- Email: [email protected]
Distal tibial mononeuropathy in diabetic and nondiabetic rats reared on wire cages: an experimental entrapment neuropathy
BRAIN RESEARCH ELSEVIER
Brain Research 698 (1995) 130-136
Research report
Distal tibial mononeuropathy in diabetic and nondiabetic rats reared on wire cages: an experimental entrapment neuropathy D.W. Zochodne
a,* Marilyn
M. Murray
b Paul van
der Sloot c, Richard J. Riopelle d
The University of Calgary, Department of Clinical Neurosciences, 3330 Hospital Drive N.W., Calgary, Alberta, T2N 4NI, Canada b c / o Dr. Richard J. Riopelle, Kingston General Hospital, Apps Research Centre, Room 200, 76 Stuart Street, Kingston, Ontario K7L 2V7, Canada " 10703 68 Avenue, Edmonton, Alberta, T6H 2B6, Canada d Kingston General Hospital, Apps Research Centre, Room 200, 76 Stuart Street, Kingston, Ontario K7L 2V7, Canada Accepted 28 June 1995
Abstract
Using electrophysiological recordings, we studied a distal tibial mononeuropathy that involves the hind foot of rats reared in cages with wire grid flooring. In an initial set of experiments, serial sciatic-tibial motor conduction recordings were made in smaller or larger rats reared in cages with wire grid or sawdust flooring. Electrophysiological features of the neuropathy were loss in the amplitude of the distal tibial nerve M potential recorded over hind limb foot muscles, temporal dispersion of the potential, often into multiple peaks, and a prolonged distal latency of the response. The changes in M amplitude were more apparent in larger rats with a greater body weight. In a second series of experiments we studied sciatic-tibial conduction over 16 weeks in nondiabetic rats and rats rendered diabetic with streptozotocin raised and wire grid or plastic flooring. Tibial mononeuropathy developed in both wire grid-reared groups, but there was evidence that it appeared earlier in diabetic rats. Electrophysiological changes of distal mononeuropathy also obscured the expected slowing of sciatic-tibial motor conduction velocity from diabetics. Tibial mononeuropathy in rats reared on wire grid flooring may be a useful animal model of human entrapment neuropathy but its presence can confound studies of experimental neuropathy. Rats used in studies of experimental neuropathy should be housed in plastic cages with sawdust or shavings flooring. Keywords: Diabetic neuropathy; Nerve injury; Nerve compression
1. Introduction
There are few animal models of 'naturally' developing entrapment mononeuropathies. Carpal tunnel syndrome, for example, is an exceedingly common human entrapment mononeuropathy of the distal portion of the median nerve at the wrist. In this disorder, as in other human entrapment or compression neuropathies, repetitive trauma to the nerve trunk, and perhaps ischemia, are thought to be important in its pathogenesis. Rat models of mononeuropathy, using artificial nerve compression or ligature ties around the nerve, provide only limited information about specific experimental paradigms and do not really model human entrapment [2,3,5]. One interesting reason to consider a model of entrapment in animals would be to study the
* Corresponding author. [email protected]
Fax:
(1)
(403)
283-8731;
E-mail:
0006-8993/95/$09.50 © 1995 Elsevier Science B.V. All rights reserved SSDI 0 0 0 6 - 8 9 9 3 ( 9 5 ) 0 0 8 7 6 - 4
influence of diabetes on its development, a factor that may accelerate human entrapment mononeuropathy [1,6]. In this work, we evaluated a distal tibial mononeuropathy of the hind foot of rats reared on wire floor cages. This mononeuropathy resembles that described in guinea pigs by Fullerton and Gilliatt [4], where distal tibial nerve demyelination results from presumed repetitive compression of its plantar branches in the hind foot. Rats are frequently used for studies of experimental neuropathies involving sciatic-tibial fibers (such as diabetes). Despite the ubiquity of these studies, distal tibial compression mononeuropathies confounding such electrophysiological recordings have not, to our knowledge, been characterized. In an initial set of experiments, we evaluated compression tibial mononeuropathy by comparing sciatic-tibial conduction and caudal conduction (a separate nerve territory not subject to compression) of nondiabetic rats raised on wire cage or plastic sawdust flooring over a 4 - 6 week period. Young rats of initial body weight 158-184 g were com-
D.W. Zochodne et al. /Brain Research 698 (1995) 130-136
pared to older rats, of initial body weight 418-460 g to determine the influence of body weight on the development of the mononeuropathy. In a second series of experiments, we compared the evolution, over 16 weeks, of tibial mononeuropathy in nondiabetic rats and rats rendered diabetic (streptozotocin) reared exclusively on wire cages. The results were compared with electrophysiological studies of similar aged nondiabetic and streptozotocin-diabetic rats raised exclusively on plastic sawdust covered flooring over an identical time duration.
2. Materials and methods
131
the return to baseline after the last peak (excluding the F wave). We verified that tibial fibers almost exclusively account for the compound action potential recorded over the foot muscles by demonstrating its near complete obliteration when the distal tibial nerve was sectioned. During the recordings, near nerve temperature of 37_+ I°C was maintained by placing a thermistor in the subcutaneous tissue over the midsciatic nerve and using a heating lamp and temperature control feedback unit. For caudal studies, we recorded from the proximal tail and stimulated distally (conduction velocity reflects faster conducting sensory fibers from the mixed nerve) or stimulated proximally and recorded distally (motor studies) with subcutaneous near nerve temperatures maintained as above.
2.1. Rats, diabetes 2.4. Histological studies The animals used were male Sprague-Dawley rats with free access to rat chow and water. Diabetes was induced by a single intraperitoneal injection of streptozotocin (65 m g / k g ) in citrate buffer with hyperglycemia verified 3 - 5 days later. Controls received citrate buffer alone. Rats were accepted as diabetic only with a glucose value at or greater than 16.0 mmol/1. Glucose levels were measured from the caudal vein by glucometer (AccuChek II m; Boehringer Mannheim; Canada; Dorval, Quebec), an oxygen rate method (Beckman Glucose Analyzer; Beckman Instruments Inc; Palo Alto, CA) or a glucose oxidase method (Ektachem DT60II analyzer; Kodak, Rochester, NY)
2.2. Flooring Wire cage flooring was a grid of intersecting stainless steel wires with each 'open square' having dimensions of 7 mm by 7 mm. Urine and feces fell through the wire grid to be collected on paper sheets beneath the cages. Each wire cage compartment housed 2 - 6 rats. Plastic cages had a flooring of coarse sawdust particles and housed 2 - 4 rats. The cages and flooring were changed approximately weekly to remove accumulated urine and feces.
2.3. Electrophysiological recordings Techniques for sciatic-tibial, and caudal motor and sensory recordings have been reported previously [10]. For sciatic-tibial studies, we recorded from the dorsal subcutaneous space of the hind foot of the rat with platinum electrodes and stimulated at the sciatic notch and the knee. Conduction velocities of motor fibers were calculated by dividing the difference in the notch and knee distances to the recording site by difference in their respective distal latencies. M potential amplitude was measured from baseline to peak using responses obtained from knee stimulation. Distal motor latencies were taken to the onset of the first negative peak from knee stimulation. M duration was measured from the onset of the first negative deflection to
Histological studies were conducted of nerves fixed in cacodylate buffered glutaraldehyde, embedded in epon and sections stained with toluidine blue. Hindfoot interosseous muscle samples were fixed in formalin, embedded in paraffin and stained with hematoxylin and eosin. Muscle histology was reviewed by a muscle pathologist masked to the origin of the samples.
2.5. Protocols Nondiabetic studies Rats were randomly assigned to housing in units with wire grid flooring, as described above, or sawdust covered plastic floored cages. Electrophysiological recordings were made at the outset then again in 2 weeks, 4 weeks and 6 weeks in smaller rats with initial weights of 158-184 g or at 2 weeks and 4 weeks in rats with larger weights (with initial weights 418-460 g). Histological samples were taken at endpoint (6 weeks) from the smaller rats. Diabetic studies Rats of initial weight 200-300 g were rendered diabetic with streptozotocin as described above or given citrate buffer alone to serve as controls. In one set of experiments, both diabetic rats and their respective controls were kept exlusively on wire grid flooring with electrophysiological recordings 1, 8 and 16 weeks following injection. In a second set of experiments, a separate group of diabetic rats and their respective controls were kept exclusively on sawdust covered plastic flooring and electrophysiological measurements were made 8 and 16 weeks following injection. In each of the experiments pools of identically aged rats were used and randomly assigned to receive either streptozotocin or citrate. 2.6. Analysis For each electrophysiological parameter, mean and S.E.M. values were determined. Selected endpoints were
132
D.W. Zochodne et al. / Brain Research 698 (1995) 130-136
compared using a two-tailed Student's t-test. In the nondiabetic studies (2.5.1.) comparisons were made between the groups reared on wire flooring versus sawdust covered plastic flooring or between values recorded at baseline (0 weeks) versus later times. In the diabetic studies (2.5.2.) comparisons were made between diabetic and nondiabetic groups reared on wire gird or sawdust flooring using a one-way analysis of variance (ANOVA) and post-ANOVA Student's t-test for the 8 and 16 week recordings. Other results in the wire grid rats were compared between diabetics and nondiabetics with an unpaired Student's t-test.
3. Results 3.1. Nondiabetic studies
Rats of initial weight 158-184 g reared on wire grid flooring developed electrophysiological abnormalities confined to sciatic-tibial motor fibers starting 2 - 4 weeks after the initial studies. There was impairment of the expected maturational rise, as observed in the sawdust-plastic reared rats, in the amplitude of the M potential recorded over foot muscles from stimulation of tibial fibers at the knee. At 4 weeks, the M potential in the wire grid flooring group was dispersed compared to sawdust-plastic reared rats. At 6 weeks, the M potential amplitude had declined in the wire grid flooring rats, compared to further maturational rises in the sawdust-plastic group. The M potential in wire grid reared rats was lower in amplitude than that of sawdustplastic reared rats at 6 weeks. There was also an increase in the distal latency and M potential duration. Sciatic-tibial conduction velocity between the sciatic notch and knee was not statistically different between the two groups. This result corresponded with normal histological studies of the sciatic nerve trunk in rats reared on wire flooring. Histological studies of the interosseous foot muscles were normal and comparable in the two groups. Fig. 1 illustrates serial sciatic-tibial recordings following knee stimulation in individual rats raised on wire grid flooring or plastic sawdust flooring. Overall results for this group are given in Table 1. Rats of initial weight 418-460 g raised on wire flooring had a decline in the amplitude of the M potential within the first 2 weeks of the experiment. This reduction persisted at 4 weeks. There was a nonsignificant trend toward dispersion. The sciatic-tibial M potential in sawdust-plastic reared rats did not change over 4 weeks. Results are given in Table 1. 3.2. Diabetic studies
Diabetic rats gained less weight than citrate controls, developed cataracts and had hyperglycemia (Table 2). Both diabetic and nondiabetic rats raised on wire grid
flooring had significantly reduced sciatic-tibial M potential amplitudes after 8 and 16 weeks compared to rats raised on sawdust-plastic flooring. M potentials were highly dispersed in both diabetics and nondiabetics raised on wire flooring (Fig. 2). Rats reared on sawdust-plastic flooring had sciatic-tibial M potentials that were unchanged between 8 and 16 weeks and were comparable to M potential amplitudes of similarly aged nondiabetic rats raised on sawdust-plastic flooring from the nondiabetic studies described above (section 3.1). The development of electrophysiological abnormalities in wire grid flooring reared rats appeared earlier in diabetic rats as indicated by a lower amplitude M potential and a nonsignificant trend toward greater dispersion at 1 week following injection. At 8 weeks, but not 16 weeks, there was a similar but nonsignificant, trend toward lower M potentials and greater dispersion in diabetic than nondiabetic wire flooring reared rats. These differential changes in diabetic rats were not seen in sawdust-plastic reared rats, i.e. diabetes was not associated with declines in the M potential amplitude or an increase in its duration. The distal latency of the M
Serial Sciatic-tibial M potentials plastic-sawdust
wire grid cages
cages
Jr 0 weeks
4 weeks 25 mv I 4ms
6 weeks
Fig. 1. Examples of sciatic-tibial M potentials serially recorded in 4 rats in the smaller weight range (2 on the left raised on sawdust-plastic flooring and 2 on the right on wire grid flooring). Note that the M potentials of rats raised on wire grids are comparable in amplitude and morphology to those from plastic-sawdust cages at the outset of the experiment. At 4 weeks, there is a failure in the expected maturational rise in the amplitude of the M potentials and broadening of the peaks. At 6 weeks, the M potentials are markedly reduced in amplitude, appearing very broad and polyphasic from temporal dispersion.
D.W. Zochodne et al. / B r a i n Research 698 (1995) 130-136
133
Table ! Electrophysiological results in nondiabetic rats reared on sawdust-plastic or wire grid flooring Measurement
0 Weeks Sciatic-tibial fibers
2 Weeks Sciatic-tibial fibers
4 Weeks Sciatic-tibial fibers
6 Weeks Sciatic-tibial fibers
Small rats
Large rats
Wire (5)
Sawdust (5)
Wire (5)
Sawdust (5)
Distal latency (ms) M amplitude (mV) M duration (ms) Conduction velocity ( m / s )
1.6 ±0.1 4.6 ±0.4 1.33 ± 0.03 26.5 ± 1.7
1.7 ±0.1 6.3 -+0.7 1.35 -+ 0.115 26.9 + 2.9
1.6l ± 0.03 11.2 +1.2 1.26 ± 0./13 35.9 ±1.0
1.7 ± 0.1 11.1 ± 1.2 1.4 ± 0. I 41.6±4.4
Distal latency (ms) M amplitude (mY) M duration (ms) Conduction velocity ( m / s )
1.71 ± 0.06 5.9 ±0.6 1.31 ± 0.03 30.6 ± 1.6
1.6 + 0.1 8.8 +0.8 * 1.26 ± 0.03 34.0 ± 5.1
1.68 ± 0.07 7.1 ±11.8 1.4 20.1 41.0 ±2.3
1.81 ± 0.05 10.7 ± 1.4 1.311 + 0.08 47.7 +5.2
Distal latency (ms) M amplitude (mV) M duration (ms) Conduction velocity ( m / s )
1.73 ± 6.3 ± 1.47 ± 33.7 ±
1.64 ± 0.08 8.8 +0.8 1.19 ± 0.03 * 41.7 ±3.4
1.59 ± 0.116 8.0 _+ 1.0 + 1.6 ± 0 . 2 40.3 +4.2
1.7 ± 11.1 10.0 ± 1.3 1.3+t1.1 53.9 + 5.11
Distal latency (ms) M amplitude (mV) M duration (ms) Conduction velocity ( m / s )
1.81 ± 0.0t~ 4.1 -+ 1.3 3.3 ± 1.7 38.1 ± 2.6
0.06 1.0 0.09 1.3
1,46 ± 0.119 12.3 _+0.8 " 1.24 ± 0.02 " * 33.2 -+ 1.7
Results are means ± S.E.M~ * Wire vs. plastic P < 0.05 using an unpaired 2-tailed Student's t-test. * * Wire vs. plastic P _< 0.05 using a Mann-Whitney 2-tailed test. + P ~<0.05 compared to week 0 using a paired Student's t-test.
potential
was
raised on wire floor cages. Sciatic-tibial m o t o r c o n d u c t i o n
l o n g e r in d i a b e t i c a n d n o n d i a b e t i c r a t s r a i s e d o n w i r e c a g e s
following
knee
stimulation
at
16
weeks
velocity was not significantly different between diabetics
c o m p a r e d to t h o s e r a i s e d o n p l a s t i c c a g e s . C a u d a l m o t o r
a n d n o n d i a b e t i c s r a i s e d o n w i r e f l o o r i n g at 16 w e e k s . In
a n d c a u d a l s e n s o r y c o n d u c t i o n w a s s l o w e d in d i a b e t i c r a t s
contrast
sciatic-tibial
motor
conduction
was
slowed
in
Table 2 Electrophysiological results in diabetic rats and nondiabetic rats Wire grid flooring
Measurement
Nondiabetic
Diabetic
Nondiabetic
Weight (g) Glucose (mmol/l)
29 281 ± 1 0 24.5 ± 10
24 611 ±10 6.5 ± 0.2
13 346 ±24 22.2 _+ 0.5
16 567 ±13 4.7 + 0.2
Distal latency (ms) M amplitude (mV) M duration (ms) Conduction velocity ( m / s )
1.88 i 0.03 4.9 ±0.4 1.50 ± 0.06 42.3 ± 1.4
1.85 ±/I.03 6.9 ±0.3 * 1.40 _+ 0.02 42.1 ± 1.1
Distal latency (ms) M amplitude (mV) ~ M duration (ms) b Conduction velocity ( m / s )
1.72 ± 0.02 1.7 ± 1.03 7.8 ±0.6 47.1 ±0.9
1.86 ± 0.08 2.1 +0.04 5.9 ±0.8 46.8 ±2.7
1,91 ± 0.04 13,8 ± 1.1 1.21 + 0.04 50.6 +2.4
1.89 _4_0.02 14.6 ± 1,1 1.20 4- 0.05 52.89 ± 2.2
Distal latency (ms) c M amplitude (mV) d M duration (ms) ¢ Conduction velocity ( m / s ) Conduction velocity ( m / s ) Conduction velocity ( m / s )
1.89 ± 0.05 1.6 ± 0.2 6.9 ± 0.6 49.1 ± 1.6 41.5 ± 1.4 49.3 ± 1.0
2.16 + 0.04 1.2 + 0.2 5.98 ± 0.6 51.6 ± 2.4 45.3 ± 1.8 * 54.2 ± 0.9 *
1.72 _+ 0.03 11.6 ± 1.4 1.20 ± 0.06 46.5 ± 1.6
1.70 + 11.03 11.6 ± 0.8 1.26 t 0.06 51.3 ± 1.3
71
1 week Sciatic-tibial fibers
8 weeks Sciatic-tibial fibers
16 weeks Sciatic-tibial fibers
Caudal motor fibers Caudal sensory fibers
Sawdust-plastic
Diabetic
Values are means ± S.E.M. " P < 0.001 (ANOVA); wire groups vs. plastic groups P < 0.0001. b p < 0.001 (ANOVA); wire groups vs. plastic groups P < 0.0001. c p < 0.001 (ANOVA); wire groups vs. plastic groups P < 0.002. d p < 0.001 (ANOVA); wire groups vs. plastic groups P < 0.01101. P < 0.01101 (ANOVA): wire groups vs. plastic groups P < 0.0001. * Diabetic vs. nondiabetic P _<0.05.
134
D. W. Zochodne et al. / Brain Research 698 (1995) 130-136
SCIATIC-TIBIAL M POTENTIALS IN 8 WEEK DIABETIC RATS sawdust-plastic flooring
wire grid flooring
5 mv
1 mv
5 ms
] 0 mv
] mv I
Fig. 2. Examples of sciatic-tibial M potentials in diabetic rats raised on either sawdust-plastic flooring (left) or wire grid flooring (right) after 8 weeks of hyperglycemia.As in the experimentswith nondiabetic rats, the M potential is considerably reduced in amplitude, broad and polyphasic with multiple peaks indicating temporaldispersion (note change in amplifier gain).
diabetic rats raised on sawdust-plastic cages at 16 weeks. Results are given in Table 2.
4. Discussion
The major findings in this work were: (i) rats develop a hindfoot tibial mononeuropathy related to wire grid floored cage housing; the neuropathy is intensified the longer the animals remain in this type of housing and probably develops more quickly in larger animals with a greater body weight; (ii) the mononeuropathy likely arises from pressure injury on the plantar surface of the paw involving the distal branches of the tibial nerve before they innervate the foot interosseous muscles; (iii) the injury appears largely demyelinative in character and is manifest as a prolonged distal latency, loss of amplitude and temporal dispersion of the tibial M potential recorded over the hindlimb foot muscles; (iv) diabetic rats appeared to be more susceptible to the early development of the mononeuropathy; (v) the mononeuropathy has the potential to obscure meaningful information about motor conduction in sciatic-tibial fibers of experimental models. Other than cage type, no other factors were .identified in these studies to account for the marked difference in M potential morphology between rats raised on wire grid
flooring versus those raised in sawdust-plastic cages. In general, any decline in M potential amplitude from motor nerve stimulation and recording over the endplate zone of a muscle could be accounted for by loss of motor axons, a loss of muscle fibers, distal conduction block from demyelination or temporal dispersion of motor fiber conduction. With significant temporal dispersion, there is failure of individual motor unit action potentials to summate but an increased chance of phase cancellation of these potentials. The resultant M potential, or compound muscle action potential, reflects this summation of individual motor unit action potentials and is reduced in amplitude. The M potentials recorded from sciatic-tibial fibers in our studies were highly dispersed, indicated by increased duration of their negative peaks with a pattern of multiple small peaks. We did not identify histological evidence of direct interosseous muscle foot damage or denervation that would indicate loss of muscle fibers or axon loss. In addition, the distal motor latencies following knee stimulation were longer in wire grid flooring reared rats than those reared in sawdust-plastic cages. These features suggest prominent demyelination rather than axonal degeneration in the early stages of this neuropathy. Electrophysiological measurements were more variable in the initial nondiabetic studies, particularly among the small rats. This variability reflected the error rate of distance measurements in sciatic-tibial fibers of smaller rats, smaller endplate zones and the smaller sample sizes studied in both of these groups. The amplitudes of the M potential at baseline in small rats and the conduction velocities at 2, 4 and 6 weeks in the two groups had somewhat different mean values between them but the differences were not statistically different. Similarly, serial changes in distal latencies and conduction velocities illustrated expected trends toward faster conduction with maturation but could not be regarded as strictly comparable because they changed for different reasons. It would be of interest to look for recovery, as might be expected in a largely demyelinative mononeuropathy, with subsequent removal of rats from wire cages to sawdust plastic cages but we did not conduct this experiment. Also, one could not exclude some superimposed axonal degeneration in longer term exposure to wire cage flooring (our histology samples were of muscles at 6 weeks only). Chronic compression or entrapment mononeuropathies of humans frequently have axonal loss later superimposed on focal demyelinative changes. Both changes were observed in guinea pigs with wire cage mononeuropathy studied for longer time periods [4]. Our results suggest tibial mononeuropathy in rats arises from demyelination of distal branches of the tibial nerve deep to the plantar surface of the hindfoot. The electrophysiological changes closely resemble those of the guinea pig mononeuropathy, where histological study of distal plantar nerves identified segmental demyelination and remyelination [4]. The predisposition we observed toward
D. W. Zochodne et al. / Brain Research 698 (1995) 130-130
earlier neuropathy (within 2 weeks) in rats with greater body weight supports the hypothesis that plantar foot pressure against individual wires in the flooring results in local demyelination. The neuropathy does not appear to be particularly distressing to rats because autotomy was not a feature of it. The mechanism of demyelination in this mononeuropathy is likely to be mechanical, with myelin stripping and intussusception, as observed by Ochoa et al. [7] following experimental pneumatic compression of nerves. Although ischemia might be supposed to occur in compressed nerves, mobile rats are unlikely to exert the same level of pressure over a nerve for more than a few minutes, and axonal degeneration requires at least 1-3 h of continuous severe ischemia to develop [8]. Crush injury of nerves does not result in ongoing ischemia after the actual mechanical event [11,12] and acute ischemia more commonly results in predominant axonal damage rather than focal demyelination [13]. The early susceptibility to neuropathy in diabetes is of interest and resembles the predisposition to entrapment neuropathies observed in human diabetic patients [1,6]. One confounding factor in evaluating the role of diabetes in the development of the mononeuropathy was the marked differences in body weight between diabetic and nondiabetic rats. The greater weight of nondiabetic rats likely enhanced injury to the tibial nerves and may have masked the later influence of diabetes. We observed significant differences between diabetics and nondiabetics early on, when weight differences were less prominent. Baseline sciatic-tibial measurements before streptozotocin or citrate lreatment might have further illustrated the greater loss of 1he M potential in diabetics in serial fashion. Since both diabetics and nondiabetics were randomly assigned from the same pool of identically aged rats, pre-injection differences using the sample sizes studied (n = 24, 29) would have been highly unlikely. Separate work in our lab suggests that sample sizes of 8-10 prior to streptozotocin or citrate injection provide closely matched baseline data. The reasons for early compression neuropathy in diabetes have not been satisfactorily explained. Increased susceptibility to prolonged (diabetic nerves are relatively resistant to the onset of ischemic conduction block) ischemia is one, somewhat unsatisfactory mechanism described above. Dyck et al. [3], studying experimental compression of the sciatic nerve, suggested a relative resistance of diabetic nerve to compression because there were fewer histological changes under a compressed segment of diabetic peroneal nerves. It is uncertain whether this model accurately reflects the mechanical stress of pressure over a wire gird. Insensitivity of the foot to repetitive damage may be an additional important mechanism of diabetic damage but it is uncertain how prominent this would be after only 1 week of diabetes. Finally, diabetic nerve may suffer greater mechanical disruption because of axoglial dysjunction, an abnormality of the architecture of the paranodal area of myelinated fibers [9].
135
An important message from our work is that tibial compression mononeuropathy may complicate studies of experimental neuropathy. In such studies, sciatic-tibial conduction is invariably chosen as an index nerve to investigate. We observed that the tibial mononeuropathy of diabetic rats obscured the expected conduction velocity slowing of sciatic-tibial fibers that was observed in other nerve territories or in sawdust-plastic cage reared rats. Unless other nerve territories are being used, rats should not be raised on wire grid floored cages for studies of experimental neuropathy.
Acknowledgements The work was supported by grants from the ALS Foundation of Canada, the Medical Research Council of Canada and Muscular Dystrophy Association of Canada. D.W.Z. has received personal support from the Ontario Ministry of Health (Career Investigator) and the Alberta Heritage Foundation for Medical Research (Clinical Investigator). Dr. Keith Brownell kindly reviewed our muscle histology samples. Brenda Boake and Heather Price provided expert secretarial assistance.
References [1] Brown, M.J. and Greene, D.A., Diabetic neuropathy: pathophysiology and management. In A.K. Asbury and R.W. Gilliatt (eds.), Peripheral Nert,e Disorders, Butterworths, London, 1~84, pp. 126153. [2] Dougherty, P.M., Garrison, C.J. and Carlton, S.M., Differential influencc of local anesthetic upon two models of experimentally induced peripheral mononeuropathy in the rat, Brain Res., 570 (1992) 109-15. [3] Dyck, P.J., Engelstad, J.K., Giannini, C., Lais, A.C., Minnerath, S.R. and Karnes, J.L., Resistance to axonal degeneration after nerve compression in experimental diabetes. Proc. Natl. Acad. Sci. USA, 86 (1989) 2103-2106. [4] Fullerton, P.M. and Gilliatt, R.W., Pressure neuropathy in the hind foot of the guinea-pig, J. Neurol. Neurosurg. P,sychiat., 30 (1967) 18-25.
[5] Kingery, W.S., Castellote, J.M. and Wang, E.E., A loose ligature-induced mononeuropathy produces hyperalgesias mediated by both the injured sciatic nerve and the adjacent saphenous nerve, Pain, 55 (1993) 297-304. [6] Mulder, D.W., Lambert, E.H., Bastron, J.A. et al., The neuropathies associated with diabetes mellitus, Neurology, 11 (1961) 275-284. [7] Ochoa, J., Fowler, T.J. and Gilliatt, R.W., Anatomical changes in peripheral nerves compressed by a pneumatic tourniquet, J. Anat., 113 (1972) 433-455. [8] Schmelzcr, J.D., Zochodne, D.W. and Low, P.A., lschemic and reperfusion injury of rat peripheral nerve, Proc. Natl. Acad. USA, 86 (1989) 1639-1642. [9] Sima, A.A.F., Brismar, T. and "Yagihashi, S., Neuropathies encountered in the spontaneously diabetic BB Wistar rat. In P.J. Dyck, P.K. Thomas, A.K. Asbury, A.I. Winegrad, D. Porte (Eds.), Diabetic Neuropathy, W.B. Saunders, Toronto, 1987. [10] Zochodne, D.W., Ward, K.K. and Low, P.A., Guanethidine adrcner-
136
D. W. Zochodne et al. / Brain Research 698 (1995) 130-136
gic neuropathy: an animal model of selective autonomic neuropathy, Brain Res., 461 (1988) 10-16. [11] Zochodne, D.W. and Ho, L.T., Endoneurial microenvironment and acute nerve crush injury in the rat sciatic nerve, Brain Res., 535 (1990) 43-48. [12] Zochodne, D,W. and Ho, L.T., Hyperemia of injured peripheral
nerve: sensitivity to CGRP antagonism, Brain Res., 598 (1992) 59-66. [13] Zochodne, D,W., Ho. L.T. and Gross, P.M., Acute endoneurial ischemia induced by epineurial endothelin in the rat sciatic nerve, Ann. J. PhysioL, 263 (1992) HI806-H810.
Brain Research 698 (1995) 130-136
Research report
Distal tibial mononeuropathy in diabetic and nondiabetic rats reared on wire cages: an experimental entrapment neuropathy D.W. Zochodne
a,* Marilyn
M. Murray
b Paul van
der Sloot c, Richard J. Riopelle d
The University of Calgary, Department of Clinical Neurosciences, 3330 Hospital Drive N.W., Calgary, Alberta, T2N 4NI, Canada b c / o Dr. Richard J. Riopelle, Kingston General Hospital, Apps Research Centre, Room 200, 76 Stuart Street, Kingston, Ontario K7L 2V7, Canada " 10703 68 Avenue, Edmonton, Alberta, T6H 2B6, Canada d Kingston General Hospital, Apps Research Centre, Room 200, 76 Stuart Street, Kingston, Ontario K7L 2V7, Canada Accepted 28 June 1995
Abstract
Using electrophysiological recordings, we studied a distal tibial mononeuropathy that involves the hind foot of rats reared in cages with wire grid flooring. In an initial set of experiments, serial sciatic-tibial motor conduction recordings were made in smaller or larger rats reared in cages with wire grid or sawdust flooring. Electrophysiological features of the neuropathy were loss in the amplitude of the distal tibial nerve M potential recorded over hind limb foot muscles, temporal dispersion of the potential, often into multiple peaks, and a prolonged distal latency of the response. The changes in M amplitude were more apparent in larger rats with a greater body weight. In a second series of experiments we studied sciatic-tibial conduction over 16 weeks in nondiabetic rats and rats rendered diabetic with streptozotocin raised and wire grid or plastic flooring. Tibial mononeuropathy developed in both wire grid-reared groups, but there was evidence that it appeared earlier in diabetic rats. Electrophysiological changes of distal mononeuropathy also obscured the expected slowing of sciatic-tibial motor conduction velocity from diabetics. Tibial mononeuropathy in rats reared on wire grid flooring may be a useful animal model of human entrapment neuropathy but its presence can confound studies of experimental neuropathy. Rats used in studies of experimental neuropathy should be housed in plastic cages with sawdust or shavings flooring. Keywords: Diabetic neuropathy; Nerve injury; Nerve compression
1. Introduction
There are few animal models of 'naturally' developing entrapment mononeuropathies. Carpal tunnel syndrome, for example, is an exceedingly common human entrapment mononeuropathy of the distal portion of the median nerve at the wrist. In this disorder, as in other human entrapment or compression neuropathies, repetitive trauma to the nerve trunk, and perhaps ischemia, are thought to be important in its pathogenesis. Rat models of mononeuropathy, using artificial nerve compression or ligature ties around the nerve, provide only limited information about specific experimental paradigms and do not really model human entrapment [2,3,5]. One interesting reason to consider a model of entrapment in animals would be to study the
* Corresponding author. [email protected]
Fax:
(1)
(403)
283-8731;
E-mail:
0006-8993/95/$09.50 © 1995 Elsevier Science B.V. All rights reserved SSDI 0 0 0 6 - 8 9 9 3 ( 9 5 ) 0 0 8 7 6 - 4
influence of diabetes on its development, a factor that may accelerate human entrapment mononeuropathy [1,6]. In this work, we evaluated a distal tibial mononeuropathy of the hind foot of rats reared on wire floor cages. This mononeuropathy resembles that described in guinea pigs by Fullerton and Gilliatt [4], where distal tibial nerve demyelination results from presumed repetitive compression of its plantar branches in the hind foot. Rats are frequently used for studies of experimental neuropathies involving sciatic-tibial fibers (such as diabetes). Despite the ubiquity of these studies, distal tibial compression mononeuropathies confounding such electrophysiological recordings have not, to our knowledge, been characterized. In an initial set of experiments, we evaluated compression tibial mononeuropathy by comparing sciatic-tibial conduction and caudal conduction (a separate nerve territory not subject to compression) of nondiabetic rats raised on wire cage or plastic sawdust flooring over a 4 - 6 week period. Young rats of initial body weight 158-184 g were com-
D.W. Zochodne et al. /Brain Research 698 (1995) 130-136
pared to older rats, of initial body weight 418-460 g to determine the influence of body weight on the development of the mononeuropathy. In a second series of experiments, we compared the evolution, over 16 weeks, of tibial mononeuropathy in nondiabetic rats and rats rendered diabetic (streptozotocin) reared exclusively on wire cages. The results were compared with electrophysiological studies of similar aged nondiabetic and streptozotocin-diabetic rats raised exclusively on plastic sawdust covered flooring over an identical time duration.
2. Materials and methods
131
the return to baseline after the last peak (excluding the F wave). We verified that tibial fibers almost exclusively account for the compound action potential recorded over the foot muscles by demonstrating its near complete obliteration when the distal tibial nerve was sectioned. During the recordings, near nerve temperature of 37_+ I°C was maintained by placing a thermistor in the subcutaneous tissue over the midsciatic nerve and using a heating lamp and temperature control feedback unit. For caudal studies, we recorded from the proximal tail and stimulated distally (conduction velocity reflects faster conducting sensory fibers from the mixed nerve) or stimulated proximally and recorded distally (motor studies) with subcutaneous near nerve temperatures maintained as above.
2.1. Rats, diabetes 2.4. Histological studies The animals used were male Sprague-Dawley rats with free access to rat chow and water. Diabetes was induced by a single intraperitoneal injection of streptozotocin (65 m g / k g ) in citrate buffer with hyperglycemia verified 3 - 5 days later. Controls received citrate buffer alone. Rats were accepted as diabetic only with a glucose value at or greater than 16.0 mmol/1. Glucose levels were measured from the caudal vein by glucometer (AccuChek II m; Boehringer Mannheim; Canada; Dorval, Quebec), an oxygen rate method (Beckman Glucose Analyzer; Beckman Instruments Inc; Palo Alto, CA) or a glucose oxidase method (Ektachem DT60II analyzer; Kodak, Rochester, NY)
2.2. Flooring Wire cage flooring was a grid of intersecting stainless steel wires with each 'open square' having dimensions of 7 mm by 7 mm. Urine and feces fell through the wire grid to be collected on paper sheets beneath the cages. Each wire cage compartment housed 2 - 6 rats. Plastic cages had a flooring of coarse sawdust particles and housed 2 - 4 rats. The cages and flooring were changed approximately weekly to remove accumulated urine and feces.
2.3. Electrophysiological recordings Techniques for sciatic-tibial, and caudal motor and sensory recordings have been reported previously [10]. For sciatic-tibial studies, we recorded from the dorsal subcutaneous space of the hind foot of the rat with platinum electrodes and stimulated at the sciatic notch and the knee. Conduction velocities of motor fibers were calculated by dividing the difference in the notch and knee distances to the recording site by difference in their respective distal latencies. M potential amplitude was measured from baseline to peak using responses obtained from knee stimulation. Distal motor latencies were taken to the onset of the first negative peak from knee stimulation. M duration was measured from the onset of the first negative deflection to
Histological studies were conducted of nerves fixed in cacodylate buffered glutaraldehyde, embedded in epon and sections stained with toluidine blue. Hindfoot interosseous muscle samples were fixed in formalin, embedded in paraffin and stained with hematoxylin and eosin. Muscle histology was reviewed by a muscle pathologist masked to the origin of the samples.
2.5. Protocols Nondiabetic studies Rats were randomly assigned to housing in units with wire grid flooring, as described above, or sawdust covered plastic floored cages. Electrophysiological recordings were made at the outset then again in 2 weeks, 4 weeks and 6 weeks in smaller rats with initial weights of 158-184 g or at 2 weeks and 4 weeks in rats with larger weights (with initial weights 418-460 g). Histological samples were taken at endpoint (6 weeks) from the smaller rats. Diabetic studies Rats of initial weight 200-300 g were rendered diabetic with streptozotocin as described above or given citrate buffer alone to serve as controls. In one set of experiments, both diabetic rats and their respective controls were kept exlusively on wire grid flooring with electrophysiological recordings 1, 8 and 16 weeks following injection. In a second set of experiments, a separate group of diabetic rats and their respective controls were kept exclusively on sawdust covered plastic flooring and electrophysiological measurements were made 8 and 16 weeks following injection. In each of the experiments pools of identically aged rats were used and randomly assigned to receive either streptozotocin or citrate. 2.6. Analysis For each electrophysiological parameter, mean and S.E.M. values were determined. Selected endpoints were
132
D.W. Zochodne et al. / Brain Research 698 (1995) 130-136
compared using a two-tailed Student's t-test. In the nondiabetic studies (2.5.1.) comparisons were made between the groups reared on wire flooring versus sawdust covered plastic flooring or between values recorded at baseline (0 weeks) versus later times. In the diabetic studies (2.5.2.) comparisons were made between diabetic and nondiabetic groups reared on wire gird or sawdust flooring using a one-way analysis of variance (ANOVA) and post-ANOVA Student's t-test for the 8 and 16 week recordings. Other results in the wire grid rats were compared between diabetics and nondiabetics with an unpaired Student's t-test.
3. Results 3.1. Nondiabetic studies
Rats of initial weight 158-184 g reared on wire grid flooring developed electrophysiological abnormalities confined to sciatic-tibial motor fibers starting 2 - 4 weeks after the initial studies. There was impairment of the expected maturational rise, as observed in the sawdust-plastic reared rats, in the amplitude of the M potential recorded over foot muscles from stimulation of tibial fibers at the knee. At 4 weeks, the M potential in the wire grid flooring group was dispersed compared to sawdust-plastic reared rats. At 6 weeks, the M potential amplitude had declined in the wire grid flooring rats, compared to further maturational rises in the sawdust-plastic group. The M potential in wire grid reared rats was lower in amplitude than that of sawdustplastic reared rats at 6 weeks. There was also an increase in the distal latency and M potential duration. Sciatic-tibial conduction velocity between the sciatic notch and knee was not statistically different between the two groups. This result corresponded with normal histological studies of the sciatic nerve trunk in rats reared on wire flooring. Histological studies of the interosseous foot muscles were normal and comparable in the two groups. Fig. 1 illustrates serial sciatic-tibial recordings following knee stimulation in individual rats raised on wire grid flooring or plastic sawdust flooring. Overall results for this group are given in Table 1. Rats of initial weight 418-460 g raised on wire flooring had a decline in the amplitude of the M potential within the first 2 weeks of the experiment. This reduction persisted at 4 weeks. There was a nonsignificant trend toward dispersion. The sciatic-tibial M potential in sawdust-plastic reared rats did not change over 4 weeks. Results are given in Table 1. 3.2. Diabetic studies
Diabetic rats gained less weight than citrate controls, developed cataracts and had hyperglycemia (Table 2). Both diabetic and nondiabetic rats raised on wire grid
flooring had significantly reduced sciatic-tibial M potential amplitudes after 8 and 16 weeks compared to rats raised on sawdust-plastic flooring. M potentials were highly dispersed in both diabetics and nondiabetics raised on wire flooring (Fig. 2). Rats reared on sawdust-plastic flooring had sciatic-tibial M potentials that were unchanged between 8 and 16 weeks and were comparable to M potential amplitudes of similarly aged nondiabetic rats raised on sawdust-plastic flooring from the nondiabetic studies described above (section 3.1). The development of electrophysiological abnormalities in wire grid flooring reared rats appeared earlier in diabetic rats as indicated by a lower amplitude M potential and a nonsignificant trend toward greater dispersion at 1 week following injection. At 8 weeks, but not 16 weeks, there was a similar but nonsignificant, trend toward lower M potentials and greater dispersion in diabetic than nondiabetic wire flooring reared rats. These differential changes in diabetic rats were not seen in sawdust-plastic reared rats, i.e. diabetes was not associated with declines in the M potential amplitude or an increase in its duration. The distal latency of the M
Serial Sciatic-tibial M potentials plastic-sawdust
wire grid cages
cages
Jr 0 weeks
4 weeks 25 mv I 4ms
6 weeks
Fig. 1. Examples of sciatic-tibial M potentials serially recorded in 4 rats in the smaller weight range (2 on the left raised on sawdust-plastic flooring and 2 on the right on wire grid flooring). Note that the M potentials of rats raised on wire grids are comparable in amplitude and morphology to those from plastic-sawdust cages at the outset of the experiment. At 4 weeks, there is a failure in the expected maturational rise in the amplitude of the M potentials and broadening of the peaks. At 6 weeks, the M potentials are markedly reduced in amplitude, appearing very broad and polyphasic from temporal dispersion.
D.W. Zochodne et al. / B r a i n Research 698 (1995) 130-136
133
Table ! Electrophysiological results in nondiabetic rats reared on sawdust-plastic or wire grid flooring Measurement
0 Weeks Sciatic-tibial fibers
2 Weeks Sciatic-tibial fibers
4 Weeks Sciatic-tibial fibers
6 Weeks Sciatic-tibial fibers
Small rats
Large rats
Wire (5)
Sawdust (5)
Wire (5)
Sawdust (5)
Distal latency (ms) M amplitude (mV) M duration (ms) Conduction velocity ( m / s )
1.6 ±0.1 4.6 ±0.4 1.33 ± 0.03 26.5 ± 1.7
1.7 ±0.1 6.3 -+0.7 1.35 -+ 0.115 26.9 + 2.9
1.6l ± 0.03 11.2 +1.2 1.26 ± 0./13 35.9 ±1.0
1.7 ± 0.1 11.1 ± 1.2 1.4 ± 0. I 41.6±4.4
Distal latency (ms) M amplitude (mY) M duration (ms) Conduction velocity ( m / s )
1.71 ± 0.06 5.9 ±0.6 1.31 ± 0.03 30.6 ± 1.6
1.6 + 0.1 8.8 +0.8 * 1.26 ± 0.03 34.0 ± 5.1
1.68 ± 0.07 7.1 ±11.8 1.4 20.1 41.0 ±2.3
1.81 ± 0.05 10.7 ± 1.4 1.311 + 0.08 47.7 +5.2
Distal latency (ms) M amplitude (mV) M duration (ms) Conduction velocity ( m / s )
1.73 ± 6.3 ± 1.47 ± 33.7 ±
1.64 ± 0.08 8.8 +0.8 1.19 ± 0.03 * 41.7 ±3.4
1.59 ± 0.116 8.0 _+ 1.0 + 1.6 ± 0 . 2 40.3 +4.2
1.7 ± 11.1 10.0 ± 1.3 1.3+t1.1 53.9 + 5.11
Distal latency (ms) M amplitude (mV) M duration (ms) Conduction velocity ( m / s )
1.81 ± 0.0t~ 4.1 -+ 1.3 3.3 ± 1.7 38.1 ± 2.6
0.06 1.0 0.09 1.3
1,46 ± 0.119 12.3 _+0.8 " 1.24 ± 0.02 " * 33.2 -+ 1.7
Results are means ± S.E.M~ * Wire vs. plastic P < 0.05 using an unpaired 2-tailed Student's t-test. * * Wire vs. plastic P _< 0.05 using a Mann-Whitney 2-tailed test. + P ~<0.05 compared to week 0 using a paired Student's t-test.
potential
was
raised on wire floor cages. Sciatic-tibial m o t o r c o n d u c t i o n
l o n g e r in d i a b e t i c a n d n o n d i a b e t i c r a t s r a i s e d o n w i r e c a g e s
following
knee
stimulation
at
16
weeks
velocity was not significantly different between diabetics
c o m p a r e d to t h o s e r a i s e d o n p l a s t i c c a g e s . C a u d a l m o t o r
a n d n o n d i a b e t i c s r a i s e d o n w i r e f l o o r i n g at 16 w e e k s . In
a n d c a u d a l s e n s o r y c o n d u c t i o n w a s s l o w e d in d i a b e t i c r a t s
contrast
sciatic-tibial
motor
conduction
was
slowed
in
Table 2 Electrophysiological results in diabetic rats and nondiabetic rats Wire grid flooring
Measurement
Nondiabetic
Diabetic
Nondiabetic
Weight (g) Glucose (mmol/l)
29 281 ± 1 0 24.5 ± 10
24 611 ±10 6.5 ± 0.2
13 346 ±24 22.2 _+ 0.5
16 567 ±13 4.7 + 0.2
Distal latency (ms) M amplitude (mV) M duration (ms) Conduction velocity ( m / s )
1.88 i 0.03 4.9 ±0.4 1.50 ± 0.06 42.3 ± 1.4
1.85 ±/I.03 6.9 ±0.3 * 1.40 _+ 0.02 42.1 ± 1.1
Distal latency (ms) M amplitude (mV) ~ M duration (ms) b Conduction velocity ( m / s )
1.72 ± 0.02 1.7 ± 1.03 7.8 ±0.6 47.1 ±0.9
1.86 ± 0.08 2.1 +0.04 5.9 ±0.8 46.8 ±2.7
1,91 ± 0.04 13,8 ± 1.1 1.21 + 0.04 50.6 +2.4
1.89 _4_0.02 14.6 ± 1,1 1.20 4- 0.05 52.89 ± 2.2
Distal latency (ms) c M amplitude (mV) d M duration (ms) ¢ Conduction velocity ( m / s ) Conduction velocity ( m / s ) Conduction velocity ( m / s )
1.89 ± 0.05 1.6 ± 0.2 6.9 ± 0.6 49.1 ± 1.6 41.5 ± 1.4 49.3 ± 1.0
2.16 + 0.04 1.2 + 0.2 5.98 ± 0.6 51.6 ± 2.4 45.3 ± 1.8 * 54.2 ± 0.9 *
1.72 _+ 0.03 11.6 ± 1.4 1.20 ± 0.06 46.5 ± 1.6
1.70 + 11.03 11.6 ± 0.8 1.26 t 0.06 51.3 ± 1.3
71
1 week Sciatic-tibial fibers
8 weeks Sciatic-tibial fibers
16 weeks Sciatic-tibial fibers
Caudal motor fibers Caudal sensory fibers
Sawdust-plastic
Diabetic
Values are means ± S.E.M. " P < 0.001 (ANOVA); wire groups vs. plastic groups P < 0.0001. b p < 0.001 (ANOVA); wire groups vs. plastic groups P < 0.0001. c p < 0.001 (ANOVA); wire groups vs. plastic groups P < 0.002. d p < 0.001 (ANOVA); wire groups vs. plastic groups P < 0.01101. P < 0.01101 (ANOVA): wire groups vs. plastic groups P < 0.0001. * Diabetic vs. nondiabetic P _<0.05.
134
D. W. Zochodne et al. / Brain Research 698 (1995) 130-136
SCIATIC-TIBIAL M POTENTIALS IN 8 WEEK DIABETIC RATS sawdust-plastic flooring
wire grid flooring
5 mv
1 mv
5 ms
] 0 mv
] mv I
Fig. 2. Examples of sciatic-tibial M potentials in diabetic rats raised on either sawdust-plastic flooring (left) or wire grid flooring (right) after 8 weeks of hyperglycemia.As in the experimentswith nondiabetic rats, the M potential is considerably reduced in amplitude, broad and polyphasic with multiple peaks indicating temporaldispersion (note change in amplifier gain).
diabetic rats raised on sawdust-plastic cages at 16 weeks. Results are given in Table 2.
4. Discussion
The major findings in this work were: (i) rats develop a hindfoot tibial mononeuropathy related to wire grid floored cage housing; the neuropathy is intensified the longer the animals remain in this type of housing and probably develops more quickly in larger animals with a greater body weight; (ii) the mononeuropathy likely arises from pressure injury on the plantar surface of the paw involving the distal branches of the tibial nerve before they innervate the foot interosseous muscles; (iii) the injury appears largely demyelinative in character and is manifest as a prolonged distal latency, loss of amplitude and temporal dispersion of the tibial M potential recorded over the hindlimb foot muscles; (iv) diabetic rats appeared to be more susceptible to the early development of the mononeuropathy; (v) the mononeuropathy has the potential to obscure meaningful information about motor conduction in sciatic-tibial fibers of experimental models. Other than cage type, no other factors were .identified in these studies to account for the marked difference in M potential morphology between rats raised on wire grid
flooring versus those raised in sawdust-plastic cages. In general, any decline in M potential amplitude from motor nerve stimulation and recording over the endplate zone of a muscle could be accounted for by loss of motor axons, a loss of muscle fibers, distal conduction block from demyelination or temporal dispersion of motor fiber conduction. With significant temporal dispersion, there is failure of individual motor unit action potentials to summate but an increased chance of phase cancellation of these potentials. The resultant M potential, or compound muscle action potential, reflects this summation of individual motor unit action potentials and is reduced in amplitude. The M potentials recorded from sciatic-tibial fibers in our studies were highly dispersed, indicated by increased duration of their negative peaks with a pattern of multiple small peaks. We did not identify histological evidence of direct interosseous muscle foot damage or denervation that would indicate loss of muscle fibers or axon loss. In addition, the distal motor latencies following knee stimulation were longer in wire grid flooring reared rats than those reared in sawdust-plastic cages. These features suggest prominent demyelination rather than axonal degeneration in the early stages of this neuropathy. Electrophysiological measurements were more variable in the initial nondiabetic studies, particularly among the small rats. This variability reflected the error rate of distance measurements in sciatic-tibial fibers of smaller rats, smaller endplate zones and the smaller sample sizes studied in both of these groups. The amplitudes of the M potential at baseline in small rats and the conduction velocities at 2, 4 and 6 weeks in the two groups had somewhat different mean values between them but the differences were not statistically different. Similarly, serial changes in distal latencies and conduction velocities illustrated expected trends toward faster conduction with maturation but could not be regarded as strictly comparable because they changed for different reasons. It would be of interest to look for recovery, as might be expected in a largely demyelinative mononeuropathy, with subsequent removal of rats from wire cages to sawdust plastic cages but we did not conduct this experiment. Also, one could not exclude some superimposed axonal degeneration in longer term exposure to wire cage flooring (our histology samples were of muscles at 6 weeks only). Chronic compression or entrapment mononeuropathies of humans frequently have axonal loss later superimposed on focal demyelinative changes. Both changes were observed in guinea pigs with wire cage mononeuropathy studied for longer time periods [4]. Our results suggest tibial mononeuropathy in rats arises from demyelination of distal branches of the tibial nerve deep to the plantar surface of the hindfoot. The electrophysiological changes closely resemble those of the guinea pig mononeuropathy, where histological study of distal plantar nerves identified segmental demyelination and remyelination [4]. The predisposition we observed toward
D. W. Zochodne et al. / Brain Research 698 (1995) 130-130
earlier neuropathy (within 2 weeks) in rats with greater body weight supports the hypothesis that plantar foot pressure against individual wires in the flooring results in local demyelination. The neuropathy does not appear to be particularly distressing to rats because autotomy was not a feature of it. The mechanism of demyelination in this mononeuropathy is likely to be mechanical, with myelin stripping and intussusception, as observed by Ochoa et al. [7] following experimental pneumatic compression of nerves. Although ischemia might be supposed to occur in compressed nerves, mobile rats are unlikely to exert the same level of pressure over a nerve for more than a few minutes, and axonal degeneration requires at least 1-3 h of continuous severe ischemia to develop [8]. Crush injury of nerves does not result in ongoing ischemia after the actual mechanical event [11,12] and acute ischemia more commonly results in predominant axonal damage rather than focal demyelination [13]. The early susceptibility to neuropathy in diabetes is of interest and resembles the predisposition to entrapment neuropathies observed in human diabetic patients [1,6]. One confounding factor in evaluating the role of diabetes in the development of the mononeuropathy was the marked differences in body weight between diabetic and nondiabetic rats. The greater weight of nondiabetic rats likely enhanced injury to the tibial nerves and may have masked the later influence of diabetes. We observed significant differences between diabetics and nondiabetics early on, when weight differences were less prominent. Baseline sciatic-tibial measurements before streptozotocin or citrate lreatment might have further illustrated the greater loss of 1he M potential in diabetics in serial fashion. Since both diabetics and nondiabetics were randomly assigned from the same pool of identically aged rats, pre-injection differences using the sample sizes studied (n = 24, 29) would have been highly unlikely. Separate work in our lab suggests that sample sizes of 8-10 prior to streptozotocin or citrate injection provide closely matched baseline data. The reasons for early compression neuropathy in diabetes have not been satisfactorily explained. Increased susceptibility to prolonged (diabetic nerves are relatively resistant to the onset of ischemic conduction block) ischemia is one, somewhat unsatisfactory mechanism described above. Dyck et al. [3], studying experimental compression of the sciatic nerve, suggested a relative resistance of diabetic nerve to compression because there were fewer histological changes under a compressed segment of diabetic peroneal nerves. It is uncertain whether this model accurately reflects the mechanical stress of pressure over a wire gird. Insensitivity of the foot to repetitive damage may be an additional important mechanism of diabetic damage but it is uncertain how prominent this would be after only 1 week of diabetes. Finally, diabetic nerve may suffer greater mechanical disruption because of axoglial dysjunction, an abnormality of the architecture of the paranodal area of myelinated fibers [9].
135
An important message from our work is that tibial compression mononeuropathy may complicate studies of experimental neuropathy. In such studies, sciatic-tibial conduction is invariably chosen as an index nerve to investigate. We observed that the tibial mononeuropathy of diabetic rats obscured the expected conduction velocity slowing of sciatic-tibial fibers that was observed in other nerve territories or in sawdust-plastic cage reared rats. Unless other nerve territories are being used, rats should not be raised on wire grid floored cages for studies of experimental neuropathy.
Acknowledgements The work was supported by grants from the ALS Foundation of Canada, the Medical Research Council of Canada and Muscular Dystrophy Association of Canada. D.W.Z. has received personal support from the Ontario Ministry of Health (Career Investigator) and the Alberta Heritage Foundation for Medical Research (Clinical Investigator). Dr. Keith Brownell kindly reviewed our muscle histology samples. Brenda Boake and Heather Price provided expert secretarial assistance.
References [1] Brown, M.J. and Greene, D.A., Diabetic neuropathy: pathophysiology and management. In A.K. Asbury and R.W. Gilliatt (eds.), Peripheral Nert,e Disorders, Butterworths, London, 1~84, pp. 126153. [2] Dougherty, P.M., Garrison, C.J. and Carlton, S.M., Differential influencc of local anesthetic upon two models of experimentally induced peripheral mononeuropathy in the rat, Brain Res., 570 (1992) 109-15. [3] Dyck, P.J., Engelstad, J.K., Giannini, C., Lais, A.C., Minnerath, S.R. and Karnes, J.L., Resistance to axonal degeneration after nerve compression in experimental diabetes. Proc. Natl. Acad. Sci. USA, 86 (1989) 2103-2106. [4] Fullerton, P.M. and Gilliatt, R.W., Pressure neuropathy in the hind foot of the guinea-pig, J. Neurol. Neurosurg. P,sychiat., 30 (1967) 18-25.
[5] Kingery, W.S., Castellote, J.M. and Wang, E.E., A loose ligature-induced mononeuropathy produces hyperalgesias mediated by both the injured sciatic nerve and the adjacent saphenous nerve, Pain, 55 (1993) 297-304. [6] Mulder, D.W., Lambert, E.H., Bastron, J.A. et al., The neuropathies associated with diabetes mellitus, Neurology, 11 (1961) 275-284. [7] Ochoa, J., Fowler, T.J. and Gilliatt, R.W., Anatomical changes in peripheral nerves compressed by a pneumatic tourniquet, J. Anat., 113 (1972) 433-455. [8] Schmelzcr, J.D., Zochodne, D.W. and Low, P.A., lschemic and reperfusion injury of rat peripheral nerve, Proc. Natl. Acad. USA, 86 (1989) 1639-1642. [9] Sima, A.A.F., Brismar, T. and "Yagihashi, S., Neuropathies encountered in the spontaneously diabetic BB Wistar rat. In P.J. Dyck, P.K. Thomas, A.K. Asbury, A.I. Winegrad, D. Porte (Eds.), Diabetic Neuropathy, W.B. Saunders, Toronto, 1987. [10] Zochodne, D.W., Ward, K.K. and Low, P.A., Guanethidine adrcner-
136
D. W. Zochodne et al. / Brain Research 698 (1995) 130-136
gic neuropathy: an animal model of selective autonomic neuropathy, Brain Res., 461 (1988) 10-16. [11] Zochodne, D.W. and Ho, L.T., Endoneurial microenvironment and acute nerve crush injury in the rat sciatic nerve, Brain Res., 535 (1990) 43-48. [12] Zochodne, D,W. and Ho, L.T., Hyperemia of injured peripheral
nerve: sensitivity to CGRP antagonism, Brain Res., 598 (1992) 59-66. [13] Zochodne, D,W., Ho. L.T. and Gross, P.M., Acute endoneurial ischemia induced by epineurial endothelin in the rat sciatic nerve, Ann. J. PhysioL, 263 (1992) HI806-H810.