Abstract
A simple technique for cerebrospinal fluid (CSF) collection was developed in F-344 rats. Cell counts and total protein concentrations were evaluated to assess sample quality. While the 50 to 70 μL samples of CSF collected on three different days showed a progressive decrease in the total erythrocyte and nucleated cell counts, no significant changes were observed in the total protein concentrations. Progressive decreases in the total erythrocyte count correlated positively with the decreases in volume of CSF collected. Our data suggest that collection of less than 50 μL of CSF will give a better quality of CSF in F-344 rats. This is the first report of cellular and protein parameters in the CSF of F-344 rats.
Introduction
Cerebrospinal fluid (CSF), a secretory product of the choroid plexus, circulates through the interconnected system of four ventricles in the brain, central canal of the spinal cord, and the subarachinoid space. CSF, which was formerly thought to function simply as a cushion for the brain has been shown to be a multifunctional fluid. It has an important role in homeostatic, hormonal, and signaling functions (Jones, 2004). In the past, the CSF of rats was assayed infrequently for neurogenic molecules because of the difficulty of collection and low sample volume yields. However, with the advent of more sensitive assay techniques such as enzyme-linked immunosorbent assay and high-performance liquid chromatography (Klockowski and Levy, 1987), the low CSF yield from rats is no longer a limitation in utilization for biomarker measurement in a variety of toxicities and neurological disorders.
Numerous techniques have been described for CSF collection in rats permanently implanted with cannulae or by guided cannulae (Curzon et al., 1985; De La Riva and Yeo, 1985; Jolkkonen et al., 1986; Sanvitto et al., 1987; Dingemanse et al., 1988; Sokomba et al., 1988; Takemoto, 1991; Westergren and Johansson, 1991; Halonen et al., 1992; Consiglio and Lucion, 2000). Ylitalo et al. (1976) and Klockowski and Levy (1987) described the collection of CSF from the cisterna magna of anesthetized rats via percutaneous puncture. CSF from rats is routinely analyzed in many pharmaco-toxicologic (Curzon et al., 1985; De La Riva and Yeo, 1985; Jolkkonen et al., 1986; Kornhuber et al., 1986b; Takemoto, 1991) and biochemistry (Ylitalo et al., 1976; Westergren and Johansson, 1991) research studies. Although, there are infrequent reports of the normal cellular and protein composition of CSF of several strains of laboratory rats (Luxton et al., 1989), the present literature lacks any such reports about Fischer-344 rats. The current study was conducted to refine a technique for percutaneous CSF collection and to characterize the protein concentration and cell counts from visually clear samples of CSF collected from F-344 rats.
Materials and Methods
Animals
Seventy-eight 16-week-old male Fischer 344 rats weighing 200 to 300 g were purchased from Harlan Sprague–Dawley Inc. (Indianapolis, IN) and housed individually in steel cages for 1 week prior to the procedure. The rats were allowed free access to a normal laboratory diet (Certified Rodent Diet 5002, Pellet, supplied by PMI Nutrition International Inc.) and chlorinated potable water. All rats were allowed at least 1 week for acclimation to the housing facilities and the diet before being used in the study. The controls in the animal room were set to maintain a temperature of 69°F to 75°F and 30% to 70% relative humidity. Rats were maintained on a 12-hour light/dark cycle. The Animal Care and Use Committee of Lilly Research Laboratories approved all study protocols. Rats were divided into two groups of 30 rats each: Group A for the purpose of determination of cell counts and group B for total protein assay. The remaining 18 rats were kept as potential replacement animals to compensate for any rats with overt blood contamination of CSF.
CSF Sampling
Immediately following asphyxiation with CO2 the fur was clipped over a portion of the caudal head overlying the cisterna magna of the rats. Thereafter, the rats were held manually so that the occipital bone was almost horizontal to the table and the rest of the body was lying at a 50° angle to the head. This position allowed for the identification of the collection site, which was a 3 mm2 rhomboid depressed area in the midsagittal plane approximately 12 mm caudal to the base of the ears, between the occipital protuberance and the spine of the atlas. The needle of a 1/2 inch × 27 gauge winged infusion set with 8 inch tubing (Terumo) connected to a 1 mL disposable plastic syringe, was inserted into the soft rhomboid surface on the scalp. As the needle passed through the dura-mater into the cisterna magna, a decrease in resistance was perceptible that confirmed the presence of the needle in the cisterna magna. Care was taken to create a gentle negative pressure in the syringe until approximately 50 to 70 μL CSF was drawn into the tubing. The needle was removed from the cisterna magna and CSF was drained by pushing the plunger of the syringe into a properly labeled 2 mL Monoject sterile blood collection tubes (Tyco Healthcare Group) without any additive, and any samples with visible evidence of blood were excluded from the analysis. To ensure euthanasia after CO2 asphyxiation, the thorax was punctured immediately following CSF collection. All CSF samples were collected by a single laboratory technician.
Cell Counts
Both total erythrocyte and total nucleated cell counts were performed by direct enumeration using a calibrated hemocytometer and a light microscope on CSF from 10 Group A rats each on 3 different days. All analyses were completed within 3 hours of CSF collection. For differential nucleated cell counts, cells from 30 μL CSF were concentrated on a 1 × 3 inch glass microscope slide by the Shandon Cytospin 3 Cytocentrifuge (Waltham, MA), were stained with Wright’sGiemsa stain and enumerated with direct light microscopy. The differential nucleated cell count normally included a minimum of 100 nucleated cells. If less than 100 nucleated cells were identified in smears of concentrated CSF, all nucleated cells were counted and individual cell types were expressed as percent of total.
Total Protein Assay
Based on the principle of total protein quantification in CSF described by others (Kornhuber et al., 1986b), CSF samples from 10 Group B rats were analyzed for total protein concentration using the Roche/Hitachi-917 analyzer on 3 different days.
Statistical Correlation
Pearson’s correlation coefficients were calculated to study the linear relationship between volumes of CSF collected at different time points with total erythrocyte and nucleated cell counts and total protein concentrations. The criterion for significance was p < 0.05 (Pagano and Gauvreau, 2000).
Results
The volume of CSF collected from rats varied approximately from 50 to 70 μL. Attempts to obtain CSF failed in 3 rats (3.8%) and 15 rats (19.2%) had visual evidence of blood in CSF samples. A total of 60 of the 78 (76.9%) CSF samples lacked visual evidence of blood and were included in the study making the total number of samples 30 each in groups A and B. Ten samples each were collected on days 1, 2, and 3 from rats in group A and days 4, 5 and 6 from those in group B.
Mean erythrocyte and total nucleated cell counts and total protein concentrations for the two groups of rats are presented in Table 1. Using the CSF collection technique we described, there were few erythrocytes and nucleated cells in CSF of rats in this study.
Erythrocytes were only slightly more common than nucleated cells. Among nucleated cells, lymphocytes were observed in the greatest numbers followed by macrophages and neutrophils in the decreasing order. No nucleated erythrocytes, epithelial cells, eosinophils, or basophils were detected during the differential nucleated cell counts. One of the 10 protein samples collected on the sixth day contained insufficient volume for analysis. As technical experience was gained over the 3 collection days in group A and B, CSF sample volumes, cell counts, and protein concentrations in CSF samples tended to decrease.
To gain insight into the potential relationships between CSF volume collected using this technique and alterations in erythrocytes and total nucleated cell counts and protein concentrations in CSF, Pearson’s correlation coefficients were calculated and a significant linear relationship was observed between the total erythrocyte count and volume of CSF collected on different days.
Discussion
Important constraints for CSF collection in rats include method of restraint and collection as well as collection volume. Methods of restraint were not explored in the current study; but would be expected to have minimal impact on sample values provided that the rat was anesthetized. Because subsequent studies were expected to be terminal, asphyxiation followed by immediate CSF collection was used in the current study. Because the CSF collection occurs very quickly after asphyxiation, any form of chemical restraint would likely provide results similar to those observed in this study.
The techniques described in the literature using an indwelling catheter (Gunther and Herger, 1984; Frankmann, 1986; Kornhuber et al., 1986a; Kornhuber et al., 1986b; Sanvitto et al., 1987) are advantageous for repeated collection of CSF in rats; however, these techniques require surgical intervention, a cumbersome collection apparatus, and may result in complications such as encephalitis and osteitis, blockage of cannulae with fibrous connective tissue, and/or longer collection times. The CSF collection technique used in the present study is relatively rapid, easy to perform, and requires neither surgical intervention nor a complex collection apparatus. Although conducted as a terminal procedure in the current study, full recovery of all of the rats was observed in a subsequent CSF collection study, following CSF collection from rats anesthetized with isoflurane.
The mean erythrocyte count concentration in the CSF samples selected for the current study was 6.21 ± 1.17 × 103/μL (n = 30), which was less than the up to 10 × 103/μL reported by Kornhuber et al. (1986b) and up to 30 × 103/μL reported by Westergren and Johansson (1991). An important reason for the low erythrocyte contamination in the current study was the exclusion of samples with visual evidence of blood contamination from our study. Kornhuber et al. (1986b) demonstrated a significant increase in the concentration of amino acids such as alanine, asparagine, aspartate, citrulline, glutamate, glycine, phenylalanine, and taurine with increase in erythrocyte counts. Although the current effort did not attempt to study correlations between erythrocyte contamination and total protein, Westergren and Johansson (1991) did not observe any correlation between erythrocyte count and albumin concentrations, as long as the erythrocyte counts were less than 30 × 103/μL. Therefore, given that samples in our study had mean erythrocyte counts of 6.21 × 103/μL, it was unlikely that the erythrocyte contamination had any impact on total proteins concentrations. For each of the three collection days in this study sufficient CSF volumes to measure total nucleated cell counts and total erythrocyte counts were obtained. However, the remaining CSF volume to also make concentrated preparations for differential cell counts was adequate from only 46% of the CSF samples (Table 1). Samples with low erythrocyte counts generally correlated with low volume (insufficient sample for differential nucleated cell count). We believe that increased negative pressure during CSF collection likely causes trauma to the blood vessels present in the structures lining the cisterna magna resulting in contamination of CSF with blood. We were able to obtain approximately 25 μL of CSF without applying negative pressure, which may be sufficient to carry out certain assays especially those using sensitive techniques such as ELISA. However, with minimal negative pressure up to 50 μL of CSF can be collected.
Footnotes
Acknowledgments
The authors are indebted to Lilly Research Laboratories, a Division of Eli Lilly and Co. for providing the funding for this project and to Tara Bensch, Qi Liu, Pamela Lee, Carla Rodebeck, and Cathy Durbin for technical support.