Neuroscience Letters, 85 (1988) 187 192 187

Elsevier Scientific Publishers Ireland Ltd.

NSL 05137

Co-localization of proenkephalin- and

prodynorphin-derived opioid peptides in laminae
IV/V spinal neurons revealed in arthritic rats

E. Weihe l, M.J. Millan 2, A. Leibold l, D. N o h r I a n d A. H e r z 2

JDepartment qf Anatomy, Johannes Gutenberg- Universitdt, Mainz (F.R.G.) and 2Department ~['
Neuropharmacology, Max-Planck-lnstitut fiir Psychiatrie, Martinsried (F. R.G. )
(Received 28 September 1987; Revised version received 30 October 1987: Accepted 30 October 1987)

Key word~': Prodynorphin; Proenkephalin; Coexistence; Spinal cord; Arthritic pain

By the use of highly selective antisera and an immunohistochemical technique the possible coexistence
of proenkephalin- (PRO-ENK)- and prodynorphin (PRO-DYN)-derived peptides was examined in 4- to
6-Ftm-thick serial sections of the L4 L~ segments of the spinal cord of non-colchicine-treated polyarthritic
rats. In control, non-colchicine treated animals, virtually no cell bodies stained for the PRO-ENK-derived
peptides, heptapeptide (MRF) and octapeptide (MRGL), nor for the PRO-DYN-derived peptides, dynor-
phin A (DYN) and ~-neoendorphin (NEO). In contrast, in polyarthritic rats, numerous large (15 30/Lm)
multipolar neurons could be visualized with each antiserum in laminae IV/V. Alternate staining of adja-
cent sections with either anti-MRF or anti-MRGL antisera, followed by either anti-DYN or anti-NEO
antisera, revealed a clear coexistence of P R O - E N K and PRO-DYN peptides. It was possible to demon-
strate co-localization of all 4 opioids in a single cell. It appeared that all cells staining for PRO-ENK pep-
tides in laminae IV/V also stained for P R O - D Y N peptides.

Rats subjected to the pain of chronic polyarthritis reveal pronounced alterations

in the functional activity of central nervous system (CNS) opioid systems. In particu-
lar, there is a very marked influence upon the levels of prodynorphin (PRO-DYN)-
derived peptides in the spinal cord, whereas pools of proenkephalin (PRO-ENK)
peptides appear to be comparatively less affected [8, 9]. Recently, we have shown that
the level of m R N A encoding D Y N is increased in the lumbar cord of polyarthritic
rats [6]. Further, by the use of immunohistochemistry we demonstrated that the
DYN neurons affected are located in laminae I/II and IV/V of non-colchicine-treated
arthritic rats (Weihe et al., submitted). In an extension of this work we wished to
determine, by use of light microscopic (LM) immunohistochemistry, whether a popu-
lation of spinal cord P R O - E N K neurons may be modulated by chronic arthritic pain.

Correspondence: E. Weihe, Anatomisches Institut, Johannes Gutenberg-Universitfit Mainz, Saarstral3e 19-

21 D-6500 Mainz, F.R.G.
188

In particular, we attempted to evaluate whether there may be a common population

of PRO-ENK- and PRO-DYN-expressing neurons visualizable in polyarthritic ani-
mals.
Chronic polyarthritis was induced in male Wistar rats by the injection of Mycobac-
terium butyricum in the tail-base as described previously [9]. Three weeks later, when
peak symptoms had developed, polyarthritic ( n - 6 ) and control rats ( n = 5 ) were
anesthetized with sodium pentobarbital (60 mg/kg i.p.) and fixed by perfusion with
Bouin's solution [16-18]. Animals were not treated with colchicine prior to fixation.
The L4/L 5 spinal cord segments were processed for LM immunohistochemistry on
paraffin sections using biotinylated secondary antisera, streptavidin-biotin-peroxi-
dase complexes (Amersham), and the chromogen diaminobenzidine (Dakopatts) for
the visualization of immunoreactions [16-18]. In order to determine the distribution
and coexistence patterns of the two opioid families, 4 to 6-/~m-thick adjacent sections
were alternately stained with primary antisera against opioid sequences contained in
either PRO-ENK, i.e. heptapeptide (MRF) and octapeptide ( M R G L ) or in PRO-
DYN, i.e. DYN A(I-17) and ~-neoendorphin (NEO). These antisera have previously
been extensively characterized by immunohistochemistry and radioimmunoassay [9,
16-18]. Up to a concentration as high as 10/~M there was no mutual cross-reactivity
in pre-absorption studies although they all detected C-terminally extended higher
molecular weight forms of their respective antigens. The anti-MRGL and anti-MRF
antisera were absolutely specific for PRO-ENK. Anti-DYN and anti-NEO were
equally selective for PRO-DYN. There was no evidence for cross-reactivity with a
variety of heterologous antigens. The localization of immunoreactions was deter-
mined in relation to Rexed's laminae.
In control rats, P R O - E N K (MRGL)-ir varicose fibers generally outnumbered
those containing ir-PRO-DYN (DYN or NEO). Ir-DYN and ir-NEO fibers were co-
distributed. P R O - E N K - and PRO-DYN-ir varicose fibers were similarly concen-
trated in laminae I/II and IV/V, as well as in lamina X, particularly in the dorsal
grey commissure. The higher density of M R G L - i r varicose fibers compared to those
containing ir-DYN or ir-NEO was particularly striking in laminae I/II. In contrast
to extremely rare fiber staining for ir-DYN or ir-NEO in lamina III, substantial
numbers of M R G L - i r varicose fibers were also present in this lamina. MRF-ir vari-
cose fibers were generally less frequent but completely co-distributed with ir-MRGL.
Perikaryal staining for P R O - E N K was virtually absent from all laminae in the L4/L5
segments of control rats. Very occasionally, faint D Y N / N E O - i r perikarya were pre-
sent in laminae I/II but not in IV/V.
In polyarthritic rats, the distribution pattern of varicose fibers in the L4/L5 seg-
ments staining for P R O - E N K - and PRO-DYN-specific opioids was similar to con-
trol animals. The density of DYN/NEO-ir varicose fibers and the intensity of their
immunostaining was higher in arthritic than in control rats. In contrast, it was not
possible to determine such differences as regards varicose fiber staining for M R G L
or MRF. A striking difference to control animals was the appearance of large
numbers of heavily stained D Y N / N E O - and M R G L / M R F - i r perikarya in laminae
IV/V. Their diameters were in the range of 15-30/zm. Up to 5 perikaryal profiles
 189

Fig. 1. Light micrographs of 4 adjacent sections of L5 spinal cord segment (lamina V) of a polyarthritic
rat alternately stained with antisera specific for PRO-ENK (b,d) and PRO-DYN opioid peptides (a,c);
neuronal profiles exhibiting coexistence are numbered; note coexistence of all 4 opioid peptide immuno-
reactivities in one single neuron (1); a~d x 400.

occurred in one dorsal h o r n section plane. The n e u r o n s were best classified as multi-
polar. A d j a c e n t sections along the entire extent of the L4/L5 segments revealed that
P R O - D Y N a n d P R O - E N K - s p e c i f i c ir-opioid peptides coexisted in this distinct
n e u r o n a l cell p o p u l a t i o n o f l a m i n a e IV/V (Fig. 1). It a p p e a r e d that all l a m i n a e IV/V
190

neurons which stained for P R O - E N K also stained for P R O - D Y N and vice versa. In
other laminae, no significant 'lighting up' of neurons staining for PRO-ENK
( M R G L or MRF) was visualized in polyarthritic rats. In contrast, intensely stained
DYN/NEO-ir perikarya were also present in laminae I/II of the dorsal horn as will
be reported in more detail (Weihe et al., submitted).
We were able to show that there was a distinct PRO-ENK-related response of
laminae IV/V spinal neurons to polyarthritis, and this paralleled that of P R O - D Y N
neurons in these laminae. However PRO-DYN, in contrast to P R O - E N K neurons,
also responded in laminae I/II. Further, whereas spinal PRO-ENK-varicose fibers
appeared not to be greatly influenced by the arthritis, the P R O - D Y N response also
comprised a marked increase in the intensity and density of varicose fiber staining.
These differences may explain why, in polyarthritis, the increase in spinal levels of
P R O - E N K peptides and PRO- ENK m R N A is less than that of PRO-DYN peptides
and its m R N A [6].
Our major finding is that P R O - E N K and P R O - D Y N neurons in laminae IV/V
which were 'lit up' by polyarthritis are identical. In this respect, it is important to
re-emphasize the high degree of selectivity of our antisera. Further, no co-localization
of PRO-ENK with PRO-DYN was seen in laminae I/II and this strengthens the ar-
gument that the M R G L / M R F antisera do not cross-react with P R O - D Y N products.
We provide, thus, unequivocal evidence for the coexistence of two different PRO-
ENK and two different PRO-DYN-derived opioid products in individual lamina IV/
V spinal cord neurons. Previously a coexistence of P R O - D Y N and P R O - E N K in the
spinal cord was suggested to occur in small neurons of the dorsal grey commissure
(L6-S1) in the rat [13]. However, in view of the mutual cross-reactivity of the antisera
against Met-enkephalin (ME) and Leu-enkephalin (LE) used [4, 17, 18], no conclu-
sive evidence could be obtained. Indeed, we are aware of only one study, that of
Guthrie and Basbaum [5], who showed co-occurrence of P R O - D Y N (as ir-DYN B)
and PRO-ENK (as ir-MRGL) in single medullary neurons of rat, which has convinc-
ingly demonstrated a coexistence of P R O - E N K and P R O - D Y N in neurons in the
CNS. On the other hand, coexistence of PRO-ENK and P R O - D Y N seems not to
be unique to the CNS since it has also been observed in adrenal medullary cells [3,
18] and in other peripheral neurons [18]. The exact nature and function of the lami-
nae IV/V neurons co-expressing P R O - E N K and PRO-DYN is unclear. From the
neuroanatomical, neurophysiological and neuropharmacological data available [I, 2.
4, 7, 8, 15, 19] it is probable that at least some are interneurons that are involved
in the processing of noxious input. It is also possible that some may be projection
neurons to the brain. Indeed, that certain spinal opioid neurons project rostrally has
been demonstrated by spinal transection studies [11]. Opioid marginal neurons were
shown to project to the parabrachial nucleus [14] and the presence of enkephatin-
containing spinoreticular neurons also has been reported [10].
In conclusion, we have demonstrated a coexistence of P R O - E N K and PRO-DYN
opioids in laminae IV/V neurons and fibers of lumbar (La/Ls) spinal cord segments
of the non-colchicine-treated polyarthritic rat. It is probable that such a co-localiza-
tion also occurs in control tissue but that levels are too low to be visualized. Alterna-
 191

tively, a second (PRO-ENK or PRO-DYN) gene might be 'induced' under chronic

polyarthritic pain. In any case, the data have important implications for our under-
standing of the functional role of opioid systems in pain modulation. Previous studies
have suggested the possibility of interactions between various co-released opioids (or
other co-stored products) in the control of nociception in the brain (cf. ref. 8) or spi-
nal cord [12]. There may be a co-ordinated synergistic action (rather than an indepen-
dent) of multiple opioid peptides derived from PRO-ENK- and PRO-DYN-synthe-
sizing neurons. This finding is of direct clinical relevance in view of the possibility
of co-joint application of particular opioids to improve their efficacy.

We would like to thank Dr. F.C. Colpaert for providing us with the arthritic rats
and Dr. E. Weber for the generous supply of octapaptide antiserum. This study was
supported by the Deutsche Forschungsgemeinschaft, Bonn.

I Botticelli, L.J., Cox, B.M. and Goldstein, A., Immunoreactive dynorphin in mammalian spinal cord
and dorsal root ganglia, Proc. Nat. Acad. Sci. USA, 78 (1981) 7783 7786.
2 Cruz, L. and Basbaum, A.I., Multiple opioid peptides and the modulation of pain: immunohistochemi-
cal analysis of dynorphin and enkephalin in the trigeminal nucleus caudalis and spinal cord of the cat,
J. Comp. Neurol., 240 (1985) 331 348.
3 Evans, J.C., Erdelyi, E., Hunter, J. and Barchas, J., Colocalization and characterization of immuno-
reactive peptides derived from two opioid precursors in guinea pig adrenal glands, J. Neurosci., I2
(1986) 3423 3428.
4 Fallon, J.H. and Leslie, F.M., Distribution of dynorphin and enkephalin peptides in the rat brain, J.
Comp. Neurol., 249 (1986) 293 336.
5 Guthrie, J. and Basbaum, A.I., Colocalization of immunoreactive proenkephalin and prodynorphin
products in medullary neurons of the rat, Neuropeptides, 4 (1984) 437 445.
6 H611t, V., Haarmann, I., Millan, M.J. and Herz, A., Prodynorphin gene expression is enhanced in thc
spinal cord of chronic arthritic rats, Neurosci. Len., 73 (1987) 9(~94.
7 Merchenthaler, I., Maderdrut, J.L., Altschuler, R.A. and Petrusz, P., lmmunocytochemical localiza-
tion of proenkephalin derived peptides in the central nervous system of the rat, Neuroscience, 17 (1986)
325 348.
8 Millan, M.J., Multiple opioid systems and pain, Pain, 27 (1986) 303 349.
9 Millan, M.J., Millan, M.H., Czlonkowski, A., H611t, V., Pilcher, C.W.T., Herz, A. and Colpaert, F.C.,
A model of chronic pain in the rat: response of multiple opioid systems to adjuvant-induced arthritis,
J. Neurosci., 6 (1986) 899 906.
10 Nahin, R.L., Micevych, P.E. and Liebeskind, J.C., Neurochemical identification of afferents onto spi-
nomedullary neurons in the rat spinal cord central gray matter, Brain Res., 401 (1987) 292 302.
11 Pickel, V.M., Miller, R., Chan, J. and Sumal, K.K., Substance P and enkephalin in transected axons
of medulla and spinal cord, Regul. Pept., 6 (1983) 121 135.
12 Ren, M.F., Lu, C.H. and Han, J.S., Dynorphin-A-(l-13) antagonizes morphine analgesia in the brain
and potentiates morphine analgesia in the spinal cord, Peptides, 6 (1985) 1015 1020.
13 Sasek, C.A. and Elde, R.P., Coexistence ofenkephalin and dynorphin immunoreactivities in the dorsal
horn gray commissure of sixth lumbar and first sacral spinal cord segments in rat, Brain Res., 381
(1986) 8 14.
14 Standaert, D.G., Watson, S.J., Houghten, R.A. and Saper, C.B., Opioid peptide immunoreactivity in
spinal and trigeminal dorsal horn neurons projecting to the parabrachial nucleus in the rat, J. Neur-
osci., 6 (1986) 1220,1226.
15 Vincent, S.R., H6kfelt, T., Christensson, I., Terenius, L., Dynorphin-immunoreactivc neurons in the
central nervous system of the rat, Neurosci. Lett., 33 (1982) 185 190.
16 Weihe, E., Hartschuh, W. and Weber, E., Prodynorphin opioid peptides in small somatosensory pri-
192

mary afferents of guinea-pig, Neurosci. Lett., 58 (1985) 347--352.

17 Weihe, E., Leibold, A., Nohr, D., Fink, T. and Gauweiler, B., Co-existence of prodynorphin-opioid
peptides and substance P in primary sensory afferents of guinea-pig, NIDA Res. M onogr, 75 (19861
295 298.
18 Weihe, E., Nohr, D., Hartschuh, W., Gauweiler, B. and Fink, T., Multiplicity of opioidergic pathways
related to cardiovascular innervation: differential contribution of all three opioid precursors. In K.O.
Stumpe and K. Kraft (Eds.), Proceedings of the Satellite Symposium to the l lth Scientific Meeting
of ISH 1986, Opioid Peptides and Blood Pressure Control, Springer, Berlin, in press.
19 Zieglg~insberger, W., Central control of nociception. In V.B. Mountcastle, F.E. Bloom and S.R. Geiger
(Eds.), Handbook of Physiology The Nervous System, Vol. IV, Williams and Wilkins, Baltimore,
1986, pp. 581 675.