NDT Advance Access originally published online on October 18, 2006
Nephrology Dialysis Transplantation 2007 22(1):88-95; doi:10.1093/ndt/gfl497
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Disproportionally low clearance of macromolecules from the plasma to the peritoneal cavity in a mouse model of peritoneal dialysis
1Department of Nephrology, University Hospital of Lund, Sweden and 2Department of Experimental Medical Science, BMC F12, Lund University, Sweden
Correspondence and offprint requests to: Prof. Bengt Rippe, Department of Nephrology, University Hospital of Lund, S-211 85 Lund, Sweden. Email: Bengt.Rippe{at}med.lu.se
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
Background. This study was performed to establish a model for quantitative measurements of a number of basic peritoneal transport parameters, particularly transperitoneal clearances (Cl) of macromolecules, during mouse peritoneal dialysis.
Methods. Mice were anaesthetized using 3% isofluorane inhalation anaesthesia. The right jugular vein and the left femoral artery were cannulated for infusion and sampling purposes and for registration of (mean) arterial blood pressure. Access to the peritoneal cavity occurred via a thin abdominal catheter (Ø 0.7 mm). About 2.5 ml of either 4% (n = 9) or 1.5% (n = 5) glucose containing PD-fluid were instilled intraperitoneally (i.p.). Dialysate volume was followed vs time using i.p. RISA (125I human serum albumin) as a volume marker, after correcting for RISA mass disappearance from the peritoneum, assessed separately (n = 11). Microsampling (10 µl) of plasma and dialysate was performed for determinations of glucose, haematocrit, radioactivity (RISA and 51Cr-EDTA) and Ficoll.
Results. The i.p. volume vs time curves [VD(t)] were, after scaling, similar to those observed in humans (and in rats). Clearance of RISA out of the peritoneal cavity (Clout) was 9.33 ± 0.83 µl/min and the clearance of RISA to plasma (Cl
P) and the RISA clearance to the peritoneal cavity (ClD) were 1.49 ± 0.13 and 0.084 ± 0.008 µl/min, respectively. The peritoneal transport coefficients for 51Cr-EDTA and glucose, as well as Clout and ClP, were 13–17% of those previously assessed in 300 g rats, whereas ClD was only 
2% of that in rat.
Conclusions. All peritoneal transport parameters measured, except ClD, scaled very well to the corresponding human data. The mechanisms of the disproportionally low clearance of macromolecules from the plasma to the peritoneal cavity in mice remain elusive and warrant further study.
Keywords: albumin; capillary permeability; Ficoll; lymph flow; peritoneal dialysis; pore modelling; transcapillary escape rate
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
A large number of animal models for peritoneal dialysis (PD) research have been developed over the years [1,2]. Recently mouse models have been introduced to take advantage of genetically modified animals [3,4]. Although the mouse offers several advantages, including low cost, easy breading and fast turnover, the small body size is clearly a limiting factor in in vivo studies. Since a 30 g mouse has a plasma volume of just around 1.3–1.5 ml, blood loss of
300 µl in an adult mouse may produce hypovolaemic shock. Furthermore, intravenous infusion volumes of more than 50 µl at a time should be avoided. Also, cannulation and instrumentation of these animals pose special problems due to the small organ size. The problems with the use of a mouse model may be overcome by, among other things, using micro-dissection techniques (dissection microscope) and by the application of microsampling (10–20 µl at a time). In two recently published studies, a mouse model of PD was shown to be feasible for physiological experimentation. In the first of these studies, peritoneal equilibration tests (PET) [5] were performed for glucose, sodium and urea, and dialysate to plasma concentration (D/P) curves were published as well as drained volumes at the end of the dwells [4]. In a later study assessment of intraperitoneal (i.p.) volume as a function of dwell time [VD(t)] using a macromolecular i.p. volume marker (RISA; 125I human serum albumin) was also performed [6]. However, critical quantitative measures of mass transfer parameters for small and large solutes were lacking.
In the present study, mainly due to the assessment of VD(t), we have been able to obtain quantitative data in terms of mass transfer area coefficients (MTAC or permeability-surface area products; PS) for small solutes and clearances of macromolecules, RISA and fluorescein isothiocyanate (FITC)-Ficoll, across the peritoneal membrane. Furthermore, it is critical for an accurate assessment of VD(t) to validate the macromolecule indicator dilution technique used. This was performed in this study by separately assessing RISA mass disappearance kinetics from the peritoneal fluid, including the initial rapid RISA loss, using a volume (and tracer) recovery technique [7]. In providing these basic data, the present study may form the basis for further usage of the mouse model for measurements of quantitative peritoneal transport parameters in PD.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
Experiments were performed on forty male C57 BL/6J mice (Møllegaard, Lille Stensved, Denmark) weighing on average 29.9 ± 1.42 (±SD) g and being 12–14 weeks of age. The mice had free access to food and water until the day of experiment. Animal experiments were approved by the Animal Ethical Committee at Lund University. Mice were anaesthetized using Isofluorane inhalation anaesthesia. Mice were first placed in a small plastic cage, containing 4% Isofluran (Isofluran Forene, Abbot Scandinavia AB, Solna) in room air using a Univentor 400 Anaesthesia unit (Univentor Ltd., Zejtan, Malta). They were then transferred to a heating pad (Temperature Control Unit HB 101/2, Panlab S.I., Barcelona, Spain), by which body temperature was kept constant at 37 ± 1°C, while maintaining anaesthesia using a small mask covering the nose and mouth. Then a tracheotomy was performed, and after intubation, anaesthesia was maintained and controlled using 2–3% Isofluran in room air on a small animal respirator (Mouse ventilator 28025, Ugo Basile, Comerio, Italy). Tidal volume was set at 0.30–0.35 ml and ventilation frequency at 90–100 min–1 (guided by blood gas analyses). A positive end expiratory pressure (PEEP) of
5 mmHg was used to prevent pulmonary atelectasis. The right jugular vein was cannulated for infusion (PE10 tubing) and the left femoral artery for sampling and for continuous monitoring of blood pressure on a polygraph (Model 7B; Grass Instruments Co., Quincy, MA). NaCl, usually mixed with tracer amounts of 51Cr-EDTA (Amersham Biosciences, UK) (7 µl/min), was given continuously via the right jugular vein. Access to the peritoneal cavity was established via an abdominal Neoflon catheter (NeoflonTM 0.7 mm x 19 mm, Becton-Dickinson Infusion Therapy AB, Helsingborg, Sweden), which was inserted into the lower left quadrant of the abdominal wall for dialysis fluid infusion and sampling.
Peritoneal dialysis protocol
2.4–2.5 ml PD solution containing either 4% (n = 9) or 1.5% (n = 5) of glucose, low in GDP (Gambrosol-trioTM), were instilled i.p. Dialysate volume was followed as a function of dwell time using i.p. RISA (125I human serum albumin, Isopharma Kjeller, Norway) taking into account the disappearance of RISA from the peritoneal cavity determined in separate experiments (see subsequent text). Microsampling of plasma and dialysate was performed repeatedly during 2 h, namely at 10, 20, 30, 40, 60, 90 and 120 min. Sample volume was 10 µl except at 60 and 120 min when larger sample volumes (40 µl) were taken for glucose determinations. Blood samples (10–40 µl) were taken at 0, 60 and 120 min for the dwell studies. Prior to dialysate sampling 100–200 µl of the dialysate was flushed back and forth via the abdominal catheter and the abdomen gently agitated to facilitate mixing. At the end of the dwell the dialysate was completely recovered from the peritoneal cavity using tarred syringes and gauze tissues, all weighed on a balance (Denver Instrument, XS-410, USA). Blood and dialysate samples were analysed for radioactivity in a gamma counter (Wizard 1480; LKB-Wallac, Turku, Finland). Blood glucose was measured instantly on a Glucometer DEX 2 (Bayer AB, Göteborg, Sweden). Glucose concentrations in dialysate were measured using an automatic analyser (YSI 2300D, YSI Inc., Yellow Springs, OH).
Intraperitoneal (i.p.) volume vs time curves [VD(t)], clearance of RISA out of the peritoneal cavity (Clout) and clearance of RISA to the plasma (Cl P)
The technique for measuring i.p. volume as a function of VD(t) by considering the (exponential) disappearance of RISA during the course of the dwell has been described in detail previously [7]. The i.p. volume can only be assessed if the i.p. mass of volume indicator is known as a function of time, i.e. if one can adequately correct for the disappearance of the volume maker from the dialysate during the dwell. In 11 separate mice the i.p. mass of tracer as a VD(t) was determined by collecting (and weighing) all i.p. fluid using tarred syringes and gauze tissues at either 1 min (n = 4), 20 min (n = 3), 40 min (n = 4). In addition, the drained tracer mass values from the dwell studies presented above at 120 min (n = 14) were added to the data. I.p. mass of tracer was calculated as total volume retrieved times the measured concentration of tracer [7]. The initial loss (during 0–1 min) of tracer due to adsorption (‘binding’) to the peritoneal surfaces was 2.8 ± 0.4% of the total tracer mass instilled [7], and the residual i.p. volume, assessed as the ‘immediately’ (within 1 min) drained volume minus the instilled volume, was 0.11 ± 0.013 ml (n = 4).
Cl P, corrected for spill-over from plasma to the interstitium, and Clout from the peritoneal cavity (to the peritoneal tissues) were calculated as described previously based on RISA kinetics [7]. In order to calculate Cl P, plasma volume was approximately estimated to be 0.05 ml/g of mouse body weight (BW) [obtained from the estimation of transcapillary RISA escape rate (TER) described below]. Spill-over correction [7] of ClP for RISA transport from plasma to the interstitium was made assuming a TER of albumin of 10% per h (see subsequent text).
Small solute mass transfer area coefficients (MTAC or PS)
Small solute MTAC, i.e. the PS for 51Cr-EDTA and glucose, were averaged from sequential measurements throughout the dwell, setting the sieving coefficient (
) at 0.55, as described in detail previously [8]. However, since the rate of net convection was small, values could be set arbitrarily between 0 and 1 without significant error. Whereas 51Cr-EDTA concentration in plasma could be kept constant, the plasma concentration of glucose increased during the course of the dwell for both 1.5%, and especially, for 4.0% glucose dwells. For 4% glucose dwells plasma glucose rose from physiological levels to 21.2 ± 2.9 mmoles/l at 60 min, and then fell to 18 ± 2.2 mmoles/l at 120 min.
Transcapillary escape rate of RISA (TER) and RISA clearance into the peritoneal cavity (Cl D)
In separate experiments (n = 7), we determined Cl D as well as the TER during a 1.5% glucose PD dwell in mice prepared and anaesthetized as aforementioned, by injecting 50(–70) µl of 125I-human serum albumin (t = 0) i.v. Actually, the RISA injection volume exceeded 50 µl in three of the experiments. Blood samples (10 µl at a time using Microcaps®) were drawn at 10, 20, 40 and 90 min and in three animals at 120 min. Haematocrit samples (40 µl) were taken at 60 and 90 (or 120) min to be able to convert blood concentrations into plasma concentrations. Tracer appearance in dialysis fluid was measured by dialysate sampling (10 µl at a time) at 10, 20, 40, 60 and 90 min (and in three animals at 120 min). To minimize the impact of free 125I all dialysate samples were precipitated with 10% trichloroacetic acid. After centrifugation, the supernatant (containing free label) was discarded and only the pellet measured. Assessment of Cl D was performed, as described in previous publications, from the mass transfer of tracer into the peritoneum per unit time divided by the average plasma concentration of tracer, assessed from the area under the plasma tracer concentration curve [8]. Plasma concentrations were calculated from blood samples by adjusting for haematocrit. The TER of RISA (during 60 min) from plasma to whole body interstitium was assessed using linear regression of log plasma RISA concentration vs time.
Clearance of FITC-Ficoll 70/400 from plasma to peritoneum
The transperitoneal clearance of Ficoll 70/400 (MW 70,000, Pharmacia, Uppsala, Sweden and MW 400 000, Sigma, St Louis, MO) labelled with FITC was assessed in eight mice during 1.5% glucose PD dwells (120 min) by infusing 50 µl of FITC-Ficoll 70/400 i.v. during the last 40 min of the 2 h dwell, and collecting 30–40 µl of plasma using a haematocrit micro-capillary in the midst of the infusion period (at 100 min). Labelling of FITC was made according to Ohlsson et al. [9]. The concentrations of FITC-Ficoll were determined using high performance size exclusion liquid chromatography (HPSEC) with devises from Waters (Waters 125, Milford, MA, USA) using an Ultrahydrogel-500 column (7.8 x 300 mm, Waters) and 0.05 M phosphate buffer with 0.015 M NaCl (pH 7.4) as the mobile phase [10]. The system contains a pump (Waters 1525), an absorbance detector (Waters 2487) and a fluorescence detector (Waters 2475). The samples were analysed at an excitation wavelength of 492 nm and an emission wavelength of 518 nm. The system was controlled using Breeze software 3.2 (Waters). The column was calibrated with narrow FITC labelled Ficoll standards (73, 59, 46, 38 and 30 Å), provided by Dr Torvald Andersson (Pharmacia, Sweden) [10]. Ficoll D/P ratio was calculated by dividing the dialysate concentration by the average plasma concentrations from the HPSEC, i.e. the concentrations in plasma after 20 min of Ficoll exposure, corresponding to the mean plasma concentrations over the experimental period.
Statistics and calculations
All data, except when indicated otherwise, are expressed as means ± SE. A two-pore model described previously in detail [11,12] was fitted to the FITC-Ficoll data by minimizing the function:
|
|
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
Haemodynamic parameters
The mean arterial pressure (MAP) in the middle of the dwell studies (i.e. at
60 min) was 76 ± 3 mmHg, while heart rate was 550 ± 14 beats/min (BPM), the MAP being approximately 3 mmHg higher at start of the dwells, and 2–3 mmHg lower at the end of the dwells with no significant (systematic) change in heart rate. A chart recording from a typical experiment is shown in Figure 1.
|
I.p. RISA mass and RISA indicator dilution determined i.p. volume curves [VD(t)]
The initial adsorption (‘binding’) of RISA used as a volume maker was low, 2.8 ± 0.4%, and the tracer mass kinetics followed a monoexponential function, which was used for the volume calculations (Figure 2). Figure 3 shows the VD(t)-curves (UF-curves) for 2.4–2.5 ml of instilled volume for either 1.5 or 4% glucose (solid lines).The curves are very similar in relative amplitude to those that can be rescaled [by (BW)0.67] from rats (scaling factor 5) [7] or from humans (scaling factor 181) using the parameters employed in the three-pore model of peritoneal permeability (Table 1) (dotted lines) [13]. However, due to the discrepant relationship between i.p. volume (VD) and peritoneal surface area (S) in mice, the VD/S ratio being only 20–25% of that in humans, the peak time of the 4% glucose UF-curve occurred already at 40–50 min, compared with a peak time of 220–240 min in humans (cf. Equation 5b in [14]).
|
|
|
Small solute permeability-surface area products (PS)
The PS (MTAC) for glucose and 51Cr-EDTA were 51.6 ± 4.0 µl/min (n = 14) and 38.3 ± 1.9 µl/min (n = 14), respectively (Table 2), which is almost identical to the corresponding values measured (or calculated) for humans from a human PD population and scaled to mouse using (BW)0.67 (conversion factor 181) [15]. However, due to the relatively much larger peritoneal surface area in the mouse, the rate of equilibration of i.p. solute concentrations with plasma was increased 4–5-fold (vs humans), as shown in Figure 4A and B. A comparison between human and mouse peritoneal mass transfer data is presented in Table 2.
|
|
Transcapillary escape rate (TER) of RISA
The TER of RISA from plasma (Figure 5A) was 9.9 ± 2.5% per h. This is slightly higher than usually measured in humans, but considerably lower than values usually obtained in rats [16]. Probably due to the fact that the injected volume of RISA tracer was >50 µl in some of the experiments, there seemed to be an initial ‘injection artifact’. This implies that there was some depression of RISA concentration in plasma between 0 and 20 min, followed by a continuous decline of plasma RISA concentration.
|
Clearance of RISA out of the peritoneal cavity (Clout), clearance of RISA to plasma (Cl
P) and clearance of RISA from plasma to peritoneum (Cl D)Clout was 9.33 ± 0.82 µl/min, while Cl
P was 1.49 ± 0.13 µl/min (n = 14), which is also in accordance with values recalculated from humans (11.3 and 0.9 µl/min, respectively; see Table 2). However, Cl D was only 0.084 ± 0.008 µl/min, which is approximately 12% of the value (0.67 µl/min) recalculated from humans using (BW)0.67 as scaling factor. As earlier obtained in rats and humans the 0–120 min accumulation function of RISA in the peritoneal cavity (Figure 5B), for RISA given intravenously, as well as the 0–120 min plasma to dialysate RISA concentration ratios (P/D) (Figure 5C), for RISA administered i.p., showed linear relationships over time.
Clearance of FITC-Ficoll 70/400 to the peritoneal cavity
The clearance of FITC-Ficoll 70/400 as a function of molecular radius (ae) is shown in Figure 6 together with the transperitoneal clearance of albumin. The data are consistent with the two-pore model, in which the small pore radius was 46.9 ± 1.74 Å and the large pore radius 183 ± 11.5 Å, while the fractional hydraulic conductance accounted for by the large pores (
L) was only 0.0081 ± 0.002 (cf. 0.06–0.08 in humans or rat) [8,11]. The transperitoneal albumin (ae = 36 Å) clearance (Cl D) in this study (Table 2) was not significantly different from that of neutral Ficoll36Å, the latter being 0.150 ± 0.032 µl/min.
|
X) determined for Ficoll was 219 ± 73 cm. The clearance (Cl |
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
The present study confirms recent reports indicating that acute PD is feasible in small animals, such as anaesthetized mice [4,6]. By the use of micro-sampling techniques and after validation of the RISA indicator dilution method for i.p. volume measurements, it was possible to assess UF-curves VD(t) and to calculate PS for different size solutes across the peritoneum during PD with reasonable precision in the mouse. Although there are previous reports on ‘rate constants’ of tracer albumin and dextran, respectively, across the mouse peritoneum [17], the present study represents the first quantitative description of transperitoneal small solute and macromolecule transport in terms of PS values and clearances. For the first time albumin mass transfer kinetics across the peritoneum in both the peritoneal-to-plasma and the plasma-to-peritoneal direction are assessed. Furthermore, we have placed mouse peritoneal transport parameters into a more general perspective by subjecting them to scaling.
The PS and clearances as well as the VD(t) curves measured were in good agreement with human data when rescaled to mice (30 g) using the scaling factor (BW)0.67 (=1/181). Indeed, there was a remarkable agreement between the relative amplitudes of the measured UF-curves, especially that for 4% glucose, and those theoretically simulated based on the three-pore model, using (average) human transport parameters for the simulations (Table 1) [13], as rescaled to mouse (Figure 3). From the shape of the VD(t) curves it seems, however, that the mouse peritoneal hydraulic conductance (LpS) may have been somewhat larger than that scaled from humans, because the (relative) curve peaks are somewhat higher in the mouse than predicted from human data [13]. Furthermore, since the instilled volume in mice was neither scaled to body surface area (BSA), which would have yielded an instilled volume of 11 ml (2000 ml/181), nor by BW, which would have yielded a volume of 0.9 ml (2000 ml/2330), from human data, this also affected (increased) the ‘height’ of the UF-curve. Thus, 2.4–2.5 ml of instilled volume is (much) larger than the BW-scaled 0.9 ml. According to Stelin and Rippe [14], the height of the VD(t) curve is determined by the ratio of the peritoneal UF-coefficient to the PS for glucose (LpS/PSg), and it is also determined by the glucose concentration times the absolute volume of the instilled dialysis fluid (Equation 4 in [14]). Furthermore, the time to reach curve peak (tpeak) is dependent on the ratio of i.p. volume to surface area (or more precisely, by VD/PSg according to Equation 5b in [14]). Because the LpS/PSg ratio [where the surface area terms (S) actually cancel] and the glucose concentrations tested remained unaltered in the mice, while the BW-related VD was higher, the expected (relative) ‘height’ of the VD(t) curve should be somewhat higher in mice than in humans. Furthermore, because the VD(t)/PSg was only 20–25% of that in humans, the tpeak would be reduced to 20–25% (40–50 min) compared with that in human PD (220–240 min for 4% glucose). This is more or less exactly consistent with the data obtained! Also, rate constants for the PET (D/P or D/D0 vs time) would be increased 4-fold in mice vs in human PD. This implies, for example, that D/D0 for glucose would reach 0.3–0.4 in 1 h instead of in 4 h (cf. in humans), in agreement with the present results (Figure 4B).
For pharmacokinetic purposes Flessner et al. [18] suggested a scaling factor of (BW)0.7 between species of widely differing BWs, which is partly a ‘compromise’ between scaling for BSA [theoretical scaling factor (BW)0.67] and for BW [(BW)1]. Mathematically, for a sphere, the scaling exponent of surface area to volume is (volume)2/3. However, a number of previous studies have indicated that the metabolic rate scales to BW with an equation having a scaling exponent of 0.70–0.75, which is slightly different from scaling to BSA, using the simple ‘sphere’ analogy [19]. At present there is no accepted explanation why the scaling exponent relating metabolic rate to BW is in general 0.7–0.75, and not close to the value of 0.67. However, the present mouse data were consistent with the latter BSA scaling exponent, rather than with 0.7 (or 0.75).
Whereas we obtained a very good correspondence of mouse data with those in humans (and rat) for a majority of transport coefficients, Cl D when properly scaled, was less than 15% of the expected value, as recalculated from either humans or rat (Table 2). Very low clearances from plasma to peritoneum of albumin and dextran can also be recalculated from the previously published data of Nagy et al. [17]. According to Flessner [20], small solute clearance from plasma to peritoneum occurs mostly from capillaries situated at a distance 0–400 µm from mesothelial surface. Thus, there is a rather steep small solute concentration gradient across this depth. However, for macromolecules the interstitial concentration gradients are much more ‘shallow’, and extend across much larger distances in the peritoneal tissue. Thus, macromolecules entering the peritoneal cavity from the plasma could derive from capillaries at a depth of up to 1–2 mm from the mesothelial surface. On visual inspection, the anterior mouse abdominal wall from the suprapubic region up to the xiphoid region, was seen to be extremely thin and translucent. Histology was done to quantify the depth of tissue available for macromolecular transport (unpublished data). From the linea alba and 4 mm laterally the thickness was only 100–200 µm with gradual growth in thickness in more lateral regions, where the abdominal wall became more muscular. It is therefore speculated that the thinner mouse peritoneum, with less vessels available for delivery of macromolecules, but not for small solute delivery, could be a major reason for the very low clearance of albumin and Ficoll from the plasma to the peritoneum observed [17]. More detailed studies of the mouse peritoneal membrane ultrastructure are, however, warranted.
In apparent contrast to the present results on Cl D, [21] found that the clearance of radiolabelled albumin from plasma to the peritoneal fluid within a small plastic chamber, affixed to the thick portion of the abdominal wall, was not different from that in rat when normalized to the peritoneal contact area. Thus, there seems to be no changes in intrinsic permeability of the capillaries of the mouse peritoneum compared with that in rat peritoneum, since Cl D, when properly scaled, is unchanged when the peritoneal thickness is within the normal range (1–2 mm).
Potentially, a lesser contact area between the dialysate and the peritoneum in mouse and in humans or rat may also have contributed to the low Cl D values. However, in that case small solute mass transfer coefficients (PS values) and Clout for macromolecules would also have been reduced, which was not observed. Indeed, all parameters measured, except Cl D, were in accordance with human and rat data, as aformentioned. Another possibility for the lower clearance of albumin and Ficoll from plasma to peritoneum, and the slightly lower TER, in mice vs rats may be the slightly lower MAP in mice. A lower MAP may imply moderately lower prevailing capillary hydrostatic pressure in the peritoneal microvessels. While this may reduce the clearance significantly, it is not likely that a 30(–40)% reduction in MAP (compared with rat) would result in Ficoll and albumin clearances being only 15% of expected values.
Despite the low macromolecular clearances determined in our experiments, the small pore radius, as well as the large pore radius, analysed according to the two-pore theory, were almost identical to those obtained in either human or rat [8,13]. However, the fraction of hydraulic conductance accounted for by the large pores (L) was only 15% of that previously determined in humans (0.008 ± 0.002 vs 0.06). This is again consistent with a lower number of capillaries contributing to macromolecular transport in mouse than in either humans or rat. These data, however, do not deny the possibility of a lower effective hydrostatic pressure gradient prevailing across the large pores, conceivably due to a lower effective capillary hydrostatic pressure, which may have contributed to the lower macromolecular clearance in the mouse.
In a recent study by Ni et al. [4], PS for urea was assessed using a simplified algorithm (the so-called Garred formula) and was determined to be only 25 µl/min. Rescaling this value to human PD, would yield a value of only 5 ml/min, whereas the present study would predict a PS for urea on the order of 18 ml/min in humans (100 µl/min in the mouse). This value is consistent with results recently obtained by Leypoldt and co-workers (personal communication). The reason for this discrepancy is not clear. However, the absence of a direct VD(t) curve and the use of a simplified algorithm for calculating PS in the study of Ni et al. [4] may partly explain the difference.
Isofluran anaesthesia (3%/vol) has been considered useful in cardiovascular and cerebral studies since it leaves the animals relatively unaffected haemodynamically [3]. Indeed, in the present study the mice investigated seemed to have been only little influenced haemodynamically by the inhalation anaesthesia used, because heart rate was above 500 BPM and blood pressure around 80 mmHg. The average heart rate in conscious mice has been measured to be 500 BPM (range 470–650 BPM) [3], and MAP in anaesthetized mice is usually in the rage of 80–100 mmHg [3]. Furthermore, in our experiments we tried to reduce the blood withdrawal to <100 µl of blood for the entire course of the experiment. Hence, the anaesthetized mice in our study seem to have remained rather stable haemodynamically [3]. Thus, we consider the (rather) low MAP assessed in our study to reflect a generally reduced MAP in mice compared with that in either rat or humans.
It is concluded that quantitative measurements of transport parameters across the peritoneum in mice are feasible with the use of micro-dissection and micro-sampling techniques. With the exception of macromolecular clearance from peritoneum to plasma, all transport parameters measured scaled remarkably well to those assessed in either rat or humans when recalculated using (BW)0.67 as a scaling factor. The disproportionally low macromolecular clearance from plasma to peritoneum in mice compared with humans and rats may be explained by the fact the large solute exchange, in contrast to that of small solutes, occurs from capillaries distributed within a wide depth of the peritoneum. The lower Cl D thus may, at least partly, be due to a much thinner peritoneal parietal tissue, with lesser capillaries that can leak macromolecules, in mice compared with rats and humans.
|
Acknowledgements |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
This study was supported by Swedish Medical Research Council Grant 08285, the Lundberg Medical Foundation, and the Vascular Wall Program at Lund Medical Faculty. The expert secretarial assistance by Kerstin Wihlborg is gratefully acknowledged.
Conflict of interest statement. None declared.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
- Lameire N, Van Biesen W, Van Landschoot M, et al. (1998) Experimental models in peritoneal dialysis: a European experience. Kidney Int 54:2194–2206.[CrossRef][Web of Science][Medline]
- Mortier S, Lameire NH, De Vriese AS. (2005) Animal models in peritoneal dialysis research: a need for consensus. Perit Dial Int 25:16–24.
[Abstract/Free Full Text] - Rao S and Verkman AS. (2000) Analysis of organ physiology in transgenic mice. Am J Physiol Cell Physiol 279:C1–C18.
[Abstract/Free Full Text] - Ni J, Cnops Y, Debaix H, et al. (2005) Functional and molecular characterization of a peritoneal dialysis model in the C57BL/6J mouse. Kidney Int 67:2021–2031.[CrossRef][Web of Science][Medline]
- Twardowski ZJ, Nolph DK, Khanna R, et al. (1987) Peritoneal equilibration test. Perit Dial Bull 7:138–147.
- Ni J, Verbavatz JM, Rippe A, et al. (2006) Aquaporin-1 plays an essential role in water permeability and ultrafiltration during peritoneal dialysis. Kidney Int 69:1518–1525.[CrossRef][Web of Science][Medline]
- Zakaria ER and Rippe B. (1995) Intraperitoneal fluid volume changes during peritoneal dialysis in the rat. Indicator dilution vs volumetric measurements. Blood Purification 13:255–270.[Web of Science][Medline]
- Rosengren BI, Carlsson O, Venturoli D, et al. (2004) Transvascular passage of macromolecules into the peritoneal cavity of normo- and hypothermic rats in vivo: active or passive transport? J Vasc Res 41:123–130.[CrossRef][Web of Science][Medline]
- Ohlson M, Sorensson J, Haraldsson B. (2000) Glomerular size and charge selectivity in the rat as revealed by FITC-ficoll and albumin. Am J Physiol Renal Physiol 279:F84–91.
[Abstract/Free Full Text] - Rippe C, Asgeirsson D, Venturoli D, et al. (2006) Effects of glomerular filtration rate on Ficoll sieving coefficients (theta) in rats. Kidney Int 69:1326–1332.[Web of Science][Medline]
- Rippe B and Stelin G. (1989) Simulations of peritoneal solute transport during continuous ambulatory peritoneal dialysis (CAPD). Application of two-pore formalism. Kidney International 35:1234–1244.[Web of Science][Medline]
- Rippe B and Haraldsson B. (1994) Transport of macromolecules across microvascular walls. The two-pore theory. Physiol Reviews 74:163–219.
[Abstract/Free Full Text] - Rippe B, Stelin G, Haraldsson B. (1991) Computer simulations of peritoneal fluid transport in CAPD. Kidney Int 40:315–325.[Web of Science][Medline]
- Stelin G and Rippe B. (1990) A phenomenological interpretation of the variation in dialysate volume with dwell time in CAPD. Kidney Int 38:465–472.[Web of Science][Medline]
- Simonsen O, Wieslander A, Venturoli D, et al. (2003) Mass transfer of calcium across the peritoneum at three different peritoneal dialysis fluid Ca and glucose concentrations. Kidney Int 64:208–215.[CrossRef][Web of Science][Medline]
- Rippe B, Lundin S, Folkow B. (1978) Plasma volume, blood volume and transcapillary escape rate (TER) of albumin in young spontaneously hypertensive rats (SHR) as compared with normotensive controls (NCR). Clin Exp Hypertens 1:39–50.[Medline]
- Nagy JA, Herzberg KT, Masse EM, et al. (1989) Exchange of macromolecules between plasma and peritoneal cavity in ascites tumor-bearing, normal, and serotonin-injected mice. Cancer Res 49:5448–5458.
[Abstract/Free Full Text] - Flessner MF, Parker RJ, Sieber SM. (1983) Peritoneal lymphatic uptake of fibrinogen and erythrocytes in the rat. Am J Physiol 244:H89–96.
- Singer MA and Morton AR. (2000) Mouse to elephant: biological scaling and Kt/V. Am J Kidney Dis 35:306–309.[Web of Science][Medline]
- Flessner MF. (1991) Peritoneal transport physiology: insights from basic research. J Am Soc Nephrol 2:122–135.[Abstract]
- Flessner MF, Devekadra R, Credit K. (2005) Intrinsic permeability of the mouse peritoneum to water and solutes. Is it similar to the rat? J Am Soc Nephrol 16:270A.
- Carlsson O and Rippe B. (1999) Peritoneal lymphatic absorption and solute exchange during zymosan-induced peritonitis in the rat. Am J Physiol 277:H1107–1112.
Accepted in revised form: 24. 7.06
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  G. Grande, C. Rippe, A. Rippe, A. Rahman, K. Sward, and B. Rippe Unaltered size selectivity of the glomerular filtration barrier in caveolin-1 knockout mice Am J Physiol Renal Physiol, August 1, 2009; 297(2): F257 - F262. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Rippe HOW TO ASSESS TRANSPORT IN ANIMALS? Perit. Dial. Int., February 1, 2009; 29(Supplement_2): S32 - S35. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Bielohuby, N. Herbach, R. Wanke, C. Maser-Gluth, F. Beuschlein, E. Wolf, and A. Hoeflich Growth analysis of the mouse adrenal gland from weaning to adulthood: time- and gender-dependent alterations of cell size and number in the cortical compartment Am J Physiol Endocrinol Metab, July 1, 2007; 293(1): E139 - E146. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||