Central nervous system toxicity and cerebrospinal fluid pharmacokinetics of intraventricularly administered bleomycin in beagles.

The neurotoxic effects and cerebrospinal fluid (CSF) pharmacokinetics of bleomycin were evaluated in beagles after chronic intraventricular administration twice a week for 8 consecutive weeks. Bleomycin was reasonably well tolerated at doses of 0.067 to 0.3 mg/week. Doses higher than 0.3 mg/week produced marked elevation of CSF protein levels and a necrotizing vasculitis within the central nervous system. Pharmacokinetic studies were performed approximately 1 month after the completion of the toxicity studies. [3H]inulin was used as a reference compound. Both [3H]inulin and bleomycin were cleared from the CSF more slowly than in previous studies and more slowly than in normal dogs, which suggests that bulk CSF absorption was reduced by the drug, probably secondary to protein-induced blockage of the arachnoid granulations through which CSF is normally absorbed. Because a "minimally toxic" dose of bleomycin (approximately 0.1 mg/week) produces a CSF C X t of only 1.9 mg/min/ml, we believe that a Phase I clinical trial would be too dangerous given the limited therapeutic potential that a dose of 0.1 mg/week could achieve.

Central nervous system toxicity and cerebrospinal fluid pharmacokinetics of intraventricular 3-[(4-amino-2-methyl-5-pyrimidinyl)ethyl]-1-(2-chloroethyl)-1-nitro soureas and other nitrosoureas in beagles.

The central nervous system toxicity and cerebrospinal fluid (CSF) pharmacokinetics of 3-[(4-amino-2-methyl-5-pyrimidinyl)ethyl]-1-(2-chloroethyl)-1- nitrosoureas, a (ACNU) were determined in beagles and compared to those for three other nitrosoureas, 1-(2-chloroethyl)-3-(2,6-dioxo-3-piperidyl)-1-nitrosourea, 1,3-bis(2-chloroethyl)-1-nitrosourea, and chlorozotocin. Of the four drugs, ACNU was tolerated best and at doses of 0.2 to 0.8 mg/week for 8 consecutive weeks. We found that the average half-time for CSF elimination of ACNU was 18 min (range, 12 to 38 min). This value exceeded the known rate of ACNU decomposition in aqueous solution (28 to 29 min), implying that the disappearance of ACNU from CSF was due to hydrolytic decomposition and cellular entry and/or transcapillary loss across central nervous system capillaries. The drug exposure integral (C X t) of ACNU in the CSF after a "toxic dose low" of 0.8 mg in the dogs would achieve the equivalent of in vitro cell kills in excess of 3 logs for rat 9L and human glioma 126 cells. As a potential therapeutic agent for meningeal neoplasia, the major limiting factor may be that the CSF elimination of ACNU is rapid compared to its equilibration time from ventricle to spinal- and cerebral convexity-subarachnoid space. Based on these results, we have instituted clinical Phase I trials of intra-CSF ACNU.

Neuropharmacological effects of methotrexate perfused through the cerebrospinal fluid system of the rhesus monkey.

Thirteen adult Rhesus monkeys were repeatedly perfused through the ventriculocisternal or ventriculolumbar spaces with Elliott's B solution containing various concentrations of methotrexate (MTX) and trace amounts of [3H]MTX and [carboxy-14C]inulin. The concentrations of MTX ranged from 4.8 to 0.15 mg/ml representing perfusion dosages of 551 mg/sq m to 16 mg/sq m. The average steady-state concentration out-concentration in (Co/Ci) value for MTX was 0.78 +/- 0.04 for the ventriculocisternal and 0.66 +/- 0.01 for the ventriculolumbar routes. MTX treatments did not significantly affect mean inulin steady-state Co/Ci values or CSF formation rate. With the exception of a monkey perfused with MTX at an inflow concentration of 4.8 mg/ml, body weight, food intake, and urine output, analyzed at weekly intervals, generally were not remarkably affected by MTX perfusions. In five monkeys perfused with MTX in concentrations of 4.8 to 0.6 mg/ml, gross neurological toxicity was observed, principally in the form of seizures and hypokinesia during perfusion series with occasional residual motor deficit. Significant cerebral damage was associated with the brains of two monkeys perfused with MTX at concentrations of 2.4 and 0.6 mg/ml and two monkeys perfused at concentrations of 1.2 and 0.3 mg/ml; there of the four animals displayed signs of gross neurotoxicity, and two animals developed permanent motor deficits. However, the extent to which neurotoxic signs could be attributed solely to MTX was difficult to judge because some changes in central nervous system morphology were associated with the mechanical aspects of the procedure. Overall behavioral performance as measured by a visual pattern discrimination reinforced by avoidance or escape from an electric shock was not significantly affected by repeated perfusions of MTX (0.6 mg/ml) in two monkeys not otherwise studied in detail.

A.E. Bennett Award Paper. A kinetic analysis of 5-hydroxyindoleacetic acid excretion from rat brain and csf.

Studies of neurotransmitter kinetics based on intraventricular injections of radio-labeled metabolites have been limited by several problems, including the inability of most investigators to recover more than 45% of the infected isotope from brain homogenates within several minutes after the injection...

The puzzle of where cerebrospinal fluid is absorbed: new pieces.

The widely held theory of cerebrospinal fluid (CSF) absorption by the arachnoid villus system cannot explain the movement of substances within the fluid, the deposition pattern of corpora amylacea on the surface of the brain, and pathological findings in neurological disorders. Experiments studying the movement of melatonin and inulin in the CSF compartment demonstrate that some CSF recycles into the ventricular system and CSF contacting tissue diffusely absorbs some. Photomicrographs of a suprapineal recess portal into the third ventricle are presented. A cycling theory of CSF assigns function to the structure of the choroid fissures and the suprapineal recess.

Entry rate kinetics of D-mannitol and carboxyl-inulin for transfer across the blood-cerebrospinal fluid barrier.

This study evaluates the entry rate kinetics of hydrophilic compounds [3H]-D-mannitol and [14C]-carboxyl-inulin across the blood-cerebrospinal fluid (CSF) barrier in a rabbit experimental model. To maintain steady state levels of these tracers in circulation, 100 microCi of [3H]-D-mannitol and 150 microCi of [14C]-carboxyl-inulin were administered as a bolus and by slow infusion for four hours via a femoral venous catheter. Entry rate kinetics of [3H]-D-mannitol and [14C]-carboxyl-inulin from plasma into cisterna magna CSF were computed using a mathematical equation described by Davson. [3H]-D-mannitol and [14C]-carboxyl-inulin maintained steady state levels throughout the experiment. Entry rates for mannitol and carboxyl-inulin were represented by a straight line, from the slope of which K(out) (or K(in)) were computed: K(in) values for mannitol and carboxyl-inulin were 0.06820 hr(-1) and 0.00023 hr(-1), respectively. Differences in the entry rate of mannitol and carboxyl-inulin may be explained by the molecular size and effective radius of these tracers.

Dynamics of distribution of 3H-inulin between the cerebrospinal fluid compartments.

Since the distribution of substances between various cerebrospinal fluid (CSF) compartments is poorly understood, we studied (3)H-inulin distribution, over time, after its injection into cisterna magna (CM) or lateral ventricle (LV) or cisterna corporis callosi (CCC) in dogs. After the injection into CM (3)H-inulin was well distributed to cisterna basalis (CB), lumbar (LSS) and cortical (CSS) subarachnoid spaces and less distributed to LV. When injected in LV (3)H-inulin was well distributed to all CSF compartments. However, after injection into CCC (3)H-inulin was mostly localized in CCC and adjacent CSS, while its concentrations were much lower in CM and CB and very low in LSS and LV. Concentrations of (3)H-inulin in venous plasma of superior sagittal sinus and arterial plasma were very low and did not differ significantly, while its concentration in urine was very high. In (3)H-inulin distribution it seems that two simultaneous processes are relevant: a) the pulsation of CSF with to-and-fro displacement of CSF and its mixing, carrying (3)H-inulin in all directions, and b) the passage of (3)H-inulin from CSF into nervous parenchyma and its rapid distribution to a huge surface area of capillaries by vessels pulsations. (3)H-inulin then slowly diffuses across capillary walls into the bloodstream to be eliminated in the urine.

Species-specific transfer of plasma albumin from blood into different cerebrospinal fluid compartments in the fetal sheep.

1. The blood-cerebrospinal fluid (CSF) transfer of endogenous sheep albumin and several exogenous species of albumin has been investigated in different CSF compartments of the immature fetal sheep brain, at an early stage of development (60 days gestation, term is 150 days) when the CSF concentration of total protein is high. 2. There were marked differences in the steady-state CSF/plasma ratios for all species of albumin (including endogenous sheep albumin) between different CSF compartments. Ratios measured in the cisterna magna were significantly higher than those in the dorsal subarachnoid space, which in turn were higher than those in the lateral ventricles. The ratios for endogenous sheep albumin were (%; mean +/- S.E.M.): lateral ventricle (LV), 4.0 +/- 0.03; dorsal subarachnoid (DSA), 6.1 +/- 1.0; cisterna magna (CM), 13.7 +/- 0.8. 3. Three hours after I.V. injection, the CSF/plasma ratios for bovine albumin (LV, 2.0 +/- 0.2; DSA, 2.4 +/- 0.1; CM, 7.2 +/- 0.7%) were significantly lower than the ratio for endogenous sheep albumin in all three compartments. The ratios for human albumin (LV, 0.7 +/- 0.2; DSA, 1.0 +/- 0.2; CM, 3.9 +/- 0.4%) were significantly lower than those for bovine albumin. 4. In all three CSF compartments, the endogenous sheep albumin ratios were higher than would be expected on the basis of transfer by passive mechanisms. Conversely, steady-state CSF/plasma ratios for [3H]sucrose and [14C]inulin were consistent with passive transfer, and there were no differences between the ratios for these markers measured in each of the three CSF regions. 5. Goat albumin and [35S]sheep albumin ratios were not significantly different, 5 h after injection, from the endogenous sheep albumin levels in each of the three CSF compartments. 6. It is concluded that in the 60-day-old fetal sheep, all of the endogenous albumin in CSF is derived from the plasma by a specific transfer mechanism that can distinguish between different species of the same protein. There is also some evidence of a small passive component of blood-CSF albumin transfer. 7. Immunocytochemical evidence suggests that the route of transfer from blood to CSF is transcellular, through the choroid plexus epithelial cells. 8. Regional variations in albumin ratios are probably due to differences in specific transfer into each CSF compartment. This is reflected in a differential immunocytochemical staining for albumin in choroid plexus epithelial cells from different regions of the brain. 9. The results are discussed in terms of differences in albumin amino acid sequences, structural homologies, and transfer by a specific transcellular mechanism.

Clearance from cerebrospinal fluid of intrathecally administered beta-endorphin in monkeys.

Five adult male monkeys (Macaca mulatta) weighing 7.1-9.9 kg were given synthetic human beta-endorphin (800 micrograms) and [14C]methoxy-inulin (50 microCi) in 400 microliters of normal saline intrathecally. Serial samples of cerebrospinal fluid were drawn through a previously positioned indwelling spinal catheter and were assayed for concentrations of beta-endorphin (determined by radioimmunoassay) and inulin (determined by liquid scintillation counter). Spinal fluid concentrations of beta-endorphin and inulin peaked and declined in a parallel manner. The clearance ratio (calculated from the reciprocal of the ratio of the areas under the respective curves of elimination of the two species) remained remarkably similar from animal to animal, giving a mean value of 1.060 +/- 0.090 (SEM). This ratio, being near unity, suggests that beta-endorphin is eliminated from spinal fluid in a fashion similar to that of inulin, which is removed exclusively by bulk absorption.

The nature of the decrease in blood-cerebrospinal fluid barrier exchange during postnatal brain development in the rat.

1. The blood-cerebrospinal fluid (CSF) exchange of a wide range of passively transported lipid insoluble compounds (0.43-5.4 nm molecular radius) has been investigated in rats at different stages of postnatal development (2 days old to adult). A novel 'litter-based' model for investigating blood-CSF barrier exchange in immature animals is described. 2. At each age investigated there was a clear inverse correlation between molecular radius and blood-CSF barrier exchange, in addition to an overall decrease in blood-CSF barrier exchange with increasing age. 3. The decrease in blood-CSF barrier exchange with age was not consistent with a reduction in pore diameters, nor does it appear to be due to an increase in the CSF sink effect with age. It seems likely to be due to a relative decrease in the number of a population of large diameter pores.

Intracerebral dialysis and the blood-brain barrier.

The aim of the study was to evaluate how implantation of a dialysis probe influences the blood-brain barrier. Leakage of endogenous serum albumin was evaluated by Evans blue/albumin staining and by immunohistochemistry. The passage from blood to dialysate of two substances that normally do not pass into the brain, [3H]-inulin and glutamate, was studied 3 and 24 h after insertion of a dialysis probe. Evans blue, given 20 min before rats were killed, was observed around the probe and surrounding brain tissue. Albumin immunoreactivity was seen at considerable distance from the probe with larger spread at 24 h than at 3 h after probe insertion. Glutamate and [3H]inulin were detected in the dialysate with no significant further increase of radioactivity after intracarotid infusion of protamine sulfate that enhances the permeability over the blood-brain barrier. When protamine was followed by infusion of glutamate, the concentrations of taurine increased in the dialysate in four of eight rats. That plasma constituents have access to the brain around the dialysis probe is essential to consider, particularly in studies using substances and drugs that do not pass an intact blood-brain barrier.

Distribution in cerebrospinal fluid, blood, and lymph of epidurally injected morphine and inulin in dogs.

We describe procedures for catheterizing the epidural space, the azygos vein, and the thoracic lymph duct of dogs without using fluoroscopy. The success rates of the procedures were 100, 80, and 50%, respectively (n = 10). To assess the validity of the model, 3H-morphine and unlabeled morphine (2 mg) were injected epidurally in ten dogs. Lumbar cerebrospinal fluid (CSF), azygos venous blood, arterial blood, and lymph were sampled before and 5, 20, 60, 120, 180, 240, 300 and 360 min after injection. During the first 20 min, morphine levels in the azygos vein were about three and ten times greater than arterial and lymphatic levels, respectively (n = 3; P less than 0.01). Morphine levels were significantly greater in the azygos vein (n = 8) and the femoral artery (n = 10) during the first 20 and 60 min than they were later, respectively (P less than 0.05). In the lymph (n = 5), the levels of morphine at 60 min were statistically greater (P less than 0.05) than levels at 4, 5, and 6 hr. At no time were the concurrent arterial and lymph levels different from each other. In the lumbar CSF, the morphine peak concentration was reached 5-60 min after epidural injection and ranged between 5 and 93 micrograms/ml. In the CSF, the levels of morphine were significantly greater during the first 20 min than later (n = 7; P less than 0.05). The washout of the lumbar CSF curve for morphine appeared to be fitted by a two-compartment open model. The t1/2-alpha and t1/2-beta values were 14.7 +/- 7.2 min and 106 +/- 45 min, respectively (mean +/- SD). Cumulative percentages of the epidural dose of morphine passed into the azygos system within the first 5, 20, 60, and 120 min after injection were calculated to be 4.0 +/- 2.1, 23.5 +/- 14.6, 49.2 +/- 34.2, and 55.9 +/- 35.3, respectively (mean +/- SD; n = 8). Both 14C-inulin and 3H-morphine were injected epidurally in one dog and intrathecally in another dog. In the CSF, morphine appears to be cleared at a rate similar to that of inulin. The fraction of morphine and inulin crossing the dura after epidural injection was calculated to be 0.31% and 0.59%, respectively.