The response to formalin in the second phase… , northwestern neurology

1116

• volume 2 no 12 • december 1999

articles

a

c

b

Fig. 3. ERK activation is NMDA receptor dependent. Capsaicin-induced pERK in the

ipsilateral dorsal horn (a) is partially suppressed by MK 801 (1.5 nmol) intrathecally

injected into spinal cord 20 min before the capsaicin (b). Scale bar, 50 µm. (c) Effect of

MK-801 on capsaicin-evoked ERK activation, measured by the number of pERK-positive

© 1999 . •

is not induced by Aβ-fiber stimulation of the attached ipsilat-

eral dorsal root (b), but is induced in the most superficial lay-

ers of the ipsilateral dorsal horn by C-fiber stimulation (c).

Scale bar, 50 µm. (d) Number of pERK-positive neurons per

section in the ipsilateral dorsal horn. Induction of pERK by C-

fiber stimulation is partially blocked by the NMDA receptor

antagonist APV (100 µM). *p 0.01, compared to control (a);

b

c

+

p 0.01, compared to C-fiber stimulation (d), n =5.

a dose-dependent manner, without significant effect

on the first phase (Fig. 5b). Over the entire course of

the second phase (10–60 min.), 0.1 µg or 1 µg PD

98059 inhibited pain behavior by 55% (p 0.01) and

71% (p 0.01), respectively (Fig. 5b). At these doses,

PD 98059 had no effect on basal mechanical or ther-

mal pain sensitivity in naive animals but inhibited

pERK immunolabel evoked by intraplantar capsaicin

(data not shown).

To investigate if timing of the MEK inhibitor

influenced the behavioral response, we administered

1 µg PD 98059 intrathecally 5 minutes after the for-

malin injection, that is, immediately after the first

phase. The response to formalin in the second phase

(10–60 min.) was still significantly inhibited (43%,

p 0.01) by such post-treatment. Interestingly, in

these animals, no significant decrease in pain behav-

ior occurred during the rising component of the sec-

neurons in the superficial laminae (I–II) of the ipsilateral dorsal horn. *p 0.01, as com – ond phase; the effect of the drug was delayed relative

pared to capsaicin group, n =4.

to its action when given before formalin injection

(Fig. 5b and c). These results imply that the early

component of the second phase is driven by input

generated in the first phase, and the falling compo-

(Fig. 4a and b). Aδ-fiber stimulation increased the number of

nent of the second phase is established after the first phase, pre-

pERK-positive neurons, but significantly less than C-fiber stim-

sumably by nociceptor afferent input evoked during the second

ulation (p 0.01; Fig. 4c and d). The C-fiber stimulus-evoked

phase. However, both phases involve ERK.

pERK was partially inhibited by the competitive NMDA recep-

tor antagonist APV (35% at 100 µM, p 0.01; Fig. 4d).

ERK activation and formalin-induced pain hypersensitivity

Formalin (1.5%, 50 µl) was injected into the plantar surface of

the hindpaw of awake rats, and the time the rats spent licking or

lifting the injected paw was measured over five-minute intervals

for an hour as an index of pain behavior. Control animals showed

a biphasic behavioral response. The first phase (0–5 minutes)

results from activation of nociceptors. This was followed, after a

short recovery, by a second phase (10–60 minutes). Based on its

sensitivity to centrally applied NMDA receptor antagonists18,27

and the differential effects of pre – and post-treatment with

intrathecal opioids18, this second phase has been interpreted as

an expression of use-dependent changes in spinal neurons, initi-

ated by activity generated during the first phase. Nevertheless,

input from the periphery also develops during the second

phase29,30, which may sustain tonic pain behavior31. Like capsaicin,

formalin injection into the hindpaw induced unilateral rapid ERK

activation in isilateral superficial laminae. The level of pERK

peaked at 3 minutes, declined at 8 minutes, but was

still higher than baseline at 60 minutes (Fig. 5a).

Intrathecal injection of PD 98059, a MEK

inhibitor that blocks phosphorylation of the ERKs32,

suppressed the second phase of the formalin test in

a

Fig. 4. Aδ- and C-fiber-dependent activation of ERK in vitro.

Levels of pERK are low in control spinal cord slices (a). ERK

DISCUSSION

Although ERK activation in most cells is driven by growth fac-

tors2,33, activity-dependent activation occurs in many neu-

rons5,6,10,34. We have found that ERK 1 and 2 are activated in the

spinal cord following peripheral stimulation. The phosphoryla-

tion of these kinases, however, is specific to noxious stimuli and

highly spatially and temporally organized.

ERK phosphorylation in the spinal cord is not simply activi-

ty dependent because natural or electrical input evoked in low-

threshold Aβ-fibers does not initiate it, even though such input

activates many cells in the dorsal and ventral horn. It is a stimu-

lus-specific, activity-dependent phosphorylation. Only noxious

peripheral stimuli (thermal, mechanical or heat) or Aδ-or C-fiber

stimulation activate ERK in the dorsal horn, in a manner that

encodes stimulus intensity. Moreover, the pERK-labeled neurons

have a highly restricted anatomical distribution. The mediolat-

eral and rostrocaudal location of labeled cells in the lumbar spinal

cord follows the somatotopic architecture of the central termi-

d

Connected themes: neuroscience network, neuroscience neurology, neuroscience new york, The response to formalin in the second phase… , northwestern neurology, neuroscience news, neuroscience news articles, neuroscience nih.