Regulation of myelinated nociceptor function by nerve growth factor in neonatal and adult rats

036 l-9230/93$6.00 + .OO

Brain ResearchBulletin, Vol.30, pp. 245-249, 1993 Printed in the USA. All rights reserved.

Copyright0 1993Pergamon Press Ltd.

Regulation of Myelinated Nociceptor Function by Nerve Growth Factor in Neonatal and Adult Rats AMY

M. RITTER,’

GARY

R. LEWIN

AND

LORNE

M. MENDELL’

Department of Neurobiology and Behavior, State University of New York at Stony Brook, Stony Brook, NY

11794

RITTER, A. M., G. R. LEWIN AND L. M. MENDELL. Regulation qf myelinated nociceptorfunction by nerve growth factor in neonatal and adult rats. BRAIN RES BULL 30(3/4) 245-249, 1993.-The role of nerve growth factor (NGF) as a survival factor for sensory neurons during embryonic life has been well documented. Here we examine the actions of NGF or antisera against NGF (anti-NGF) on physiologically identified sensory neurons with myelinated axons later in life, after the dependence on NGF for survival ends. We find that the effects of NGF and anti-NGF are specific for sensory neurons which are nociceptors. Treatments were found to affect the biophysical properties, the development, or the physiological function of myelinated nociceptors. They also affect the animal’s behavioral response to noxious stimulation, depending upon when the treatments were given: neonatally, from 2-5 weeks of age, or chronically, beginning at birth. Thus, we find that the actions of NGF are specific for nociceptors but that the function of this neurotrophic factor changes according to the developmental age of the animal. Nerve growth factor

Sensory neurons

Nociception

Hyperalgesia

Development

Rat

METHOD

IF rats are exposed to antibodies against nerve growth factor (NGF) in utero about 80% of dorsal root ganglion (DRG) neu-

Neonatal rats (Sprague-Dawley) were treated with NGF or anti-NGF, administered every day for the first 7 days and then every other day for the duration of the treatment, at doses of 5 pi/g SC for anti-NGF (9,14) and either 2 pg/g SC or 1 Kg/g IP for NGF [see (13) for details]. Intracellular recordings were made under cr-chloralose anaesthesia (75 mg/kg) (13,16), and extracellular recordings were made under urethane anaesthesia (1.25 g/kg) (9,14) after animals had reached maturity at 5 weeks of age or older. Mechanical thresholds were evaluated by determining the stiffness ofthe von Frey hair necessary to evoke limb withdrawal in response to stimulation of the dorsum of the foot.

rons die (6), indicating that during this early developmental period most of these neurons are critically dependent upon NGF for survival. This dependence on NGF declines over time, so that by the time animals have reached adulthood NGF deprivation does not result in any cell death among sensory neurons (5,14). Despite this apparent insensitivity to NGF deprivation in the adult, about one-half of DRG cells maintain high affinity receptors for NGF (12) and continue to transport this factor from the periphery (19). It appears that as animals mature, the function of peripherally derived NGF changes and the population of sensory neurons which utilizes it becomes more restricted. Both in the neonatal animal and in the adult, the DRG forms a heterogeneous group of cells which innervate a variety of targets in the periphery, and even within a target, such as skin, these afferents subserve a number of different physiological functions (1). We have been interested in the role that NGF plays in the function of sensory neurons in adult animals and also in neonatal animals during the period of time that sensory neurons are still developing but beyond the period of time that NGF is still required for survival. Our approach has been to treat animals either chronically or for restricted periods with antisera against NGF (anti-NGF) or with exogenous NGF, and then to examine the physiology of identified primary afferents.

RESULTS We found that irrespective of the type of treatment given or the aspect of sensory neuron function examined, the effects of either anti-NGF or NGF on myelinated afferents are specific for high-threshold mechanoreceptors (HTMRs). No group of lowthreshold cutaneous afferents responding to innocuous stimulation of the skin, i.e., hair follicle afferents, touch domes, etc., was affected by these treatments in any discernible way. Moreover, these treatments had no effect on the receptive field organization of individual afferents or on the organization of the sural nerve territory. Brush field mappings of the territory in treated animals were the same as in controls, and of the 111 cutaneous HTMRs in treated animals which were rigorously

’ Present address: Laboratory of Neural Control, NINDS/NIH, Bldg 36, Bethesda, MD 20892. ’ To whom requests for reprints should be addressed.

245

246

RITTER,

characterized and mapped, none was found with a receptive field extending outside the sural nerve territory. CASES WHERE TREATMENTS

PRODUCE

RECIPROCAL

RESULTS

It might be expected that the results of treatment with NGF and anti-NGF would be symmetrical. If treatment with antiNGF depressed some aspect of primary afferent function, then treatment with NGF would elevate it. We do not find this to be true in all cases. However, one instance in which it is clearly the case is in regulation of the width of the somal action potential (AP) of high-threshold mechanoreceptive afferents, both cutaneous HTMRs and high-threshold afferents with subcutaneous receptive fields. Action potentials (APs) in these cells are distinctive in shape with a large amplitude, long after hyperpolarization (AHP) and an inflection on the falling limb of the spike [(7): Fig. I A, top record]. This is reminiscent of the shape of the AP of immature spinal neurons (18) and somatic spikes of this type in dorsal root ganglion cells are insensitive to TTX ( 1520). In contrast, the somal APs of low-threshold afferents are briefer, smaller in amplitude, have short duration AHPs, lack the inflection on the falling phase, and are TTX-sensitive (Fig. lA, bottom record). Somal APs recorded from high-threshold afferents in animals treated chronically (birth to 5 weeks) with NGF had falling phases which were significantly longer in duration than those in controls. The opposite was true for spikes of highthreshold afferents in animals chronically treated with anti-NGF: the length of the falling phase of these spikes was briefer than in controls (Fig. 1B). The narrow spikes of low-threshold afferents were unaffected by either treatment. Thus, pharmacological doses of NGF can affect the membrane properties of nociceptive sensory neurons. That anti-NGF treatment produced the opposite change in spike shape as NGF treatment suggests that in vivo NGF may have some role in regulating the electrical properties of these cells. NGF treatment can also affect the physiology of cutaneous HTMRs. When NGF was administered from birth to 2 weeks of age, thinly myelinated cutaneous HTMRs (A6 HTMRs) became more sensitive to mechanical stimuli. Mechanical thresholds of slowly adapting HTMRs, as measured with calibrated von Frey hairs, dropped from a mean of 4.3 f 2.3 g (SD) in controls to 2.5 -+ 1.2 g in NGF-treated animals. This is illustrated in the cumulative sum distribution of Fig. 2, in which the curve for thresholds from NGF-treated animals is shifted to the left of that from controls. This sensitization gradually wore off: about 5 weeks after treatment had ceased, mechanical thresholds approached those of controls. Conversely, if anti-NGF was administered earlier than postnatal day (PND) 4 and extending at least until PND I 1, then cutaneous HTMRs became less sensitive to mechanical stimuli, and mean thresholds increased to 9.5 f 6.7 g. The cumulative sum distribution of thresholds is shifted to the right, toward higher values of threshold than those in controls (Fig. 2). Like the effect on spike duration, the effects of NGF and antiNGF on mechanical threshold were reciprocal. Treatment with NGF during the neonatal period resulted in a reduction of the thresholds of cutaneous A6 HTMRs, and treatment with antiNGF over a similar time period raised the mechanical thresholds of this population of afferents. However, effects on HTMR threshold could be evoked only during the first few weeks after birth. a period of time during which primary afferent physiology is still developing (4). Treatment of rats after the second postnatal week with either NGF or anti-NGF did not produce any effect on the mechanical threshold of cutaneous Afi HTMRs.

LEWIN AND MENDELL

CASES WHERE TREATMENT DID NOT PRODUCE RECIPROCAL RESULTS

NGF and anti-NGF treatment did not produce opposite effects in all instances. Treating animals from the end of the second week after birth until maturity (5 weeks of age) with NGF produced a behavioral hyperalgesia to mechanical stimulation while not affecting the physiology of A6 HTMRs. Thresholds for foot withdrawal from a calibrated von Frey hair were only 10% of those in control animals, thus making these animals much more sensitive to noxious mechanical stimuli. However, if animals were treated with anti-NGF over the same time period, no change in behavioral threshold to mechanical stimuli was observed. Another nonreciprocal effect arose after treatment with NGF and anti-NGF within the first 2 postnatal weeks. Administration of anti-NGF resulted in a severe depletion in the number of cutaneous A6 HTMRs in the sural nerve. At the same time, the proportion of a type of low-threshold mechanoreceptor, D-hair afferents, increased to the same extent that A6 HTMRs were lost (9,14). This change could occur if treatment was begun on or later than PND 2, without the 20% death of sensory neurons that occurs if treatment is begun on the day of birth (9). We interpret this to mean that developing HTMRs are induced to become D-hairs in the absence of NGF. However, treatment with NGF from birth to PND 14 did not cause a switch in the opposite direction: the proportion of cutaneous A6 HTMRs did not increase. nor did the proportion of D-hairs decrease. DISCUSSION

Those NGF or anti-NGF treatments which produced effects by acting within the first 2 postnatal weeks were acting over a period of time during which the innervation of skin is still developing. When sensory axons grow out to skin, they initially form a dense plexus in the epidermis which persists for several days (3). Most fibers then withdraw to innervate structures in the dermis; for sensory neurons innervating the skin of the hindlimb, this shift takes place between PND 5-10 (1 I). In the mature animal, some fibers remain in the epidermis, and it is thought that some of these intraepidermal axons are the anatomical substrates of A6 HTMRs (8). In embryonic mice it has been found that levels of NGF mRNA are more concentrated in the epidermis than in the dermis (2). We hypothesize that relatively high levels of NGF in the epidermis are necessary to stabilize the projections of developing cutaneous A6 HTMRs. When these levels are lowered by treatment with anti-NGF, fibers are forced to withdraw into the dermis and, subsequently, innervate hair follicles. This hypothesis can account for many of the effects that we see with NGF or anti-NGF treatments. First, the hypothesis does not maintain that NGF is responsible for the formation of nociceptors, merely that it provides a hospitable environment in which these afferents can develop. This is consistent with the fact that addition of NGF did not create a surplus of nociceptors. Second, the effects on mechanical thresholds may be envisaged as being brought about by an anatomical change in peripheral terminals. Mechanical thresholds were raised after anti-NGF administration only if the treatments were given over a period of time during which HTMRs can be converted into D-hairs. Thus, the population of HTMRs that were affected are those few which did not convert. It might be that these few HTMRs that remain have a less dense projection into the epidermis after treatment. In addition, the fibers may not penetrate as far into the epidermis, may have simpler endings, or less of whatever it is that confers mechanical sensitivity to an afferent. Indeed, pre-

NGF

AND N~ICEPTIV~

247

~NCTl~N

A. Somal

Spikes

of

in Controt

DRG Cells Rats

dV,‘dt --+-r----fb

Hair CV= 10.1 m/s dV/dt

EL Mean Spikes

’

1 .O

i=

08. 0.6

= lE

s-i , i

Fall of

Times

of

Somat

~yeiina~~d

HTMRs

0.4 0.2 0.0

ANTI-NCF

NGF

CONTROL

Treatment

Group

FIG. t. Effects of anti-NGF and NGF treatments on the somal spikes of HTMRs. (A) lntmcellular recordings from untreated control animals of the somal spike of an A&HTMR (top trace) and of the somal spike of a low-threshold afferent (third trace), in this case a Dhair afferent. Beneath each trace is a diffe~ntiated record. dV/dt, illustrating in the case of the HTMR the presence of a hump on the descending limb of the spike. For both spike records, ~Iibmtion (square wave at the finning of the spike trace) = 10 mV, 1 ms.(B) Bar graph illustrating mean duration of the falling phase of somal spikes of HTMRs after NGF or anti-NGF treatments. Error bars represent standard error of the mean.

limina~ results indicate that the few calcitonin gene-related peptide (CGRP) immunoreactive fibers which remain in the epide%nIis after anti-NGF treatment do not penetrate as far as do those in control animals (Lewin, McMahon, Tonra, and MendelI, unpublished obse~ations}. Conversely, we speculate that treatment with NGF while afferents are still developing may induce HTMRs to form more profusely branching and, thus, more sensitive endings in the epidermis but not cause the formation of any additional HTMRs. If NGF levels are not maintained at this high level, then after treatment ends these arbors may gradually be trimmed back, which would account for the temporary nature of the mechanical sensitization after NGF treatment. The fact that it is not possible to affect mechanical ~resholds of cutaneous HTMRs with NGF or anti-NGF treatments past

the period of time during which skin innervation is maturing (i.e., beyond PND 14) indicates that this may represent a time window during which this particular type ofafferent is sensitive to NGF. Also, the function

of NGFduring

this period is not the

same as it is in less mature animals, i.e., the changes we see are independent of anti-NGF-induced cell death. The rofe of NGF in the regulation of nociceptor function in the adult is not as clear. Regulation of action potential width could be obtained with chronic NGF treatment, but it is not known whether treatment beginning past PND 14 would duplicate this efhxt on nociceptors. After PND 14, NGF treatment was capable of producing a profound ~havioral hyperalgesia in the absence of any tissue damage or inflammation. This treatment did not affect the peripheral physiology of A6 afferen& yet it appears that NGF is still capable of interacting with

248

RITTER,

LEWIN

AND

MENDELL

Cumulative Percent of Units

1

10

Threshold (g) *

Neonatal

NGF

-+-

Control

*

Neonatal

Anti-NGF

N=31 FIG. 2. Cumulative sum distributions of the mechanical thresholds of slowly adapting cutaneous HTMRs in untreated

control rats and rats treated with NGF or anti-NGF. Thresholds were obtained with calibrated von Frey hairs applied to the most sensitive spot in the receptive field of each unit. Each point represents the percentage of units (ordinate) having a threshold at or below the value indicated on the abcissa, which is determined by the force produced by the von Frey hair. In these experiments, NGF was administered between PND O-14. Data from animals treated with anti-NGF were obtained from animals treated between PND 0- 14, PND 2- 14, PND 4-I I. or from PND 0 to 5 weeks of age. nociceptive systems at this late age, perhaps due to regulation of the central connectivity of these afferents (10). It is possible that NGF may exert its effects by acting on C-fibers, which would be consistent with the fact that NGF can regulate substance P levels even in the adult animal (17). Such a possibility is currently under investigation. Although these experiments do not clearly define the function of NGF in mature animals, it is obvious that this function is different from that in neonatal animals.

CONCLUSIONS

These results demonstrate that the actions of NGF are very specific for a physiologically defined subclass of sensory neuron, namely mechanical nociceptors, and that its effects on nociceptors in the intact animal go well beyond that of simply regulating cell survival. Furthermore, the role that NGF plays in the regulation of nociceptor function changes according to the developmental stage of the animal.

REFERENCES I. Burgess, P. R.; Perl, E. R. Cutaneous mechanoreceptors and nociceptors. In: Iggo, A., ed. Handbook of sensory physiology. vol. 11: Somatosensory system. New York: Springer; 1973:29-78. 2. Davies, A. M.; Bandtlow, C.; Heumann, R.; Korsching, S.; Rohrer, H.; Thoenen, H. Timing and site of nerve growth factor synthesis in developing skin in relation to innervation and expression of the receptor. Nature 326:353-358; 1987. 3. Fitzgerald, M. J. T. Perinatal changes in epidermal innervation in rat and mouse. J. Comp. Neurol. 126:37-42; 1967. 4. Fitzgerald, M. Cutaneous primary aBerent properties in the hindlimb of the neonatal rat. J. Physiol. 383:79-82; 1987. 5. Gorin, P. D.; Johnson, E. M., Jr. Effects of long-term nerve growth factor deprivation on the nervous system of the adult rat: An experimental autoimmune approach. Brain Res. 198:27-42; 1980. 6. Johnson, E. M., Jr.; Gorin, P. D.; Brandeis L. D.; Pearson, J. Dorsal root ganglion neurons are destroyed by exposure in utero to maternal antibody to nerve growth factor. Science 219:9 16918; 1980. 7. Koerber, H. R.; Druzinsky, R. E.; Mendell, L. M. Properties of

8.

9. 10. Il. 12. 13.

somata of dorsal root ganglion cells differ according to peripheral receptor innervated. J. Neurophysiol. 60: 1584- 1595; 1988. Kruger, L.; Perl, E. R.; Sedivec, M. J. Fine structure of myelinated mechanical nociceptor endings in cat hairy skin. J. Comp. Neurol. 198:137-154; 1981. Lewin, G. R.; Ritter, A. M.; Mendell, L. M. On the role of nerve growth factor in the development of myelinated nociceptors. J. Neurosci. 12:1896-1905; 1992. Lewin, G. R.; Winter, J.; McMahon, S. B. Regulation of afferent connectivity by nerve growth factor in the adult rat spinal cord. Eur. J. Neurosci. 4:700-707; 1992. Reynolds, M. L.; Fitzgerald, M.; Benowitz, L. I. GAP-43 expression in developing cutaneous and muscle nerves in the rat hindlimb. Neuroscience 41:200-211; 1991. Richardson, P. M.; Verge Issa, V. M. K.; Riopelle, R. J. Distribution of neuronal receptors for nerve growth factor in the rat. J. Neurosci. 6:2313-2321; 1986. Ritter, A. M. The effects of nerve growth factor and its antisera on physiologically identified sensory neurons. PhD Dissertation, Department of Neurobiology and Behavior, SUNY-Stony Brook; 1991.

NGF AND NOCICEPTIVE F-UNCTION 14. Ritter, A. M.; Lewin, G. R.; Kremer, N. E.; Mendell, L. M. Requirement for nerve growth factor in the development of myelinated nociceptors in vivo. Nature 350500-50 I ; 1991. 15. Ritter, A. M.; Mendell, L. M. The somal spikes of physiologically identified high threshold mechanoreceptors is insensitive to TTX. Pain Suppl. 5:s 110; 1990. 16. Ritter, A. M.; Mendell, L. M. The somal membrane properties of physiologically identified neurons in the rat: Effects of nerve growth factor. J. Neurophysiol. (in press). 17. Schwartz, J. P.; Pearson. J.; Johnson, E. M., Jr. Effect of exposure

249 to anti-NGF on sensory neurons of adult rats and guinea pigs. Brain Res. 244:378-381; 1982. 18. Spitzer, N. C. Ion channels in development. Annu. Rev. Neurosci. 21363-97; 1979. 19. Thoenen, H.; Barde, Y.-A. The physiology of nerve growth factor. Physiol. Rev. 60: 1284-1335; 1980. 20. Yoshida, S.; Matsuda, Y.; Samejima, A. Tetrodotoxin-resistant sodium and calcium components of action potentials in dorsal root ganglion cells of the adult mouse. J. Neurophysiol. 4 I: 1096- 1106; 1978.