliet MM, Fabriek BO, Daniëls WP, Dijkstra CD, and van den Berg TK (2006) The rat macrophage scavenger CD163: expression, regulation, and role in inflammatory mediator production. Immunobiology 211: 419-425.[CrossRef][Medline]

Putnam FW (1976) The trace components of plasma: an overview. Prog Clin Biol Res 5: 1-24.[Medline]

Roach TA, Macdonald VW, and Hosmane RS (2004) A novel site-directed affinity reagent for cross-linking human hemoglobin: bis[2-(4-phosphonooxyphenoxy) carbonylethyl]-phosphinic acid. J Med Chem 47: 5847-5859.[CrossRef][Medline]

Safadi FF, Thornton P, Magiera H, Hollis BW, Gentile M, Haddad JG, Liebhaber SA, Cooke NE (1999) Osteopathy and resistance to vitamin D toxicity in mice null for vitamin D binding protein. J Clin Invest 103: 239-251.[Medline]

Schary WL and Rowland M (1983) Protein binding and hepatic clearance: studies with tolbutamide, a drug of low intrinsic clearance, in the isolated perfused rat liver preparation. J Pharmacokinet Biopharm 11: 225-243.[CrossRef][Medline]

Ship NJ, Toprak A, Lai RP, Tseng E, Kluger R, and Pang KS (2005) Binding of acellular, native and cross-linked human hemoglobins to haptoglobin: enhanced distribution and clearance in the rat. Am J Physiol Gastrointest Liver Physiol 288: G1301-G1309.[Abstract/Free Full Text]

Smallwood RH, Mihaly GW, Smallwood RA, and Morgan DL (1988) Effect of a protein binding change on unbound and total plasma concentrations for drugs of intermediate hepatic extraction. J Pharmacokinet Biopharm 16: 529-542.[CrossRef][Medline]

Takami M (1993) Catabolism of heme moiety of hemoglobin.haptoglobin in rat liver cells in vivo. J Biol Chem 268: 20335-20342.[Abstract/Free Full Text]

Tirona RG, Tan E, Meier G, and Pang KS (1999) Uptake and glutathione conjugation kinetics of ethacrynic acid in rat liver: in vitro and perfusion studies. J Pharmacol Exp Ther 291: 1210-1219.[Abstract/Free Full Text]

Voet D and Voet JG (1995) Biochemistry, 2nd ed, Wiley, New York.

Weinstein MB and Segal HL (1984) Uptake of free hemoglobin by rat liver parenchymal cells. Biochem Biophys Res Commun 123: 489-496.[Medline]

Winslow RM (1992) Hemoglobin-Based Red Cell Substitutes, Johns Hopkins University Press, Baltimore.

Zuwala-Jagiello J and Osada J (1998) Internalization study using EDTA-prepared hepatocytes for receptor-mediated endocytosis of haemoglobin-haptoglobin complex. Int J Biochem Cell Biol 30: 923-931.[CrossRef][Medline]

This Article Abstract Full Text (PDF) All Versions of this Article: dmd.107.019174v1 36/5/937    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Chow, E. C. Y. Articles by Pang, K. S. Search for Related Content PubMed PubMed Citation Articles by Chow, E. C. Y. Articles by Pang, K. S.