- Email: [email protected]
Effect of applied electrical fields on sprouting of intact saphenous nerve in adult rat
Brain Research, 303 (1984) 331-336 Elsevier
331
BRE 10110
Effect of Applied Electrical Fields on Sprouting of Intact Saphenous Nerve in Adult Rat B. POMERANZ, M. MULLEN and H. MARKUS
Departments of Zoology and Physiology, Universityof Toronto, Toronto, Ont. M5S 1A1 (Canada) (Accepted November 22nd, 1983)
Key words: electrical fields - - nerve sprouting - - nerve lesion - - nerve growth - - regeneration sensory nerve - - electrotherapy ---sciaticnerve -
-
Saphenous nerve sprouting was measured behaviorally and histologically after chronic sciatic denervation in the adult rat. The effect of electrical stimulation (either weak DC fields, or stronger AC fields) on the rate of sprouting of the intact saphenous was studied. Sprouting was enhanced by DC fields (1 pA) if the cathode was placed distal to the growth tips, but was unaffected by anode stimulation. Sprouting was also enhanced by AC fields (1000 #A per pulse) given at 20 Hz and 0.1 ms duration. In the discussion we postulate that separate mechanisms might mediate the AC and DC results. The DC effects are the first demonstration in mammals of results previously observed in lower vertebrates. INTRODUCTION There are two principal sources of cutaneous innervation in the rat hindpaw: the sciatic and saphenous nerves. If the sciatic nerve is p e r m a n e n t l y ligated and severed, the saphenous will sprout and send new branches into sciatic territory reaching a reliable b o u n d a r y on the ventrai skin of the paw. This is d e m a r c a t e d by very clear b o r d e r s at a p p r o x i m a t e l y 3 weeks after the lesion. Sprouting of saphenous beyond this b o u n d a r y proceeds very slowly after the third week for reasons which are not u n d e r s t o o d 5. In this p a p e r we investigated the effects of electrical stimulation of the paw on the rate of saphenous nerve sprouting after chronic sciatic denervation. The existing literature on the effects of D C electric stimulation on growth and regeneration have, until recently, dealt solely with n o n - m a m m a l i a n species or embryonic material. Using ultrasensitive current probes it was found that a m p u t a t e d limbs in salamanders produce a steady flow of ionic current leaving the stump end (in the o r d e r of 20 # A / c m 2) (ref. 3). This led to investigations d e m o n s t r a t i n g that weak D C electric stimulation e n h a n c e d regeneration of nerves in these amphibians 4 and recent experiments showed similar effects in severed l a m p r e y spinal
cord 2. Moreover, D C electric fields have been shown to p r o m o t e neurite formation on cultured neurons from Xenopus 12 and frog e m b r y o s 7 (in vitro). W e undertook the present study to d e t e r m i n e if similar effects occurred in m a m m a l s , and in vivo. METHODS Male adult Wistar rats (350-400 g) were housed in single cages on soft litter beds to minimize a u t o t o m y (self mutilation of anesthetic toes) 16. F o r the sciatic denervation p r o c e d u r e , the animals were anesthetized with sodium pentobarbital. A medial thigh incision exposed the sciatic nerve high in the left thigh. The nerve was ligated in two places (1.5 cm apart) with 4-0 silk and was cut between the ties. (Post mortems after completion of the e x p e r i m e n t indicated that regeneration of the sciatic nerve did not occur.) A f t e r denervation, the skin was sutured with dissolvable chromic gut (4-0), and the animal was given ampicillin i.m. (25 mg/ml - - 0.5 ml). Forty-eight hours after surgery, animals were r a n d o m l y assigned to one of 5 groups to receive t r e a t m e n t for the next 21 days (6 days on, 1 day off, for 3 weeks). The 5 groups consisted of: (1) D C cathodal, 1 ,uA stimulation; (2) DC anodal, 1 MA stimulation; (3) A C 20 Hz, 0.1 ms
Correspondence: B. Pomeranz, Department of Zoology, University of Toronto, Ont. M5S 1A1, Canada. 0006-8993/84/$03.00 © 1984 Elsevier Science Publishers B.V.
332 duration, 1 mA stimulation (biphasic square wave pulses); (4) Sham electrodes, no electrical stimulation; and (5) Control group, receiving no electrodes, no stimulation. The electrodes consisted of 34-gauge stainless steel insect pins inserted to a depth of 0.5 mm. For electrical stimulation (groups 1 through 3), the relevant electrode (for example, the cathode in group 1) was inserted into the most lateral digit of the left hindpaw, and the indifferent electrode (for example, the anode for group 1) was inserted 3 cm from the tip of the tail. Group 4 received 2 similarly placed electrodes, but no stimulation. All stimulations lasted 30 min each day. All animals, including controls (groups 4 and 5) were kept comfortably restrained in plastic holders to which they had been trained, and from which their hindlimbs protruded. The parameters for groups 1 and 2 were selected from the literature on limb regeneration in salamanders4; the parameters for group 3 were obtained from a study on sensory nerve sprouting in rats 11.
Behavioral measurements Our preliminary studies confirmed previous reports 5 that the saphenous sprouting reaches a stable boundary in the foot by 21 days after sciatic denervation. Hence on day 23 after lesioning the sciatic, the thresholds to elicit flexor withdrawal reflexes were measured at each of 20 regions on the ventral aspect of the left paw (see Fig. 1A). Measurements were taken on awake, alert, relatively unstressed animals, using forceps affixed to a strain gauge transducer, so that force (g) could be read directly from a storage oscilloscope. Details of this method have been recently described 13. The force at which the behavioral withdrawal reflex occurred was scored for each of the 20 points on the paw. A 5-minute rest period was allowed between each measurement to avoid stressing the rat. Only one measurement was made at each site to prevent sensitization of the tissues. If the reflex failed to occur by a force of 500 g, the test stimulus was discontinued (forceps released) and the area was scored as 500+. A blind regime was employed in order to prevent experimenter bias, despite the fact that the withdrawal reflex and strain gauge readings were reliable objective measures. Care was taken to avoid stress during both daily stimulation and in the test situation. Twelve animals were assigned to each
of the 5 groups. Some data points, however, were not collected on day 23 as some animals were rejected from the study before day 23 because of autotomy 16.
Histological studies To determine the extent of saphenous sprouting, histology of the skin was done. Moreover, a correlation was made between behavioral responsiveness and the presence of nerve fibers in the skin to validate the behavioral method as an indicator of sprouting. For histological purposes a series of experiments was performed at 21 days after sciatic denervation for an additional 20 animals: 10 in each of two groups, DC cathodal and control. Behavioral responsiveness was determined on day 21 as described above. Each animal was then euthenized for histological silver staining of the nerves in the skin of the areas previously tested in vivo. Euthenasia was performed by overdose of ether followed by cervical dislocation. The skin of the ventral hindpaw was immediately removed and bound to the cutting block (using Tissue-Tek II O.C.T. compound) and fast frozen with dry ice. In this way alone did the modified Protargol-Peroxide technique differ from that of Loots et al. 9, who employed the standard paraffin fixation method. RESULTS
Behavioral measurements Of the 12 animals assigned to each of the 5 behavioral test groups, no less than 10 survived in any given group. The data were analyzed by a non-parametric statistical test as the results were not normally distributed because of the fact that many points were scored as 500+ (non-responsive). Hence the Dunn ztest approximation based on Kruskal-Wallis ranked sums for multiple comparisons (efficiency equal to 95.5% as compared to the Ftest) was used. The criterion for significance was P < 0.05 two-tailed. As reported by previous workers 5 the control group showed behavioral responses as far lateral as regions I, J and K (Fig. 1 and Table I). Of the 5 groups, only 2 groups showed significant behavioral responses as far lateral as L, M and N; these were the DC negative and 20 Hz AC groups (Table I, underlined data). The DC positive, needles only, and control groups showed no significant responses in regions L, M and N. Hence the DC cathodal and 20 Hz
333 lateral
A
Points Measured
B
DC Negative
C
DC Positive
the DC cathodal group had a mean of 0.13 (+ 0.09 S.D.). Thus the treated rats showed a 10-fold increase in fiber count in zones L, M and N. This was significant at P < 0.005 using Student's t-test, twotailed. We then made a correlation between the behavioral results in zones L, M and N for each animal and the number of nerve terminals counted histologically in zones L, M and N for each animal. This gave an r value of 0.65 which was significant at P < 0.05. Fig. 2 shows the correlation graph. This significant positive correlation supports the use of the behavioral test as an indication of sprouting. Furthermore in zones O, P and Q, where no behavioral responses could be elicited (in controls or experimental rats), there were no fibers observed. DISCUSSION
Control
Electrode Alone
AC 20 Hz
Fig. 1. Ventral aspect of the left hindpaw. A shows the 20 zones tested for withdrawal reflex. B-F show a typical example from each of the 5 experimental groups. In these 5 figures a blank area indicates that the zone was not responsive because of sciatic denervation. A dotted area indicates that saphenous mediated responses were present. Note that only B and F show responses in zones L, M, N indicating that responses spread beyond the normal boundary.
treatments both caused spread of saphenous responses from zones I, J and K into the more lateral zones L, M and N.
Histology Using the silver stained material from regions L, M and N, we counted the number of nerve fibers in each section. We found marked differences between control and DC cathodal groups: controls showed a mean of 0.013 (_+ 0.015 S.D.) fibers per section while
As stated above, there have been several reports on the salutory effects of weak DC electric fields on nerve growth in non-mammalsZ,4,7,12. This is the first paper to report enhanced nerve growth in a mammal after application of a DC electric field. Moreover, this is the first such report on a mammal in vivo. The importance of placing the cathode of the DC field distal to the growth tip has been observed in previous studies on non-mammalian preparations2,4,7,12 and is also evident in our results. Indeed the failure of the DC anode (placed distal to the growing sprouts) to enhance regeneration is an excellent control ruling out non-specific effects such as handling, stress, and needle injuries. Clearly the electric field is the cause of the enhanced nerve growth, and not these other artifacts. The possibility that electrolysis products released at the negative electrode might explain these results is rather unlikely in view of the large distances between the site of the electrode (the tip of the 5th toe) and the growth tip of the saphenous sprouts (in zones I, J and K). This distance is between 1.5 and 2.0 cm. Cell membrane barriers would impede diffusion and blood flow would quickly wash away these substances before they reached the nerve sprouts. Although we considered toe implantation of A g AgC1 electrodes connected to the tissues by saline wicks4, we were concerned with the mechanical (trauma) artifacts this might cause. Another important similarity between our results
120 12/12
Controls n = 12
460 2/12
500 0/12
500 0/10
E
500 0/12
500 0/10
500 0/10
500 0/12
500 0/11
80 12/12
120 10/10
50 10/10
150 12/12
120 11/11
F
130 12/12
210 10/10
40 10/10
170 12/12
100 11/11
G
40 12/12
150 10/10
30 10/10
140 12/12
140 11/11
H
410 12/12
370 10/10
340 10/10
370 12/12
360 11/11
1
K
L
240 12/12
270 10/10
120 10/10
330 12/12
120 12/12
260 10/10
160 10/10
280 12/12
460 3/12
500 0/10
360 5/10
480 1/12
320 210 380 11/11 11/11 9/11
J
M
420 4/12
470 1/10
250 7/10
470 2/12
420 8/11
N
0
440 3/12
440 3/10
250 8/10
400 5/12
500 0/12
500 0/10
500 0/10
500 0/12
290 500 1 1 / 1 1 0/11
P
500 0/12
500 0/10
500 0/10
500 0/12
500 0/11
Q
480 1/12
500 0/10
500 0/10
500 0/12
500 0/11
R
500 0/12
500 0/10
500 0/10
500 0/12
500 0/11
* Top row indicates mean force in g to elicit nociceptive reflex. Standard deviation was not calculated as many groups were not normally distributed. ** Bottom row indicates the fraction of animals showing a reflex response. Numerator is n u m b e r of responding rats, denominator is total rats studied. N.B. Underlined data show where the responses extended lateral to and beyond the normal boundary.
80 12/12
500 0/10
170 10/10
Needles n = 10
140 10/10
270 4/10
60 10/10
90 10/10
AC 20 Hz n = 10
500 0/10
500 0/12
500 0/12
140 12/12
150 12/12
D
DC positive n = 12
C 500 1/11
B
90 390 11/11 8/11
Zone A
DC negative 100" n=ll lull**
Treatment group
Behavioral responses
TABLE I
500 0/12
500 0/10
440 1/10
500 0/12
500 0/11
S
T
500 0/12
500 0/10
380 2/10
500 0/12
500 1/11
r~ 4~
335 in mammals and the previous results on non-mammalian preparations is the strength of the electric field which was found to be effective in enhancing sprouting. In the present study this was in the order of 1/zA/cm 2 or 1 mV/cm. The small current densities are two orders of magnitude too small to directly initiate action potentials in the nerves and hence other mechanisms for enhanced growth might be involved in the D C experiments. Three speculations might be considered. (a) Calcium currents have been reported at the growth cone in the lamprey spinal cord 1°. Perhaps weak D C fields might exert an effect on this calcium current. (b) Perhaps there is an electrophoretic effect towards the cathode which might attract positively
charged molecules. Perhaps nerve growth factor which is highly positively charged15, or surface glycoproteins z2 move towards the cathode. Recent evidence shows that Con A receptors accumulate at the cathodal end of the cell implicating glycoproteins 12. (c) One hypothesis suggests that Merkel cells continually secrete a substance which induces nerves to sprout6; perhaps the D C field enhances this effect. The principles which govern the growth enhancement by weak D C fields probably do not apply to our results with 20 Hz A C pulses. Bipolar square pulses were used in the A C experiments, ruling out the contribution of net D C fields. Moreover, in the 20 Hz experiment sufficient current was delivered to initiate nerve impulses. A previous abstract reported the efficacy of nerve impulses to promote nerve
A 0.30
0.25
A
0.20
A
~ 0.15
A A >
z 0.10
A A
0.05
A A A
VV 0
VV V V
•
0
1
2
3
Number of Sites Responding Fig. 2. Correlation of histological nerve fiber counts in zones L, M and N (vertical axis), with behaviorally elicited flexor withdrawal reflex from these zones (horizontal axis). Each triangle represents one rat; the results from zones L, M and N were combined for each animal. Open triangles are from the 10 rats which had been given DC negative treatment. Closed triangles are from the 10 control animals receiving no treatment. The r value was 0.65 for this correlation, showing that the behavioral test is a valid indication of the extent of histological sprouting.
336 sprouting of sensory nerves to the skin on the back of an adult rat 11. In that study the nerve was directly stimulated at 20 Hz for 10 min on 3 occasions (on day 0, 4 and 8). In that experiment (as in our 20 Hz group), sufficient current was used to activate action potentials on the nerve. H e n c e our results suggest that at least two different mechanisms may be involved in the effects of electrical fields on nerve growth: one involving a weak D C field with cathode distal to the growth tip, the other requiring stronger A C pulses sufficient to p r o m o t e action potentials on the growing nerve. Unfortunately we cannot d e t e r m i n e from our histological data whether enhanced sprouting involved myelinated or unmyelinated fibers. I n d e e d , there is some uncertainty as to whether silver stains can reveal unmyelinated fibers in the light microscope (J. D i a m o n d , personal communication). The possible role of weak electric fields in the regulation of developmental and regenerative processes has been postulated 1. H o w e v e r previous reports
have been on the salutory effects of applied electric fields in lower vertebrates, or in embryos, or in vitro; hence the applicability to nerve regeneration in adult mammals in vivo has been open to question. I n d e e d lower vertebrates and embryos have long been known to have a greater regenerative capacity than adult mammals. H o w e v e r , recent evidence in young rats with mid-arm amputations, showed that DC electrical stimulation enhances limb regenerationS.14. Our finding in adult rat is the first to show that application of D C fields enhance nerve growth in mammals. These concepts can no longer be restricted to lower vertebrates. ACKNOWLEDGEMENTS This research was s u p p o r t e d by N S E R C of Canada. W e appreciate the able technical assistance of B. Dressier, P. F o r t n e r , D. Krushelnycky, E. Pomeranz and T. Sosath. W e thank Dr. J. D i a m o n d for showing us the histology technique.
REFERENCES 1 Borgens, R. B., The role of ionic current in the regeneration and development of the amphibian limb. In J. F. Fallon and A. I. Caplan, (Eds.), Limb Development and Regeneration, Alan Liss, New York, 1981, pp. 597-608. 2 Borgens, R. B., Roederer, E. and Cohen, M. J., Enhanced spinal cord regeneration in lamprey by applied electric fields, Science, 213 (1981) 611-617. 3 Borgens, R. B., Vanable, J. W. and Jaffe, L. F., Bioelectricity and regeneration: large currents leave the stumps of regenerating newt limbs, Proc. nat. Acad. Sci. U.S.A., 74 (1977) 4528-4530. 4 Borgens, R. B., Vanable, J. W., Jr. and Jaffe, L. F., Bioelectricity and regeneration. Initiation of frog limb regenera-, tion by minute currents, J. exp. Zool., 200 (1977) 403-416. 5 Devor, M., Schonfeld, D., Seltzer, Z. and Wall, P. D., Two modes of cutaneous reinnervation following peripheral nerve injury, J. comp. Neurol., 185 (1979) 211-220. 6 Diamond, J., The regulation of nerve sprouting by extrinsic influences. In F. O. Schmitt, (Ed.), The Neurosciences 4th Study Program, M.I.T., Cambridge, MA, 1979, pp. 937-955. 7 Hinkle, L., McCaig, L. D. and Robinson, K. R., The direction of growth of differentiating neurons and myoblasts from frog embryos in an applied electric field, J. Physiol. (Lond.), 314 (1981) 121-136. 8 Libbin, R. M., Person, P., Papierman, S., Shah, D., Nevid, D. and Grob, H., Partial limb regeneration of the above-el-
bow amputated forelimb, J. Morphol., 159 (1979) 439-452.. 9 Loots, G. P., Loots, J. M., Brown, M. M. and Schoeman, J. L., A rapid silver impregnation method for nervous tissue: a modified Protargol-peroxide technique, Stain Technol., 54 (1979) 97-100. 10 MacVicar, B. and Llinas, R., Calcium spikes in regenerating giant axons of the Lamprey spinal cord, Soc. Neurosci. Abstr., 8 (1982) 914. 11 Nixon, B., Jackson, P., Diamond, A., Foerster, A. and Diamond, J., Impulse activity evokes collateral sprouting of intact nerves into available target tissue, Soc~ Neurosci. Abstr., 6 (1980) 171. 12 Patel, N. and Poo, M. M., Orientation of neurite growth by extracellular electric fields, J. Neurosci., 2 (1982) 483-496. 13 Pomeranz, B., Markus, H. and Krushelnycky, D., Spread of saphenous somatotopic projection map in spinal cord and hypersensitivity of the foot after chronic sciatic denervation in adult rat, Brain Research, in press. 14 Smith, S. D., The role of electrode position in the electrical induction of limb regeneration in the subadult rat, Bioelectrochem. Bioenergetics, 8 (1981) 661-670. 15 Thoenen, H. and Barde, Y. A., Physiology of Nerve Growth Factor, Physiol. Rev., 60 (1980) 1284-1335. 16 Wall, P. D., Devor, M., Inbal, R., Scadiwl, J. W., Schonfeld, D., Seltzer, Z. and Tomkiewicz, M. M., Autotomy following peripheral nerve lesions: experimental anesthesia dolorosa, Pain, 7 (1979) 103-113.
331
BRE 10110
Effect of Applied Electrical Fields on Sprouting of Intact Saphenous Nerve in Adult Rat B. POMERANZ, M. MULLEN and H. MARKUS
Departments of Zoology and Physiology, Universityof Toronto, Toronto, Ont. M5S 1A1 (Canada) (Accepted November 22nd, 1983)
Key words: electrical fields - - nerve sprouting - - nerve lesion - - nerve growth - - regeneration sensory nerve - - electrotherapy ---sciaticnerve -
-
Saphenous nerve sprouting was measured behaviorally and histologically after chronic sciatic denervation in the adult rat. The effect of electrical stimulation (either weak DC fields, or stronger AC fields) on the rate of sprouting of the intact saphenous was studied. Sprouting was enhanced by DC fields (1 pA) if the cathode was placed distal to the growth tips, but was unaffected by anode stimulation. Sprouting was also enhanced by AC fields (1000 #A per pulse) given at 20 Hz and 0.1 ms duration. In the discussion we postulate that separate mechanisms might mediate the AC and DC results. The DC effects are the first demonstration in mammals of results previously observed in lower vertebrates. INTRODUCTION There are two principal sources of cutaneous innervation in the rat hindpaw: the sciatic and saphenous nerves. If the sciatic nerve is p e r m a n e n t l y ligated and severed, the saphenous will sprout and send new branches into sciatic territory reaching a reliable b o u n d a r y on the ventrai skin of the paw. This is d e m a r c a t e d by very clear b o r d e r s at a p p r o x i m a t e l y 3 weeks after the lesion. Sprouting of saphenous beyond this b o u n d a r y proceeds very slowly after the third week for reasons which are not u n d e r s t o o d 5. In this p a p e r we investigated the effects of electrical stimulation of the paw on the rate of saphenous nerve sprouting after chronic sciatic denervation. The existing literature on the effects of D C electric stimulation on growth and regeneration have, until recently, dealt solely with n o n - m a m m a l i a n species or embryonic material. Using ultrasensitive current probes it was found that a m p u t a t e d limbs in salamanders produce a steady flow of ionic current leaving the stump end (in the o r d e r of 20 # A / c m 2) (ref. 3). This led to investigations d e m o n s t r a t i n g that weak D C electric stimulation e n h a n c e d regeneration of nerves in these amphibians 4 and recent experiments showed similar effects in severed l a m p r e y spinal
cord 2. Moreover, D C electric fields have been shown to p r o m o t e neurite formation on cultured neurons from Xenopus 12 and frog e m b r y o s 7 (in vitro). W e undertook the present study to d e t e r m i n e if similar effects occurred in m a m m a l s , and in vivo. METHODS Male adult Wistar rats (350-400 g) were housed in single cages on soft litter beds to minimize a u t o t o m y (self mutilation of anesthetic toes) 16. F o r the sciatic denervation p r o c e d u r e , the animals were anesthetized with sodium pentobarbital. A medial thigh incision exposed the sciatic nerve high in the left thigh. The nerve was ligated in two places (1.5 cm apart) with 4-0 silk and was cut between the ties. (Post mortems after completion of the e x p e r i m e n t indicated that regeneration of the sciatic nerve did not occur.) A f t e r denervation, the skin was sutured with dissolvable chromic gut (4-0), and the animal was given ampicillin i.m. (25 mg/ml - - 0.5 ml). Forty-eight hours after surgery, animals were r a n d o m l y assigned to one of 5 groups to receive t r e a t m e n t for the next 21 days (6 days on, 1 day off, for 3 weeks). The 5 groups consisted of: (1) D C cathodal, 1 ,uA stimulation; (2) DC anodal, 1 MA stimulation; (3) A C 20 Hz, 0.1 ms
Correspondence: B. Pomeranz, Department of Zoology, University of Toronto, Ont. M5S 1A1, Canada. 0006-8993/84/$03.00 © 1984 Elsevier Science Publishers B.V.
332 duration, 1 mA stimulation (biphasic square wave pulses); (4) Sham electrodes, no electrical stimulation; and (5) Control group, receiving no electrodes, no stimulation. The electrodes consisted of 34-gauge stainless steel insect pins inserted to a depth of 0.5 mm. For electrical stimulation (groups 1 through 3), the relevant electrode (for example, the cathode in group 1) was inserted into the most lateral digit of the left hindpaw, and the indifferent electrode (for example, the anode for group 1) was inserted 3 cm from the tip of the tail. Group 4 received 2 similarly placed electrodes, but no stimulation. All stimulations lasted 30 min each day. All animals, including controls (groups 4 and 5) were kept comfortably restrained in plastic holders to which they had been trained, and from which their hindlimbs protruded. The parameters for groups 1 and 2 were selected from the literature on limb regeneration in salamanders4; the parameters for group 3 were obtained from a study on sensory nerve sprouting in rats 11.
Behavioral measurements Our preliminary studies confirmed previous reports 5 that the saphenous sprouting reaches a stable boundary in the foot by 21 days after sciatic denervation. Hence on day 23 after lesioning the sciatic, the thresholds to elicit flexor withdrawal reflexes were measured at each of 20 regions on the ventral aspect of the left paw (see Fig. 1A). Measurements were taken on awake, alert, relatively unstressed animals, using forceps affixed to a strain gauge transducer, so that force (g) could be read directly from a storage oscilloscope. Details of this method have been recently described 13. The force at which the behavioral withdrawal reflex occurred was scored for each of the 20 points on the paw. A 5-minute rest period was allowed between each measurement to avoid stressing the rat. Only one measurement was made at each site to prevent sensitization of the tissues. If the reflex failed to occur by a force of 500 g, the test stimulus was discontinued (forceps released) and the area was scored as 500+. A blind regime was employed in order to prevent experimenter bias, despite the fact that the withdrawal reflex and strain gauge readings were reliable objective measures. Care was taken to avoid stress during both daily stimulation and in the test situation. Twelve animals were assigned to each
of the 5 groups. Some data points, however, were not collected on day 23 as some animals were rejected from the study before day 23 because of autotomy 16.
Histological studies To determine the extent of saphenous sprouting, histology of the skin was done. Moreover, a correlation was made between behavioral responsiveness and the presence of nerve fibers in the skin to validate the behavioral method as an indicator of sprouting. For histological purposes a series of experiments was performed at 21 days after sciatic denervation for an additional 20 animals: 10 in each of two groups, DC cathodal and control. Behavioral responsiveness was determined on day 21 as described above. Each animal was then euthenized for histological silver staining of the nerves in the skin of the areas previously tested in vivo. Euthenasia was performed by overdose of ether followed by cervical dislocation. The skin of the ventral hindpaw was immediately removed and bound to the cutting block (using Tissue-Tek II O.C.T. compound) and fast frozen with dry ice. In this way alone did the modified Protargol-Peroxide technique differ from that of Loots et al. 9, who employed the standard paraffin fixation method. RESULTS
Behavioral measurements Of the 12 animals assigned to each of the 5 behavioral test groups, no less than 10 survived in any given group. The data were analyzed by a non-parametric statistical test as the results were not normally distributed because of the fact that many points were scored as 500+ (non-responsive). Hence the Dunn ztest approximation based on Kruskal-Wallis ranked sums for multiple comparisons (efficiency equal to 95.5% as compared to the Ftest) was used. The criterion for significance was P < 0.05 two-tailed. As reported by previous workers 5 the control group showed behavioral responses as far lateral as regions I, J and K (Fig. 1 and Table I). Of the 5 groups, only 2 groups showed significant behavioral responses as far lateral as L, M and N; these were the DC negative and 20 Hz AC groups (Table I, underlined data). The DC positive, needles only, and control groups showed no significant responses in regions L, M and N. Hence the DC cathodal and 20 Hz
333 lateral
A
Points Measured
B
DC Negative
C
DC Positive
the DC cathodal group had a mean of 0.13 (+ 0.09 S.D.). Thus the treated rats showed a 10-fold increase in fiber count in zones L, M and N. This was significant at P < 0.005 using Student's t-test, twotailed. We then made a correlation between the behavioral results in zones L, M and N for each animal and the number of nerve terminals counted histologically in zones L, M and N for each animal. This gave an r value of 0.65 which was significant at P < 0.05. Fig. 2 shows the correlation graph. This significant positive correlation supports the use of the behavioral test as an indication of sprouting. Furthermore in zones O, P and Q, where no behavioral responses could be elicited (in controls or experimental rats), there were no fibers observed. DISCUSSION
Control
Electrode Alone
AC 20 Hz
Fig. 1. Ventral aspect of the left hindpaw. A shows the 20 zones tested for withdrawal reflex. B-F show a typical example from each of the 5 experimental groups. In these 5 figures a blank area indicates that the zone was not responsive because of sciatic denervation. A dotted area indicates that saphenous mediated responses were present. Note that only B and F show responses in zones L, M, N indicating that responses spread beyond the normal boundary.
treatments both caused spread of saphenous responses from zones I, J and K into the more lateral zones L, M and N.
Histology Using the silver stained material from regions L, M and N, we counted the number of nerve fibers in each section. We found marked differences between control and DC cathodal groups: controls showed a mean of 0.013 (_+ 0.015 S.D.) fibers per section while
As stated above, there have been several reports on the salutory effects of weak DC electric fields on nerve growth in non-mammalsZ,4,7,12. This is the first paper to report enhanced nerve growth in a mammal after application of a DC electric field. Moreover, this is the first such report on a mammal in vivo. The importance of placing the cathode of the DC field distal to the growth tip has been observed in previous studies on non-mammalian preparations2,4,7,12 and is also evident in our results. Indeed the failure of the DC anode (placed distal to the growing sprouts) to enhance regeneration is an excellent control ruling out non-specific effects such as handling, stress, and needle injuries. Clearly the electric field is the cause of the enhanced nerve growth, and not these other artifacts. The possibility that electrolysis products released at the negative electrode might explain these results is rather unlikely in view of the large distances between the site of the electrode (the tip of the 5th toe) and the growth tip of the saphenous sprouts (in zones I, J and K). This distance is between 1.5 and 2.0 cm. Cell membrane barriers would impede diffusion and blood flow would quickly wash away these substances before they reached the nerve sprouts. Although we considered toe implantation of A g AgC1 electrodes connected to the tissues by saline wicks4, we were concerned with the mechanical (trauma) artifacts this might cause. Another important similarity between our results
120 12/12
Controls n = 12
460 2/12
500 0/12
500 0/10
E
500 0/12
500 0/10
500 0/10
500 0/12
500 0/11
80 12/12
120 10/10
50 10/10
150 12/12
120 11/11
F
130 12/12
210 10/10
40 10/10
170 12/12
100 11/11
G
40 12/12
150 10/10
30 10/10
140 12/12
140 11/11
H
410 12/12
370 10/10
340 10/10
370 12/12
360 11/11
1
K
L
240 12/12
270 10/10
120 10/10
330 12/12
120 12/12
260 10/10
160 10/10
280 12/12
460 3/12
500 0/10
360 5/10
480 1/12
320 210 380 11/11 11/11 9/11
J
M
420 4/12
470 1/10
250 7/10
470 2/12
420 8/11
N
0
440 3/12
440 3/10
250 8/10
400 5/12
500 0/12
500 0/10
500 0/10
500 0/12
290 500 1 1 / 1 1 0/11
P
500 0/12
500 0/10
500 0/10
500 0/12
500 0/11
Q
480 1/12
500 0/10
500 0/10
500 0/12
500 0/11
R
500 0/12
500 0/10
500 0/10
500 0/12
500 0/11
* Top row indicates mean force in g to elicit nociceptive reflex. Standard deviation was not calculated as many groups were not normally distributed. ** Bottom row indicates the fraction of animals showing a reflex response. Numerator is n u m b e r of responding rats, denominator is total rats studied. N.B. Underlined data show where the responses extended lateral to and beyond the normal boundary.
80 12/12
500 0/10
170 10/10
Needles n = 10
140 10/10
270 4/10
60 10/10
90 10/10
AC 20 Hz n = 10
500 0/10
500 0/12
500 0/12
140 12/12
150 12/12
D
DC positive n = 12
C 500 1/11
B
90 390 11/11 8/11
Zone A
DC negative 100" n=ll lull**
Treatment group
Behavioral responses
TABLE I
500 0/12
500 0/10
440 1/10
500 0/12
500 0/11
S
T
500 0/12
500 0/10
380 2/10
500 0/12
500 1/11
r~ 4~
335 in mammals and the previous results on non-mammalian preparations is the strength of the electric field which was found to be effective in enhancing sprouting. In the present study this was in the order of 1/zA/cm 2 or 1 mV/cm. The small current densities are two orders of magnitude too small to directly initiate action potentials in the nerves and hence other mechanisms for enhanced growth might be involved in the D C experiments. Three speculations might be considered. (a) Calcium currents have been reported at the growth cone in the lamprey spinal cord 1°. Perhaps weak D C fields might exert an effect on this calcium current. (b) Perhaps there is an electrophoretic effect towards the cathode which might attract positively
charged molecules. Perhaps nerve growth factor which is highly positively charged15, or surface glycoproteins z2 move towards the cathode. Recent evidence shows that Con A receptors accumulate at the cathodal end of the cell implicating glycoproteins 12. (c) One hypothesis suggests that Merkel cells continually secrete a substance which induces nerves to sprout6; perhaps the D C field enhances this effect. The principles which govern the growth enhancement by weak D C fields probably do not apply to our results with 20 Hz A C pulses. Bipolar square pulses were used in the A C experiments, ruling out the contribution of net D C fields. Moreover, in the 20 Hz experiment sufficient current was delivered to initiate nerve impulses. A previous abstract reported the efficacy of nerve impulses to promote nerve
A 0.30
0.25
A
0.20
A
~ 0.15
A A >
z 0.10
A A
0.05
A A A
VV 0
VV V V
•
0
1
2
3
Number of Sites Responding Fig. 2. Correlation of histological nerve fiber counts in zones L, M and N (vertical axis), with behaviorally elicited flexor withdrawal reflex from these zones (horizontal axis). Each triangle represents one rat; the results from zones L, M and N were combined for each animal. Open triangles are from the 10 rats which had been given DC negative treatment. Closed triangles are from the 10 control animals receiving no treatment. The r value was 0.65 for this correlation, showing that the behavioral test is a valid indication of the extent of histological sprouting.
336 sprouting of sensory nerves to the skin on the back of an adult rat 11. In that study the nerve was directly stimulated at 20 Hz for 10 min on 3 occasions (on day 0, 4 and 8). In that experiment (as in our 20 Hz group), sufficient current was used to activate action potentials on the nerve. H e n c e our results suggest that at least two different mechanisms may be involved in the effects of electrical fields on nerve growth: one involving a weak D C field with cathode distal to the growth tip, the other requiring stronger A C pulses sufficient to p r o m o t e action potentials on the growing nerve. Unfortunately we cannot d e t e r m i n e from our histological data whether enhanced sprouting involved myelinated or unmyelinated fibers. I n d e e d , there is some uncertainty as to whether silver stains can reveal unmyelinated fibers in the light microscope (J. D i a m o n d , personal communication). The possible role of weak electric fields in the regulation of developmental and regenerative processes has been postulated 1. H o w e v e r previous reports
have been on the salutory effects of applied electric fields in lower vertebrates, or in embryos, or in vitro; hence the applicability to nerve regeneration in adult mammals in vivo has been open to question. I n d e e d lower vertebrates and embryos have long been known to have a greater regenerative capacity than adult mammals. H o w e v e r , recent evidence in young rats with mid-arm amputations, showed that DC electrical stimulation enhances limb regenerationS.14. Our finding in adult rat is the first to show that application of D C fields enhance nerve growth in mammals. These concepts can no longer be restricted to lower vertebrates. ACKNOWLEDGEMENTS This research was s u p p o r t e d by N S E R C of Canada. W e appreciate the able technical assistance of B. Dressier, P. F o r t n e r , D. Krushelnycky, E. Pomeranz and T. Sosath. W e thank Dr. J. D i a m o n d for showing us the histology technique.
REFERENCES 1 Borgens, R. B., The role of ionic current in the regeneration and development of the amphibian limb. In J. F. Fallon and A. I. Caplan, (Eds.), Limb Development and Regeneration, Alan Liss, New York, 1981, pp. 597-608. 2 Borgens, R. B., Roederer, E. and Cohen, M. J., Enhanced spinal cord regeneration in lamprey by applied electric fields, Science, 213 (1981) 611-617. 3 Borgens, R. B., Vanable, J. W. and Jaffe, L. F., Bioelectricity and regeneration: large currents leave the stumps of regenerating newt limbs, Proc. nat. Acad. Sci. U.S.A., 74 (1977) 4528-4530. 4 Borgens, R. B., Vanable, J. W., Jr. and Jaffe, L. F., Bioelectricity and regeneration. Initiation of frog limb regenera-, tion by minute currents, J. exp. Zool., 200 (1977) 403-416. 5 Devor, M., Schonfeld, D., Seltzer, Z. and Wall, P. D., Two modes of cutaneous reinnervation following peripheral nerve injury, J. comp. Neurol., 185 (1979) 211-220. 6 Diamond, J., The regulation of nerve sprouting by extrinsic influences. In F. O. Schmitt, (Ed.), The Neurosciences 4th Study Program, M.I.T., Cambridge, MA, 1979, pp. 937-955. 7 Hinkle, L., McCaig, L. D. and Robinson, K. R., The direction of growth of differentiating neurons and myoblasts from frog embryos in an applied electric field, J. Physiol. (Lond.), 314 (1981) 121-136. 8 Libbin, R. M., Person, P., Papierman, S., Shah, D., Nevid, D. and Grob, H., Partial limb regeneration of the above-el-
bow amputated forelimb, J. Morphol., 159 (1979) 439-452.. 9 Loots, G. P., Loots, J. M., Brown, M. M. and Schoeman, J. L., A rapid silver impregnation method for nervous tissue: a modified Protargol-peroxide technique, Stain Technol., 54 (1979) 97-100. 10 MacVicar, B. and Llinas, R., Calcium spikes in regenerating giant axons of the Lamprey spinal cord, Soc. Neurosci. Abstr., 8 (1982) 914. 11 Nixon, B., Jackson, P., Diamond, A., Foerster, A. and Diamond, J., Impulse activity evokes collateral sprouting of intact nerves into available target tissue, Soc~ Neurosci. Abstr., 6 (1980) 171. 12 Patel, N. and Poo, M. M., Orientation of neurite growth by extracellular electric fields, J. Neurosci., 2 (1982) 483-496. 13 Pomeranz, B., Markus, H. and Krushelnycky, D., Spread of saphenous somatotopic projection map in spinal cord and hypersensitivity of the foot after chronic sciatic denervation in adult rat, Brain Research, in press. 14 Smith, S. D., The role of electrode position in the electrical induction of limb regeneration in the subadult rat, Bioelectrochem. Bioenergetics, 8 (1981) 661-670. 15 Thoenen, H. and Barde, Y. A., Physiology of Nerve Growth Factor, Physiol. Rev., 60 (1980) 1284-1335. 16 Wall, P. D., Devor, M., Inbal, R., Scadiwl, J. W., Schonfeld, D., Seltzer, Z. and Tomkiewicz, M. M., Autotomy following peripheral nerve lesions: experimental anesthesia dolorosa, Pain, 7 (1979) 103-113.